In the 1940s political scientists sought a new approach to studying and understanding politics. They began to apply advances in the behavioral sciences to political inquiry, shifting their research from institutions and values to the role of the individual political actor. In particular, this change in perspective closely aligned with the behavioral revolution in the psychological sciences initiated by John B. Watson ( 1913
This new focus reflected a fundamental change in how political scientists understand and study politics. Political inquiry on institutions, such as legislative bodies or government agencies, was viewed as ambiguous and subjective. Behavioralism turned away from the study of institutional values, such as the role of capitalism or democracy. The new discipline focused on rooting out subjectivity and bias by applying the natural sciences to the social sciences, with statistical research a key methodology. Behavioralist research viewed individuals as actors and did not emphasize their institutional context or the values they held. This reflected a deviation from the theoretical institutionalism of eighteenth- and nineteenth-century political inquiry, which examined the structure and operations of institutions, whether a governing body or firm. The institutionalists engaged in expository and rationalist research on politics, in which the actions of the collective institution were the unit of analysis. The behavioralists, by contrast, studied the individual isolated from the actions and values of the institutions.
No single research methodology, however, was prevalent among the behavioralists. This lack of uniformity was due to the difficulty political scientists faced in applying models, theories, and methods of the natural sciences. Also, numerous questions about how a political scientist could study politics without values remained. That such ambiguity and tension existed in the 1940s reflects the inability of behavioralism to overcome the methods and values of institutionalism. As a result some political scientists lost faith in the ability of statistical models to explain the reality of the individual political actor. This skepticism regarding the ability of political scientists to measure the individual objectively and without values was present in the common critiques of behavioralism.
Heinz Eulau' s edited volume Behavioralism in Political Science ( 1969 ) was the central work to present behavioral studies of American government at midcentury. Still, the articles collected in this volume echo the lasting influence of institutionalism on political science and the inability of behavioralism to completely eradicate values and institutions from research on politics. Eulau points to the “intellectual battle lines” between rational and statistical inquiry in political science and between individual behavioral and opposing institutional analysis, as well as tensions regarding the role and acceptability of values ( 1969, 3 ). That Eulau considers behavioral methodology, particularly statistical analysis, debatable relative to the values of institutional analysis is indicative of the subjectivity of the behavioralist revolution. Specifically, little consensus could be found among the behavioralists on the various positions they espoused. Concerns regarding the validity of methodology of both the behavioralist and institutional researcher prove to be interrelated in nature, given that the political researcher dictates both the methodology and the acceptance, or lack thereof, of values when studying the individual political actor.
David Easton, in an article originally published in 1953, also notes that the application of behavioralist theories to political inquiry could be “disturbing and disappointing, if not in absolute results at least in terms of what is possible” ( 1969, 32 ). Echoing other skeptics, Easton lists a range of potential explanations for this negative judgment of the field, including the paucity of existing political methodology. He also discusses the “inching pace” of research, which challenges the relevance of prior rational and institutional inquiry considered “traditionalist” ( Easton 1969, 33 ). These and other concerns regarding the state of political science research at the time not only reflected the state of the field but also held implications for the future of the social sciences in the context of the statistical emphasis of scientific behavioralism. Indeed the debate between institutionalism and behavioralism that first arose in the 1940s was not exclusive to the discipline of political science but became significant in all the social sciences.
Consistent with Eulau and Easton, David Truman, in an article originally published in 1955, notes the need for advances in research methodology, calling for better survey measures and better generalizability in quantitative research ( Truman 1969 ). He also delineates developments in behavioral political science that require future investigation: the scope of a “pattern between or among formal institutions,” the nature of descriptive research concerning institutions, and concerns about data collection and research techniques in response to the “limiting assumptions imposed by conventional institutional categories” ( Truman 1969, 53, 56 ). In response to institutional political analysis, Truman establishes the reactionary nature of behavioralism, which is evident in methodological developments that challenged the role of values in political inquiry in addition to questioning whether substantive research conclusions could be reached by including institutional research values.
Truman also acknowledges the skepticism and potential limitations associated with behavioralism. In political science strictly statistical research tends to provoke doubt. Thus he argues that “the salvation of political science lies in concentrating on the dynamic psychology of individuals, in analyzing individual actors in depth” ( Truman 1969, 60
For Easton and Truman, a central concern was political scientists' ability to reach substantive research conclusions using behavioralist research methodology. In 1925, however, Charles Merriam had noted the potential for “another angle of approach” that views “political behavior as one of the essential objects of inquiry” ( Dahl 1969, 70 ). Robert Dahl, in an article first published in 1961, traces behavioralism in the social sciences to a variety of sources, including but not limited to Merriam and the Chicago behavioral school, German scholars arriving in the United States in the 1930s, research related to World War II, the Social Science Research Council, and financial support of research institutions. Nevertheless Dahl maintains a cautious attitude when assessing behavioralism, noting that further research is needed before evaluating the field as too rigid in its emphasis on methodology as opposed to theory. Dahl points to the need for “unity between empirical political studies and a concern for general theory” ( 1969, 89 ). This point is at the heart of much of the debate as expressed by the authors collected in Eulau' s 1969 volume.
