Kripke’s claptrap on proper names — 4

by pieterseuren

Kripke was well aware of the problem of reference to nonexistent entities when he presented his lectures in 1972. He said: “I deliberately ignore delicate questions arising from the possible nonexistence of an object” (p.21) and “Unfortunately, this […] involves us in the problem of singular attributions of existence, one I cannot discuss here” (p. 110). In fact, in 1973 he devoted the six John Locke lectures he was invited to give in Oxford to precisely this problem, trying to come to terms with it, but these lectures were anything but convincing. The audience dwindled rapidly (I was one of the dwindlers) and the lectures were never published—that is, not until a month ago when, to my utter surprise, OUP New York brought them out as a little book under the title Reference and Existence: The John Locke Lectures. I haven’t been able yet to get hold of a copy, as the book is only just out, but OUP allows for a preliminary peek of the first eleven pages on the internet. There one reads—which is also what I remember—that Kripke tries to account for reference to nonexistent entities, whether by means of a proper name or a DRP, by invoking the notions of either ‘pretended’ or ‘mistaken’ reference. Yet even if one pretends to make a reference, or is mistaken in making one (in that one thinks the reference object exists whereas it doesn’t), the reference must be with regard to something, and, as far as I could make out, Kripke does not specify the ontological status of that nonexistent ‘something’. Anyway, I may revert to this topic after I have read the book.

The upshot is that this whole theory of proper names as rigid designators is a philosophical farce that has nothing to do with the reality of language or indeed with serious philosophy. So why not call it by its proper name: claptrap? This rather sobering conclusion is reinforced by the fact that Kripke’s cherished a posteriori necessity should, by his own reckoning, be instantiated not only by the use of proper names but also by that of DRPs. Suppose I, a newcomer to the village, come across a bothersome drunkard in the local pub on my first Saturday night out. Then, the next morning I discover to my horror that this drunkard is in fact the local vicar. This allows me to say in truth, on Sunday morning: The drunkard I met last night in the pub is the local vicar. Note that I have used two DRPs, not two proper names, in making my true identity statement. The statement is made true by the fact that the reference object of the first DRP happens to coincide with that of the second. Given this fact, my statement must, in Kripke’s book, be (a) necessary because, on metaphysical grounds, the man in question could not have been another man, and (b) a posteriori because I discovered the identity empirically. So Kripke does not need proper names as rigid designators at all to promote his notion of a posteriori necessity: once reference is fixed for any DRP, every true identity statement (except one with two identical terms) is, or should be, by his reckoning, necessary a posteriori. And reference is fixed in language use, not in the underlying language system. If Kripke chooses to insist on his metaphysics, then the correct generalisation is that once reference is fixed for two referring terms, no matter how or when or for how long, then true statements asserting the identity of the reference object of the two DRPs must, for Kripke, be both necessarily and a posteriori true. Kripke’s basic fault is that he wants to assign the property of “necessary a posteriori truth” to sentence types, not to utterance tokens (statements). But reference is a property of utterance tokens, not of sentence types. Therefore, truth and falsity, which can only be established after all reference relations have been fixed, are properties adhering not to sentence types but to utterance tokens, in particular, of course, statements (see my Language in Cognition, OUP, 2009, pp. 97–101).

Describing a language is a type-level, not a token-level activity. Grammars, lexicons and phonological descriptions specify languages as tools, they do not specify all possible uses of these tools, though they may specify that certain linguistic forms are reserved for special classes of use. Kripke, following the mainstream of American philosophy, fails to make that distinction, which leads to terrible confusions. At least as far as proper names are concerned, he seems to assume that reference values are to be registered in a description of a language. On p. 77 we read:

When I say that a designator is rigid, and designates the same thing in all possible worlds, I mean that, as used in our language, it stands for that thing, when we talk about counterfactual situations. I don’t mean, of course, that there mightn’t be counterfactual situations in which in the other possible worlds people actually spoke a different language. […] But still, in describing that world, we use English with our meanings and our references. (italics original)

On p. 101 he quotes the American philosopher Ruth Marcus: “Being present at that discussion, I remember that she advocated the view that if you really have names, a good dictionary should be able to tell you whether they have the same reference.” From the subsequent, rather woolly, text one is led to infer, with some difficulty, that he rejects this view, not, however, because of its linguistic absurdity but maybe because he seems to think that his notion of a posteriori necessity avoids Marcus’s unwanted conclusion. Be that as it may, there can’t be any doubt that the description of a language, that is, of its lexicon, grammar and phonology, does not require a specification of all possible reference objects, whether of DRPs or of proper names: such a requirement would make it impossible to describe a language. It is not part of English that the name Socrates is mostly used, in academic contexts, to refer to that specific well-known fifth-century BCE Athenian philosopher. People who don’t know this can’t be helped by a remedial English language course but they can be by taking lessons in the history of philosophy. The description of a language shows (or presupposes) the systematic means that can be used in a language to achieve reference in each situation of actual speech. Specifying referents of proper names for entities that are known world-wide is the task of encyclopedias, not of dictionaries, which take proper names only in so far as they are language-specific or because of their grammatical gender (for example, Sahara is masculine in French, Spanish, Italian and Portuguese) or other linguistic details. In general and as a matter of principle, reference fixing is situation- and discourse-restricted and bets on the interlocutor’s correct reconstruction of the speaker’s intended reference. It is part of the use of a language, not of the language itself.

The moral of the story is that one should not be intimidated by the pomp and circumstance with which the products of new academic gods are presented to the world at large. On the contrary, the halo bestowed on new academic gods more often than not blunts their self-criticism and thus makes them produce ponderous claptrap. At the same time, that very halo makes it hard for less august mortals to see through the feebleness of the claptrap and even harder to make others see through it. Therefore, the greater the halo, the greater the reservations one should have.