Child L2 phonology acquisition under the influence of multiple varieties

Abstract

Input variability is vividly present even in L1 acquisition contexts (Foulkes and Docherty 2006), let alone in an FL/ L2 context where learners are exposed to input in one form from fellow students, to a different variety from the local teacher, and possibly another variety from the institutional model which typically represents the “native-standard norm” (Cook 2008; Regan 2013). However, little is currently known about (second) language acquisition in relation to input multiplicity (cf. Siegel 2010). In fact, it is unclear how L2 acquisition models such as the Speech Learning Model (Flege 1995) or Optimality Theory (Prince and Smolensky 1993) cope with input comprising multiple varieties. Against this backdrop, this study set out to investigate the nature of child L2 phonology acquisition under the influence of multiple varieties and its interface with sociolinguistic factors in Hong Kong (HK). The study looks at L2 English phonology acquisition by Hong Kong Cantonese children when various varieties are present. Specifically, it targets youngsters exposed to Filipino-accented English from live-in housekeepers in addition to the school and community input encompassing UK, US, and HK varieties. Results show that the 31 kindergarteners in their third year of studies aged 4;6 to 6, and the 29 first year secondary school students aged 11 to 14 who had received/were still receiving Filipino-accented English significantly outperformed 34 age-matched controls who were not exposed to such input on a picture-choosing task and a sound discrimination AX3 task targeting Filipino English plosives /p,t,k/ and fricatives /f,v/ (plosive onsets are often unaspirated while /f,v/ are sometimes rendered as [p,b] respectively in this variety (Tayao 2008)). These findings confirm predictions made by L2 speech acquisition theories in that the acquisition of L2 phonology is possible given a sufficient amount of exposure to the target language input. iii However, participants did not produce this variety in the production part of the experiment (a picture naming and a pair matching task) despite showing signs of perceptual knowledge. In addition, a separate instrument (verbal-guise technique) tapping into informants’ attitude towards Filipino-accented English reveals ambivalent attitudes towards this variety, making it challenging for one to resort to speech accommodation (Beebe and Giles 1984) or speech design models (Bell 1984; 2001) for an adequate explanation. This study highlights the complexity involved when multiple varieties are present in the acquisition context, which is arguably the norm rather than the exception in this current age of unprecedented geographic, social, and occupational mobility (Chambers 2002). It also reminds us of the importance of scrutinising from several perspectives the nature of input in L2 phonology (Moyer 2011; Piske and Young-Scholten 2009). Without a clear understanding of the diversity present in the input, it is difficult to make any solid claims about learners’ phonological competence in a given target language. In addition, the seemingly conflicting results on the perceptual and production parts of the study underline how essential it is to analyse the acquisition outcome from several perspectives through task triangulation.