Estruturas conceituais e estratégias de investigação: modelos representacionais e instanciais, analogias e correspondência

Abstract

In the present article some strategies of scientific inquiry are discussed which have a pair of features in common: first, they are based on the notions of model, and second, they depend on a counterpoint between two or more structures. Three strategies of this kind are considered. First, I consider modeling in the usual scientific sense. In order to do this, one distinguishes between two notions of model: the logical or instantial one, and the proto-theoretical or representational one. While the differences between both senses of the term are highlighted, an attempt is also made to understand the sense in which the two are related to one another. In particular, in order to investigate the structure-reality relationship of representational modeling, one needs to invoke the structure-structure relationship of logical modeling. The analysis of models contributes also to clarify the relational nature of empirical adequacy between scientific theories and data, through the notion of empirical substructure or model of data. Second, analogies are considered. Their operation is analyzed here in terms of models and partial isomorphisms, which allows one to grasp the idea of a translation of problems (and their solutions) across distinct domains of inquiry. The important role played by analogies in the thought of Maxwell (who also notably valued models) is highlighted. Third, I discuss the principle of correspondence, which historically played a remarkable role in the first period of quantum theory. The principle acquired two distinct forms, different in meaning, both of which can be clarified through the use of models. The Bohr-Sommerfeld model of the atom receives special attention, both in order to provide a context for discussing the principle of correspondence, and to exemplify the features of representational models as well.