Abstract

It has already been argued by Frazer and Boland (1983) that, interpreted in an instrumentalist fashion, Milton Friedman’s well known and much criticized 1953 paper on “The Methodology of Positive Economics”1 proved to be convergent with Popper's falsificationist philosophy of science2. I think that this comparison is flawed. For one can assuredly contest this interpretation in view of the fact that Popper always opposed any kind of instrumentalist philosophy of science3. It is not even clear that what Friedman has to say on the intricate question of the status of theories and on the function of tests has anything to do with what Popper criticizes under the general category of “instrumentalism”. What I will do here is not so much to directly challenge the Frazer-Boland interpretation as to replace it with a completely new comparison of both methodologies in order to get a new look at Friedman's methodological stance. Of course, one has to focus only on the arguments that really are comparable and leave aside those that could easily be proved to be fundamentally different, if not plainly incompatible. But, as I will try to show, the exercise is worthwhile, for this new comparison will force us to reinterpret, at least partially, what seem to me to be the two central tenets of Friedman's methodological doctrine. Of course, in order to find convergence between Friedman's and Popper's views on economic methodology, we have to read..