================= HES POSTING ================= 
 
Robert Leeson wrote: 
 
> By formalist revolution, I mean the econometrics movement from the 1930s 
> and the simultaneous revival of interest in the Walrasian approach.  Do 
> these two developments constitute a revolution?  I think they do, in that 
> the leading players self-consciously sought to effect a revolution in 
> research methods. 
> 
> Tinbergen et al sought to uncover the structual equations of capitalism; so 
> as to provide a guide for policy makers to alter key magnitudes in order to 
> mitigate the business cycle.  Even when the immediate post-war econometric 
> forecasts were disastourously wrong, they were still perceived by Klein 
> (JPE 1946, 1947) to be vastly superior to the unrigourous "armchair" 
> statistics of those (eg the NBER) the revolutionaries sought to replace. 
> Thus Keynesian economics rose and fell to the accompaniment of econometric 
> failure. 
> 
> The Walrasians were equally contemptuous of Marshallian economics; 
> Marshall's fuzziness had "paralised the best brains in economics for 
> decades" according to Samuelson.  For Tobin, Keynesian economics was "a 
> short-cut GE approach". 
> 
> The leading opponents of the formalist revolution were J. M. Keynes and 
> Milton Friedman. 
 
I have, per understandings with other members of the HES-LIST advisory 
board, refrained from commenting on and participating in the discussion 
lest it appear to be a paper, discussion, reply to critics, rejoinder 
exercise. Leeson's remarks on formalism, however, are both sufficiently off 
the original editorial subject, and sufficiently muddled, to require reply. 
 
First, "revolution" is a word not to be taken lightly. It represents a 
large literature in the history of science, and more recently in the 
history of mathematics. Leeson's use of it in a forum for historians of 
economics needs to attend to its baggage. Unlike humpty dumpty, Leeson 
cannot make words mean what he wishes. Similarly for "formalism". The 
recent Golland paper in JHET on the subject should alert him to the 
problems with the word. Most economists when writing or speaking 
colloquially use the word to mean, variously, "mathematical", 
"axiomatized", "rigorous", "abstract", "non-empirical", et hoc genus omne. 
The idea of formalization however has meant different things, and has had 
different interpretations, over the 20th century, and today it has 
different meanings in the discipline of philosophy from those in 
metamathematics, or mathematics, or the history of matematics, or the 
history of science, or economics, or the history of economics, etc. 
Whatever it means in one or another disipline, it does not mean 
"econometrics" which in most historical reconstructions is antithetical to 
the formalist line in mathematics. To call Samuelson either a Walrasian, or 
a formalist, is inappropriate;  and it goes against all evidence to argue 
that Tinbergen had connections to formalist argumentation. 
 
The connection with the original editorial becomes obvious. Saying anything 
one wishes about the history of economics to support a particular view of 
present-day economics prevents economists and historians from taking 
seriously the scholarly integrity of the subdiscipline of the history of 
economics. 
 
E. Roy Weintraub, Professor of Economics 
Director, Center for Social and Historical Studies of Science 
Duke University, Box 90097 
Durham, North Carolina 27708-0097 
 
Phone and voicemail: (919) 660-1838 
Fax: (919) 684-8974 
E-mail: [log in to unmask] 
URL: http://www.econ.duke.edu/~erw/erw.homepage.html 
 
============ FOOTER TO HES POSTING ============ 
For information, send the message "info HES" to [log in to unmask]