《Effects of L2 Proficiency on Lexical and Syntactic Transfer in Early L3 Acquisition英语专业本科生毕业论文(设计).doc》由会员分享，可在线阅读，更多相关《Effects of L2 Proficiency on Lexical and Syntactic Transfer in Early L3 Acquisition英语专业本科生毕业论文(设计).doc（32页珍藏版）》请在三一办公上搜索。
1、 本科生毕业论文（设计）Effects of L2 Proficiency on Lexical and Syntactic Transfer in Early L3 Acquisition题 目： 第二语言熟练度对第三语言习得中词汇及句法迁移的影响 系 别： 外国语言文学系 专 业： 英 语 学生姓名： 学 号： 指导教师： （职 称）二 年 月ABSTRACTThe research of third language acquisition is gaining increasing attention as acquiring a third language (L3) becomes
2、 a common phenomenon. And both generative and applied linguists have been focusing on the “cross linguistic influence” (CLI) on L3 acquisition. Typology factor is in the high light while L2 proficiency is one of the few discussed influences. It is generally acknowledged that in the early stage of le
3、arning a second foreign language (L3) typologically similar to the first one (L2), transfer plays a very important role. The author thus examines the difference of transfer rate of two separate Groups of distinct English proficiency (Group1 is high, Group 2 low) during the acquisition of some lexica
4、l and syntactic features in French. The self-designed empirical test consists of three sectors which illustrate different findings: in the acquisition of the same features between English and French, the two Groups do not have significant difference in transfer rate. However, in the acquisition of c
5、lose and completely different features between English and French, the two Groups did have significant differences: namely, the Group with higher English proficiency has significantly higher percentage of positive transfer rate in the acquisition of the French features close but not the same as Engl
6、ish. Furthermore, this Group outperforms the low English proficiency Group in the acquisition of distinct features of French which does not present in English. The stimulus recall after the tests also gave us some insights for teaching, such as how to make use of typology for transfer, and how to tu
7、rn the negative into the positive.Key words: third language acquisition, transfer, CLI, second language proficiency 摘要 随着学习第三语言的人们越来越多, 三语习得也开始备受关注。跨语言影响（CLI）是普通语言学和应用语言学家都相当关注的领域。其中迄今为止讨论最多的因素还是语言相似性,而对于语言熟练度因素则相关研究甚少。同时,在早期三语（与二语相近的语言）习得中,人们普遍认为存在迁移。因此,本文重在探讨二语熟练度对早期三语习得中迁移情况的影响。被试是两组二语（英语）熟练度截然不同
8、的三语学习者（组1为熟练度高,组2熟练度低）接受自行设计的法语测试,测试被分为三部分,结果分别证明了：在习得法语与英语完全一样的特征时,两组的迁移率在统计上没有显著差异。但是在以下两种情况下,两组有显著差异：在习得法语与英语相近但是有些许不同的特征时,组1正迁移率在统计上显著高于组2；在习得法英完全不同的语言特征时,组1分数在统计上显著高于组2。由此可见,熟练度在一定程度上确实与迁移相关。产生此现象的原因通过试后的“刺激回忆”录音可以推断是在于二语熟练度高的同学善于运用语言学习意识分析两种语言在结构层面上的相似和不同从而帮助理解和习得,而不是简单地猜测两者完全的等同。这正是教学中可以提高的地方
9、。关键词：三语习得, 迁移, 跨语言影响, 二语熟练度CONTENTSChapter One Introduction.11.1 Research background.1 1.2 Significance of the study.11.3 Organization of the thesis1Chapter Two Literature Review22.1 A general view22.2 Typology factor in CLI of multilingual lexical processing22.3 Proficiency factor32.4 Review of dom
10、estic studies.42.5 Summary of previous parts of review.5Chapter Three Theoretical Framework.6 3.1 The theoretical foundation of the hypotheses.63.1.1 The dominant source of transfer in L363.1.2 The effects of transfer in different situations.73.2 The definition of transfer.7Chapter Four Methodology9
11、4.1 Subjects.94.2 Research questions & Hypotheses.9 4.2.1 Research Questions.9 4.2.2 Hypotheses.104.3 Administration of test104.4 Instrumentation.104.5 Empirical test.11 4.5.1 Task design.11Chapter Five Results and Discussion125.1 General situations of the results.125.2 Results and discussion of res
12、earch question 1.125.3 Results and discussion of research question 2.145.4 The insights for teaching gained from the stimulus recalls.19Chapter Six Conclusion .216.1 Conclusion216.2 Limitations22 Selected Bibliography .23Appendix.26Chapter One Introduction 1.1 Research backgroundBoth generative ling
13、uists and applied linguists have now been interested in the new domain of third language acquisition (L3A), which used to be underestimated as a sub-question of SLA just as we did to the second language. Generative linguistic perspective has been focusing on language processing while applied linguis
