The goal of this study is to provide a better understanding of the influence from first language (L1) phonology and morphosyntax on second language (L2) production and perception of English regular past tense morphology. The following research questions guided the present study: 1) Do L1 phonology and morphosyntax affect L2 learners’ production of the English past tense morpheme -ed? 2) Do L1 phonology and morphosyntax affect L2 learners’ perception of the English past tense morpheme -ed? 3) Does the phonetic form of the regular verbs affect L2 learners’ production of the English past tense morpheme -ed? 4) Does the phonetic form of the regular verbs affect L2 learners’ perception of the English past tense morpheme -ed? 5) Does L2 learners’ perception of the -ed morpheme correlate with how they produce it? To answer these questions, this study compared speakers of Turkish (a language that encodes tense morphologically and permits final consonant clusters) with speakers of Korean (a language that encodes tense morphologically but does not allow final consonant clusters) and speakers of Chinese (a language that does not mark tense and does not license final consonant clusters) on story completion, sentence repetition, self-paced listening (SPL), and perception judgment tasks.Sixty-two L2 learners of English (18 L1 Turkish, 21 L1 Korean, 23 L1 Chinese) and 24 native English speakers participated in this study. The results from the story completion task showed that both the Turkish and Korean groups were significantly more accurate than the Chinese group, although none of the three L2 groups produced the past tense morpheme -ed at native-like levels. In the sentence repetition task, the three L2 groups performed alike, and they all showed high levels of morphology suppliance (all over 90%), but only the Korean group performed in a way similar to the native control group. Moreover, in both of the oral production tasks, the learners in the three L2 groups alike were more accurate with regular verbs ending in a single consonant than regular verbs ending in consonant clusters. These results suggest that L1 morphosyntax might be an important factor in the production of English regular past tense morphology and that there seems to be a general phonological effect on final constraint clusters. The present data also indicate that the phonetic form of the regular verbs plays a role in the production of English past tense morphology.Two tasks examined perception of past tense marking. The results from the perception judgment task revealed that none of the three L2 groups perceived the -ed morpheme in a native-like manner. In addition, while the Turkish group was equally accurate in perceiving the -ed inflection in cluster and non-cluster contexts, both the Korean and Chinese groups were less accurate with regular past forms taking non-syllabic allomorphs [t] and [d]. The results from the SPL task were somewhat inconclusive, but demonstrated that participants were most sensitive to grammatical errors targeting syllabic allomorph [əd]. These results suggest that L2 learners’ perception of English past tense morphology is affected by a combination of L1 phonological constraints as well as more general properties related to the phonetic form of the regular verbs.Finally, Spearman’s correlation tests showed no statistical correlations between the perception and production of the -ed morpheme for any of the learner groups. While these results could be explained by differences in test materials and scoring procedures, they may also reflect a disconnect between perception and production of past tense in individual learners.
Read
