Jump to ContentJump to Main Navigation
The New Power PoliticsNetworks and Transnational Security Governance$

Deborah Avant and Oliver Westerwinter

Print publication date: 2016

Print ISBN-13: 9780190604493

Published to Oxford Scholarship Online: August 2016

DOI: 10.1093/acprof:oso/9780190604493.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 21 January 2019



Networks and Transnational Security Governance

(p.1) 1 Introduction
The New Power Politics

Deborah Avant

Oliver Westerwinter

Oxford University Press

Abstract and Keywords

Why do we need a new approach to security governance? This chapter explains the puzzles that motivate this project and the proposed framework for solving them. It sketches two important ways security governance varies: by how effective it is and whose concerns it serves. It then outlines the network mechanisms likely to affect governance: the relative position of different actors, the distribution of ties, and the quality of ties. The volume’s logic rests on a relational understandings of power, which is introduced and analyzed. The final part of the chapter describes the volume’s chapters and then highlights the projects’ overarching theoretical contributions which include its focus on network and power dynamics, its highlighting of the importance of multiplex nodes, and its novel way of understanding the US role in security governance.

Keywords:   network, security governance, power, global security, network theory

How is it that the United States could work together with China and Iran (among others) to combat piracy at a moment when US tensions with China were on the rise and many US politicians were unwilling to even negotiate over nuclear issues with Iran? How do we account for instances when the United States acts in ways that threaten global stability rather than working to generate it the way theories of hegemony would expect? How can we explain the clearly consequential involvement of commercial and human rights groups in security arrangements that affect sovereign concerns? Why are many issues to do with weapons of mass destruction governed through informal transgovernmental arrangements rather than institutions that involve the highest levels of government? Traditional analyses of global security politics do not ask on these questions. Assuming the benefits of US hegemony, they focus on its stability, the rise of new powers, and the intransigence of various rogue states. Nor can traditional theories answer them. They cannot explain why there is “governance” (or not) of important security issues—from combatting piracy to curtailing nuclear proliferation to reducing the contributions of extractive industries to violence and conflict. They often miss collective efforts resulting from non-state action or taking less traditional forms. And they cannot explain why governance sometimes serves broad concerns (as liberal theories expect) and other times narrow ones (as realist theories expect). This volume uses network theory to develop a framework within which to answer these questions that puzzle traditional security analysts.

In the last fifteen years, there has been a surge of writing about new forms of governance constituted by various mixtures—often termed networks—of state and non-state actors (Keck and Sikkink 1998; Reinicke 1998; Reinicke and Deng 2000; Slaughter 2004; Kahler 2009). In these new modes of regulation, authority to govern does not reside exclusively with states but is wielded by a (p.2) variety of actors at different levels, including national bureaucrats, intergovernmental organizations (IGOs), nongovernmental organizations (NGOs), transnational corporations, and business associations (Abbot and Snidal 2009; Mattli and Woods 2009; Avant et al. 2010; Lake 2010). Building on the insights of these studies and network theory’s foundational logic, we propose a framework for the study of a power politics that goes beyond states alone— what we call the new power politics. We examine networking—or building relationships—as a dynamic force through which agents can shape their future fates and governance outcomes. We also look at the relations surrounding particular issues as “networks” and explore how the positions of actors in these networks, the distribution of ties, and the quality of ties can offer insights into (1) why governance arises around some security issues and not others and (2) the concerns governance serves.

Our approach engenders a fresh look at power and how it works in global politics. Unlike some global governance analyses, we see power as integral to governance (see also Barnett and Duval 2005). Rather than conceptualizing power as a commodity that actors own, as traditional analyses often do, we see power as emerging from, and depending on, relationships. This view of power is dynamic. While an actor’s relational position shapes her power, it is also something she can take action to shape strategically. Charli Carpenter’s analysis in this volume ties agenda setting and policy change related to the compensation of civilians in wartime to just this kind of action. The new power politics is thus about maintaining, developing, and shaping relationships as well as about working through these relationships to affect governance outcomes.

By and large, we agree with the claims of traditional security studies scholars that the governance of transnational issues is the product of power politics rather than its transformation into “powerless” cooperation. We maintain, though, that traditional conceptions of both power and politics are too narrow. The new power politics we analyze is the management of relationships that allow agents (whether states, state agencies, companies, NGOs, or individuals) to affect their fates and inclinations through collective mobilization (see also Goddard and Nexon 2016). Useful analyses of when and why power politics leads to more or less effective governance and governance that serves many or few in the contemporary world requires attention to relationships among these varied actors.

