L

Language Evolution In 1866, the Société de Linguistique de Paris banned the presentation of any theory purporting to explain the evolution of the human language faculty. During the years following the publication of Charles Darwin’s Origin of Species there had been an explosion in the number of (increasingly) far-fetched accounts offered and the Société felt its reputation for scientific rigour was at stake. The area is indeed fraught with problems. A

major problem is that language faculties don’t fossilize. A palaeontologist will never find a section of an early hominid’s language module among the strata of a cliff, or a grammatical structure hidden in a piece of flint. There is no fossil record of any kind through which to sift, and as a consequence many theories are simply not testable. In addition to this problem, other traditional methods of evolutionary research, such as the analysis and comparison of analogous traits within related species, are unavailable. The combinatorial, recursive system that is language appears to be uniquely human. While there are many complex animal communication systems (see Hauser 1996), none have anything in common with the core features of human language. Even our closest relative, the chimpanzee, has neither the cognitive ability to acquire language nor the physical prerequisites for producing it if they could. Reflecting the concerns of the Société de Lin-

guistique de Paris, (retrospectively at least), Stephen Jay Gould (1989: 14) warned against Kiplingesque ‘just-so stories’ in accounts of language evolution: ‘the universals of language are

so different from anything else in nature … [that] origin as a side consequence of the brain’s enhanced capacity, rather than a simple advance in continuity from ancestral grunts … seems indicated’. For someone who has said so much about language, Noam Chomsky was notoriously reticent (at one time at least) on the question of its evolution. Chomsky (1988: 167) stated that ‘in the case of such systems as language or wings, it is not easy even to imagine a course of selection that might have given rise to them’. Indeed, he was largely thought to advocate a view that a language faculty could have emerged by way of a single genetic mutation (see PiatelliPalmarini 1989). Despite the numerous problems, over the past

twenty years or so there has been a massive resurgence of interest in the evolution of not only language, but also communication and cognition. Modern theories of language evolution are constrained by advances in evolutionary psychology, and are helped by the increasing number of researchers – from disciplines as diverse as neuroscience, behavioural genetics and archaeology – becoming involved in the field (see Christiansen and Kirby 2003 for a review). Various aspects of human language have been

proposed as the core around which the human language faculty evolved. For Terrence Deacon (1997), it is the fact that language uses ‘symbols’ – humans are, to use the title of Deacon’s book, the ‘symbolic species’. For Jean Aitchison (1996), it is ‘the naming insight’ deployed by children in the early stages of the acquisition of language. For Andrew Carstairs-McCarthy (1999), it is subject-predicate structure which, he argues, has its origins in syllable structure. Michael

Corballis (2002) proposes that language evolved from manual gestures. Steve Mithen (2005) suggests it was the propensity early hominids possessed for making music that led eventually to the evolution of language. In 2002, Chomsky himself joined the discussion (Hauser et al. 2002). This paper sought to sever the link between the evolution of language and the evolution of communication and argued that language – in the sense of a ‘narrow syntax’ (2002: 2) which generates linguistic representations and maps them on to the language of thought – could have evolved as a by-product of other human computational abilities, such as number or navigation. For those working in pragmatics, it is inter-

esting to note that all these proposals are based on a view of ancestral hominid linguistic communication that is a simple coding-decoding affair (see Dessalles 2007 for an exception). Such a view of communication is also adopted in one of the seminal papers on the topic, Pinker and Bloom (1990). However, it is problematic. If linguistic communication began as a pure coding-decoding process, we need to explain why it changed in character so drastically at a certain stage, and became inferential (see Sperber 2000). Most accounts ignore the fact that it is not just language that is unique, but also the way we use it. An act of linguistic communication is a meeting of minds. Human communicators are simultaneously engaged in a massive amount of multilayered intention attribution and expression. A further assumption in much work is that the

evolution of the human linguistic ability preceded the development of the kind of mental capacity necessary to handle these multilayered metarepresentations. This view is also problematic, since without some capacity for inferential communication it is hard to see how a language faculty could have been adaptive. Natural selection is a process by which advantageous heritable characteristics or traits increase in prevalence within successive generations of a particular organism, and disadvantageous ones decrease. But any organism, who acquired through mutation some novel linguistic codingdecoding ability, would have had no advantage within his or her community. These abilities would have been entirely useless, as this organism

would have had no one to talk to and no one to listen to. In the development of other cognitive mechanisms, such as face recognition and colour categorization, there is a clear domain of information in the environment that is present before the mechanism evolves. What in the environment could have resulted in mutations toward a language faculty being selected for? As Sperber (1990: 757) puts it:

Either we deny with Piatelli-Palmarini and others that the language faculty has been selected, or we deny that the domain of information relevant to the language faculty was empty before the emergence of the faculty itself.