Studies in Second Language Acquisition of Chinese 9781783092093

Table of contents :
Contents
Contributors
Preface
1. Influence of L1 Thinking for Speaking on Use of an L2: The Case of Path Expressions by English-Speaking Learners of Chinese
2. Pragmatic Development: An Exploratory Study of Requests by Learners of Chinese
3. Peer/Group Interaction in a Mandarin Chinese Study Abroad Context
4. Task-based Language Teaching of Chinese in a Study Abroad Context: A Learner Perspective
5. The Relationship between the Effectiveness of Recasts and Working Memory in the Learning of Different Linguistic Structures
6. The Effect of Mixed-sensory Mode Presentation on Retaining Graphic Features of Chinese Characters
Epilogue
Index

Citation preview

Studies in Second Language Acquisition of Chinese

SECOND LANGUAGE ACQUISITION Series Editor: Professor David Singleton, University of Pannonia, Hungary, and Fellow Emeritus, Trinity College, Dublin, Ireland This series brings together titles dealing with a variety of aspects of language acquisition and processing in situations where a language or languages other than the native language is involved. Second language is thus interpreted in its broadest possible sense. The volumes included in the series all offer in their different ways, on the one hand, exposition and discussion of empirical findings and, on the other, some degree of theoretical reflection. In this latter connection, no particular theoretical stance is privileged in the series; nor is any relevant perspective – sociolinguistic, psycholinguistic, neurolinguistic, etc. – deemed out of place. The intended readership of the series includes final-year undergraduates working on second language acquisition projects, postgraduate students involved in second language acquisition research, and researchers and teachers in general whose interests include a second language acquisition component. Full details of all the books in this series and of all our other publications can be found on http://www.multilingual-matters.com, or by writing to Multilingual Matters, St Nicholas House, 31–34 High Street, Bristol BS1 2AW, UK.

SECOND LANGUAGE ACQUISITION: 77

Studies in Second Language Acquisition of Chinese

Edited by ZhaoHong Han

MULTILINGUAL MATTERS Bristol • Buffalo • Toronto

Library of Congress Cataloging in Publication Data Studies in Second Language Acquisition of Chinese/Edited by ZhaoHong Han. Second Language Acquisition: 77 Includes bibliographical references and index. 1. Chinese language—Study and teaching—Foreign speakers. 2. Second language acquisition. I. Han, Zhaohong—editor of compilation. PL1065.S74 2014 495.101'9–dc23 2014001669 British Library Cataloguing in Publication Data A catalogue entry for this book is available from the British Library. ISBN-13: 978-1-78309-208-6 (hbk) Multilingual Matters UK: St Nicholas House, 31–34 High Street, Bristol BS1 2AW, UK. USA: UTP, 2250 Military Road, Tonawanda, NY 14150, USA. Canada: UTP, 5201 Dufferin Street, North York, Ontario M3H 5T8, Canada. Website: www.multilingual-matters.com Twitter: Multi_Ling_Mat Facebook: https://www.facebook.com/multilingualmatters Blog: www.channelviewpublications.wordpress.com Copyright © 2014 ZhaoHong Han and the authors of individual chapters. All rights reserved. No part of this work may be reproduced in any form or by any means without permission in writing from the publisher. The policy of Multilingual Matters/Channel View Publications is to use papers that are natural, renewable and recyclable products, made from wood grown in sustainable forests. In the manufacturing process of our books, and to further support our policy, preference is given to printers that have FSC and PEFC Chain of Custody certification. The FSC and/or PEFC logos will appear on those books where full certification has been granted to the printer concerned. Typeset by Techset Composition India (P) Ltd., Bangalore and Chennai, India. Printed and bound in Great Britain by the CPI Group (UK Ltd), Croydon, CRO 4YY.

Contents

Contributors

vii

Preface

ix

1

Influence of L1 Thinking for Speaking on Use of an L2: The Case of Path Expressions by English-Speaking Learners of Chinese Shu-Ling Wu

2

Pragmatic Development: An Exploratory Study of Requests by Learners of Chinese Xiaohong Wen

30

3

Peer/Group Interaction in a Mandarin Chinese Study Abroad Context Li Jin

57

4

Task-based Language Teaching of Chinese in a Study Abroad Context: A Learner Perspective ZhaoHong Han and Joo-oeck Maeng

80

5

The Relationship between the Effectiveness of Recasts and Working Memory in the Learning of Different Linguistic Structures Shaofeng Li

6

1

103

The Effect of Mixed-sensory Mode Presentation on Retaining Graphic Features of Chinese Characters Yongan Wu

126

Epilogue

142

Index

150

v

Contributors

ZhaoHong Han is Professor of Language and Education at Teachers College, Columbia University, USA. Her research interests straddle second language learnability and teachability. Her most recent book (coauthored with Teresa Cadierno) is Linguistic Relativity in SLA: Thinking for Speaking (2010, Multilingual Matters). Li Jin is Assistant Professor in the Department of Modern Languages at DePaul University, Chicago, IL. She obtained her PhD in Second Language Acquisition/Instructional Technology at the University of South Florida in 2007. Her research interests include Chinese as a foreign language pedagogy, technology-enhanced foreign language teaching and learning, and study abroad curriculum. Shaofeng Li is Senior Lecturer of Applied Language Studies at the University of Auckland, where he teaches postgraduate and undergraduate courses in second language acquisition. His primary interest is in investigating the interactions between learning conditions (e.g. implicit versus explicit; task type) and individual difference variables such as language aptitude and working memory. His other research interests include form-focused instruction, quantitative research methods, and language testing. His recent and forthcoming publications appear in Language Learning, Modern Language Journal, Applied Linguistics, Language Teaching Research, ELT Journal and The RELC Journal. Joo-oeck Maeng is Professor of Chinese at the Hankuk University of Foreign Studies in South Korea. His research interests include second language acquisition of Chinese and Chinese pedagogy. Xiaohong Wen is Professor of Applied Linguistics and Director of Chinese Studies at the University of Houston. She has research interests in affective factors, acquisition of Chinese as a foreign language and research-based curriculum design and instruction. Her publications include two books, Studies of Chinese Language Acquisition by English Speakers (2008), and Learning and vii

viii Studies in Second L anguage Acquisit ion of Chinese

Teaching Chinese as a Second Language (2012), published by the Peking University Press, as well as 30 articles in refereed journals. Shu-Ling Wu is Assistant Professor at the Defense Language Institute. She received her PhD from the University of Hawaii at Manoa (2011). Her areas of research include second and heritage language acquisition, applied cognitive linguistics, and language testing and measurement. She published an article on L2 learning of motion event expressions in Language Learning (2011). Yongan Wu, PhD, completed his master’s and doctoral degrees at the University of Oklahoma, majoring in English education, and is now Assistant Professor of Chinese at the University of North Florida. His research interests include second language acquisition and cross-cultural studies.

