Abstract

Relatively recent work on Carnap, based on his published papers and books as well as on his unpublished correspondence and other material, has suggested that Carnap and Kuhn might not have been miles apart when it comes to the issue of theory-change. Two prevailing thoughts are that a) Kuhnian ‘paradigms’ might be taken to be very similar to Carnapian ‘linguistic frameworks’ and b) Kuhnian ‘ incommensurability ’ between competing paradigms is consonant with Carnap’s thesis that when a linguistic framework is replaced by another, there is a change of language and the analytic-synthetic distinction needs to be redrawn within the new framework. Irzik and Grunberg have gone on to note that Carnap endorsed “semantic incommensurability ”. They base their claim on the theses that a) Carnap endorsed meaning holism; b) Carnap endorsed the thesis of ‘theory-ladenness of observation’. They are certainly right in saying that “without semantic holism semantic incommensurability would be groundless; without theory-ladenness it would be severely restricted to the theoretical terms”. But, I think, they are wrong in claiming that Carnap endorsed either meaning holism or the theory-ladenness of observation. The aim of this paper is to show how Carnap avoided the alleged problem of incommensurability. Better put, Carnap’s view about the language of science is such that this problem does not arise. Drawing on his published and unpublished material, I highlight some connections between his work on semantics and his mature work on the structure of scientific theories, which was based on his re-invention of the Ramsey-sentences.