David G. Stern

This paper examines a number of river images which have been attributed to Heraclitus, the ways they are used by Plato and Wittgenstein, and the connection between these uses of imagery and the metaphilosophical issues about the nature and limits of philosophy which they lead to. After indicating some of the connections between Heraclitus’, Plato’s and Wittgenstein’s use of river images, I give a preliminary reading of three crucial fragments from the Heraclitean corpus, associating each with a different river image. Each of these images implies an overall conception of the nature of change and continuity and of the relationship between language and world. I then turn to the use of these images by Plato and Wittgenstein, and explore the relationship between their uses of river imagery and their conceptions of philosophy. To put it slightly differently, this paper is about Plato’s and Wittgenstein’s use of certain Heraclitean ideas.

The title of this chapter is borrowed from John McDowell's ‘One strand in the private language argument’ (1998b). In that paper, he argues that much of what is best in Wittgenstein's discussion of private language can be seen as a development of the Kantian insight that there is no such thing as an unconceptualized experience - that even the most elementary sensation must have a conceptual aspect. On McDowell's view, a sensation is a ‘perfectly good something - an object, if you like, of concept involving awareness. What is a nothing … is the supposed pre-conceptual this that is supposed to ground our conceptualizations’ (1998b: 283). McDowell's Sellarsian objections to the notion of the Given in that paper are an insightful and illuminating development of Wittgenstein's discussion of the topic. However, McDowell's recoil from the notion of an unconceptualized experience, a conception of sensation on which it turns out to be ‘simply a nothing’ (ibid.), leads him to reject Wittgenstein's cryptic proposal that a sensation is ‘not a something, but not a nothing either’ (PI 304). Instead, McDowell embraces the opposed view on which every experience is a ‘perfectly good something’ (1998b: 283), something of one kind or another, for it must be possible to bring it under the appropriate concepts. What McDowell misses here, I believe, is that a central aim of Wittgenstein's discussion of our supposed ability to refer to inner objects is to attack the very idea of ‘pre-linguistic awareness … as a substratum on which the capacity for concept-carried awareness is constructed’ (ibid.).