D. Hands

This paper examines methodological issues raised by revealed preference theory in economics: particularly contemporary revealed preference theory. The paper has three goals. First, to make the case that revealed preference theory is a broad research program in choice theory—not a single theory—and understanding this diversity is essential to any methodological analysis of the program. Second, to explore some of the existing criticisms of revealed preference theory in a way that emphasizes how the effectiveness of the critique depends on the particular version of revealed preference under consideration. Finally, three additional criticisms are presented that are aimed specifically at contemporary revealed preference theory

InStabilizing Dynamics Roy Weintraub provides a history of stability theory from the work of Hicks and Samuelson in the late 1930s to the Gale and Scarf counterexamples in the 1960s. Unlike his earlier work in the history of general equilibrium theory this recent contribution is not an attempt to fit the Walrasian program into the narrow framework of some particular philosophy of natural science. Rather, the theme inStabilizing Dynamicsis broadly social constructivist. Simply put, the constructivist view of science is “that scientific knowledge itself is constructed socially, in communities of scientists: Knowledge is constructed, not found”.

Reflection without Rules offers a comprehensive, pointed exploration of the methodological tradition in economics and the breakdown of the received view within the philosophy of science. Professor Hands investigates economists' use of naturalistic and sociological paradigms to model economic phenomena and assesses the roles of pragmatism, discourse, and situatedness in discussions of economic practice before turning to a systematic exploration of more recent developments in economic methodology. The treatment emphasizes the changes taking place in science theory and its relationship to the movement away from a rules-based view of economic methodology. The work will be of interest to all economists concerned with methodological issues as well as philosophers and others studying the relationships between economics and contemporary science theory.

Abstract: Don Ross’ Economic Theory and Cognitive Science (2005) provides an elaborate philosophical defense of neoclassical economics. He argues that the central features of neoclassical theory are associated with what he calls the Robbins-Samuelson argument pattern and that it can be reconciled with recent developments in experimental and behavioral economics, as well as contemporary cognitive science. This paper argues that Ross’ Robbins-Samuelson argument pattern is not in the work of either Robbins or Samuelson and in many ways is in conflict with their own versions, and defenses, of neoclassical theory.

Discussions of Karl Popper's falsificationist philosophy of science appear regularly in the recent literature on economic methodology. In this literature, there seem to be two fundamental points of agreement about Popper. First, most economists take Popper's falsificationist method of bold conjecture and severe test to be the correct characterization of scientific conduct in the physical sciences. Second, most economists admit that economic theory fails miserably when judged by these same falsificationist standards. As Latsis states, “the development of economic analysis would look a dismal affair through falsificationist spectacles.”.

This book brings together ten previously published essays on the philosophy of economics and economic methodology. The general theme is the application of Karl Popper's philosophy of science to economics -- not only by Popper himself but also by other members of the "Popperian school." There are three major issues that surface repeatedly: the applicability of Popper's falsificationist philosophy of science; the applicability of I. Lakatos's "methodology of scientific research programs" to economics; and the question of Popper's "situational analysis" approach to social science.