The Development Of Chinese Lexical Representation And The Mechanism Of Word-Generating Errors By CSL Learners Of English Background

Any theory of second language acquisition is incomplete without a representation component, because, as pointed out by Levelt (1989), representation and acquisition cannot be studied independently of each other. However, compared to acquisition, representation has received little attention from second language researchers (Jiang 2000). This is especially true with Chinese vocabulary acquisition as a second language (L2). As the inner mechanism of acquisition, representation unavoidably affects the lexical processing. From the perspective of Chinese vocabulary acquisition, word-generating errors are among the most commonly-seen lexical processing errors. They are interlanguage forms created by the learners with morphemes and regulations available in their brains, reflecting the developing status of their mental lexical representation. Combining the study of Chinese lexical representation and that of vocabulary acquisition would promote our in-depth understanding of the process and mechanism of Chinese vocabulary acquisition, and thereby help to improve the Chinese L2 vocabulary teaching techniques.The current study applies empirical approaches to test the applicability of the L2 lexical representation development model proposed by Jiang (2000) in Chinese vocabulary learning, and then explores the development of the main types of lexical knowledge based on word-generating error analysis. The mechanism of the errors is therefore discussed.Study I and study II explore respectively the mediation of L1 and L1 lemma in Chinese vocabulary cognition. Based on Jiang's model, four stages of Chinese mental lexical development are proposed: formal association stage, formal/lemma association stage, L1 lemma mediation stage and reconstructing stage.Word-generating errors are results of the inadequate representation of lexical knowledge which can be categorized into three types based on the error analysis: word meaning, word boundary and morphology. Study III, study IV and study V are thereby designed to probe into the development of these three types of lexical knowledge representation, and to explain the mechanism of the word-generating errors from a cognitive perspective. Error type I, the synonym/related words substitution error, appears when two or more Chinese synonyms or related words share the same English translation because the cognizing of Chinese words need to activate the links between Chinese words and their L1 translations. Error type II, the words and morphemes substitution error, is due to the little or inadequate word boundary information presented in the lexical entries. Error type III, the typical self-generated word error, is caused when the learners create new words with morphology and morpheme information they have achieved.Based on the above empirical research results, a specific representation model for learners' self-generated words is set up, and the developing mode of this model is further discussed from a theoretical perspective. The implications for Chinese L2 vocabulary teaching are also discussed.