Vygotsky and cognitive science: Language and the unification of the social and computational mind By William Frawley (review)

REVIEWS161 Vygotsky and cognitive science: Language and the unification ofthe social and computational mind. By William Frawley. Cambridge, MA: Harvard University Press, 1997. Pp. xi, 333. Reviewed by James Stanlaw, Illinois State University In May 1997, IBM's supercomputer played world chess champion Gary Kasparov in a six game match. Kasparov is thought by many to be the best chess player who ever lived, even far exceeding, for example, the exploits of America's favorite son, the eccentric Bobby Fischer. This was only the second time a reigning world champion played a computer under official tournament conditions (a year previously Kasparov soundly defeated 'Deep Blue' 33 to 2j). Frawley's new book, while being neither a work on computer chess nor artificial intelligence, actually addresses some of the implications of this match (though perhaps unknowingly). F's basic question is simply this: how much of what we call 'thinking' or 'the mind' is due to computational processes, and how much is due to social context, cultural conditioning, or personal meta-knowledge? Many linguists and cognitive scientists subscribe to the first view. Ray Jackendoff , for instance, has said that he 'takes very seriously the motivating premise of contemporary cognitive science—the mind can be thought of as a biological information-processing device—' (1987: xi). The way IBM's computer plays chess in many ways mirrors this computational model. It makes its moves by brute force, calculating billions of positions in seconds and choosing the best move on the basis of these enumerations. Other linguists (Lakoff 1987), anthropologists (Keller & Keller 1996), and cognitive psychologists (Suchman 1987) do not see the mind as a manipulator of internal units, representations, or symbols. Kasparov—to maintain the chess analogy—brings to the board a century and a half of modem chess theory and knowledge of a vast literature from the international chess community. The best players are also said to 'play the man as much as the board', and Kasparov's fighting spirit and intimate knowledge of his opponents' weaknesses and personalities no doubt have contributed greatly to his success. And, as with most grandmasters, many of his moves are intuitive, based on considering the whole board as a holistic Gestalt rather than the result of actually seeing many positions ahead. F claims that the dichotomy between raw symbolic computation and social contextualization is simply wrong—both in the field of cognitive science itself and in real everyday existence. Kasparov, after all, does visualize and compute some specific tactics, and Deep Blue has the latest theories and opening repertoires programmed in by outside experts knowledgeable of both contemporary chess and computer science. The social mind and the computational mind must come together; that is, it must be that 'some parts of social language are computationally effective' (1). But what does this 'sociocomputationism' look like? To F, the answer lies in the 1930s Russian school of sociocultural theories of language, thinking, and personality most represented in the works of A. R. Luria (1976) and, more importantly, Lev Vygotsky (1962, 1978, 1986). The first part of F's book gives an excellent overview of some of the seminal issues involved in theoretical cognitive science and in epistemology as it relates to language. In a masterful summary of the philosophy of mind and language, F examines everything from Plato's problem (How can I know so much—like what a perfect triangle is—when I am given so little?) to Chomsky's Orwell's Problem' (Why do I know so little—like how the world political economy acts to suppress certain persons—when I am given so much?) to Wittgenstein's problem (How can I manage and reconcile the internal virtual representation of myself with the external real me?). Two primers on cognitive science and Vygotskyan psychology follow as well as a discussion of why Vygotsky needs to be taken seriously not only by psychologists but by linguists and others as well. As Vygotskyan theory has been receiving more attention in the past decade, and the implications ofhis work are being felt in a variety offields (cf. Wertsch 1985a, 1985b), this discussion of how computation must cooperate with culture—Vygotskyan style—is provocative. Simply put, F argues that...