note to Liz Bates

[Tom Givon]

Liz,
I don't know if you are familiar with a series of works by Posner and
his associates (Petersen, Raichle etc.) on the development of a reading
module in the pre-striate areas of the left occipital lobe. The location
is of course quite suggestive, being within the object regognition
(ventral) stream (cf. Miskin and Ungerleider etc.). But the adaptation of
that loication to a reading task seems to be, in itself, highly domain
specific. Your 'weaker' position on domain specificity need not be quite
as strong. In terms of evolution, it is very clear that all language-related
modules were initially specialized to do other things. And even that while
they process some aspects of language now, they may continue to do those
"daytime" tasks. But it is still an open question whether in their capacity
of language processors, they have or have not been restructured (reconfigured)
in a highly domain-specific way. I think this is another area where one ought
to resist rigid positions. At least two areas that are quitessential "processors"
in language -- phonology and grammar -- exhibit enough unique characteristics
to suggest that at least the mode of processing (if not the location) is
rather unique and domain specific. This is not as extreme as the Chomskyite
dogma. But I think, in evolutionary terms, it may be viewed as an intermediate
stage, somwhere between a totally domain-general module and a totally domain-
specific module. Since the evolution of phonology and grammar are, most likely,
the latest evolutionary additions to the array of capacities that combine in
supporting human communication, finding them organized in such an "early"
fashion should not be all that surprising.
Best, TG