One of the secondary themes at the Evolang conference in Barcelona this week is that the investigators into the evolution of language are getting too big for their britches. They are overly boastful and do not know what they are really talking about. This theme was repeated this morning (Friday, March 14, 2008) by South African linguist Rudolf Botha as he chastised linguists for talking about evolution in much too breezy a fashion, without really having the data to back up their assertions. He insisted that accounts of language evolution need to be supported by a fuller and more systematic use of the pertinent theory, especially evolutionary theory. In particular, he lamented the off-hand appeal to pre-adaptations, or exaptations as they are known in formal evolutionary theory.
An exaptation is a feature that was formed through natural selection for one function and then was co-opted to serve another role. For example, it is common to read that some aspect of speech that clearly must have a long evolutionary history, such as the fine control humans have over the operation of their tongue, originally evolved for some other, unknown purpose and only in more recent times was used to speak. Researchers who believe that speech is only, say 100 thousand years old or even less, are forced to conclude that almost every biological element supporting speech is an exaptation. (This point has been discussed before on this blog. See: Speech’s Big Bang).
Botha cited as an example Noam Chomsky's proposal that Merge arouse as an exaptation (see: Chomsky's Theory of Language Origins) without troubling himjself to offer any evidence at all to back up the proposal.
Botha was NOT saying that the exaptation arguments are false, only that they need to be more solidly based on real biological evidence. Many linguists defend their vague approach by saying that the “real work” of science does not allow for overly restrictive demands of theory.
Botha pointed out that biologists routinely expect that an argument of exaptation include:
- an account of exactly what got co-opted and where that feature came from,
- an explanation of the causal mechanisms responsible for the co-opting, and
- an account of the co-opted feature’s new biological function.
In defending demands for this level of accounting Botha pointed to several studies in biology that manage to do all of these things. For example an article by biologists David Lytle and Robert Smith (available here) shows how a 150-million-year-old-behavior in water bugs originally evolved to facilitate migration but has since been co-opted in two distantly related species to permit flash-flood escape. The result of their study, which included experiments, was a piece of scientific beauty, and similar thoroughness and theoretical soundness should be more common in the study of language origins.
Botha gave as an example of work that went part way along the proper path work in a book by Carstairs-McCarthy (here) that proposed that the neural organization underliying syllable structure was co-opted to provide a syntax of words. The author provided some evidence at least, but still did not show how the facts force the conclusion of an exaptation.Only with full rigor will accounts of language origins be able to sustain their claims.
Comments