January 12, 2004

The Language Instinct: The New Science of Language and Mind

SWET members are all professional language users; we are paid to express our thoughts, or those of others, in coherent terms. Our common professional tool is language, and it is a matter of professional pride to many of us that we ‘understand’ language and use it ‘properly.’ Any work on language as an abstract concept therefore should interest at least some SWET members, even if we are not students of formal linguistics. Such a work is The Language Instinct by Steven Pinker, first published in 1994, purporting to do for linguistic theory what Richard Dawkins’s The Selfish Gene (1976) did for evolutionary biology (Dawkins is quoted on the cover of Pinker’s book as expressing approval of Pinker’s work). Although this is not a new book, any book on linguistics raises some interesting points, many of which would appear to have some relevance for SWET members. In addition, I have rarely read a book on science which has aroused such emotion in me and made me think so hard about the role of science and scientists’this ‘review’ is therefore subjective in that the reviewer and his reactions play some part in the story.

Pinker has acquired a reputation as a popularizer of psychological concepts, chiefly those concerned with language. This book, in a whirlwind 400 or so pages, takes the reader on a tour of (inter alia) linguistic development in children, creolization, a modified form of Chomskian deep structure, phonetics, and comparative linguistics, including the evolution of languages. The treatment of these subjects is intended to lead the reader to Pinker’s central premise, which is (grossly simplified) that linguistic ability is a genetic predisposition of Homo sapiens, and Homo sapiens alone, in the same way that building hexagonal brood cells is a genetic predisposition of honeybees.

Along the way, he seems to take delight in debunking the work of those who have gone before him in linguistics (and other fields). For example, everyone ‘knows’ that Eskimos have twenty (or is it fifty?) words for snow. Pinker restates this ‘fact’ as an urban legend and traces it back to a report by the anthropologist Franz Boas which mentioned four unrelated root words for ‘snow,’ subsequently reported by Benjamin Whorf as seven, and then later inflated by others. The explosion of this myth occurred only in 1986, claims Pinker. Nor is this his only attack on Whorf’s conclusions’such well-known linguistic ‘facts’ as ‘the Hopi have no words which correlate with Western notions of time’ are also examined and purportedly dismissed. I say ‘purportedly’ because on examination of other writings on the subject, it is not clear to me that Whorf was entirely wrong in his analysis of Hopi thought processes (for an example of such writings, see Nicholas Yee, ‘[url=www.nickyee.com/ponder/whorf.html]What Whorf Really Said.[/url]’

Pinker’s analysis of ‘talking’ chimpanzees is also an exercise in setting the record (as he sees it) straight; in order to make his point that language ability is a human-only genetic predisposition, he must disprove the claims that chimpanzees can be taught to use language. First, he examines the evidence for chimpanzees that have purportedly been taught to use language by humans and concludes that in many cases the experimenters were self-deceived in their interpretation of sounds or gestures, or even deliberately excluded or misrepresented data that went against their preconceived conclusions and subsequently reported that language was being used. He also compares the efforts of these chimps with those of human children learning language, analyzes the respective mistakes (children’s language acquisition is one of his specialties), and concludes that human infants are more predisposed to a natural grammar than are chimpanzees.

The anthropologist Margaret Mead makes an appearance as a feet-of-clay idol, supposedly exposed as the victim of Samoan teenagers’ hoaxing in her book Coming of Age in Samoa (1928). He cites ‘Derek Freeman’s 1983 bombshell [Margaret Mead and Samoa]’ to prove his point. Not being familiar with the reactions to Freeman’s work, I did a Web search and was somewhat surprised to find that these conclusions are not unreservedly accepted, twenty years or more later, even though Mead herself is not recognized as a fountain of absolute truth (the 2002 Encyclopædia Britannica, for example, simply mentions that Mead’s ‘position [on cultural determinism] . . . caused some later 20th-century anthropologists to question both the accuracy of her observations and the soundness of her conclusions’). Like Pinker himself, Freeman seems to take more pleasure (and has done so for many years, apparently) in upsetting the academic applecart than in advancing the course of truth’the jury still appears to be out on Freeman’s criticism of Mead, and his work, though important in that it has forced a re-evaluation of the field methodology of Mead and other early anthropologists, may not be as much of a bombshell as Pinker would like us to believe.

