The doctrine of simple individuals has its equal and opposite reaction in the view that an individual is simply a bundle of properties, that the identity of an individual is entirely dependent on the identity of its properties. This view also seems to me to be in some sense wrong and I shall attack it in passing. If all my remarks have seemed excessively polemical it is because I have been anxious to make it as clear as possible what the (...) motivation behind this paper is. I am mainly concerned with the problem concerning the "it" which underlies or has properties and I shall want to argue that the identity of an individual--what it is that makes an individual that individual--does not depend merely on its being that individual or having that piece of substratum, but it depends partially and in a complicated way on the identity of its properties. The analysis will have to exhibit the nature of that dependence. Most of the discussion will revolve around proper names and their function since it is through linguistic considerations, I think, that we get the clearest illumination. The result, in case any one is interested, will be Absolute Idealism, or something like it. (shrink)

The title is an allusion to the fact that, traditionally, propositions have served at least two distinct functions in philosophy, even though these functions have not usually been distinguished. Propositions have been invoked as the ‘meanings’ or ‘intensions’ of sentences and as the objects of propositional attitudes. Thus the proposition that Socrates is wise is the meaning of the English sentence, ‘Socrates is wise,’ and is what Charles believes when he believes that Socrates is wise. ‘Means that’ and ‘believes that’ (...) take the same kind of object. I shall argue, first, that there are two quite different functions here and it takes two quite different kinds of proposition to serve these distinct functions. The second thing I want to do is examine the suggestion that we can treat propositions as sets of possible worlds. There is a serious difficulty here, and so I shall want to turn that idea on its side: I shall define possible worlds as sets of G-propositions and then define E-propositions as sets of possible worlds. (shrink)

Chapter I announces the aim of the book, which is, to deal with the question: What is a language? It also registers complaints against current semantical methods. The sections here are closely related to Quine's Two Dogmas, but the author finds himself dissatisfied, not just with analyticity, but also with logical truth, truth, designation. The difficulties are of two orders. In one case they would be dissolved by having general definitions of the terms in question. In the other case we (...) should have to discuss use of a language and that would take us into pragmatics. It turns out that to get our "general" definitions we have to define "language," "logical truth," and "G-designation" all at once, so to speak. Chapter II lays down criteria of adequacy for a definition of "language": two languages are identical if and only if they have the same vocabulary and every expression has the same significance in one language as it has in the other. Now the author has to cope with "significance." In Chapter III, he follows Quine in holding that an extensional languages recognizes at most individuals, truth-values, and classes; and an intensional or modal language recognizes at most individual concepts, properties, and propositions. The author wants to have it both ways: like lots of people, he wants individuals and properties and propositions. To have it both ways, he has to work out a new semantical method ; and since his properties and propositions are not the familiar intensions, he has to call them G-properties and G-propositions. If these entities are named in the language then they are G-designated by their names. (shrink)

I wish now to return to the elementary characterization of languagehood in section III and its rationalization in section IV and say something by way of conclusion. The account given may or may not have a large number of fascinating and important consequences, but I shall confine myself to a couple of minor points and one not so minor.Let us suppose that there are either an infinite number of extra-linguistic entities (which seems plausible) or an infinite number of possible expressions (...) (inscriptibles). Then, since the Descriptive Part of a language is just a finite mapping of expressions onto entities, there will be an infinite number of languages. The point is pretty obvious anyway. A language is presumably a linguistic structure (whatever that is) thatcould be used by intelligent creatures in exchanging information and in doing logic. There is no reason at all to suppose that the number of such structures is finite.The second minor point is derivative. There is no essential differencequa languages between the so-called natural languages and the so-called artificial languages. Consider hammers. Some hammers are actually used to drive nails. Some perish in warehouse fires before they ever have that honor. But all are equally hammers. It's the same with languages. Some are actually used, the vast majority, including the “artificial” languages defined in semantics books, are never used. But apart from differences in richness and complexity, all are equally languages.8But the major point I wished to bring out in this paper is the contrast between a language and its use, or, if you like, between a language and speech behaviour. I labored the distinction in my introductory remarks and in the body of the paper I offered a sketch of a purely semantical definition of ‘language’. Insofar as the apparatus goes any distance towards bringing order out of chaos then this circumstance vindicates the distinction. We say that Alexander is able to ride Bucephalus and is able to speak Greek. Just as in the first case we distinguish between what Alexander is able to do, namely ride Bucephalus, and what he is able to ride, namely, Bucephalus, so in the second we should distinguish between what Alexander is able todo, namely, speak Greek, and what he is able tospeak, namely, Greek. Strictly speaking, of course, we have to recognize the linguistician's commonplace distinction, within use, between the ability or competence to speak (use) a language and specific exercises of that ability, or performances9. This is the distinction between ‘He speaks English’ and ‘He is now speaking English’.Another example, closer to present concerns. We would have no trouble distinguishing between the ability to play the Bach Chaconne and performances of the Chaconne on the one hand, and on the other, the Chaconne itself; between what the violinist is able todo (viz., play the Chaconne) and what he is able to play (viz., the Chaconne). And once we accept the distinction we would expect to be able to describe the Chaconne —which is a sequence of sounds - independently of the behavior involved in playing it. An unedited score in fact does this. It is the violinteacher who talks about double stopping and so on.10 In the same way, it seems to me, if we accept the distinction between a language and its use we commit ourselves to discovering ways of giving a phonological-syntactical-semantical characterization of a languagewithout reference to behavior, without employing the word ‘uses’ - and this demand has exactly the same rationale as the demand that we keep psychology out of logic. Then in pure pragmatics, the word ‘uses’ comes into its own. (shrink)