Retour vers liste

Détail de la contribution

Auteur: Yoolim KIM

Redefining the L1: Can Vowel Categories Change?

Abstract/Résumé: Previous research on second language acquisition explores the effects of the first language (L1) on second language (L2) acquisition (Flege, 1986; 1988). Focus has conventionally been on the influence of the L1 on the L2 and the resulting occurrence of a “foreign accent” (Flege, 1992). For example, native Koreans learning English often replace /ɔ/ with /o/ because /o/ is unavailable in Korean. Thus, words such as ‘ball’ and ‘hall’ are pronounced [bol] and [hol]. However, it has been noted that in some individuals the L1 is also produced with an accent. There is increasing evidence suggesting that a developing L2 can affect the L1. Chang (2010), who studies the malleability of L1 sounds, refers to this collective phenomenon as “phonetic drift”. My paper reports on a pilot study of vowel production in Korean-English bilinguals which supports the notion of shifting phonemes in both the L1 and the L2. 11 bilingual speakers of Korean and English recorded passages in both languages, producing 11 English vowels and 8-10 Korean vowels. The intersections of first and second formants were plotted for each vowel to produce formant charts for each individual’s vowels. The analysis compared participants’ English and Korean vowels to a standard set of Korean and English vowel measurements developed by Yang (1996). I also compared my bilingual speakers’ vowel formants against average vowel formants of monolingual Korean speakers. I found that all the bilinguals’ vowels, but especially their Korean vowels, were more back than average standards. The results show an overall trend where English vowels [ɪ, a, ə, æ, ʌ, ɛ] overlap and cluster. These findings illustrate that pronunciation errors can occur in the L1 because L2 vowels are interacting with L1 vowels in a way that is causing them to cluster and/or collapse into a narrower vowel range. There are consequential effects of multiple sound systems from different languages interacting with one another. Some individuals appeared to create a relatively new sound system consisting of sounds from both languages, while also including sounds unique to that individual. The data suggest that L2 learners borrow sounds from their existing sound systems, and in the process meld borrowed sounds to form an idiolectal vowel (or a "melded" vowel) inventory. The perceptual result of these melded vowels is accented speech production in both the L1 and the L2. There is also the possibility certain vowel production phenomena are simply inherent to Korean-English bilinguals, introducing a possible “bilingual effect.” This paper aims to challenge the notion that vowel categories, especially those belonging to the L1, are permanent. Perhaps all vowels, regardless of the state of bilingualism, are in constant flux, vulnerable to change as an effect of language interaction.