It is one of the fundamental paradoxes of world politics: while the norm of state sovereignty designates members of the international community as nominal equals, relations between states are fundamentally unequal.Footnote 1 Significant disparities exist in the contemporary international system, whether in terms of military capabilities, economic wealth, or political and cultural influence.Footnote 2 The vast majority of states exert little to no influence over the shape of world politics. Indeed, many states possess minimal control over their own foreign and domestic policies, which are more likely to be conditioned by the demands of powerful allies, rich economic partners, and international institutions than their own preferences.
Traditionally, scholars have downplayed the prevalence and importance of relations of inequality in world politics.Footnote 3 States were assumed to be sovereign entities that operated as equals in an anarchic international system. Order was maintained through the mechanism of the balance of power, in which no individual state obtained a disproportionate share of material capabilities. A growing number of scholars, however, have questioned this basic formulation.Footnote 4 Rather than see sovereignty as indivisible, they point to the various ways in which strong states exploit power asymmetries to establish authority over the weak. Far from being an exception, relations of hierarchy permeate the international system. Alexander Cooley, for example, points to the ways in which American overseas military bases fundamentally constrain the sovereignty of host societies.Footnote 5 Evelyn Goh likewise highlights how the United States has enmeshed states in East Asia in a hierarchical regional order.Footnote 6 The effects of hierarchy on world politics, moreover, can be profound. David Lake argues that military and economic dependence on the Untied States encourages subordinate states to spend proportionately less on defense.Footnote 7 Patrick McDonald likewise contends that the relative peace between democratic countries over the past hundred years is due to the ‘evolution in hierarchic relationships’ constructed by the United States.Footnote 8
Hierarchic relations of sovereign inequality are hardly a new phenomenon. In the eighteenth and nineteenth centuries, European states expanded overseas, establishing relations of imperial dominance over distant societies.Footnote 9 During the Cold War, the superpowers competed to secure political influence over allies and proxies around the world.Footnote 10 Yet while scholars acknowledge the importance of hierarchy, fundamental questions remain unanswered. How should International Relations scholars conceive of hierarchy? How do the politics of hierarchy differ from those under anarchy? What analytical approaches are best able to explain when hierarchies emerge and how they are sustained?
In this article, I argue that the existing literature on hierarchy, for all its diverse insights, misses what makes hierarchy unique in world politics. First, it employs a variety of competing definitions of hierarchy, many of which are overly broad. While it is tempting to define hierarchy as any form of inequality or social stratification, the most distinctive feature of hierarchy in world politics is that it is a relationship of sovereign control between nominal equals. A dominant actor possesses the authority to dictate to a subordinate, although the latter retains some degree of residual autonomy. This does not mean that other forms of inequality – whether in terms of power, resources, or status – are unimportant. Yet these forms of inequality can flourish even under conditions of anarchy, while anarchy itself can help produce and reinforce them. A narrow definition captures what makes hierarchy in world politics distinct: it involves the sharing of sovereignty between actors that retain some degree of political independence.
Second, the literature has erred by basing its theoretical development around a simple and misleading distinction between anarchy in international politics and hierarchy in domestic politics. Authority relationships in world politics, however, are dramatically different than those found in domestic politics. Hierarchy in domestic politics typically involves political authorities delegating the tasks of governance to administrators in the context of established legal regimes and formal procedures. In hierarchy in world politics, in contrast, dominant states seek to orchestrate their desired outcomes by enlisting and appropriating the authority of intermediaries within subordinate societies. As a consequence, hierarchy in world politics operates in more indirect, incomplete, and contested ways than it does in domestic politics. It is shaped less by written laws or standard procedures, than by more subtle forms of manipulation and the reproduction of informal practices.
Third, the literature has sought to explain hierarchy in world politics by focusing primarily on the attributes of dominant states. Some authors emphasise the interests and capabilities of dominant states: the extent to which they would benefit from hierarchy, and their capacity to supply it.Footnote 11 Other authors focus on the identities of dominant states: the degree to which they conceive of themselves of commanding positions of authority over other societies or peoples.Footnote 12 Yet these perspectives obscure the fundamental importance of subordinate actors in helping to fashion hierarchy in world politics. Because hierarchy in world politics depends on the enlistment and appropriation of alternate sources of authority, it cannot simply be reduced to the attributes of the dominant. Hierarchy in world politics is fundamentally a relational phenomenon. The position of actors within broader networks of political, economic, and social exchange shape not only who can claim the right to command, but also who will tend to obey.
Rather than reduce the politics of hierarchy to those of anarchy, I argue we need to develop a new set of relational network concepts to understand when and how hierarchy is constructed in world politics. Hierarchy is not just the byproduct of mutually beneficial bargains struck between unitary rational actors, but is enabled by different configuration of ties between actors in world politics. Dominant states tend to reside in central positions in networks of exchange, which provides them with the material and ideational resources required to enlist capable intermediaries. Effective intermediaries, in turn, tend to be positioned as brokers between external patrons and subordinate societies, which helps them orchestrate and legitimate foreign domination. In some situations, shifting configurations of social ties can result in the creation of new actors or the construction of competing sites of political authority, which can be appropriated in unanticipated ways by dominant states. In short, by adopting a relational network approach,Footnote 13 scholars can better understand the complicated processes that allow relations of hierarchy to both emerge and endure, despite the constraints imposed by the otherwise anarchic structure of world politics.
The remainder of this article proceeds as follows. The first section examines competing conceptions of hierarchy in world politics, and makes the case for a narrow conception of hierarchy as the appropriation of political authority. The second section compares how authority functions in world politics with how it does so in domestic politics, and argues that hierarchy amidst anarchy is a distinct system whose dynamics cannot be reduced to those one observers within states. The third section builds on this observation and describes how different relational network concepts – such as access, brokerage, and yoking – can help illuminate the processes by which authority is enlisted and appropriated in world politics. Throughout the article, I discuss the case of the East India Company’s subsidiary alliance system with Indian princely states to illustrate these points. Yet I conclude with applications to contemporary world politics, notably the United States’ continued leadership of the hierarchic, liberal international order.
