Defending Literal Meaning

The notion of literal meaning was once so unproblematic that it would easily fit Arnauld’s description: it was one of those notions that are “so clear that they cannot be explained through others, because there are none which are clearer and simpler than them.” But in our century, its fate has not been different from that of so many other notions previously presumed to be “fundamental ,” and it has been severely challenged. Literal meaning has come under attack in successive waves. First by philosophers, then by linguists and, more recently, by psychologists. Having discussed elsewhere some of the philosophically and linguistically based attacks against literal meaning, I will address myself here to a recent psychological argument that purports to provide empirical evidence to the effect that literal meanings play no role in the process of language comprehension and hence cannot be claimed to have “psychological reality” (Gibbs, 1984). Since there is no way to disentangle completely conceptual and empirical issues, it will be necessary first to sketch an alternative conception of literal meaning (Section 2) which will be able to cope with most of the theoretical points raised by Gibbs and others against literal meaning (Section 3). Then, I will consider the empirical evidence that allegedly shows that this notion has no psychological validity and try to rebuff this allegation (Section 4). The next step will be to provide some positive evidence for the “psychological reality” of literal meanings (Section 5). I will conclude by outlining the implications of the discussion for such issues as the theory of metaphor and the distinction between semantics and pragmatics (Section 6).