Transnational Security Governance Puzzles

What is transnational security governance, how does it vary, and why is this surprising to traditional international relations theory? We accept the (p.3) common definition of governance as the ability to steer or manage a collective (Commission on Global Governance 1995; Keohane and Nye 2000). Governance has been used to describe processes in governments, within or among corporations (Gourevitch and Shinn 2007), NGOs, and, increasingly, in the transnational or global spaces that we focus on in this volume (Rosenau 1992; Held et al. 1999; Duffield 2002). Central elements to transnational governance include: creating issues, setting and managing agendas, negotiating rules or standards, implementing policy, and monitoring, enforcing, and adjudicating outcomes (Abbott and Snidal 2009; Avant et al. 2010a). Leadership to create a collective and management to maintain it are important parts of transnational governance (Young 1991; Tallberg 2010), as are shaping expectations, generating collective action (including the provision of “public goods”1), and otherwise “ordering” a shared enterprise.

At the global level, arguments about governance have taken at least two different forms. Analyses of regimes, defined as “principles, norms, rules and decision making procedures around which actor expectations converge in a given issue area” (Krasner 1983, 2), examined the conditions under which a thin veneer of governance developed among states (Stein 1982; Krasner 1983; Keohane 1984; Snidal 1985). In the early 1990s, though, James Rosenau and others noted an increasing likelihood that one could see steering in the transnational arena outside the participation of states. To paraphrase his seminal work with Ernst-Otto Czempiel, it was increasingly common to have “governance without government” (Rosenau and Czempiel 1992). Transnational steering could be generated by a wide variety of actors, including but not limited to states. This spawned a large literature looking at the nature and extent of both global order and order in particular transnational issue areas.

We see security issues as those that actors identify as relevant to their security and mobilize collective resources around (Buzan 1991). Our contributors are attentive to contemporary security challenges. Many of these focus on the management of violence as is familiar within traditional definitions of security (Walt 1991). Other papers extend to human security or broader efforts to shape security outcomes in ways that avoid the use of violence. Affecting the degree of order around a security issue, then, is security governance.

(p.4) How Transnational Security Governance Varies

This volume focuses on two ways in which this collective ordering of security may vary. First, and most fundamental, is the degree or effectiveness of governance (Young 1994, 30; 1999, 11–15). In principle, governance can be present or absent (Young 1980). A situation of true anarchy, confused and inchoate could be characterized by an absence of governance. More commonly, governance is more or less effective. Effectiveness refers to the degree to which collective steering or management—or what looks this way in retrospect—is present, either generally in a system of governance, or specifically in a given issue area. Unlike a situation that is entirely ungoverned, in some situations rules or standards may exist but not be acted on and lead to confused expectations and little collective order. These would be situations of weak governance. As expectations grow clearer, processes for resolving distributional conflicts are known, and some level of collective order is apparent, the effectiveness or functionality of the governance situation grows. Most effective are situations in which expectations are clear and the capacity for collective action is widespread.

The ability to take action and “solve” global or transnational problems—or alternatively, the degree to which governance is ineffective or dysfunctional—has been a core point of debate among scholars of global governance. Traditionally, the mostly realist scholars focused on security have been pessimistic about the potential for effective governance of the issues they study. As Robert Jervis (1982, 358) explained, security regimes are both especially valuable and especially difficult to achieve. Our survey of the world, however, reflects more governance than realists expect on a variety of security issues. Some of this takes the familiar form of agreements among states (Montgomery, this volume) or hegemonic leadership by the United States (Cooley and Nexon, this volume). More, however, is the result of forms of governance that the realist literature has left unexamined; from the transgovernmental linkages among state bureaucrats that govern weapons of mass destruction (Eilstrup-Sangiovanni, this volume) to the multi-stakeholder initiatives that govern private military and security services and conflict diamonds (Avant, Westerwinter, this volume) to informal arrangements to counter piracy or generate state responsibility for civilians harmed in their military actions (Carpenter, Percy, this volume). The effectiveness of governance on security issues as well as its form is puzzling to realist analysis.

But extrapolating from liberal analyses (more prominent in other issue areas) does little better. As often as not, governance is not led by the United States or other major powers, as liberal theories of hegemony expect, but by much smaller states or even non-state actors. Governance does sometimes result from the shared interests that liberals expect to generate cooperation, as in the case of (p.5) antipiracy (Percy, this volume). In many issues, though, shared interests are the result of the process itself. This is particularly the case in the governance of private military and security services (Avant, forthcoming).