Durable inequalities: Competing conceptions of hierarchy in world politics
One of the fundamental assumptions of almost all theories of world politics is that the structure of the international system is anarchic. Because there is no authority that can make and enforce laws, states are compelled to look out for their own interests, most notably security from external threats. As a result, hierarchy has not traditionally been a focus of International Relations scholars. Kenneth Waltz, for example, emphasises that international politics is an inherently ‘decentralized system’ in which each political unit ‘is the equal of all the others … none is entitled to command; none is required to obey’.Footnote 14 While he acknowledges that international systems can be seen as ‘more or less anarchic’, Waltz characterises the current international system as only ‘being flecked with particles of government and alloyed with elements of community’.Footnote 15 Because of the logic of self-help, states will rarely volunteer to have their sovereignty reduced. Similarly, because of the logic of balancing, states will rarely accumulate sufficient power to dominate others.Footnote 16
In practice, however, norms of state sovereignty are frequently violated.Footnote 17 States are often extinguished as political units, and powerful actors occasionally accumulate sufficient capabilities to ‘roll up’ the international system.Footnote 18 Moreover, composite political actors have always played an important role in international history, whether empires, suzerainties, dynasts, or federations.Footnote 19 As Barry Buzan and Richard Little note, ‘there has seldom if ever been a time when all political units have acknowledged that they were independent’.Footnote 20 John Hobson and Jason Sharman likewise argue, ‘in the period since 1648 neither the European nor the global political systems has been purely anarchical, but have instead exhibited varying combinations of ‘sub-systems’ of hierarchy alongside anarchy’.Footnote 21 Rather than being governed by a single ordering principle, in other words, the international system tends to be hybrid in character, with anarchy governing relations between certain sets of actors and hierarchy predominating relations between others.Footnote 22 As Alexander Barder concludes, ‘we should understand the history of international relations as a patchwork of various forms of hierarchical relations’.Footnote 23
While scholars acknowledge that relations of domination and subordination permeate world politics, they disagree about which should be characterised as hierarchy.Footnote 24 Some prefer a narrow conception of hierarchy that focuses on unequal sovereignty claims. According to David Lake, ‘hierarchy exists when one actor, the ruler … possess authority over a second actor, the ruled’.Footnote 25 John Ikenberry likewise defines hierarchy as ‘ordered relations between units where power and authority is centralized and the units in the system are functionally differentiated’.Footnote 26 Hierarchy, in this view, is the product of strategic interaction. Dominant states claim a right to dictate some aspect of a subordinate polity’s foreign or domestic policies.Footnote 27 These subordinates, whether by coercion or consent, are obligated to comply. Examples of this narrow conception of hierarchy are numerous. In classical overseas empires, European powers claimed the exclusive right to govern colonial possessions, although local intermediaries were often granted some residual rights over internal matters.Footnote 28 Overseas basing represents another type of authority relationship in which a host country leases sovereignty over some portion of its territory to a foreign power.Footnote 29 Relations of sovereign inequality need not be exclusively confined to states. The international community has routinely assumed sovereignty over countries following cases of humanitarian intervention, as in Kosovo and East Timor.Footnote 30
Others prefer a broad conception of hierarchy, one that focuses not just on inequalities in political authority, but also inequalities in power, wealth, status, or prestige. Ian Clark describes the existence of a ‘hierarchy of states’, which can range from states that possess considerable material capabilities to those that lack them.Footnote 31 Ann Towns likewise emphasises the role of ‘international social hierarchies’, which can shape how states with high or low status understand and legitimise their policies.Footnote 32 For these authors, hierarchy is not a bilateral relationship that actors enter into willingly, but a social position that constitute who the actors are and what roles and responsibilities they possess. A number of studies highlight the importance of these diffuse social hierarchies in world politics. Gerry Simpson highlights how ‘great powers’ are frequently granted elevated status and special rights by the international community, while ‘outlaw’ and ‘rogue states’ are marginalised due to their perceived illegitimacy.Footnote 33 Ayşe Zarakol likewise details how ‘outsider states’ that have experienced military defeat embrace ‘strategies which satisfy social-status cravings’, even when these strategies provide few material benefits.Footnote 34 Relations of inequality permeate the international system, shaping which states are empowered to act and what choices they make, including decisions concerning war and peace.Footnote 35
A final set of authors embrace an even more wide-ranging conception of hierarchy as distinct system for organising world politics. Adam Watson contends that international systems can vary between those that are anarchic and those that are hierarchic, which he conceptualises as universal empire.Footnote 36 David Kang likewise argues that a hierarchic system is one where ‘a dominant power … does not cause others to balance against it’.Footnote 37 Hierarchy is neither a relationship between actors nor a type of stratification within a system. If anarchy is a system that lacks an authority that can make and enforce laws, then hierarchy is its opposite: a system in which such a singular authority exists. Scholars have pointed to a number of historical systems that approach this ideal. Victoria Tin-bor Hui describes how the Qin rules embraced a ‘logic of domination’ and ‘swept across the ancient Chinese system’.Footnote 38 Daniel Deudney likewise highlights how Rome took advantage of its inclusive domestic regime and divided opponents to ‘[absorb] the entire array of polities and systems in the Mediterranean basin’.Footnote 39 Surveying a range of historical cases, Wohlforth, Kaufman, and Little conclude, ‘it is very common for … international relations in a system to be [more] hierarchical than anarchical’.Footnote 40
For all their differences, these competing conceptions of hierarchy share some common assumptions. They all accept that hierarchy implies the existence of some degree of differentiation and inequality among actors. As Janice Bially Mattern and Ayşe Zarakol observe, ‘there is significant scholarly convergence on the idea that hierarchies are intersubjectively constituted systems structured by vertical stratification’.Footnote 41 At the same time, these conceptions are also wildly incommensurate. Hierarchy is either a bilateral relationship of sovereign inequality between a dominant and a subordinate, a kind of social stratification in which actors are differentiated by role, or a particular system in which a single power predominates. Where we should locate hierarchy, what factors contribute to hierarchy, and which actors constitute hierarchy vary considerably across these approaches (Table 1). If the literature on hierarchy is going to develop into a unified research programme, we must make some difficult decisions about which of these conceptions should be preferred.
Competing conceptions of hierarchy.
