Deutsche Varietäten in Internetkorpora – eine kleine Entwicklungsgeschichte

Abstract

The starting point for this article is the research on the dictionary Variantenwörter-buch des Deutschen, which was done between 1998 and 2004 at the Universities of Basel, Duisburg and Innsbruck. This dictionary contains all known national and regio-nal variants of the German standard language. During the research work it became clear that the identification of such variants needed an empirical validation. Since still no useful linguistic corpora of the German language existed at that time, experi-ments with Internet search engines were made, first with AltaVista, later also with Google. Thanks to the frequency numbers given in the search results, it was possible to get reliable hints for national variation. Today the situation has changed. Useful linguistic corpora are now available. In this paper, it is shown on the one hand that the results of the web search engines became less reliable over the time, on the other hand that still no well-structured corpus exists, which satisfies all needs of the re-search on national variants.