Rationality, Norms, and the Sociological Reconstruction of Economic Theory

Abstract

This paper characterizes the interface of economic and sociological theory, focusing on Parsons ’ analysis of the “problem of order” and on Weber ’s and Marx ’ analyses of the emergence of machine capitalism. I demonstrate the consistency of both perfect-information, Arrow-Debreu models and models of competitive markets under machine capitalism with the assumptions of neoclassical theory, most especially the contention that all actors always act rationally and maximize against constraints. I also demonstrate that imperfect-information models and models of capitalist production that are not competitive generate anomalies that manifest certain fallacies within neoclassical theory, especially the contention that we can explain social action successfully if we assume that all actors are instrumentally rational all of the time.

Methodologically, I show that my analyses have consequences for an evaluation of Friedman ’s contention that the empirical veracity of the assumption of rationality is irrelevant because the predictions that derive from models where rationality is assumed are empirically warranted. These empirical predictions presume other attributes of the “real world,” like competitive constraints, that are not always present, and thus the predictions are often false. We will conclude that the veracity of the assumptions in our models is often crucial to the model’s empirical warrantability. Even when the assumption of rationality is justified, it should not be taken a priori; it requires explanation. Actors may conform to a variety of normative orientations, and since the conceptualization of these alternative, positively stated normative orientations is impossible within utilitarian theory, this suggests that neoclassical theory must be reconstructed sociologically.