Text introducers of proverbs and other idioms

Abstract: This paper focuses on the textual behaviour of both non-sentential and sentential idioms in Czech and English. Twenty idioms of each type in both languages have been verified against the British National Corpus and the Czech National Corpus, each of the same size (100 million words). The idioms have been inspected as to possible coocurrence with metalanguage phrases or words, called ‘introducers’ in the paper, accompanying them in texts. The introducers identified in initial searches have then been further investigated in both corpora. A classification and statistics of occurence of these and typology of use and meaning have been suggested.