CHAPTER 17
CONSTRUCTIVISM
IAN HURD
The basic insight behind the constructivist approach can be understood by unpacking a quick observation made by Alexander Wendt. He says that "500 British nuclear weapons are less threatening to the United States than 5 North Korean nuclear weapons" (Wendt 1995, 73). In this little observation are found traces of the features that distinguish constructivism from other approaches to international relations, including its critique of materialism, its emphasis on the social construction of interests, its relationship between structures and agents, and its multiple logics of anarchy. On its surface, the empirical puzzle of the threat embodied by North Korean missiles is easy to explain: As Wendt (1995, 73) says, "the British are friends and the North Koreans are not." This of course begs an understanding of the categories of friend and enemy, and it is through this opening that Wendt and other constructivists have addressed both important substantive aspects of international relations (for instance, "how do states come to see others as friends and as enemies?") and the philosophical background it presupposes (for instance, "how can we study social and relational phenomena like 'friend' and 'enemy in international relations?").
This chapter examines the features that distinguish constructivism from other approaches to international relations and then looks at some controversies within constructivist scholarship today and between constructivists and others. There are many excellent short histories of the constructivist school (e.g. Barnett 2005; Reus-Smit 2005), and my goal is to avoid repeating them and instead explain what I think the term constructivism means in international relations. To do so, I also
For very useful comments on earlier drafts, I thank Karen Alter, Chris Reus-Smit, and Duncan Snidal.
constructivism
299
define other approaches, including materialism, realism, and rationalism, in order to show how constructivism differs. This involves some controversy, because the lines that separate them are not at all clear. In what follows, I take realism to be at its core about materialism (that is, the theory that states respond to material needs, incentives, and power) and rationalism to be about instrumentalism (that is, the theory that states pursue individual advantage by calculating costs and benefits). Constructivism, by contrast, emphasizes the social and relational construction of what states are and what they want. All these approaches might be used to focus on power politics, cooperation, conflict, or any other substantive phenomena. It is, therefore, wrong to associate a substantive interest in power exclusively with realism, because all the "paradigms" of international relations are interested in power, as either motivation, cause, or effect. I differentiate realism as a particular theory about material power in international relations, in contrast with constructivism's emphasis on the social meaning attached to objects or practices.1
In asking for an explanation of the importance in world politics of social concepts like friend and enemy, the constructivist challenge opened two paths. One was more empirical and used the tools provided by Friedrich Kratochwil (1989), Nicholas Onuf (1989), Wendt (1992), and other constructivists to explain anomalies of other approaches. The other was more conceptual and concerned how these social concepts might work in the world and how they could be studied and used in study. From constructivism's starting point as a reaction to materialism, individualism, and rationalism, the empirical branch of research was like a downstream flow; it applied the insights of constructivism to understand interesting patterns, behaviors, and puzzles. The philosophic branch went upstream—it sought to understand the reasons for, and implications of, the differences between constructivism and other approaches to social phenomena.
1 The Distinguishing Features of Constructivism
This section outlines four features of constructivism that distinguish it from other approaches and show how constructivism addresses both philosophical and empirical issues that were inaccessible through the prevailing models of international relations in the 1980s. The four are not necessarily exclusive to constructivism, but each has a constructivist variant that is distinct from both the materialism of
1 J. Samuel Barkin (2003), by contrast, defines realism as a concern with "power" and then notes that this is consistent with social construction. I agree that classical realists incorporated nonmaterial forces, but by my definition that makes them less "realist."
300      ian hurd
realism and the rationalism of neoliberalism, and carries distinct implications for how world politics is studied.
1.1 An Alternative to Materialism
The original insight behind constructivism is that meaning is "socially constructed." This is also the source of the label "constructivism." Wendt (1992, 396-7) says "a fundamental principle of constructivist social theory is that people act toward objects, including other actors, on the basis of the meanings that the objects have for them."2 In a socially constructed world, the existence of patterns, cause-and-effect relationships, and even states themselves depends on webs of meaning and practices that constitute them (e.g. Kratochwil 1989). These meanings and practices might sometimes be relatively stable, but they are never fixed and should not be mistaken for permanent objects.3 As ideas and practices vary over time or space, patterns that once looked solid and predictable may change as well. For instance, sovereignty is a social institution in the sense that a state can be sovereign only when it is seen by people and other states as a corporate actor with rights and obligations over territory and citizens (and they act accordingly). The practice of sovereignty has changed over time, and the powers and identities of actually existing states have changed as well (see, e.g., the essays in Biersteker and Weber 1996). To take a more concrete example, since 1945 the idea has spread that massive human rights violations by states against their citizens may legally justify international intervention. Sovereignty is thereby changing, and the autonomy of some rulers (that is, rights violators) is reduced while that of others (potential interveners) is increased. Sovereignty is an important organizing force in international relations that rests on the shared ideas of people and the practices people engage in.
