----------------- HES POSTING ----------------- Robert, I agree fully with the line of thought you have expressed. Although I did not write that the perfect knowledge assumption is merely a simplifying assumption, I agree with Brewer and you that it functions in this way also. Indeed, I would go further and discuss the language that is used to express the perfect knowledge assumption after Knight. If the simplifying assumption is stated in mathematical form, a whole lot of unnecessary effort will go into employing an industry of experts to teach the language and to test competency. This has far reaching effects, due to the thought processes people use in deciding whether to study economics and the costs of supplying credentials. Most students of economics give no deeper thought to why they are learning "economics" than their bank accounts and other immediate interests. For them the purpose of attending the university and majoring in economics is to pass exams and to write acceptable theses. On the supply side, the cost of making and evaluating exams on whether the simplifying assumptions are used wisely is lowest if those exams are about the mathematics. Mathematical theses are also easier to evaluate. As a consequence, the industry of teaching economists becomes biased in the direction of mathematical language learning. This leads to a number of side effects. The two major ones, I think, are these. First, the teachers and their supervisors give the use of mathematics in communicating ideas to peers (i.e., in economics journals) undue support. Second, the use of mathematics as a language makes it easier to accept the counter-intuitive idea that the distinctly human choices of the past can be captured by the statistical analysis of numerical data. This, in turn, promotes a spin- off industry of econometrics and the accompanying set of journals devoted to this field. Meanwhile the issue of exactly why we use the simplifying assumption of perfect knowledge in the first place gets shrouded by a haze of mathematical and econometrics subfields. Even most of those who intuitively recognize the sterility of these fields are unable to identify what went wrong. They find it sufficient to merely state the obvious fact that the assumptions made in mathematical and econometric models are unrealistic. Moreover, they typically express their more realistic assumptions in mathematical form! The question asked by historians of economics about what constitutes "neoclassical economics" has been influenced by the historical development of professional economics since the mathematical language began to achieve an important position in economic education. Some HESers today associate neoclassical economics entirely with the mathematical expression of various ideas and the econometric models used to test hypotheses. Others, like myself, conceive also of a deeper neoclassical economics which has been crowded into a relatively small corner of the profession but which appears to be undergoing a revival of sorts. To get back to my point in the opening paragraph, the use of the perfect knowledge assumption as a counterfactual is, in my view, quite different from its use as a simplifying assumption. I think that my message stressed its use as a counterfactual, while Brewer's message stressed its function as a simplifying assumption. In my view, both uses are important in our efforts to understand economic interaction. Pat Gunning, Sultan Qaboos University, Oman ------------ FOOTER TO HES POSTING ------------ For information, send the message "info HES" to [log in to unmask]