Preface

With the rapid ascension of China as a world economic and political power has come a surge of interest in learning Chinese as a second or foreign language. According to the People’s Daily, there are currently over 40 million learners of Chinese worldwide. In the United States alone, the number of Chinese programs in the pre-K-12 sector has increased by more than 200% since 2004, and over 1600 American public and private schools are teaching Chinese, up from 300 or so a decade ago, according to the New York Times. Chinese is also overtaking German to become the third most tested language – after Spanish and French – in the Advanced Placement Test. The scope of learners has been rapidly expanding, going beyond the traditional population of heritage speakers to speakers of other languages. Such exponential growth of interest in learning Chinese is, however, markedly undercut by a lack of qualified instructors, who not only have strong language skills, but who also know how to teach the language effectively and efficiently. The burgeoning demand for quality instruction is, in turn, woefully underserved by a lack of research-based understanding of the nature and process of learning and, correspondingly, a lack of understanding of what kinds of pedagogical conditions may improve the efficacy of learning, a gap this book is helping to fill. Research on second language (L2) acquisition of Chinese has existed, but for a long time has been confined to a narrow range of topics and perspectives. Recent years have begun to see steady changes (Everson & Shen, 2010; Wen, 2008; Zhao, 2011) both in the scope of inquiry and in methodology, as is, in part, mirrored in the present volume. This book presents a collection of studies exploring aspects of the learning of Chinese as an additional language, cutting across the domains of morphosyntax, pragmatics, cognitive capacity, interactional learning and instructed learning, via a variety of conceptual frameworks and methodological strategies. Chapter 1 (Wu) traces the acquisition of L2 lexicalization patterns for encoding motion events. Drawing on Talmy’s (1985) typological framework of verb-framed versus satellite-framed languages, the study examined intra-typological variation in lexicalization patterns between English (the L1) and Chinese (the L2) and the development of ix

x Studies in Second L anguage Acquisit ion of Chinese

Chinese motion expressions in English-speaking learners. With a crosssectional design and through a meticulous analysis of oral production data, the study identified areas of ease and difficulty in acquiring the L2 thinking-for-speaking patterns. Chapter 2 (Wen) looks into the pragmatic development of the speech act of request in English-speaking learners of Chinese. Using the standard procedure of discourse completion, the study elicited data from non-heritage speakers of Chinese, leading to findings suggesting, inter alia, that pragmatic development is constrained by linguistic development. Chapter 3 (Jin) investigates peer interaction in the context of a study abroad program. From a sociocultural perspective, and with the help of surveys, interviews, observations and journal entries, the study qualitatively explored both the effects of peer interaction on the learning of Chinese and learners’ perception of their experience, uncovering both the complex learning affordances provided by peer interaction and an interplay of multiple factors, such as proficiency, learning style and personal goals, in determining learners’ attitude toward, and participation in, peer interaction. Chapter 4 (Han and Maeng) preliminarily investigates learners as task-doers. The study surveyed participants’ reactions to a summer study abroad task-based language teaching program. Results showed, among other things, that attending the program boosted the learners’ confidence and reduced their fear in speaking Chinese. The chapter ends with a lengthy list of questions inviting further investigation. Chapter 5 (Li) investigates the relationship between the effectiveness of recasts and working memory in the learning of different linguistic structures. The study employed an experimental design involving an experimental and a control group, and measured treatment effects and working memory capacity. Correlational results revealed that the different structures were differentially amenable to recasts, and that working memory selectively correlated with the learning of one of the target structures. Chapter 6 (Wu) examines the effects of input modality on learning retention. Aural and visual input was compared to visual input only and their relative impact on short-term and longer term retention of Chinese characters in beginning learners. Factorial analyses of presentation mode and retention assessed via recognition and production tasks indicated a superiority of a dual-mode presentation of input over a mono-mode presentation of input, highlighting the value of aural input to learning characters. This book is intended to serve both as a sounding board and a springboard. It provides a display of current thoughts and activities in the domain of L2-Chinese research and, at the same time, provokes further questions, thereby opening up further avenues for future research. Some of these questions are taken up in the epilogue of the book. To the extent that it covers a wide spectrum of issues relating to L2 learning of Chinese, the book offers insights that will prove useful to researchers (including graduate students) as well as curriculum developers, materials writers and teachers.

Pref ace

xi

The ideas presented in some of the chapters were incubated at The First Roundtable in Second Language Studies: L2 Acquisition of Chinese, held at Teachers College, Columbia University and organized by Yayun Sun, Kun-wan Choong, Shaoyan Qi and Hye-won Shin. The production of this volume would not have been possible without the thoughtful participation of its contributors, nor without the assistance of Eun-Young Kang and John Andras Molnar. It is always a pleasure to publish with Multilingual Matters. ZhaoHong Han New York

References Everson, M. and Shen, H.H. (eds) (2010) Research among Learners of Chinese as a Foreign Language. Honolulu, HI: National Foreign Language Resource Center, University of Hawai’i at Manoa. Talmy, L. (1985) Lexicalization patterns: Semantic structure in lexical forms. In T. Shopen (ed.) Language Typology and Syntactic Description, Vol. 3: Grammatical Categories and the Lexicon (pp. 57–149). Cambridge: Cambridge University Press. Wen, X.H. (2008) Studies of Chinese Language Acquisition by English-Speakers: From Theories to Practice. Peking: Peking University Press. Zhao, Y. (2011) A tree in the wood: A review of research on L2 Chinese acquisition. Second Language Research 27 (4), 559–572.

1

Influence of L1 Thinking for Speaking on Use of an L2: The Case of Path Expressions by English-Speaking Learners of Chinese Shu-Ling Wu

The relationship between spatial language and cognition has been the subject of an active and growing line of research (e.g. Bowerman, 1996; Choi & Bowerman, 1991; Gathercole, 2009; Hickman & Robert, 2006; Narasimhan & Brown, 2009; Slobin, 1998; Slobin et al., 2010; Talmy, 1985, 2000; see also Guo et al., 2009). Space is fundamental to human cognition. Children learn to describe spatial relations and motion events at a very young age (Berman & Slobin, 1994; Bowerman, 1996). Although all languages provide ways to talk about spatial relationships and motion, features of the same spatial information can be mapped to different linguistic units and be selectively represented in foregrounded or backgrounded constituents across different languages. As illustrated by Langacker’s (1987, 1991) discussion of construal, a speaker can subjectively decide what aspects of a situation he or she selects to portray. For example, when describing the motion event of a plane taking off, individuals may choose to profile different points of the trajectory, as in the plane flew off the runway, where the speaker stresses the source of the motion; whereas in the plane flew into the sky, where the speaker highlights the goal. Moreover, previous studies provide converging evidence that speakers of different languages conceptualize spatial relationships and motion in a language-specific manner (e.g. Bowerman, 1996; McNeill & Duncan, 2000; Slobin, 1996b, 2004; Slobin et al., 2010; Talmy, 1985, 2000). For instance, the spatial category that the English placement verb put can describe is subdivided into two placement verbs, legen ‘lay’ and stellen ‘make stand’ in German (Slobin et al., 2010). The verb legen ‘lay’ is used when the object is placed horizontally, whereas stellen ‘make stand’ is used when the object is placed 1

