This longitudinal study explores regressive phonological cross-lingustic influence in 14 multilingual speakers (7adolescents and 7 adults) by examining VOT in their L1, L2, and L3 throughout the first year of L3 learning. For the adolescent group, significant changes to both their L1 and L2 over time were found, while the adults' background languages remained relatively stable on the group level. However, for both groups, much individual variation was uncovered.
The present study investigates regressive phonological cross-linguistic influence (CLI) in 9 adolescents (aged 12-13) and 7 adults (aged 21-39) with the same language combination (L1 German, L2 English, L3 Polish), who were recorded completing a range of production tasks in all three languages for four times over the course of the first year of L3 learning. It thus broadens the scope of previous research on phonological CLI in two ways: 1) by tracing the development of all languages upon the arrival of a new language in a multilingual's system longitudinally, and 2) by investigating CLI patterns in two age groups when input and learning environment are comparable. Previous L2 age studies have mostly only made retrospect assumptions about (target) language development, so that longitudinal data including the entire language repertoire of multilingual speakers are needed to substantiate claims made in that regard. It is hypothesized that, 1) for both groups, the L2 will be more vulnerable to L3 influence than the L1 due to differences in stability as suggested by the Phonological Permeability Hypothesis (Cabrelli 2016), and 2) that if changes of the L1 do occur (so called phonetic drift), they will be more visible for the adolescents than the adults. The acoustically analyzed features include rhotics, /v-w/ contrast, and VOT. First analyses of the production of /v/ and /w/ (centre of gravity) through linear mixed modelling show significant changes over time for L1 and L2 in both groups, which can tentatively be interpreted as regressive CLI. Further analyses will be conducted to show if this also holds true for the other features under scrutiny, as well as to investigate potential group differences in developmental patterns of CLI.
Cabrelli, J. (2016). Testing the Phonological Permeability Hypothesis: L3 phonological effects on L1 versus L2 systems. International Journal of Bilingualism: 1-20.
