Operational Codes and Foreign Policy Roles Conceptual Insights and Empirical Results
Stephen G. Walker and Mark Schafer
In this chapter, we assess the conceptual insights and empirical results of the operational code analysis (OCA) and foreign policy role studies in this volume with criteria identified as markers of progress in scientific research programs. Do these studies represent advances in the form of solving conceptual or empirical problems, expanding the scope of analysis into new domains of inquiry, and identifying new avenues of research (Lakatos 1970)? Can an alliance of OCA and role theory as a “theory complex” link the domestic and foreign domains of world politics (Laudan 1977)? Does an alliance also address the gap in foreign policy theory and practice between the academic and policy communities (George 1993)? Are conceptual and empirical anomalies resolved and conflicting theories reconciled across research programs in foreign policy analysis (FPA) and world politics (Lakatos 1970; Laudan 1977)? Are the solutions to these problems also consistent with the solutions to problems of evidence, inference, and conceptualization in psychological explanations of world politics (Greenstein 1987)? Do they represent significant progress for crossing Simon’s bridge and linking the sciences of the mind and the social sciences (Lupia, McCubbins, Popkin 2000)?
Almost 15 years ago we stated, “Operational code analysis may make more contributions to foreign policy theory in the next decade than in the last decade” (Walker and Schafer 2006, 246). Specifically, (pp. 246-248) we anticipated progress by OCA along four paths: (1) as the core of a cog- nitivist research program that offers an agent-centered account of foreign policy and world politics; (2) as a method for system-level research programs to augment and enrich their structure-oriented explanations of world politics; (3) as an approach offering alternative or complementary accounts of strategic interactions between states at the dyadic level of analysis in world politics; (4) as an approach to identify non-linear processes and emerging properties of complex systems with agent-based simulations of strategic interactions.
In this chapter, we revisit these forecasts as well as the prospects for substantive, theoretical, and methodological progress in the operational code research program regarding the advancement of actor-specific, situation-generic, and abstract-theoretical knowledge of foreign policy and world politics (George 1993).
The chapters in Part II, The Operational Codes of World Leaders, focus on the cognitive explanation offered by a leader’s key operational code beliefs in an agent-centered account of foreign and domestic policy decisions regarding the exercise of power by specific leaders in autocratic (Putin) and democratic (Bush, Obama, Trump) political systems. This approach to actor-specific knowledge of political leaders is also used in this volume to augment or qualify system-level and structure-oriented explanations of various features of world politics, including variations in the beliefs of individual leaders (Kumaratunga in Sri Lanka, Morales in Bolivia, Al-Qaeda and ISIS leaders in the Middle East and North Africa) to account for the variations in civil violence in Sri Lanka, the international rivalry between Bolivia and Chile, and the organization and lethality of militant terrorist groups in the MENA region.
The chapters in Part III, The Psychological Characteristics of US Presidents, integrate models and methods from agent-centered and system-level research programs to focus on the decisions and actions of the United States as a key agent in 20th century world politics. This focus extends from the personality traits and operational code beliefs of US presidents through the economic and military power of the United States and the evolution of international norms over the course of the 20th century to analyze dyadic relations between the United States and other states between 1900 and the dawn of the 21st century. This effort yields situation-generic knowledge about the most powerful agent in world politics during military dispute situations (MIDS) and insights about linkages between presidential leadership patterns in the foreign policy domain and congressional electoral outcomes in the domestic policy domain. Generic patterns of learning and threat perception by US presidents as a group in the 20th century are also revealed along with their origins from variations in presidential personality traits.
The chapters in Part IV, Computational Models of Foreign Policy Roles, synthesize abstract-theoretical knowledge across different levels of aggregation and analysis in world politics. Their focus is on strategic interactions among dyads and evolutionary patterns over time among states in different regional and world systems. In particular, the link between key operational code beliefs about the nature of the political universe, the optimal strategies for exercising social power in the form of positive and negative sanctions, and control over historical development are linked to the emergence of the foreign policy roles of friend, partner, rival, and enemy in world politics. In turn, the interaction patterns between roles enacted by pairs of states as role dyads are examined to link their dynamics to the outcomes between them of mutual cooperation or mutual conflict and domination or submission. The strategies of appeasing, bandwagoning, balancing, or bullying are specified as different computational models for enacting different foreign policy roles measured by the metrics of operational code indices applied to data from computer simulations or historical case studies of interactions between states.
