Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 26 March 2017

Role Theory as an Empirical Theory of International Relations: From Metaphor to Formal Model

Summary and Keywords

Role theory as an empirical theory of international relations has an underlying logical structure with the ability to generate different models of cooperation and conflict in world politics at multiple levels of analysis: system-oriented models of incentives and role constraints; actor-centered models of role conceptions and expectations; action-focused models of cues and role enactment. The emphasis at each of these levels of analysis is on strategic interaction, which makes role theory a theory of international relations between ego and alter as well as a theory of their respective foreign policy decisions.

The logical and empirical applications of role theory’s models to world politics have morphed from metaphor and analogy into formal models of prediction and explanation that meet the criteria of testability associated with an empirical theory of international relations. These criteria include the logical rules of deductive inference and the correspondence rules of empirical falsifiability associated with the systematic comparison of empirical cases.

The pattern of migration and evolution of role resembles the earlier pattern of importing game as a metaphor and introducing the logical structure of game theory into the field of international relations. Binary role theory employs the concepts of role theory and a set of game theory models to analyze conflict and cooperation in world politics. The role metaphor and the concepts of binary role theory provide a substantive “theory of payoffs” for game theory. The latter’s formal models help transform the logical structure of role theory from a metaphor or analogy to a logically coherent and empirically testable theory of international relations.

Keywords: binary role theory, national role conceptions, international roles, role transition, empirical international relations theory


The concept of role has traveled from the theater to the social sciences and then into the field of international relations initially as a metaphor and then as an analogy for understanding world politics (Sarbin & Allen, 1968; Harnisch, Frank, & Maull, 2011). The assumption governing this analytical move is that understanding the relations between states is enhanced by recognizing their respective roles in a system of states. Within the context of a social system or domain of human behavior, “roles are repertoires of behavior, inferred from others expectations and one’s own conceptions, selected at least partly in response to cues and demands” (Walker, 1992, p. 23). Roles define parts for actors in a social system just as they define parts for actors in a play.

This metaphorical usage of role provides a language and a perspective for describing cooperation and conflict between states as a social system. Its application as an analogy for describing international relations takes the form of drawing inferences about properties of roles in one context from similar properties of roles in another context. For example, the properties of friend and enemy roles may be similar (analogous) in the domains of interpersonal and international relations. In both domains cooperation behavior may define the enactment of the role of friend while conflict behavior may define the enactment of the enemy role.

Role theory has accompanied the migration of its core concept across the social sciences into the field of international relations. It consists of a set of concepts that are logically inter-related to specify propositions that may explain as well as describe international relations. Role theory’s propositions should ideally explain: (1) the origins and variations in roles over time for different actors, (2) the differences in roles between actors, (3) the dynamics of interactions between and among actors with different roles, (4) the consequences of changes in roles for relations among actors. The word actors in these four predicates is a placeholder for the constituent units of the role system of interest, for example, the individuals in an interpersonal system or the states in an international system. As an empirical theory these propositions should also be testable, that is, the models that they specify should be susceptible to narrative or statistical tests in the form of observations that support, qualify, or refute hypotheses derived from them.

Snidal (1985, p. 36) points out, “Metaphor, analogy, model, and theory are complementary in social scientific research; they are each appropriate at various stages. It is often useful to go back and forth among them. As research advances, however, metaphor and analogy are of increasingly limited usefulness. The greater rigor and deductive power of the model, together with the interpretive richness and open-endedness of its corresponding theoretical framework, make that combination ultimately more productive.” He argues that metaphors and analogies rely primarily on the weaker logic of induction for description while theories and models employ the stronger logic of deduction for explanation. The former’s inductive logic requires a relatively loose fit between concepts and observations while the latter’s deductive logic imposes a tighter correspondence between a model’s internal relations and empirical observations. An analogy’s abstract properties may not hold for some observations without rejecting the analogy. However, any observations at odds with a model normally require its rejection or revision (Snidal, 1985, pp. 32–34).

The deductive logic of a theory and its richer interpretive structures may also contain multiple models compared to an analogy or metaphor. “The theory contains a deductive structure plus an interpretation of fundamental assumptions and theoretical constructs. This richer interpretive structure (as compared to the tighter correspondences in the model) provides for greater richness of explanation. Through it, the theory maintains a greater open-endedness and a surplus meaning which guides revision and extension of the model” (Snidal, 1985, pp. 34–35). Therefore, assessing role theory as an empirical theory of international relations involves identifying its internal deductive logic and the degree of fit between its models and empirical observations of world politics.

Previous Assessments of Role Theory

Previous reviews of role theory by scholars of international relations have characterized role theory as conceptually rich but methodologically poor (Walker, 1979, 1987; Thies, 2010; Breuning, 2011). The first wave of its conceptual applications to international relations between 1970 and 1990 rise to the level of metaphor and analogy but (with a few exceptions) fall short of meeting the level of specifying rigorous models derived from a coherent set of analytical propositions (Thies, 2010). At the same time reviews of this early research have suggested that role theory is a promising candidate for enriching the understanding of world politics and for offering a solution to the level-of-analysis problem and the agent-structure problem in the field of international relations. The substantive basis for these claims is role theory’s application to agent-centered problems of foreign policy analysis by North American scholars and to structure-oriented problems of managing international relations by European scholars (Thies, 2010; Breuning, 2011).

North American scholarship has focused primarily on the national role conceptions of leaders as cognitive links with foreign policy decisions while European efforts have focused as well on the role demands of European Union institutions as structural constraints on interactions among EU members. Role conceptions and role demands are concepts used to analyze the actual behaviors of states and their interactions as role enactments (Walker, 1987; see also Breuning, 2011; Harnisch, 2011). The three concepts offer a language for describing the behaviors as instances of the enactment of role conceptions and the adaptation to role demands. The question is whether these correlations with role enactments are coincidental or truly causal.

A sophisticated answer to this question assumes that the concepts are ideally employed on behalf of a sophisticated analytical framework (Mearsheimer & Walt, 2013). This point is the focus of Snidal’s (1985) emphasis on making the conceptual integrity of models derived from a rich theory as rigorous as possible. This criterion in practice means that the relationships among the variables in the model are logically exhaustive and mutually exclusive, that is, they cover all of the possible relationships and do not contradict one another. The theory also has to be effective, that is, relevant to the phenomena under analysis and modeled specifically enough to generate research hypotheses testable empirically with available instruments and the observations generated by them (Randall, 2012, p. 16; see also Feynman, 1995).

While metaphorical and analogical applications may act as heuristics, the hypotheses need to fit into a broader set of propositions linking the concepts in a theory. Theoretical progress in modeling is required as well as empirical progress in measurement, in order to move from a rough approximation in modeling and measuring an empirical domain to an increasingly precise correspondence between empirical observations and the logical relationships modeled within a theory. As Black (1962, p. 242) puts it: “Perhaps every science must start with metaphor and end with algebra; and perhaps without the metaphor there would never have been any algebra.” Scholars of international relations theory do not all agree with Black’s characterization of science nor the corresponding notion of theory that it implies for the study of world politics (Dunne, Hansen, & Wight, 2013). However, Snidal (1985, p. 34, no. 7) subscribes to it with the caveat that “the word ‘model’ should be substituted for ‘algebra,’” in order to emphasize that mathematical statements are not necessarily superior to verbal statements of a model.

Role Theory as an Empirical Theory

In this assessment of role theory the focus will be upon attempts by international relations scholars to employ it as an empirical theory rather than as a metaphor or analogy. These examples emphasize the construction of formal models of state behavior and interactions based on concepts and propositions from role theory. Their empirical research designs illustrate the features identified previously as helpful in moving toward more precision in deriving multiple models from role theory and measuring the fit between them and empirical observations. They are also relevant examples of the kind of studies that previous assessments of role theory and international relations have recommended as necessary or at least desirable (Thies, 2010; Breuning, 2011).

