Smith and Walras: Two Theories of Science

Abstract

Despite appearances, Walras's theory of science is very different from Smith's. Walras's point of departure is Plato, of whom Smith is sharply critical. Walras sees economics as a wholy deductive system whose assumptions need not be true and whose propositions do not have to be confirmed by data before being applied. Smith stresses empirical testing, although he recognizes that it can never completely free our knowledge of the influence of conventions whose origin is in the mind itself. As a result of these differences, the kind of statement regarded as a theorem by Walras need not be regarded so by Smith. Smith et Walras: deux theories de la science. Malgre les apparences, la theorie de la science de Walras differe considerablement de celle de Smith. Ia perspective walrasienne prend ses racines dans Platon que Smith pour sa part critique vertement. Walras considere la science economique comme un systeme entierement deductif erige sur des postulats qui n'ont pas a etre vrais et compose de propositions qui n'ont pas a etre confirmees par l'observation avant de servir. Smith met l'accent sur la verification empirique meme s'il reconnalt que la connaissance ne peul jamais se liberer completement de l'influence des conventions qui ont leur origine dans 1'esprit lui-meme. A cause de ces diffe6rences, la sorte de proposition consideree comme tin theoreme par Walras n'est pas necessairement tenue pour tel par Smith. The recent two hundredth anniversary of the Wealth of Nations, coinciding as it did with the centenary of the Elements of Pure Economics, prompted economists to ask anew the old question of what Adam Smith and Leon Walras had in common and where they differed. Of the many facets of such an inquiry into the diverse origins of modern economics, this article is concerned with only one: the relationship between the theory of science proThe first draft of this article was prepared as a part of my comments on William Jaffe's insightful and thought-provoking paper, presented during the Tenth Annual Meeting of the Canadian Economics Association in Quebec City, and subsequently published in this JOURNAL. In giving the article its present form, I draw upon comments from the chairman of the session, Vincent W. Bladen, and from Professor Jaffe himself. For further suggestions and advice, I am indebted to my colleagues Samuel Hollander and David Nowlan. Any remaining errors and insufficiences are, of course, my own. Canadian Journal of Economics/Revue canadienne d'Economique, XI, no. 3 August/aofit 1978. Printed in Canada/Imprim6 au Canada. 0008-4085/78/0000-0387 $01.50/!1978 Canadian Economics Association This content downloaded from 207.46.13.64 on Sun, 04 Sep 2016 04:33:35 UTC All use subject to http://about.jstor.org/terms