To the extent that they expect governance at all, traditional realist analyses should expect that the processes reflect and serve the concerns of dominant states. This is the second dimension of variation we focus on: who governance serves. We examine who governance serves in two different ways. First, it can serve some ideas and/or interests rather than others. Much of the literature on global governance, for instance, focuses on the way it reflects liberal ideas and/or the interests of dominant powers, such as the United States or the United States and the European Union (Drezner 2007). Second, it can serve many or only a few. Mattli and Woods (2009, 12) differentiate between common interest regulation (where processes for inclusion and accounting for the interests of those governed are in place and there is broad demand for these processes) and captured regulation (where processes for inclusion are not present and demand for inclusion is absent). The more potential for participation and mechanisms for the weak to check the powerful, the more inclusive the process is; processes characterized by capture are exclusive. Those focused on the accountability of governance (Kahler and Lake 2003; Grant and Keohane 2005) or the ability of the weak to check the powerful (Burris et al. 2005) also look generally at who governance serves.

Realists generally assume any governance process is likely to be captured by the major powers—the United States and Europe (Drezner 2007), perhaps China. Only to the degree that what serves the concerns of dominant powers is also valued by others—something implicitly assumed in hegemonic stability theory—should more general concerns be served (Drezner 2007; Brooks and Wohlforth 2008). Liberals and rationalists have often looked to states for mechanisms through which weaker players or more general interests might be served. Some thus look to state-dominated international treaties or “hard” law for the best results (Abbot and Snidal 1998). Others expect international agreements to remove power from governments and thus suffer from democratic deficit (Grant and Keohane 2005). Our analyses show that governance processes do not line up so neatly. Sometimes we observe governance processes in which even the hegemon can be captured by narrowly interested clients (Avant, Cooley and Nexon, this volume). Furthermore, as Cooley and Nexon show, these interests need not be within the United States at all. Interpersonal connections between Georgian elites and the George W. Bush administration led the United States to take action that risked stability in Eastern Europe (Cooley and Nexon, this volume). Informal processes can be tools for generating capture for hegemonic states (Eilstrup-Sangiovanni, this volume) but also for seemingly weaker non-state actors (Westerwinter, this volume). Informal processes can yield results (p.6) that serve broad (Eilstrup-Sangiovanni, Avant, Percy, this volume) or narrow (Cooley and Nexon, Westerwinter, this volume) concerns.

In sum, there is broad variation in the effectiveness of security governance across issues. When there is governance there is also variation in who it serves. Neither traditional realists nor liberal and rationalist approaches can explain the pattern of variation we see. We thus turn to the logic behind social network theory for a framework through which we might begin to solve these puzzles of transnational security governance.


We define a network as a set of nodes and a set of ties representing the presence or absence of relationships between nodes.2 Nodes are individuals, though they may be aggregated into corporate entities, such as states or NGOs. Ties are relationships. They signify interactions and can allow for the flow of material and nonmaterial resources, such as money, weapons, information, or advice. Network structures are “persistent patterns of relations among agents that can define, enable, or constrain those agents” (Hafner-Burton, Kahler, and Montgomery 2009, 561). Rather than reified dots and lines, the contributions to this volume consider social networks as the substrate of past social interactions, as “the congealed residues of history” (Padgett and Powell 2012, 3).

Given our broad understanding of networks, any organizational form can be represented and examined as a network. As will be familiar to many, transnational advocacy networks and global policy networks can be analyzed in terms of the patterns of relationships that exist among the actors involved and how these patterns affect outcomes, such as advocacy success or the provision of public goods (Keck and Sikkink 1998; Reinicke 1998). But markets and hierarchies can also be represented in terms of their structural network features (Podolny and Page 1998). A spot market, for example, can be conceived of as an assemblage of isolated nodes with solely transitory ties; an oligopolistic market as a network with several hubs; and a monopoly as a star-shaped network where the majority of nodes are connected asymmetrically to one highly central unit.

This perspective is analytically distinct from approaches that understand networks as a particular form of organization different from markets and hierarchies (Powell 1990). It flows from a relational ontology that sees actors as constituted by, rather than independent of, relations. As put by Harrison White, “without persons being presupposed as actors, attention necessarily shifts to confluences of (p.7) observable processes-in-relations. Out of these emerge actors and locations of social action” (White 1997, 59–60, italics in original). The analytical wager of this approach is that “significant aspects of causation in social and political life inhere in transactions themselves” (Nexon 2009, 27). From this approach we look at how the structure and/or quality of network relations affect the dynamics of collective action—and thus governance—rather than claiming that a “network” is a particular form of governance.

Studies focused on networks as a particular organizational form tend to bundle together various relational features. They assume that “networked” organizations are structurally flat, benign, power free, and characterized by informality and high levels of trust (Powell 1990; Sørensen and Torfing 2007). We concur that flat relations, informality, and trust can be important features of networks. Rather than assuming that these features are inherent to particular groupings, though, we investigate the actual features of particular networks as likely to affect governance dynamics and outcomes. As many have demonstrated, informal networks can nonetheless be quite centralized (Carpenter 2011; Westerwinter 2013). An informal centralized network should lead to different expectations than an informal flat one. To understand the way these variables should matter, we turn to the literature on networks for insights into which variations might be important for understanding variation in transnational security governance.