2

Studies in Second L anguage Acquisit ion of Chinese

vertically. In addition to semantic variation in the category of placement verbs, cross-linguistic variation in path expressions is also identified. Choi and Bowerman (1991) observed that English and Korean children used and understood path expressions according to the categories of each language. For example, Korean-speaking children, like Korean adults, distinguished motion between tight-fit (kkita, for ‘putting a ring on a finger’ or ‘putting a book in a case’) and loose-fit events (nehta, for ‘putting an apple in a bowl’ or ‘putting a book in a bag’). By contrast, English children distinguished the spatial distinctions between in (putting objects into a container) and on (putting objects into contact with a flat surface), regardless of the differences between tightfit and loose-fit. Such language-specific differences have been observed from as early as 16–20 months. For L2 researchers, the research question of interest is to what degree the predispositions for spatial organization or event construal established during first language (L1) development play a role during adult second language (L2) acquisition. The study reported in this chapter is intended to contribute to the understanding of how L1 conceptualization of motion may exert influence on L2 use and acquisition of path expressions. It compares the behavior of 80 English-speaking L2 learners of Chinese to that of two baseline groups of 40 native speakers (NSs) of Chinese and 40 of English. The theoretical background and a review of the relevant theoretical and empirical research will be presented first. After that, the differences between Chinese and English in terms of path encodings and potential sources of challenges that are associated with learning L2 Chinese path expressions will be discussed. Next, the design and results of the study will be presented and explained. Based on the findings, the various contributing factors that influence the use of L2 Chinese path expressions by adult L1 English speakers will be discussed.

Background Languages differ widely in the ways in which they describe the path information of a motion event. The path in this case is, by definition, the course or trajectory of the motion. Slobin (2004) specifies that path components comprise: (a) direction of the motion, such as into or up; (b) deixis or direction with regard to the viewpoint of the speaker; or (c) contour, such as zigzag or curved. According to Talmy’s (1991, 2000) binary typology of motion events, languages can be divided into either satellite-framed languages (S-languages) or verb-framed languages (V-languages), depending on how the path is encoded for events involving movement. S-languages, such as English and Chinese, characteristically conflate motion and manner/cause in the main verb and encode the path of movement in a satellite attached to the main verb, whereas V-languages like Spanish and other Romance languages typically encode the path in the main

Inf luence of L1 Think ing for Speak ing on Use of an L2

3

verb. For instance, when describing the same motion scene of an owl flying out of a hole in a tree, speakers of typologically different languages depict the path in distinct manners. Consider the following motion sentences describing an owl’s movement: (1) Satellite-framed language (S-language) (a) English An owl flew Figure Motion + Manner (b) Chinese 猫头鹰 飞 Ma ¯otóuyı¯ng fe¯i owl fly Figure Motion + Manner

out. Path 出来 chu ¯-lái out-hither Path

了。 le PFV

(2) Verb-framed language (V-language) Spanish Salió un buho. exited an owl Motion + Path Figure As shown in Examples 1a and 1b, in S-languages, the path component is encoded in a satellite, namely, out in English and chu ¯-lái in Chinese, while the manner and motion are conveyed via the main verb. The English path satellite out and the Chinese directional complement (DC) chu-lai ‘out-hither’ each relate to the main verb as ‘a dependent to a head’, according to Talmy (2000: 102), and thus are typical examples of path satellites. By contrast, Spanish, a V-language, conflates path together with motion in the main verb salió ‘exit’. Talmy’s distinction not only captures the systematic differences between languages of the two types in terms of how path information is encoded, but also provides implications for studies of L1 and L2 learning. Extending on Talmy’s insights, Slobin (1996a, 1996b, 1998, 2004, 2009) and his colleagues have conducted cross-linguistic and developmental studies to explore the impact of motion event typology on language use. Analyses of the narrations produced by speakers of S-languages (English and German) and V-languages (Hebrew, Spanish and Turkish) showed that speakers of S-languages demonstrated a higher degree of elaboration in their descriptions of path of motion than did speakers of V-languages. Specifically, speakers of S-languages tended to specify details concerning the moving figures along the paths by attaching several path satellites to a single main verb. However, it would require separate path verbs in V-framed languages to express such complex path information. It was observed that speakers of V-languages tended to analyze the paths into fewer components, and they preferred to describe the static scenes in which the movement took place and to leave the

4

Studies in Second L anguage Acquisit ion of Chinese

details of the paths to be inferred. These distinctive typological preferences could be observed from as early as three years of age. Slobin (1987, 1996a, 2003, 2009) has proposed a thinking-for-speaking (TFS) hypothesis – a particular kind of language–cognition interface phenomenon – to account for the cross-linguistic differences in the way speakers of different languages describe paths of the same motion events. TFS is ‘a special form of thought that is mobilized for communication’ (Slobin, 1996a: 76). He suggests that, when acquiring their L1, children of individual languages gradually become attuned to different aspects of motion events, so as to develop language-particular patterns of TFS. As illustrated in the owl examples, in order to succinctly describe the owl’s movement in the evanescent time frame in which utterances are constructed, speakers of S-languages attend to the manner of the motion as well as direction. Moreover, Chinese NSs focus not only on the directionality of the path, but also on the deictic relationship between the moving figure and the interlocutor/observers. In contrast, speakers of V-languages do not attend as much to either the manner or the deictic relationship inherent in such a motion event, and such event components tend not to be linguistically encoded in their languages. That is, the conceptualization of path information is filtered through languagespecific semantic structures and through habits of TFS developed in L1 acquisition. Such TFS patterns vary considerably from language to language and can be observed in all forms of linguistic production and reception (i.e. thinking for speaking/writing and listening/reading for understanding), as well as a range of mental processes, such as translation and mental imagery (Slobin, 2000, 2003; see also Oh, 2003). A proposal of interest to L2 researchers is that Slobin (1993, 1996a) claims that the TFS patterns developed in L1 acquisition might be resistant to restructuring in L2 acquisition, especially for adult L2 learners.

