英语反身代词和宾语代词的二语习得研究

  

L2 Acquisition of English Reflexives and Objects（英语反身代词和宾语代词的二语习得研究）探讨了英语人称代词的母语和二语习得研究。这里的人称代词包含反身代词和第三人称代词。在对相关的文献详细综述后，L2 Acquisition of English Reflexives and Objects（英语反身代词和宾语代词的二语习得研究）进行了一项实证研究，即比较高级英语水平的中国学生在英语反身代词以及宾语代词两个领域的习得。实验结果表明，他们对前者局部约束性的掌握远好于对后者有形性的掌握。这一现象可以通过语法分析理论得到解释：参数的重新设定要靠语言依据来激活；如果二语输入中缺少能够激活参数值重新设定的语言依据，那么即使处在高级英语水平阶段，二语学习者的中介语依然会呈现母语的参数值特征。

 

Contents
Acknowledgements
前言
Abstract
Chapter 1 Universal Grammar and Second Language Acquisition
1.1 Introduction
1.2 UG and SLA
1.2.1 Principles and Parameters of UG
1.2.2 SLA
1.2.2.1 The Binding of Reflexives
1.2.2.2 The Licensing of Null Objects
1.2.3 Three Major UG-based SLA Theories
1.2.3.1 The Full Transfer Full Access Hypothesis
1.2.3.2 The Minimal Trees Hypothesis
1.2.3.3 The Failed Functional Features Hypothesis
1.3 Summary
Chapter 2 Cross-Linguistic Variation on the Binding of Reflexives and Licensing of Null Objects
2.1 Introduction
2.2 Cross-linguistic Variation on the Binding of Reflexives
2.2.1 The Parameterized Approach
2.2.2 The Move-to-Infl Approach
2.2.3 The Relativized SUBJECT Approach
2.2.4 A New Approach to the Cross-linguistic Variation on the Binding of Reflexives
2.3 Cross-linguistic Variation on the Licensing of Null Objects
2.4 The Binding of Reflexives and the Licensing of Null Objects
2.5 Summary
Chapter 3 Studies on the Acquisition of Reflexive Binding and Null Objects
3.1 Introduction
3.2 Reflexive Binding in L1 Acquisition
3.3 Reflexive Binding in L2 Acquisition
3.3.1 Hirakawa(1990)
3.3.2 Yuan(1994)
3.3.3 MacLaughlin(1998)
3.3.4 Thomas(1995)
3.3.5 Jiang(2009b)
3.3.6 White et al.(1997)
3.4 Null Objects in L1 Acquisition
3.5 Null Objects in L2 Acquisition
3.5.1 Yuan(1997)
3.5.2 Park(2004)
3.6 Summary
Chapter 4 Main Study
4.1 Introduction
4.2 The Study
4.2.1 Subjects
4.2.2 Materials and Procedure
4.3 Results
4.3.1 Reflexive Binding
4.3.1.1 Group Results
4.3.1.2 Individual Results
4.3.1.3 Summary
4.3.2 Null Objects
4.3.2.1 Group Results
4.3.2.2 Individual Results
4.3.2.3 Summary
4.3.3 Correlation between Reflexive Binding and Null Objects
4.4 Summary
Chapter 5 Discussion
5.1 Introduction
5.2 A-Binding/A’-Binding Asymmetry
5.2.1 Native Language Transfer in L2 Acquisition
5.2.2 Learning Strategies in L2 Acquisition
5.2.3 A New Approach Involving an Interaction of the L1 Grammar and the L2 Input
5.2.3.1 Morphological Complexity of the English Reflexive
5.2.3.2 Binding to Non-Subjects
5.2.3.3 Overt Agreement Morphology