A contrasting approach to "social construction" in world politics is the position known as "materialism," which suggests that material objects (bombs, mountains, people, oil, and so on) have a direct effect on outcomes that is unmediated by the ideas people bring to them. Neorealism and neoliberalism are explicitly materialist approaches to world politics. They seek to explain international patterns and behaviors as the result of purely material forces, particularly the military hardware, strategic resources, and money that they see as constituting "power." For example, John Mearsheimer (1995, 91) argues that "the distribution of material capabilities among states is the key factor for understanding world politics." Among neoliberals, Joshua Goldstein and Robert Keohane (1994) identify states' material interests as distinct from people's ideas about the world, and their research on the causal effects of ideas uses as its baseline the materialist hypothesis. Neorealists and neoliberals in
2 This insight appears also in the work of Hedley Bull and the English School as well as of some classical realists.
3 This is the mistake of "reification."
constructivism
301
the 1980s shared a commitment to materialism in which socially mediated beliefs were not important autonomous forces, and they argued among themselves over the likely implications of such a world for patterns such as cooperation, institution-making, arms races, and balancing (see, e.g., the essays in Baldwin 1993).
The ideas that give shape to international politics are more than just the beliefs of individuals. They include ideas that are intersubjective (that is, shared among people) and institutionalized (that is, expressed as practices and identities). Intersubjective and institutionalized forms of ideas "are not reducible to individual minds" (Wendt 1999, ch. 4; Legro 2005, 5). Jeffrey Legro (2005, 6) summarizes the constructivist understanding of ideas: "ideas are not so much mental as symbolic and organizational; they are embedded not only in human brains but also in the 'collective memories,' government procedures, educational systems, and the rhetoric of statecraft." This makes it clear that the constructivist insight is not that we replace "brute materialism" with "brute idealism" (cf. Palan 2000). Rather, constructivism suggests that material forces must be understood through the social concepts that define their meaning for human life.
A purely materialist approach has difficulty explaining why the United States should see British missiles as any less threatening than North Korean missiles. The "self-evident" friendliness of Britain toward the United States as compared to the apparent hostility of North Korea is not self-evident from a purely material perspective. After all, the physical consequences of an attack by the nuclear weapons of either country would be devastating. The brute material threat to the United States posed by a British nuclear weapon is at least comparable to, and probably much greater than, that of a North Korean weapon. The difference between the two is the conviction among many American leaders that the North Koreans are more likely to act aggressively toward the United States than are the British. This conviction is based on interpretations of history, rhetoric, and behavior, and it generates the expectation that war with North Korea is more likely than war with the British, and in turn leads to different policy strategies in response to their weapons.
For constructivists, beliefs, expectations, and interpretations are inescapable when thinking about international affairs, and their importance shows that the materialist position is untenable. While the shift from a materialist to a socially constructed view of international relations was controversial in the early 1990s, it has now been broadly accepted. The constructivist insight has been largely internalized by the discipline.4 Even materialist theories of international relations now generally openly include at least two kinds of ideas (though mostly individual rather than collective ideas): first, "non-material" factors such as (for Mearsheimer 2001, 58) "strategy, intelligence, [and] resolve," and, secondly, socially constructed interests. However, they usually also claim that the practical importance of the social content
4 Jennifer Sterling-Folker (2000) argues that this was made easier by the fact that many putatively materialist theories of international relations already incorporated social content. See also Wendt (1992); Williams (2005).
302
ian hurd
of international relations is minimal when compared to the influence of brute material factors, and so the research agendas of neorealism and neoliberalism have at once conceded the constructivist insight while maintaining their core claims.
As the socially constructed nature of world politics has been broadly accepted, it has become clear that what remains contestable between constructivists and others is how (not "whether") this insight affects the study of world politics, both in its methodology and in its substance. The debate over the construction of state interests and their sources follows from this debate.
1.2 The Construction of State Interests
The scholarly interest in the "national interest" has always been central to international relations and foreign-policy analysis. The constructivist approach has been productive in this area because of its focus on the social content involved in the production of international relations, including state interests.
While most scholars now acknowledge that state interests are at base ideas about needs, many nonconstructivists maintain that the content of those interests is for practical purposes unchanging and includes some combination of the desires for survival, power, wealth, and security. They contend that the socially constructed nature of interests does not alter the fact that the primary interests that drive states are prefigured by the material resources and situation of the states, and so states are either constructed by material forces or can be treated as if their construction is irrelevant to their interests and behavior (e.g. Brooks and Wohlforth 2007). States are "minimally constructed."
By contrast, constructivists would argue that the apparent "hostility" of North Korean missiles shows that American leaders respond to the social relationship between the United States and the military resources of others, friend or enemy, rather than to the hardware itself. These social relations are not fixed, and the American national interest therefore cannot be ascertained, let alone pursued, without considering them. The United States has an interest in resisting North Korea, because American leaders perceive a hostile relationship with it, while it has no interest in containing the UK, because it perceives a mutually beneficial relationship. Constructivists often find it useful to examine the historical construction of "national interests" (e.g. Finnemore 1996; 2003; Weldes 1999).
It is sometimes said that the difference between constructivism and other approaches is that the former is concerned with the construction of interests while the latter take interests as fixed and given (see, e.g., Goldstein 2005, 126). This is not true. Nor is it true that only constructivists suggest that state interests might be influenced by forces at the level of the international system. Constructivists do not have a monopoly on the study of how interests are made or of systemic influences on interests. Many nonconstructivists are interested in how states come to
constructivism
303
hold the interests that structure their decision-making. Andrew Moravcsik (1999), for instance, provides a liberal theory of how state interests are constructed out of the economic interests of domestic industries and coalitions. Stephen Krasner (1999) argues from a realist perspective that individual rulers present as the national interest the policies they believe will ensure their personal survival as rulers. (Both present these as "material" factors though they rest on ideas about needs.) Game theorists sometimes endogenize the formation of interests so that interests change as a result of interactions (e.g. Gerber and Jackson 1993). On system-level influences, Jon Pevehouse (2005) uses broadly rationalist tools to examine how the constitution of states is affected by their membership and participation in regional organizations.
