Russell: The Journal of Bertrand Russell Archives

New work on Russell's early philosophy 163 I I wish to thank John Rowe, Deanofthe Faculty ofArts, University ofWestern Ontario, for support of research through a SSHRC Faculty of Arts grant. 2 While discussing Hursthouse's paper I shall adopt Russell's convention in thePrinciples of using italics for names of concepts. 3 ct· p. 39n.3. Unless otherwise identified, all page references are to the volume under review. ROUGHLY COVERING THE period which begins with The Principles of Mathematics (1903) and extends through the unpublished 1913 manuscript Theory ofKnowledge, this is an excellent volume on the development of Russell's logical theory and philosophy of language. There is a very nice essay by Rosalind Hursthouse giving the philosophical motivation for Russell's concept ofdenoting in the Principies , and thus clarifying the theory of "On Denoting". The problem which struck Russell in the Principles may be introduced as follows: Consider the sentence, 'Humanity belongs to Socrates'. Preanalytically, this seems clearly to be "about" the concept humanity.2 But the concepts humanity,. man, and all men are all in some sense "equivalent". So how comes it that 'Man is mortal' is not in any sense about man? Russell's solution was to hold that among concepts there is a twofold division. There are those which, like man, all men, some men, etc., are denoting concepts, and there are those (e.g. humanity) which are not: "A concept denotes, when, if it occurs in a proposition, the proposition is not about the concept, but about a thing connected in a certain peculiar way with the concept" (Principles, p. 53, quoted by Hursthouse, p. 37). Following Blackburn and Code 1978,3 Hursthouse argues that this "peculiar connection " of denoting concepts with their denotation is just Frege's determining which, on Frege's view, holds between the sense and reference of names and descriptive phrases. Although Russell differs from Frege in holding that names are not denoting phrases, i.e. they do not indicate denoting concepts, he agrees with Frege on the question of denoting phrases involving 'all', 'some', 'the', 'a', 'any', 'every',etc. The denoting concepts of such denoting phrases denote their denotations. What is given up in "On Denoting" is, first of all, the mistaken conception of generality which holds that, e.g., 'all men' refers to the Bertrand Russell's Early Philosophy, Part I. Edited by Jaakko Hintikka. Synthese, vol. 45, no. I (Sept. 1980). 188 pp. US$23.00. by William Demopoulos1 Socrates :::> Socrates 164 Russell winter 1981-82 class of men. This undercuts the motivation for a notion of denoting at least so far as 'all' and 'some' sentences are concerned. But in the case of sentences involving definite descriptions, it still appears that by their use we indicate concepts which denote their denotation. Russell's classic resolution of the problem is of course the theory of descriptions-the contention that definite descriptions achieve their purpose, not by their association with a denoting concept but "because such things as Scott have unique properties" (p. 41). This ofcourse is only the briefest summary ofHursthouse's argument. Even so, I think one unclarity"is evident: IfHursthouse is right, how can denoting possibly be indentified with Frege's determining? Denoting arises only because of Russell's confusion regarding quantification. Although Frege is committed to there being some relation between the sense and reference of a designating expression, it seems completely unwarranted to suggest that this relation is the relation of denoting. So far as I can see, neither denoting nordetermining is an arbitrary relation between its terms. But for Frege, this feature is rather straightforwardly explained by the compositional structure of designating expressions. This explanation clearly does not suffice for the sense in which denoting is nonarbitrary . This suggests that Hursthouse's paper should yield an important qualification of the argument of Blackburn and Code. The longest paper (seventy-two pages) in the issue is Nicholas Griffin's very ambitious contribution, "Russell on the Nature of Logic (19031913 )". It is not possible to do justice to Griffin's paper in a short review such as this one. Especially useful is Griffin's discussion (the first in print, to my knowledge) ofRussell...