RUSSIAN NATIONAL CORPUS AS A SOURCE OF DATA IN THE WORK ON THE AUTHOR’S DICTIONARY

The paper shows the possibilities of using corpus data in compiling author’s dictionaries. The research is based on the materials of the multi-volume “Dictionary of the language of Russian Poetry (20th century)”, representing the language of ten outstanding poets: I. Annensky, A. Akhmatova, A. Blok, S. Yesenin, M. Kuzmin, O. Mandelstam, V. Mayakovsky, B. Pasternak, V. Khlebnikov, M. Tsvetaeva. Work on the dictionary as on a consolidated author’s reference book of a commenting type began in the mid-1990s. Since the opening of the Russian National Corpus, an important part of this work has been the appeal to corpus resources. The lexical array described in the dictionary is hetero geneous, which determines the different degree of demand for corpus data in relation to different groups of vocabulary. The article emphasizes the fact that the role of such data is especially significant when describing words requiring the increased attention of the dictionary compiler. These are units claiming the status of individual authorial neologism; rare words — obsolete, regional and others that require explanation; lexical borrowings, including barbarisms; proper names which often appear in the texts of the named poets in variants unusual for a modern native speaker; also words with an indistinct stylistic status. The study provides examples of dictionary entries for such units. The commenting part in these entries was prepared based on corpus data.