Published online by Cambridge University Press: 01 January 2020
Some contemporary Russellians, defenders of the view that the semantic content of a proper name, demonstrative, or indexical is simply its referent, are prepared to accept that view's most infamous apparent consequence: that coreferential names, demonstratives, indexicals, etc. are intersubstitutable salva veritate, even in intentional contexts. Nathan Salmon and Scott Soames argue that our recalcitrant intuitions with respect to the famous apparent counterexamples are not semantic intuitions, but rather pragmatic intuitions. Strictly and literally speaking, Lois Lane believes, and even knows that Clark Kent is identical to Superman, since she believes and knows that Superman is identical to Superman. Salmon and Soames attempt to soften our reaction to this shocker by allowing that it is typically misleading to utter the sentence ‘Lois Lane knows that Clark Kent is identical to Superman,’ since it pragmatically implicates, without semantically entailing, that Lois Lane would accept the sentence ‘Clark Kent is identical to Superman.’ Our compulsive tendency to claim that ‘Lois Lane knows that Clark Kent is Superman’ is false, rather than merely misleading, is due to a confusion between semantics and pragmatics, between truth conditions and conditions of appropriateness of utterance.
1 See Salmon, Nathan Frege's Puzzle (Cambridge, MA: The MIT Press 1986)Google Scholar; Soames, Scott ‘Direct Reference, Propositional Attitudes, and Semantic Content,’ in Salmon, N. and Soames, S. eds., Propositions and Attitudes (New York: Oxford Unversity Press 1988) 218-20Google Scholar; and Grice, H.P. ‘Logic and Conversation,’ in Martinich, A.P. ed., The Philosophy of Language 2nd ed. (New York: Oxford University Press 1990) 149-60Google Scholar
2 Mark, Richard, Propositional Attitudes (New York: Cambridge University Press 1990), 125Google Scholar
3 See chapter 3.
4 The terminology here should not be taken to imply that the sentences accepted must be natural language sentences, or that the notion of acceptance here is exactly the everyday notion of acceptance; these are technical notions that Richard develops in his own way. See Propositional Attitudes, chapter 3, especially 181-90.
5 In this paper I will pretend that the ‘linguistic parts’ of RAMs in one's RS are pieces of natural language. This ignores some complications present in Richard's final view - see Propositional Attitudes, 181-90.
6 See Propositional Attitudes, 136 ff.
7 See Kripke, Saul ‘A Puzzle About Belief,’ in Salmon, and Soames, Propositions and Attitudes 102-48Google Scholar, and Richard's, discussion in Propositional Attitudes, 179-80.Google Scholar
8 See Propositional Attitudes chapter 3. Another ‘contextualist’ theory which shares many attractive features with Richard's is that defended in Crimmins, Mark and Perry, John ‘The Prince and the Phone Booth: Reporting Puzzling Beliefs,’ The Journal of Philosophy 86 (1989) 685–711.CrossRefGoogle Scholar
9 See Propositional Attitudes Chapter 3.
10 I'm not claiming that (3) is true; as will be seen below, since Odile = Amanda and (1) seems true, it's plausible that (3) is false. My claim is simply that Charlie and my intentions are sufficient for restriction R2's being operative; after all, we think we're talking about two different people, and if we were talking about two different people, (3) would be true and R2 would be in place.
11 I thank Mark Richard and an anonymous referee for helpful discussions on this point.
12 See Lewis, David ‘Scorekeeping in a Language Game,’ in his Philosophical Papers vol. 1 (New York: Oxford University Press 1983) 233-49, at 245-6Google Scholar on contextual determination of standards of precision.
13 Perhaps a description theory of demonstratives would avoid this consequence in the case of pointing.
14 See Propositional Attitudes 151-3.
15 The phrase is from Kripke’ s ‘A Puzzle About Belief’ (122).
16 See Propositional Attitudes pp. 173-180.
17 One might object here along the lines of the objection at the end of section III: there is no single correlation function that makes the latter two conjuncts true, but the natural way to interpret this sentence is indeed in terms of a single correlation function. But I disagree that the natural way to interpret this sentence is according to a single correlation function, since the speakers in the dialogue don't know that Odile is identical to my student.
18 This argument will be blocked at some point by the theory of the present section, depending on how quantification into the left-hand position of belief sentences is handled. The objection is just that the argument shouldn't be blocked- it seems valid.
19 This point has also been made (independently) by Crimmins, Mark ‘Context in the Attitudes,’ Linguistics and Philosophy 15 (1992) 185-98CrossRefGoogle Scholar; see 192. Richard, replies in ‘Attitudes in Context,’ Linguistics and Philosophy 16 (1993) 123-48, at 127-9CrossRefGoogle Scholar. Whatever the merits of Richard's reply, it does not apply to my examples in this section, not directly anyway.
20 Incidentally, Richard's theory also has the result that ‘Jane believes that A is large and red’ turns out false in every context, whereas it would be natural to count it true in the context I discuss in the text. Moreover, it is hard to see how Richard's theory could be revised to respect this intuition.
22 It might be thought that turning correlation functions into relations ruins Richard's account of the Pierre puzzle all by itself. But Richard can make an alternate response to the Pierre puzzle: when we move from saying ‘Pierre believes that London is pretty’ to ‘Pierre believes that London is not pretty’ we switch contexts; in each of the original contexts, one of the sentences is false, because that context has restrictions that rule out the correlation relations that would make that sentence false. Pierre is not irrational because in such contexts, it is not the case that both ascriptions are true. Granted, there are some contexts in which both turn out true, but this was true on Richard's original theory (see Propositional Attitudes, 180).
(i) ƎxƎy(x=y and x is a planet and y is a planet and John believes that he saw x rise, then y rise, then x set, then y set)
therefore,
(ii) ƎxƎy(x=y and x is a planet and y is a planet and John believes that he saw x rise, then y rise, then y set, then x set)
Richard regarded it a virtue of his theory that it made this argument invalid. However, the argument is valid on the new theory (assuming as Richard does that no restrictions are allowed on variables). However, I don't think this is conclusive evidence against the relational theory. Richard says of this argument that it is ‘far from transparently valid’ (Propositional Attitudes, 153). But I think it is also far from transparently invalid.
23 Propositional Attitudes, 152. Since Richard claims to have intuitions about Al, I take it that he assumes the quasi-logical (ii) to have the same truth conditions as some sentence of English, perhaps ‘There is some famous author such that Odile believes that s/he is dead'; similarly for (ii’).
24 Here I thank an anonymous referee for a suggestion.
25 For an idea of the issues that would be involved in denying (AS2), see Kaplan, David ‘Quantifying In,’ Synthese 19 (1968-69) 178–214, section XICrossRefGoogle Scholar; specifically, note the relationship between his (45), (46), and (47). (AS2) could be tinkered with to accommodate truth value gaps.
26 I thank Mark Richard for helpful comments here.
27 See Quine, W.V.O. ‘Quantifiers and Propositional Attitudes,’ The Journal of Philosophy 53 (1956) 177-87CrossRefGoogle Scholar and Kaplan, ‘Quantifying In.’
27 See Quine, W.V.O. ‘Quantifiers and Propositional Attitudes,’ The Journal of Philosophy 53 (1956) 177-87CrossRefGoogle Scholar and Kaplan, ‘Quantifying In.’
28 I would like to thank Mark Aronszajn, David Cowles, Ed Gettier, two anonymous referees, and especially David Braun and Mark Richard for their help with this paper.