ADULT ENGLISH SPEAKERS' ACQUISITION OF CHINESE COUNT-MASS CLASSIFIERS

This empirical study investigates English speakers' comprehension of count mass classifiers in L2 Chinese. The participants for the study were 62 English-speaking adults learning Chinese in Taiwan. The results of the study indicate the following: (1) In early stages of L2 acquisition, English speakers do not seem to honour the count mass distinction in Chinese, contrary to suggestions made by studies investigating English-speaking and Chinese-speaking children (Soja 1992, Soja et al. 1991, Chien et al. 2003). (2) The predominant use of the general classifiers ge suggests that it does not require the noun which it denotes to be of a particular type. (3) There are between-groups developmental differences probably due to lexical learning. (4) An asymmetrical development between the interpretation of count-mass classifiers and the CLP structures with de modifying count classifiers raises a question concerning the source of variability in L2 acquisition. Two theories, namely the Full Transfer/ Full Access hypothesis and the Partial Availability of UG hypothesis, in relation to parameter-resetting in SLA within the framework of Principles and Parameters are set out to explain the divergence. It coheres with what Hawkins and Hattori (2006) suggest that the apparent native-like L2 performance may not be equal to the underlying properties in the grammar of native speakers.