ABSTRACT
This study examines the interactions between three vocalic subsystems of multilingual speakers and explores the role language status might play in explaining variability across them. Eight 14-year-olds raised in Germany, who had learnt English at school for 6 years and Polish for 1 year, participated in the study. They were divided into three groups: children of German parents, children with one Polish-speaking parent but with German as the main home language, and children with two Polish-speaking parents and Polish as their main home language. The young multilinguals read a word list and performed a delayed repetition task in all their languages, with both tasks containing tokens of /i, ɪ, u, ʊ, ϵ, æ/. The results show a great degree of individual variability in the production of the vowels in all three languages and point to language status as one factor shaping the phonological subsystems of these multilinguals. The findings are interpreted within the framework of the Dynamic Systems Theory.
Acknowledgements
We are very grateful to the pupils, their Polish teacher and the Headmaster of the school in Berlin for allowing us to collect the data reported in this study. Also, we are grateful to the editors and the three anonymous reviewers for their helpful comments and suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. In contrast to the psycholinguistic perspective presented here, within socio-linguistic perspectives of L3 learning, language status is commonly defined as language prestige. As such it has been identified as an important factor affecting multilingual societies and their learners (see e.g. Cenoz, Citation2013 for a review).
2. We did not apply normalisation procedures for two reasons. First of all, we are working on young teenage populations and normalisation procedures are generally geared more towards adult speakers. The problem of speaker normalisation seems to be a longstanding issue and questions remain about optimal procedures for normalisation and the exact relationship between vocal tract growth and formant-frequencies (see Vorperian & Kent, Citation2007). Second, this was intended as an exploratory small-scale study with a limited number of participants, and we were mostly interested in comparisons of multilingual vowel spaces for particular individuals. We are planning to apply vowel normalisation in a follow-up larger scale longitudinal study to allow for more generalisable comparisons across participants.