Korean Learners’ Production of English Consonant Clusters and the Role of L2 Proficiency

Abstract

This study investigated the production of word-medial English biconsonantal clusters with an onset liquid by the high and low proficiency Korean EFL learners. It was found that Korean learners’ production was influenced by the factors such as L2 proficiency level, the place/manner of articulation of the first member of a cluster (C1 place/manner), the liquid type involved, and the word frequency. First, while the high group did not show any significantly different production accuracy according to C1 manner, the low group exhibited significantly lower production accuracy when C1 manner was a stop than when it was a nasal. Since most errors occurred with respect to the nasalization or lateralization of a stop consonant before a liquid, this result indicates that the low group’s low accuracy was due to the influence of L1 phonological rules realizing a stop before a liquid as a nasal or a lateral. Second, both the low and high proficiency groups showed lower production accuracy for the homorganic clusters such as /tl/, /tɹ/, /nl/, and /nɹ/ although the low group’s accuracy for /nɹ/ was inexplicably as high as the high group’s. This can be accounted for in terms of the phonetic similarity of the members of a cluster. Finally, it was found that the low proficiency group was influenced by the loanword pronunciation for the words such as outlet and online and showed low production accuracy differently from the high proficiency group. (Hannam University, University of Seoul)