The case for the broadest, systemic conception is reasonable. If anarchy is a type of structural ordering principle, as Waltz emphasised, then it makes sense to locate hierarchy at the same level of analysis.Footnote 42 The main problem with this approach, however, is that no system in history, no matter how centralised, has approached that of a pure hierarchy. For all its martial prowess, Rome still had to fend off external challenges, whether from rival city states such as Carthage or Germanic tribes along its northern frontiers.Footnote 43 Fissiparous tendencies and internal rivalries among Roman generals also meant ‘the later empire came to resemble an anarchy as much as a hierarchy’.Footnote 44 While concentrations of power are common, there has simply never been a world state.Footnote 45 Systems dominated by a single actor are interesting not because they are no longer anarchic, but because other forms of hierarchy can flourish within them. In such systems, dominant powers often make claims to high status roles and extend their authority over others.Footnote 46 Thus the broad systemic approach to hierarchy, while sound in theory, devolves to one of the rival conceptions of hierarchy in practice.
The remaining conceptions of hierarchy in world politics differ primarily in scope: should hierarchy refer narrowly on bilateral authority relations between dominant and subordinate actors or broadly on social stratification among groups of actors? An advantage of the latter approach is that it conforms to customary usage. When sociologists describe ‘social hierarchies’, for example, they refer to the stratification of societies based on particular attributes, whether race, class, or gender.Footnote 47 International Relations scholars have described how broad markers of the ‘standard of civilization’, including race and gender, produce similar inequalities in international society.Footnote 48 For such approaches, hierarchies are everywhere, and everything stems from hierarchy.Footnote 49
Yet there are downsides to using the term ‘hierarchy’ to refer to all forms of stratification in world politics. To begin with, it distracts us from what makes anarchy distinctive in the first place, which is not an absence of stratification in general, but an absence of political authority in particular. Waltz never claimed that states in the international system would be completely equal. He accepted that the international system would be stratified based on material capabilities and even role identities like ‘major power’ (although he assumed the latter could probably be reduced to the former).Footnote 50 Yet because sovereignty is indivisible and states prioritise self-help, relations of authority are dismissed either as anomalies or as byproducts of the balance of power. ‘Whatever elements of authority emerge internationally’, Waltz argued, ‘are barely once removed from the capability that provides the foundation for the appearance of those elements’.Footnote 51 The conception of hierarchy as sovereign inequality, therefore, cuts to the heart of traditional theories, highlighting the very kind of stratification that Waltz and others assume will be rare.
Using the term hierarchy to refer broadly to all forms of stratification in world politics also assumes that different types of stratification tend to reinforce one another. Those with high status are able to acquire greater wealth and power, and vice versa, to such a degree that it makes sense to refer to a common hierarchy of states. There is some evidence to support this contention. In his careful study of early nineteenth-century efforts to abolish the slave trade, for example, Edward Keene shows that Britain’s decision to deprive other polities of sovereignty was shaped more by the their status as either ‘civilized’ or ‘barbarous’ than their relative power or past behaviour.Footnote 52 Yet we should not assume that different types of stratification always go together in world politics. States with high status may be compelled to assume more responsibility for maintaining order, which can deprive them of material capabilities and economic resources over time.Footnote 53 Conversely, the absence of a single authority responsible for managing the global market can make it difficult to assign responsibility for growing economic inequality within and between states.Footnote 54 Hierarchy can reduce stratification in some cases, while anarchy can reinforce it in others.
Given the way in which International Relations theorists have used the term anarchy, therefore, there is a persuasive case for the narrow conception of hierarchy as type of authority relationship. Just because this conception focuses solely on sovereign inequality, however, does not mean that every case of hierarchy looks the same. As Lake emphasises, hierarchy can vary considerably in the extent of authority claimed by the dominant state.Footnote 55 Some cases of hierarchy permit a subordinate to retain substantial residual rights of control. Examples include protectorates, overseas basing agreements, spheres of influence, and suzerainties. In other cases, a dominant state claims a significant degree of sovereignty over subordinates. Examples include military occupations, mandates, trusteeships, as well as informal and formal empires. Nor does a narrow focus on sovereign inequality mean that it will always be easy to identify when and where hierarchy exists.Footnote 56 Dominant states may limit a subordinate’s sovereignty through informal and indirect means. Subordinate actors may claim that their policies reflect their own preferences, not the demands of some external power. Even using this narrow conception, therefore, care must be taken to distinguish between cases where actors align their policies due to common interests and cases where a dominant state is exercising genuine authority over a subordinate.
Orchestrated control: the unique features of hierarchy amidst anarchy
If we accept the narrow conception of hierarchy as an authority relationship between dominant and subordinate states, this raises the obvious question of how authority is created in world politics. Authority is typically defined as rightful rule: a dominant party has the right to command, while a subordinate party is obligated to comply.Footnote 57 Yet the ties that bind states together in world politics tend to be less robust than those that bind individuals together in domestic politics. How then is this strong social obligation created in the seemingly asocial realm of international politics? One approach is to reduce right to might. Dominant states have the authority to command by virtue of their material capabilities. Kenneth Waltz provides a version of this argument when he claims that authority in international politics ‘quickly reduces to a particular expression of capability’.Footnote 58 Strong states use their superior capabilities to seize the sovereignty of subordinate polities through coercive means.Footnote 59 When force is efficacious and conquest pays, relations of sovereign inequality should be commonplace.Footnote 60
An alternative is to argue that authority can be granted through consent. A subordinate can choose to surrender their rights in exchange for protection. This is the essence of David Lake’s ‘relational contracting’ approach. Hierarchy is most likely to emerge when scale economies in the production of security exist and when opportunism is likely to undermine cooperation under anarchy.Footnote 61 In these situations, dominant states reach agreements with subordinates where the latter agree to transfer some degree of their sovereignty in exchange for the benefits of protection and other side payments.Footnote 62 Relations of hierarchy act as a type of ‘contract’ where a dominant state provides protection and ‘social order’ in exchange for a subordinate actor’s ‘consent’, although Lake is careful to acknowledge examples where side payments and other forms of compensation may be insufficient and coercion may be required.Footnote 63
Viewing authority relations in world politics through the lens of coercion or consent has an intuitive appeal. After all, states under anarchy use these very same instruments to advance their interests on a routine basis. Yet coercion and consent are not mutually exclusive mechanisms of control, nor are they the only way in which authority can be imposed.Footnote 64 In many cases, a dominant actor possesses authority by virtue of their status or prestige, not simply a capacity to intimidate or bribe. Not surprisingly, powerful states frequently strive to establish legitimacy over subordinates so that they do not have to rely on material inducements alone.Footnote 65 Along the same lines, dominant states often derive their authority not from overt exercises of power, but by shaping the broader context in which choices are made.Footnote 66 By setting the agenda, powerful states can constrain the options available to subordinate polities in ways that ensure their continued compliance.
