Edited 21 August, 2003
On the social science wars
Daedalus, winter 2004
In the spring of 2003, as the founding editor of Perspectives on Politics, I helped to launch the first new journal sponsored by the American Political Science Association (APSA) in over a century. The new journal grew out of general disaffection that had been floating around the discipline for years. In political science (as in other fields in the social sciences from economics to anthropology) there has been a kind of simmering war being waged between quantitative and qualitative researchers – and the new journal was, in part, meant to heal the rift.
The APSA acknowledged dissatisfaction after analyzing a 1998 survey of members and ex-members. Over two-fifths of current members who responded, and half of former members who responded, criticized the Association’s flagship journal, the American Political Science Review (APSR); it headed the list of APSA activities with which respondents were unhappy. Respondents wrote that the APSR “covers one small corner of the discipline,” that it is “virtually useless for my teaching preparations and research specializations,” or that it is not “reflective of the range of research methods and approaches in the discipline.” The Association’s report concluded that many members saw the Review as “too narrow, too specialized and methodological, and too removed from politics.”
In short, some of the most prominent members of the discipline, as judged by their appearance in its most selective and prestigious journal, were developing a new type of “science” that left other members of the discipline angry and unimpressed.
Several years later the Association’s governing Council approved the creation of a new journal, and eventually selected me to serve as its founding editor.
The new journal’s mission was to publish “integrative essays” that would be less specialized than normal research articles and that might “appl[y]… political science to questions of public policy.” The committee charged with implementing the Council’s directive added additional mandates: the new journal should also include “state of the discipline type essays, book reviews, reviews of literature in other disciplines with relevance to political science, conceptual and methodological essays, [and] a policy forum for debates on current policy issues, among other new materials” [italics added]. Those other new materials might, for example, include “articles similar to those found in Science magazine.” The implementing committee concluded, in something of an understatement, that Perspectives on Politics “should publish a very wide range of scholarship” – and by implication should foster a broader conception of “science” in political science than the APSR had been doing..
Although I was not involved in shaping the journal’s mandate or design, I share its originators’ goals. Like similar journals in other social science disciplines (for example, the Journal of Economic Perspectives, and Context, a new sociology journal), Perspectives on Politics is a response, in part, to a widespread perception that the drive to be “scientific” risks distorting our purposes – and that too many scholars are moving into narrower and narrower specializations, divorced from the concerns of non-specialists.
The respective merits of breadth and depth are a complicated and old issue. To some, specialization is an essential virtue in the face of a wide range of worthy topics and the deepening of knowledge about each; it is evidence of the maturation of the social sciences. Only by specializing does an individual have a chance to develop sufficient substantive and methodological knowledge to develop sharp hypotheses, test them definitively against alternatives, and pinpoint their contribution to theoretical frameworks. Science consists in the cumulation of small advances built on previous small advances, so that whole becomes a good deal greater than sum of its parts.
There is no intrinsic substantive content to this claim about scientific advancement; it can hold for the study of canonical political philosophers, for a particular area of the world, for explaining how a particular institution conducts its business, or for the revelation of hidden discrimination against disadvantaged groups or marginalized populations. It is also not intrinsically opposed to engagement with political or policy concerns; small bits of cumulative, specialized knowledge may be just as important for determining how to combat terrorism or reform tax law as for understanding the median voter theorem in legislative decision-making.
An alternative framework sees increased specialization as insufficient to, or even the downfall of, the social sciences. In this view, the compilation of small, cumulative findings is boring to read and teach, and narrows one’s intellectual capacities. True science consists in understanding how the small parts of the world fit together into a larger pattern, and investigating why one set of changes occurs rather than other imaginable ones. Absent a broad vantage point, the ability to consider a problem from multiple perspectives, and the recognition of one’s own inevitably partial and biased conceptual lenses, one cannot determine how and why the world works as it does. True science also entails knowing when to give up on a given framework rather than to continue trying to refine it; one cannot imagine alternative paradigms without breadth of vision.
Here too there is no intrinsic link between the call for integrative breadth and any particular topic of study, normative stance, or degree of policy relevance. And here too, the indeterminate signifier of “science” is given a content intended to confer status on a particular set of practices.
Finally, there are many political scientists who do not aspire to the mantel of “science;” they see their enterprise as closer to the humanities or history, in that they seek to give meaning to a phenomenon rather than to provide a causal explanation for it. They too are involved in the methods wars that are roiling the APSA, and are an important constituency for Perspectives on Politics.
