A SIMPLE INTEGRATED FRAMEWORK FOR INVESTIGATING GENETIC AND CULTURAL EVOLUTION OF LANGUAGE

Abstract

Two areas concerning language evolution have received increasing interest in recent years: the genetic evolution of the language faculty and the cultural evolution of language. However, these study areas are somewhat separate, and there is a need to integrate these efforts and explore the relevant gene-culture coevolutionary interactions (Mesoudi et al., 2011). The relationship between genes and language is extremely complex; therefore, there is much controversial discussion on various aspects, including the role of learning (Suzuki & Arita, 2008). Fundamentally, several researchers argue that cultural evolution has little influence on the genetic evolution of language although, various results have shown that genetic biases are essential for a proper understanding of language evolution (Fitch, 2011). For example, Chater et al. (2009) have shown, by using a computational model, that there are strong restrictions on the conditions under which the Baldwin effect can embed arbitrary linguistic constraints and the effect only emerges when language provides a stable target for natural selection. This paper proposes an integrated computational framework for investigating possible scenarios of genetic and cultural evolution of language. Specifically, we would like to demonstrate cultural evolution to investigate phylogenetic dynamics of language and, at the same time, genetic evolution of phenotypic plasticity to investigate the Baldwin effect in language evolution, while keeping the framework as simple as possible. The main idea is to express the linguistic space in the polar coordinate system and graduallymove the individuals and languages using genetic and cultural evolutionary dynamics in the space. In the model, the linguistic space is expressed in the polar coordinate system as shown in Fig. 1. Each language is represented as a point, and its radius and angle correspond to its expressiveness that makes a benefit of successful information sharing, and numerical defined structural feature, respectively. Each agent is also represented as a point and a fan-shaped field surrounding the point. This field corresponds to the set of languages that the focal agent can acquire for information sharing with other agents. The location of the point, the agent’s innate language, and the size