More importantly, by reducing the creation of authority to the politics anarchy, both of these approaches downplay what makes hierarchy unique in world politics. The assumption is that hierarchy in world politics resembles domestic politics, in that a dominant actor’s authority is absolute and uncontested.Footnote 67 Because dominant actors possess strong claims to authority, patterns of hierarchy will be reflections of their particular capabilities and preferences. Yet what makes hierarchy amidst anarchy so distinctive is precisely that it does not resemble domestic politics (Table 2). Dominant states rarely make as complete and unchallenged claims to sovereign authority as do central governments in domestic politics.Footnote 68 Hierarchy amidst anarchy is a curious relationship in which dominant and subordinate actors share sovereignty. Claims to authority are often ambiguous and overlapping, rather than clear and distinct.Footnote 69 Reducing hierarchy to the capabilities or preferences of dominant states ignores the prominent role of subordinates in producing and reinforcing these particular patterns of authority.
Varieties of authority relations.
To illustrate these points, consider the varied ways a dominant State A might shape the behaviour of potential subordinate State B. One possibility would be to rely on the politics of anarchy, to forgo authority claims over State B and to interact with it as a sovereign member of international society. In this situation, any ordered relationship that might emerge will primarily be the product of mutual decisions to cooperate. State A can try to convince State B to change its policies, either by appealing to common interests or a shared identity, yet it cannot command State B to comply. The gains from an potential agreement need not be equal, of course, and the brooding shadow of violence can influence bargaining outcomes. But to the extent that State A and State B harmonise their policies, they should do so by choice, as partners or even allies.
A second possibility is that State A might choose to conquer State B, absorbing it into its own state. In this situation, State A is relying entirely on the politics of domestic hierarchy. State B no longer exists as a sovereign entity and its territory and people will be governed directly by State A. In this situation, order is maintained as it is in any other state. The people of State B become subjects. They are governed by administrators who serve as agents of State A. Authority is delegated from State A to these agents, who act based on written laws and accepted procedures. State B harmonises its policies to those of State A because it has no choice. It is no longer independent, and its political elites no longer have any claim to power, except perhaps as administrators who are directly accountable to a higher authority.
A final possibility is that State A claims authority over State B, but that the latter is permitted to retain some amount of residual sovereignty. State B does not entirely cease to exist; its political elites retain some modicum of their former status and power. Yet their capacity to act is constrained and shaped by the authority claimed by State A. It is this intermediate situation that makes hierarchy under anarchy unique. Rather than treat sovereignty as indivisible, dominant and subordinate actors make simultaneous claims to political authority, the boundaries of which may be ambiguous or even overlap. State A does not govern State B directly, but rather seeks a degree of control over its subordinate. Yet State B is not powerless in this situation. The acceptance and participation of its elites in this system of shared sovereignty is often a prerequisite for hierarchy to emerge and endure.
There are limits to this typology, of course. In practice, hierarchy under anarchy might more closely resemble either the politics of domestic hierarchy or international anarchy. In imperial possessions ruled directly by metropolitan administrators, for example, governance might resemble that found in the domestic politics of the colonial powers themselves. Conversely, in situations where the degree of sovereignty being conceded is minimal and the mechanisms for reclaiming rights is credible, as is the case in some overseas basing agreements, the factors driving outcomes may be similar those shaping cooperation under anarchy.Footnote 70 Yet recognising that hierarchy amidst anarchy is governed by its own distinct political dynamics has a number of important implications:
First, in hierarchy amidst anarchy, the subordinate actor plays an essential role. On the one hand, actors that share authority with a dominant state risk diluting their own legitimacy. Politicians who grant basing rights to the United States, for example, are often accused of being sellouts to a foreign power. On the other hand, subordinate actors can gain prestige for being associated with a dominant power. Warrant chiefs in colonial Nigeria often laid claim to extensive powers by virtue of their connection to imperial authorities. Politicians in subordinate actors, in other words, reside in liminal positions. They act both as collaborators with external powers, and as intermediaries between these patrons and their own constituents. For hierarchy amidst anarchy to endure, dominant states must devise ways to share authority that neither critically undermines nor excessively empowers potential subordinates.
Second, in hierarchy amidst anarchy, the relationship between dominant and subordinate actors is distinct. Because authority is shared, ambiguous, and often overlapping, dominant states ensure their subordinates’ compliance less through the rule of law or the delegation of formal responsibilities. Control more closely resembles what Abbott, Genschel, Sindal, and Zangl call ‘orchestration’, or the enlistment of an ‘intermediary through material and ideational support’.Footnote 71 Dominant states can provide subordinates with resources to improve their capacity to perform desired tasks, or can provide guidance and approval to enhance their effectiveness when doing so.Footnote 72 Yet in hierarchy amidst anarchy, orchestration can also work in reverse. Subordinate actors can provide specific local assets that dominant states require to maintain their authority, but are unable to generate on their own. They can also endorse a dominant state’s claim to authority, enhancing its legitimacy despite its outsider status. Hierarchy amidst anarchy, in short, is a kind of mutual orchestration. Dominant states leverage resource and expertise asymmetries to control their clients, while subordinate actors exploit local knowledge and legitimacy deficits to constrain their masters.