In general, the task in these studies is to extend the reach of role theory to solve two basic analytical problems: the level-of-analysis problem and the agent-structure problem. These two problems address aspects of a larger task, namely, the integration of foreign policy theory and international relations theory within the context of a general theory of human behavior and social relations for the domain of world politics (Walker, Malici, & Schafer, 2011). The level-of-analysis problem is the problem of making and integrating generalizations about relations between variables across different levels or units of analysis, ranging from the individual decision maker through the domestic society to the international system as levels of analysis (Singer, 1961; Waltz, 1959, 1979; Elman, 1996). The agent-structure problem is the problem of understanding and generalizing about constitutive and causal relations between parts and wholes as levels of analysis (Wendt, 1987; Archer, 1988; Carlsnaes, 1992).

How is role theory a candidate for a potential theoretical solution to these problems in the domain of international relations? Can a single theory integrate and explain processes occurring between states as well as inside the individuals, groups, organizations, and societies that constitute states? Which level of analysis has ontological or causal primacy: are they autonomous from one another or constituted so that one cannot exist without the other? Do agents generate structures or vice versa? Proponents of role theory make the argument on behalf of its answers to these questions by pointing to the theory’s descriptive, organizational, and explanatory capacities for understanding world politics (Thies, 2010; Breuning, 2011; Walker, 1987, 2013). They maintain that:

Descriptively, the concepts associated with role theory analysis provide a vocabulary of images which can focus upon foreign policy behavior at the national level of analysis, shift down to the individual level of analysis, and also move up to the systemic level of analysis; … they also take on multidimensional scope in their application to foreign policy behavior. … Organizationally … the concepts associated with role analysis permit the analyst to adopt either a structure-oriented or a process-oriented perspective. … [T]he explanatory value of role analysis appears to depend on whether its concepts are theoretically informed (a) by an appropriate set of self-contained propositions and methods, or (b) by the specification of an appropriate set of auxiliary limiting conditions and rules linking these conditions with role concepts

(Walker, 1987, pp. 2–3).

There are three general kinds of such role theory models, differentiated by their respective concepts that focus as sources of roles on three different levels of analysis: role demand models, role conception models, and role enactment models.

Role demand models are structure-oriented on external system variables. Role conception and expectation models are agent-centered on the interaction of internal cognitive variables. Role enactment models are action-focused on behavioral interaction variables. The simplest versions of all three kinds of models offer explanations for the sources of role enactment patterns by and between ego and alter as two actors that constitute a role set as a dyad As the scope of application increases from one dyad to triads and larger ensembles that constitute more complex social systems, other key concepts from role theory are employed to model their relations, for example, role strain, role competition, and role conflict (Sarbin & Allen, 1968; Holsti, 1970; Walker, 1979, 1992; Walker, Malici, & Schafer, 2011).

These concepts refer respectively to the level of effort (role strain), the level of attention and resources (role competition), and the degree of contradiction (role conflict) experienced by ego to enact multiple roles toward multiple alters (and vice versa) both simultaneously and over time (Walker, 1987, 1992; see also Holsti, 1970; McCall & Simmons, 1978; Biddle, 1979; Stryker & Serpe, 1982; Stryker & Statham, 1985). They also make role theory’s central mechanism of role location more explicitly dynamic by accounting for role transition between ego and alter who engage in both role-making and role-taking as they exchange cues or receive them from the audience of other actors in a social system (Thies, 2010; Breuning, 2011; Harnisch, 2011).

Role location refers to the process of exchanging cues (verbal or physical signals) between members of a role set that are associated with different roles and communicate what roles each member of the dyad or larger ensemble of actors expects and attributes to one another. If these cues do not assign identical or complementary roles to ego and alter, then role conflict exists between or among the members of a role set and is often accompanied by role competition and role strain. Sarbin and Allen (1968) identify four strategies to manage these tensions: ego can shift between roles over time or toward different alters (alternation strategy), enact roles simultaneously (interpenetration strategy), enact them so that they are indistinguishable from one another (merger strategy), or adopt another role (altercasting strategy). Role transition occurs when the number of roles enacted by ego or alter expands (role differentiation) or one pair of roles between ego and alter replaces another pair (role evolution) (Walker, 1992, pp. 31–34; Burr, Leigh, Randall, Day, & Constantine, 1979, pp. 80–88; see also Malici & Walker, 2016).

Collectively, this inventory of role concepts and their logical relationships should ideally offer coherent insights and answers to fundamental questions that students of international relations share. In order to do so, the answers to those questions need to be isomorphic and commensurate with role theory (Walker, Malici, & Schafer, 2011, pp. 245–266; Walker, 2013, pp. 27–44). The internal deductive logic and empirical correspondence rules of these models must also fit the substantive phenomena that they address (Snidal, 1985, pp. 32–26). Therefore, role theory’s structure-oriented, agent-centered, and action-focused models need to be able to represent the substantive core of world politics.

The degree to which these criteria are met distinguishes the application of role theory as a model from its application as a metaphor or analogy. Snidal (1985, p. 30) makes the point this way:

Increased rigor of specification is not to be confused with expression in a different form—specifically with mathematical as opposed to purely verbal statement. … Real rigor requires tightening the correspondence between the metaphor and the issue at hand. Analogies and models are to some extent simply more controlled metaphors, although they are further distinguished by their respective logics of inference.

The substantive core of world politics is defined here as the rational exercise of power among the various actors in the domain of world politics (Walker, Malici, & Schafer, 2011, pp. 10–19). The key concepts in this definition are rationality and power, in which rationality generally denotes “behavior that is appropriate to specified goals in the context of a given situation” (Simon, 1985, p. 294) and power refers to the exercise of positive and negative sanctions by one actor to control the actions of another actor in a social system (Baldwin, 1989). The rational exercise of power by actors toward one another generates patterns of strategic interaction between them, in which each one’s use of sanctions as an agent or initiator of sanctions is goal-oriented and cognizant of the other one’s expectations as a patient or recipient of sanctions (Baldwin, 1989, 2002).

At the core of politics are both social and psychological phenomena in the form of strategic interactions (behavioral sanctions) and cognitions (rational expectations) that collectively constitute a system of politics. The exercise of power is potentially rational in a political system to the extent that sanctions and cognitions are shared information among members of a dyad as the system of interest. Role theory is a social-psychological theory with a substantive core that focuses on the phenomena of the communication of information between two actors via behaviors (cues) that signal the expectations (roles) each actor has regarding the other in their interaction as a social system. This focus on communication in the form of role location offers a theoretical perspective that addresses how and why power is exercised among the members of a political system (Walker, 2013; Deutsch, 1966, 1968). Therefore, models of role enactment derived from the concepts of role theory that specify the exercise of power between states should be able to answer questions about the substantive core of world politics of interest to students of international relations (Walker, Malici, & Schafer, 2011, pp. 245–266).

First Wave: Modeling Foreign Policy Decisions

The first wave of applications of role theory to international relations tended to focus simply on one or another of the concepts and processes linking ego, alter, and audience at the top of Figure 1 (Breuning, 2011; see also Thies, 2010; Holsti, 1970). There was little attempt to assemble and integrate them into the more complex model of foreign policy and world politics depicted at the bottom of this figure. Also absent was an explicit attempt to model the processes of role contestation and role location that connected the concepts in role theory or focus on the interactions that lead to processes of role strain, role competition, role conflict, and role transition implicit in the role location process.

Role Theory as an Empirical Theory of International Relations: From Metaphor to Formal ModelClick to view larger

Figure 1. Simple and complex visualizations of role theory concepts and processes.*

Notes: *Concepts: RC =Role Conceptions; RX = Role Expectations; RD = Role Demands; RE = Role Enactment.