14、tics on what “cross linguistic influences” (CLI) have on L3 learning, summarized as six main factors, including L2 status and proficiency. However, few investigate specifically how L2 proficiency influences L2 transfer and L3A, and some latest researches related to the influence of proficiency on tr
15、ansfer were for the generative linguistic purpose：to testify L3As difference from L2A. 1.2 Significance of the studyThe purpose of this research is two-fold: 1) to probe the effects of L2 proficiency on lexical and syntactic transfer in early L3A, in other words, whether second foreign language lear
16、ners with high proficiency of the first foreign language have more positive(or negative) transfer in early L3 lexical and syntactic acquisition. 2) To further investigate when and how different types of positive or negative transfer between an L2 (typologically similar to L3) and L3 occur, and how l
17、earners with high proficiency achieve to make use of the positive transfer and suppress the negative transfer (if they were), and ultimately arouse the bilingual learners awareness of their learning process.1.3 Organization of the thesisThis thesis is organized in 6 chapters. The first chapter is th
18、e general introduction, Chapter 2 the literature review. The remaining three chapters are devoted to the current study, including the design and implementation of the study, the results and analysis of the major findings and the last, the conclusion and limitation of the study as well as the future
19、prospects.Chapter Two Literature Review2.1 A general viewIn recent years, a number of researchers have started to look seriously at the phenomenon of L3/multilingualism as a separate domain of inquiry. Apart from the expanding studies done within applied linguistics, generative linguists are also sh
20、owing growing interest in it. While empirical studies in TLA (third language acquisition) are not as extensive as in SLA, a number of researchers have done important researches, especially within the last 15 years, on the influence of the L1 and L2 on the acquisition of an L3 (Cenoz 2001; Williams a
21、nd Hammarberg 1998; Ringbom 1983, 1987; Dewaele 1998; Klein 1995; Kellerman 2001; De Angelis and Selinker 2001; Singleton 1987). From the previous general descriptive studies on L3A, one can sum up six major factors that influence L3A process: Recency effect (Williams and Hammerberg 1998, Hammarberg
22、 2001); Proficiency (Ringbom 1987); L2 status (Willaims and Hammarberg 2001); Status of L3 in the society (Hammarberg 2001); Typological proximity; and Psychotypology (Bouvy 2000).2.2 Typology factor in CLI of multilingual lexical processingNumerous studies have shown that cross-linguistic influence
23、 from an L2 is favored if the L2 and L3 are typologically similar, especially if the L1 is typologically distant (Ringbom (1987); Williams and Hammarberg 1998; Stedje 1977; Dewaele 1998; Cenoz (2001; 2003); Kellerman (1995), cited from (Leung 2007). Ahukanna et al. (1981, as discussed in Carvalho 20
24、06) assessed interference from two previously learned languages (English was learners L2 and, Igbo, L1) in learners acquisition of French. Their results showed that English, a language more similar, caused more cross-linguistics interference than Igbo. Singleton (1987) points out a very influential
25、idea that psycho typology can not only explain why some linguistic errors (due to transfer) occur but also that a lack of consciousness on the part of the learner to typological similarities can lead to a lack of transfer where transfer could occur. Recently, Singleton and OLaoire (2004, as discusse
26、d in Leung 2007) looked at L1 EnglishL2 Irish (very advanced)L3 French using two elicitation instruments. They found that L1 English, which is considered typologically closer to French in lexical terms than Irish is, acts as the dominant source of lexical borrowing. They first interpreted these resu
27、lts strong support for (psycho)typology and against the L2 factor. But they had to acknowledge that, unlike English, the participants L2 Irish, though very advanced, was only a non-native language. In the 2005 study, therefore, Singleton and OLaoire looked at bilingual L1 IrishEnglish participants.
