Diachronic Corpora and the Study of Language Change

Abstract

How can morphological, lexical, and syntactic change be found and analyzed in diachronic corpora? Historical linguistics is inherently corpus based, and many of the techniques that have been used in historical linguistics for centuries are precursors of corpus-linguistic techniques today. The use of electronic corpora makes many research questions easier to answer and results transparent and replicable. There are many decisions that have to be made when building a synchronic historical corpus and even more decisions and are necessary in building a diachronic corpus. Using two diachronic corpora[1], I will show that it is necessary to use a powerful corpus architecture which allows many independent annotation layers. I will demonstrate this by looking at morphological change (perfect auxiliaries), syntactic change (relative clauses) and lexical change. The focus of this talk is methodological.

[1] The corpora, DDB and Ridges, are in German but this is not essential for the arguments I will make; knowledge of German is not required. The corpora are freely available at http://korpling.german.hu-berlin.de/ddb-doku/index.htm, http://korpling.german.hu-berlin.de/ridges/index_en.html