What distinguishes a specifically constructivist story on interests is that the influences on interest formation are social. Legro (2005, 4) represents the constructivist view: "new foreign policy ideas are shaped by preexisting dominant ideas and their relationship to experienced events."5 This follows directly from the insight on social construction above. Wendt (1992, 397) says "actors acquire identities—relatively stable, role-specific understandings and expectations about self—by participating in... collective meanings." Interests are in part products of those identities. The social constitution of interests encompasses all the ways that actors' interests and identities might be influenced by their interactions with others and with their social environment. This includes the processes of socialization and internalization (Hurd 1999), the drive for social recognition and prestige (Wendt 1999, ch. 5), the effects of social norms on interests and on behavior (including the desire to create norms that legitimize one's behavior) (Hurd 2007a), and the presence or absence of a sense of "community" (Adler and Barnett 1998).
1.3 Mutual Constitution of Structures and Agents
The constructivist attention to the social construction of interests and identities introduces the more general problem of the relationship between structures and agents. By "structures" I mean the institutions and shared meanings that make up the context of international action, and by "agents" I mean any entity that operates as an actor in that context. Returning to Wendt's illustration, the relationship of enmity that makes the United States fear North Korean nuclear weapons is not a fixed and stable fact. It is, instead, a result of ongoing interactions both between the two states and among the states and their social context. These interactions may reinforce the relation of enmity or they may change it. They may also reinforce or change the broader social structures in which the actors exist, including norms and other forms of shared meaning regarding sovereignty, threat, and interests.
5 Contrast this with realism, of which Moravcsik (1999,680, n. 6) says "the distribution of ideas and information is a function of the underlying distribution of material power resources."
304      ian hurd
The co-constitution of states and structures goes beyond recognizing that there are interaction effects between the unit and the system level. Kenneth Waltz emphasized interaction effects but in a way that maintained states as unchanging units. In Theory of International Politics, he suggested that two states interacting in anarchy are "not just influencing the other" by their actions; "both are being influenced by the situation their interaction creates" (Waltz 1979, 74). Consistent with his materialist premise, Waltz looked for how this changed the material incentives facing states as they weighed policy alternatives.
A constructivist approach to co-constitution, by contrast, suggests that the actions of states contribute to making the institutions and norms of international life, and these institutions and norms contribute to defining, socializing, and influencing states. Both the institutions and the actors can be redefined in the process. The recognition of mutual constitution is an important contribution to the theory of international relations, because many interesting empirical phenomena in international relations are understandable only by a methodology that avoids assuming a neat separation between agents and structures. In studying international norms, it quickly becomes clear that states are concerned simultaneously with shifting their behavior to match the rules and reconstructing the rules to condone their behavior (Hurd 2007a). For instance, when states claim they are using force only in self-defense, they cannot avoid reinforcing Articles 2(4) and 51 of the UN Charter (which forbid aggressive war) and at the same time are redefining the rules by specifying how they wish the concepts of "sovereignty," "self-defense," and "aggression" to be understood. International norms are simultaneously the products of state actions and influences upon state action. Thus, the idea that states and the international environment are mutually constituted is inherent in the constructivist approach.
1.4 Multiple Logics of Anarchy
The constructivist approach leads to a different interpretation of international anarchy from the one offered by neorealists or neoliberals, and, to the extent that the concept of anarchy organizes international life, it therefore leads to different understandings of world politics more generally.
"Anarchy" is the term used in international relations to describe a social system that lacks legitimated institutions of authority (Milner 1991). It is a formal condition of a system in the sense that it describes any system that is not organized through hierarchical structures of authority and command. Waltz (1979), in defining the neorealist school, derived from the structural condition of anarchy a set of predictions about the behavior of units, including balancing behavior, self-help strategies, and a self-interested identity. Wendt's critique of Waltz showed that these patterns did not follow simply from the structural condition of anarchy; they came from the additional assumption that units see each other as rivals over
constructivism 305
scarce goods. "Rivalry" is a social relationship that can best be understood, in international relations and elsewhere, by examining its social construction. This requires acknowledging that the relationship is not fixed, natural, or permanent. Wendt proposed a spectrum of international anarchies based on variation in the ideas that states have about themselves and others. With enmity at one end and friendship at the other, and with indifference in the middle, the formal condition of anarchy is by itself not very informative about the behavior of the units. After all, he says, "an anarchy of friends differs from one of enemies" (Wendt 1995, 78). This allows for the possibility of community (Adler and Barnett 1998; Cronin 1999), hierarchy (Simpson 2004), rivalry (Wendt 1992), and other social relations within a formally anarchic structure. Inter-state conflict is also conditioned by the social qualities of international anarchy, as illustrated by the efforts of states to appear to operate within the confines of the norms on war.6 Such diverse behaviors, and others, are compatible with the anarchical structure of the international system, and can be addressed through the constructivist approach. (I discuss below the constructivist possibility that the system is not anarchic.)