28、They successfully replicated the 2004 results: English but not Irish showed the privileged status in cross-lexical transfer. This is strong testimony to the role of (psycho-) typology in the lexical domain, which means that the language with typological proximity usually acts as the dominant source
29、of transfer despite the order of acquisition.The general conclusion of the studies was that, while all the previously acquired languages potentially affect the development of the L3 system, linguistic typology and psychotypology play a crucial role in determining the privileged source of transfer.2.
30、3 Proficiency factorWhile acknowledging the above positive role typology plays, we cannot deny other factors which interact with each other affecting the state of transfer, name one, proficiency. Ringbom (1987) ranks proficiency higher than formal similarities between languages as important for tran
31、sfer, commenting that, “transfer is frequent even without formal similarity, provided that the learner is fluent in the language from which he transfers.” Singleton (1987) conducted a case study to investigate the relationship between one learners perception of the degree of typological relatedness
32、between L1 (English), L2s (Irish, Latin, and Spanish) and L3 (French). Data from self report (stimulated recalls) revealed that the learner was indeed sensitive to the relatedness between Latin, one of the previously studied languages, and French, the target language. However, although Latin was clo
33、ser to French than English, English was more prominent possibly because the proficiency factor overrides distance. Cummins (1976) proposed a very important hypothesis on the previous languages influence on child L3 acquisition, threshold hypothesis: bilinguals need to achieve a certain level of prof
34、iciency before any cognitive benefits become evident. In effect, there are two thresholds; at the lowest level up to the first threshold, bilingual children have low levels of competence in both languages, and are, as such, limited bilinguals; this can result in negative cognitive effects. Between t
35、he first and second thresholds are bilingual children who have age-appropriate competence in one but not both languages, thus one language will be relatively weak. The cognitive effects are neither positive nor negative; in fact, cognitively, the bilingual child will show little difference to a mono
36、lingual child. However, bilingual children at the uppermost level, above the second threshold, have age-appropriate competence in both languages; sometimes known as proficient bilinguals, these children are likely to demonstrate positive cognitive effects.2.4 Review of domestic studies Within China,
37、 few studies can be found devoted to this field. There are two important papers close to what the author intends to explore. In Wan (2005)s study, he concluded that learners with high L2 English proficiency tend to bring their previous knowledge of foreign language learning into the acquisition of t
38、he third language (Japanese), and they are more possible to apply the strategy of avoiding using difficult grammatical rules. The second researcher Zhu(2000) conducted a study of the influence of the first foreign language (French) on the writing of the second foreign language (English), and indicat
39、ed that, on the syntactic level, there is no significant difference between subjects of higher and lower French proficiency. But on lexical level there is such difference, and the high proficiency Group has higher possibility to produce negative transfer because they have larger French vocabulary, a
40、nd this becomes the source of negative transfer since most French words are spelled differently from English.2.5 Summary of the previous parts of reviewFrom the above three sections of review one can see that, researches are done either 1) to find out the dominant source of lexical transfer, or 2) t
41、o study the pattern of L3 lexical or syntactic acquisition (of similar or different L3 features presented in previous learned languages). Hardly any specific study has been done to examine the condition that an adult learners L1 is much more proficient than L2 but typologically much more distant tha
42、n L2. Will the transfer dominantly be from L1, the one with proficiency, or L2, the one with much closer distance? From the previous studies, we can infer that the dominant source of transfer in this situation should be the L2. If L1 transfer happens from time to time, it may be due to the fact that
43、 the proficiency already overrides the distance, but the occasional overriding of proficiency can never take the dominant source. This inference is yet to be testified.This study is thus designed to compare the two Groups pf learners (Group 1 of native Chinese with high proficiency of L2 English and
44、 Group 2 of native Chinese with weak L2 English proficiency) who may have the corresponding features described in the last two phases of Cummins models respectively (The level of subjects in Group 1 corresponds to the level between the first and second threshold, while Group 2 to the level above the
45、 second threshold), with the difference that the subjects are adults, so the reason for positive/negative effects may not be the “age appropriate competence” proposed by Cummins; it could be something else to be explored.Chapter Three Theoretical Framework3.1 The theoretical foundation of the hypoth
46、esesThe theoretical foundation of the hypotheses initiated in the current study is based on several models of the precious studies in L3A.3.1.1 The dominant source of transfer in L3Numerous studies have concluded that language with typological proximity usually acts as the dominant source of transfer despite the order of ac