Third, given these dynamics, the factors that shape the operation of hierarchy amidst anarchy are much different than those influencing sovereign states. When it comes to forging cooperation under anarchy, the power and interests of the states in question are of prime importance. Yet given the central role of authority in hierarchy, power and interests are just part of the picture. Subordinate actors must be able to surrender authority, and dominant states must be able to appropriate it, without sacrificing their broader political legitimacy. Given the requirements of mutual orchestration, the position of the actors vis-à-vis one another is also key. Dominant states must be positioned to provide the resources and expertise required to sustain hierarchy over subordinate actors. Conversely, subordinate actors must be positioned to grant local assets and legitimacy to a dominant state.
To illustrate these points, consider the East India Company’s use of the subsidiary alliance system during the late eighteenth and early nineteenth centuries. Under this arrangement, Indian princes surrendered control of their foreign affairs and paid an annual subsidy in exchange for the stationing of Company troops within their territories.Footnote 73 At first glance, this system resembles Lake’s conception of hierarchy as a type of contract intended to promote security. Yet on closer examination, the subsidiary alliance system depended on much more than just shared interests, and Indian princes and Company officials often went to great lengths to bolster one another’s authority. The Company enhanced the capabilities and prestige of Indian princes by recognising lines of succession, providing loans, and extending extraterritorial protections to members of royal families. Its troops would not just defend state borders, but also help Indian states collect revenue and suppress internal revolts. Conversely, Indian princes sought to legitimate the presence of Company officials by appointing them to positions at court and granting them honorific titles. They also provided crucial local assets to support the Company’s troops including cheap provisions, timely intelligence, and specialised military forces such as irregular cavalry. The subsidiary alliance system remained fundamentally unequal, and Indian princes bristled at the constraints on their sovereignty. Yet it highlights how dominant and subordinate actors can engage in mutual orchestration to render authority sharing under hierarchy both stable and legitimate.
Positional power: Networks and authority in world politics
Hierarchy amidst anarchy is a distinct type of political relationship, different than the one found among states under anarchy or within the domestic politics of states. The sharing of authority is fraught with difficulties, however. Dominant states may be unable to identify and recruit reliable intermediaries with whom to share sovereignty. Even when potential subordinates are present, these actors may be unable to legitimate external domination in the eyes of local populations. When dominant states lack access to local assets or legitimacy, the cost of establishing and maintaining control will tend to be high, perhaps even prohibitive. Understanding the positions of potentially dominant and subordinate actors relative to each other is critical, therefore, for assessing the likelihood that hierarchy emerges and endures.
Traditional theories of hierarchy have not paid significant attention to how actors maintain authority in practice. David Lake’s relational contacting approach, for example, looks primarily at the motive actors have to establish hierarchy. While Lake acknowledges that dominant states will pay different ‘governance costs’ to maintain hierarchy, he does not provide a systematic theory of where governance costs come from or what states can do to influence them.Footnote 74 Governance costs are essentially structural and static: a product of the material balance of power and the technologies of repression. Besides, because hierarchy reflects the mutual interests of the contracting parties, the costs of maintaining it should be relatively low to begin with. Alexander Cooley and Hendrik Spruyt’s theory of ‘incomplete contracts’ adopts a similar approach.Footnote 75 They assume that the primary challenge states face when sharing sovereignty is how to write the particular contract – which rights should be retained, how can the retention of these rights be made credible, and so forth. The maintenance of hierarchy is reduced to a challenge of institutional design.
As described in the previous section, however, the main challenge in maintaining hierarchy is not harmonising interests or crafting contracts, but the practice of sharing sovereignty itself. Hierarchy amidst anarchy requires actors to occupy positions that facilitate the legitimate appropriation of authority and supports the dynamics of mutual orchestration. This raises the question of how best to theorise the positions of dominant states and subordinate actors. A number of scholars have explored the power of positionality through the lens of sociological network theory.Footnote 76 The essential building block of this approach is the social tie, which Charles Tilly defines as ‘a continuing series of transactions to which participants attached shared understandings’.Footnote 77 These social ties can vary in content. They can include economic transactions, such as the exchange of goods or capital; political transactions, such as the exchange of diplomats; and cultural transactions, such as the exchange of religious missions or media. Social ties can also differ in character. They can be strong ties in which exchanges are large in magnitude, frequent in appearance, and sustained over time; or weak ties in which exchanges are rare, unimportant, and fleeting.Footnote 78
Even more important than the content or character of individual ties is how they combine together to form social networks.Footnote 79 These networks are important because an actor’s position can have a profound impact over its potential power. Actors that occupy central positions within a broader network may have access to more abundant material and ideational resources than those that are marginal or weakly connected. Actors that possess ties with a wide range of different and disparate groups may possess greater influence than those that interact only with small, like-minded cliques. A number of studies have built on these insights. Stacie Goddard finds that actors that occupy brokerage positions are more capable of acting as transformative political entrepreneurs.Footnote 80 Emilie Hafner-Burton and Alexander Montgomery describe how a state’s position with networks of preferential trade agreements shapes their ability to benefit from globalisation and to impose economic sanctions.Footnote 81 Daniel Nexon and Thomas Wright discuss how imperial powers find it easier to prevent resistance in possessions that are disconnected from one another, as in hub-and-spokes empires.Footnote 82
It is beyond the scope of this article to propose a complete theory of the relationship between networks of social ties and political authority in world politics. In previous work, I traced how social ties influenced the ability of imperial powers to organise and sustain campaigns of peripheral conquest.Footnote 83 I build on this approach here to highlight some of the particular network dynamics that might facilitate the creation of hierarchy (Figure 1). This approach differs from those who see networks as a distinct ‘way of governing’.Footnote 84 Rather than juxtapose networks and hierarchies as competing organisational forms, it contends that networks can provide the material and ideational resources that allow for the construction of relations of sovereign inequality. I focus in particular on three potential mechanisms: the first two – access and brokerage – highlight how static configurations of ties can provide dominant and subordinate actors with the capacity to create hierarchy; the final one – yoking – explores how dynamic shifts within and across networks can create new actors that get folded into relations of hierarchy.
Networks and hierarchy.
Appropriating authority: Dominant states, centrality, and access
A relational network approach to hierarchy starts with the premise that actors often derive authority by virtue of their position. Dominant states tend to be situated in positions that enable them to exercise control, while subordinate states reside in positions that constrain their ability to resist. States possess their own material and ideational resources, to be sure. But the ability of states to generate, mobilise, and employ these resources is significantly shaped by their position within broader social networks.
