In considering the evolution of human language, I think it is helpful to contrast these two books, and the ideas presented in them:
Terrence Deacon’s “The Symbolic Species: The Co-evolution of Language and the Brain”
Stephen Pinker’s “The Language Instinct: How the Mind Creates Language (P.S.)”
Neither book is exceptionally new, and in fact, Pinker has cranked out a number of books since The Language Instinct. However, I think The Language Instinct is the best of Pinker’s volumes for this discussion. In it, he lays out the basic evolutionary psychology argument in a way that is most directly contrasted with the ideas in Deacon’s. Also, The Language Instinct has a great chapter called (if memory serves) “The Language Mavens” which is worth reading whether or not you agree with or even like the rest of Pinker’s book.
Pinker’s argument, first laid out in a paper with Bloom in Cosmides, Toobey and Barkow’s volume “The Adapted Mind: Evolutionary Psychology and the Generation of Culture,” is that there is strong evidence that language is an adaptation (i.e., it is costly, complex, etc.) and thus, as an adaptation, is something subject to natural selection. Since language is mainly a cerebral brain function, we can assume that there are brain structures (mainly in the cerebrum) that facilitate language, and that these structures (language “modules,” in the parlance of evolutionary psychologists) are the actual traits under selection. Furthermore, we can assume that these traits are just like any other traits shaped by evolutionary forces, in that there are sets of genes that code for them. In essence, it is these genes that are actually under selection.
Pinker strengthens his argument by citing interesting aspects of language. Most compelling is probably Pinker’s discussion of pidgins and creoles, which he uses to argue that young children seem to be linguistic geniuses. In settings where the adults are speaking a pidgin (a partially formed derivative semi-language made of two or more languages brought together in once community by historical events such as slavery) children, as they grow up, turn the pidgin into a full blown language (a “creole”) with all the usual bells and whistles, including a reasonably extensive lexicon. This is an argument for the “built in” nature of language function because how could this happen were it not deeply encoded by genetic programs? (Or at least, that seems to be the argument.)
Deacon presents a very different argument. The main problem that Deacon has with Pinker is that Pinker’s model (and this applies broadly to evolutionary psychology in general) is biologically impossible. Neural systems, especially in the cerebrum, are not coded for at any important level of detail by sets of genes. Rather, they develop as the brain develops in response to the context in which they are growing. There is abundant evidence to support this view of neurological development. In fact, this view of neural development is pretty much the established central dogma for neurobiology. Deacon was speaking of this sort of thing before this had become so well established, and in areas such as anthropology and evolutionary psychology, where there appears to be a decade or so time lag between thinking about how brains have evolved and knowing about how brains actually develop and function. The point is, Pinker’s model can’t be true because it relies on biological systems that don’t exist.
Essentially, the fact that brains don’t work the way Pinker requires them to work is a sufficiently powerfully killer fact to derail the Pinker/Bloom/Cosmides/Toobey/Barkow paradigm that we need look no further. However, Deacon’s work offers more. It turns out that there are Darwinian processes working at the neurological level, in that the development of the brain systems that we see in adults involves overproduction followed by culling, which is precisely a Darwinian process. Even more interesting (but not, I think, as clear or well established) is Deacon’s assertion that languages themselves, due to the process of Darwinian selection in the symbolic realm, are adapted to be learned by children. Deacon asserts that a language ideal for use by adults may be very different than the language system we actually end up with.
So, children are not really linguistic geniuses. Rather, languages are shaped to be learned by children who are in some ways incredible learning machines, but with roughly the same limitations as adults.
An excellent discussion of Deacon and Pinker can be found on John Hawks’ blog: http://www.johnhawks.net/weblog/topics/minds/deacon.html John mentions that Chomsky’s view of language requires no intermediate forms, which seems implausible yet “is supported by some prominent evolutionists, such as Stephen Jay Gould, who views it as likely that other brain functions requiring symbolic logic were the targets of selection, and that language later arose as an artifact of culture.” Maybe, but I would like to add that in my view, the “fully blown or nothing” view of language is politically motivated and not necessarily well founded by evolutionary biology. Why political? John explains: “[the possibility of intermediate languages] makes it possible that today’s people still vary in their biological capabilities with respect to language, and that selection may still be happening.”
My interpretation of Deacon vs. Pinker leads me to support Deacon’s view over Pinkers in most respects, while John is more equivocal. This may be in part because I have had hours of conversation with Deacon about this, and I share his view (to the extent that one can do this at all) of the symbolic/semiotic side of the argument.
However, John Hawks also points out differences between Deacon’s model and current paleo-neurobiology, based on research done more recently than the publication of The Symbolic Species, with which he (Hawks) is very familiar. So I strongly recommend that the reader visit his site and put some of that stuff in your brain as well.