----------------- HES POSTING ----------------- 
 
Dear colleagues, 
 
We should remind ourselves that neoclassical economics does not  
always require the assumption of perfect knowledge. Although the  
neoclassical theory of, for example, perfect competition reproduced  
in textbooks does require perfect knowledge (of the relevant  
variables), neoclassical economics has for quite some time now  
assumed in many other contexts probabilistic knowledge and, as  
someone said on this list a couple of days ago, asymmetric  
information (to a lesser extent, as some neoclassicals stick to the  
representative agent). Even a notion of uncertainty (sometimes  
called Knightian uncertainty, sometimes called ambiguity) that  
goes beyond Knightian risk or uncertainty in standard Subjective  
Expected Utility theory is beginning to make its way into the  
mainstream.   
 
In reply to Michael Perelman, I'd say that formalization does not  
require perfect knowledge either, at least as I understand these  
terms, which is not necessarily the same way Michael does. On  
the other hand, it is true that some specific varieties of  
formalization, particularly some axiomatic approaches to behavior  
in neoclassical economics, impose quite restrictive assumptions  
regarding knowledge - but not perfect knowledge.   
 
Of course, all this depends on what one means by perfect  
knowledge. In the sense I use the expression, perfect knowledge is  
distinct from probabilistic knowledge and from uncertain knowledge  
of probabilities (the latter is characteristic of what some people  
have called termed Knightian uncertainty, ambiguity or just  
uncertainty). These other types of knowledge have been suitable to  
an axiomatic approach.   
 
Cheers, 
David Dequech 
 
------------ FOOTER TO HES POSTING ------------ 
For information, send the message "info HES" to [log in to unmask]