Syntax out of learning: The cultural evolution of structured communication in a population of induction algorithms

Abstract

A new approach to the origins of syntax in human language is presented. Using computational models of populations of learners, it is shown that compositional, recursive mappings are inevitable end-states of a cultural process of linguistic transmission. This is true even if the starting state is no language at all. It is argued that the way that knowledge of language is transmitted through a learning bottleneck profoundly influences its emergent structure. This approach provides a radical alternative to one in which the structure of language is viewed as an innate, biological adaptation to communicative pressures.