Representation models as devices for scientific theory applications vs. the semantic view of scientific theories: The case of models of the nuclear structure.

摘要

Analyses of the nature and structure of scientific theories have predominantly focused on formalisation. The Received View of scientific theories considers theories as axiomatised sets of sentences. In Hilbert-style formalisation theories are considered formal axiomatic calculi to which interpretation is supplied by a set of correspondence rules. The Received View has long been abandoned. The Semantic View of scientific theories also considers theories as formal systems. In the Semantic conception, a theory is identified with the class of intended models of the formal language, if the theory were to be given such linguistic form. The proponents of the Semantic View, however, hold that this class of models can be directly defined without recourse to a formal language. Just like its predecessor, the Semantic View is also not free of untenable implications. The uniting feature of the arguments m this work is the topic of theoretical representation of phenomena. The Semantic View implies that theoretical representation conies about by the use of some model, which belongs to the class that constitutes the theory. However, this is not what we see when we scrutinise the features of actual representation models in physics. In this work particular emphasis is given to how representation models are constructed in Classical Mechanics and Nuclear Physics and what conceptual resources are used in their construction. The characteristics that these models demonstrate instruct us that to regard them as families of theoretical models, as the Semantic View purports, is to obscure how they are constructed, what is used for their construction, how they function and how they relate to the theory. For instance, representation models are devices that frequently postulate physical mechanisms for which the theory does not provide explanations. Thus it seems more appropriate to claim that these representation devices mediate between theory and experiment, and at the same time possess a partial independence from theory. Furthermore, when we focus our attention to the ways by which representation models are constructed we discern that they are the result of the processes of abstraction and concretisation. These processes are operative in theoretical representation and they demand our attention if we are to explicate how theories represent phenomena in their domains.