Thinking Big

Arguing Comparative Politics. By Alfred Stepan. Oxford University Press, 2001. 369 pp.

Most readers of the Journal of Democracy will be familiar with some aspects of Alfred Stepan's academic output, but they may not appreciate the breadth of his scholarship or the way in which his many contributions fit together. Arguing Comparative Politics provides a synoptic overview of the development of his ideas, with a vivid introduction that adds a personal dimension and links together a succession of articles and chapters published over the last 30 years. Instead of reviewing each of the chapters in turn, it may be more useful to react to the overarching themes raised by the collection as a whole.

The volume opens with this declaration of faith: "All aspects of research are, in a fundamental sense, problem driven" (p.1). As a description of current research training in British (and, I suspect, U.S.) political science departments, this assertion rests on shaky empirical foundations. As a more personal affirmation of beliefs about both why comparative politics is worth pursuing and what constitutes a valid contribution to the enterprise, this statement sums up the outlook not only of the author but also of an entire cohort of comparativists who built their careers and set their intellectual agendas during the later stages of the Cold War, a context that shaped much of the research agenda. By contrast, the "problems" that engage the research energies of the younger generation are more technical (and perhaps more depoliticized even) than those studied by their predecessors, and almost certainly the career incentives and constraints have changed. The conflation of fact and value in Stepan's opening declaration is not something that a newly minted assistant professor would risk, at least not unless it could be backed up with a large data set.

In the same vein, while Stepan feels free to generalize about the importance of "the state" or the imprecision of the concept of "civil society," younger scholars with reputations still to build are understandably fearful of painting on such vast canvases. Even the very notion of "arguing" comparative politics may seem a little alien and polemical to many contemporary practitioners. "Carefully testing a comparative hypothesis" may be a safer bet, even if the result is only to confirm (on a "scientific" basis) some relatively unsurprising or even obvious conclusion. How to solve the problem has become more critical than whether a problem is worth solving.

Stepan's problems may be important, but they are often loosely specified and not very tractable. This may be why they are "argued" in the lawyer's sense of making a case and marshaling good evidence, rather than "proven" like a mathematical theorem or "demonstrated" as in a laboratory experiment. In fact, whereas a lawyer would at least try to establish a specific assertion beyond a reasonable doubt, Stepan's arguments mostly involve increasing the plausibility of some quite broad and even subjective interpretive claims. At his best, Stepan widens our horizon of possibilities and challenges conventional wisdom. That really is "arguing" comparative politics, even if by some standards it falls short of political "science."

Consider the following selections from his argumentative propositions:

• Contrary to Samuel P. Huntington's thesis, the professionalization of a country's military officers can just as well result in their politicization and the expansion of their role rather than leading them to concentrate solely on external defense.
• The hardliner-versus-softliner distinction underlying the democratic "transitions" literature requires a kind of differentiation that is only possible in authoritarian regimes in the strict sense—not under sultanism, totalitarianism, or early post-totalitarianism.
• Instead of thinking abstractly about the logic of liberalization versus democratization, it may be more productive to think historically about the numerous alternative "paths" that have led to similar outcomes in various settings.
• Instead of reasoning deductively about federalism in general, we may more closely approach the comparative evidence by distinguishing between federalism as "coming together" and federalism as "holding together."
• Again contrary to Huntington, none...