The conceptual insights, empirical metrics, and mathematical models of OCA as a research program are the methodological tools for “bridging the gap” (George 1993) in this volume between actor-specific, situation-generic, and theoretical-abstract knowledge about FPA and international relations (IR) as domains of scientific inquiry. The OCA research program’s aggregation and synthesis of the personalities and beliefs of leaders and the roles of states with foreign policy strategies and dyadic relations between agents in world politics represents substantive progress in the integration of FPA and IR as fields of study in political science.
Have these substantive advances followed sound principles of theoretical and methodological progress in the natural and social sciences? If there has been progress, what is the quality of these contributions? There are various standards offered by different philosophers of science and practicing political scientists for assessing progress in a scientific research program (Kuhn 1962; Lakatos 1970; Laudan 1977; Elman and Elman 2003; Jackson 2011). Their criteria include “the generation of novel facts, the solution of conceptual and empirical problems... escaping the endogeneity trap and... [surviving]... severe testing” (Walker 2003, 274). The form that the employment of these criteria take can vary, but they share a common goal of deeper understanding and explanation via “the systematic application of a set of theories and concepts so as to produce a ‘thoughtful ordering of empirical actuality’” (Jackson 2011, 20 citing Weber 1999), which is the distinguishing characteristic of science as a knowledge-production enterprise (Jackson 2011, 16-23). A conceptual framework’s logical form or model can be represented as a temporal narrative, statistical array, or mathematical equation, and the fit with empirical contents can be with inductive, deductive, abductive or reflexive methods of inference (Jackson 2011).
Theoretical progress primarily involves strengthening and extending the analytical reach of the conceptual framework in a research program.
It can take the form of refining, extending, or replacing an existing conceptual framework. Over the past 15 years, the conceptual framework of the operational code research program has experienced all three forms of theoretical progress. The concept of an operational code has been refined from being simply a belief system that expresses a code of conduct in the exercise of social power to a multi-dimensional construct of philosophical beliefs about the political universe and instrumental beliefs about the utility of different kinds of social power manifested by different types of agents (individuals, groups, coalitions, and institutions). It has been extended as well to include an agent’s pattern of cognitive beliefs about the exercise of power and an agent’s pattern of actual social conduct in the exercise of social power.
If these theoretical amendments result in the discovery and corroboration of novel facts, then their succession marks theoretical progress (Lakatos 1970; Walker 2003). The empirical contents of operational code theory are represented by the chapters in this volume as a series of theoretical models focusing on cognitive and social phenomena. As a cognitive theory (T,), the model was expressed as the cognitive consistency between philosophical and instrumental beliefs and the congruence between the beliefs and decisions of agents. As a cognitive-motivational theory (T2) that incorporated personality as well as cognitive features, the model predicted novel facts about relationships between the exercise of social power and a leader’s beliefs, motivations, and interpersonal style.
This progression from a cognitive to a motivational model occurred at the expense of the hierarchical relationship of cognitive consistency in (T,) between philosophical and instrumental beliefs. There emerged a dissonance pattern between types of beliefs and behavior and then a third operational code model of “hot” cognition (T3), which guided operational code research until the focus of the operational code research program expanded from the study of individual leaders to include the explicit analysis of states and dyads as units of analysis (Walker 2003; Walker and Schafer 2010). Today a fourth model of role enactment (T4) informs the application of metrics and mathematics from OCA as the mechanics for applying binary role theory to the political universe (Walker 2013; Malici and Walker 2017). While the succession of theoretical models and novel facts resembles the pattern of theoretical progress identified by Lakatos (1970), it adheres more closely to the criteria of theoretical progress identified by Laudan (1977):
The multiple criteria for defining and detecting progress, i.e., the inflation of solved empirical and conceptual problems and the deflation of conceptual problems, capture important activities within the operational code research program. The solution of conceptual problems is largely ignored in a Lakatosian account of the program’s trajectory.... More generally, there is less flexibility in the Lakatos account for simultaneously investing in several theories, dividing one’s intellectual labor across theories, and giving equal weight to the tasks of empirical
and conceptual problem-solving (Walker 2003, 270-271).
There is also little evidence in the operational code research program to support Kuhn’s (1962) thesis that scientific progress follows a revolutionary pattern rather than an evolutionary pattern. There has been no paradigm-shift characterized by a dramatic change in the program’s theory and methods. There is instead a pattern of cooperation rather than competition and the evolution of a “theory complex” or alliance among models and methods at different levels of aggregation and analysis marking theoretical and methodological progress in OCA (Walker 2003, 274-276).
