predictions

[Brian MacWhinney]

Dear FUNKnet,

It seems to me that it is a free country and anyone can say anything that
they want, as long as it is not libelous.  I'm sure that when Fritz was
being interviewed he told the reporter that linguists and psycholinguists
disagreed sharply on the interpretation of the genetic data.  And probably
the reporter just decided to ignore his remarks on that issue.  And
undoubtedly Fritz, like many of us who have been in a similar position, was
shocked to see how his story was reported.

I'm rather more interested in Fritz's statement that he would never claim
that there would be a gene for the subjunctive.  It would have seemed to me
that UG might very well predict that there should be a gene that controls
ability to mark or not mark mood shift on verbs, as triggered by operations
involved in subordinate clauses.  Aren't there various proposals for how
things like this get handled in LF that involve certain formal
manipulations, some of which may be more marked and complex than others?  I
would expect that a weak local neural dissociation of the type reported in
the Jaeger et al. article in Language for the past tense would also evident
for the subjunctive.  Probably there are areas of the brain that light up
during processing of the verb in "If I were to go to the store".  Assuming
that such results could be obtained, why wouldn't we also expect to find
genes that control this "skill" or "module".  Perhaps there are families
that use the subjunctive in certain contexts, but omit it others.  Wouldn't
this suggest a dominant heritability pattern, as is argued for the past
tense gene in the family studied by Gopnik and Vargha-Khadem?

Considered more broadly, can we extract at this point in 1997 a specific
set of predictions from UG regarding which linguistic abilities should be
genetically linked and which should not?  Perhaps there are even some
functionalists who would like to offer parallel predictions.  If the
theories are not sharp enough to offer these predictions a priori, then
what scientific sense can be made of post hoc attempts to seek for
confirmation in sporadic reports from isolated, incompletely reported, gene
lines?

Fodor presented a framework for characterizing cognitive modules and it
appears that Chomsky adopted that framework during the 1980's.  But no one
has used those efforts to either generate genetic predictions or at least
formulate a method for systematically generating genetic predictions.


--Brian MacWhinney