Rethinking for speaking in L2 acquisition In light of Talmy’s motion event typology and Slobin’s TFS framework, L2 research has begun to explore the effects of L1 TFS on L2 acquisition (see Han & Cadierno, 2010). Robinson and Ellis (2008) have characterized the development of L2 competence in expressing motion events as one of ‘rethinking for speaking’. Building on this, Odlin (2008) and Jarvis and Pavlenko (2008) define the cross-linguistic influence that involves linguistic relativistic effects as a kind of ‘conceptual transfer’. As Jarvis and Pavlenko (2008) note, the lexicalized and grammaticalized concepts or preferred frames of an L1 can sensitize the speakers to specific distinctions and facilitate categorization and comprehension along the lines of their L1 habitual modes of thought. Consequently, when learning a new language, they would need to restructure their existing conceptual representations and learn to conceptualize the world in a different way when speaking and

Inf luence of L1 Think ing for Speak ing on Use of an L2

5

comprehending, such as when attending to new ways of categorizing objects, events and phenomena or making new attributions to familiar objects. A few L2 studies examining L2 learners’ production in motion events have begun to explore whether L2 learners are able to adjust to new TFS patterns in their descriptions of path expressions when their L2 is typologically different from their L1. Cadierno (2004) investigated how speakers of L1 Danish, an S-language, come to express motion events in L2 Spanish, a V-language. Sixteen Danish learners (eight at intermediate level and eight at advanced level) and 16 NSs of Spanish participated in this study. The results showed that, compared to the written narrations from the Spanish NSs, intermediate L2 Spanish learners exhibited a higher degree of complexity and elaboration of paths of motion in their interlanguage production through the use of redundant and anomalous path particles as well as unnecessary ground adjuncts. This phenomenon has been referred to as ‘satellization’ of L2 Spanish motion constructions (Cadierno, 2004), which is parallel to the learners’ L1 Danish production. This suggests that learners’ L1 TFS patterns influence their acquisition of a typologically different L2 (see also Cadierno & Ruiz, 2006; Navarro & Nicoladis, 2005). On the other hand, while most L2 studies have reported signs of L1 TFS influence on L2 motion expressions, a recent longitudinal study by Stam (2010) reported evidence of a learner’s shift toward new L2 TFS patterns in her use of path expressions. Stam (2010) looked into a Spanish learner’s development of L2 English speech and gesture production over nine years and observed that the learner’s TFS about path positively changed. She was gradually aligned with L2 encoding preferences and was able to consistently express path with a satellite at the end of the study period, although her expressions of manner remained L1-like over the observed period. In sum, despite the increasing number of studies on this topic, the extent to which L1 TFS patterns influence L2 learning of path expressions, and the questions of whether L2 learners can shift toward the new L2 patterns and how fast and thorough such a shift may occur, remain to be determined. So far, L2 research on motion events has mostly been concentrated on comparing L1s and L2s that are typical examples of Talmy’s binary typology. The potential challenges associated with intra-typological variation are largely overlooked (see, however, Cadierno, 2010; Hasko, 2010). Hence, this study offers an intra-typological comparison of Chinese and English path expressions (cf. Beavers et al., 2010; Wu, 2011) and addresses aspects of L2 Chinese learning that have been unexamined heretofore in the areas of applied cognitive linguistics and second language acquisition.

Intra-typological variation in path of motion between Chinese and English Chinese, like English, is considered to be an S-language (Peyraube, 2006; Shen, 2003; Talmy, 1985, 1991, 2000). Chinese uses post-verbal DCs to encode

6

Studies in Second L anguage Acquisit ion of Chinese

the path, a strategy that is somewhat similar to English path satellites (e.g. up, down, out). This closed set of direction indicators is summarized in Table 1.1. A prominent difference in path expressions between Chinese and English is the use of the hither and thither satellites (i.e. 来 lái and 去 qù). The hither/thither path denotes the moving figure’s deictic path or movement from the perspective of the speaker: 来 lái ‘moving toward the speaker’ or 去 qù ‘moving away from the speaker’. The encoding of deictic path is common in the expression of motion events by Chinese speakers. According to Chen (2007: 53), 55% of the motion event descriptions found in his elicited narrations encoded deictic paths. Note that when the motion expression is plain without an object or place noun (e.g. object noun: he took out a book; place noun: he ran to school), it becomes necessary to encode the deictic path as the second DC for the utterance to be grammatical. For instance, to say ‘he walked in’ in Chinese, one has to say Verb (走 zoˇu ‘walk’) + DC1 (进 jìn ‘into’) + DC2 (来 lái ‘hither’). Omitting the deictic DC2 来 lái would generate a non-target-like form (i.e. *Ta ¯ zoˇu jìn le., lit. ‘He-walk-into-PFV’). Hanks (1996, 2009) points out that properly uttering deictic expressions involves several simultaneous dimensions of processing. Interlocutors need to resort to perception, proximity and ongoing interaction in the context of an utterance. For NSs of Chinese, such processing is automatic. The frequent encoding of deictic path predisposes speakers to habitually direct their attention to such dimensions, whereby they can quickly construct deictic expressions. Observing the frequent use of deictic paths in Chinese motion expressions, Slobin (2004) has suggested that deixis seems to be more closely tied to conceptions of path for Chinese speakers. In English, by Table 1.1 DCs in Chinese 来 lái 去 qù 上 shàng 下 xià 起 qıˇ 进 jìn 出 chu¯ 回 huí 过 guò

到 dào 开 ka¯i aThe

‘hither’ (moving toward the speaker) ‘thither’ (moving away from the speaker) ‘up’ (moving upward, profiling the goal region)a ‘down’ (moving downward) ‘up’ (moving upward, profiling the source region)a ‘into’ (moving from outside to inside) ‘out’ (moving from inside to outside) ‘back’ (returning to an original place) ‘across’ (moving from one side to the other) ‘past’ (moving past a point) ‘over’(moving from one point to another) ‘to’ (arriving at a point) ‘apart; away’ (departing from a point)

English path satellite ‘up’ is subdivided into two categories in Chinese: 上 shàng and 起 qıˇ.

Inf luence of L1 Think ing for Speak ing on Use of an L2

7

contrast, the hither/thither perspective is either omitted or implied in the lexical verb forms, such as bring/take, come/go or return/leave. The encoding of hither and thither information (i.e. lái and qù) that appears as the DC2 would require English-speaking learners to adjust themselves to an unfamiliar TFS pattern, in which the deictic path needs to be frequently attended to (Wu, 2011). Moreover, cross-linguistic divergences in spatial semantic categories exist between Chinese and English. For example, the semantic classification represented by the English path satellite up is subdivided into two categories in Chinese: 上 shàng and 起 qıˇ. The Chinese morpheme 上 shàng conceptually highlights the region where the moving figure will be located after moving, while 起 qıˇ highlights the original region where the moving figure came from (Chu, 2004; Liu, 1988; Liu et al., 1983; Yao & Liu, 1997; see also Dai, 2005). Consider the following examples: (3) (a) 请 站 上 来。 Qıˇng zhàn shàng lái. please stand up hither ‘lit. *Please stand up to here. / Please stand over here.’ (b) 他 站 Ta ¯ zhàn he stand ‘He stood up.’

