Making Things with Words: Wittgenstein on Inference and Representation

Abstract

Some interpreters argue that for Wittgenstein, logic is not concerned with proof and inference, but with truth and entailment, and that Wittgenstein regards inference as a theme of psychology. That construal is often supported by Tractatus 5.132, where Wittgenstein states that only the propositions which serve as the premise and the conclusion can justify the inference, hence that no mediation by an inferential act and laws of inference is needed. Wittgenstein considers the idea of mediation to be Frege’s view, and he also rejects Frege’s distinction between thought and assertion. Other scholars, for their part, argue for the interpretation that Wittgenstein’s own view is not far from what Frege held. Still others claim that for Wittgenstein, inference and representation are equally basic notions. The present paper first discusses interpretations of Wittgenstein’s view on inference, proposed by Martin Gustafsson, Colin Johnston, Gilad Nir, Göran Sundholm, and Kurt Wischin. It then compares the views presented in the Tractatus with Frege’s semantic views, including Frege’s pragmatic ideas in his later writings, as well as with J.L. Austin’s speech act theory. It argues that contrary to what Tractatus explicitly tells us, Wittgenstein is close to Frege in his conception of inference and representation. Moreover, the paper pays special attention to features in the Tractatus that resemble Austin’s idea of doing things with words.