ABSTRACT
In taking stock of the interaction between mathematical theory and the practice of the social and policy sciences, it becomes all too clear that our methods are, in the main, inadequate to the problems they confront. It is a sad fact that the premier form of mathematics employed in these areas is statistics, a field not at all suited to the expression of substantive theory. As all who have used statistical theory can attest (and as all who avoid it have always suspected), statistics places extremely severe restrictions on the kinds of hypotheses that can be stated and tested. On the other hand, even when nonstatistical mathematical theory has been applied in the social and policy sciences the results have generally been mediocre.