These four elements are the distinguishing features of constructivism in international relations theory. They are related to each other in the sense that, if one adopts the first idea (that is, that world politics is partly socially constructed), then the other three logically follow as implications for studying international relations. However, each of the other three is also consistent with nonconstructivist premises. For instance, one need not be a constructivist to study the origins of national interests, nor does finding that anarchy may differ across time and place necessarily mean that one is using a constructivist approach. This has helped to generate controversy over what is and is not constructivist research in international relations. The irreducible core of constructivism for international relations is the recognition that international reality is socially constructed. This has implications for the concept of anarchy, for the agent-structure relationship, and for national interests, but all three of these areas of research are also approachable through nonconstructivist means.
2 Controversies within Constructivism
In defining constructivism in this way, widely diverse research falls within its scope. This includes work with major differences on issues such as the unit of analysis, the possibility of positivist paths to knowledge, and the nature of the international
6 On shared norms that govern inter-state war, see Price and Tannenwald (1996); Price (1998); Sands (2005). On humanitarian intervention, see Welsh (2002).
306      ian hurd
system. In this section I highlight some of the controversies that arise over these issues and illustrate both the breadth of constructivist scholarship and the antagonisms it engenders among scholars.
2.1 State-centrism
The constructivist approach does not imply any particular unit of analysis as fundamental in the study of international relations. As a result, it is compatible with a kind of pluralism about the unit that has been both productive and contentious among international relations scholars.
The process of social construction cannot be understood by focusing exclusively on forces or actors at any of the three "levels of analysis" conventionally used in international relations theory (Waltz 1959; 1979). For any given puzzle in international relations, there are undoubtedly important elements of the answer to be found at all levels of analysis. In addition, one can examine how actors and structures at all levels of analysis are socially constructed. Constructivists have therefore provided interesting research on the constitution of individual state identity, on the making of meso-level norms and practices, and on the constitution of the international system (see, respectively, Lynch 1999; Shannon 2000; Reus-Smit 1999). The emphasis on forces or actors at one level over others may be defensible on pragmatic grounds given the interests of the particular scholar, but the co-constitution of actors and structures means there is no impetus in constructivism for a zero-sum debate over "which" level provides the most leverage over puzzles. There is no point in constructivist research to arguments over whether, for instance, domestic politics "matters" or not in international relations.
There is, however, room for debate over what can be taken as given by assumption at the start of a piece of empirical research. For instance, to take states as given in order to study how their interactions are structured by and contribute to a particular set of international norms implies setting aside the (prior) social construction of the state as an institution. This is potentially problematic, since the historical construction of states as sovereign may well be an important element of any story about how states interact with norms. The analytic separation of actors, practices, and structures as distinct entities can be problematic, though it may sometimes be useful. The dilemma of what to problematize and what to take as given is inherent in all research, and by focusing on the complexities of mutual constitution the constructivist approach encourages scholars to be open about what is lost by their particular choices and assumptions. This at least makes possible debate over the trade-offs implicit in these choices.
2.2 Science and Positivism
The recognition of social construction in world politics leads directly into a controversy over epistemology and the use of scientific methods in the field of
constructivism 307
international relations. This divides constructivism into a positivist and a postpos-itivist camp, distinguished by their positions on epistemological questions and the methods they believe are useful given those epistemological positions.
Positivist epistemology maintains that the socially constructed international system contains patterns that are amenable to generalization and to falsifiable hypotheses. These patterns are the product of underlying laws that govern social relations, where the laws can be identified by careful scientific research. While the methods that are appropriate to study world politics may not be those of laboratory science (for instance, controlled experiments with a strict separation between observer and event), the ultimate goal of the social scientific project is the same as for the physical sciences—explaining cause-and-effect relationships that are believed to exist independently of the observer's presence. Positivist constructivists are careful to include constitutive explanations among the cause-effect relations they seek to understand, but they approach the study of social constitution with the same tools of social science (e.g. Wendt 1999; 2000; Finnemore 2003; Barnett 2005).
A competing view, represented by postpositivists, is that in social life data are not fully objectifiable, observers cannot be fully autonomous of the subject under study, and social relationships cannot be separated into discrete "causes" and "effects."7 What social "laws" a scholar might observe are, therefore, inherently contingent rather than existing naturally and objectively in the world.8 As a result, according to David Campbell (2007, 209-10), social inquiry "has to be concerned with the social constitution of meaning, the linguistic construction of reality, and the historicity of knowledge. This reaffirms the indispensability of interpretation, and suggests that all knowledge involves a relationship with power in its mapping of the world." Claims to knowledge about world politics both reflect and act as structures of power, and there are no "Archimedean points from which to assess the validity of analytical and ethical knowledge claims" (Price and Reus-Smit 1998, 262). In this view, the purpose of theorizing is not to identify and test hypotheses about lawlike regularities. Instead, one objective for research is to interpret how social meaning and power produce the apparent stability in the social world (Devetak 2005,169).
