subjectmatter; at leasttheformer is subjecttofocalbrain damage and childhoodimpairment.Experts at chess and (in
my own case) playing a musicalinstrumentexhibita great deal of structured and automatized behavior. But we do not
believe there are innate specializations for reading, driving, chess, and clarinet-playing. So why single out language as
innate?
I think this argument has to be answeredfirst of all by rejecting Fodor's classification of all these characteristics as
together symptomaticofa processingmodule.Anywell-practicedand overlearnedability seemstobeautomaticand to
have specialized structure to some degree. So we may grant that reading, driving, chess, and clarinet-playing can all
behave online like Fodorian modules. However, the issue of innateness of language does not concern how language is
processed online: it concerns how language is acquired. Hence the comparable question to ask about these other
abilities is: What must we give learners in advance in order for them to be able to overlearn these abilities, and how
much of this learning follows automatically from cognitivecapacities theywouldhave anyway? In order to answer this
question, we must determine exactly what they have overlearned. We have some idea of how language is structured,
how hard it wouldbe to learn fro mscratch, and how children do actuallylearn it. We have no co mparable analysis for
these other activities. And, as stressed in section 4.5, we have no comparableanalysisfor any of the perceptual, motor,
or cognitive capacities in which these could be embedded.
Inshort, Elmanetal. are righttoask why languageshouldbedifferentfromother overlearnedabilities, manyofwhich
are unlikely to have a direct narrow innate basis. However, I think the question is not just rhetorical: before a proper
comparison can be made, there is much empirical work to be done on other abilities as well as on language.
Afinalstep isrequiredinthegeneticgrounding ofUniversal Grammar. Ifthereisa geneticbasisfor languagelearning,
not present in apes, where did it come from? The only reasonable possibility is through evolution. Chapter 8, will take
up the question of possible evolutionary routes to modern language. For the moment, let me just note that Chomsky
points out thebalance between learning and evolution(1965: 59):“There is surely no reason today for takingseriously
a position that attributes a complex human achievement entirely to months (or at most years) of experience, rather
than to millions of years of evolution....”That is, the more properties of language we can attribute to evolution, the
easier language acquisition is for the child.
But Chomsky immediately hedges his bets on evolutionary justification of Universal Grammar, and continues:“or to
principles of neural organization that may be even more deeply grounded in physical law.”Though logically possible,
this alternative declines to follow the argument through to its inexorable