How Networks Vary

There are many ways that networks can vary. We focus on three broad variations likely to be important in transnational security governance. First is the position of actors within the network. In sociopolitical life, most networks reflect unevenly distributed ties among nodes. This creates differences in their positions. Actors with a large number of direct connections to others occupy central network positions (Freeman 1978; Bonacich 1987). Centrality is a key mechanism through which nodes garner influence in network theory and thus one prominent source of power in the analyses in this volume. Actors can also vary in the degree to which they connect otherwise unconnected others. Such brokers are located at critical junctures and bridge “structural holes” that separate different actors or regions of a network (Granovetter 1973; Burt 1992). Brokers can also connect different types of networks. When one actor simultaneously participates in more than one network, that actor is a multiplex node. Multiplex nodes can link interpersonal to interorganizational or interstate networks, or link networks in different issue areas. Multiplexity is implicit in arguments about regime complexes where some actors are part of several institutions that govern a policy problem and thereby connect these institutions to each other (Raustiala (p.8) and Victor 2004; Orsini et al. 2013). Centrality and broker positions have been key to a variety of arguments about international relations outcomes (Nexon and Wright 2007; Nexon 2009; Goddard 2009, 2012; Carpenter 2011; Murdie 2014). Multiplexity has not received as much attention in analyses of global politics but we demonstrate that the degree to which nodes overlap and whether they introduce competitive or complementary dynamics is a key concept for understanding the security governance issues we address in this volume. It is likely to have increasingly important effects on contemporary global politics and is worthy of much more study.

Our second variable is the distribution of ties within the network. A primary distinction between networks is whether they are decentralized or centralized. In a decentralized network, nodes are similar to each other in terms of their pattern of connectedness with the majority of nodes having roughly the same number of links (Barabási 2003; Butts 2009). A decentralized network resembles the “flat” structure that analyses of networks as distinct types of organization often assume (Powell 1990). One might imagine a spontaneous online community to resemble a flat network. Studies of Anonymous, for instance, suggest that it is highly connected with no particularly central nodes (Kushner 2014). Centralized networks, by contrast, are dominated by one or a small number of highly central nodes (hubs) while the majority of actors have few relations. The coexistence of a few hubs with a large number of sparsely connected nodes establishes a hierarchy among nodes (Barabási 2003, 70). Flat networks can be good for spreading information but are unpredictable in generating sustained action. Centralized networks can be efficient for generating collective action by establishing, and reinforcing a focal point. They can be less suitable for the fast dissemination of information particularly in situations in which the network hubs are unable or unwilling to operate effectively.

The third broad category of variation across networks that we highlight is the quality of ties. This can refer to a range of phenomena. Some actors may be linked by friendship, others by enmity with obviously different consequences for interaction. Ties can interact to promote brokerage (enhancing additional connections) or closure (tighten existing connections) (Burt 2005; Obstfeld et al. 2014). Both brokerage and closure are forms of social capital but activate different social dynamics among network actors. Closure leads to strong ties, indicating high frequency and strong coupling, while brokerage is frequently among weaker ties, indicating lower frequency and more fragile coupling. Both brokerage and closure have a dark side. Closure can generate rigidity and “echo” (echo blocks new information and builds “us” versus “them” categories) that makes networks fragile; brokerage can yield instability. Finally, relations can be formal or informal. Formal ties can, for example, be constituted by contractual (p.9) relationships among actors as defined in international agreements. Informal ties include connections grounded in actors’ memberships in the same international or transnational institutions and exchanges of information and advice. Although formal ties have traditionally been seen as more effective in enabling political actors to overcome collective action problems and shaping behavior (Abbot and Snidal 1998), many of the contributors to this project focus on the importance of informal ties for governance.

To summarize, network theory places relationships among actors and the properties of these relationships at the center of studying political processes. We focus on variation in the position of actors, the distribution of ties, and the quality of ties to better understand variation in transnational security governance across a broad range of empirically relevant policy problems.

Networks and Power

Our network approach broadens conventional analyses of power in several ways. To begin, a network perspective places a premium on relational understandings of power; power that derives from and is exercised through actual or potential relationships rather than residing only in a node. Also, since it is agnostic with respect to what entities constitute the nodes in a network, network analysis allows for a multiplicity of actor configurations ranging from states to IGOs, companies, NGOs, and even individuals to be seen as power wielders. Finally, it makes no stipulation that the use of power be detrimental to those it affects.

