Whoa, Brian!

[Tom Givon]

Dear FUNK people,

I was going to stay out of the recent traffic; till I saw Brian MacW.'s
unfortunate interpretation of what I do or don't think about arbitrary
grammatical structures. So, hopefully as briefly as I can:

1. In ch. 6 of "On Understanding Grammar" (1979) I observed that quite
   often, perhaps most often, synchronically-arbitrary morphosyntactic
   structures arise in the course of perefectly adaptive ('functionally
   motivated') grammaticalization. This is a fact of life one has to learn
   to live with. It does not mean that the *process* of grammaticalization,
   whethger diachronic, developmental or evolutionary, is 'unmotivated'
   or counter-adaptive. But only that in the emergence of complex
   structures, a certain level of arbitrariness comes with the terri-
   tory. Unless, of course, you want to throw a tantrum and ignore it,
   which I concede quite a few FUNKers seem to enjoy doing.

2. Subsequently, in a chapter on biological evolution in "Mind, Code &
   Context" (1989), I noted that a similar phenomenon has long been observed
   in biological evolution of complex organisms, going under the traditional
   name of "excess structure" (S.J. Gould's 'spandrels'). Again, many
   people with an anti-evolutionary philosophical bias have been citing
   such facts to suggest that bio-evolution is not adaptive (not 'function-
   guided'). Many Chomskyites have been embracing this as an anti-adaptive
   argument that would support 'autonomous/arbitrary syntax'. I think, as
   in the case of the diachronic 'spandrels', this attitude is rather
   unfortunate, and no serious evolutionary biologist--not even Steven
   J. himself--would espouse it (tho I have seen some people with PhD in
   Biology espouse it... I guess I don't count them as being serious.).

3. If, as I have been suggesting since 1977, grammar is a partially-
   automated speech-processing system, then one may as well note
   that the rise of automaticity--whether developmentally or phylo-
   genetically--always allows for *some degree* of rigidification,
   ritualization, symbolization, and thus seeming arbitrariness. These
   are some of the *consequences* of establishing rigid neuro-processing
   pathways. It is part of the cost the organism pays for the *substantial*
   benefits of automaticity:
       *higher processing speed
       *lower error rate
       *lesser dependence on context

4. One may also suggest that many features of culture are likewise
   ritualized, to the point where they seem--to the superficial
   observer--rather arbitrary. Again, many anthropologists, since
   Whorf, Boas and Bloomfield but also more recently (cultural
   relativism seems to have regained center-stage in anthropology...)
   take cultural diversity & ritualization as evidence of total
   arbitrariness, non-universality & non-adaptiveness of culture.
   Again, I see this as a cop-out, an escape from a serious empirical
   study of the great wonders of complexity of the human spirit.

All in all, it seems to me that social scientiest would rather not
tackle complexity seriously & empirically. Instead, they keep wanting
to find some escape hatches, so that they could go on with *reductionist*
interprertations of complex domains. In linguistics, this is obvious in
the insistence that the system is either 100% 'emergent' or 100% 'innate',
100% 'motivated' or 100% 'arbitrary'. For as long as this kind of
intellectual escapism prevails, I doubt it that the field would advance
much farther beyond chasing its own tail; which is, in my most-humble
opinion, what has been going on since Bloomfield/Skinner & Chomsky set
up the terms of this sterile debate.


Cheers,  TG