All things being equal, dominant states will be more capable of claiming authority when they are located in a central position within a relevant network, meaning they possess strong ties with a large number of other actors.Footnote 85 ‘Network centrality’ provides dominant states with a number of advantages. They will have access to a greater volume of material or ideational resources due to the sheer number of exchanges in which they participate. They will also be able to draw on a wider variety of resources, given the diversity of sources that they can access. Equally important, they will be less vulnerable to periodic disruptions in exchange patterns and more able to threaten to cut off exchanges to gain leverage over other actors.Footnote 86 States located in central positions, in short, possess what Hafner-Burton, Khaler, and Montgomery call ‘social power as access’.Footnote 87 They can leverage this access to accumulate resources, acquire information, and bolster their prestige in ways that actors that have weak ties with fewer actors cannot.
Dominant states can leverage this access to construct hierarchy through a variety of mechanisms. They can exploit asymmetries in information to identify valuable and reliable intermediaries. They can take advantage of abundant and diverse resources to recruit these actors as collaborators. They can exploit asymmetries in resources or expertise to orchestrate specific outcomes with their chosen subordinates. Yet centrality can help dominant states maintain hierarchy in more subtle and indirect ways as well. They can exploit favourable network positions to set the agenda, promoting rules that tend to be ambiguous and flexible when applied to them, but specific and obligatory when applied to others. ‘Structural privilege brings influence over policy scripts’, Jason Beckfield observers, ‘powerful states and societies seek to maintain their privileged positions’.Footnote 88 They can also use their positions to socialise subordinate societies, championing specific norms or ideas that help bolster their authority.Footnote 89 ‘The learning and socialization processes’, Hafner-Burton, Khaler, and Montgomery conclude, ‘may be manipulated and co-opted by powerful network actors’.Footnote 90 Rather than undermining the legitimacy of dominant and subordinate actors, the sharing of authority can come to be seen as a normal and natural arrangement.
The aforementioned example of the subsidiary alliance system highlights how dominant states can exploit central positions within social networks to promote hierarchy.Footnote 91 To raise the local military manpower to station in subsidiary states, the East India Company exploited connections to landed elites and military nobles in Awadh, Bihar, and Benares, as well as ties with military recruiters, known as jemadars.Footnote 92 To generate the capital for loans to local princes, the Company took advantage of ties to coastal traders and urban moneylenders, who supported and financed expansion.Footnote 93 To manage new political relationships with interior states, the Company leveraged its connections with former Mughal bureaucrats and administrators, whether the banias of Bengal or dubashes of Madras, to serve as interpreters, advisors, secretaries, and intelligence agents.Footnote 94 To help legitimate its authority, the Company sponsored the award of traditional robes of investiture to local bureaucrats and district collectors, creating the appearance of continuity with Mughal practices.Footnote 95 In these ways, the Company took advantage of its central position within subcontinent-wide networks to generate the local resources and expertise required to orchestrate outcomes with Indian princes.
Enabling authority: Subordinate actors, intermediaries, and brokerage
The position of dominant states is only part of the picture, however. Because hierarchy amidst anarchy requires the sharing of authority, the capacities of subordinate actors are important as well. Politicians in subordinate polities can help orchestrate outcomes of their own, supporting a dominant state with local assets and bolstering the legitimacy of external domination in the eyes of local populations. The central claim of this section is that the effectiveness of intermediaries is determined less by their individual attributes, but more by their position relative to both their distant patrons and their local constituents.
All things being equal, politicians in subordinate polities will be more effective intermediaries when they occupy bridging positions between actors within a network. Being situated as ‘brokers’ can provide politicians in subordinate states with a number of assets.Footnote 96 They can gain leverage by facilitating exchanges of resources that are scarce in one part of the network but plentiful in another. They can collate information to gain an advantage over rivals who do not appreciate the big picture, or share information selectively in order to manipulate different audiences.Footnote 97 Actors situated as ‘brokers’, Daniel Nexon observes, can ‘opportunity hoard’ by ‘monopolizing cross-cutting ties’.Footnote 98 Yet brokers do not just possess material advantages. Perhaps most importantly, they can also appropriate and combine symbols from a wide variety of sources in creative ways in order to make ‘multivocal’ rhetorical claims with broad appeal.Footnote 99 As Goddard concludes, politicians that bridge disparate networks of actors ‘wrench power from their positions’.Footnote 100 They can knit together groups in ways that bolster their influence when required, or keep groups separated to maximise their leverage when necessary.
Politicians situated as brokers in subordinate societies will make particularly effective collaborators, for a number of reasons. They can mobilise assets from their liminal position to help orchestrate desired outcomes with a dominant state. They can funnel intelligence to a dominant state, allowing it to make better use of its own resources. Perhaps most importantly, they can draw upon diverse sources of authority to help find inventive narratives by which to legitimate the claims of dominant powers. Brokers are also more attractive than the alternatives: collaborators that lack ties with their own societies will be unable to provide useful assets to an external power, while those that are too strongly tied to their own constituents may be unwilling or unable to legitimate foreign domination. Yet brokers are not mere patsies, they can also limit the ambitions of dominant states. Because they can manage flows of resources and information, they can constrain a dominant state’s ability to completely usurp a subordinate’s sovereignty. Brokers can act both as important enablers of hierarchy, but also as checks on its scale and scope.
