The talking ape: How language evolved (review)

Not a year goes by without at least one new book—often more—on the evolution of language. Whatever drives this process is certainly not a flood of new knowledge. Indeed, the number of theories on how language evolved stands in inverse proportion to the number of hard facts available. Consequently, any book on this topic must be evaluated mainly on the level of skill and sense with which it clothes this partial skeleton of data with indirect evidence from a variety of disciplines.

Robbins Burling, perhaps best known to linguists for his work on Tibeto-Burman languages, has published some papers on language evolution, but The talking ape is his first full-length treatment. The book seeks to cover the entire field, although there is relatively little on species ancestral to humans—species that must, on his account, have performed a large, if not the largest, part in the creation of language. His first chapter dismisses attempts to derive language directly from some prior system of animal communication. He allots greater significance to the lexicon than to syntax, opts for a gradual rather than a rapid development of language, and suggests that social rather than technological pressures were the forces that originally selected for it.

In his second and third chapters B develops his distinction between language and prelinguistic communication, particularly in light of modality distinctions between audible and visible modalities. He resists the temptation to see signing as a necessary bridge linking language with prelanguage, and stresses that modality is unimportant as compared with the capacity of language ‘to convey referential and propositional meaning’ (49).

B’s fourth chapter takes up the issue of the relationship between language and mind. He has moved from the position that language developed in the mind as a tool for thinking with—one Noam Chomsky, apparently, still holds—to one in which language required a sophisticated, nonlinguistic conceptual system before it could emerge. He seems to assume little substantive difference between human and nonhuman concepts, although at times his wording reflects some uncertainty on this point.

His relatively brief fifth and sixth chapters discuss the nature of signs, following Peirce 1940 in distinguishing iconic, indexical, and symbolic varieties, and review signs from a variety of species and modalities—songs of whales and birds, snarls and tail-wags of dogs, chimpanzee ‘nursing pokes’ (signs indicating an infant’s desire to feed), children’s ‘arms-up’ (‘Lift me!’) gestures, and those human signs that seem intermediate between animal systems and language, like thumbs-up or -down, exclamations (‘tsk-tsk’), and the like. While stopping short of claiming that these form an actual continuum, he presents them, especially in Table 2, as stepping stones on the way to language, even though, as he admits, the properties that distinguish them (innate versus learned, analog versus digital) are absolute rather than scalar.

In Ch. 7 B takes on the early development of language. He cautiously endorses the old notion (now back in fashion, Mithen 2005), that language and music share a common origin, with a subsequent split into ‘meaningful’ and ‘meaningless’ systems. He sees the earliest utterances as holistic, largely consisting of imperatives, perhaps linked to specific situations. Such utterances would differ from animal calls only in that they had to be innovated and learned. But an increase in their number would drive phonological differentiation: ‘Distinctions could be added until a whole system of contrastive phonology was available’ (139).

Chs. 8 and 9 deal with the evolution of syntax. Ch. 8 discusses the relative contributions of hard-wiring and experience to the acquisition of language in the context of whether syntax evolved suddenly or gradually. Following Pinker and Bloom (1990), B sees syntax as resulting from a gradual process of natural selection. In Ch. 9 he discusses intermediate stages. Single words form stage one, word-strings stage two; stage three is regular word order, which B sees as deriving from some kind of semantic/pragmatic iconicity. Stage four consists of regular phrase structure; language then requires only ‘morphological...