^{page 106 note 1} I should like to thank my colleague Mr David Hirschmann for his help in reading and criticising earlier versions of this paper.

^{page 106 note 2} Inaugural lecture at University College London, I December 1964.

^{page 111 note 1} Journal of Philosophy, LVII (1960)Google Scholar, reprinted in Philosophical Turnings, 1966.Google Scholar

^{page 112 note 1} In the translation by D. F. Pears and B. F. McGuiness, London, 1961.

^{page 113 note 1} An Introduction to Wittgenstein's Tractatus (1967) p. 69 n.Google Scholar

^{page 114 note 1} Op. cit., p. 66.

^{page 115 note 1} The distinction between saying and showing does not on this account rule out self-referring expressions. A picture may depict itself, e.g. the Model Village in Bourton-on-the-Water in Gloucestershire depicts the village in which it stands, and so depicts itself (up to a limit of 1/64 of the actual size of the model). ‘Theoretically’, it could do so indefinitely without absurdity. The advantage of the distinction between saying and showing is that it makes it very clear that not being able to state that P is significant, and not being able to refer to P are, for P, two very different things.

^{page 117 note 1} McGuire, J. E., ‘Atoms and “The Analogy of Nature”’, Studies in the History and Philosophy of Science, vol. 1, no. 1, 1970.Google Scholar

^{page 119 note 1} Locke, , Essay, bk 11, chap. viii.Google Scholar

^{page 120 note 1} Cf. Anscombe, , op. cit., p. 67Google Scholar ‘What I have called the externality of the correlations between the elements of a picture and actual objects is an important feature of Wittgenstein's account. Giancarlo Colombo, S. J. … commented on Wittgenstein's theory of the “isomorphism”, as it is called, between language and the world, that it was difficult to see why a described fact should not be regarded as itself a description of the proposition that would normally be said to describe it, rather than the other way round.’ It is just these sorts of consequences of the ‘externalist isomorphism’ of Wittgenstein's Tractarian account which show why we have to modify it severely, at least if it is to be capable of illuminating development.

^{page 122 note 1} Gombrich, E. H., Art and Illusion, p. 89.Google Scholar

^{page 122 note 2} Art and Illusion, p. 321Google Scholar, ‘Every observation, as Karl Popper has stressed, is a result of a question we ask of nature, and every question implies a tentative hypothesis’. Professor Richard Gregory in his Eye and Brain makes very much the same use of the notion of the brain employing unconscious hypotheses.

^{page 122 note 3} Kuhn, T. S., The Structure of Scientific Revolutions.Google Scholar

^{page 122 note 4} Lakatos, I. and Musgrave, A., Criticism and the Growth of Knowledge, pp. 59–79.Google Scholar

^{page 123 note 1} Ibid., p. 77.

^{page 123 note 2} Ibid., p. 81.

^{page 125 note 1} Cf. Moore, , Commonplace Book, e.g. pp. 73–5, 117, 138, 156, 326.Google Scholar

^{page 125 note 2} Cf. Sense and Sensabilia, p. 65.Google Scholar

^{page 128 note 1} In this respect William Blake was justified when he opposed ‘dear Mother Outline’ to ‘brushing and daubing’ as part of his attack on what he saw as the evils of Empiricism. Cf. The Complete Writings of William Blake, ed. Keynes, Geoffrey (Nonesuch Press, 1957) p. 553.Google Scholar

^{page 129 note 1} Daitz, Edna, ‘The Picture Theory of Meaning’, Mind, 62 (1953) pp. 184–201CrossRefGoogle Scholar, repr. Flew, (ed.), Essays in Conceptual AnalysisGoogle Scholar, and Copi, and Beard, , Essays on Wittgenstein's TractatusGoogle Scholar.

^{page 130 note 1} Berger, John, The Success and Failure of Picasso, p. 169.Google Scholar

^{page 130 note 2} See Plate 4.