Many have argued that power is intrinsically relational (Simon 1953; Dahl 1957; Nagel 1968; Baldwin 1979; Barnett and Duval 2005). Robert Dahl’s “intuitive idea” that “A has power over B to the extent that he can get B to do something that B would not otherwise do” is, for example, firmly based on the idea that “power is a relation” (Dahl 1957, 202–203). As Baldwin (2013, 274) notes, the change in actor B’s behavior “may be understood broadly to include beliefs, attitudes, preferences, opinions, expectations, emotions or predispositions to act.” Barnett and Duval (2005, 8) define power as “the production, in and through social relations, of effects that shape the capacities of actors to determine their own circumstances and fate.”

Relational perspectives on power stand in contrast with conventional international relations scholars who often conceive of power as an attribute, something an individual actor can have (or not) and which can be used to affect others’ behavior and accomplish ends. Often, power resources are treated as if they were power itself (Baldwin 2002, 180). This “power-as-resource” conception figures prominently in Waltz’s (1979, 98) structural realism where military and economic capabilities are defined as “attributes of units” (i.e., states) and (p.10) has since then exerted sustained influence on the thinking and theorizing about power and influence in the discipline of international relations.3

Moving away from a commitment to states as the only relevant nodes allows network frameworks to fit more easily with analyses of power that see many different sources and dimensions of power. Moving away from the fixation on national power as the key to global governance leads to a more open examination of why it is that any particular actor A can generate different behavior from actor B than we would otherwise expect. This acknowledges a greater variety of potential political spaces (Berenskoetter 2007, 17). There are various sources of power and different types of actors have more access to some than others. Multinational corporations have greater access to money, states to military force, NGOs to principles, but each of these can be important for generating changes in others’ behavior.

Finally, network analysis is open to the idea that power may be “power to” rather than “power over.” This notion of power to generate something new, connected to collective action is downplayed in traditional security studies, which has been more attentive to zero sum notions of power. As Berenskoetter (2007, 4) points out, though, it is fundamental in Hannah Arendt’s analysis, which focused on power as something creative and productive, and also present in Talcott Parsons’ work. The “power to” is agnostic about the zero sum (or not) qualities that generate capacities but is focused on their potential for generating creativity and collective action (Baldwin 2013, 278–279).

Thinking of power as relational, attached to different nodes with various bases, and with a broad range of effects pulls our analysts away from general statements about generic power and toward a more nuanced analysis of how different manifestations of power work in particular issue areas. This is consistent with those who have broken power down by its relational specificity and mode of operation (Barnett and Duvall 2005) or with those who emphasize the way different bases of power (or “capital”) operation in various fields (Bourdieu 1986). Most important, it is open to understanding power in its many and various forms (Berenskoetter and Williams 2007).

This volume’s contributors highlight several particular insights. First, an actor’s relative position in a group of governors can be seen as an independent basis of power in international relations (Kahler 2009). This network power resides in the persistent patterns of relations among the actors involved in governing an issue (Hafner-Burton, Kahler, and Montgomery 2009). Relational structures (p.11) are a source of power because they provide actors in specific positions with social capital, information, coalitional opportunities, and other resources which can be exploited to pursue collective action (Knoke 1990; Burt 1992; Lin 2001). In the anti–land-mines network, for example, NGOs in central positions used the informational and coalitional advantages emanating from their positions to shape the agenda of negotiations over an intergovernmental land mine treaty and influence both the contents of the treaty as well as the development of the institutional regime created to implement it (Bower, this volume). Similarly, multinational corporations that participate in several regimes to govern different conflict minerals used their multiplex positions to disseminate governance structures across regimes (Haufler, this volume).

Power often works through diffuse mechanisms. The structural and productive power that Barnett and Duvall (2005) highlight are two important ways in which diffuse power works. Montgomery (this volume) demonstrates that diffuse power can have greater impact than direct power in many situations. And Avant (this volume) shows that diffuse power can even affect the preferences of the United States and its support for transnational governance processes.

How an actor behaves in a relational setting depends on how that actor sees the relationships it is involved in. An actor may think that she has more friends—or enemies—than others do. There may be more or less agreement about the configuration of power among the actors in a network. These perceptions may shift with issue areas or recent interactions. Thus, while Kahler (2009) usefully describes network power vis-à-vis its bargaining leverage, social power, and exit capacities, we must bear in mind that network relations are, in part, a function of different actors’ perceptions. For instance, as Carpenter explains in this volume, advocates for the making amends campaign were not successful in making this a global campaign to change global norms of war until the organizers were able to change the perception of the Campaign for Innocent Victims in Conflict (CIVIC). Reconstituting its perceived relationships with both actors (at the UN) and issues (protecting civilians) in the global human security network was critical to CIVIC’s success in global agenda setting.

