I just re-read Lexical Competence, by Diego Marconi (MIT Press, 1997), a book that deserves to be read more widely than it has been. Here is my spin on it:
Lexical competence is competence in understanding words, a component of general semantic competence (competence in understanding language).
Marconi reviews arguments to the effect that lexical competence is not accounted for by standard model-theoretic or truth-conditional accounts of language. Such accounts give meaning/truth conditions of complex senstences in terms of meaning/truth conditions of atomic sentences, and meaning/truth conditions of atomic sentences in terms of meaning/interpretation of words, but they do not account for the meaning of the words. What they do say–meaning postulates and the like–falls far short of giving the meaning of words.
Given this state of affars, Marconi proposes to shift the focus of our semantic theorizing from meaning to competence. Instead of giving a theory of the meaning of words, Marconi offers a theory of our competence in using words.
He argues that lexical competence divides into inferential and referential competence. Referential competence is our ability to apply words to objects in either direction: to say ‘apple’ in response to an apple (naming) and to pick an apple in response to ‘apple’ (application). Inferential competence is our ability to draw materially good inferences involving words: if x is an apple, then x is a fruit, etc. According to Marconi, referential competence does not reduce to inferential competenece and vice versa. Some of his evidence is that it is possible to have one without the other. He reviews some nice neuropsychological evidence of double dissociation bewteen the two aspects of lexical competence.
One side effect of Marconi’s proposal is to shift our attention from the notion of reference (a somewhat mysterious relation between words (or concepts) and the world) to the notion of referential competence, which, being a relatively observable relation between speakers, words, and the world, appears to be more empirically tractable than reference.
I have some qualms about some aspects of Marconi’s model of semantic competence. Although i read it a long time ago, I’m pretty sure it influenced my thinking about concepts. With Sam Scott, I have written a paper (forthcoming in Philosophy of Science) arguing that concepts should be split into linguistic and nonlinguistic concepts. I think the evidence discussed by Marconi can be reinterpreted within our framework. Referential competence, at least of a basic kind, is subserved by nonlinguistic concepts, whereas inferential competence, at least of a sophisticated kind, is primarily subserved by linguistic competence. If this is correct, it leads to a revision of Marconi’s model. It would be interesting to go through the relevant empirical evidence (including the evidence reviewed by Marconi) and see how well one can develop a model of it within the framework we propose. (If anyone is interested in pursuing this, please contact me.)
More generally, I am unaware of anyone who has developed the theme of lexical competence since Marconi’s book came out about a decade ago, but I’m not an expert in this area. Does anyone know of any recent work along Marconi’s lines?