Content area
Full text
Charles Ragin, Fuzzy-Set Social Science (Chicago: Chicago University Press, 2000)
Perhaps the most pressing methodological issue in comparative politics today is how to reconcile small-N and large-N research programs. A great deal of work, and quite a lot of worry, has been expended on this point in the past several years. Regrettably, most of the work-and, I suspect, all of the worryhas been on the small-N side. Evidently, large-N researchers do not have much interest in how their studies and their methods accord with the approaches that have characterized comparative politics up until now. Or perhaps they have little to say on these questions.
Charles Ragin has a great deal to say about these questions, and he is wellpositioned to do so. Fuzzy-Set Social Science is the latest in a long and impressive line of books and articles that explores a middle ground between "case-study" and "variable-centered" approaches. Indeed, Ragin's position in this debate is different from most. This is visually demonstrated by a schematic diagram which plots the number of studies (on the Y axis) within the subfields of comparative sociology and comparative politics against the number of cases employed in each study (on the X axis) (p. 25). The heuristic figure shows a stark, and dramatic, U-shaped curve. In other words, virtually all the work being conducted in these fields has a small or large N. There is no middle ground. Surely, this constitutes a serious bias in the way social scientists apprehend and analyze the world.
Ragin aims to occupy this no-man's land, and sets forth an ambitious agenda for doing so. The book consists of two large arguments-one about concept formation and the other, carried over from previous work, about case analysis. Since the latter, referred to as qualitative comparative analysis, or QCA, has been published previously (see Ragin 1987; Drass and Ragin 1992), I shall simply quote the author's general explanation: "In a nutshell," Ragin says, "QCA provides analytic tools for comparing cases as configurations of set memberships and for elucidating their patterned similarities and differences" (p. 120). Rather than being captive to causes that are linear and additive, QCA allows us to explore causal complexity (albeit with a smaller sample).
The "fuzzy-set" idea, which gives the book its title,...