5.2.4 The Implications of the Findings for UG-based SLA Theories
5.3 The Finite/Non-finite Asymmetry in the A-Binding Domain
5.4 The A-bound/A’-bound Null Object Asymmetry in the A’-binding Domain
5.5 Conclusion
References
Appendix I The Main Test
Appendix II The Proficiency Test
List of Tables
1.1 Parametric values for Turkish and German:Cevdet’s three interlanguage stages(Tur=Turkish;S1=Stage 1;S2=Stage 2;S3=Stage 3;Ger=German)
1.2 Accurate judgments on relative clauses(in%)
2.1 Illustration of the GCP and PAP(from Manzini and Wexler 1987)
2.2 Three types of binding patterns
3.1 Responses of”yes”(in %)
3.2 The number of subjects which are classified into each binding pattern
3.3 Responses of”true”(in%)
3.4 Number of subjects in each task
3.5 Total omission of subjects and objects(in%)(Park 2004:5)
3.6 The licensing conditions of null subjects and null objects
3.7 Summary of findings of the L2 acquisition of English
4.1 Subjects’ background information(SD=standard deviation)
4.2 Sentence types used to test L2 learners’ interpretation of English reflexives
4.3 Sentence types used to test L2 learners’ knowledge of English overt objects
4.4 Mean accuracy scores on local and LD antecedents
4.5 Mean accurate judgments of LD antecedents on finite versus non-finite clauses
4.6 Three types of binding patterns
4.7 Classification of respondents according to the 80% consistency criterion
4.8 Criteria for rejection of LD referential antecedents(RAs)and LD quantified antecedents(QAs)
4.9 Number of Chinese speakers with full,partial,or no rejection of LD RAs and LD QAs according to the criteria levels set in Table 4.8
4.10 Overall mean accuracy scores on null objects and converted percentages
4.11 Mean accuracy scores on A-bound and A’-bound null objects
4.12 Error rates for Chinese speakers and native controls on each null-object sentence(percentage of errors per sentence)
4.13 Number of participants giving consistent responses to null objects
4.14 Accuracy scores on reflexive binding and null objects for each Chinese speaker
5.1 Cevdet’s three successive developmental stages
List of Figures
4.1 Mean accuracy scores on local antecedents by sentence type
4.2 Mean accuracy scores on LD antecedents by sentence type
4.3 Mean accurate judgments(in%)of local binding and LD binding involving quantified vs.referential antecedents in the finite sentences
4.4 Percentage of errors per null-object sentence
4.5 Accuracy scores on reflexive binding and null objects for each L2 subject