If power emerges from and depends on the formation and continuation of relationships, it implies that as relationships build, power can grow; and if they change or break down, power can shift and dissipate. This dynamic view of power implicitly sits behind the notion of networking, or strategically creating relationships to help one reach her goal (Lake and Wong 2009). Its implications for thinking about power in global politics are significant. The actions an agent takes not only affect the task at hand but also affect that agent’s relationships and thus future possibilities to shape governance processes. Different actions, drawing on different sources of power can thus have consequences for future possibilities. Using coercion, for instance, may undermine some relationships (p.12) in ways that diminish an actor’s network position and thus power. Using persuasion or appealing to social norms, may enhance an actor’s institutional authority or network position (though may constrain the conditions under which she can use coercion). This approach thus emphasizes that wielding one form of power may have consequences for an actor’s ability to wield another form of power, enhancing or diminishing overall power. It also highlights the agency of those who relate to a powerful actor for that actor’s continued power.

Networks, Power, and Governance

Many of the recent analyses of networks in global governance take networks as they find them (Kahler 2009). While this is useful for explaining whether governance is effective or not and who it serves, it ignores the degree to which agents actively shape their network position and the overall qualities of the networks they participate in. We are interested in both the dynamic elements of networking and the network qualities at any particular moment. Both affect governance. The power of network logic can generate new tools for evaluating, and elaborating on, existing arguments in the field of security analysis. Network logic also offers new insights into strategic networking and the way in which network structure affects governance outcomes. Table 1.1 describes the way these conceptions structure this volume.

To begin, Alexander Montgomery offers novel insights on the value of centrality. Using network techniques, he demonstrates new insights into how centrality matters for both governance effectiveness and whom it serves. First, with regard to the literature on conflict management, network techniques uncover a new mechanism—diffuse socialization—that is more important than dispute resolution in efforts to govern conflict. Second, with regard to the literature on international hierarchy, he shows that its core insights might apply more broadly. Although the United States’ centrality allows it to exercise domination through its alliance system, so does the centrality of other players in other alliance systems.

The next three chapters look explicitly at strategic networking and its impact on who governance serves. Charli Carpenter analyzes the way norm entrepreneurs constructed relationships within the context of particular structural constraints. Constructing connections with central actors and central issues gave these entrepreneurs greater agenda-setting effectiveness. Activists wanting to generate a norm that militaries should make amends to civilians harmed by (even lawful) military operations constructed relational ties first with central actors in the United States and then with central issues in the human security network in order to attract the support of powerful gatekeeping hubs.

(p.13) Alexander Cooley and Daniel Nexon analyze how multiplex nodes—particularly those that sit simultaneously in personal and high-level governmental networks—can shape government perceptions in perverse ways. In the case they examine, the strategic construction of personal relationships between Tbilisi and members of the Bush administration nearly hijacked US foreign policy. Interpersonal connections between Georgian elite and US administration officials led the United States to support the aims of a narrow set of Georgian elite in a way that exacerbated the potential for conflict between Georgia and Russia, something the United States did not want.

Deborah Avant examines how multiplex nodes and the logic through which they connect can affect US preferences. Connecting with the United States or “netting the empire” was key to both frustrating (small arms) and achieving (private military and security companies, PMSCs) effective governance and to serving narrow (small arms) or broad (PMSCs) aims. She finds that brokerage strategies are more supportive of effective transnational governance and closure strategies more easily used to halt governance efforts. Forming connections to affect US preferences and authority claims is an important avenue for both domestic and transnational actors to influence the direction of transnational security governance.

The final five chapters focus on how network variables affect governance outcomes in particular issues. Mette Eilstrup-Sangiovanni argues that informal transgovernmental networks (TGNs) to control the proliferation of weapons of mass destruction (WMD) rather than formal IGOs are most likely to arise when powerful central actors face domestic veto players and expect enforcement to be rare. TGNs are more likely to be effective when the participating numbers are small. In their execution, though, she argues that TGNs generate more benefits to central (and powerful) actors and thus may be associated with narrower interests. She uses the Proliferation Security Initiative and Missile Technology Control Regime to illustrate her claims.

Adam Bower argues that the ban on antipersonnel mines codified in the 1997 Mine Ban Treaty is not reducible to the properties of the actors involved or the quality of particular ideas, but is rather the product of the relationships among the actors involved. Specifically, key actors with central positions had disproportionate influence in reshaping the policy agenda concerning antipersonnel mines. His account explains how the goal of a complete prohibition won-out in the face of sustained opposition, and why global military powers led by the United States were unable to generate support for an alternative framing that retained antipersonnel mines as legitimate weapons of war. His analysis suggests that central positions can generate effective governance even in the face of traditionally powerful resistance.