Processes: Role Contestation; Structural Adaptation; Role Contestation; Role Location.

These “blind spots” (Breuning, 2011) or omissions were partly due to the initial attraction of national role conception as a concept in the subfield of foreign policy analysis at the expense of other role theory concepts, which led to a focus on national role conceptions and their influence on foreign policy decisions and actions at the expense of applying other concepts in role theory to the social interactions between states (Breuning, 2011; see also Holsti, 1970). With a few exceptions (e.g., Walker, 1979) this application made the monad (ego) the unit of analysis rather than the ego–alter dyad. The dominant choice reflected Holsti’s (1970) assumption that the absence of government and the presence of anarchy reduced the external influence of others in the enactment of national roles. While a state’s power position (status) in the regional and global international systems was recognized to influence the capacity of a state to enact different roles and multiple roles, the domestic influences of societal cultures and national leaders were assumed to be more potent sources of the variety of national roles across states (Holsti, 1970; Wish, 1987; see also Thies, 2010; Breuning, 2011).

A focus on role demands at both levels of analysis (either domestic or external) tended to leave alter (the actions and expectations of others) out of the modeling equation except as part of the occasion for enacting a role. The assumption seemed to be that role conceptions either swept them aside or otherwise made them irrelevant. The result was a relative lack of attention to the processes of role conflict, role competition, or role strain and the strategies for resolving or managing them and, therefore, an incomplete understanding of the process of role location. The myopic effects of this focus also included problems in linking more precisely role conception and role enactment. While Holsti and others were able to identify role conceptions in the rhetoric of national leaders with both qualitative and quantitative content analysis methods, the link with specific patterns of foreign policy behaviors was more tenuous. Historical narratives provided thick descriptions of foreign policy actions while quantitative measurements of them were relatively scarce and their results were somewhat inconclusive. These results led to the impression that role theory was conceptually rich but methodologically weak (Walker, 1979, 1987; Thies, 2010; Breuning, 2011).

More important, the focus on role conceptions at the expense of other role concepts limited the migration of role theory as a whole into foreign policy analysis as a subfield and hindered its expansion into the field of international relations. Foreign policy analysis as a subfield focuses on the actions of states while international relations focuses on the interactions of states. To make the transition and bridge these two levels of analysis, a focus on dyads as the simplest kind of international system appears to be desirable (Walker, Malici, & Schafer, 2011; see also Richardson, 1987). This step increases the potential number of causal arrows in Figure 1 that are open to theorizing and available for systematic empirical investigation.

In addition to the processes shown in the more complex model at the bottom of this figure, it is possible to make the model more dynamic with two-way interactions and dynamic changes over time between the concepts. Including additional diagonal arrows as well as existing horizontal and vertical arrows linking the concepts in both directions are also possible, which would identify processes of role-making and role-taking that link behavioral and cognitive concepts across ego and alter (Walker, Malici, & Schafer, 2011, pp. 248–257). This shift away from a monadic toward a dyadic focus as a unit of analysis is one characteristic (along with a focus on processes and more sophisticated empirical research designs) that the trajectory of the current second wave of role theory studies since 1990 in world politics manifests (Harnisch, Frank, & Maull, 2011; Thies, 2010; Breuning, 2011; Bengtsson & Elgstrom, 2012; Cantir & Kaarbo, 2012; Harnisch, 2012; Shih, 2012; Thies, 2012; Thies & Breuning, 2012; Walker, 2013; Harnisch, Berstick, & Gottwald, 2016; Cantir & Kaarbo, 2016; Malici & Walker, 2016).

This transition actually began during the first wave within the subfield of foreign policy analysis with the conceptual and empirical applications of role theory in the Comparative Research on the Events of Nations (CREON) Project (M. Hermann, 1987; C. Hermann, 1987; Singer & Hudson, 1987; see also Hermann, Kegley, & Rosenau, 1987). The CREON scholars foreshadowed a move in the direction of dyadic analysis and the potential engagement of role theory more directly with the field of international relations. The CREON applications of role theory were characterized by an explicit dual focus on agent and situation and a distinction between role contestation processes over internal decisions and structural adaptation processes regarding external situations. These scholars also employed explicit formal models and quantitative indices to represent decision-making processes and properties of foreign policy actions. These features made it relatively easy to identify the deductive logic of their models and the fit between these models and empirical observations.

Role Conceptions and Role Contestation

Margaret Hermann (1987) modeled the relationship between foreign policy role orientations and the quality of foreign policy decisions. The analysis addressed the personality traits of individual leaders as one source of a state’s role conceptions. She identified six personal characteristics of leaders as bases for the foreign policy role orientations of their states. They included the traits of nationalism, belief in the ability to control events, need for power, need for affiliation, conceptual complexity, and distrust of others. She measured these personal characteristics with a quantitative content analysis of interviews given by leaders and combined them into a typology of expansionist, active independent, influential, mediator–integrator, opportunist, and developmental role orientations based on each leader’s combination of high or low scores on those traits (M. Hermann, 1987, pp. 126–127).

Hermann then modeled the decision-making processes of decision units with different power distributions among the members of the decision unit and leaders with different role orientations. This effort led to a series of proposals about the linkages between a decision unit’s way of coping with information uncertainty and the role orientation of a predominant leader. In effect, she was modeling the process of role contestation in Figure 1 linking the dispositions of leaders and characteristics of the decision-making process resolving disagreement or conflict over the leader’s preferred role orientation (M. Hermann, 1987, pp. 134–135). This effort offered answers to the question of the origins of national roles by reference to the role orientations of predominant leaders and their interactions with advisers configured into different power distributions as a decision-making unit. She and other CREON associates went on to analyze empirically the leader-adviser dynamics in a variety of decision units (M. Hermann & C. Hermann, 1989; Preston, 2001; M. Hermann, 2001; Kaarbo, 2012).

Role Demands and Structural Adaptation

At the same time Charles Hermann (1987) developed models of linkages between a leader’s role conceptions, different types of situations, and the resulting foreign policy associated with enacting a role. Hermann’s main interest was to model how a government chooses among multiple foreign policy role conceptions, depending on the situation that is the occasion for decision. Whereas core regime beliefs and role conceptions are the internal predispositions for role enactment, different features of the situation are external predispositions for role enactment:

The CREON project assumes that foreign policy behavior results only after a nation’s authoritative decision makers have perceived a problem. As defined by the decision makers, every problem has a source (who caused the problem) and a subject (who is deprived by the problem). In addition, some problems have actual or potential facilitators and aggravators. Source, subject, facilitator, and aggravator are systemic roles in the problem. … Any international entities may be perceived by the actor as occupying one of these roles. The decision rules about core beliefs depend in part upon which nations are occupying these roles for a specific problem.

(C. Hermann, 1987, p. 225).

These systemic roles are occupied by different alters without explicit regard for their agency (behavior) and corresponding role conceptions and core regime beliefs. So this CREON model treated external actors primarily as sources of role demands, that is, situational features that constrain the selection and enactment of ego’s roles, rather than as alters whose actions are cues that may generate role conflict, role strain, or role competition processes (see C. Hermann, 1987, pp. 224–227).

Collectively, the system roles of source, subject, facilitator, and aggravator assigned to other international entities instead defined the situation as a problem and an occasion for ego to make and implement a foreign policy decision. The CREON research program identified five such situations—confrontation, intervention, assistance needed, assistance resource, and collaboration—defined by the properties of alters in a situation: (1) Do they pose a threat to ego’s basic values? (2) Are they stronger or weaker than ego? (3) Do they display positive or negative affect toward ego? (4) Are they salient to ego? The answers to these questions depend on ego’s core political beliefs that match a role conception and a corresponding pattern of role enactment, which is judged appropriate for the situation as specified or inferred by the answers to these questions and the problem (the basic values) at stake for ego (C. Hermann, 1987, pp. 227–235).