Two criteria for assessing methodological progress include “escaping the endogeneity trap” and surviving “severe testing” (Walker 2003, 274; see also Dessler 2003; Keohane and Martin 2003). “Escaping the endogeneity trap requires that for beliefs to matter, they must be relatively independent of structural features of the decision-making environment and have a significant causal impact on decisions and outcomes” (Walker 2003, 274). This formulation of the endogeneity trap resembles the arguments advanced by political psychologists regarding the scope conditions for when the individual differences between leaders make a difference in political decisions and/or political outcomes (George 1969; Hermann 1974; Holsti 1976; Walker 1982; Greenstein 1987 ). In this view, the cognitive and extra-cognitive personality traits of leaders compete with situational variables for influence over the exercise of power in politics.
The dispositional variables are more likely to be influential and perhaps indispensable in understanding and explaining decisions and outcomes when the environment is loosely structured and/or a leader is located in a strategic position in the decision-making process. In a loosely structured, uncertain environment where information is either very scarce or too plentiful to process prior to the point of decision, then who leads matters, e.g., in a crisis situation where a single leader or a small group makes decisions for war or peace under the stress of high stakes, surprise, and a short time to decide (Hermann 1972; Brecher and Wilkenfeld 1997).
Under such conditions, a decision maker’s emotions and idiosyncratic biases may directly influence decisions by re-enforcing or substituting for unreliable information from the environment as the basis for making choices that turn out to be foreign policy mistakes (Greenstein 1987; Schafer and Crichlow2010; Walker and Malici 2011). Under more favorable conditions where decision-making is made with more systematic and less ad hoc procedures and decision units, the influence of a leader’s personality and beliefs may be exercised indirectly by the leader’s control over the organization of the advisory system and who are advisors
(George 1980; Hermann and Preston 1994: Preston 2001; Schafer and Crichlow 2010; Winter 2003).
Escape from the endogeneity trap in the operational code research program is managed with the adoption of “bounded rationality” as a Goldilocks “just right” solution to the problem (Walker 2009). Both the dispositions of the agent (A) and conditions in the environment (E) influence decisions, actions, and outcomes regarding the exercise of power by their interaction (A E). Moreover, both (A) and (E) as a system of interaction are agents in a social system and internalize both their relative personality traits as instrumental beliefs for ego defense and their respective environmental conditions as philosophical beliefs for object appraisal in their cognitive belief systems. Collectively, they mediate self- other relations by constructing through their interaction the map of a complex adaptive social system constituted from complex adaptive subsystems of beliefs and personality traits (Smith 1968). The complexity of their interactions is measured statistically by the VICS metrics of OCA applied to texts of mental and social events and modeled mathematically with the rules of games specified by binary role theory (Walker and Schafer 2010; Walker. Malici, and Schafer 2011; Walker 2013; Malici and Walker 2017).
This solution is entailed as well by Simon’s (1957) concept of bounded rationality, in which “Policy makers may act rationally, but only within the context of their simplified subjective representations of reality” (Simon 1957; Tetlock 1998, 876; cited in Walker 2003, 255). It is also consistent with Hawking and Mlodinow’s (2010, 42-46) position: “There is no picture- or theory-independent concept of reality. Instead we will adopt a view that we will call model-dependent realism: the idea that a physical theory or world picture is a model... and a set of rules that connect the elements of the model to observations.... According to mod- el-dependent realism, it is pointless to ask whether a model is real, only whether it agrees with observation. If there are two models that both agree with observations... then one cannot say that one is more real than another” (cited in Walker 2013, 228, n. 18).
Jackson (2011, 197) identifies four kinds of models employed by IR theories, which are distinguished by their respective ontological and epistemological stances regarding whether their methodological commitments are (1) dualistic or monistic ontologically with respect to the relationship between knower and known; (2) phenomenological or transfactual epistemologically with respect to the relationship between knowledge and observation. The issue of whether the theoretical models in the OCA research program will survive “severe testing” becomes defined by these criteria for testing them as bounded rationality models (Jackson 2011, 196-201).
