Department of Archaeology and Prehistory, University of Sheffield, U.K.
r.m.harrison@sheffield.ac.uk        p.nystrom@sheffield.ac.uk

abstract

Throughout primate evolution the brain has increased in both size and specialisation in order to accommodate an increasingly complex range of behaviours. Control of many of these behaviours has become represented asymmetrically in the cortex (Kaas 1987), for example, language capabilities and motor control of the dominant hand. It has been suggested that the evolution of language is related to increasingly lateralised use of the limbs due to the strong tendency for control of these behaviours to be located in the same hemisphere of the brain (e.g. Deacon 1992a; Gibson 1993). In this paper we examine the evidence which may support the perception that there is a relationship between the evolution of limb preference and language and whether it is really justified to draw a correlation between these two behaviours. We will also assess the potential benefits to the study of the origins of language if such a correlation exists.

Over time the language areas of the brain, notably Broca’s and Wernicke’s areas, have become increasingly lateralised, with the left hemisphere being dominant in the control of language functions in the majority of modern humans. Even though Broca’s area is now considered to consist of several overlapping regions involving different capabilities (Aitchison 1998), it is still believed to perform important language functions in modern humans (Deacon 1992b). The planum temporale is an important component of Wernicke’s language area for both the production and comprehension of spoken and gestural human languages. This structure, which is predominantly more expanded in the left hemisphere, also exhibits a correlation with handedness, notably a differential activation in listening tasks according to handedness (Tzourio et al. 1998). There is evidence for asymmetry of the planum temporale in fossil endocasts of Australopithecus, Homo habilis, H. erectus and H. sapiens neandertalensis (Holloway 1980; LeMay 1976; Tobias 1987). Apes, particularly chimpanzees, show a human-like asymmetry in the language areas, including the planum temporale, and in areas associated with cognition (Gannon and Kheck 1999; Gannon et al. 1998).

Thus, it appears that the neural framework for "chimpanzee language" (Gannon et al. 1998) may have been in place in the common ancestor of chimpanzees and humans, i.e. long before the appearance of fully developed human language. However, Corballis (1991) argues that there is a fundamental discontinuity between humans and other species, which relates primarily to a generative assembling device "responsible for constructing representations in generative fashion from small vocabularies of primitive units" (Corballis 1991: 219). In contrast, Aitchison (1998) states that language is composed of many constituents, each of which displays a different degree of continuity. The evolution of a complex adaptive trait, such as language, must be considered to have been a gradual process. Therefore, chimpanzee vocalisation may be considered as a precursor to the highly specialised form of vocalisation of modern humans, and could provide insight into the early development of human language. It is possible that the same speculation can be made of handedness, with manual dexterity becoming increasingly complex over time.

The lateralisation of motor control of the dominant hand has been the subject of much investigation, with a special focus on nonhuman primate limb preference. Results from meta-analyses suggest that preferences exist at an individual rather than species-level (e.g. McGrew and Marchant 1996, 1997), and that individual preferences get increasingly stronger with certain tasks in higher grades of primates. Even though right hand preference in modern human populations appears to be cross-cultural, there is evidence that the strength of this preference varies both according to the task being performed and at an individual level (Healey et al. 1986; Marchant et al. 1995; Peters and Murphy 1992). Unfortunately, the focus of many studies on human handedness inventories and the performance of fine motor tasks such as writing does not facilitate an easy comparison with nonhuman primates.

However limb preference is not unique to H. s. sapiens. Preferences are found in other primate species, with higher grades of primates showing increasing levels of individual preferences. Even though modern humans also show variability in limb preference, in addition to a low but constant level of variability in the neural control of dominant limb and language (Ramussen and Milner 1977), it can be questioned if there is an advantage for modern humans to have both handedness and language represented in the same hemisphere (Corballis 1991). Many explanations can be offered for this widespread, though not exclusive, association. Possibly handedness and language control are located in the same hemisphere by random chance and there is no connection between them. Alternatively there may be a highly developed functional correlation which necessitates them being located in the same hemisphere of the brain. Asymmetries also occur in the areas of the cortex which relate to motor control of speech, the periSylvian cortex (Galaburda 1984), with the Sylvian fissure being longer in the right hemisphere, particularly in right handed individuals (Bradshaw 1980; LeMay and Culebras 1972). Thus, the motor control of language production might be related to the need to have a dominant hand capable of performing highly specialised tasks. A possible impetus behind such a specialisation is the manufacture and use of tools which, at its extreme levels, requires fine motor control of both the dominant hand in the manufacture of tools and of vocal communication in the transmission of information regarding the processes of production.

 Conference site: http://www.infres.enst.fr/confs/evolang/