A New Science of Politics?

To hear political scientists Emerson M. S. Niou and Peter C. Ordeshook tell it, theirs is "a discipline mired in imprecision, vagueness, obscure logic, ill-defined constructs, nontestable hypotheses, and ad hoc argument." And it was in reaction to this intellectual flabbiness, they assert in International Security (Fall 1999), that "rational choice" theory—the mathematically oriented approach of which they are leading proponents—has come into academic vogue in recent years. Niou, of Duke University, and Ordeshook, of the California Institute of Technology, were not simply explaining how this came to be: they, and five other contributors, were vigorously defending rational choice against a pointed indictment by Stephen M. Walt, a fellow political scientist at Harvard University’s Kennedy School of Government. His 44-page attack on rational choice theory and its growing influence appears in the spring issue of International Security. The theory at the center of this controversy grows out of economics. It assumes that social and political outcomes are the collective product of individual choices by rational individuals. Rational choice theorists construct mathematical models to represent real-world situations, then use them to show what the only logical outcomes are. The approach has roots in the 1950s, but it lately has become fashionable in academia, Walt notes. "Elite academic departments are now expected to include game theorists and other formal modelers in order to be regarded as ‘up to date,’ graduate students increasingly view the use of formal rational choice models as a prerequisite for professional advancement, and research employing rational choice methods is becoming more widespread throughout the discipline." By one estimate, 40 percent of the published articles in the American Political Science Review now take the rational choice approach.

Unfortunately, Walt maintains, the rational choicers’ elephantine methodological labors have brought forth, in the political science subfield of international security studies, only the tiniest mice of substance. "Formal rational choice theorists have refined or qualified a number of existing ideas, and they have provided formal treatments of a number of familiar issues," he says, but they have produced little in the way of "powerful new theories." Their elaborate formal exercises often yield only "rather trivial" or unoriginal results. A 1991 study, for example, found that "nations generally enter into alliances in the expectation of improving their security position." Another 1991 study, he charges, merely "reinvented the central elements of deterrence theory without improving on it." Little given to empirically testing their propositions against events in the real world, certainly not in any convincing way, rational choicers, says Walt, "have been largely absent from the major international security debates of the past decade."

But Walt does not appreciate the way in which the scientific enterprise must proceed, respond Niou and Ordeshook. He is, they assert, "someone concerned not with science and empirical regularity as those terms need to be understood for the development of cumulative knowledge, but instead with the commentary and informal discussion we find in newspapers and popular journals that has too long appeared under the label ‘political science.’ Such discussion and commentary may be entertaining and even sometimes enlightening, but it remains mere journalism until it can be given the solid scientific grounding that formal theorists pursue."

As for the charge that much of the rational choicers’ work only shows what everybody already knew, Niou and Ordeshook aver that that is necessary: "Showing that a prior conclusion follows logically from some set of initial assumptions is a form of reproducibility that science demands—it tells us that the models in question are not mere fantasy and may not even be fundamentally flawed."

Such "basic science," according to Bruce Bueno de Mesquita and James D. Morrow, a senior fellow and a senior research fellow, respectively, at the Hoover Institution, Stanford University, responding to Walt in International Security, may well take a long time to produce "practical application[s]." But, they say, it has already produced one: "Bueno de Mesquita’s ‘expected utility’ model... [which] predicts the outcome of complex political settings." They claim, citing one Central Intelligence Agency official, that "the U.S. government . . . uses the model to assist with important foreign policy matters." Walt, however, comments that "such assertions should be taken with many grains of salt."

Rational choicers deal with domestic political questions as well. In the 1950s, New Republic (Oct. 25, 1999) senior editor Jonathan Cohn writes, RAND Corporation economist Kenneth Arrow developed his influential Possibility Theorem—for which he won a Nobel Prize— showing the unexpected ways in which a multicandidate election can frustrate the true preferences of voters. Inspired by Arrow’s work, notes Cohn, the late William Riker envisioned a full-blown political science akin to neoclassical economics. At the University of Rochester, he built a department and, eventually, a school of thought on that vision, starting in 1962. Rational choice scholars explore such things as the problem of "free riders" (who enjoy the benefits but don’t share the burdens of membership in political groups) and the behavior of voters. One of their insights is that voters have no obvious reason to vote, since any one voter’s chances of affecting the outcome are so slim. In resorting to "psychic gratification" and other explanations for why millions do in fact vote, argue Yale University political scientists Donald Green and Ian Shapiro in Pathologies of Rational Choice Theory (1994), rational choice theorists undermine their basic assumptions about the "rationality" of human behavior.

The growing controversy in the corridors of political science departments is about more than just the validity of rational choice arguments. What chiefly bothers Walt, for instance, is not the rational choice approach per se (he finds some limited value in it), but rather the "imperialist" tendencies of rational choice scholars. And he is not alone in this complaint. "Critics say it’s the scholars’ strong-arm mentality, not their strong scholarship, that has propelled rational choice this far," writes Cohn. Rational choicers, however, claim that the outstanding quality of their work has led to their rise. "We’re a handful of people," Bueno de Mesquita told Cohn. "The reason it appears to be this dominant thrust is because the clarity of work attracts attention." Certainly, other research traditions in political science are not immune to criticism. In the study of international politics, writes John Lewis Gaddis, the noted historian of the Cold War, in Diplomatic History (Winter 1993), historians and political scientists under the spell of traditional "realist" theory came to assume that "because all nations seek power and influence . . . they did so for equally valid reasons; that in turn led to a kind of ‘moral equivalency’ doctrine in which the behavior of autocracies was thought to be little different from that of democracies." Perhaps the effort to turn political science into a "hard" science is a vain one. In any case, it is clear that the controversy over the growing influence of rational choice theory is important. It will affect, as Walt says, not only political science but what political scientists can contribute to broader public debates on significant issues.