¹ This is an edited version of my contribution to Knowing Our Own minds, ed. Macdonald, C., Smith, B. and Wright, C. (Oxford: Oxford University Press 1998).Google Scholar

² Austen, Jane, Emma (London: Penguin Books, 1987), p. 398.Google Scholar This nice passage was drawn to my attention by Tanney, Julia, who uses it for her own purposes in her ‘A Constructivist Picture of Self-Knowledge’, Philosophy 71 (1996), 405–22.CrossRefGoogle Scholar

³ Since it cannot be attributed, as with phenomenal avowals, to the fact of sincerity-cum-understanding guaranteeing truth, it is an interesting question what this weak authority should be taken to consist in. It might be suggested that it is nothing other than the presumptive acceptability of testimony generally. And certainly that proposal would be enough to set our problem: for the presumptive acceptability of original testimony—testimony for which the source is not itself testimony - extends no further than to subject matters which an informant is deemed competent to know about. So the question would recur: how is it possible for subjects to know about their intentional states in ways that involve no consideration of the evidence on which a third-party must rely? Actually, however, I think the suggestion is wrong. What distinguishes the presumptive acceptability of attitudinal avowals from anything characteristic of testimony generally, is that the authority which attaches to them is, in a certain sense, inalienable. There is no such thing as showing oneself chronically unreliable in relation to the distinctive subject matter of attitudinal avowals. I may have such poor colour vision that you rightly come to distrust my testimony on matters of colour. I may, unwittingly, have a very bad memory and, learning of this, you may rightly come to a state of wholesale suspicion about my testimony on matters of personal recall. But no corresponding wholesale suspicion concerning my attitudinal avowals is possible. You may not suppose me sincere and comprehending and yet chronically unreliable about what I hope, believe, fear and intend. Wholesale suspicion about my attitudinal avowals - where it is not a doubt about sincerity or understanding - jars with conceiving of me as an intentional subject at all.

⁴ From p. 149 of McDowell, John, ‘Intentionality and Interiority in Wittgenstein’, in Meaning Scepticism, ed. Puhl, K. (Berlin: de Gruyter, 1991), pp. 148–69.Google Scholar

⁵ To take another nineteenth-century example, it is, in a sense, the entire subject matter of George Eliot's novella, The Lifted Veil. The Cartesian character of that writer's notions about the mental is explored in depth in Catherine Wright's ‘The Unseen Window: Middlemarch, Mind and Morality’ (Ph.D. dissertation, University of St Andrews, 1991).Google Scholar

⁶ In particular, I do not think that we have any satisfactory concept of what it would be to be in touch with others' mental states telepathically. I do not mean, of course, to rule it out that someone might prove, by dint of his own occurrent suspicions and afflictions, to be a reliable guide to the states of mind of another. But that possibility falls conspicuously short of the idea that a subject might share direct witness of another's mental states.

⁷ For further indications, see my ‘Wittgenstein's Later Philosophy of Mind: Sensation, Privacy and Intention’, in Meaning Scepticism, ed. Puhl, , pp. 126–47.Google Scholar

⁸ See e.g. Philosophical Investigations §§34, 146, 152, 154, 205, 303, 330–2, 427, 577, 673; also part II §vi p. 181, and §xi pp. 217–18. The distinction is prominent in the Remarks on the Philosophy of Psychology as well, where Wittgenstein uses the terminology of dispositions versus states of consciousness; see, for instance, vol. II, §§45, 48, 57 and 178.

⁹ This idea is elaborated a little at pp. 237ff. Of my ‘Wittgenstein's Rule-Following Considerations and the Central Project of Theoretical Linguistics’, in Reflections on Chomsky, ed. George, Alexander (Oxford: Basil Blackwell, 1989), pp. 233–64.Google Scholar

¹⁰ But see also Remarks on the Philosophy of Psychology, vol. I, §§450, 501, 593, 599 and 832.

¹¹ The sometime popularity of this interpretation is traceable to its being advanced by several of the first reviewers: Strawson, P. F., for instance, in his critical study of the Investigations in Mind 63 (1954), 70–99Google Scholar; and Malcolm, Norman in his ‘Wittgenstein's Philosophical Investigations’ in The Philosophical Review 63 (1954), 530–59.CrossRefGoogle Scholar

¹² After P. T. Geach's emphasis of such difficulties for moral expressivism, Austin's performatory account of knowledge, etc. See Geach's, ‘Assertion’ in The Philosophical Review 74 (1965), 449–65.CrossRefGoogle Scholar

¹³ For instance Investigations part II, section ix: ‘a cry, which cannot be called a description, which is more primitive than any description, for all that it serves as a description of the inner life.

A cry is not a description. But there are transitions. And the words, “I am afraid”, may approximate more, or less, to being a cry. They may come quite close to this and also he far removed from it.’.

¹⁴ Philosophical Investigations §109.

¹⁵ Philosophical Investigations §654. It doesn't matter that this is said in the context of discussion of a different issue (recollection of the content of a prior intention).

¹⁶ As Baker and Hacker style it.

¹⁷ Philosophische Grammatik, (Oxford: Blackwell, 1969), section X, §§133 and 134.Google Scholar

¹⁸ From p. 142 of my ‘Wittgenstein's Later Philosophy of Mind: Sensation, Privacy and Intention’, in Meaning Scepticism, ed. Puhl, .Google Scholar

¹⁹ I suppose this is a programme of what McDowell has disparagingly called ‘constructive philosophy’.

²⁰ Both these ideas are explored in my Truth and Objectivity (Cambridge, MA: Harvard University Press, 1994).Google Scholar

²¹ The phrase, of course, is from Philosophical Investigations §133.