Author: Jiajia Su

Publisher:

Published: 2019

Total Pages:

ISBN-13:

"This thesis investigates the L2 acquisition of plural marking in Chinese by English speakers and Korean speakers within the framework of the Feature Reassembly Hypothesis (FRH) (Lardiere 2009). According to the FRH, acquisition difficulty arises from differences in how features are assembled in lexical items and under what conditions features are realized in the L1 and L2; acquisition difficulty can be overcome when positive evidence is available in the input, but might persist when relevant evidence is obscure or unavailable.The Chinese plural suffix -men, the Korean plural suffix -tul, and the English plural suffix -s share some properties and differ on others: (i) plural marking is optional in Chinese and Korean, but obligatory in English; (ii) plural-marked nouns have a specific interpretation in Chinese and Korean, but not in English; and (iii) plural marking is restricted to human nouns in Chinese, but not in Korean or English. In addition, there are some co-occurrence differences: (i) when a noun co-occurs with a demonstrative, plural marking is obligatory on the noun in Korean, obligatory on the demonstrative in Chinese, and obligatory on both demonstrative and noun in English; and (ii) when a noun co-occurs with a classifier (Cl), plural marking is prohibited in Chinese, and prohibited in Korean with the exception of human classifiers and human nouns. Moreover, classifiers can be reduplicated in Chinese to express plurality: yi Cl Cl reduplication has an abundant reading, and Cl Cl reduplication has a distributive reading. These cross-linguistic differences in plurality are analyzed in terms of differences in the way features are assembled and differences in conditions on feature realization.An experiment was conducted to test L2 knowledge of Chinese plural marking. 15 advanced and 17 intermediate English-speaking learners of Chinese, 16 advanced and 19 intermediate Korean-speaking learners of Chinese, and 25 native Chinese speakers were tested using a Grammaticality Judgment Task and a Truth Value Judgment Task. The results show that: (i) all the L2 groups have acquired the [Number+plural, D+specific, Animacy+human] features associated with the Chinese plural suffix -men; (ii) the two English groups and the advanced Korean group have acquired the co-occurrence condition with demonstratives and the co-occurrence condition with classifiers; (iii) only the two advanced groups have acquired the [Number+plural] feature of yi Cl Cl reduplication, none of the L2 groups have acquired the [Q+abundant] feature of yi Cl Cl reduplication; and (iv) all the L2 groups have acquired the [Number+plural, Q+distributive] features of Cl Cl reduplication. The results are consistent with the FRH: (i) differences in how features are assembled in lexical items and differences in conditions on feature realization between the L1 and L2 can lead to acquisition difficulty; and (ii) acquisition difficulty resulting from L1 transfer can be overcome, but successful acquisition is not guaranteed. As for the process of feature reassembly, the study suggests that feature reassembly can be triggered by positive evidence, direct negative evidence such as grammar teaching, and generalized indirect negative evidence based on the statistical distribution of input"--