起 qıˇ up

来 lái hither

了。 le. PFV

In Example 3a, using 上 shàng to encode the path will suggest the agent moves upward to a profiled goal. The goal implied in this sentence could be a stage or any place in a higher position. By contrast, using 起 qıˇ to encode the path will profile only the source region of the movement. As in Example 3b, the motion of standing up highlights the vertical lift from a seat, and there is no implied goal for the agent to move to. L1 Chinese speakers are sensitive to the distinction between 上 shàng and 起 qıˇ, because learning the language as an L1 has trained them to be so. An Englishspeaking learner of L2 Chinese, by contrast, will need to reallocate attentional resources and reclassify the related spatial concepts in their L2 TFS in order to promptly describe motion events like those in Examples 3a and 3b. Wu (2011) has observed that learners often overlooked such distinctions between the two DCs 上 shàng and 起 qıˇ. The high percentage of misuse of 上 shàng and 起 qıˇ was especially pronounced at the intermediate level of proficiency, suggesting that the process of re-categorization required for learning the specifications between 上 shàng and 起 qıˇ posed considerable difficulty for the learners. Another path satellite showing different spatial categorization between Chinese and English is the DC 过 guò. The spatial notions defined in the

8

Studies in Second L anguage Acquisit ion of Chinese

Chinese DC 过 guò can correspond to three English satellites, including across, past and over. Consider the following examples: (4) (a) 他 走 过 马路。 Ta ¯ zoˇu guò ma ˇlù. he walk across road ‘He walks across the road.’ (b) 他 走 过 Ta ¯ zoˇu guò he walk past ‘He walks past me.’

我 woˇ my

身边。 she¯n-bia¯n. body-side

(c) 请 把 椅子 搬 过 Qıˇng baˇ yıˇzi ba ¯n guò Please BA chair move over ‘Please move the chair over here.’

来。 lái. hither

The original basic spatial schema denoted by 过 guò is illustrated in Example 4a. 过 guò describing the direction of moving from one side of the road to another is equivalent to the English satellite across in this sentence. In Example 4b, the spatial notion defined by 过 guò has transformed to include the direction of moving ‘past’ a point, me, on the trajectory. Additionally, the DC 过 guò can also serve as a semantically general expression which can be used to describe a figure’s movement from one point to another. In Example 4c, the speaker made the request to someone to move the chair over. 过 guò in this sense is roughly equal to over. The different spatial senses of 过 guò, structured by interrelated schemas, represent a complex abstraction, which Chinese NSs learn how to command through experiencing the different form-meaning mappings in a variety of communicative contexts in the course of their L1 acquisition. Regarding the learning of path expressions where divergences in spatial categorization exist between L1 and L2, Wu (2011) has shown that, when a single spatial category in learners’ L1 is subdivided in the L2, as is the case with English up versus Chinese 上 shàng and 起 qıˇ, English learners of L2 Chinese at the intermediate level of proficiency still had difficulty distinguishing between the new sub-categories in the L2. The use of 过 guò, by contrast, represents another case in which several spatial categories in the learners’ L1 English are collapsed into a single path expression in the L2. It would be interesting to explore whether English-speaking learners of L2 Chinese will be more apprehensive or at ease when using all the sub-categories of 过 guò in their L2. Aside from the aforementioned differences in the ways Chinese and English divide up space, the other DCs such as, 下 xià ‘down’ or 出 chu ¯ ‘out’, respectively, have only one corresponding English path satellite and

Inf luence of L1 Think ing for Speak ing on Use of an L2

9

generally denote similar spatial semantics to their English counterpart. It is likely that learning to use these DCs would be less challenging for Englishspeaking L2 learners. Conversely, the re-categorization and reallocation of attentional resources needed for acquiring Chinese DCs, such as 来 lái, 去 qù, 上 shàng, 起 qıˇ and 过 guò, could be sources of challenge in developing L2 ways of TFS. As suggested by Slobin (1996a), L2 learners may require a relatively long period of time to restructure their L1 manners of TFS in order to be able to express motion events fluently in the L2, and exactly how long such transformation could take may vary considerably depending on the length and amount of L2 exposure.

Differences between heritage and foreign language learning As noted by Bowerman and Choi (2001: 497), ‘children construct spatial semantic categories over time on the basis of the way they hear words used in the input’. In the process of learning to use path expressions in an L2, the degree of L2 exposure and frequency of L2 use inside and outside the classroom may play a role in how well they are able to restructure their existing spatial categories and their ability to learn to conceptualize spatial relations in L2-specific ways. Studies have shown that heritage language learners1 (HLLs), as a result of their prior and prolonged exposure to the language and culture, exhibit different learning profiles from foreign language learners (FLLs) (see Au & Romo, 1997; E.J. Kim, 2003; H.-S. Kim, 2001; Kondo-Brown, 2005; Lee et al., 2005; McGinnis, 1996; Wu, 2011). For instance, HLLs were found to be more confident in their listening ability, and to have more heterogeneous competencies in speaking, reading and writing skills than their FLL counterparts (e.g. H.-S. Kim, 2001; McGinnis, 1996). In terms of L2 morphology and syntax, differences between HLLs and FLLs have also been identified. For example, Au and Romo (1997) observed that first-year university Korean HLLs outperformed FLLs on grammaticality judgments involving the use of subject and object case markers. Hence, in accordance with cognitive linguistic and usage-based accounts of L2 acquisition (Robinson & Ellis, 2008), the present study adopted a separate investigation of HL and FL learning backgrounds.

The Study This cross-sectional study investigated L2 Chinese learners’ interlanguage development for describing path of motion. It aims to explore learners’ L2 thinking-for-writing development of their use of DCs, so as to better understand the rethinking-for-speaking difficulties in learning path expressions

10

Studies in Second L anguage Acquisit ion of Chinese

where intra-typological variations exist between the L1 and the L2. Specifically, the following research questions were addressed: (1) With the L1–L2 differences and similarities in spatial semantic categories for path expressions, how does the learners’ ability to use DCs relate to their overall L2 Chinese proficiency? (2) What differences are there between L2 learners’ and NSs’ use of DCs? (3) Do HLLs and FLLs exhibit different learning profiles?