Returning to the case of the subsidiary alliance system, it is perhaps unsurprising that the East India Company relied heavily on indigenous intermediaries within Indian states. Imperial historians have long emphasised the importance of local collaborators, which Ronald Robinson famously described as the ‘non-European foundations of European imperialism’.Footnote 101 Yet the concept of brokerage helps shine additional light on who became valued collaborators and what assistance they provided. The most valued collaborators were rarely the Indian princes, who bristled at the restrictions placed on their sovereignty, nor Europeanised elites, who lacked local credibility. Rather, valued contributors tended to be ambitious mid-range bureaucrats, who had ties both to the Company and to some local base of political power. In the case of Awadh, for example, the Company depended on various members of the prince’s administration, including his chief minister and finance minister.Footnote 102 The Company also cultivated relations with rural district administrators or amils, who given their wealth and sizable military forces, acted like semi-autonomous rulers.Footnote 103 These various intermediaries did not just passively accept the Company’s claims to sovereignty; they helped to actively orchestrate it at the local level. The Chief Minister of Hyderabad brokered an agreement with local grain merchants or banjaras to ‘follow the [Company’s] army’ and keep it well fed.Footnote 104 Almas Ali Kahn, the most prominent rural official in Awadh, passed along intelligence, provided provisions for Company regiments, and even contributed his household troops to help the Company extract revenue from refractory landholders.Footnote 105 Through these various acts, Indian brokers eroded princely sovereignty and helped legitimate its appropriation by the Company.
Constructing authority: Actors, boundaries, and yoking
A focus on networks can help illuminate when dominant states are well positioned to appropriate authority, as well as which politicians in subordinate polities are well situated to enable this appropriation. To this point, however, we have assumed the actors that enter into relations of hierarchy are independent polities with clearly defined boundaries that choose to share sovereignty with one another. Whether we describe our protagonists as ‘states’, ‘polities’, or more generally ‘actors’, they begin in a condition of anarchy in which they possess complete sovereignty, until the moment they transition into relations of hierarchy. While a reasonable starting point, there are limits to this approach. Actors often transition not from anarchy to hierarchy, but from one kind of subordinate relationship to another. Even more important, the very act of forming and reforming hierarchies can shift boundaries between actors and can influence who constitutes an actor in the first place.
Network approaches have long emphasised that entities do not have fixed boundaries, and that processes of interaction and exchange can refashion who actors are and when they can exercise agency.Footnote 106 These approaches adopt an ‘anticategorical imperative’, which rejects the idea that units with fixed sets of attributes – whether utility-maximising individuals, social classes, or territorial nation-states – are what constitutes social life.Footnote 107 Instead, shifting patterns of interaction – and the meanings we attach to them – can create actors, a process Andrew Abbott calls ‘yoking’.Footnote 108 If network ties increase in strength and density, this can reinforce existing boundaries, harden current identities, and bolster the political authority of those who claim to speak for the group. Conversely, if network ties diversify in content and character, this can erode current boundaries, diffuse different ideas, and create opportunities for entrepreneurs to fashion novel identities and alternative sources of political authority. ‘Rhetoric can create new social categories from what were once different sets of networks’, Goddard observes, and thus ‘actors themselves can be created and transformed during the course of strategic interaction’.Footnote 109
Shifting patterns of networks ties can have many consequences, of course, not just the creation of new actors. John Padgett highlights how the connection of ties across different domains, a process he calls ‘network folding’, can lead to the emergence of innovative social practices and novel organisational forms.Footnote 110 Yet because yoking can transform who claims political authority and where political authority is located, it can play an important role in the emergence of hierarchy. Moments when traditional repositories of political authority are eroding, in particular, may be ripe for hierarchy, as emerging actors seek external support for their claims to greater authority. Competition between new actors for recognition or prestige may also make them more willing to accept restrictions on their sovereignty. Yet dominant states do not just exploit new actors; they can help create them as well. Dominant states can adopt actions that empower and legitimate new actors, essentially manufacturing intermediaries that are directly invested in their rule. Dominant states can also claim that their existing hierarchic relations form a broader community with a distinct political identity, thus creating a powerful new justification for their continued dominance.
This creation and manipulation of new actors played a prominent role in our case of the subsidiary alliance system. Traditional accounts of the rise of British power in India emphasise the political vacuum created by the decline of the Mughal Empire. Yet recent studies argue that Mughal decline did not result in a prolonged period of political chaos and economic depression, but the proliferation of a diverse variety of new political actors.Footnote 111 Mughal successor states took advantage of subcontinent wide networks of trade and finance to engage in the centralisation of state administrations, a process Burton Stein calls ‘military fiscalism’.Footnote 112 This drive to increase revenues and raise armies empowered a range of ancillary actors – including revenue farmers, moneylenders, merchants, and mercenaries.Footnote 113 Yet paradoxically, efforts to strengthen successor states increased divisions within and between them, providing opportunities the East India Company to embed itself into these diverse networks and to extend its political sovereignty.Footnote 114 In some cases, the Company exploited disputes between rival factions within existing successor states. Between 1798 and 1839, it intervened in no less than 11 succession crises, often backing rival claimants in exchange for a subsidiary alliance treaty.Footnote 115 In other cases, the Company encouraged local authorities to create new successor states. When the Maratha ruler of Sindhia refused to accept a subsidiary alliance in 1803, the Company engaged in a concerted campaign to bolster and appropriate the authority of petty rulers and military chieftains, including the Jats, Bundelas, and Rohillas.Footnote 116 In the subsequent war, as Randolf Cooper argues, the Company did not destroy Sindhia’s armies so much as it ‘recycled and absorbed’ them.Footnote 117 The redrawing of boundaries across north India allowed the Company to construct new actors willing to surrender sovereignty to a foreign power.
Conclusion: Hierarchy loosed upon the world?
The fundamental premise of this article is that International Relations scholars have misunderstood the phenomenon of hierarchy amidst anarchy. Some have been tempted to dismiss its existence. ‘Sovereignty over the same territory cannot reside simultaneously in two different authorities’, Hans Morgenthau once argued, ‘sovereignty is indivisible’.Footnote 118 Yet this approach ignores the myriad ways in which authority is shared, even among actors in the otherwise anarchic structure of world politics. Others have sought to reduce hierarchy to anarchy. Hierarchy in world politics, David Lake maintains, ‘rests on the benefits to subordinates of the political order provided by the dominant state and credible reassurances that they will not be abused’.Footnote 119 Yet equating hierarchy to credible bargains designed to advance mutual interests misses the subtle and distinct ways in which dominant states appropriate authority and claim legitimacy in world politics.