The CREON conceptualization of role enactment was “an action of verbal or physical communication designed as an attempt to influence others” and is explained as “the most likely response of a national government to a problem, involving external entities, that is recognized by a regime” (C. Hermann, 1987, p. 235). Its properties co-vary with the role conceptions of ego and the role demands represented by the system roles of source, subject, facilitator, and aggravator that define the situation prompting a response by ego. Such an act of communication is a foreign policy event whose properties “frequently have been posed as who does what to whom, when, and how?” (C. Hermann, 1987, p. 236; Callahan, Brady, & Hermann, 1982).

The CREON project formally modeled as tree diagrams the combination of role conceptions, basic values, situations, and probable behavior properties for roles associated with selected core political system beliefs in the CREON regime orientations for the Cold War period (C. Hermann, 1987, pp. 227–235 and Appendix 4: 282–289). Singer and Hudson (1987) tested the predictions of the CREON external predispositions model with a sample of 622 foreign policy events attributed to six African nations (Ivory Coast, Guinea, Ghana, Uganda, Kenya, and Zambia) for the 1959–1968 decade. They operationalized and coded the context of each event for prior affect, relative capabilities, salience, and presence of facilitators or aggravators as a confrontation, intervention, assistance needed, assistance resource, or collaboration situation plus each event’s behavioral properties of affect, commitment, recipient, and instrument.

This CREON study reported rates of postdictive success versus the probability of success by chance for each property of an African state’s role enactment behavior as follows: affect (51.3% v. 33%), commitment (20.9% v. 25%), instrumentality (67.4% v. 33%), and recipient (60.9% v. 10%). The percentage and number of correct predictions per event for N = 622 events varied as follows: zero correct (6.4%), one correct (28.0%), two correct (32.3%), three correct (25.2%), and four correct (4.0%). The range of predictive success (defined as three correct out of four properties) for the six African nations ranged between 25.0% for Kenya and 49.2% for Zambia with the success rate for the remaining three countries hovering around 33% (Singer & Hudson, 1987, pp. 212–217).

Overall, the authors judged the external predisposition model’s performance as promising with scores of 33% for getting three or more out of four predictions right and 65% for getting two or more out of four right, which are well above the random success rate of 5.6% for getting three out of four properties correct: “We view these results as promising since the model only examines external variables. This can also be regarded as evidence that these external variables do play an important role in the shaping of African foreign policy behavior” (Singer & Hudson, 1987, p. 214).

Unfortunately, the CREON project did not endure long enough to combine the internal predisposition model of role orientations based on personality traits and decision-making units with the external predisposition model of situational variables. This task would probably have required an explicit shift from the monad (ego) to the dyad (ego–alter) as the spatial unit of analysis. CREON’s panel design for collecting foreign policy events would have likely needed to expand from events randomly collected by year to a time series design either for shorter time periods on a daily basis or aggregated to weeks, months, quarters, and years, depending on the frequency of observations for the role conceptions and role demands of both ego and alter. The technology was not readily available in the 1980s for undertaking this effort. It did not become available until the advent of desktop computers and digital sources of events and the public statements of leaders that followed.

With a few exceptions in North America, (Walker, 1992; Chavetz, Abramson, & Grillot, 1996; Le Prestre, 1997) the focus of attention on role theory shifted in the 1990s geographically to Europe along with an application of role theory within the context provided by the institutions of the European Union. This research did not typically employ formal modeling or quantitative measurement techniques and shifted from a psychological focus on role conceptions to a sociological focus on institutional role expectations and role demands. There was some movement away from the monad as a unit of analysis, but the shift often leaped over the dyad into an organizational context of many actors interacting to generate common social, economic, and security policies for members of the European community with varying degrees of success (Elgström & Smith, 2006; Aggestam, 2006; Bengtsson & Elgström, 2011).

These developments are well-covered in the volume edited by Harnisch, Frank, and Maull (2011), who reach some of the same conclusions regarding priorities for extending role theory as a theory of international relations that were reached by commentators located in North America (Rosenau, 1987; Thies, 2010; Breuning, 2011). The main salutary effect of the shift from North America to Europe was a deepening of role theory’s scope in two directions. On the one hand, there was increased attention paid to domestic context as the source of roles and influences such as public opinion on the process of role selection within states. On the other hand, there was a more explicit recognition that role enactment was a process of social interaction and not just a psychological process of reaction to stimuli and situational constraints as role demands from the environment. These developments represented a shift away from the internal processes of decision-making and the influence of psychological variables on decision-making.

Finally, a focus on the process of social interaction implied an extension of the temporal dimension of role theory to focus on the sequence of exchanges between ego and alter that make role theory dynamic. Rosenau (1987, p. 81) called these sequences “role scenarios” or “action scripts,” which couple ego and alter together in their respective enactments of role and counter-role. This move emphasizes the characteristic features of interaction and interdependence that make role theory a systems theory. It thereby becomes a theory of international relations as well as a theory of foreign policy analysis by spanning the micro-macro divide between parts and wholes in two respects: it bridges the spatial distinction between individual states and the system of states and also the temporal distinction between actions and reactions with the concept of role scenarios as episodes involving the emergent patterns of interactions between and among states over extended time periods.

These patterns involve a social and political logic in the form of a tit-for-tat pattern of reciprocity or complementarity as the exercise of social power to establish, maintain, or transform a pattern of order between ego and alter. Rosenau (1987, pp. 64–65) conjectured in 1987 that with “the advent of a fifth generation of computers … it should be possible to extend scenario analysis well into time n. … Just as the computer can now play full and imaginative games of chess, taking into account a multiplicity of possible moves and situations that might sustain the game to time n, so might the potentials of complex international situations be played out across time.”

Rosenau (1987, p. 65) concluded that this vision of the future of role theory’s application to the study of international relations would require the specification of formal models of social interaction consistent with the distinguishing features of both role theory and the substantive core of world politics: “To be sure, chess is played by a much simpler set of rules than is world politics, and to this extent the analogy breaks down. But we are not lacking in knowledge of the foreign policy rules and surely the utility of this knowledge can be enhanced by the technology now available.” Was Rosenau’s (1987, 1990) vision of role theory’s promise as a formal model of foreign policy and international relations prescient?

Second Wave: Modeling International Relations

Recent theoretical and empirical developments in the second wave of role theory’s application to international relations do appear to be consistent with Rosenau’s vision. The publication of the 2011 edited volume Role Theory and International Relations, edited by Sebastian Harnisch, Cornelia Frank, and Hanns W. Maull, reflects a vigorous expansion in the use of role theory by European scholars into the 21st century that extends beyond the EU institutions. A special issue of Foreign Policy Analysis, edited by Thies and Breuning (2012), offers evidence of a similar renewed interest in role theory by North American scholars. The second wave of role theory studies since 1990 on both sides of the Atlantic has focused increasingly on interaction patterns between role dyads over extended time periods. These studies address such problems as changing U.S. relations with allies in an era of American hegemony (Harnisch et al., 2011), the rise of China and its changing relations with states in the Asia-Pacific region (Harnisch, Berstick, & Gottwald, 2016), fluctuating relations between other regional hegemons such as Brazil and its neighbors in Latin America (Wehner, 2015), and the evolution of U.S.–Iran relations in the Middle East (Malici & Walker, 2016).