Participants A total of 160 subjects participated in this cross-sectional study. Eighty were L2 Chinese learners, 40 were NSs of Chinese and 40 were NSs of English. All participants were recruited at a large public university in the United States. NS baseline data was collected to serve as a basis for comparison with the data produced by L2 learners and to test the effectiveness of the research design for eliciting the target structures. The learner participants were recruited in a stratified random fashion along with two variables of interest, namely, learner background (HLLs versus FLLs) and proficiency level (Low versus High). Forty HLLs and 40 FLLs were recruited to investigate whether these two groups of learners differ in terms of their acquisition and use of Chinese DCs. All the learner participants identified English as their current strongest language. The criteria for classification as an HLL were: (1) that the learner identified his or her strongest language before the age of five as Mandarin Chinese or another Chinese dialect; and (2) that he or she had one or both parents who speak Mandarin Chinese or another Chinese dialect as their native or dominant language, and that he or she reported exposure to the language at home. Based on information gathered from the background information questionnaire (see Appendix A) administered to the 40 HLLs, 75% (n = 30) of them met the first criterion, and 100% of them (n = 40) met the second. Out of the 30 HLLs who identified their strongest language before five as Mandarin Chinese or another Chinese dialect, 22 spoke Cantonese, five spoke Mandarin, one spoke Min, one spoke Amoy, and one spoke Teochew. Regarding the second variable of interest, the 80 learners were drawn from two proficiency groups. Forty of the learners were sampled from lowerdivision college courses in Chinese, which included various sections of 200level Chinese language courses. The other 40 were from upper-division courses, comprised of mostly 300- and 400-level Chinese language courses as well as graduate courses related to Chinese Studies. Moreover, to ensure that all learners participating in this study had learned the target motion expressions, only learners who had studied Chinese at college for at least three semesters were recruited.

Inf luence of L1 Think ing for Speak ing on Use of an L2

11

Proficiency was then further determined by an elicited imitation test (EIT) (cf. Erlam, 2006; Jessop et al., 2007; Vinther, 2002). The EIT utilized in this study was a Mandarin version of the EIT designed by Ortega et al. (2002). Ortega et al. (2002) tested parallel EITs across four languages, including English, German, Spanish and Japanese, and found that the EITs successfully offered a good indication of learners’ global L2 proficiency levels. The Mandarin EIT (see Appendix B) developed by Zhou and Wu (2009) closely followed the test specifications set by Ortega et al. (2002), and its validity as a proficiency measurement tool was supported in Wu and Ortega (2013). The test comprised 30 Mandarin sentences, ranging from seven to 19 syllables. The 30 items together contained a wide range of vocabulary and grammatical structures, and thus presented varying degrees of challenge. During the 10-minute test, participants were instructed to repeat each sentence they heard as well as possible. To avoid rote repetition, an interstimulus interval of 2.5 seconds was inserted between the end of each sentence and the start of each repetition. A five-point scoring rubric (developed by Ortega et al., 2002) was used to evaluate learners’ performance on each repetition. The median score of the 80 L2 learners’ EIT scores, 50, was used as the cutoff point for discrimination of the high and low proficiency groups in both HLLs and FLLs. Learner participants who scored below 50 on the Mandarin EIT were classified as having low proficiency, and those who scored above 50 were considered as having high proficiency. As shown in Table 1.2, four groups were generated, using background and proficiency as grouping variables. Results of an independent-sample t-test showed that low proficiency HLLs (M = 36.21, SD = 7.04) and FLLs (M = 35.04, SD = 11.46) did not differ from each other in terms of their EIT scores, t (35.845) = 0.391, p = 0.698, Cohen’d = 0.12. There was also no significant difference in scores for high proficiency HLLs (M = 76.23, SD = 16.27) versus FLLs (M = 72.00, SD = 18.40), t (40) = 0.779, p = 0.441, Cohen’d = 0.24. Hence, low proficiency HLLs and FLLs were treated as separate groups within a single low proficiency level, as were high proficiency HLLs and FLLs. Differences Table 1.2 Learner participants by background and by EIT score Groups Low HLLs FLLs High HLLs FLLs

N

Minimum

Maximum

Mean

SD

14 24

22 15

47 49

36.21 35.04

7.04 11.46

26 16

51 52

112 115

76.23 72.00

16.27 18.40

Notes: The highest possible individual score was 120, based on 30 items polytomously scored from 0 to 4. The EIT score of 50 was the cutoff point for the low and high groups.

12

Studies in Second L anguage Acquisit ion of Chinese

drawn from comparisons between HLLs and FLLs at the same proficiency level will then be attributed to the learners’ language background.

Procedures Three instruments were developed and administered for this study (see Table 1.3 below). The learner participants first completed a picture-cued written task and then filled out a background information questionnaire, designed to measure their prior contact with Chinese. Finally, they completed an EIT as a general proficiency measure for Chinese. On average, the learner participants spent between 22 and 32 minutes completing the procedure. The NS participants completed only the main task (i.e. the picture-cued written task). The picture-cued written task (see Appendix C) was comprised of 16 motion event scenarios, each of which required participants to use a variety of different DCs to describe the paths of the movements shown in consecutive pictures. Three of the actual test items (Items 5 and 7 of Part I, Item 4 of Part II) are provided in Appendix C. This instrument was originally developed for use by Wu (2011), with four items replaced by newly designed ones for better elicitation of the target DCs. Each scenario featured two characters, ‘person A’ and ‘person B’, who were the same for all scenarios. Participants were instructed to describe B’s movements or to make a request to B from the perspective of A. This design was meant to clearly present the deictic viewpoint for each context. Additionally, in order to elicit different types of DC constructions, four items designated only verbs as cues, five items gave both designated verbs and Place NPs, and seven items included both designated verbs and Object NPs. Because the focus of this study was the learners’ development of L2 TFS about path rather than their ability to write Chinese characters, they were allowed to use either Chinese characters or Pinyin2 to complete the sentences. For the instrument given to Chinese NSs, all of the English instructions were translated into Chinese. For the instrument given to the English NSs, the Chinese Place/Object prompts were removed from the pictures, and only their English translations were presented. Note that, in order to elicit natural English motion sentences, the designated Chinese verb for each sentence was removed. This was because the direct English translation for these Chinese verbs might not fit the motion scenarios as well as the original Chinese verbs did. Table 1.3 Tasks and procedure Step

Instrument

Time for completion

(1) (2) (3)

Picture-cued written task (see Appendix C, Parts I & II) Background information questionnaire (see Appendix A) Elicited imitation test (see Appendix B)

12 to 20 minutes 1 to 2 minutes 9 to 10 minutes

Notes: Total time required is 22–32 minutes.

Inf luence of L1 Think ing for Speak ing on Use of an L2

13

Results of the data produced by the Chinese NSs were used as a basis for scoring the L2 learners’ production. Each of the items produced by the L2 learners was dichotomously scored with 0 or 1. Zero was given when the DCs supplied were non-target-like (e.g. the DCs supplied did not match the movement delineated in the picture, or no DC was supplied). A score of one was given when the DCs supplied were target-like. Two raters worked independently to score all responses. The initial scorings showed 96% agreement, and all discrepancies were resolved by discussion.

