CORPUS AS A TOOL FOR STUDYING THE RUSSIAN LANGUAGE USAGE IN REGIONAL PRESS

The paper describes the results of corpus-based analyses of lexical-semantic and stylistic differences of functioning new words in Russian- language Belarusian newspapers in comparison with the Russian regional press. A sampling of adjectives not included in grammar dictionary of morphological analyzer and fixed in texts not earlier than the 1990-ies was retrieved with the help of corpus tools. Semantic, stylistic and frequency characteristics of these adjectives were compared in different sub-corpora of the RNC: Grodno region newspapers, Russian regional newspapers, main corpus and large newspaper corpus. A significant amount of new words are encountered both in Belarusian newspapers and in Russian regional press, but in Russian newspapers some of these words have a wider range of collocations indicating their more complex semantic structure. Lexemes that met only in Grodno subcorpus or only in the Russian regional subcorpus can be subdivided into several groups: (1) region-specific vocabulary relating to local realities, these words are represented in Grodno subcorpus more widely, (2) special terminology, and (3) conversational vocabulary, the latter two groups are represented to a greater extent in Russian media. In addition, there were distinguished  the following groups: (4) low-frequency new words that are not fixed in the RNC and in dictionaries, (5) authorial neologisms, (6) speech errors associated with impaired accuracy of usage and lexical compatibility. The corpus-based analyses revealed some regional differences in functioning of specific lexical items or specific groups, and similarity in using models for word formation and semantic derivation.