Item

Abstract

Objectives: Languages change under conditions of contact, but the kinds of changes possible are quite varied. There is a wealth of documentation of contact induced borrowing, which can be interpreted to mean that contact causes a reduction in linguistic diversity. But there are also documented cases where language contact leads to an increase in diversity, that is, where contact causes an increase in the differences between languages. These contradictory outcomes have been explained in terms of two broad social functions of language: (1) the facilitation of interaction between groups, and (2) the demarcation of group boundaries. However, it remains unknown what ecological, social, and demographic environments are associated with these funcitons and their outcomes. In this paper we build a model of human intercultural accessibility, and use it to investigate what factors can predict linguistic diversification. Materials/Methods: From a sample of 400 Austronesian language/culture pairings (Greenhill, Blust and Gray 2008) we build a geographic model of ease of interaction. This is based on type of land- and sea-scape, and accomodates our knowledge of ethnographically-attested navigational technology. This measure of human accessibility is compared to linguistic patristic distance, i.e. the amount of evolutionary change distinguishing each pair of languages. Results: Under a neutral model of linguistic diversification, human accessibility predicts a considerable amount of the variation in linguistic distance. We add other social and demographic factors to the model, and attempt to account for the deviation from this neutral model of language evolution. Large scale tendencies in the direction and magnitude of language change can be predicted on the basis of these spatial-demographic models.