Some Problems of Formal Methodology of Science

Most of empirical theories can be applied in different areas, to different phenomena, which, in general, cannot be combined into a single model structure of the sort examined in metamathematics. What we can at best achieve is to represent the different phenomena being possible referents of a theory by different model structures. This observation seems to underlie the well known Suppes (cf. [9], § 12, 2) account of empirical theories (shared by Simon, Sneed and others) according to which an empirical theory should be given in the form of a set-theoretic predicate that defines a class of model structures. As far as I am concerned, I shall stick to the traditional point of view and dealing with a theory I shall assume that it is defined by its components such as its language, theorems (laws), methods of proof, methods of gathering empirical data, and criteria of distinguishing the intended referents of the theory from those phenomena which do not belong to its range.