A commentary on Terrence W. Deacon, The Symbolic Species: The Co-Evolution of Language and the Brain (W.W. Norton & Company, 1997).

copyright © Mark Turner, 1997
Bostonia, Spring 1998, Number 1, 72-73.

Imagine a mutant being, genetically gifted to paint like Vermeer, born into a culture where no one else can even doodle with a stick. That is the classic Chomskyan view of the origin of language: by genetic accident, astounding special language abilities were inserted into the human brain. In The Symbolic Species, Terrence Deacon, professor of anthropology at Boston University, offers an alternative picture. Language, he argues, is not an instinct and there is no genetically installed linguistic black box in our brains. Language arose slowly through cognitive and cultural inventiveness. Two million years ago, australopithecines, equipped with nonlinguistic ape-like mental abilities, struggled to assemble, by fits and starts, an extremely crude symbolic system - fragile, difficult to learn, inefficient, slow, inflexible, and tied to ritual representation of social contracts like marriage. We would not have recognized it as language.

This view reminds me of something Paul Valéry said in The Art of Poetry: "Poetry can be recognized by its ability to get us to reproduce it in its own form: it stimulates us to reconstruct it identically." Poetry so thoroughly harmonizes with the predispositions of the human brain that it flows into the brain and occupies it, sowing there the seed of its own replication. In Deacon's view, language has evolved to become poetry for the newborn brain.

The second, subordinate means by which language improved, in Deacon's view, had to do with changes in the brain. Crude and difficult language imposed the persistent cognitive burden of erecting and maintaining a relational network of symbols. In that demanding environment, genetic variations that rendered brains more adept at language were favored. Language began as a cognitive adaptation. Genetic assimilation then eased some of the burden. Cognitive effort and genetic assimilation interacted as language and brain co-evolved.

The notion that a magical genetic black box for language was inserted into brains otherwise essentially like our own has become the mainstream view, but it was originally an afterthought on the part of Chomskyan grammarians who were primarily interested in analyzing the formal structures of language as we know it. Chomsky acknowledges the tenuousness of the evidence for speculating about the origin of language. "You can," he says in lectures, "tell any story you like." Actually, he shows special dislike for some stories, including the proposal (by, for example, Stephen Pinker and Paul Bloom) that language originated by natural selection on initial genetic variation. Elsewhere, I have argued that Pinker and Bloom's genes-first adaptationist account is weak because it does not offer a plausible "environment of evolutionary adaptiveness" for language, that is, an environment in which the first lone genetically grammatical person would have enjoyed reproductive advantage.

The most obvious environment in which genetic endowments useful for language would have been favored is a community of people whose members had already invented rudimentary language by drawing on their pre-existing cognitive abilities, the crucial one being, in my view, "parable" - the conceptual projection of small stories. Rudimentary grammatical categories (like "verb") arose by projection from basic elements of stories (like "action").

Pinker and Bloom cannot propose that language is a cognitive invention that underwent genetic assimilation (they do see a place for genetic assimilation once normal hearers began striving to comprehend the dazzling linguistic productions of genetically blessed speakers) because they think genetic specialization for language must have begun the process ("There must have been a series of steps leading from no language at all to language as we now find it, each step small enough to have been produced by a random mutation or recombination"). Against the Vermeer-from-a-magic-hat grammarians and against the genes-first adaptationists, Deacon argues that language was a cognitive and cultural invention that underwent genetic assimilation. He adduces a wealth of anthropological, paleontological, and neurobiological evidence. Language, he argues, was "acquired with the aid of flexible ape-learning abilities." It was grafted onto an apelike brain. It is not walled off from other cognitive functions such as interpreting and reasoning. Grammatical form is not independent of conceptual meaning. There is no linguistic black box and there was no insertion.

Genetic assimilation (known as "Baldwinian evolution") could not, Deacon argues, put patterns of universal grammar into brains. To be subject to natural selection, basic grammatical operations would have to be realized invariantly in neural circuitry across the entire human population, but they are not. "Those aspects of language that many linguists would rank most likely to be part of a Universal Grammar are precisely those that are ineligible to participate in Baldwinian evolution! If there are innate rules of grammar in the minds of human infants, then they could not have gotten there by genetic assimilation, only by miraculous accident." Instead, genetic assimilation built new wetware, largely in the area of prefrontal cortex, that assisted attention, memory, and association, consequently easing the burden of language. These neurobiological changes were "a direct consequence of the use of words. . . . [T]he major structural and functional innovations that make human brains capable of unprecedented mental feats evolved in response to the use of something as abstract and virtual as the power of words." "An idea," says Deacon, "changed the brain."

The Symbolic Species is a dose of vitamin C for a field with a chronic head cold. Theoretical linguistics, as a scientific enterprise, is locked into an unfortunate pattern in which theory and evidence define each other so thoroughly that opposing camps feel comfortable dismissing rather than confronting one another. Two linguists of the greatest eminence and learning can disagree about fundamental questions (does syntax depend on meaning?) without submitting to a generally respected method for deciding which, if either, of them is right. The best hope for springing linguistics from this disciplinary bind lies in bringing to bear other human sciences like anthropology, cognitive science, and neuroscience. Deacon leads the way.