Page 257 - Encyclopedia of Philosophy of Language

viduals and kinds in the world. Reference is a relation between an expression and an object or kind. Para- digms of 'singular reference' are proper names, demonstratives, and definite descriptions; 'singular' because a name—'Chomsky,' for example—refers in a particular use, if it refers at all, to just one object. 'General terms,' 'table,' 'electron,' and the like, refer not to particular objects but to kinds or (as some prefer to think of it) to the properties that all tables or electrons share.
Little in the philosophy of language is uncon- troversial, and the theory of reference is no exception. In a provocatively sceptical work, Stephen Schiffer (1987) has reminded us that the problems with which philosophers have struggled may be artifacts of their conception of language. He thinks that all that is bedrock is:
we humans have noise and mark making proclivities, and, like earthworms and flounders, we survive for some finite period of time in the environments into which we are born.
(Schiffer 1987:1)
The rest is philosophers' theory which, in Schiffer's view, is mostly pretty bad theory. Schiffer is right at least to the extent that we cannot take talk of meaning, or truth, or reference for granted. The ideas that 'Kripke' is the name of Kripke, that 'tiger' refers to tigers, that the sentence the moon is made of green cheese is false, despite their commonsense appearance, involve theoretical ideas about language. So they must be judged by the usual standards (inchoate though they be) that apply to all theories. Because humans have been immersed in language so long, philo- sophical reflection on language must take its point of departure from human responses to this long experi- ence, to what is called 'folk theory.' Philosophers' theories of language are mostly rival attempts to sys- tematize, debug, and extend this folk theory. We should not too prematurely follow Schiffer and resign ourselves to the failure of all these attempts.
As a species we have invested a huge chunk of our cognitive resources in language. Learning a language is probably the most intensive and the most critical achievement of any child, and no child comes empty minded to the job; the human larynx certainly, and the human brain most probably, are adapted to that task. So it seems reasonable to conclude that our noise and marking activities are adaptive, and are so in virtue of some feature(s) of the marks and noises so produced. It is of course a further step to the idea that the function of language is to represent our physical and social world. But we need a theory of those fea- tures that make our investment in language pay, and our only starting point is the theory we have, a theory in which reference and truth play a central role in capturing the relations between language and the world.
2. The Theory of Reference: An Outline of an Ambitious Program
What is reference, and what is its place in an overall account of the nature of language? These questions are best answered by articulating and evaluating an ambitious semantic program, namely one which aims to explain meaning via reference and sentence struc- ture, and then to give an account of reference that appeals to nothing semantic. The ultimate aim of this theory of reference is to explain the relationship between linguistic representation and the world rep- resented. That requires that it rely on no semantic notions, else it presupposes the very relationship to be explained. This program (hereafter the ambitious program) has four enabling assumptions; those who think it too ambitious take it that at least one of these is false.
2.1 The Need for a Theory of Meaning
The first assumption is that what is needed, and can legitimately be looked for, is a theory of meaning. General truths about the nature and organization of language are to be discovered. Thus what is needed, in some broad sense, is a theory. Looking for a theory of meaning makes the working assumption that the folk-theoretic notion of meaning—that notion which informs our unreflective modes of speaking about meaning—is not hopelessly compromised. Thus defenders of the ambitious program suppose that our intuitive notions of meaning will need to be clarified and revised rather than abandoned in favor of some- thing completely different. Neither part of this assumption is uncontroversial. Schiffer, for example, challenges the need for theory, and Quine, among others, is sceptical of the prospects of finding a defi- nition of meaning which is not either vacuous or circular.
2.2 Meaning and Truth-conditions
The second assumption is that one fundamental element of sentence meaning is a sentence's truth- conditions. A minimum condition of successful trans- lation, for example, is that the translating sentence have the same truth-conditions as the sentence trans- lated. Moreover, if the point of language is to enable us to represent our physical and social world, and if truth is our fundamental mode of evaluating rep- resentation, then two sentences sharing truth- conditions share a very significant property. They will be true and false in just the same circumstances. An explanation of a sentence's truth-conditions does not tell us all we need to know about a sentence's meaning. There is some interesting difference between Edward is not very clever and Edward is a dork but both sentences share something very important. The idea that sen- tence meaning is captured through a theory of truth- conditions is not quite as coarse grained as might seem. Even ifCordates are vertebrates and Renatesare
Reference: Philosophical Issues
235