The Ministry of Education in Japan expressed a concern about local English teachers’ accent in the 1951 edition of the Suggested Course of Study in English, while presenting Received Pronunciation and General American as the model pronunciation for teachers and learners. Since then, Japanese-accented English (JAE) has been constantly criticised by both TESOL practitioners and ‘folk linguists’ in Japan as a gross and shameful deviation from standard speech, with native-speakerism being prevailed in and outside classrooms across the country. However, few empirical findings have reported on L1 English speakers’ perceptions of and ideologies towards JAE. By reviewing and re-analysing the data collected in New Zealand (NZ) at the beginning of the current century, this article sheds light on the identification and indexicality of phonological features of JAE among the Inner Circle speakers of English.
Despite their high motivation for pronunciation improvement, no late learners of English, regardless of their L1 backgrounds, were perceived as a native English speaker by the participants in the test environment. At the same time, JAE was identified at a higher rate than any other variety of L2-accented English, evidently due to the NZ listeners’ massive exposure to Japanese visitors. In the identification process, the conflation of /l/ and /r/ functioned not only as a marker but as a part of stereotype of JAE in a Labovian sense; once listeners had perceived it, they were more likely to assume the speaker to be from a Japanese-speaking background, although the same conflation was commonly observable in other East Asian accents as well. Arguably, this indexicality may have been specific to NZ people at that time.

抄録全体を表示

This essay introduces current approaches to English grammar and considers how the rapprochement (Smith 1996) between prescriptive grammar and descriptive grammar is possible. Traditional grammar has been prescriptive whereas modern linguistics has challenged descriptive, analytical measures. Modern linguistics, once aiming at analytical and theoretical definitions, is gradually changing as new disciplines such as sociolinguistics and pragmatics emerge. Examining language not only in its structure but also in its use in context illuminates what variants for different contexts exist and why they are in constant flux. A Comprehensive Grammar of the English Language by Quirk et al. is a pioneering work that describes the structure and use of the English language. The subsequent publication by Biber et al., The Longman Grammar of Spoken and Written English, reinforces its descriptive motivation by reflecting corpus information and referring to four different contexts: conversation, written fiction, newspaper, and academic prose. Unlike prescriptive grammar books filled with do's and don't's, these grammar books presume that different variants are used by different speakers in different contexts. As a sample grammatical category, this essay focuses on tag questions. English tag questions demand complex grammatical knowledge because they include the coreferential use of a personal pronoun, the proper formation of the operator reflecting the tense, person, number, and gender of the main clause, enclitic negative forms such as isn't and wouldn't, the use of do-support when necessary, and the polar contrast between the positive and negative sense. Hewings and Hewings (2005) propose that any grammar writing should come from three sources: actual usage, native speakers' intuition, and information from digital corpora. Based on this idea, this essay collects examples of tag questions from various grammar books, investigates native speakers' intuitive judgment through fieldwork, and examines samples from corpus data. From these observations, it is evident that writing grammar from multiple sources rather than from the dichotomy between correct and incorrect is crucial in understanding how grammar works and why variants develop.

抄録全体を表示