Classification of the Indo-European Languages Using a Phylogenetic Network Approach

Abstract

Discovering the origin of the Indo-European (IE) language family is one of the most intensively studied problems in historical linguistics. Gray and Atkinson (2003) inferred a phylogenetic tree (i.e., additive tree or X-tree, Barthelemy and Guenoche 1991) of the IE family, using bayesian inference and rate-smoothing algorithms, based on the 87 Indo-European language data set collected by Dyen et al. (1997). When conducting their classification study, Gray and Atkinson assumed that the evolution of languages was strictly divergent and the frequency of word borrowing (i.e., horizontal transmission of individual words) was very low. As consequence, their results suggested a predominantly tree-like pattern of the IE language evolution. In our opinion, only a network model can adequately represent the evolution of the IE languages. We propose to apply a method of horizontal gene transfer (HGT) detection (Makarenkov et al. 2006) to reconstruct a phylogenetic network depicting the evolution of the IE language family.