The effect of experience on the acquisition of a non-native vowel contrast

This study examines the effect of second language experience on the acquisition of the English vowel contrast /ε/–/ææ/ by native speakers of Dutch. It reports on the results of production and perception tasks performed by three groups of native Dutch learners of English in Belgium, differing in experience with English, as measured through study choice and progress. Whereas experience has often been defined in terms of ‘Age of Acquisition’ or ‘Length of Residence’ in studies on L2 immersion in, for instance, immigrant settings, in traditional foreign language contexts experience needs to be defined in terms of amount of instruction and (non-)naturalistic exposure. The results revealed an asymmetry in production and perception: all learners seemed to have created a new phonetic category for English /ææ/, but failed to phonetically implement it in a native-like way. No new category for the vowel /ε/ was created, but learners who had opted for English studies at tertiary level produced a clear contrast between the two English vowels. All learner groups performed well on a discrimination and identification task involving the vowels /ε/ and /ææ/. The results are discussed in light of currently used speech learning theories, such as Flege, 1987 and Flege, 1995, Best’s Perceptual Assimilation Model for naı¨ve listeners (1995; Best et al., 2001) and its adaption for learners in an Second Language Acquisition context (Best and Tyler, 2007). It is argued that predictions formulated within these models also hold for learners in a Foreign Language Acquisition context.

► In FLA, experience needs to be defined in terms of instruction and/or exposure. ► L2 education does not always lead to an improved production of L2 vowel contrasts. ► Predictions of current SLA speech learning models also hold for FLA.