Auf der Suche nach historischen Phrasemen – oder: Wörterbücher als Korpora

Abstract

The treatment of phrasemes is problematic both in a lexicographic view and a corpus analytic view, even though the efforts of the collocation research provide enormous amounts of output. Finding and automatically extracting phrasemes, that are longer than two or three word collocations, is still not satisfactorily possible, because there are no markers in the text, that indicate the borders of the expressions. If in addition the perspective is an historical one, the problems arise as the amount of available historical texts decreases. For an historical phraseographic analysis realistic examples are extremely help- and needful, so that other sources for assembling examples have to be found. The following text proposes some approaches, how dictionaries can be used as corpora, and shows, which results (examples, illustrations, links etc.) can be expected. The point of origin of the idea is to use dictionaries not in the classical way – that means: searching the headword and reading the related article –, but to use the whole dictionary as a corpus. This inverse use of dictionaries allows us to find a lot of implicit lexikographic information, that in a normal way of using dictionaries could only be found by coincident.