Two recent studies illustrate both the spatial and temporal scope and the modeling in Rosenau’s vision for analyzing role scenarios as formal models in world politics. They analyze patterns between role dyads over time, which is a dynamic extension of role theory’s core process of role location. Together they encompass two processes of role transition. One process is a long-term process in which a state experiences role transition in the form of socialization from the status position of one master role into another one within the structure of an international system, for example, moving from small-power to great-power status in the regional or global system (Thies, 2013). The other is a relatively short-term process of transition between auxiliary roles, for example, moving from cooperation to conflict roles as issues develop and power is exercised between members of a role dyad within the context of their respective master roles in the regional or global system of interest (Walker, 2013).

The two studies share a common reliance on the formal logic of game theory as a method for specifying strategies of role evolution and role transition that characterize the dynamics of role theory’s core process of role location over time. This feature distinguishes them from other dynamic studies in the second wave, which employ narrative methods without the logic of formal models to trace role transition processes. Both authors also use analytical historical narratives as empirical tests of their formal models. One relies upon a rich qualitative presentation of the narrative details (Thies, 2013). The other extracts properties from words and deeds embedded in the narrative details, in order to assess the fit between models and empirical observations (Walker, 2013).

The socialization model of role transition stipulates that if granted diplomatic recognition by other states, all states have the master role of sovereign state as a baseline status in the international system of interest. Otherwise, they are relegated to the status of quasi-states, rogues, or joiners who are not full-fledged members of the club of nations (Thies, 2013, pp. 22–23). Four master roles for states (novice, small power, major power, and great power) refer to the different positions in the hierarchical power structure of a sovereign state system. The transition from one master role to another one is a structure-oriented change within a system of states in which their respective positions are defined by their relative power capabilities. Novices are emerging states with low or uncertain capabilities, which transition over time into small, major, or great power members with corresponding moderate, higher, or greatest capabilities. As the power position of a state increases, it becomes capable of enacting more as well as different auxiliary situational roles (Thies, 2013, pp. 44–46).

While the transition in master roles may follow a linear pattern, it is also reversible as states can decline temporarily or permanently in power over time. States may also skip positional master roles, depending on how much and how rapidly their capabilities change vis-à-vis other states. For example, losers in major wars may go from great powers to small powers or novices may be thrust into great power roles by the disappearance or decline of other states in the power structure of the international system. The power position of a state is also somewhat relative, and its corresponding master role depends on the system of interest under analysis. A focus on the global system may define a state as a major power in its relations with other states while having the capabilities to act as a great power in its relations with states within its regional system.

Depending on the master role of a state, different auxiliary roles are potential or actual patterns in its interactions with other states. One example is the rise of the United States from its entry into the global system as a novice member in the 1783–1815 period after the successful Revolutionary War with Britain and its ascendance to the role of great power by the end of the 19th century. Another example is Israel’s entry into world politics as a novice member of the state system in the 1948–1956 period following the successful War for Independence in 1948 and its eventual rise to the role of a major member of the Middle East regional system after the Six-Day War in 1967. Figure 2 shows the time periods in which the United States and Israel occupied different master roles and enacted various auxiliary roles toward multiple alters (others) and audiences (third parties) at regional and global levels of the international system (Thies, 2013).

Role Theory as an Empirical Theory of International Relations: From Metaphor to Formal ModelClick to view larger

Figure 2. Role location and role transition patterns of United States (1783–1945) and Israel (1948–1967).*

Notes: *Master roles: novice, small, major, and great powers. Alters: Great Powers (GPs), United Kingdom (UK), France (FR), Latin America (LA), Asia (AS), Soviet Union (SU), Syria (Syr), United States (US), Arabs (AR), Egypt (Egy), Non-state Actors (NA).

Source: Extracted from Thies (2013).

Thies (2013, pp. 36–38) models the processes of role location and role transition that generate the patterns in Figure 2 with a socialization game. In this game ego (United States or Israel) pursues an auxiliary role that is either appropriate or inappropriate in relation to its master role (power position) in the regional or global system of interest. Among the audience of other states in the system, at least one responds as alter either by accepting or rejecting the auxiliary role. If alter’s response is “accept,” then the enactment of the auxiliary role is successful. If the auxiliary role is new but appropriate for ego’s master role, then acceptance also signals that auxiliary role differentiation has occurred. If this accepted role is inappropriate for ego’s current master role, then its acceptance by alter signals as well the transition from one master role to another.

If alter’s response is “reject,” then bilateral role conflict is present. If multiple alters in the audience disagree, that is, some accept and others reject ego’s role, then multilateral role conflict occurs (Walker & Simon, 1987, pp. 142–143). Thies (2013, pp. 38–39) identifies four responses as alternatives for ego to manage bilateral role conflict: force, fizzle, resolve, and reject (Thies, 2013, pp. 38–40). Force refers to ego’s use of power to overcome alter’s rejection of the role; fizzle occurs when ego accepts that it will not achieve the role; resolve is an intensified belief in the pursuit of the role by ego; reject is when ego stops desiring the role. While Thies (p. 39) describes these four responses mainly in cognitive and motivational terms, that is, in terms of beliefs and desires, they also manifest themselves in behavioral terms as acts of words or deeds directed toward alter. In fact, the process of role location involves emotional as well as behavioral, cognitive, and motivational aspects (M. Hermann, 1987; C. Hermann, 1987; Singer & Hudson, 1987; Walker, 2013).

Thies also attributes these possible patterns to alter as ego’s socializer, which makes the process of role location a dyadic process of social interaction. He models the social interaction between ego and alter as a socialization game with a structure determined by Nature and presented in simplified form as a decision tree in Figure 3. In the decision tree Ego chooses as step (1) either pursue or don’t pursue a role. If Ego chooses pursue, then Alter chooses as step (2) either accept or reject. If Alter chooses reject, then Ego chooses as step (3) one of the four outcomes (force, fizzle, resolve, reject) (Thies, 2013, pp. 37–40). Because the decision tree model depicts the socialization outcomes only for ego, it is desirable to expand the model to include the possible outcomes as well for alter. The possibilities for both ego and alter are shown in the decision matrix for the role location game in Figure 3.

Role Theory as an Empirical Theory of International Relations: From Metaphor to Formal ModelClick to view larger

Figure 3. Outcomes of the socialization and role location games for ego and alter.*

Notes: *Outcomes for Ego: Force, Fizzle, Resolve, Reject. Ego, Alter joint outcomes: Ego, Alter settle (+, +), Ego submits, Alter dominates (+, ?), Ego dominates, Alter submits (?, +), Ego, Alter deadlock (?, ?).

Sources: Adapted from Thies (2013) and Walker (2013).

The Thies socialization game in extended form as a decision tree has the same logical structure as the general role location game in normal form as a decision matrix. Each game begins from an initial state either at the top node of the tree or in one of the cells in the matrix. The player with the next move in both models can choose “stay” or “move” from the top node or initial cell. If the choice is “stay,” then the game is over unless the other player chooses “move.” The game continues until both players choose “stay” at a particular node or cell.

In the socialization game “Nature” in the form of the power structure of the international system of interest specifies the appropriate master role and corresponding auxiliary roles for ego. Sequential game theory in matrix form can depict this information from Nature as payoffs ranking different possible outcomes for ego and alter from highest (4) to lowest (1), which specifies their choices when they play a 2 x 2 game as a sequence of alternating binary moves that has the logical structure of a decision tree when depicted in extended form (Brams, 1994, pp. 27–34; see also Brams, 2002). In both games the power distribution between the players is a constraint that specifies the outcome of the game. In the socialization game it is the power distribution and ego’s corresponding power position that specifies whether a role is appropriate or inappropriate to pursue (Thies, 2013, pp. 37–38, 42–44). In the role location game the power distribution (or some other role demand in Nature) can specify each player’s preference rankings for the four outcomes of settle, deadlock, dominate, or submit (Walker, Malici, & Schafer, 2011, pp. 28–33).

