Accent Difference Makes No Difference to Phoneme Acquisition

Abstract

ELT materials tend to use prestige variety speakers as models, an underlying assumption being that this is needed in order to acquire the phonology necessary to parse English speech (Rose & Galloway, 2019). Global Englishes Language Teaching (GELT) (Galloway & Rose, 2018) provides the potential for movement away from such ‘native speaker’ ideologies, but lacks empirical evidence. In this study, the use of GELT input in comparison with prestige varieties of English was investigated. Sixteen first-year L1 Japanese university students in an English Medium Instruction programme participated in a self-paced listening study via a learning management system (LMS). All participants were tested on their perception of the English vowels /æ/, /ʌ/, /ɜː/ and /ɔː/. After this pretest, they were separated into two groups: using edited TED talks, the experimental group (G) (N=8) watched videos of Global English varieties, and the control group (P) (N=8) watched videos of prestige English varieties. Both groups acquired losses, i.e., immediate posttest scores were mainly lower than pretest scores on vowel identification. Scores were predicted by the variation in interval between lessons and posttest, but not by the varieties of English used. This provides support for the view that GELT is as valid a language teaching approach as using prestige varieties.