Politics plays a part in this self-casting as an enfant terrible; as mentioned earlier, a modified form of Chomsky’s deep structure is introduced and approved (though it is only acknowledged as such some time into the explanation), and it is obvious that Pinker is following the Chomskian party line on various issues, such as the disagreement with cultural anthropology and linguistics. Obviously, if language is to be regarded as a natural faculty of humanity, as Pinker believes, it cannot also be the tool that shapes humanity (to a greater or lesser extent, both Whorf and Mead can be said to have held this view)‘it must be humanity that is to be the shaper of language. Unfortunately, Pinker has not proved this point in this book (to my satisfaction, at least), regardless of how many Eskimo words for snow exist, whether Hopi Indians have a word for ‘tomorrow’ or not, or whether Samoan teenagers engaged in promiscuous sex in the 1920s.

Obviously, science, especially a ‘soft’ science like Pinker’s field of developmental linguistics, is not a matter of black and white facts, but encompasses many gray areas. Scientists are entitled to their own theories and opinions, and yesterday’s received wisdom is often shown to be out of date, founded on erroneous interpretations of data, or even on false data. No one argues with this; it is the way that science proceeds. What worries me about this particular book is that controversial assertions (and it appears that some of these assertions are more, not less, open to contradiction now than when Pinker repeated them) are repeated as fact’exactly what Pinker himself claims to be attacking in many cases. Pinker’s evidence is carefully selected, often asserted with little supporting evidence, and thin ice is rapidly glossed over.

Despite all this, I do not find myself disagreeing radically with Pinker’s conclusions. But like so many secular evolutionists, he is religiously fanatical about his unbelief, and the ideological bias shows through all his work in a way which irritates me in a way that Richard Dawkins’s atheism, for example’eloquently expressed, rationally argued (and incidentally a nonbelief I do not share)‘does not.

It is when he comes to the ‘language mavens’ that I find Pinker to be at his most irritating. He argues that the complaints of ‘language mavens’‘such as some correspondents on the SWET-L mailing list’who decry the ‘misuse’ of language, are as meaningless as those of naturalists who state that pandas are using the ‘wrong’ paw with which to eat their bamboo shoots.

The anti-Whorfian bias is explicitly stated here, meaning in this case that human beings should be in control of the language that they use, rather than language ‘rules’ governing the way in which humans speak. Put this way, the thesis seems reasonable, but Pinker’s tone here (as elsewhere) tends toward the smug and patronizing, with more than a hint of the smartass about it.

Indeed, this air of superiority is irritating to me, when compared with the writing of some modern great explainers, such as Dawkins, Stephen Jay Gould (The Panda’s Thumb, 1980), Stephen Hawking (A Brief History of Time, 1988), James Gleick (Genius, 1992), or Oliver Sacks (Awakenings, 1976, The Man Who Mistook His Wife for a Hat, 1986).

To be frank, Pinker is not a good writer. His style, by turns pompous and mock’self-deprecating, grates on my nerves, and his manner of presenting facts (and opinions masquerading as facts) does not lend itself to persuasive argument. The fact that this book has been selected as the introductory textbook on linguistics at a number of highly ranked institutions gives me pause for thought, not because of the book’s conclusions but because of the subjective way in which they are presented. To see The Language Instinct as an objective introduction to modern linguistic theory (as the cover blurbs would suggest) is a mistake’though the material is there, it comes loaded with a political and cultural agenda not immediately apparent but which must be taken into account in order to make sense of the writer’s ultimate intentions.

From Newsletter No. 102 (July 2003)