Both games may be games of imperfect information, in which one or both players may not correctly judge the distribution of power. They may also interpret incorrectly each other’s decision to “stay” or “move,” which makes the final outcome of either the socialization game or the role location game relatively uncertain. However, the role location game circumscribes the range of outcomes by limiting decisions by ego (e) and alter (a) to a sequence of four moves {e, a, e, a} from one of the cells {+,+; +,−; −,−; −,+} in the game matrix as an initial state. Each player chooses CO+ or CF− in the sequence to enact four basic roles: (1) friend (unconditional cooperation); (2) partner (conditional cooperation); (3) rival (conditional conflict); (4) enemy (unconditional conflict). Examples of valence patterns within the sequence that signal the enactment of roles are (1) unconditional cooperation (+ − +); (2) conditional cooperation (+ + +) or (+ − −); (3) conditional conflict (− − −) or (− + +); (4) unconditional conflict (− + −).

In a sequence of four alternating moves by ego and alter, for example {e, a: + + − +}, the initial two moves signal a cooperation (+) or conflict (−) role for ego and alter. The third and fourth moves signal the final outcome and existing power distribution between them as symmetrical (e = a) or asymmetrical (e > a or e < a). In this example both ego and alter chose cooperation (+) as their respective initial moves while ego chose (−) and alter chose (+) as the final two moves. The role enacted by ego is exploitative partner (+ + −) and alter’s role is unconditional friend (+ − +) in this sequence; the power distribution (− +) between them is asymmetrical (e > a) as ego dominates (−) and alter submits (+) to constitute the final outcome of this sequence.

It is possible to check this description of the roles enacted by ego and alter against the role conceptions of ego and alter and the role demands from Nature that hypothetically specify or constrain the role enactment of ego and alter. Is the power distribution signaled in the enactment of these auxiliary situational roles consistent with the relative power (= or ≠) positions of their master roles? With the positive (+) or negative (−) valences of their national role conceptions? If these binary role enactment characteristics are congruent (match up) across behavioral, state, and systemic levels of analysis, then binary role theory can describe and explain foreign policy decisions and international relations of cooperation and conflict between ego and alter for a specific dyad and by extension for the ego-alter Nth dyads in an international system of interest (Walker, Malici, & Schafer, 2011, pp. 257–266).

In a recent study of British relations with the Axis powers of Germany, Italy, and Japan during the decade prior to World War II, Walker (2013) analyzed the fit among binary role theory’s three models of role demands, role conceptions, and role enactments (Walker, 2013). Combining the three sets of elements into a Power-Identity-National interests (PIN) formal model of role enactment, the one-sided, empirical analysis focused on describing and explaining ego’s patterns of congruence and incongruence among formal game models constructed primarily from British role enactment patterns, role demands, and role conceptions rather than focusing directly as well on alter’s role conceptions and behavior and the pattern of relations between Britain and each Axis power. Also identified were instances of role strain, role competition, and role conflict experienced by Britain, and the belated role transition from partner to rival in British foreign policy roles during the decade of the 1930s.

The authors of both the Thies (2013) and Walker (2013) studies have since moved toward refining their models and testing them more systematically in collaboration with colleagues (Thies & Nieman, 2015; Malici & Walker, 2016). Both research programs have focused on the problem of rising powers over extended periods of time in different regions. Thies and Nieman (2015) have investigated transitions in India’s foreign policy role between independence and the present era as part of a larger study of Brazil, Russia, India, China, and South Africa as the five rising BRICS powers (Thies & Nieman, 2017). They use a combination of qualitative methods to identify national role conceptions and quantitative methods (a Bayesian Markov chain Monte Carlo Poisson change-point model) to identify changes in India’s conflict behavior during militarized interstate disputes.

The Thies and Nieman (2015) analysis continues to be relatively one-sided in its emphasis on ego’s foreign policy roles although the authors are explicitly dyadic in their focus on militarized interstate disputes with discrete alters. They have also formalized a quantitative statistical model of role enactment to test against empirical patterns of role enactment and transition plus continued to lengthen the temporal scope of their analysis to span decades of role enactment patterns. More important, they have explicitly selected an important puzzle in international relations theory as their substantive problem, namely, analyzing the phenomenon of rising powers in regional and global international relations as a problem of role transition. This choice suggests a potentially fruitful bridge between role theory and important problems identified by other theories of world politics, for example, the rise and decline of hegemons, the assimilation of new states into the existing world order, and the emergence of new regimes in old states following civil or foreign conflicts (Walker, 2013, pp. 7–13; see also Gilpin, 1981; Modelski, 1987; Doran, 1991; Kupchan, 1994; Cederman, 1997; Schweller, 1998).

Thies (2015) has explored this bridge in the contemporary era with the application of his socialization model to Sino-American relations. He focused on the 1995–1996 Taiwan Straits crisis between China and the United States and employed the conceptual framework from his earlier studies of the United States and Israel as rising powers (Thies, 2013). In a two-sided analysis he analyzed the attempts of each party to influence the selection of roles by the other party and found that their attempts at altercasting one another into other roles failed during this crisis. Both China and the United States enacted roles from their own repertoire of role conceptions as ego rather than meet the role expectations communicated from the other as alter, which Thies concluded created conditions conducive to the renewal of their previous rivalry during the Cold War.

This set of problems is also the substantive focus identified by Malici and Walker (2016) in their two-sided analysis of the rise of Iran in world politics. They have analyzed three critical historical junctures in the history of the U.S.–Iran dyad: the nationalization of Iran’s oil industry and the CIA coup to reverse it and restore the shah as the ruler of Iran (1951–1953); the Iranian revolution and seizure of U.S. hostages (1978–1981); the temporary detente between Iran and the United States (1997–2002). The authors do an explicitly two-sided analysis of role conflict and role transition in U.S.–Iran relations with an automated content analysis of the national role conceptions of U.S. and Iranian leaders and the statistical analysis of an event history for the interactions between the two states during these three critical junctures. They employ the PIN model to analyze the fit among formal models based on the role demands of power and interests, role conceptions and expectations based on each state’s beliefs about the exercise of power, and role enactment valences based on words and deeds exchanged between the United States and Iran.

The two-sided analysis with the PIN model identifies the relative impact of role demands, role conceptions and expectations, and cues exchanged between ego and alter in resolving role conflicts and making role transitions in U.S.–Iran relations. The formal models employed in this analysis also identify the potential results of counter-factual accounts of their relations and offer some guidance to policymakers for managing and resolving role conflicts when they arise. Their study also demonstrates the logic of links between the micro-level beliefs of key leaders and the macro-level sequences and outcomes of strategic interaction episodes between states (Malici & Walker, 2016).

Conclusion: Toward a Third Wave

Absent on both sides of the Atlantic in the second wave is an explicit and extended theoretical and empirical focus by role theorists on strategic interactions modified by role conceptions and structural constraints inside and outside the state. Structure-oriented and agent-centered models have received the lion’s share of attention in the second wave while action-focused models have been relatively scarce. The main omissions in the second wave studies of role transition and role location studies are two-sided, extensive formal modeling of strategic interactions between ego and alter and similar attention to agents inside the state at the meso-level of social interactions that generate the adoption of national role conceptions and expectations (Cantir & Kaarbo, 2012, 2016). The formal modeling studies have mainly tested the congruence of the contents in the public statements of the leaders with the behavioral patterns between their states (Thies, 2013; Walker, 2013).

