Chapter One Introduction
1.1 Motivations of the Study
Many linguists believe that language universals and linguistic typology can beapplied to second language acquisition. Comrie (1981) suggests that the use ofcommon tj^ology in the study of second language acquisition (SLA) can confirmwhether these universals exist. Hawkins (1987) believes that the implicationaluniversals can be used to predict the order of acquisition. These language universalsnot only are falsifiable in practice, but may also explain the process and nature oflanguage acquisition. Greenberg (1991: 37) also states that modem language typology"has been able to develop specific results in the form of a typical implication, whichso far has played a major role in providing research hypotheses for SLA study".Studies in SLA can not only find common ground between the first language (LI) andthe second language (L2)，but investigate the difference between the two languagesand the fundamental causes. What's more，it also allows us to explore the impact ofLI on SLA process as well as the language development and the transfer trajectoryfrom LI to L2.During the last four decades, the subject of relative clauses attracts extensiveattention in linguistic studies. Most of these studies focus on English children's LIacquisition of English relative clauses. However，research about the acquisition ofrelative clauses by L2 learners (especially primary learners) whose mother tongue isChinese is comparatively fewer.
……….
1.2 Purposes and Significance of the Study
The main purpose of the present study is to investigate the acquisition of Englishrelative clauses by Chinese learners in junior high school and discuss the differencesbetween two teaching methods (explicit and implicit instruction) for learners tocomprehend the relative clauses. To be specific, the following objectives will beachieved.Firstly, the present research not only aims at investigating the acquisition ofrelative clauses but also intends to integrate the element of word order into the studybecause there exist plentiful differences between the word order of Chinese andEnglish. The combination of the second language acquisition and word order studywould be a novel research perspective.Secondly, the study plans to conduct instructions among junior high schoolstudents who are paid less attention because most of the studies on the acquisition ofrelative clauses target at students or adults whose English are more proficient.However, the investigation of the students whose knowledge of relative clauses is atan earlier stage would be more meaningful.Thirdly, from the aspect of research method, most of the existing experiments arecarried out through language tests (e.g. sentence translation, sentence combinations,grammatical judgment) but the learners' active usage of the knowledge is neglected.Therefore, a spontaneous output test (the composition writing task) is designed toinduce the subjects to produce the language materials for analyzing.
………..
Chapter Two Literature Review
2.1 Studies on Second Language Acquisition of Relative Clauses
This part attempts to review previous studies on second language acquisition ofRCs abroad and home and gives a brief summary and comment based on theliterature.Referring to the definition of relative clauses, linguists define it differently fromvarious aspects. In their Dictionary of Language and Linguistics，Hartman & Stork(1972) states that the relative clause is a subordinate clause guided by a relativepronoun or adverb. This definition targets at English relative clauses. Andrews (1985)，from a more comprehensive aspect, defines relative clause as a subordinate clausewhich modifies all parts of the sentence except itself and is semantically consistentwith the modified constituent. The modified part is the head noun while thesemantically equivalent subordinate clause is the relative clause. Andrews' definitionis no longer restricted to English but appropriate for other languages.All in all, relative clauses hold different features in different languages. Comrie(1981) believes in Asian languages RCs should be regarded as attributive clauseswhile in European languages, RCs keep the original name for the former ones whichdo not involve argument structure and do not have filler-gap dependency. In thisthesis，we talk about English relative clauses.
……….
2.2 Studies on Differences of Word Order Between English and Chinese
Word order 1 plays an essential role in conveying important grammatical information. Abundant morphological changes can be found in some languages (e.g. Indo-European languages) in which the grammatical meanings are expressed withoutemphasizing the importance of word order in a sentence. Modem standard Englishlays a great emphasis on language form, thus most of its sentence types have a fixedword order. However, unlike English, a Chinese sentence is meaningless without theassistance of word order in most cases. This part attempts to review Chinese andEnglish word order differences, especially in RCs (An, 2006). It is well-known that from the syntactic level, the fimdamental sentence structureof English is SVO (Subject-Verb-Object) while Chinese has both similarities anddifferences with SVO and SOV (Subject-Object-Verb，e.g. Japanese) language，whicharouse great interests of researchers at home and abroad. For instance, Shyu (2001)concerns SOV word order in Mandarin Chinese. With the assumption of basic SVOword order [1]，Shyu proposes movement account for deriving SOV word order [2].
……..
Chapter Three Experimental Design and Methodology........ 21
3.1 Research Questions........ 21
3.2 Research Methodology........ 21
3.2.1 Methods in the Tests........ 22
3.2.2 Methods in the Instruction........ 24
3.3 Subjects........26
3.4 Test Materials ........27
3.5 Research Procedures........ 29
3.6 Data Collection and Scoring........ 31
Chapter Four Results and Discussion........ 33
4.1 Results and Discussion for the Two Tests........ 33
4.2 Analyses and Discussion for the Compositions.........69
Chapter Five Conclusion ........ 73
5.1 Major Findings of the Present Study........ 73
5.2 Implications of the Present Study........ 74
5.3 Limitations and Recommendation for the Future Research........ 76
Chapter Four Results and Discussion
4.1 Results and Discussion for the Two Tests
In this part，the results from the pre-test and the post-test are presented anddiscussed. From the vertical view，differences between the three test groups areexamined while the discrepancies between the RC sentence types and the relationshipbetween the two tasks are analyzed in the horizontal way. In addition, the data fromthe two tests are put together to make fiirther comparisons. In the Multiple Choice Task, the subjects in the control group and theexperimental group were required to choose any option or options they supposed werecorrect. Through the scoring methods mentioned in 3.6, the general performance ofthe control and the experimental groups is illustrated in Table 4.1 (A detailedstatistical table is shown in Appendix 7). There exist negative values in the scoresbecause of the particular marking methods. The range of the total score value is from-21 to 21. What's more, two decimal places are reserved in all the cases.
……….
Conclusion
The present study is a research on junior high school students' acquisition ofrelative clauses through the implicit and explicit instructions. In this chapter, themajor findings of the research are firstly presented and then the implications andlimitations of the study are then put forward. The present research aims to investigate the current state of knowledge about RCsby Chinese junior high school students, and through two different instructing methods,which are the implicit instruction and the explicit instruction, the study attempts toexplore the differences of the two instructing approaches and to make suggestions forthe future teaching of RC knowledge.Generally, the subjects in the experiment hold a comparatively same initial state ofthe knowledge of RCs before the instructions. However, the subjects' knowledgeabout the RC sentence types are different, which reveals the difficulty order of thefour RC sentence types. When concerning the results of the Multiple Choice Task,which represents the general knowledge of the sentence types, we discover that theOO type is the easiest to comprehend, then followed by the SS type. The SO and OStypes are more difficult to understand. The difficulty order (from the hardest to theeasiest) can be shown as: SO/OS > SS >OO. When we consider the result of theSentence Structure Analysis Task，which indicates the subjects' comprehension of thestructure of RC sentences, it is detected that the structure of the OS type is easiest tograsp while the other three are relatively more difficult to understand.
…………
Reference (omitted)