The bounded rationality models in the OCA research program take a monistic stance in which the knower and the known are both endogenous to the model. They take a phenomenalist stance regarding what is observable by assuming that internal dispositions and thoughts can be detected from external patterns of speech and actions that communicate information about personality traits and beliefs. The criteria for testing these kinds of models in the OCA research program are testing hypotheses with frequentist patterns of statistical association and inference and testing analytical narratives with sequentialist patterns of computational rules of congruence. The statistical tests address how frequently do patterns of beliefs and actions occur and how likely they have not occurred by chance. The analytical narrative tests address how closely the distribution pattern of beliefs or sequence of actions are congruent (match up) with the pattern of distribution or sequence in the analytical narrative (Walker, Schafer, and Young 1998, 2003; Schafer and Walker 2006; Walker. Malici, and Schafer 2011).
Crossing Simon’s Bridge: The Road Ahead
In this volume, we have compiled several previously unpublished studies from the OCA research program under the general rubric of “crossing Simon’s bridge.” Crossing Simon’s bridge refers to bridging the sciences of the mind and the social sciences. The phrase articulates a unifying theme connecting the chapters in Parts II, III, and IV of the book, which move from a focus on the minds of individual political leaders in Part II to placing leaders in political situations in Part III, and examining the ensuing patterns of political actions and outcomes in Part IV. While we have not followed the same leaders along this path, we have employed a common set of analytical tools to mark the location and progress of different kinds of political agents across these levels of political aggregation and analysis. The tools are the models and metrics of OCA that bridge the gap between actor-specific knowledge of the personalities and beliefs of individual leaders, the situation-generic knowledge of political situations, and the abstract-theoretical knowledge of political systems.
These models and metrics provide a lexicon, a common language for detecting and analyzing the phenomena of beliefs, actions, and outcomes, which are the major ontological dimensions of the “intentional stance” in understanding and predicting human behavior as a competence theory (Dennett 1989; Jackson 2011). In contrast, the physical stance is a sub-personal performance theory, which emphasizes comprehension rather than competence. Employing the physical stance is to “predict the behavior of a system, determine its physical constitution (perhaps all the way down to the microphysical level) and the physical nature of the impingements upon it, and use your knowledge of the laws of physics to predict the outcome for any input” (Dennett 1989, 16). The design stance is a variant of the physical stance wherein “one ignores the actual (possibly messy) details of the physical constitution of an object, and, on the assumption that it has a certain design, predicts that it will behave as it is designed to behave under various circumstances” (Dennett 1989, 17).
Dennett (1989,43-82) locates the intentional stance within three levels of psychological explanation: the folk psychology of ordinary language, the personal intentional system of social psychology, and the sub-personal performance system of cognitive psychology. Folk psychology refers to the explanations offered by lay persons (rather than scientists) when they use beliefs and desires as every day terms to explain human behavior while the intentional stance links those terms as an intentional system of bounded rationality. “Intentional system theory... is envisaged as a close kin of, and overlapping with, such already existing disciplines as decision theory and game theory, which are similarly abstract, normative, and couched in intentional language. It borrows the ordinary terms “belief” and “desire” but gives them a technical meaning within the theory” (Dennett 1989, 58). The physical stance’s methodology employs instruments from the natural sciences applicable to studying the sub-personal characteristics of the human brain such as the trilogy of cognitions, emotions, and motivations associated with neuroscience (Ledoux 2002).
The intentional stance is for analyzing the personal and social levels of the human self, which we understand here in terms of the concept of bounded rationality and through access to the instruments of language and other forms of observing interpersonal communication. The building blocks for “crossing Simon’s bridge” in the OCA research program are beliefs and desires as specified by the concept of bounded rationality in the domain of world politics. The internal boundaries of bounded rationality at the individual level of analysis are extracted from texts and text-actions as patterns of instrumental beliefs or sequential actions attributed to the self while the external boundaries are extracted from texts or text-actions as patterns of philosophical beliefs or sequential actions attributed to others. The corresponding boundaries at the state level of analysis are the patterns of role conceptions, role expectations, and role enactments attributed to the self as Ego and to others as Alters.
Greenstein (1987, 14-20) identifies three steps in aggregating a psychological explanation of personality and politics across individual and state levels of analysis to systemic outcomes: (1) single case studies of individual leaders; (2) typological studies to classify leaders; (3) aggregative analyses to contextualize the performance of types of leaders. He specifies two general strategies of bridging the gap between them: “building up from direct observation of small-scale political processes.... and working back from theoretical analyses of systems and their psychological requirements” (Greenstein 1987, 127-139; Greenstein and Lerner 1971). These building blocks and strategies of aggregation are reflected in the organization of this volume. The operational code beliefs of the individual leaders in Part II are analyzed as (1) single case studies and classified into (2) a typology of belief systems and motivational profiles developed by Holsti (1977) and refined by Walker (1983). Strategies of aggregative analyses (3) are followed in Parts III and IV to contextualize the performance patterns of different types of leaders or other focal actors (e.g., groups, coalitions, organizations, institutions, states).