Of course, there is no need to insist that the study of politics be a “science,” just as there is no logical reason to pit breadth against depth: they are separate rather than conflicting values in science. But any reader, writer, teacher, professional association’s budget, or journal editor must make trade-offs at the margins. Perspectives comes down on the side of integrative breadth rather than cumulative depth, but less from a deep commitment about the right way to conduct science than from a perception of the need for a counterweight to most high-status academic journals.
I pointed out earlier that all social sciences are facing this trade-off between breadth and depth in their publications, teaching, and graduate training. Most have begun a journal with a mission similar to Perspectives; in fact, political science was a bit slow on the uptake so we have been able to learn from the experiences of the other disciplines. The underlying conflict over the changing and contested meaning of “science” has, however, taken a different form in each of the four disciplines I know best.
The nastiest fights in political science at present are over methodological frameworks – not over competing political values or appropriate hierarchies of power. That might seem surprising in a discipline that has at its core the analysis of the exercise of power, but it perhaps stems from the very breadth of the discipline. Political science encompasses the canon of political philosophy from Thucydides through Hannah Arendt, and also moves through qualitative research via case studies and historical or institutional analysis to highly technical quantitative analysis and formal reasoning. No other social science covers such a wide epistemological range so deeply; therefore it perhaps makes sense that we argue over how to do our work more than over what our work is about.
The discipline of sociology, in contrast, has largely avoided methods wars, but at the cost of arguably even more painful disputes. In recent years, battles among sociologists have revolved around the roles of race and gender in determining professional standing, and the presumed association of race and gender with differing definitions of science. In the late 1990s, for example, the American Sociological Association became embroiled in a bitter dispute over the editorship of the American Sociological Review. The nominations committee proposed an African American candidate with a slate of editorial board members who collectively presented a strong emphasis on qualitative and/or postmodernist research, sustained attention to issues of hierarchy and stratification, and a commitment to the view that the pursuit of scientific objectivity and precision was a mistaken way to understand the social world. The governing council of the ASA chose a different set of candidates (one of whom was also African American), amid vehement accusations of racism both of specific named individuals, and the ASA as an organization. There have been equivalent battles over gender issues in the ASA, incorporating the same underlying struggle over the meaning of science and the goals of social science analyses.
Economists are much less likely to debate methods for conducting research or to challenge the ascriptive characteristics of researchers; their central fight is over the legitimacy of critiques of neoclassical orthodoxy. Dissident Europeans have begun a movement for “post-autistic economics,” and in the United States a tiny tempest in a teapot at Harvard University has been deemed significant enough to be reported in the Weekly Standard and the Economist. A two-semester course of micro- and macro-economics is taught by a senior member of the department, and is the mandatory gateway course for all students who seek to study economics. It is, everyone agrees, totally conventional; that is its purpose. A chaired professor in the Economics department proposed an alternative gateway course in microeconomics that would teach the same textbook, but then explicitly analyze the assumptions underlying neoclassical economics; the department voted almost unanimously not to permit it except as an elective. It is hard to conceive of a sociology or political science department collectively deciding that all students must take one particular two-semester course which is always taught in the same way before taking any other course in the discipline. Science is clearly defined by mainstream economists; the question in that discipline is whether the mainstream can be overturned rather than how broadly it is to be defined.
The discipline of anthropology has, like political science, engaged in disputes over methodology, but in this case the dominant position rejects the validity of positivism and conventional understandings of science. For several decades, the most prominent anthropologists have argued that researchers need to attend ever more to the subjectivity of the researcher, the power dynamics and subtle interplays of communication and emotion between subjects and researchers, and the broad context within which any research endeavor takes place. Good “science” in anthropology, in this view, is a move away from the misguided search for “objective truth,” precisely defined and carefully tested causal hypotheses, and the cumulation of small findings; it is a move toward placing the researcher’s inevitable biases and partialities at the center of the research process. As in all disciplines, there is disagreement among anthropologists – but there, the backlash against the hegemonic paradigm is swinging in the opposite direction from the equivalent backlash in political science.
In the end, I am reasonably optimistic about the upshot of the social science wars, at least in political science. The APSR is becoming more eclectic in its definitions of good science and other journals may follow its lead. Perspectives on Politics is opening channels for communication across subfields and rival frameworks. And the best graduate students and junior faculty are learning how combine diverse epistemologies and modes of analysis in new and flexible ways – and that is good news for the future of political science.
Jennifer Hochschild, a Fellow of the Academy since 1996, is Professor of Government at Harvard University, with a joint appointment in Afro-American Studies, and also the founding editor of Perspectives in Politics. Her books include The American Dream and the Public Schools (with Nathan Scovronik, 2003), Facing Up to the American Dream: Race, Class, and the Soul of the Nation (1995), and What's Fair: American Beliefs about Distributive Justice (1981.)