(p.14) Oliver Westerwinter examines how the informal communication networks among states, the diamond industry, and NGOs affected the development of a more rigorous formal monitoring system in the Kimberley Process (KP) that governs conflict diamonds. Particularly, those with central and broker positions were able to use these to push for monitoring that increased the effectiveness and broadened the interests served by the agreement despite resistance from materially powerful but structurally poorly positioned actors. Like Bower, his analysis suggests that centrality, even informal centrality, can lead even weaker players to generate effective governance that serves broader interests.

Virginia Haufler compares the governance of conflict diamonds in the Kimberly Process to the governance of other conflict minerals. While the push to regulate rough diamonds occurred quickly, was negotiated in a multi-stakeholder forum that included almost the entire diamond market, and resulted in a single comprehensive public-private regulatory institution leading to relatively effective governance, pressure to regulate other conflict minerals (tin, tantalum, tungsten, gold) has meandered through in a stop-and-start manner, has not always involved multi-stakeholder processes, and has not resulted in the creation of a single global regime. She explains this puzzle by looking at the way in which associational networks (such as membership in business policy organizations) and contractual networks (relationships within an industry value chain) interact. The KPs resulted from associational relationships embracing corporate social responsibility (CSR) and a dominant central player (De Beers) who reinforced that orientation. Without value chain support, work on other conflict minerals has not been as effective despite a model for action in the KPs. The relationships among networks can be important for reinforcing or undermining effective governance.

Finally Sarah Percy explains why effective multinational military cooperation against piracy has been possible—even among states that are at times hostile toward one another. An informal but centralized system, she argues, has allowed brokers to mediate timely communication and coordination among navies of states that share a goal but lack positive relationships. Cooperation has also been facilitated by the shared values and discourses among navies and a relatively clear international legal framework for surrounding piracy. In contrast to cases examined by other contributions (Avant, Cooley and Nexon, Carpenter, Westerwinter, this volume) the actors that occupy central and broker positions in the counter-piracy network have not used their position to strategically manipulate governance processes and outcomes in their favor. Percy ties this to agreement on the goal (countering piracy) and beliefs that any strategic use of network positions would jeopardize achieving this goal. (p.15)

Table 1.1 Overview of Variables and Cases

Network Variables

Governance Outcomes

Effective or Ineffective

Benefits Broad or Narrow Concerns

  • Position of actors

    • Central

    • Broker

    • Multiplex

  • Montgomery—central position, conflict reduction

  • Carpenter—construction of relationships with central actors, making amends

  • Bower—central position, land mines

  • Westerwinter—central and broker positions, governance of conflict diamonds

  • Haufler—multiplex nodes, conflict diamonds versus conflict minerals

  • Montgomery—central position, alliance concerns

  • Cooley and Nexon—multiplexity, US preferences, Georgia

  • Avant—multiplexity, US preferences, small arms versus PMSCs

  • Bower—central position, land mines

  • Westerwinter—central and broker positions, governance of conflict diamonds

  • Distribution of ties

    • Decentralized/flat

    • Centralized/hierarchical

  • Haufler—multiplex nodes, conflict diamonds versus conflict minerals

  • Percy—informal centralized, antipiracy

Percy—informal centralized, antipiracy

  • Quality of ties

    • Brokerage versus closure

    • Formal versus informal

  • Avant—brokerage versus closure, small arms versus PMSCs

  • Eilstrup-Sangiovanni—informal, WMD

  • Percy—informal centralized, antipiracy

  • Eilstrup-Sangiovanni—informal, WMD

  • Westerwinter—informal, governance of conflict diamonds


Much of what we see in security governance is puzzling to traditional analyses. We see more and different governance than realists expect and different routes to governance than liberals anticipate. Collective ordering is not free of power but the new power politics involves more types of actors with various bases of (p.16) power, and works through forms that are left out of many theories. Focusing on the nodes and ties present in different issue areas and using some basic theoretical tools from network logic, we were able to generate explanations for a wide range of security governance behavior and outcomes. Network theory can yield new insights even in well-studied areas of global governance such as how international organizations affect conflict management and whose interests shape alliance concerns (Montgomery, this volume). Beyond that, the volume’s analyses yield important new insights relevant to our understanding of transnational security governance.