The “building up” strategy of aggregation analysis is pursued in Part III from small-scale phenomena exhibited by individual leaders by aggregating to state-level phenomena in the form of performance patterns associated with the US presidency and the electoral system plus US dyadic behavioral patterns regarding militarized disputes and international norms. The disaggregation strategy of “working back” from theoretical analyses of systems to the psychological requirements of their leaders is followed in Part IV wherein different principles of international order represented by different combinations of foreign policy roles are identified for states as agents. Then the psychological requirements for the enactment of those roles are specified in the form of different types of social and cognitive operational codes.
The overall patterns of aggregation (building up) and disaggregation (working back) are both hierarchical and recursive, i.e., it is possible for subsystems to aggregate as the building blocks for a larger system and also to disaggregate from the larger system into the smaller subsystems. These patterns of composition and decomposition between parts and wholes have potential spatial and temporal dimensions, which fluctuate and reveal them as complex adaptive systems that persist or change as they adapt to their environments (Simon 1969; Gell-Mann 1994; Axelrod and Cohen 1999). At the state level of aggregation and analysis, dyads, triads, and larger ensembles of states aggregate spatially into even larger regional systems of world politics, which Buzan and Waever (2003) identify as security complexes. They also aggregate temporally into relations of amity, collaboration, discord, and enmity and their corresponding roles of friend, partner, rival, and enemy, which are the building blocks that define their relations.
These roles are identified by their respective strategies of unconditional cooperation (friend), conditional cooperation (partner), conditional conflict (rival), and unconditional conflict (enemy). Binary role theory specifies the models of interaction games between pairs of states who are members of a regional security complex, defined collectively by Buzan and Waever (2003, 491) as “a set of units whose major processes of securitization, desecuritization, or both are so interlinked that their security problems cannot reasonably be analyzed or resolved apart from one another” (see also Lake and Morgan 1997; Walker 2007). In these games, the players (Ego and Alter) exchange cues that signal their respective role identities and their transitions from one role to another over time. Searching for these patterns in regional systems of world politics along the following paths of inquiry are likely future trajectories in the operational code research program, as its scope expands from analysis of the cognitive operational codes of individual leaders to the analysis of the social operational codes of states, dyads, triads, and larger ensembles of agents and systems in world politics.
- • One future path of inquiry is to expand the cognitive operational code profiles of individual leaders. The PsyCL data set presented in this volume is an expanding data resource for retrieving the personality traits and belief systems of both US and foreign leaders as the sources of foreign policy roles.
- • A second future path of inquiry is either to collect event data sequences independently or from other scholars and analyze the role enactment patterns that emerge spatially and temporally to define relations among members of a domestic or regional security complex (see Goertz, Diehl, and Balas 2016; Walker, Schafer, and Beieler 2016).
- • A third future path of inquiry is to compare the role conceptions and role expectations attributed to one another by pairs of predominant leaders or other focal actors whose public statements contain this cognitive information with the actual roles enacted between them as members of a domestic or regional security complex.
Data collection methods for these paths of inquiry are exemplified in Part II of this book. The information about role enactment and role selection from these sources may be analyzed either statistically with VICS metrics or computationally with formal TOM models. Prototypes of data analysis methods for statistical research designs are in Part III and for computational research designs in Part IV of this volume. Event data collections by other scholars offer opportunities to analyze the role enactment patterns of state-level actions and the social operational code profiles of states and institutions within states (Goertz, Diehl, and Balas 2016; Walker, Schafer, and Beieler 2016).
Finally, the evolution of OCA beyond the beliefs of individual leaders to focus on the social operational codes of larger decision units as focal actors also offers a fourth path of inquiry, which is to expand beyond the intentional stance and employ the design stance to do an OCA of artificial systems (Simon 1969). The social roles and corresponding strategies that constitute the games in binary role theory are artificial social systems of interaction that are subject to human design and simulation. Such games will also occur “naturally” as patterns of interaction between agents as the political universe “programs” itself the way that the physical universe does in the absence of intervention by human engineers (Lloyd 2006).