I have argued instead that hierarchy in world politics is a distinct type of political relationship governed by its own unique dynamics. Contrary to those who conceive of hierarchy as any form of social stratification, I have argued that it is best thought of as a relationship of sovereign inequality. Yet relations of domination and subordination in world politics tend to be much more fragile and incomplete than those within domestic hierarchies. Rulers do not simply delegate instructions to the ruled based on legal frameworks or formal procedures. Dominant states are rarely viewed as legitimate in the eyes of local populations. Local politicians that work with external powers are easily dismissed as collaborators and sellouts. The mere act of sharing political authority in world politics threatens to render it both illegitimate and ineffective.
The maintenance of hierarchy amidst anarchy, therefore, depends on the capacity of dominant states and subordinate intermediaries to engage in processes of mutual orchestration. They must be able to mobilise their material and ideational resources to bolster one another’s claims to political authority in their respective spheres. To do this, they draw on their positions within existing networks of political, economic, and social exchange. Dominant states are able to maintain control because of their ability to access capable intermediaries who can facilitate the appropriation of local authority. These intermediaries are able to render external domination legitimate in the eyes of local populations because of their privileged position as brokers within subordinate societies. In some cases, these processes of mutual orchestration are facilitated by the emergence of new actors, competing sites of political authority that can rationalise hierarchy in creative and unexpected ways. In short, networks are not distinct from hierarchies in world politics, but rather can facilitate and enable them.
While I have used relational concepts to explore authority appropriation in world politics, it is worth emphasising that network approaches can be useful in other contexts as well. For those who favour a conception of hierarchy as social stratification, network analysis provides tools that can differentiate actors based on status or rank, as well as identify shared communities or cliques.Footnote 120 Rather than treat high- and low-status groups as static binary categories, network approaches can capture more diverse patterns of social stratification while highlighting which actors are situated in contested or liminal positions. Similarly, those who prefer a systemic conception of hierarchy can employ network tools to measure the degree of density, centralisation, and polarisation across the system as a whole.Footnote 121 Instead of assuming that systems dominated by a single actor all look the same, network approaches can highlight how systems can vary in the scale, scope, and domains in which dominance is being exercised. In this way, a relational network approach can clarify – and perhaps even bridge – differences among the contending approaches to hierarchy.Footnote 122 The same sets of overlapping network ties can reinforce inequalities in authority within dyadic relations, in status among competing groups, and in power and wealth across the system as a whole.
While scholars have paid greater attention to hierarchy in theory, it remains unclear how important hierarchy will continue to be in practice. International institutions such as the United Nations have elevated norms of juridical sovereignty, calling into question the legitimacy of prior forms of external domination such as colonial rule or international trusteeship.Footnote 123 The emergence of the United States as the leading unipolar power has also raised questions about whether external domination remains the most common means for creating order in world politics. Scholars such as John Ikenberry emphasise the distinct character of American leadership, which while hierarchic in some respects, also relies to a significant degree on the consent of other states.Footnote 124 Perhaps theories of hierarchy as a type of bargain are more suited to a moment when the international order appears to be less coercive and more ‘constitutional’ in character.Footnote 125
Whether the United States relies primarily on consent is a matter of debate, of course, and Washington has struggled at times to legitimate hierarchic relations with its varied subordinates. In both Iraq and Afghanistan, observers questioned whether national governments possessed genuine sovereignty. In Japan and South Korea, American military bases became focal points for nationalist protest. Emerging powers such as China have questioned the allocation of voting rights in the World Bank, while India has called for reform of the insular club of permanent members of the United Nations Security Council. Iran and others have called into question the hypocritical behaviour of the United States and other recognised nuclear weapons states when it comes to disarmament pledges made under the Nuclear Non-Proliferation Treaty. Just because the United States consults with its allies and partners on certain occasions does not mean the fundamental dilemmas created by its appropriation of other’s sovereignty are easily resolved.
The relational network approach developed in this article suggests that it will the United States position within overlapping networks of political, economic, and social exchange, more than its military might or commitment to play by the rules, that will have the greatest impact on the future of hierarchy. The importance of America’s capital markets and the dollar in the global economy provides Washington with leverage to pressure countries to open their markets to foreign firms and capital.Footnote 126 The prominent position of the United States within security and economic institutions allows it to craft binding rules for other states, while preserving its own autonomy.Footnote 127 The network approach also suggests that overt exercises of power may be less essential to the preservation of American dominance than more subtle forms of orchestration. The United States can put informal pressure on international institutions such as the World Bank, to take one example, to ensure that its preferences are reflected in official reports and staffing decisions.Footnote 128 The United States can leverage the support of critical local intermediaries to promote bilateral economic deals or justify continued access to military bases.Footnote 129
There are a number of changes, however, that might limit the United States capacity to leverage these networks in the future. Globalisation and the rise of new economic hubs have reduced the centrality of the United States in global networks of exchange. China’s creation of the Shanghai Cooperation Organisation and the Asian Infrastructure and Investment Bank, suggest that authority in world politics may not be embedded exclusively in American-led institutions in the coming years. The growing power of nationalist and anti-American movements has likewise constrained the ability of potential brokers to legitimate liberal policies in the face of domestic opposition. Recent events in the Philippines and Turkey highlight how the diminished influence of cosmopolitan intermediaries in subordinate states can complicate the United States’ efforts to preserve hierarchy, even if the underlying interests of both parties remain the same. Meanwhile, the yoking together of other networks has led to the creation of a whole range of new actors that can challenge the authority of the United States, whether global anti-basing movement or transnational terrorist groups. Even if the United States remains committed to maintaining the current liberal international order (which itself can no longer be taken for granted), there may be deeper shifts in networks of global exchange that will disrupt patterns of hierarchy in world politics in the coming decades.
Acknowledgements
The author would like to thank Ahsan Butt, Michael Byers, Alex Cooley, Jack Donnelly, Roxanne Euben, Stacie Goddard, Robert Jervis, Peter Katzenstein, Jonathan Kirshner, David Lake, Barak Mendelsohn, Craig Murphy, Iver Neumann, Daniel Nexon, Jack Snyder, Lora Anne Viola, Alexander Wendt, and participants in workshops at Cornell University, Wellesley College, and the Norwegian Institute of International Affairs for their helpful comments on earlier versions of the article. The funding for this article was provided by the Norwegian Research Council under the project ‘Evaluating Power Political Repertoires (EPOS)’, project no. 250419.