Results The results of the analyses of the data produced by Chinese and English NSs will be presented first to address areas of intra-typological variations between the two NS baseline groups’ encodings of path. After the L1 data is presented, the results of L2 production will be described. The differences in performance among the different learner groups will be described and compared to the data from the Chinese and English NSs.

Encodings of path in L1 production Responses for the 16 items of the picture-cued written task by the two NS groups were tallied and summarized in Table 1.4. The responses were listed hierarchically from most frequent to least frequent. The first column presents the English NSs’ descriptions of the 16 motion scenes. Their encodings of path satellites are presented in the second column, and those used by the Chinese NSs are shown in the third column. The final column, Place/ Object cue, lists the cue given for each item. Most of the responses collected from the Chinese and English NSs consistently showed use of path satellites when describing the path of different motion scenes, confirming that Chinese and English are S-languages. Note that, because there was no designated English verb for each motion scenario in the instrument given to the English NSs, they used path verbs such as leave or return to denote the path more frequently than did the Chinese NSs. There were only responses to two items including more than two kinds of path encodings (i.e. English NSs’ responses to Item 7 of Part I and Chinese NSs’ responses to Item 6 of Part II; see Table 1.4 for details). Taking Item 2 of Part I as an example, the motion scenario occurring between persons A and B describes, ‘A noticed his neighbor B passing by. B was finishing jogging, and his house was nearby’, and participants were asked to describe B’s movement from A’s perspective. Three answers were given: ‘B jogged home’, ‘B jogged toward the house’ and ‘B ran to the house’. These different answers suggest the different perspectives in selecting the manner verb and how they interpreted the meaning of the destination. For these two items, only

14

Studies in Second L anguage Acquisit ion of Chinese

Table 1.4 Summary of path expressions produced by NSs Part I. Motion scene

Eng path

Chn path

Place/object cue

(1) walk down the stairs

down

xià ‘down’

stairs

(2) leave the classroom walk out of the classroom (3) jump up (4) move out of the dormitory leave the dormitory (5) stand up (see Appendix C, Part I, Example 1) (6) run into the library (7) jog home jog toward the house run to the house (see Appendix C, Part I Example 2) (8) bring the letter

out of

chu¯ ‘out’

classroom

up out of

qıˇ-lái ‘up-hither’ chu¯ ‘out’

none dormitory

up

qıˇ-lái ‘up-hither’

none

into toward to

jìn ‘into’ huí ‘back’ huí-qù ‘back-thither’

library none

lái ‘hither’

letter

Part II. Motion scene

Eng path

Chn path

Place/object cue

(1) put the books in the box

in

book

(2) take out the homework (3) bring a chair put a chair on the stage (4) bring the pizza upstairs bring the pizza up (see Appendix C, Part II Example 1) (5) put the chair back to the original place return the chair (6) park the car in the garage move to the left

out on

homework chair

up

jìn ‘into’ zài ‘Prep. in’ chu¯-lái ‘out-hither’ shàng-lái ‘up-hither’ guò-lái ‘over-hither’ shàng-lái ‘up-hither’

back to

huí-qù ‘back-thither’

chair

in

jìn ‘into’ dào ‘to’ jìn-qù ‘into-thither’ guò-lái ‘over-hither’ dào ‘to’ dào-qù ‘to-thither’

none

to (7) bring another chair (8) deliver books to the classroom

to

pizza

chair classroom

common responses that had more than 10 incidences were included as the baseline data for comparison with the data produced by L2 learners. Moreover, there was one item in which path was not exclusively described by means of a DC in the Chinese NSs’ production. In Item 1 of Part II, 70% of the Chinese NSs used the DC 进 jìn ‘into’ to express the motion request ‘put the books in the box’, whereas 30% of the Chinese NSs (12 out of 40) used the preposition 在 zài, followed by the goal NP and a

Inf luence of L1 Think ing for Speak ing on Use of an L2

15

localizer 箱子里 xia ¯ngzi-lıˇ ‘box-inside’, in which the localizer 里 lıˇ helps convey where the object NP books should be moved to. Given that the preposition 在 zài represents another syntactic category distinct from DCs, this item was therefore excluded from the analyses. Out of the remaining 600 responses (k = 15) produced by the 40 Chinese NSs, 95% contained the target DC constructions, suggesting the task design successfully elicited the target structures. Comparing the encoding of the path component between Chinese and English, one can find that the Chinese NSs often took a deictic viewpoint when describing motion events, especially when there was a change of location for an object NP. For the six items that included an Object NP cue (i.e. Item 8 in Part I & Items 2, 3, 4, 5 & 7 in Part II), 93% of the Chinese NSs’ responses encoded a deictic path via 来 lái ‘hither’ or 去 qù ‘thither’ to specify whether the object was to be moved toward the speaker or away from the speaker. English NSs’ responses to these six items, by contrast, did not show such a habitual TFS pattern of expressing deictic orientation relations. The hither/thither perspective, however, was sometimes implied in English verb roots such as bring and return (e.g. Item 8 of Part I & Item 5 of Part II). It appears that the two languages have their own linguistic strategies to express deictic orientation, but Chinese speakers more frequently do so by means of supplying the DCs 来 lái ‘hither’ and 去 qù ‘thither’. The distinction of upward direction by 上 shàng and 起 qıˇ was another pronounced difference between Chinese and English NSs’ path descriptions. Chinese NSs used 上 shàng to describe upward movement with a goal, as for Item 4 of Part II (see Appendix C), for which participants requested person B to deliver a pizza to a residence on the fifth floor. Because this upward movement has a profiled goal, Chinese NSs chose 上 shàng to describe this event (i.e. 请把披萨送上来。 Qıˇng baˇ pı¯sà sòng shàng lái., lit. ‘please-baˇpizza-send-up-hither’). On the other hand, for Item 3 of Part I, they were required to describe person B’s movement of jumping up in the air in a basketball game (i.e. B 跳起来了 B tiào qıˇ lái le., lit. ‘B-jump-up-PFV’), and for Item 5 of Part I (see Appendix C), they were to describe person B’s movement of standing up to answer a question (i.e. B 站起来了 B zhàn qıˇ lái le., lit. ‘B-stand-up-PFV’). These two events of upward motion do not have a profiled goal, and all Chinese NSs used 起 qıˇ to encode the paths, profiling the source of the movement in question. English NSs’ responses to these three items, by contrast, did not involve such distinctions. The particle up was used in all of these three motion scenarios (i.e. Item 4 of Part II: Please bring up the pizza.; Item 3 of Part I: B jumped up.; and Item 5 of Part I: B stood up.) This suggests that, while Chinese NSs are sensitive to the distinction of spatial categories by the DCs 上 shàng and 起 qıˇ when describing upward movement, such spatial distinctions do not exist for English NSs. Another example showing differences in path conceptualization between Chinese and English is the use of 过 guò ‘over’. For Item 7 of Part II, the majority of the English NSs made the request ‘Please bring another chair’ in

16

Studies in Second L anguage Acquisit ion of Chinese

response to the scenario, and Chinese NSs used guò-lái to describe the path, ‘请把椅子搬过来 Qıˇng ba ˇ yıˇzi ba ¯n guò lái., lit. “Please-move-chair-overhither”. Please move the chair over here.’ When perceiving the same motion scenario, English NSs only encoded the sense of moving toward in the verb bring, whereas most Chinese NSs had explicitly denoted the deictic path and used the DC 过 guò to describe the path. By contrast, only one English NS used over to describe the path in this item. This shows that speakers of different languages may encode path information in different manners, regardless of the availability of similar linguistic devices.

