An Empirical Research Into The Acquisition Of Relative Clauses By Chinese Learners

Due to its high frequency in daily life and unique property of recursion, Relative Clauses (RCs) have caught many researchers'attention. Studies been done on acquisition sequence of relative clause are centered on two angles: one is the functions that NPs serves in sub-clauses, manifested in different RC types, such as, SU, DO, OIO, OPREP, GEN and OCOMP; another is the embeddness of a relative clause, inserted in two position, including the matrix subject position and the matrix object position.Even though many researchers have done at home and abroad from different perspectives, there is still no consistent result achieved. Apart from that, most of researches just focus on a few of relative clause types at the expense of ignoring others. Two variables, the position and sentence pattern, play very important role in forming relative clauses. According to 2 positions by 6 sentence patterns, we can obtain 12 relative clauses. Therefore, this thesis, from the perspectives of comprehension and production, is devoted to putting forward an acquisition sequence of 12 relative clauses patterns by testing those seven hypotheses including Keenan and Comrie (1977)'s Noun Phrase Accessibility Hypothesis (NPAP), Kuno (1974)'s Perceptual Difficulty Hypothesis (PDH), Maewhinney and Pleh (1988)'s Perspective Shifting Hypothesis (PSH),Hamilton(1994)'s SO Hierarchy Hypothesis (SOHH), O'Glady's Structural Distance Hypothesis (2003), Sheldon's (1974) Parallel Function Hypothesis, Gibson et al (2005)'s Information Flow Hypothesis (IFH).The 60 non-English major freshmen participants are chosen from three departments of Sichuan International Studies University, i.e. Department of Chinese Language and Literature, Department of Chinese as a foreign language, and Department of Hotel Management. With the help of the Statistical Package of Social Science (SPSS 13.0), to test for the statistical significance of the observed differences, the test scores are submitted to a two-way repeated measures analysis of variance (ANOVA) using two within participants factors, RC types (SU versus DO versus OPREP) and matrix position (matrix subject versus matrix object position). The following acquisition sequence of relative clauses is achieved.(1) In grammaticality judgment test, the acquisition sequence of relative clauses is achieved, i.e. OGEN (1.8667) > SGEN (1.8000) > OS (1.7000) > SS (1.6000) > OO (1.5667) > SO (1.5000) > OOPREP (1.4333) > OIO (1.3667) > SOPREP (1.3000) > SIO (1.2333) > OOCOMP (.9833) > SOCOMP (.8167). (Here">"means"more accessible than")(2) In sentence combination test, the acquisition sequence of relative clauses is achieved, i.e OGEN (1.8333) > OS(1.7333) > SGEN (1.6667) > SS (1.6000) > OOPREP (1.5667) > OO (1.5000)> SOPREP (1.4333) > OIO (1.3333) > SO (1.3331) > SIO (1.2667) > OCOMP (1.2333) > SOCOMP (1.0667). (Here">"means"more accessible than")The results can be explained in this way:Firstly, in terms of comprehension, the acquisition sequence of relative clause is determined by the position of NPs, i.e., the positioning of NPs plays more important role than the RC types in the process of acquisition sequence. For Chinese learners, comprehension of a sentence doesn't mean to understand the structure of a sentence, but lies in the amount of vocabularies and schematic knowledge in their mind.Secondly, in terms of production, the difference between RC types is prominent while the effects of embeddness manifested themselves only within each RC type. This means that the production of a relative clause is mainly determined by syntactic complexity, i.e., the branching direction, Chinese, being the left branching language; and English, being the right branching language.