The investigation of tag questions in a computerised corpus of child bilingual language.

Due to the relative structural complexity of tag questions in English, much of the research published on these discourse features involves studies carried out exclusively on speakers of English: Tottie and Hoffman (2006) compared tag questions used by British and American English speakers; Stenstrom et al (2002) and Torgersen and Costas (2009) reported on the use of invariant tags (such as ‘innit’, ‘okay’, ‘right’ and ‘you get me’) by adolescents in London; and Moore and Podesva (2009) investigated canonical tag questions used by high school girls in the Northwest of England. Going beyond purely monolingual English discourse, the present paper examines the speech of two bilingual children (Brazilian/British) and analyses their use of tag questions, whether in monolingual Portuguese, monolingual English or mixed (code-switched) utterances. Unlike Mills’ study of her bilingual German/English son (1981), which involved the manual analysis of data (notes and tape recordings), the methodology used in the current study draws on Corpus Linguistics and involves the automated retrieval and analysis of tag questions using the CLAN tools (MacWhinney, 2010b). Transcribed according to the CHAT system (MacWhinney, 2010a), the corpus contains over 150 recordings (approximately 30 hours) of naturalistic interactions taking place between the two main informants, a girl ‘M’ and her brother ‘J’ (aged 5 and 3 years at the beginning of data collection in 2001), and their monolingual and bilingual family relatives. Longitudinal in nature, the recordings were collected over four years, primarily in Brazil where the siblings were born and where they lived until moving to England in 2004. In addition to standard CHAT conventions, special codes were designed and inserted throughout the corpus to allow for the automatic analysis of various linguistic phenomena such as insertional code-switching, mixed forms, meta-linguistic comments, errors and tag questions, the particular focus of this presentation. Firstly, methodological aspects of the study including the coding system and the use of CLAN commands (FREQ, COOCCUR and KWAL) to carry out the analyses will be described in detail. Secondly the results will be discussed in the light of previous research on tag questions. MACWHINNEY, B. (2010a). The CHILDES Project, Tools for Analyzing Talk – Electronic Edition. Part 1: The CHAT Transcription Format. Carnegie Mellon University. Available online: http://childes.psy.cmu.edu/manuals/chat/pdf . MACWHINNEY, B. (2010b). The CHILDES Project, Tools for Analyzing Talk – Electronic Edition. Part 2: The CLAN Programs. Carnegie Mellon University. Available online: http://childes.psy.cmu.edu/manuals/clan/pdf . MILLS, A., E. (1981). It’s easier in German, isn’t it? The acquisition of tag questions in a bilingual child. Journal of Child Language. 8, 641-647. MOORE, E. & PODESVA, R. (2009). Style, indexicality and the social meaning of tag questions. Language in Society, 38(4), 447-485. STENSTROM, A-B., ANDERSEN, G. & HASUND, I. K. (2002) Trends in teenage talk: corpus compilation, analysis, and findings. Studies in corpus linguistics. 8. Amsterdam: J. Benjamins TORGERSEN, E. & COSTAS, G (2009) A corpus-based study of invariant tags in London English. In: Corpus Linguistics 2009, 2009-07-21 - 2009-07-23, University of Liverpool, UK. (Unpublished) TOTTIE, G. & HOFFMAN, S. (2006). Tag Questions in British and American English. Journal of English Linguistics. 34(4), 283-311.