The building blocks for designing artificial systems of world politics are the roles specified as games by binary role theory. These roles perform the same function for building the social operational codes at the state level of analysis as beliefs do for building cognitive operational codes at the individual level of analysis. As building blocks for aggregations of behavioral patterns between and among states, roles can be deployed in a physical stance and fit Dennett’s (1989, 16) earlier description of the physical stance by substituting “social” or “psychology” for “physical” or “physics” as follows:
If you want to predict the behavior of a [social] system, determine its [social] constitution (perhaps all the way down to the [microsocial] level) and the [social] nature of the impingements upon it, and use the knowledge of the laws of [psychology] to predict the outcome for any input (Dennett 1989, 16).
Investigating and designing these patterns and programs as complex artificial systems informed by binary role theory from a physical stance is a task for another day, which is already under way in the operational code research program (Walker no date). In the meantime, world leaders and other focal actors will continue to be analyzed in the operational code research program as the “programmers” of world politics from an intentional stance defined by binary role theory models and the metrics of OCA (Walker and Malici 2011; Dennett 2017; Malici and Walker 2017, 178-184; Walker 2007).
Axelrod, R., M. Cohen. 1999. Harnessing complexity. New York, NY: Free Press. Brecher, M., J. Wilkenfeld. 1997. A study of crisis. Ann Arbor, MI: University of Michigan Press.
Buzan, B., O. Waever. 2003. Regions and powers: The structure of regional security.
Cambridge: Cambridge University Press.
Dennett, D. 1989. The intentional stance. Cambridge, MA: MIT Press.
Dennett, D. 2017. From bacteria to Bach and back: The evolution of the human mind. New York, NY: W.W. Norton.
Dessler, D. 2003. Explanation and scientific progress. In Progress in international relations theory, eds. C. Elman, M. Elman, 381-401. Cambridge, MA: MIT Press. Elman, С, M. Elman. Eds. 2003. Progress in international relations theory. Cambridge, MA: MIT Press.
Gell-Mann, M. 1994. The Quark and the Jaguar. New York: Henry Holt and Company.
George, A. 1969. The ‘operational code’: A neglected approach to the study of political leaders and decision making. International Studies Quarterly 23: 190-222. George, A. 1980. Presidential decision making in foreign policy: The effective use of information and advice. Boulder, CO: Westview.
George, A. 1993. Bridging the gap: Theory and practice in foreign policy. Washington, DC: United States Institute for Peace.
Goertz, G., P. Diehl, A. Balas. 2016. The puzzle of peace. New York. NY: Oxford University Press.
Greenstein, F. 1987. . Personality and politics. Princeton, NJ: Princeton University Press.
Greenstein, F., M. Lerner. 1971. A source book for the study of personality and politics. Chicago, IL: Markham.
Hawking, S., L. Mlodinow. 2010. The grand design. New York, NY: Bantam Books.
Hermann, С. 1972. International crises. New York, NY: Free Press.
Hermann, M. 1974. Leader personality and foreign policy behavior. In Comparing foreign policies, ed. J. Rosenau, 201-234. New York, NY: John Wiley.
Hermann, M., T. Preston. 1994. Presidents, advisors, and foreign policy. Political Psychology 15(1): 75-96.
Holsti, O. 1976. Foreign policy viewed cognitively. In The structure of decision, ed. R. Axelrod, 18-54. Princeton, NJ: Princeton University Press.
Holsti, O. 1977. The 'operational code’ as an approach to the analysis of beliefs systems. Final Report to the National Science Foundation. Grant SOC 75-15368. Durham: Duke University.
Jackson, P. 2011. The conduct of inquiry in international relations. New York, NY: Routledge.
Keohane, R., L. Martin. 2003. Institutional theory as a research program. In Progress in international relations theory, eds. C. Elman, and M. Elman, 71-109. Cambridge, MA: MIT Press.
Kuhn, T. 1962. The structure of scientific revolutions. Chicago, IL: University of Chicago Press.
Lakatos, I. 1970. Falsification and the methodology of scientific research programmes. In Criticism and the growth of knowledge, 91-196. Cambridge: Cambridge University Press.
Lake, D., P. Morgan. Eds. 1997. Regional orders: Building security in a new world. University Park, PA: Pennsylvania State University Press.
Laudan, L. 1977. Progress and its problems: Toward a theory of scientific growth. Berkeley, CA: University of California Press.