The epistemological divide between positivists and postpositivists runs deep and may represent a decisive fissure among constructivists, and the matter is particularly sharp over the issue of ethics. (See Price, this volume.) For postpositivists, the ethical implications of international relations theory begin immediately once a scholar adopts or argues for an interpretative stance within which claims can be made. Without the positivist's faith in an independently existing reality of world politics, the postpositivist is attentive from the start about the ethical consequences of the concepts and assumptions that frame the research. The positivist, by contrast, works from the assumption that he or she is insulated by the belief that describing
7 I am grateful to Elizabeth Shakman Hurd for her comments on this section.
8 Richard Price and Christian Reus-Smit (1998) argue for a middle position of "contingent generalizations." On the capacity of international relations theory to constitute the international world, see Ashley (1986); Campbell (1998); Williams (2005).
308      ian hurd
objectively existing relations makes ethical issues a separate question. For the posi-tivist, the question of what "is" can be separated from what "ought."
The postpositivist position within constructivism is no less empirical (though not "empiricist" (Campbell 2007, 208-9)) than the positivist tradition. It is, however, empirical in a way that reflects the methods appropriate to its epistemology For instance, Campbell's study (1998,13) of the Bosnian wars examines how:
the settled norms of international society—in particular, the idea that the national community requires the nexus of demarcated territory and fixed identity—were not only insufficient to enable a response to the Bosnian war, they were complicit in and necessary for the conduct of the war itself. This is because inscribing the boundaries that make the installation of the nationalist imaginary possible requires the expulsion from the resultant "domestic" space of all that comes to be regarded as alien, foreign, and dangerous.
For Campbell, the Bosnian violence was exacerbated by outsiders' insistence that there exists an underlying "law" of ethnic intolerance that counsels that the ethnic groups of Bosnia must be physically separated from each other. A more ethical response is possible, he suggests, by critiquing the assumption that individuals have unitary ethnic identities that map cleanly onto unitary territorial nation states.
2.3 Anarchy or Authority?
Constructivists disagree among themselves on the nature of the international system. This is reflected in the debate over whether the system can be characterized as an "anarchy."
Most constructivists have operated within what Ashley (1988) called the "anarchy problematique," a position that they share with neoliberals and neorealists. This view acknowledges the existence of a formal condition of anarchy among states and makes anarchy a crucial element of the international structure. It sees hierarchy as the alternative to anarchy, where hierarchy refers to a system in which the units "stand vis-a-vis each other in relations of super- and subordination" (Waltz 1979, 81). On this level, constructivists often agree with the neorealists and neoliberals that anarchy is the fundamental organizing principle of the international system, even though they may disagree with their claims about the implications of that condition for state behavior (Cronin 1999, for instance, argues that there is "community under anarchy"). They argue that the social construction of cultural content within an anarchic system produces variation in the structural constraints and opportunities for units and therefore leads to variation in outcomes and in the patterns of state behavior. As a formal condition, anarchy remains.
However, constructivism also opens the possibility that changes in the social relations among states could transform the anarchical system into something that is not anarchic (Wendt 1999, 307-8). The key concept here is authority. Authority refers
constructivism 309
to a relation of legitimated power (Ruggie 1998, 65; Barnett and Finnemore 2004, ch. 1; Hurd 2007b). It creates a social hierarchy within which subordinates feel an obligation to follow the directives of the authoritative rule or actor. Authority and anarchy are therefore mutually exclusive. While some constructivists have remained within the anarchy problematique, others have found empirical evidence of the existence of institutions of legitimated power. International authority can be found in international organizations, in firms, and in practices such as international law. It exists in both public and private institutions. Public forms might include the UN Secretary-General (Barnett 2001), the UN Security Council (Hurd 2007&), the discourses of international law (Johnstone 2005), and norms on legitimate intervention (Finnemore 2003). Private forms include the legitimated power of firms and institutions in international markets (Sinclair 1999; Hall and Biersteker 2003). In settings where states recognize a rule, institution, or actor as having the right to make authoritative decisions on their behalf, we must recognize that authority rather than anarchy exists.
Just as the epistemological disagreement among constructivists over positivism may create a fundamental disjuncture between two world-views, the controversy over the existence of anarchy defines two camps. The presence or absence of authority divides constructivists between a "conventional" strand, which shares the anarchy problematique with neorealists and neoliberals, and a "post-anarchy" strand that rejects the anarchist view on empirical grounds. The disagreement is basically empirical—that is, it is over whether authority exists or not—and so it might be more amenable to resolution than is the epistemological divide that separates pos-itivists and postpositivists (see Hurd 2008 on the possibility of empirical "tests" of international authority). The conventional view allows that the content of anarchy might change (due to coordinating institutions, a shared culture, or other factors) but the basic structural condition of anarchy as the foundation of the international system does not. By contrast, the post-anarchy view is a fundamental challenge to the shared premise that anarchy is the continuing basis for international politics, and it has affinity with the English School, which has always been more attached to the image of an "international society" than international anarchy (see, e.g., Clark 2005).
3 Continuing Challenges in International Relations Theory
The rise of the constructivist approach has encouraged new strands of empirical and philosophical research in international relations, and has led to interesting
310      ian hurd
problems at the boundary between constructivism and other approaches. Two strands of research, on the relations between strategic behavior and international norms and between rationalism and constructivism, serve as examples of promising research in constructivist international relations theory.