Therefore, the question is whether there will be a third wave for role theory with an extended and enduring focus on two-sided strategic interactions within and between ego and alter as agents in world politics. The recent use of game theory to model role contestation as well as role location and role transition processes suggests the possibility of a third wave of domestic interaction-focused studies of foreign policy decision-making as well as international interactions informed by formal models and quantitative methods (Walker, 2013, pp. 20–23; Walker, 2014; Walker, Schafer, & Beieler, 2016). If taken individually, role demands, role identities, and role enactments models are each potential candidates for an effective theory of international relations. The definition of an effective theory depends on the spatial and temporal scale of the analysis that is relevant to the problem at hand, as physicist Lisa Randall points out:

Everyone, including physicists, is happy to use a simpler description when the details are above our resolution. Physicists formalize this intuition and organize categories in terms of the distance or energy that is relevant. For any given problem we use what we call an effective theory. The effective theory concentrates on the particles and forces that have “effects” at the distances in question. Rather than declaring particles and interactions in terms of unmeasurable parameters that describe more fundamental behavior, we formulate our theories, equations, and observations in terms of the things that are actually relevant to the scales we might detect.

(Randall, 2012, p. 16. Italics in the original)

Taken collectively, the role demands, role conceptions, and role enactments models of role theory may constitute a potentially integrated and more fundamental role theory of politics regarding the exercise of power in international relations. Such a theory in physics is called a Grand Unified Theory (GUT), or “M-theory,” in which various models or “M’s” are actually different approximations of a more fundamental theory of physics. It may exist as a single formulation or as a network of models (Walker, 2013, p. 192; see also Hawking & Mlodinow, 2010, pp. 87–119).

The application of role theory as an empirical theory of international relations has followed an evolution in the empirical scope of its application from foreign policy analysis to the analysis of international relations. Accompanying the evolution in the scope of role theory’s empirical application to international relations has been an evolution in its logical structure from a metaphor or analogy to a formal model. In the first wave of applications between 1970 and 1990 the focus was primarily on employing a role conceptions model to describe the national role conceptions attributed to the leaders of states as they made foreign policy decisions. In the second wave of applications between 1990 and 2010 a role demands model was also employed to analyze the external and internal conditions that constrained states in the enactment of national role conceptions. A third wave of applications in the future may focus on the processes of role conflict, role location, and role transition to apply a role enactments model for the analysis of interactions between and among states.

Scientific progress in a research program or domain of inquiry is marked by the ability of a theory to reduce the number of unsolved empirical and conceptual problems in that domain (Laudan, 1977; see also Elman & Elman, 2003). The solution of empirical problems is determined by testing hypotheses that can be derived and falsified within the context of a theory. The solution of conceptual problems involves either clarifying ambiguities or inconsistencies internal to the theory or resolving any external conflicts between two theories where both theories appear to be well-founded (Walker, 2003, p. 269; see also Laudan, 1977). The evolution in role theory’s empirical scope from solving foreign policy analysis problems to solving international relations problems meets the first criterion. The conceptual evolution of role theory from metaphorical to formal models meets the second criterion of reducing internal conceptual problems by clarifying its internal logic. The logic of binary role theory’s PIN model formally links power, identities, and interests and simultaneously extends its scope while integrating variables from other theories that are considered to be well-founded by theorists in other research programs (Walker, Malici, & Schafer, 2011; Walker, 2013; see also Elman & Elman, 2003).

In meeting the criterion for solving external conceptual problems between theories, however, it is important to understand that it not necessary or desirable to discard old theories with the development of more powerful theories or instruments of observation. “It is only when the theory at the higher level of analysis becomes ineffective that it becomes desirable to disaggregate to a lower level of analysis in order to understand more deeply a phenomenon of interest” (Walker, 2013, p. 190; see also Randall, 2012, p. 15). It appears that there will continue to be effective theorizing and fruitful empirical work with structure-oriented, agent-based, and action-focused models of role theory applied at multiple levels of analysis in studies of foreign policy and international relations for the foreseeable future (Thies, 2010; Breuning, 2011; Harnisch, Frank, & Maull, 2011).


Aggestam, L. (2006). Role theory and European foreign policy. In O. Elgström & M. Smith (Eds.), The European Union’s roles in international politics (pp. 11–29). London: Routledge.Find this resource:

Archer, M. (1988). Culture and agency. New York: Cambridge University Press.Find this resource:

Baldwin, D. (1989). Paradoxes of power. New York: Blackwell.Find this resource:

Baldwin, D. (2002). Power in international relations. In W. Carlsnaes, T. Risse, & B. Simmons (Eds.), Handbook of international relations (pp. 273–297). London: SAGE.Find this resource:

Bengtsson, R., & Elgström, O. (2011). Reconsidering the European Union’s roles in international relations. In S. Harnisch, C. Frank, & H. Maull (Eds.), Role theory in international relations (pp. 113–130). London: Routledge.Find this resource:

Bengtsson, R., & Elgström, O. (2012). Conflicting role conceptions: The EU in global politics. Foreign Policy Analysis, 8(1), 93–108.Find this resource:

Biddle, B. (1979). Role theory: Expectations, identities, and behaviors. New York: Academic Press.Find this resource:

Black, M. (1962). Models and metaphors. Ithaca, NY: Cornell University Press.Find this resource:

Brams, S. (1994). Theory of moves. Cambridge, U.K.: Cambridge University Press.Find this resource:

Brams, S. (2002). Game theory in practice. In M. Brecher & F. Harvey (Eds.), Millennial reflections on international studies (pp. 81–96). Ann Arbor: University of Michigan Press.Find this resource:

Breuning, M. (2011). Role theory research in international relations: State of the art and blind spots. In S. Harnisch, C. Frank, & H. Maull (Eds.), Role theory in international relations (pp. 16–35). London: Routledge.Find this resource:

Burr, W., Leigh, G., Randall, D., Day, R., & Constantine, J. (1979). Symbolic interaction and the family. In W. Burr, R. Hill, F. I. Nye, & I. Reiss (Eds.), Contemporary theories about the family (pp. 42–111). New York: Free Press.Find this resource:

Callahan, P., Brady, L., & Hermann, M. (1982). Describing foreign policy behavior. Beverly Hills, CA: SAGE.Find this resource:

Cantir, C., & Kaarbo, J. (2012). Contested roles and domestic politics. Foreign Policy Analysis, 8, 5–24.Find this resource:

Cantir, C., & Kaarbo, J. (2016). Domestic role contestation, foreign policy, and international relations. New York: Routledge.Find this resource:

Carlsnaes, W. (1992). The agency-structure problem in foreign policy analysis. International Studies Quarterly, 26, 245–270.Find this resource:

Cederman, L. (1997). Emergent actors in world politics. Princeton, NJ: Princeton University Press.Find this resource:

Chavetz, G., Abramson, H., & Grillot, S. (1996). Role theory and foreign policy. Political Psychology, 17(4), 727–757.Find this resource:

Deutsch, K. (1966). The nerves of government. London: Free Press.Find this resource:

Deutsch, K. (1968). The analysis of international relations. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:

Doran, C. (1991). Systems in crisis. New York: Cambridge University Press.Find this resource:

Dunne, T., Hansen, L., & Wight, C. (2013). The end of international relations theory? European Journal of International Relations, 19(3), 405–425.Find this resource:

Elgström, O., & Smith, M. (Eds.). (2006). The European Union’s roles in international politics. London: Routledge.Find this resource:

Elman, C. (1996). Horses for courses. Security Studies, 6, 7–53.Find this resource:

Elman, C., & Elman, M. (2003). Progress in international relations theory. Cambridge, MA: MIT Press.Find this resource:

Feynman, R. (1995). Six easy pieces. Reading, MA: Addison-Wesley.Find this resource:

Gilpin, R. (1981). War and change in world politics. New York: Cambridge University Press.Find this resource:

Harnisch, S. (2011). Role theory: Operationalization of key concepts. In S. Harnisch, C. Frank, & H. Maull (Eds.), Role theory in international relations (pp. 7–15). London: Routledge.Find this resource:

Harnisch, S. (2012). Conceptualizing in the minefield: Role theory and foreign policy learning processes. Foreign Policy Analysis, 8(1), 47–69.Find this resource:

Harnisch, S., Berstick, S., & Gottwald, J. (2016). China’s international roles. New York: Routledge.Find this resource:

Harnisch, S., Frank, C., & Maull, H. (Eds.). (2011). Role theory in international relations. London: Routledge.Find this resource:

Hawking, S., & Mlodinow, L. (2010). The grand design. New York: Bantam Books.Find this resource:

Hermann, C. (1987). Superpower involvement with others: Alternative role relationships. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 219–246). Durham, NC: Duke University Press.Find this resource:

Hermann, C., Kegley, C., & Rosenau, J. (1987). New directions in the study of foreign policy. Boston, MA: Allyn and Unwin.Find this resource:

Hermann, M. (1987). Foreign policy orientations and the quality of foreign policy decisions. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 123–140). Durham, NC: Duke University Press.Find this resource:

Hermann, M. (2001). How decision units shape foreign policy. International Studies Review, 3, 47–81.Find this resource:

Hermann, M., & Hermann, C. (1989). Who makes foreign policy decisions and how? International Studies Quarterly, 33, 361–388.Find this resource:

Holsti, K. (1970). National role conceptions in the study of foreign policy. International Studies Quarterly, 14, 233–309.Find this resource:

Kaarbo, J. (2012). Coalition politics and cabinet decision-making. Ann Arbor: University of Michigan Press.Find this resource:

Kupchan, C. (1994). The vulnerability of empire. Ithaca, NY: Cornell University Press.Find this resource:

Laudan, L. (1977). Progress and its problems. Berkeley: University of California Press.Find this resource:

Le Prestre, P. (Ed.). (1997). Role quests in the post–Cold War era. Montreal: McGill-Queen’s University Press.Find this resource:

Malici, A., & Walker, S. (2016). Role theory and role conflict in U.S.–Iran relations. New York: Routledge.Find this resource:

McCall, G., & Simmons, J. (1978). Identities and interactions. New York: Free Press.Find this resource:

Mearsheimer, J., & Walt, S. (2013). Leaving theory behind: Why simplistic hypothesis testing is bad for IR. European Journal of International Relations, 19(3), 427–457.Find this resource:

Modelski, G. (1987). Long cycles in world politics. London: Macmillan.Find this resource:

Preston, T. (2001). The president and his inner circle. Ann Arbor: University of Michigan Press.Find this resource:

Randall, L. (2012). Knocking on heaven’s door. New York: Harper-Collins Ecco.Find this resource:

Richardson, N. (1987). Dyadic case studies in the comparative study of foreign policy. In C. Hermann, C. Kegley, & J. Rosenau (Eds.), New directions in the study of foreign policy (pp. 161–177). Boston: Allen and Unwin.Find this resource:

Rosenau, J. (1987). Roles and role scenarios in foreign policy. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 44–65). Durham, NC: Duke University Press.Find this resource:

Rosenau, J. (1990). Turbulence in world politics. Princeton, NJ: Princeton University Press.Find this resource:

Sarbin, T., & Allen, V. (1968). Role theory. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (2d ed., pp. 488–567). Reading, MA: Addison-Wesley.Find this resource:

Schweller, R. (1998). Deadly imbalances. New York: Columbia University Press.Find this resource:

Shih, C. (2012). Assigning role characteristics to China. Foreign Policy Analysis, 8(1), 71–91.Find this resource:

Simon, H. (1985). Human nature in politics. American Political Science Review, 79, 293–304.Find this resource:

Singer, J. (1961). The level of analysis problem in international relations. World Politics, 14, 77–92.Find this resource:

Singer, E., & Hudson, V. (1987). Role sets and African foreign policy behavior. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 199–218). Durham, NC: Duke University Press.Find this resource:

Snidal, D. (1985). The game theory of international politics. World Politics, 38(1), 25–57.Find this resource:

Stryker, S., & Serpe, R. (1982). Commitment, identity salience, and role behavior. In W. Ickes & E. Knowles (Eds.), Personality, roles, and social behavior (pp. 199–218). New York: Springer.Find this resource:

Stryker, S., & Statham, A. (1985). Symbolic interaction and role theory. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (3d ed., pp. 311–378). New York: Random House.Find this resource:

Thies, C. (2010). Role theory and foreign policy. In R. Denmark (Ed.), The international studies encyclopedia (pp. 6335–6356). Chichester, U.K.: Wiley-Blackwell.Find this resource:

Thies, C. (2012). International socialization processes and Israeli national role conceptions. Foreign Policy Analysis, 8(1), 25–46.Find this resource:

Thies, C. (2013). The United States, Israel, and the search for international order. New York: Routledge.Find this resource:

Thies, C. (2015). China’s rise and the socialisation of rising powers. Chinese Journal of International Politics, 8(3), 281–300.Find this resource:

Thies, C., & Breuning, M. (2012). Integrating foreign policy analysis and international relations through role theory. Foreign Policy Analysis, 8(1), 1–4.Find this resource:

Thies, C., & Nieman, M. (2015). Emerging powers, identity, and conflict behavior: India as an emerging power in the international system. New Orleans, LA: International Studies Association.Find this resource:

Thies, C., & Nieman, M. (2017). Rising powers and foreign policy revisionism: Understanding BRICS identity and behavior through time. Ann Arbor: University of Michigan Press.Find this resource:

Walker, S. (1979). National role conceptions and systemic outcomes. In L. Falkowski (Ed.), Psychological models in international politics (pp. 169–210). Boulder, CO: Westview Press.Find this resource:

Walker, S. (1987). Role theory and foreign policy analysis. Durham, NC: Duke University Press.Find this resource:

Walker, S. (1992). Symbolic interactionism and international politics. In M. Cottam & C. Shih (Eds.), Contending dramas: A cognitive approach to international organizations (pp. 19–38). New York: Praeger.Find this resource:

Walker, S. (2003). Operational code analysis as a scientific research program. In C. Elman & M. Elman (Eds.), Progress in international relations theory (pp. 245–276). Cambridge, MA: MIT Press.Find this resource:

Walker, S. (2014, September 4–7). Game theory and digraph models of role contestation. Presented at the Annual Meeting of the European Consortium for Political Research, Glasgow.Find this resource:

Walker, S., Malici, A., & Schafer, M. (2011). Rethinking foreign policy analysis. New York: Routledge.Find this resource:

Walker, S., Schafer, M., & Beieler, J. (2016). Belief systems and foreign policy roles: Role contestation in U.S. foreign policy decisions. In C. Cantir & J. Kaarbo (Eds.), Domestic role contestation, foreign policy, and international relations (pp. 122–139). New York: Routledge.Find this resource:

Walker, S., & Simon, S. (1987). Role sets and foreign policy analysis in Southeast Asia. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 141–159). Durham, NC: Duke University Press.Find this resource:

Waltz, K. (1959). Man, the state, and war. New York: Columbia University Press.Find this resource:

Waltz, K. (1979). Theory of international politics. Reading, MA: Addison-Wesley.Find this resource:

Wehner, L. (2015). Role expectations as foreign policy: South American secondary powers’ expectations of Brazil as a regional power. Foreign Policy Analysis, 11(4), 435–455.Find this resource:

Wendt, A. (1987). The agent-structure problem in international relations theory. International Organization, 41, 335–370.Find this resource:

Wish, N. (1987). National attributes as sources of national role conceptions. In S. Walker (Ed.), Role theory and foreign policy analysis (pp. 94–103). Durham, NC: Duke University Press.Find this resource:

Walker, S. G. (2013). The Cognitive Architecture of British Appeasement Decisions: Symbolic and Strategic Interaction in World Politics. New York: Routledge.Find this resource: