Acoustic Properties of Korean Stops as L1 Produced by L2 Learners of the English Language

Objectives: The present study examines whether Korean children and adults who have learned English as a second language (L2) produce Korean stops differently from Korean monolingual (KM) children with respect to voice onset time (VOT) and fundamental frequency (F0) and whether the age of L2 exposure is a critical factor in determining the L2-to-L1 influence. Methods: Eighteen Korean-English bilingual (KEB) children and adults and eighteen KM children produced nine monosyllables / p(h)epsilon, p epsilon, p*epsilon, t(h epsilon), te, t*epsilon, k(h)epsilon, k epsilon, k*epsilon/twice in a row. A total of 1,890 tokens were acoustically analyzed with Praat to obtain VOT and F0 values and were subjected to statistical analysis for group comparison. Results: Two three-way ANOVAs and follow-up statistics revealed that the VOT values of aspirated and lenis stops were significantly longer in KEB children and adults than in KM children and that normalized F0 values of lenis stops were significantly higher in bilingual than in monolingual individuals. Nonetheless, no significant group difference was found between KEB children and adults. Conclusion: The acquisition of L2 English appears to disturb a tertiary contrast of Korean stops in terms of VOT and F0, in a way that bilingual individuals produced lenis stops more aspirated-like, with increased VOT and F0 values, and aspirated stops with increased VOT values. The L2-to-L1 influence may not be significantly different between early and late L2 learners, if their length of residence in a L2 speaking community is similar.