Besides serving us at the growing edge of science and beyond, metaphor figures even in our first learning of language; or, if not quite metaphor, something akin to it. We hear a word or phrase on some occasion, or by chance we babble a fair approximate ourselves on what happens to be a pat occasion and are applauded for it. On a later occasion, then, one that resembles the first occasion by our lights, we repeat the expression. Resemblance of occasions (...) is what matters, here as in metaphor. We generalize our application of the expression by degrees of subjective resemblance of occasions, until we discover from other people's behavior that we have pushed analogy too far and exceeded the established usage. If the crux of metaphor is creative extension through analogy, then we have forged a metaphor at each succeeding application of that early word of phrase. These primitive metaphors differ from the deliberate and sophisticated ones, however, in that they accrete directly to our growing store of standard usage. They are metaphors stillborn. It is a mistake, then, to think of linguistic usage as literalistic in its main body and metaphorical in its trimming. Metaphor, or something like it, governs both the growth of language and our acquisition of it. What comes as a subsequent refinement is rather cognitive discourse itself, at its most dryly literal. The neatly worked inner stretches of science are an open space in the tropical jungle, created by clearing tropes away. W. V. Quine is the Edgar Pierce professor emeritus of philosophy at Harvard University. His many influential works include Methods of Logic, Word and Object, and The Roots of Reference. "On the Nature of Moral Values," his previous contribution to Critical Inquiry, appeared in the Spring 1979 issue. (shrink)