Ledoux, J. 2002. Synaptic self: How our brains become who we are. New York, NY: Viking Press.
Lloyd, S. 2006. Programming the universe. New York, NY: Alfred Knopf.
Lupia, A., M. McCubbins, S. Popkin. Eds. 2000. Elements of reason: Cognition, choice and the boundaries of rationality. New York, NY: Cambridge University Press.
Malici, A., S. Walker. 2017. Role theory and role conflict in U.S.-Iran relations. New York, NY: Routledge.
Preston, T. 2001. The president and his inner circle. New York, NY: Columbia University Press.
Schafer, M., S. Walker. 2006. Beliefs and leadership in world politics. New York, NY: Palgrave.
Schafer, M., S. Crichlow. 2010. Groupthink r.y. high-quality decision making in international relations. New York, NY: Columbia University Press.
Simon, H. 1957. Models of man. New York, NY: John Wiley.
Simon, H. 1969. Sciences of the artificial. Cambridge, MA: MIT Press.
Smith, M. B. 1968. A map for the analysis of personality and politics. Journal of Social Issues 24: 15-28.
Tetlock, P. 1998. Social psychology and world politics. In Handbook of social psychology, eds. D. Gilbert, S. Fiske, G. Lindzey, 869-912. New York, NY: McGraw-Hill.
Walker, S. 1982. Psychological explanations of international politics: Problems of aggregation, measurement, and theory construction. In Biopolitics, political psychology, ami international politics, ed. G. Hopple. London: Francis Pinter.
Walker, S. 1983. The motivational foundations of political belief systems. International Studies Quarterly 27 179-202.
Walker, S. 2003. Operational code analysis as a scientific research program: A cautionary tale. Progress in international relations theory eds. C. Elman and M. Elman, 245-276. Cambridge: MIT Press.
Walker, S. 2007. Generalizing about security strategies in the Baltic Sea region. In Security strategies, power disparity, and identity, ed. O. Knudsen, 149-176. Burlington, VT: Ashgate.
Walker, S. 2009. The psychology of presidential decision making. In The Oxford handbook of the American presidency, eds. G. Edwards, W. Howell, 550-576. New York, NY: Oxford University Press.
Walker, S. 2003. Operational code analysis as a scientific research program: A cautionary tale. In Progress in international relations theory, eds. C. Elman, M. Elman, 245-276. Cambridge, MA: MIT Press.
Walker, S. no date. Thinking small: Binary role theory and the dynamics of world politics. Unpublished manuscript.
Walker, S. 2013. Role theory and the cognitive architecture of British appeasement decisions. New York, NY: Routledge.
Walker, S. 2003. Operational code analysis as a scientific research program: A cautionary tale. Progress in international relations theory eds. C. Elman and M. Elman, 245-276. Cambridge: MIT Press.
Walker, S., M. Schafer. 2006. Structural international relations theories and the future of operational code analysis. In Beliefs and leadership in world politics, eds. M. Schafer and S. Walker, 237-248.
Walker, S., M. Schafer 2010. Operational code theory. In The international studies encyclopedia, ed. R. Denemark, 5492-5514. Chichester, UK: Wiley-Blackwell.
Walker, S., A. Malici. 2011. US. presidents and foreign policy mistakes. Stanford, CA: Stanford University Press.
Walker, S., A. Malici, M. Schafer. 2011. Rethinking foreign policy analysis. New York, NY: Routledge.
Walker, S., M. Schafer, M. Young. 1998. Systematic procedures for operational code analysis. International Studies Quarterly 42: 175-190.
Walker, S., M. Schafer, J. Beieler. 2016. Belief systems and foreign policy roles: Role contestation in U.S. foreign policy decisions. In Domestic role contestation, foreign policy, and international relations, eds. C. Cantir, and J. Kaarbo, 122-139. New York, NY: Routledge.
Weber, M. 1999. Die 'objektivitat' sozialwissenschaftlicher un sozialpolitischer erk- enntnis. In Gesammmelte aufsatze zur wiissenschaftslehre, ed. E. Flitner, 146-214. Potsdam: Internet-Ausgabe, www.uni-potsdam.de/u/paed/Flitner/FlitnerAVeber/.
Winter, D. 2003. Measuring the motives of political actors at a distance. In The psychological assessment of political leaders, ed. J. Post, 153-177. Ann Arbor, MI: University of Michigan Press.