A Statistical Analysis of Can and May in British and Indian English

This paper compares the distributional properties of can and may in two varieties of Englishes (British English and Indian English) and provides a statistical analysis of the distributions. For this purpose, two parallel corpora were selected in the analysis: the British Component of International Corpus of English (ICE-GB) and the Indian Component (ICE-India). After the sentences with these two auxiliaries were extracted, seventeen factors were encoded into each of the sentences. Then, a logistic regression model was applied and it was statistically analyzed which factors played a role in deciding on the choice and how they affected the choice in two different varieties of English. Through the analysis, the following facts were observed: (i) Even though the English of the Outer Circle was heavily influenced by that of the Inner Circle (Kachru, 1992), the English of the Outer Circle is clearly different from that of the Inner Circle, and (ii) Bates and MacWhinney’s Competition Model (1982, 1989) is also applicable to modelling the similarities and differences of the Englishes in the Inner Circle and the Outer Circle.