taylor_2016_the_meanings_of_words.pdf

Lexical semantics, or: What does it mean to know the meaning of word? John R. Taylor Draft. To appear in B. Dancygier (ed.), The Cambridge Handbook of Cognitive Linguistics What does it mean to know a word? Or, to put it slightly differently: What does a person who knows a word actually know? And what are the criteria by which we determine that a person does indeed know a word? One aspect of knowing a word is, of course, to know its perceptible form – how it is pronounced and (for literate speakers of a written language) how it is written. Another thing a person must know is what a word means. (There are some further aspects, which I will touch on in due course.) But here’s the rub. Pronunciations are public affairs, and with the technological advances of the last century they can be recorded, played back at will, and spectrographically analysed. Meanings, though, are invisible, lodged inside each speaker’s head, and possibly not even available to introspection, let alone interpersonal validation (Jackendoff 2012). To equate meanings with “concepts” or “ideas” – to say, for example, that the meaning of shrub is the concept SHRUB, or the idea of a shrub – is to skirt around the issue. For unless we are able to say what concepts are, and can give a full and clear characterization of the concepts associated with the words of a language, we are merely playing semantics (in the non-technical sense of the word). Indeed, there is quite a long line of scholars (Bloomfield 1933, Lyons 1968, Quine 1960, Wittgenstein 1978) who have advised against the study of meanings as things in the head. Some have doubted whether it is profitable even to speak of word meanings at all (Austin 1961). The student of word meanings is therefore faced with a major methodological problem. Words, we are convinced, do have the property of meaningfulness, otherwise we should not be able to use them to create meaningful utterances. Yet there is no direct way of accessing thi