ACCENTOLOGICAL CORPUS AS A TOOL FOR STUDYING RUSSIAN WORD STRESS

The article addresses several issues related to the development and use of the Accentological corpus as a tool for studying word stress. The current state of the corpus, its composition and structure, development prospects, replenishment with new material are described in this article. Particular attention is paid to the subcorpus of “naive” poetry within the Accentological corpus as a source of accentological data. The capabilities of this resource, its effective use are demonstrated in the study of variants of nominal inflection.

A corpus-based study of inflectional stress variation of singular and plural forms of short adjectives and case forms of nouns has been conducted. While studying case forms of nouns several areas of active competition of accent variants have been examined: feminine nouns with –a ending (doska ‘desk’, stena ‘wall’), feminine nouns with a soft stem (kist’ ‘brush’, chel’ust’ ‘jaw’) and masculine nouns with a hard stem (veter ‘wind’,sharf ‘scarf ’).

The results of the corpus based study have shown that the material of accentological corpora can be used as a reliable source of accentological data. Increasing the volume of the corpus makes these data statistically reliable, as well as expanding the range of the studied forms and contributes to the detection of new points of variability.