Econ v. Sociology: Good Question

Herb Gintis doesn't soft-pedal his criticism, but he still wants to bridge the gap between economics and sociology:

[Hans] Joas and [Jens] Beckert's treatment of action theory is a showcase in misunderstanding the rational actor model, but in a way which is shared by the overwhelming majority of sociologists. Joas and Beckert begin by characterizing the rational actor model as claiming that "actors' decisions can be understood from their motivation to optimize their utility." (p.269) Even the most elementary exposition of the model, by contrast, stresses that if individuals have consistent preferences, we can represent their behavior as maximization, but this is not their "motivation" any more than the fact that we can represent the motion of light waves by minimizing a Hamiltonian means that "light waves want to travel minimal paths." So, is simply absurd to think that economists believe it is rational to be "motivated to optimize." Joas and Beckert then assert that "the debate on action in sociology tends to focus primarily on rational choice theory on the one hand and on normative theories of action on the other... The competition between economics and sociology is largely founded on radically opposed action theories. The most influential sociological alternative to rational actor theory has been the normative model of action." (p. 270-271) Now, I do not want to deny that this is the case, but I do want to deny, categorically, that there is any conflict between rational action and normative action. There is nothing in the rational actor model that precludes normative elements in the actor's preference function. Nor is there anything in the normative model that precludes individuals from making welfare-improving trade-offs between normative principles (e.g., fairness, honesty, trustworthiness) and self-regarding concerns (material goods and services, leisure, prestige). Indeed, a raft of experiments in behavioral game theory exhibit the consonance of self-regarding interests, other-regarding interests, and character virtues in the preferences of rational agents. The failure of sociologists to understand this point is probably the chief reason core sociological theory has never had a chance to develop (economics has its own problems, of course, but that is a matter better discussed elsewhere).

Siegwart Lindenberg's contribution, "Social Rationality versus Rational Egoism," shows exactly how the economist's rational actor model can be enriched to serve sociological purposes without abandoning the analytical clarity and power of economic theory. I tend to stress that the rational actor model requires no more that preference consistency, and refer for support to the success of the model in game-theoretic models of animal behavior. Now, animals are hardly rational according to most meanings of the term "rational," but they do tend to satisfy preference consistency, and hence their strategic interactions can be cogently modeled as rational action---even if the creatures are pond scum or dung beetles. However, sociologists demand a richer concept of rationality, and Lindenberg offers an expanded concept of rationality, one that conserves yet goes beyond what economists care about. He argues of human individuals that (a) they are resourceful and goal-motivated; (b) they are constrained and confronted with scarcity of time, energy, and resources; (c) they form expectations and learn; (d) they differentially value distinct states of the world; (e) they are motivated to achieve; (f) in novel situations, they search for the proper frame for the event in their reservoir of know event types. (p. 636) Of course, these characteristics are plausible and in no way conflict with the rational actor model.

So, why then does the chasm between sociology and economics persist? Good question.

From Gintis' Amazon review of Jonathan H. Turner's Handbook of Sociological Theory (link).