A Corpus Analysis of Russian Masculine Nouns with Plural /-i/ and /-a/ Endings

1.      Introduction

    This paper’s primary topic is the preference between modern Russian masculine nouns with plural nominative endings /-i/ and /-a/. According to previous studies, some Russian masculine nouns have morphological equivalents for the plural nominative form, such as bunkery – bunkera (бункеры – бункера). I will apply a statistical method to the data from the Russian National Corpus (RNC) and analyse the preference for the ending forms of the lexicon.

2. Corpus data

2.1. Targeting words of Corpus analysis

    According to Graudina et al. 2001, the following 35 words have equivalent endings /-i/ and /-a/:

al’t (альт), bunker (бункер), veer (веер), vektor (вектор), venzel’ (вензель), d’âkon (дьякон), korrektor (корректор), krendel’ (крендель), lekar’ (лекарь), mesâc (месяц), nevod (невод), otkup (откуп), otpusk (отпуск), pekar’ (пекарь), pisar’ (писарь), pojnter (пойнтер), poluton (полутон), prožektor (прожектор), pudel’ (пудель), reduktor (редуктор), sektor (сектор), stapel’ (стапель), tender (тендер), tenor (тенор), fel’d”eger’ (фельдъегерь), fligel’ (флигель), hlev (хлев), ceh (цех), škiper (шкипер), šnicel’ (шницель), šompol (шомпол), štepsel’ (штепсель), štorm (шторм), âstreb (ястреб), âčmen’ (ячмень)

2.2. Russian National Corpus

    When I used the query function for the RNC, the total number of tokens was about 230 million. To collect targeted data, I used a lexico-grammatical search, entering the targeted lexicon. Concerning grammatical features, I selected the nominative and accusative cases and plural number. To eliminate the generative case forms with /-a/ ending, I manually checked all the sentences. The results from the RNC are shown below in Table 1 and 2.

Table 1. The preference of each ending form in the RNC

Table 2. The preference of each ending form

3. Conclusion

    I establish four subcategories along with the preference for /-i/ and /-a/ ending forms of the word group. I applied the binomial test for the preference data of the lexicon. The significance level is 5%.

1)   both /-i/ and /-a/ ending forms are used but /-i/ ending forms are significantly preferred: bunker (бункер), d’âkon (дьякон), korrektor (корректор), lekar’ (лекарь), pekar’ (пекарь), pojnter (пойнтер), prožektor (прожектор), tender (тендер), hlev (хлев), ceh (цех), štorm (шторм)

2)   both /-i/ and /-a/ ending forms are used but /-a/ ending forms are significantly preferred: veer (веер), venzel’ (вензель), nevod (невод), otkup (откуп), otpusk (отпуск), pisar’ (писарь), poluton (полутон), stapel’ (стапель), tenor (тенор), fligel’ (флигель), šompol (шомпол), štepsel’ (штепсель)

3)   both /-i/ and /-a/ ending forms are used and significant preference between the two forms is not provided: krendel’ (крендель), sektor (сектор), škiper (шкипер), šnicel’ (шницель), âstreb (ястреб)

4)   /-a/ ending forms do not seem to be used and should be regarded as ungrammatical: al’t (альт), vektor (вектор), pudel’ (пудель), reduktor (редуктор), âčmen’ (ячмень)[1]

References

Graudina L. K. et al. (1976, 2001), Grammatičeskaâ pravil’nost’ russkoj reči. [The Grammatical Accuracy of Russian Speech]. Moscow.