Poster presented at the 4th International Conference on the Evolution of Language. Cambridge, MA.

Evidence for Semantic Complexity in Large-Brained Mammals



Christopher M. Conway


Abstract

Many theories of language evolution implicitly hold that an increase in semantic complexity must have been a key driving force in the emergence of syntax. After all, you don't need complex syntactic machinery if you only have a few concepts to express. In other words, complex syntax would seem to presuppose a high degree of semantic or conceptual complexity. Semantic complexity (SC) may result from high levels of cortical specialization in the brain (Schoenemann, 1999). Because cortical specialization is a natural by-product of increased brain size, a reasonable assumption is that increases in brain size led to increasing SC in the hominid lineage. One natural implication of this line of thinking is that non-human animals with large brains ought to display a high degree of SC. To explore this hypothesis, this paper investigates recent studies of large-brained mammals (dolphins, elephants, and whales) and finds that certain observable behaviors (e.g., social and communicative interactions) are consistent with the notion that advanced conceptual systems are present. For example, dolphins display impressive comprehension of human-taught, artificial languages (Herman, et al., 1984; Herman, et al., 1993), arguably superior to that of language-trained apes. Additionally, dolphins can understand referential gestures (i.e., pointing; Herman, et al., 1999), display conceptual knowledge of their own body parts (Herman et al., 2001), and in the wild, appear to use "signature whistles" to address one another (Janik, 2000). Elephants and whales also reveal behaviors that could be evidence of SC. Both have complex societies characterized by social facilitation (Christal & Whitehead, 2001; Schulte, 2000); elephants have sophisticated means of multi-modal communication (Langbauer, Jr., 2000; McComb, et al., 2000); and whale song, which has recently been suggested to display complex hierarchical structure (Suzuki, Buck, & Tyak, 1999), can change via a form of "cultural transmission" in a manner that is unprecedented in the animal world (Noad et al., 2000).

The evidence reviewed here, although not unambiguous, indirectly supports the notion that large brained animals have an increased conceptual representation of the world. The paper will conclude by addressing possible objections to this proposal. For instance, why don't animals with brains larger than humans have a correspondingly more advanced semantic repertoire? I will argue that language, once developed, further acts to increase SC. Thus, a kind of feedback loop exists, in which a sufficient amount of SC enables syntax and language to emerge, which in turn, injects additional conceptual richness into an organism's mental life. Another objection is that if these animals have a sufficient level of SC, why do they not display advanced language? I will suggest that SC alone does not guarantee language. Language presumably requires certain selection pressures beyond increasing brain size, such as the means and motivation to convey complex concepts to conspecifics. For instance, for language to emerge, organisms must not only have SC, but must be able to encode and process complex sequential information (Conway & Christiansen, in press); additionally, pressures for communicative interactions, perhaps motivated by threat of predators, may also be necessary.


rule.gif (155 bytes)

Home | People | Research | Links | Contact | Publications | Presentations
Cognitive Neuroscience Laboratory

Please email suggestions/errors to mhc27@cornell.edu