This study examines Korean English learners’ production patterns of word-medial consonant-liquid clusters C1C2, which are not phonotactically allowed in Korean. We conducted a production experiment using English stimulus words that were systematically constructed mainly based on word frequency and cluster position in relation to the morpheme boundary. While this study corroborates several findings from previous studies, it also unveils new insights. First, the highest production accuracy is observed when C1 is a stop, followed by a nasal, and then a liquid, when no morpheme boundary intervenes the cluster. Second, for words with C1 as a stop, words with the cluster contained within a single morpheme exhibit higher accuracy compared to those with the cluster spanning across morphemes. Third, the predominant error type varies with the nature of C1; liquid assimilation errors are most frequent with C1 liquid, nasalization with C1 nasal, and vowel epenthesis with C1 stop. Fourth, word frequency or familiarity showed no substantial impact on either production accuracy or error types. The findings suggest that non-native cluster production is shaped by the interaction of phonetic, phonological, and morphological factors.
1. Introduction
2. Previous Findings and Hypotheses
3. Methodology
4. Results
5. Discussion
References
(0)
(0)