Distribution of mid vowels in Norwegian learners of French: The impact of transfer

Abstract

In French, the output quality of the mid vowels /E, Ø, O/ is to a large extent governed by the “Loi de Position”, whereby the close-mid variant [e, ø, o] is selected in open syllables, and the open-mid variant [ɛ, œ, ɔ] is selected in closed syllables, e.g. pot [po] ‘jar’ vs. port [pɔːʁ] ‘harbour’ (for more details, cf. Féry, 2003). Vowel lengthening is phonetic and triggered by a limited group of consonants, i.e. [v, z, ʒ, ʁ]. In Norwegian, on the other hand, the output quality depends on phonological duration: long vowels are realised [e, ø, o] while short vowels are realised [ɛ, œ, ɔ], e.g. få [foː] ‘get-inf’ vs. fått [fɔt] ‘get-perf.part’ (for more details, cf. Kristoffersen, 2000). The two languages also differ when it comes to positioning in the vowel space: while the series of unrounded vowels [e, ɛ] is well dispersed in both languages, the two series of rounded vowels – [o, ɔ] and [ø, œ] – are far less spread in Norwegian than in French (Kloster-Jensen, 1955; Kristoffersen, 2000). Norwegian learners of French are thus faced with challenges on two different levels of analysis, phonological and phonetic, a situation which is further complicated by current pedagogical methods that fail to train the perception and production of fine-grained vocalic variation. In view of this, one should expect learners of French with little exposure to authentic speech, to use their L1 system when practicing the foreign language.

This paper presents learner data from Tromsø and Oslo, which represent students at an A2 vs. B1/B2 level, respectively. Using the interview guide established by the research programme IPFC: Interphonologie du français contemporain (Detey & Kawaguchi, 2008), the dataset contains both read and spontaneous speech. The main objective of this work – which draws on Andreassen and Lyche (2014) – is to identify the level and types of transfer involved in the acquisition of mid vowels. To achieve this goal, we first test the hypothesis whereby the beginner-level students in Tromsø maintain the Norwegian length contrast and thereby close mid vowels in syllables with target lengthening consonants. Second, we test the hypothesis whereby the same students maintain the Norwegian acoustic system and thereby limit the distance between the variants of rounded vowels. On the basis of these results, and by comparing with data from the advanced students in Oslo, we aim at sketching the first version of the acquisition path of a phenomenon that is widespread, yet considered too marginal to be explicitly taught in the Norwegian classroom.

References Andreassen, H. N., & Lyche, C. (2014). Transfert combiné dans l'acquisition des voyelles moyennes par des apprenants norvégophones. Paper presented at the Journées FLOraL: atelier interphonologie et corpus oraux, Paris, 4-9 December. Detey, S., & Kawaguchi, Y. (2008). Interphonologie du Français Contemporain (IPFC): récolte automatisée des données et apprenants japonais. Paper presented at the Journées PFC: Phonologie du français contemporain: variation, interfaces, cognition, Paris. Féry, C. (2003). Markedness, faithfulness, vowel quality and syllable structure in French. French Language Studies, 13(2), 247-280. Kloster-Jensen, M. (1955). Précis de prononciation française. Oslo: Aschehoug. Kristoffersen, G. (2000). The phonology of Norwegian. Oxford: Oxford University Press.