The L2 Acquisition of Chinese Classifiers
- Jiahuan Zhang, University of Cambridge
- Teams Room 1, University of Aberdeen
This research explores crosslinguistic influence (CLI) on adult L2 learners’ acquisition of Chinese classifiers. Participants were 17 Arabic-, 15 English- and 18 Japanese-L1 students from Egypt, Australia and Japan respectively, who recently completed intermediate Chinese language courses at their home universities. Participants were divided into two groups, non-classifier language speakers (Egyptian and Australian) and classifier language speakers (Japanese). All were asked to take a Chinese language test that included three written tasks based on a picture description: composition, gap-fillings and multiple-choice questions. A post-test questionnaire was conducted to gather demographic information of the participants and to explore individual differences on classifier acquisition such as learner beliefs and learning strategies. Statistical analysis was carried out by means of R Studio. Results suggest that: 1) Although the statistical analysis suggests that L1 is not a significant predictor of test performance, the Japanese group numerically scored higher than the Arabic and the English groups on the test; 2) the task types are a significant predictor of test performance, with a significant interaction found to occur between L1 and tasks; 3) individual differences are not a significant predictor of test performance; 4) there tends to be a sequential mastery of different types of classifiers. This study takes the position that albeit without statistical significance, similarities between L1 and L2 have the potential to promote L2 acquisition, whereas dissimilarities could hinder it. On top of the L1 effects, different types of tasks interact with the L1, which may shed new light on classifier proficiency from the perspective of language testing. Additionally, individual differences seem not to affect classifier acquisition as a whole (at least in this study). More importantly, the sequential mastery of different classifier types implies a universal sequence of acquiring classifiers, which presents a robust account for the weakness of L1 influence. This study has the potential to uncover the source of the difficulties in classifier learning, which could be a useful reference for the selection of teaching methods and textbook editing.