First, relations are dynamic. Actors can improve their network position by constructing new linkages. While this is implied in the logic of “networking,” our contributors explicitly attend to the strategic construction of network position. Constructing perceived associations with key hubs and gatekeepers can build actual ties and lead to agenda-setting effectiveness (Carpenter, this volume). And constructing interpersonal linkages with key nodes in powerful networks can be an important way through which otherwise less powerful players can attempt to generate governance outcomes that serve their interests (Cooley and Nexon, this volume). Although many of the contributions focus on network dynamics over time, the chapters in the networking section look most explicitly at the strategic qualities of these dynamics.

Second, multiplexity—a situation where one node sits simultaneously in more than one network—can be a key strategic resource affecting power politics and transnational governance. Network theorists have long held that multiplexity can lead to the spread of ideas from one network to another and the emergence of new institutional forms (Padgett and Powell 2012). This idea implicitly animates the logic of interdependence theory (Keohane and Nye 1977). Fewer analyses, however, have examined the power implications of the mechanisms through which this occurs. Our contributors examine several mechanisms through which multiplex nodes can strategically affect the identity and inclinations of the networks of which they are a part. Interpersonal ties can be used to affect organizational inclinations (Cooley and Nexon, this volume); transnational as well as domestic ties can affect national inclinations (Avant, this volume); and informal ties can affect formal institutional structures (Westerwinter, this volume).

The power agents deploy has consequences for future power. Power rooted in violence or economic capacity and used alone to bribe or punish others may be useful for forestalling action and for producing short-run effects, but this use of power often erodes relationships in ways that hinder future collective action. Generating collective action is more likely when prospective governors root their action in expertise, principles and relations, and leverage military and/or economic capacity in their service. Leveraging various power sources in (p.17) combined strategies can generate greater influence over governance processes through layered relational ties. Thus, using power as traditionally envisioned can be degenerative, eroding future capacity, while “softer” forms of power often hold a generative quality leading to greater future capacities (Avant, Eilstrup-Sangiovanni, this volume).

Also relations can be power tools for both the weak and the strong. Networks have often been portrayed as a means to empower otherwise weak actors, such as NGOs or small states, to voice their interests and influence governance outcomes (Keck and Sikkink 1998; Reinicke and Deng 2000). As the chapters of this volume show, the picture is more complex. Central positions in formal and informal networks enabled NGOs to shape the contents and development of the international regime to ban land mines (Bower, this volume) and successfully bargain for a formal monitoring system in the KP that went beyond what many states and diamond industry representatives were initially willing to agree to. However, often networks are just another way for already powerful players, such as major powers and multinational companies, to achieve their goals. The United States, for example, used its privileged network position in order to shape the form and content of the transnational regulation of private security companies (Avant, this volume) and large multinational corporations exploit the leverage that emanates from their multiplex positions at the intersection of economic and political networks to shape the evolution of the institutional complex that governs conflict minerals (Haufler, this volume). In short, relations are not always the power tool of the otherwise weak but can be used by both weak and strong players. The question of who uses networks to exert influence is more fruitfully treated as a matter for careful empirical investigation rather than theoretical assumption.

We may have been too quick to toss out the importance of functional explanations for security concerns. While it is clearly the case that many obvious problems are not solved and there are roles for entrepreneurs in constructing problems, the perceived need for governance can be important for generating it. While this point is most clearly made in Percy’s chapter on antipiracy, it is also important for the story Avant tells about governing private military and security services and Bower tells about governing land mines. Establishing a problem and gathering stakeholders around it can shift relational dynamics in ways that can engender greater collaboration.

Finally, our analyses have significant implications for how to understand the role of the United States and other central players in the contemporary global system. Traditional realist and liberal analyses have assumed that hegemons use their power (military or financial) to lead others to governance outcomes through force or encouragement. While no doubt true at times, the relational ontology that informs our project leads us to imagine other possibilities. Hegemons like (p.18) the United States are also networks of individuals and agencies. The increased connections noted by globalization theorists have generated more and more multiplex nodes in these networks. Thinking of the United States as a central and powerful network means the United States can work through multiplex nodes to shape the inclinations and actions of others (Eilstrup-Sangiovanni, this volume) and the reverse (Avant, Cooley and Nexon, this volume).


(1) In economic terms, public goods are defined as goods that are non-excludable and non-rivalrous in consumption. Although there are difficulties with using this term that have been explored elsewhere (Avant et al. 2010b, 365–367), it is widely used in the global governance literature and so we use it here. We do not assume that a public good is “good” or distinguish public “goods” from public “bads”—both are simply non-excludable and non-rival in consumption.

(2) For general overviews of network theory, see Wasserman and Faust 1994 and Scott 2000.

(3) A recent example which continues this theoretical lineage is Drezner’s (2007) notion of state power as the size of its national economy measured in gross domestic product. Note, however, that Drezner complements the size of the national economy with a state’s level of vulnerability, which introduces a relational dimension in his understanding of power.