¹ Cf. Castañeda, Hector-Neri, ‘“He”: A Study in the Logic of Self-Consciousness’, Ratio 8 (1966), 130–157.Google Scholar

² It may be worried that clinical conditions like schizophrenia and so-called multiple personality syndrome present a challenge to this claim. I think not. Either such conditions involve a genuine multiplicity of selves or they do not. If they do (which I very much doubt), then they only imply a novel form of communication between selves. If they do not, then consider the nature of the delusion that the self is under. The self seems, say, to ‘hear voices’, not realizing that it is just receiving messages from its own subconscious. But then the self correctly identifies these ‘seemings to hear’ as its own conscious experiences, and only fails to identify as its own the unconscious thoughts which give rise to those experiences. So this is not a case of a self failing to identify as its own any of its own present conscious thoughts or experiences, and hence not a counter-example to my claim.

³ I myself would be happy to endorse a Russellian analysis of such a definite description, rather than regard it as involving a ‘referential’ use in Donnellan, Keith's sense: see his ‘Reference and Definite Descriptions’, The Philosophical Review 75 (1966), 281–304CrossRefGoogle Scholar. I borrow the phrase ‘fix the refer ence’, of course, from Kripke, Saul: see his Naming and Necessity (Oxford: Basil Blackwell, 1980), 53ff.Google Scholar

⁴ Cf. Kripke, , Naming and Necessity, 57–58.Google Scholar

⁵ Consequently, it might still be held that ‘that person’ here functions as a device of ‘direct reference’ in David Kaplan's sense, according to which ‘The “direct” of “direct reference” means unmediated by any propositional component, not unmediated simpliciter’ and hence that ‘Whatever… mechanisms there are that govern the search for the referent, they are irrelevant to the propositional… content’: see his ‘Afterthoughts’, in Themes from Kaplan, Almog, J. et al. (eds), (New York: Oxford University Press, 1989), 569Google Scholar. It seems to me that my use of the adjective ‘direct’ is more natural than Kaplan's, so I shall stick to it and trust this footnote to dispel any possible confusion.

⁶ Dickens, Charles, Hard Times (Harmondsworth: Penguin Books, 1969), 224.Google Scholar

⁷ See Kripke, , Naming and Necessity, 56.Google Scholar

⁸ It is true that explicit endorsement of such a view is rarely seen these days, but many important contemporary theorists of personal identity espouse views which are recognizably its heir and which are subject to the same line of criticism—most notably, perhaps, Parfit, Derek in his Reasons and Persons (Oxford: Clarendon Press, 1984).Google Scholar

⁹ For an interesting discussion of ‘I am here now’ as an example of the contingent a priori, see Forbes, Graeme, Languages of Possibility: An Essay in Philosophical Logic (Oxford: Basil Blackwell, 1989), Ch. 6.Google Scholar

¹⁰ For further argument to this effect, see Section IV of my ‘Real Selves: Persons as a Substantial Kind’, in Human Beings Cockburn, D. (ed.), (Cambridge University Press, 1991).CrossRefGoogle Scholar

¹¹ For an account of my own positive views on the nature of the self, see my ‘Substance and Selfhood’, Philosophy 66 (1991), 81–99.Google Scholar

¹² Evans, Gareth, The Varieties of Reference (Oxford: Clarendon Press, 1982), 173n.Google Scholar

¹³ The locus classicus for the notion of a ‘basic’ action is Danto, Arthur C., ‘Basic Actions’, Americal Philosophical Quarterly 2 (1965), 141–148Google Scholar. I use the expression ‘at will’ in the way Williams, Bernard does in his ‘Deciding to Believe’, reprinted in his Problems of the Self (Cambridge University Press, 1973): see p. 148.CrossRefGoogle Scholar

¹⁴ In my ‘Substance and Selfhood’, op. cit. (pp. 87–88), I argue that what actually qualifies a particular body as mine is precisely the fact that certain parts of it are directly subject to my will and/or loci of my sensations, and this provides an additional reason for rejecting the suggestion that I can only individuate my own bodily parts as bodily parts of mine, since that would now engender a circularity in the proposed account of what makes a certain body and its parts ‘mine’. For instance, by that account it is (partly, at least) because I can move this hand ‘at will’ that it qualifies as one of my hands—but then if it is suggested that by ‘this hand’ I have to mean something like ‘my left hand’, the explanatory force of that account is completely nullified.

¹⁵ I am grateful to David Over, Andy Hamilton, Tim Crane and to audiences at the University of Liverpool and Davidson College for comments on earlier versions of this paper.