Empirically Trivial Theories and Inductive Systematization

Abstract

In recent discussions about the role of theoretical terms in inductive systematization, a claim has frequently been repeated to the effect that a theory having no non-tautological deductive observational consequences cannot achieve any inductive systematization among observational statements. This claim has been made by Bohnert (1968, p. 280), Hooker (1968, p. 157–158), Stegmiiller (1970, pp. 423, 428–429), and Cornman (1973), for the purpose of showing that the examples given by Hempel (1958, pp. 214–215), (1963, pp. 700–701), and Scheffler (1963, pp. 218–222) do not suffice to prove the indispensability of theoretical terms for inductive systematization. The purpose of this note is to criticize this claim in the form it is made by Stegmiiller, who has given, as far as I know, the only explicit argument for it. Elsewhere (see Niiniluoto, 1973) I have tried to discuss more generally the notoriously vague notion of ‘inductive systematization achieved by a theory’ and some main problems connected with it.