3.1 Strategic Behavior and Norms
It is a mistake to characterize constructivism as focused on norms as opposed to neorealism and neoliberalism, which are alleged to be focused on power and interests.9 This is a common trope, and it is highly misleading. It obscures what is perhaps the most interesting and challenging puzzle in international relations theory— disentangling the relationship between strategic actors and social/normative influences. Most constructivists agree that states act in the pursuit of what they see as their interests, and all are as concerned with "power and interest" as are realists (and liberals). What differentiates these approaches are the sources that they identify for state interests, and the content of those interests. There is no reason that the study of international norms by constructivists is inherently mutually exclusive with the study of strategic behavior. The social construction of actors may well create instrumental, goal-seeking agents who pursue their goals in part by comparing costs and benefits, and their behavior cannot be understood apart from that process of construction. In other words, it is a mistake to separate the study of the logic of consequences from the logic of appropriateness (cf. March and Olsen 1998). The more strictly that separation is enforced, the less insightful is the empirical research that can result.
This conclusion is the logical consequence of my opening definitions, where I suggested that materialism, rather than rationalism, should be seen as the opposite of constructivism.10 Constructivism generally agrees with rationalism that states perceive some needs and interests and they act in order to satisfy them. To this, constructivism adds two things: an interest in explaining how state needs and interests come to be, and the possibility that different constructions of states could lead to radically different types of states and patterns of state behavior. Constructivism problematizes states and their interests and identities, but it has no problem accepting that states generally pursue "interests." It is with materialism that constructivism has the more fundamental disagreement—there is a clear distinction between the position that actors respond directly to material incentives and the view that meaning and interpretation necessarily mediate between material forces and
9 Mearsheimer (1995, 86) identifies "power and interest" as variables associated exclusively with realist theory, so that when others make reference to them he concludes that they have become realists. Fred Halliday (2005, 32-3) says that constructivist scholars "run the risk of ignoring interests and material factors, let alone old-fashioned deception and self-delusion."
10 Michael Barnett (2005), by contrast, sees rationalism as the opposite of constructivism.
constructivism 3"
social actors. Behavior is motivated, and is studied, only through lenses acquired in and through social interaction.
3.2 Constructivism and Rationalism
The relationship between strategic behavior and international norms raises more general questions about the relationship between constructivism and rationalism, and this theme has recently received a great deal of attention. At issue are questions including whether the two stand as competitors to each other or complements, the nature of the disagreement between them, and the useful scope of each.
The two approaches are often presented as competitors to each other. There are two versions of this claim. One suggests that rationalism and constructivism predict different behaviors from states and these differences should be measurable and testable. Jeffrey Lewis (2003) takes this approach to studying EU decision-making and he performs his "test" by assuming that strategic, instrumental behavior by states is evidence in favor of rationalism, while evidence of norm internalization supports constructivism. He treats the two as mutually exclusive and zero sum. The second version of the competitive relation argues that rationalism and constructivism are based on ontological commitments that are irreconcilable. These might be about holism or individualism, inherent or constructed rationality, or social construction versus essentialism. To the extent that these are fundamental commitments about what world politics is made of, they are unbridgeable.
There are also at least two versions of the claim that rationalism and constructivism are complementary to each other. One version sees the two as asking different questions about international relations and therefore as being fundamentally uninvolved with each other. This view suggests a division of labor in which constructivism is suited to answering questions about how actors acquire their interests and identities and rationalism specializes in explaining the pursuit of interests by already constituted actors. Sterling-Folker (2000, 97; cf. March and Olsen 1998), for instance, argues that rationalist institutionalism seeks to explain "short-term behavioral cooperation in the moment," while constructivism aims to explain "its development into communal cooperation in the future." In her view, the two cannot be competitors over the same turf because they are targeted at distinct questions. This approach presumes that the real world contains separable realms that are amenable to each approach and that the two realms do not overlap. Conflicts between the two are therefore avoidable as long as the boundary between the two realms is respected. A second version sees the two as providing different views on the shared questions. Duncan Snidal and Alexander Thompson (2002, 200), for instance, examine the ways in which international institutions constrain states and, finding both rationalism and constructivism useful, conclude that the two "provide different lenses through which to view the same empirical phenomena
312      ian hurd
and outcomes." On this view, the two are relevant to the same subject matter, but their different emphases allow, when combined, for greater insight into a problem than is provided by each alone.
The relationship between rationalism and constructivism is ultimately an artifact of one's definition of the two approaches. Defining either one requires also defining the other, and so the relation between them is epiphenomenal of these definitions. By categorizing constructivism as a research agenda concerned with the social construction of actors, structures, and practices in international relations, I presume from the start that there are some kinds of research that are inaccessible to rationalist methods and assumptions, and this automatically brings up aspects of the complementary view. I am therefore skeptical of the competitive versions of the constructivism-rationalism distinction. The competitive empirical tests proposed by Lewis are undermined by the fact that the behavioral distinctions between the two are extremely faint. My definition of constructivism does not support the view that strategic behavior by states is evidence for the rationalist view and against the constructivist view. As James Fearon and Wendt (2002) suggest, there may be no measurable variables in behavior that neatly differentiate the two approaches. A more substantive gap exists over the ontological questions regarding the nature of international actors and forces. A theory of ontology is unavoidable, not optional, and disagreements about ontology are real, profound, and consequential (Wight 2006). They might also, however, be best approached by setting against each other the research that follows from different ontological positions, rather than arguing for or against a theory of ontology in the abstract. It is the consequences (both ethical and analytical) of different ontological assumptions that are worth arguing about. Therefore, while there are indeed competing ontological positions between constructivists and others in international relations, the productive way forward would seem to be to assess the insights they generate when applied in research rather than compare them directly. This supports in practice the pluralism in research methods encouraged by the complementary views above, though it does not give up on the possibility that there are underlying differences in ontology between rationalism and constructivism.