 

 

Universal Grammar and Second Language Acquisition
1.1 Introduction
It will be presupposed in this book, without argument, that UG constrains L1 acquisition, as well as adult native-speaker knowledge of language. The strongest argument for this claim comes from “Pla-to’s problem”: how small children finally get to master a language which is so complex that it is impossible for it to be derived merely from limited language evidence and little explicit instruction, for ex-ample, how they know that certain structures and interpretations are not permitted. In other words, there is a mismatch between the input (the utterances that the child is exposed to), and the output (the un-conscious grammatical knowledge that the child acquires). This mis-match also gives rise to what is known as the problem of the poverty of the stimulus or the logical problem of language acquisition. Given such underdetermination, the claim is that it would be impossible to ac-count for the L1 acquirer’s achievement without postulating a built-in system of universal linguistic principles and grammatical properties (Baker and McCarthy 1981; Hornstein and Lightfoot 1981). UG, then, is proposed as an explanation of how it is that language acquirers come to know, unconsciously, properties of grammar that go far be-yond the input in various respects. The idea is that such properties do not have to be learned; they are part of the “advanced knowledge” that the child brings to bear on the task of acquiring a language.
L2 Acquisition of English Reflexives and Objects
According to Chomsky (1980), UG constitutes the child’s initial state and the knowledge that the child is equipped with. The primary linguistic data are critical in helping the child to determine the precise form that the grammar must take. As the child takes account of the input, a language-specific lexicon is built up, and parameters of UG are set to values appropriate for the language in question. The gram-mar may be restructured over the course of time, as the child becomes responsive to different properties of the input. In due course, the child arrives at a steady state grammar for the mother tongue.
More and more researchers realized that, as linguistic theories such as Government-Binding (Chomsky 1981), Minimalism (Chomsky 1995) or Optimality Theory (Archangeli and Langendoen 1997) have developed, there have been changes in how universal principles and parameters have been formalized, that is to say, changes in what UG is assumed to consist of. For example, the numerous and very specific principles of the early days of generative theory, such as many of the original Island Constraints (Ross 1967), have been replaced with more general, invariant economy principles (e.g. Chomsky 1991), as well as computational operations, such as Move and Merge (Marantz 1995). Parameters have gradually become more constrained, now being largely associated with the lexicon: properties of items that enter into a computation, for example, may vary in feature composition and feature strength, with associated syntactic consequences.
Such ongoing changes in the definition of UG are a reflection of development and growth within linguistic theory. Nevertheless, re-gardless of how UG is formalized, there is a consensus (within the generative linguistic perspective) that it’s better to assume some in-nate and specifically linguistic constraints on grammars and grammar acquisition, otherwise it would be hard to imagine how certain ab-stract, subtle and complex properties of language are to be acquired.
L2 learners face a task parallel to that of L1 acquirers, namely the
Chapter 1 Universal Grammar and Second Language Acquisition
need to arrive at a linguistic system which accounts for the L2 input, allowing the learner to understand and speak the L2. Given this ap-parent similarity, the question of whether UG also mediates L2 acqui-sition, and to what extent, has been investigated and debated since the early 1980s. On one side of the debate, observable developmental dif-ferences between L2 and L1 learners have led some researchers to the conclusion that UG is not involved at all in the construction of L2 mental grammars by older learners, so that L2 acquisition is “funda-mentally different” from L1 acquisition (Bley-Vroman 1989; Clahsen and Muysken 1986; Felix 1985; Newmeyer 1998). In that case, older L2 learners can use their knowledge of the L2, but that knowledge is de-rived from other cognitive abilities and, hence, is fundamentally dif-ferent from that of native speakers. On the other side of the debate, is the position that the mental grammars that L2 speakers develop? ILGs?are indeed UG-constrained. For those who maintained this claim in early generative work on L2 acquisition, the explanation of L2?L1 differences has been a major task. According to Hawkins (2001b), Eubank’s (1994, 1996) theory of Valueless Features is a UG-based at-tempt to explain why L2 initial-state grammars might differ from L1 initial-state grammars, as is Vainikka and Young-Scholten’s (1994, 1996) Minimal Trees theory. The former argues that ILGs are constrained by UG, with inertness of features only a temporary property; the latter argues that all of UG is available in L2 acquisition, although some properties (functional categories) emerge after others. More recent work has shifted attention from issues like “whether ILGs are identi-cal to native-speaker grammars” to a closer examination of the nature of ILGs, with particular focus on whether ILGs are of the type sanc-tioned by UG. If it can be shown that L2 representations fall within the range sanctioned by UG, then whether or not L2 and L1 speakers have the same representations for a particular target language is irrelevant (Dekydtspotter and Sprouse 2000; Dekydtspotter, Sprouse and An-derson 1998; Epstein, Flynn and Martohardjono 1996; Schwartz and
