Abstract

Since the publication of The Structure of Scientific Revolution , relativism, resulting from the concepts of paradigm and paradigm shift, becomes one of the central problems in the philosophy of science. The author of this dissertation agrees with Thomas Kuhn in his criticism of the Logical Positivism, but denies that relativism arrives as a direct consequence. In this dissertation, the author tries to tackle the problem by analyzing some underlying basic concepts, e.g. paradigm, language and rationality. Firstly, he points out that a hard-core model of paradigm, as understood by I. Lakatos, is not totally appropriate for an all-rounded understanding of a paradigm. Rather, among the elements certain "family resemblance" comes to the surface, they are therefore regarded as belonging to the same paradigm. A paradigm is not a hindrance to but an exemplar for our learning. And a paradigm shift needs not result in a total switch over, it could be a revolution with inheritance. Secondly, the incommesurability of languages does not mean the breakdown of communication between different linguistic communities. True bilinguals can achieve cross-paradigm understanding. In addition, due to the creative activities within a language, the problem of untranslatability could be solved. Finally, rationality, rather than assuming a hierarchical structure, could be taken as a reticulated model. As a result, various disagreements at the axiological level could be rationally resolved. ;The author also tackles in passing some issues concerning language and rationality in the Chinese tradition of science, and compares this tradition with that of the western