4 Conclusion
To be a constructivist in international relations means looking at international relations with an eye open to the social construction of actors, institutions, and events. It means beginning from the assumption that how people and states think and behave in world politics is premised on their understanding of the world
constructivism
313
around them, which includes their own beliefs about the world, the identities they hold about themselves and others, and the shared understandings and practices in which they participate. It should be clear, therefore, what constructivism is not: It does not mean setting aside the ideas that material power is important or that actors make instrumental calculations of their interests; nor does it necessarily assume the a priori existence of sovereign states, epistemological positivism, or the anarchy problematique. Rather, it means that what goes on in these categories and concepts is constructed by social processes and interactions, and that their relevance for international relations is a function of the social construction of meaning.
One sign of constructivism's success in the past twenty years is the degree to which other approaches have come to recognize the socially constructed content of some of the concepts they use. The goods of realist competition, for instance, include status, prestige, reputation, and hegemony, all of which make sense only in terms of either legitimated power or shared understandings. They are, therefore, the stuff of constructivism as well. This has had the result of blurring the boundaries between the approaches, making them hard to define in exclusive terms, and raising the possibility that to attempt to define them creates artificial distinctions. The differences between realism, rationalism, and constructivism may be contested, but we move forward in arguing about them only by first being clear what we mean by the terms.
References
Adler, E., and Barnett, M. 1998. Security communities in theoretical perspective. Pp. 3-28 in Security Communities, ed. E. Adler and M. Barnett. Cambridge: Cambridge University Press.
Ashley, R. K. 1986. The poverty of neorealism. Pp. 255-300 in Neorealism and its Critics, ed.
R. O. Keohane. New York: Columbia University Press. -1988. Untying the sovereign state: a double reading of the anarchy problematique.
Millennium: Journal of International Studies, 17: 227-62. Baldwin, D. A. (ed.) 1993. Neorealism and Neoliberalism: The Contemporary Debate. New
York: Columbia University Press. Barkin, J. S. 2003. Realist constructivism. International Studies Review, 5: 325-42. Barnett, M. 2001. Authority, intervention, and the outer limits of international relations
theory. Pp. 47-68 in Intervention and Transnationalism in Africa: Global-Local Networks
of Power, ed. T. Callaghy, R. Kassimir, and R. Latham. Cambridge: Cambridge University
Press.
-2005. Social constructivism. Pp. 251-70 in The Globalization of World Politics: An
Introduction to International Relations, 3rd edn., ed. J. Baylis and S. Smith. Oxford: Oxford University Press.
-and FiNNEMORE, M. 2004. Rules for the World: International Organizations in Global
Politics. Ithaca, NY: Cornell University Press.
314
ian hurd
Biersteker, T. J., and Weber C. 1996. State Sovereignty as Social Construct. Cambridge:
Cambridge University Press. Brooks, S. G., and Wohlforth W. 2007. World out of Balance: International Relations
Theory and the Challenge of American Primacy. Princeton, NJ: Princeton University Press. Campbell, D. 1998. National Deconstruction: Violence, Identity, and Justice in Bosnia. Minneapolis: University of Minnesota Press. -2007. Poststructuralism. Pp. 203-28 in International Relations Theories: Discipline and
Diversity, ed. T. Dunne, M. Kurki, and S. Smith. Oxford: Oxford University Press. Clark, I. 2005. Legitimacy in International Society. Oxford: Oxford University Press. Cronin, B. 1999. Cooperation under Anarchy: Transnational Identity and the Evolution of
Cooperation. New York: Columbia University Press. Devetak, R. 2005. Postmodernism. Pp. 161-87 in Theories of International Relations, 3rd
edn., ed. S. Burchili et al. New York: Palgrave. Fearon, J., and Wendt, A. 2002. Rationalism v. constructivism: a skeptical view. Pp. 52-72
in Handbook of International Relations, ed. W. Carlsnaes, T. Risse, and B. A. Simmons.
London: Sage.
FiNNEMORE, M. 1996. National Interests in International Society. Ithaca, NY: Cornell University Press.
-2003. The Purpose of Intervention: Changing Beliefs about the Use of Force. Ithaca, NY:
Cornell University Press. Gerber, E. R., and Iackson, J. E. 1993. Endogenous preferences and the study of institutions.
American Political Science Review, 87: 639-56. Goldstein, J. S. 2005. International Relations, 6th edn. New York: Pearson Longman. -and Keohane, R. 0.1994. Ideas and foreign policy: an analytical framework. Pp. 3-30
in Ideas and Foreign Policy: Beliefs, Institutions, and Political Change, ed. J. Goldstein and
R. O. Keohane. Ithaca, NY: Cornell University Press. Hall, R. B., and Biersteker, T. J. (eds.) 2003. The Emergence of Private Authority in Global
Governance. Cambridge: Cambridge University Press. Halliday, F. 2005. The Middle East in International Relations: Power, Politics and Ideology.
Cambridge: Cambridge University Press. Hurd, I. 1999. Legitimacy and authority in international politics. International Organization, 53: 379-408.
