1 I have addressed it elsewhere, however: in ‘Is, Madam? It Seems!’ in Denis Dutton, ed., The Forger's Art: Forgery and the Philosophy of Art (Berkeley, CA: University of California Press 1983), 188-224; ‘Goodman on Forgery,’ The Philosophical Quarterly 33 (1983), 340-53; and ‘Forgery,’ Lawrence and Charlotte Becker, eds., The Encyclopedia of Ethics, Volume I, 2nd ed. (London: Routledge 2001), 556-61.

2 Whether this reading of Descartes is correct is another matter. I accept it here, if only for the sake of argument.

3 Gilbert Ryle, Dilemmas (Cambridge: Cambridge University Press 1954), 94-5

4 Other philosophers willing to invest in counterfeit coins include John Passmore, Philosophical Reasoning (New York: Basic Books 1961), 108; Colin Grant, ‘Polar Concepts and Metaphysical Arguments,’ in H.D. Lewis, ed., Clarity Is Not Enough (London: Allen and Unwin 1963), 272; and, in modified form, D.W. Hamlyn, Theory of Knowledge (New York: Doubleday 1970), 18-19.

5 Jay Rosenberg, The Practice of Philosophy (Englewood Cliffs, NJ: Prentice Hall 1978), 18-19. The argument contra Descartes is repeated in the third edition of Rosenberg's book (Upper Saddle, NJ: Prentice Hall 1996), 24-5. I imagine that it can be found in the second edition (1984) as well, though I haven't been able to obtain a copy of that edition of the book.

6 Anthony Kenny, ‘The Verification Principle and the Private Language Argument,’ in O.R. Jones, ed., The Private Language Argument (London: Macmillan 1971) 204-28, at 216

7 This is a point that Gerald Massey has emphasized in a number of papers in arguing for a general asymmetry between judgments of formal validity and formal invalidity. Actually, Massey's thesis needs to be qualified and restricted to be perfectly accurate, but pursuing the matter would lead us far afield. For more on the matter, see my ‘Most Assur’d of What He is Most Ignorant,’ Erkenntnis 44 (1996), 341-68, esp. 345-51.

8 And many philosophers, myself among them, think that it doesn't exhaust validity. See, for instance, the article of mine mentioned in the immediately preceding footnote.

9 A quick word on Kenny's claims about the privacy of experience. As Kenny uses the term in the quoted passage, a private experience is apparently an occurrent mental state for which there's no behavioral or publicly available evidence. If he's suggesting that it's impossible for all experiences to be private in this sense — he never explicitly says as much — I think he's wrong, and for reasons similar to those I’ve advanced against Ryle and Rosenberg. The claim that people feel pain but never publicly exhibit pain behavior isn't self-contradictory, and a world in which everyone always suppresses the urge to scratch, talk about, or otherwise manifest evidence of his itches or pains would be a very odd world, but not an impossible one. Perhaps its denizens wouldn't have the concept of pain — though I think they could — but that's another matter. The simplest way to see that it's possible for all experiences to be private in Kenny's sense is to consider an admittedly strange world, one in which a single cognitively mature individual exists for a short time. During his brief life, he experiences pain once, twice, or a number of times, but for personal reasons (curiosity about the sensation), ethical reasons (stoicism), religious reasons (a prohibition on certain kinds of behavior), or no reason at all (whim) decides not to exhibit pain behavior.

10 A remark that needs qualification; see n. 7 above.

11 Nelson Goodman, Language of Art (Indianapolis, IN: Bobbs-Merrill 1968), 122

12 This, incidentally, is a mistake common to many if not all of the polar concepts arguments that flourished in the late 1940s and throughout the 1950s. Typically, such arguments conclude that since M and N are polar concepts, there has to be something that's an M, if anything at all is an N, or even (in some arguments) if the concept of N shows up in the language. In effect, Ryle's argument, if not Rosenberg's or Kenny’s, is a polar concepts argument: the existence of genuine coins is inferred from the existence of counterfeit coins. Come to that, Ryle actually offers two polar concepts arguments, since, by analogy, he infers the existence of genuine or veridical perceptions from the existence of non-genuine or deceptive perceptions. Geared though it is to prove something of more widespread, longstanding, and central philosophical interest than the counterfeit coins argument, this second argument unfortunately fares no better. The reason is much the same: the concept of a deceptive perception may be polar to and parasitic on that of a veridical perception, but a deceptive perception is not, of necessity, a copy of a veridical perception, and it needn't be like or resemble a veridical perception, in a relational sense of either of those terms.

13 Much the same holds for perception, incidentally: to perceive, in the sense that Descartes is interested in, requires only an intentional object.

14 William E. Kennick, ‘Art and Inauthenticity,’ The Journal of Aesthetics and Art Criticism 44 (1985), 3-12, at 5.

15 This is also part of the reply I would make to the charge that there are forgeries that aren't forged XYs, something that is impossible on my definition (because I define ‘forgery’ in terms of ‘forged XY’). Consider, for instance, the following counterexample. An unscrupulous art dealer represents the work of a lesser Delft artist as a genuine Vermeer. Even if it's granted that forgery has been committed — and we probably do want to say that the art dealer is guilty of forgery — the work of the lesser painter is surely not a forged Vermeer painting, and we wouldn't want to call the original work forged.

But the assurance behind the ‘surely’ here is misplaced. At best, calling the work of the honest lesser painter a forged Vermeer is slightly anomalous, not strongly counterintuitive, and the oddity itself can be muted to some extent by filling in the context and explaining why the painting is a forgery in the first place, i.e., by telling the full story of the innocent painter and the sly art dealer. Given that the anomaly is minor and can, to some extent, be explained away; given the systematic advan tages of a unified account of forgery; and given that any residual reluctance to call the original work forged can be partly dissipated by remembering that forged XYs are, in almost all cases, original ZYs (a forged Vermeer painting, for example, is an original lesser Delft artist painting) — given all that, I think it best to consider the painting a forged Vermeer, however misleading such a claim may appear out of context. And, except for the third point just noted, the same holds if there's no original innocent artist at all, as would be the case, for instance, if a piece of driftwood were passed off as a genuine Moore sculpture.

16 Arthur Danto, The Transfiguration of the Commonplace (Cambridge, MA: Harvard University Press 1981), 39-41. Danto actually tells a slightly more complicated tale, but as far as forgery is concerned, no essential detail is missing in the above.

17 So much so that in one of my earlier papers on forgery, ‘Is, Madam? Nay, It Seems!’ I said that copying is part of the concept of plagiarism (203). This is a view that, as the rest of this paragraph will make evident, I now think incorrect.

18 I regret not being able to discuss several interesting and important topics in the near vicinity of this paper, including ghostwriters, noms de plume, piracy, impersonation, and cheating in general.

19 My thanks to three anonymous referees for a number of useful comments on an earlier draft of this paper. Special thanks, however, must go to Walter L. Weber for hounding me to complete it.