I see that Derek Bickerton has a paper in the latest issue of Biolinguistics (PDF here). Bickerton is an interesting man who has done major work on pidgins and creoles, and he has thought hard and creatively about languag origins. He is interesting too because he combines an excellent knowledge of generative linguistics with a questioning ear. As he puts it, "Fears [are] widespread… that biolinguistics may turn out to be merely a more scientific-sounding term for generative minimalism…" [p. 73]. That is to say, he fears that biolinguistics, like generative linguistics, will continue to ignore biology and biological methods. He examines several issues in the paper, but in this posting I shall limit myself to natural selection.
The same issue of Biolinguistics reviews a new book by Bickerton and the review authors imply that Bickerton agrees with them in downplaying natural selection. Nobody reading the Bickerton paper is likely to make that mistake, but it is striking the number of generative linguists who doubt or outright deny natural selection's role in language origins and functioning.
If there is any presupposition I've brought to this blog and maintained throughout, it is the role of natural selection. Bickerton's paper reminds me of several assumptions that flow from this basic one:
- Languages can vary
Natural selection is a process that tests generations according to the variation within members of a population. The varieties that are less well-adapted to the needs of the moment fail the test. A rival process is called drift, weeding out individuals randomly.
Why should I prefer selection to drift? Drift is very common, perhaps accounting for more traits in any individual than selection. Drift might account for, say, the difference in patterns of zebra striping in Grant's and Burchell's species. It does not account for complicated systems like vision. Why not? The odds against random events alone producing an elaborate organ like the eyeball are too great to be endured. Add a functioning link like the optic nerve, plus a useful region of the brain like the visual cortex and there aren't enough atoms in the universe to support the probabilities. The only way a complex system can evolve is for a simpler system to emerge and for more complex varieties to be selected as they arise and prove superior. (Bickerton points out that evolution can also simplify, but that is irrelevant in this discussion.)
Chomsky famously denies that there is significant variation in language, or at least the kind of language that he is interested in:
The term 'language' as used in this context means internal language, sometimes called "I-language," the computational system of the mind/brain that generates structured expressions, each of which can be taken to be a set of instructions for the interface systems within which the faculty of language is embedded. [quoted from here]
In short, we all share an I-language which has no significant variation around the world. I have never believed in that proposition, which is why this blog is about the origins of speech, a behavior, rather than language, an abstraction.
Originally, I suspected that this denial of variation eliminated Chomsky from contributing anything fundamental to the story of language origins, but I have modified my views. These days I believe that language rests on perception and attention, most of which is older than the Homo genus and that, as far as human evolution is concerned. has not varied signifcantly. Perception thus does provide us all with a universal grammar that is readily translatable.
Variation, however, did matter after we began using language to share perceptions. Language today lets us say much more than perceivables like cat or little, so the details of how we evolved the ability share perceptions would show variety.
- Language is an adaptation
It was in early 1970 that I first realized language must have an evolutionary history, and therefore must have had some reason for being selected. I saw right away that this would be a hard one. Language seems so useful that almost any reason might explain it, except that it would seem useful to other animals too. Why don't they also use language?
The answer I currently favor is that when African woodlands gave way to open savannas animals of many species had to adapt or get the heck out of there. Our ancestors adapted by becoming more communal, trusting in one another to a greater degree than was previously prudent. Language was one more adaptation to this new, cooperative world.
The generative approach sees language's public role as secondary (the primary role being thought and planning) and dependent on the I-language. Again, I would not disagree if you think of the I-language as the perceptual system, but Chomsky reverses the process. Instead of translating perceptions into words, the interface that makes I-language public translates words into perceptions. ("the sensorimotor systems … externalize expressions in production and assign them to sensory data in perception")
Bickerton reports that during the peer-review process for a paper an anonymous reviewer objected to his "plain radical externalism" . If natural selection has played a role in linguistic history, the story will inevitably be about the adaptation to externals rather than the perfecting of processes according to some non-environmental measure.
- Language evolved in stages
Bickerton's contributions to linguistics includes the notion of a protolanguage, a form of speech that preceded contemporary language. Chomsky and others have rejected the idea on the grounds that language is an all-or-nothing proposition. Bickerton laments,
I can think of nothing more likely to create a barrier between biolinguists and a majority of biologists than the former's insistence that language emerged ready-made "pretty much as we know it today" .
Along with the blindness to evolutionary processes, the denial of stages makes the study of linguistics clinically useless. Language problems are common enough in the pediatric world and can persist through life. Strokes introduce many specific language problems. Many people are looking to understand what goes wrong, but generative insistence on the whole pie at once offers no help in supplying medical or teaching hypotheses.
Another point Bickerton makes is that by denying that language evolved over a period of time, linguists trivialize the role of language in the evolution of humans:
The evolution of language must have taken place during the evolution of humans, as a part of that evolution, and indeed, given its importance in their subsequent development, as arguably the most important part of that evolution. In fact, surprisingly little of the literature, biolinguistic or other, makes any serious attempt to place language evolution in the context of human evolution. But even in that company, Hauser et al. (2002) stands out as being perhaps the only work on the evolution of language that includes not a single word about how humans evolved. [83-84]
If you say that the only peculiarly human feature of language evolved at a stroke, and that this stroke came fairly late in the story (e.g., when Homo sapiens was already a hundred thousand years old) then all the other parts of human evolution that are older—the tool making, the use of fire, the enormous growth in brain size, the ability to adapt to niches across Africa, Asia, and Europe, the intense cooperation and communal life styles, the introduction of child and adolescent stages in the life-cycle, and the ability to survive on the open savanna—happened entirely without language. Of course, it may be that such was the historical case, but it sounds unlikely and should not be asserted a priori.
Alternatives to Natural Selection
If you don't like natural selection and are unwilling to call yourself a creationist, what do you do? You can blame it all on drift, but creationism is more probable. Another attempt at a solution points to all the new things that have been added to our understanding of heredity and development in recent years. Epigenetics, evo-devo, and self-organization processes are especially popular as alternatives natural selection. Bickerton argues that this response is a "category mistake" . That is to say, it is wrong to suppose epigenetics is an alternative to natural selection, just as it would be wrong to suppose that the horse is an alternative to transport.
The idea of natural selection was Darwin's great contribution. He argued that traits were selected, but he did not know what determined those traits. Early in the 20th century, Mendel's theory of genes was discovered by biologists, and a key idea in evolutionary theory became that the gene was what was selected. The so-called Modern Synthesis put the whole emphasis on genes. But in the last few years it has become apparent that the Modern Synthesis is going to need modification to take into account things like epigenetics, behavioral pushes, and automatic organization. The gene is only part of the evolutionary process. The story is more complicated than the Modern Synthesis supposed, but that is not a strike against natural selection.
Epigenetics etcetera are alternatives to genes, not to natural selection. Biologists used to think that genes were the only sources of individual variety. Now we know there are many sources, but evolution is still the same old story, a testing of individuals to determine which varieties are passed on to the next generation. Thus, theories of language that deny it has variety, or that it is an adaptation to external conditions, or that it evolved in stages have nothing to say to anyone who thinks language has a biological component and wonders how that component may have arisen.