1 Russell, Bertrand ‘On Denoting’ and The Philosophy of Logical Atomism,’ reprinted in Logic and Knowledge, ed. Marsh, Robert C. (New York: Capricorn Books 1971), 39–50, 177-281Google Scholar

2 Donnellan, Keith ‘Reference and Definite Descriptions,’ Philosophical Review 75 (1966) 281–304CrossRefGoogle Scholar; ‘Putting Humpty Dumpty Together Again,’ Philosophical Review 77 (1968) 203-15; and ‘Speaker Reference, Descriptions, and Anaphora,’ in French, P.A. Vehling, T.E. and Wettstein, H.K. eds., Contemporary Perspectives in the Philosophy of Language (Minneapolis: University of Minnesota Press 1979), 28–44Google Scholar

3 Donnellan uses the expression ‘indefinite definite description’ for what I call ‘indeterminate descriptions.’ I have avoided Donnellan's terminology because of Russell's distinction between indefinite descriptions and definite descriptions.

4 Kripke, Saul ‘Speaker's Reference and Semantic Reference’ in French, P.A. Vehling, T.E. and Wettstein, H.K. eds., Contemporary Perspectives in the Philosophy of Language (Minneapolis: University of Minnesota Press 1979), 6–27Google Scholar

5 Strawson, P.f. ‘On Referring,’ Mind 59 (1950) 320-44CrossRefGoogle Scholar

6 I am uncertain whether David Kaplan holds this view of demonstratives or the first view suggested. See ‘On the Logic of Demonstratives,’ reprinted in French, P.A. Vehling, T.E. and Wettstein, H.K. eds., Contemporary Perspectives in the Philosophy of Language (Minneapolis: University of Minnesota Press 1979), 401-12Google Scholar, and his unpublished paper Demonstratives UCLA 1977.

7 Kaplan, Demonstratives. It is important to distinguish between the context, which helps to determine which proposition is expressed by a given sentence, and the circumstance, which is used to evaluate the truth of the proposition expressed. Kaplan has made this point abundantly clear in the use of indexicals and demonstratives.

8 This view is presented more formally in Sections 4 and 5 in my paper ‘On the Logic of Belief’ (forthcoming in Noûs). The view I am suggesting here is not designed to solve the problem of non-standard modification. If I say at a costume party, John is talking to the unicorn’ the situation may qualify the indeterminate description ‘the unicorn’ in such a way as to make it clear that the object referred to is not a unicorn, e.g., ‘the person dressed as a unicorn.’ In such a case the object denoted does not satisfy the predicate in the indeterminate description, and hence will not be a member of g(ɸ)(ω). To account for such cases, a distinction between situations which modify the description in standard ways and situations which modify the description in non-standard ways is needed. However, we can ignore this difficulty for the present purposes.

9 Kripke, 14-15. Here I only indicate how this distinction shows that Donnellan's examples in his early paper do not undermine Russell's view. For a more complete account I refer the reader to Kripke's paper.

10 Donnellan, ‘Speaker Reference, Descriptions, and Anaphora,’ 28

11 Ibid., 33

12 Ibid., 33-4

13 Kripke makes a similar point with a different example on page 21.

14 Donnellan, ‘Speaker Reference, Descriptions, and Anaphora,’ 34

15 Ibid., 32

16 Donnellan uses both indeterminate singular and indeterminate plural descriptions as objections to Russell's view. Since my interest here is only with singular definite descriptions, I do not discuss Donnellan's objections which are based on plural definite descriptions. However, I believe my comments about singular descriptions can be extended to include plural descriptions.

17 Donnellan, ‘Speaker Reference, Descriptions, and Anaphora.’ 37

18 Ibid., 38-9

19 Kripke, 26, n. 32

20 An alternative view to the one I suggest is to view the expression ‘the man’ in (11) as occurring inside the scope of an existential quantifier. Thus (11) is understood as [There is an x such that x is a man who came to the office today and the man identical to x tried to sell me an encyclopedia.) While such a view is compatible with the position I outlined at the beginning of this paper, I believe there are good reasons for holding that two propositions are expressed in (11). I am indebted to Tom Downing for our discussions on this issue.

21 Donnellan, ‘Speaker Reference, Descriptions, and Anaphora,’ 42

22 This is not to say that situations never qualify a determinate description.

23 I am assuming that David is not pointing to the event of Joe's emptying Caster, but rather is pointing to the object Caster. This assumption avoids a number of questions about the nature of demonstrations. However, these questions need not be settled here in order to determine the truth value of E(l4).

24 It is interesting to note, though not relevant to the question of which proposition is actually expressed by (14) in the situation, that the reasons one has for asking the question What is Joe doing?’ will determine the relative importance (to the questioner) of the two propositions David asserts. If one is looking for an ashtray and wondering whether Joe has one, then it is the false proposition E(l6) one would tend to act on and not the true proposition E(l5). On the other hand, if one is interested in whether or not Joe is helping clean up the place, then the fact that Joe is emptying Caster instead of Bester would be of little concern.

25 I would like to thank John Stone, Ted Guleserin, Mike White, and Tom Downing for their helpful comments on earlier versions of this paper. Research for this paper was supported by a Faculty-Grant-in-Aid from Arizona State University.