Tid: Torsdag 14 februari 2013 kl. 15:00 – 17:00
Plats: C307, Södra huset, Frescati


Similarities and differences in a group of closely related languages are shaped by heterogenous factors, the most important being genealogical divergence and language contact both within the group and beyond it. Furthermore, languages do not converge or diverge in uniform ways; languages that might be similar in one respect may be different in others. The choice of what features we look at therefore determines the patterns of divergence and convergence that we see. It follows that we would like to look at a great number of features rather than only a few in order to attain a comprehensive picture of convergence and divergence and the factors involved. In my talk, I describe my work using a word-aligned, POS-tagged parallel corpus to investigate the distribution of a variety of phenomena across the Slavic languages, including verbal aspect, tense, mood, reflexive pronouns and lexical domains. In order to do this, I use both manually derived data sets as well as automatically generated ones, aiming at a framework that allows me to look at a large number of features at the same time.

Ljuba Veselinova