-2007a. Breaking and making norms: American revisionism and crises of legitimacy.
International Politics, 44:194-213. -2007b. After Anarchy: Legitimacy and Power in the United Nations. Princeton, NJ:
Princeton University Press. —— 2008. Theories of international authority. In The UN Security Council and the Politics
of International Authority, ed. B. Cronin and I. Hurd. Johnstone, I. 2005. The power of interpretive communities. Pp. 185-204 in Power in Global
Governance, ed. M. Barnett and R. Duvall. Cambridge: Cambridge University Press. Krasner, S. 1999. Sovereignty: Organized Hypocrisy. Princeton, NJ: Princeton University
Press.
Kratochwil, F. V. 1989. Rules, Norms, and Decisions: On the Conditions of Practical and Legal Reasoning in International Relations and Domestic Affairs. Cambridge: Cambridge University Press.
Legro, J. 2005. Rethinking the World: Great Power Strategies and International Order. Ithaca, NY: Cornell University Press.
constructivism
315
Lewis, J. 2003. Institutional environments and everyday EU decision making: rationalist or
constructivist? Comparative Political Studies, 36: 97-124. Lynch, M. 1999. State Interests and Public Spheres: The International Politics of Jordan's
Identity. New York: Columbia University Press. March, J. G., and Olsen, J. P. 1998. The institutional dynamics of international political
orders. International Organization, 52: 943-69. Mearsheimer, J. J. 1995. A realist reply. International Security, 20: 82-93. -2001. The Tragedy of the Great Powers. New York: Norton.
Milner, H. 1991. The assumption of anarchy in international relations theory: a critique.
Review of International Studies, 17: 67-85. Moravcsik, A. 1999. "Is something rotten in the state of Denmark?" Constructivism and
European integration. Journal of European Public Policy, 6: 669-81. Onuf, N. G. 1989. World of our Making: Rules and Rule in Social Theory and International
Relations. Columbia: University of South Carolina Press. Palan, R. 2000. A world of their making: an evaluation of the constructivist critique in
international relations. Review of International Studies, 26: 575-98. Pevehouse, J. 2005. Democracy from Above: Regional Organizations and Democratization.
Cambridge: Cambridge University Press. Price, R. 1998. Reversing the gun sights: transnational civil society targets land mines.
International Organization, 52: 613-44. -and Reus-Smit, C. 1998. Dangerous liaisons? Critical international theory and constructivism. European Journal of International Relations, 4: 259-94. -and Tannenwald, N. 1996. Norms and deterrence: the nuclear and chemical weapons
taboos. Pp. 114-52 in The Culture of National Security: Norms and Identity in World Politics,
ed. P. Katzenstein. New York: Columbia University Press. Reus-Smit, C. 1999. The Moral Purpose of the State: Culture, Social Identity, and Institutional
Rationality in International Relations. Princeton, NJ: Princeton University Press. -C. 2005. Constructivism. Pp. 188-212 in Theories of International Relations, 3rd edn.,
ed. S. Burchili et al. New York: Palgrave. Ruggie, J. G. 1998. Constructing the World Polity: Essays on International Institutionalization.
New York: Routledge.
Sands, P. 2005. Lawless World: America and the Making and Breaking of Global Rules from FDR's Atlantic Charter to George W. Bush's Illegal War. New York: Viking.
Shannon, V. P. 2000. Norms are what states make of them: the political psychology of norm violation. International Studies Quarterly, 44: 293-316.
Simpson, G. 2004. Great Powers and Outlaw States: Unequal Sovereigns in the International Legal Order. Cambridge: Cambridge University Press.
Sinclair, T. 1999. Bond-rating agencies and coordination in the global political economy. Pp. 153-67 in Private Authority and International Affairs, ed. A. C. Cutler, V. Haufler, and T. Porter. Buffalo: State University of New York Press.
Snidal, D., and Thqmpson, A. 2002. International commitments and domestic politics: institutions and actors at two levels. Pp. 197-230 in Locating the Proper Authorities: The Interaction of Domestic and International Institutions, ed. D. Drezner. Ann Arbor: University of Michigan Press.
Sterling-Folker, J. 2000. Competing paradigms or birds of a feather? Constructivism and neoliberal institutionalism compared. International Studies Quarterly, 44: 97-119-
3l6      ian hurd
Waltz, K. N. 1959. Man, the State, and War: A Theoretical Analysis. New York: Columbia University Press.
-1979- Theory of International Politics. Boston: Addison-Wesley.
Weldes, J. 1999. Constructing National Interests: The United States and the Cuban Missile Crisis. Minneapolis: University of Minnesota Press.
Welsh, J. M. 2002. From right to responsibility: humanitarian intervention and international society. Global Governance, 8: 503-21.
Wendt, A. 1992. Anarchy is what states make of it: the social construction of power politics. International Organization, 46: 391-425.
-1995. Constructing international politics. International Security, 20: 71-81.
-1999. Social Theory of International Politics. Cambridge: Cambridge University Press.
-2000. On the Via Media: a response to the critics. Review of International Studies, 26:
165-80.
Wight, C. 2006. Agents, Structures and International Relations: Politics as Ontology.
Cambridge: Cambridge University Press. Williams, M. C. 2005. The Realist Tradition and the Limits of International Relations.
Cambridge: Cambridge University Press.