Scientific change and the meanings of terms : an examination of P.K. Feyerabend's incommensurability theses

Feyerabend's incommensurability thesis concerning scientific change engenders a number of logical problems. While it is possible to examine Feyerabend's theory in relation to his historical arguments, the defects implicit in his arguments for the theory render more appropriate an analytical approach. These defects arise from the conjunction of presuppositions and theses that form the background to Feyerabend's claims for an incommensurability thesis. This background contains Feyerabend's criticisms of the traditional empiricism of the twentieth century and its reductionist account of scientific development, his objections to any attempt to rationalize science, his claim that there are fundamental conceptual and ontological changes in science, and his adoption of a meaning variance thesis which envisages wholesale changes in the meanings of all descriptive terms when one theory is replaced by another. While the criticism against traditional empiricism can be upheld, it does not necessitate the conclusion that alternative theories are incommensurable. Feyerabend's attack on Lakatos' rational Reconstructionism is not conclusive: he overlooks the possibility that there do exist standards of criticism, which can be termed "rational", operating within the sciences. The suggestion, supported by Hanson and Kuhn, that there are fundamental conceptual changes in science is open to criticism. The case against radical meaning variance is more complex as it requires the support of a theory of meaning. It is not clear that Feyerabend can, using Wharf's controversial ideas about language, provide a suitable theory of meaning to support his claims. A more satisfactory theory of meaning, based on views of Frege and Wittgenstein, while not denying some changes in the meanings of scientific terms, does not entail the consequence that there are necessarily radical changes in meaning from theory to theory. Although the objections to traditional empiricism are sound and a moderate thesis of meaning variance is acceptable, these do not give rise to the view that competing theories are incommensurable. Historical evidence shows the need to take into consideration the gradual, rather than revolutionary, nature of scientific development. This is compatible both with a moderate thesis of meaning variance and with a modification of the network model developed by Duhem, Quine and Hesse.

Grantor dc:publisher.institution

Department of Philosophy

Year dc:date.issued

1976

Author dc:creator

• Mizroch, Carol Arlene

Advisor dc:contributor.advisor

• Keaney, B P

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/16619

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/16619