The Role of Defeasible Reasoning in the Modelling of Scientific Research Programmes

摘要

Classical studies in the theory of science concentrated on a logical analysis of scientific theories, with the unfortunate consequence of leaving aside many interesting epistemo-logical problems (the scientific discovery problem, or the representation of sociological and pragmatical drives, among others). In the 70's a new epistemological view emerged, which suggested a departure from logical positivism. However, either the formalisms were, still close to standard logic, or the presentations were too asystematic, and the formalization of scientific procedure was still obscure. In this paper we present a defeasible reasoning system that is an adequate model for representing scientific theories. Scientific reasoning is regarded mostly as a design process. The system incorporates incomplete, or uncertain evidence about a given situation as information provided by more or less trustable sources, to extend the reasoning context. This context is then used as a basis for a defeasible reasoning process. Then we show how our system can provide an adequate basis for a, scientific reasoning model. In particular, we concentrate on the epistemology of the scientific research programmes.