Item

Abstract

Traditionally, there are two motivations for an interest in reconstructing the history of language families. The obvious one is that through reconstructing the history of languages we can make inferences about the history of their speakers. This is particularly fruitful when carried out within an interdisciplinary framework, allowing triangulation with archaeological and genetic data, and providing methods of testing dates and migration paths. The second reason to be interested in the reconstruction of language families (and the one of particular interest to us as linguists) is that it enables us to reconstruct the processes of language change, with implications for linguistic typology, theories of grammar, psycholinguistics, and cognitive science. Rigorously formulated hypotheses about processes of language change provide us with testable predictions within an evolutionary framework. In this paper I will discuss methods for testing models of the evolution of language in order to (i) detect language contact between sister languages, (ii) investigate the mode of evolution of phonological systems (with focus on gradual versus punctual changes), and (iii) measure evolutionary dependencies between structural features of languages.