Lexical concepts must be richand specific enough to playtheir necessary role in thismix. It may be possible to reduce
the richness of lexical concepts by enriching the theories of the other three components, but in the end it is necessary
to show that the contributions of the four components add up to the full interpretation.
11.2, The prospects for decomposition into primitives
Let us begin by focusing on the necessity for lexical concepts to be learned. Nearly everyone thinks that learning
anythingconsists of constructing it fro mpreviously known parts, using previously known means of co mbination. If we
trace the learning process back and ask where the previously known parts came from, and wheretheirpreviously
known parts came from, eventually we have to arrive at a point where the most basic parts arenotlearned: they are
giventothelearner genetically, byvirtueof thecharacterof braindevelopment.We wentthrough thistropein Chapter
4; I hope nothing further need be said. Applying this view to lexical learning, we conclude that lexical concepts must
have a compositional structure, and that the word learner's f-mind is putting meanings together from smaller parts.^171
The major exception to this view of lexical learning is Jerry Fodor, who takes the position (e.g. 1975; 1998) that all
lexical concepts (or at least all morphologically simple lexical concepts) are innate. This includes, for instance, the
meanings oftelephone, carburetor, merlot, soffit, andyarmulka:even if some of these words are absent from someone's
vocabulary, their meanings are present in his or her f-mind, just waiting to be triggered. Fodor arrives at this position
because he maintains that lexical concepts are monadic: theyhaveno parts, hence no parts that could be previously
known. Hence he is forced to say that all lexical concepts are innate. He never addresses the question of how they
could all be coded genetically, nor how evolution could have come up with them. I take it that this is a position we
want to avoid.
Fodor believes that lexical concepts have no parts because he assumes an extremely limited view of what the parts
could be and how they could combine. His argument that lexical concepts are monadic stems from a paper entitled
“Against Definitions”(Fodor et al. 1980). The title of the paper gives the game away. Observing (correctly) that one
usually cannotfind airtight definitions that work all of the time, Fodor concludes that word meanings cannot be
decomposed.
334 SEMANTIC AND CONCEPTUAL FOUNDATIONS
(^171) By attributing the composition of word meanings to the learner's f-mind, I am distancing myself from the claim that the word learner is putting pieces togetherconsciously,
which is surely false. We are as usual talking in functional terms here.