In an article originally published in 1957, Leo Strauss ( 1969 ) discusses the role of theory, as opposed to statistical and behavioralist political science, in rational political inquiry. Strauss holds that statistical political research is skeptical toward unscientific political philosophy. Still he notes that “there is a fundamental difference between facts and values, and that only factual judgments are within the competence of science” ( 1969, 98 ). Applying statistical research to political inquiry has its limitations; as a result, Strauss posits the need for considering values in the social sciences. As he succinctly puts it: “a society cannot be defined without reference to its purpose” ( 1969, 103 ). The need for value judgments in political research cannot be overcome by methodological advances of behavioralism. Political philosophy is therefore a discipline to be considered as a context for, and complement to, behavioralist political inquiry.
As an approach to classical political inquiry and institutionalism, behavioralism was both reformist and reactionary. Behavioralist political scientists were unable completely to root out the role of society, institutions, and values in research. To address the divide between older forms of institutionalism and behavioralism of the 1940s, political scientists formulated neo-institutionalism in the 1970s. John Meyer and Brian Rowan ( 1977 ) wrote a foundational article on neo-institutionalism, which was followed by an edited volume by Meyer and Richard Scott ( 1983 ), among other works. Neo-institutionalism attempted to address both the individual and institutions in political science research while embracing empirical evidence. Walter Powell and Paul DiMaggio note that neo-institutionalism maintains “common skepticism toward atomistic accounts of social processes and a common conviction that institutional arrangements and social processes matter” ( 1991, 3 ). Neo-institutionalists do not discount either the individual focus of behavioralists or the institutional focus, but rather seek a synthesis of perspectives to address the divide between institutionalism and behavioralism. By means of this synthesis, they attempted to advance modern political science research.
A critical development in political science in the 1940s, behavioralism challenged existing conceptions of politics as being strictly institutional and emphasized objective inquiry using statistical tools. This approach shaped political science in the ensuing decades to employ objective research via survey techniques and statistical methods such as regression analysis. It led political scientists to realize the inadequacy of studying only the individual while simultaneously incorporating values and institutions in research. It also revealed how subjectivity can enter into seemingly objective research in the design and interpretation of that research. Ultimately, after 1977, behavioralism led to the rise of neo-institutionalism. From that point forward, modern political science research combined the statistical emphasis of behavioralism with the values of institutionalism.
SEE ALSO Institutionalism and Neo-Institutionalism ; Logic of Appropriateness ; Politics .
Dahl, Robert A. “The Behavioral Approach in Political Science: Epitaph for a Monument to a Successful Protest.” In Behavioralism in Political Science, edited by Heinz Eulau, 68–92. New York: Atherton, 1969.
Easton, David. “The Condition of American Political Science.” In Behavioralism in Political Science, edited by Heinz Eulau, 22–37. New York: Atherton, 1969.
Eulau, Heinz, ed. Behavioralism in Political Science. New York: Atherton, 1969.
Meyer, John W., and Brian Rowan. “Institutional Organizations: Formal Structure as Myth and Ceremony.” American Journal of Sociology 83, no. 2 (1977): 340–63.
Meyer, John W., and W. Richard Scott. Organizational Environments: Ritual and Rationality. Beverly Hills, CA: Sage, 1983.
Powell, Walter W., and Paul J. DiMaggio. “Introduction.” In The New Institutionalism in Organizational Analysis, edited by Walter W. Powell and Paul J. DiMaggio, 1–38. Chicago: University of Chicago Press, 1991.
Strauss, Leo. “What Is Political Philosophy? The Problem of Political Philosophy.” In Behavioralism in Political Science, edited by Heinz Eulau, 93–108. New York: Atherton, 1969.
Truman, David B. “The Impact on Political Science of the Revolution in the Behavioral Sciences.” In Behavioralism in Political Science, edited by Heinz Eulau, 38–67. New York: Atherton, 1969.
Watson, John B. “Psychology as a Behavioralist Views It.” Psychological Review 20 (1913): 158–77. Partial repr., Psychological Review 101 (1994): 248–53. http://pages.pomona.edu/~rt004747/lgcs11read/Watson13.pdf .
North Carolina State University