The Effect Of L1-L2 Syntactic Similarity On L2 Syntactic Processing

Theoretical accounts of second language (L2) acquisition vary in their predictions regarding whether or not late L2 learners can exhibit native-like online L2 syntactic processing. The Unified Competition Model(UCM; MacWhinney,1997, 2005), in contrast with Shallow Structure Hypothesis(SSH; Clahsen & Felser,2006a, 2006b), proposes that nonnative speakers can show native-like processing depending on factors such as syntactic similarity between L1 and L2. The issue that cross-language syntactic similarity affects L2 processing is not new and has been investigated extensively in the past. It has recently come back to the attention in current bilingual research based on both behavioral and electrophysiological data. However, the conclusions are mainly drawn from the groups with Indo-European language pairs as their L1 and L2, neglecting English sentence processing by native Chinese learners to some extent. In addition, it is still unclear whether there is an interaction between cross-language syntactic similarity and L2 proficiency level in online L2 sentence processing.Using ERP(event-related potentials) and a grammaticality judgment task(GJT), the present study explores the effect of cross-language syntactic similarity on L2 syntactic processing by examining how native speakers of Chinese process three different syntactic constructions varying in degrees of cross-language syntactic similarity between English and Chinese:subject-verb word order that is instantiated similarly in English and Chinese,, passive voice marking that differs in English and Chinese, and'negative inversion'that is unique to English.The results of this study generally support the predictions of UCM for even late Chinese learners of English with low level of proficiency can exhibit native-like sensitivity to syntactic feature that is similar in Chinese and English, as evidenced by the P600 effect for subject-verb word order violations. The available data suggest that similarity between L1 (Chinese) and L2 (English) syntactic structures affects syntactic processing and representation in L2. And the effect of cross-language similarity was found to interact with L2 proficiency levels by comparing the L2 advanced and low groups. Low-proficiency L2 learners are only sensitive to violations of the similar construction. The high-proficiency L2 learners, in contrast, showed sensitivity not only to the similar construction, but also to the different syntactic feature, though the P600 observed for the violations of the latter is smaller in amplitudes and distribution. Based on the present study and the related studies, it seems reasonable to infer the L2 processing of bilinguals with high level of proficiency is less influenced by cross-language syntactic similarity assuming that crosslinguistic competition does not play an important role as L2 becomes more resonant, possibly overcoming the firm establishment of L1 system according to the UCM, and that cross-language syntactic similarity may more obviously affect the L2 syntactic processing by bilinguals with low level of proficiency. With regard to unique constructions, experimental results were in accordance with the UCM predictions showing no sensitivity to its violation when the strength of cues were relatively low-particularly for bilinguals with low level of proficiency who may not acquire the L2 features yet. Bilinguals with high level of proficiency didn't exhibit native-like online sensitivity despite the high accuracy in the behavioral performance. The findings in this study could inform English teaching and learning in China. Adult L2 education practice, especially when the students were at beginning stages of L2 learning or of low proficiency, may benefit from an instructional approach that takes the L1 Chinese into consideration and designs different levels of training in teaching accordingly like targeted practice for the syntactic constructions that are different in L1 Chinese and L2 English or unique in L2 English.