L2 Acquisition of English Reflexives and Objects
Sprouse 2000). This detailed focus on the grammatical properties of ILGs remains characteristic of current research.
The aim of this book is to investigate whether ILGs are UG-constrained in detail in the domain of “pronominality”, a term that will be used here to cover reflexive binding and also the (im)possibility of null objects. “Pronominality” is a term used for an-alytical convenience since it covers a domain of enquiry in SLA that involves questions relating to the role of the L1, and development of knowledge which is underdetermined by input. It should be stressed however that “pronominality” has no status in linguistic theory. There are, so far as we know, no theoretical links between the binding do-mains of reflexives and the licensing of null objects.
The book is organized as follows: Chapter 1 shows what the book is actually dealing with. Chapter 2 presents the linguistic assumptions about the differences between English and Chinese1 in terms of the binding of reflexives and the licensing of null objects, and also sug-gests why it is interesting to consider these two phenomena together. Chapter 3 reviews the findings of some selected previous L1 and L2 acquisition studies on reflexive binding and null-object phenomenon. Chapter 4 reports the details of an empirical study. The discussion of the data and the implications they have for SLA theories are presented in Chapter 5.
1.2 UG and SLA
1.2.1 Principles and Parameters of UG
UG is a biologically-determined module of mind which defines a set of principles for the construction of mental grammars, and allows some variation in the way that experience with language can be con-verted into grammatical representations. These possibilities for varia-
Chapter 1 Universal Grammar and Second Language Acquisition
tion are referred to standardly as “parameters”, and by hypothesis the options available are highly restricted. In this theory, principles are by hypothesis true for all languages and do not have to be acquired. Pa-rameters, however, which allow for variation from language to lan-guage, require experience with particular samples of language. On the basis of experience one option rather than another is fixed.
Most parameters are assumed to be binary, that is, they have only two settings, the choices being predetermined by UG. L1 acquisition consists partly of setting parameters, the appropriate setting being triggered by the input that the child is exposed to. A central claim of parameter theory is that a single parameter setting brings together a cluster of apparently disparate syntactic properties (Chomsky 1981). This, for example, was part of the rationale for the Null Subject Pa-rameter, which related the possibility of null subjects to other syntac-tic and morphological properties found in null subject languages (Chomsky 1981; Jaeggli 1982; Rizzi 1982). The insight behind the pro-posal for parameters is that they should severely reduce the acquisi-tion task. Rather than learning a number of seemingly unrelated properties individually, the child has only to discover the appropriate setting of a parameter and a range of associated syntactic properties follows automatically. Some L1 acquisition research has provided ev-idence in favour of clustering, showing that properties which are ar-gued to be consequences of a particular parameter setting emerge at about the same time (e.g. Hyams 1986; Snyder and Stromswold 1997).
Following current proposals, parametric differences between grammars relate to properties of lexical items, particularly so-called functional categories (Borer 1984; Chomsky 1995; Ouhalla 1991; Pol-lock 1989). Linguistic theory distinguishes between lexical catego-ries-verb (V), noun, adjective, adverb (Adv), preposition?and func-tional categories, including complementizer (Comp or C), inflection (Infl or I) (often split into agreement (Agr) and tense (T)), negation
L2 Acquisition of English Reflexives and Objects
(Neg), determiner (Det), number (Num), as well as others. Functional categories have certain formal features associated with them (such as tense, number, person, gender and case). Functional categories and features form part of the UG theory.
According to White (2003), there are three potential sources of cross-linguistic variation relating to functional categories:
1) Languages can differ as to which functional categories are re-alized in the grammar. On some accounts, for example, Japa-nese lacks the category Det (Fukui and Speas 1986).
2) The features of a particular functional category can vary from language to language. For instance, French has a gender feature, while English does not.
3) Features are said to vary in strength: a feature can be strong in one language and weak in another, with a range of syntactic consequences. For example, Infl features are strong in French and weak in English, resulting in certain word-order alterna-tions between the two languages.
The lexicons of different languages, then, vary as to which functional categories and features are instantiated and what the strength of var-ious features may be. Such variation has a variety of syntactic effects.
1.2.2 SLA
UG as a system of innate principles and parameters provides constraints on grammars in the course of L1 acquisition. However, there has been considerable controversy about whether the same in-nate ability is involved in L2 acquisition by adult learners. For this reason, in testing for UG it has become clear that it is not sufficient to show that L2 speakers in performance produce or understand utter-ances in the same way as native speakers. While on the surface this would be consistent with those speakers having UG-constrained