From proto-mimesis to language: Evidence from primatology and social neuroscience

Abstract

How can we reconcile the conception of language as a conventional-normative semiotic system with a perception/action-based account of its structure and meaning? And why should linguistic meaning – as opposed to linguistic expression – be so closely related to motor activity and its neural underpinnings, as suggested by recent findings? A conceptual framework and evolutionary scenario building on the concept of bodily mimesis [Zlatev, J., 2005. What’s in a schema? Bodily mimesis and the grounding of language. In: Hampe, B. (Ed.), From Perception to Meaning: Image Schemas in Cognitive Linguistics. Mouton de Gruyter, Berlin, pp. 313–343] imply answers to these questions. The article presents evidence for a particular evolutionary stage model by reviewing recent evidence on the capacity of non-human primates for intersubjectivity, imitation and gestural communication, and from neuroscientific studies of these capacities in monkeys and human subjects. It is argued that “mirror neuron” systems can subserve basic motoric and social capacities, but they need to be considerably extended in order to provide an efficient basis for bodily mimesis, and even more so for language. It is argued that while language may be ultimately “grounded” in perception and action, it is essential not to try to reduce it to them.

Introduction

Language has been understood in widely different ways: as a “hidden code” (Sapir, 1928) a “mental organ” (Chomsky, 1975), a “computational device” (Jackendoff, 1983), an “instinct” (Pinker, 1994), etc. As all metaphors these notions may capture something of the phenomenon they apply to, but also necessarily hide others. What all of the theoretical accounts alluded to above hide is the essentially social nature of language, with normativity as a central characteristic (Itkonen, 1978, Itkonen, 2003). As famously pointed out by Wittgenstein, using language implies knowing certain criteria of correctness and such knowledge cannot be private, but must be shared within a community (Wittgenstein, 1953). This inherently social, intersubjective and normative nature of linguistic signs (both lexical and grammatical) is irreducible to individual minds/brains (Burge, 1979). Furthermore, language is used not only for “packaging” pre-existing ideas, but for formulating them (Vygotsky, 1962 [1934], Sinha, 1988, Tomasello, 1999, Hutto, 2007). Thus, language can be succinctly defined as a conventional-normative semiotic system for communication and thought (Zlatev, 2007a, Zlatev, 2007b). But language is also a bodily skill, and as much recent (and not so recent) evidence shows, not only its expressions (i.e. acts of producing speech, writing or signing), but much of its content appears to be closely based on motor activity and its supporting neural underpinnings (Glenberg and Kaschack, 2002, Arbib, 2005, Gibbs, 2005, Rohrer, 2005).

How can these two different, and apparently contradictory, perspectives on language – the social-semiotic and the “embodied” – be unified? And why does linguistic meaning (as opposed to expression) need to be related to motor activity? In this chapter, I endeavor to show that answers to both of these questions follow from the concept of bodily mimesis, and accounts of human cognitive evolution (Donald, 1991, Zlatev et al., 2005) and human ontogenetic development (Zlatev, 2003, Zlatev, 2005, Zlatev, 2008), according to which language is essentially post-mimetic. This means that language has been constructed as a super-stratum on top of capacities for intersubjectivity, imitation, re-enactive imagination and gestural communication, which constitute different aspects of bodily mimesis, and which are particularly well-developed in human beings.

I will first present the general conceptual framework and then show how a good deal of recent evidence from primatology can be interpreted in terms of it. Furthermore, I will review evidence from social neuroscience showing both similarities and differences in the primate and human neural systems for bodily mimesis. On the basis of this evidence, I will suggest a possible evolutionary scenario according to which language emerged from bodily mimesis. Finally, I will return to the two questions posed in the preceding section.