On a Milestone of Empiricism

Abstract

Quine has presented powerful arguments against the intelligibility of any statement involving quantification into the scope of a non-extensional sentence connective purporting to express logical necessity, logical possibility, or strict implication. He has also presented a fully general argument against the possibility of satisfactory non-extensional connectives. It is easy to be lured into thinking that if definite descriptions (and also, perhaps, proper names) are analysed in accordance with Russell’s Theory of Descriptions — an idea to which Quine is highly sympathetic — then facts about substitution and deduction in primitive notation yield all that is needed to circumvent Quine’s formal arguments. I have claimed as much myself, and so have Carnap, Church, Fitch, Føllesdal, Kripke, Marcus, Myhill, Prior, Sharvy, Smullyan, and others. But as far as logical modality is concerned this is a mistake. Only Quine’s general argument is thus undermined, and seeing why leads to an interesting logico-semantic hypothesis, suggested by Quine and Gödel, and, as far as I know, hitherto unproved. Furthermore, it would seem that the only way to answer Quine’s original challenge to quantified modal logic is to make the controversial assumption that names of the same object are synonymous and hence intersubstitutable in all contexts. But even this does not provide a full answer; and in fact what has happened is that ideas in the work of Føllesdal, Hintikka, Kanger, Kripke, and Quine, have resulted in a metamorphosis of subject matter.

Thus the theory of description matters most. It is the theory of the word for those For whom the word is the making of the world, The buzzing world and lisping firmament.- Wallace Stevens