Encodings of path in L2 production Use of path expressions and global proficiency Turning to the overall performance of L2 learners in their use of path expressions, the results for the picture-cued written task are given in Table 1.5 and Figure 1.1. A two-way ANOVA procedure was adopted to analyze the trustworthiness of the overall effects of participant background (HLLs versus FLLs) and proficiency level (Low versus High). The data yielded a significant main effect for proficiency level, F(1, 76) = 42.459, p < 0.001, partial η2 = 0.358. That is, the L2 Chinese learners’ ability to use DCs to describe paths of the different motion events on this task was positively correlated with their global Chinese proficiency, as learners in the high proficiency group performed significantly better than learners in the low proficiency group for both HLLs and FLLs. The effect for participant background was not significant, F(1, 76) = 2.05, p > 0.05, partial η2 = 0.026. That is, there was no significant difference between HLLs and FLLs in terms of their overall DC scores that measured learners’ ability to choose appropriate DCs to encode path of motion. The interaction between proficiency levels and participants’ background was also not significant, F(1, 76) = 0.093, p > 0.05, partial η2 = 0.001. The relationship between learners’ ability to use DCs and their L2 Chinese proficiency (measured by EI scores) was further investigated using Table 1.5 Scores of DC constructions in picture-cued written task Groups Low HLLs FLLs High HLLs FLLs

N

Minimum

Maximum

Mean

SD

14 24

1 0

8 10

4.71 4.00

2.09 2.57

26 16

1 3

13 13

9.04 7.94

2.93 3.13

Notes: Total possible individual score was 15, based on 15 items scored dichotomously as 0 or 1.

Inf luence of L1 Think ing for Speak ing on Use of an L2

17

Figure 1.1 Effect of background by level

the Pearson correlation coefficient. This measure suggested a significant positive correlation between these two variables, r = 0.705, n = 80, r2 = 0.497, p < 0.001, further confirming that more proficient L2 Chinese learners generally demonstrate a better command of the path expressions.

Adjustment to L2-specific ways of path expression Comparison of the path expressions used by the 40 English NSs and 40 Chinese NSs has shown that speakers of the two languages conceptualize directionality for movement differently. This section will focus on learners’ ability to adjust to L2-specific ways of path expression. With respect to the use of deictic paths, learners’ responses to the six object-moving motion events are summarized in Table 1.6. Out of the 240 responses produced by the 40 Chinese NSs, 93% encoded deictic paths, suggesting that attention to deixis is closely tied to Chinese NSs’ TFS when describing directionality for displacement of an object. In comparison, the low proficiency L2 learners did so only 24% of the time for HLLs and 19% for FLLs. This suggests that, while some learners began to incorporate the use of deictic encodings when describing movements, they did not appear to have acquired the L2 habitual TFS pattern which gives due attention to deictic paths. However, there was a pronounced positive difference between learners at the two proficiency levels. High proficiency HLLs encoded deictic paths 60% of the time, and FLLs did so 46% of the time. Note that, although the high proficiency learners’ use of deictic paths had gradually aligned with that of the Chinese NSs, their rate of verbalization of deictic path remained much lower than that of the Chinese NSs, who averaged 93%. The relatively lower rates suggested that learners’ L1 TFS still exerted an influence on their

18

Studies in Second L anguage Acquisit ion of Chinese

Table 1.6 Appropriate use of the deictic paths in object-moving motion events Group

NSs Low HLLs FLLs High HLLs FLLs

N

Part I Item (8) lái

Part II Item (2) lái

Part II Item (3) lái

Part II Item (4) lái

Part II Item (5) qù

Part II Item (7) lái

Mean %

rawa

%b

raw

%

raw

%

raw

%

raw

%

raw

%

40

40

100

40

100

32

80

37

93

34

85

40

100

93

14 24

3 5

21 21

5 5

36 21

2 8

14 33

5 4

36 17

3 2

21 8

2 3

14 13

24 19

26 16

14 8

54 50

18 8

69 50

15 8

58 50

19 7

73 44

14 7

54 44

14 6

54 38

60 46

Notes: aThe column marked raw denotes raw frequency counts of appropriate use of lái/qù for each item. bPercentage (%) indicates the rate of appropriate use of lái/qù.

L2 TFS. The challenge associated with encoding the deictic DCs was not a linguistic issue, but rather a cognitive one as it involves how readily they could learn to perceive and describe motion events in a L2-like manner. Moreover, echoing Wu (2011), the results show that HLLs had demonstrated a better command than FLLs at both levels in terms of their ability to incorporate the hither/thither perspective during L2 TFS. Turning to the use of DCs for upward motions, as indicated in Table 1.7, Chinese NSs unanimously used 起 qıˇ in Items 3 and 5 of Part I to describe upward movement profiling the source, and predominantly used 上 shàng in Item 4 of Part II to describe movement profiling the goal at the rate of 88%. Low proficiency HLLs on average had 21% appropriate encoding of the upward DCs, and FLLs had 17%. Again, high proficiency learners’ performance showed a positive difference for both HLLs (56%) and FLLs (52%). HLLs slightly outperformed FLLs across both levels. Further examination of the responses where DCs were used inappropriately indicated that misuse of 上 shàng for 起 qıˇ accounted for the majority of the non-target-like production among all groups (e.g. target answer: 站起來 zhàn-qıˇ-laí ‘stand-up-hither’; non-target-like forms: 站上來 zhàn shàng-laí ‘stand-up-hither’ or 站上 zhàn shàng ‘stand-up’). Roughly 50% of the non-target-like production by low proficiency HLLs was due to misuse of上 shàng for 起 qıˇ (49% for FLLs, and 55% and 71%, respectively, for high proficiency HLLs and FLLs). The findings are parallel with Wu (2011) in that learners tended to choose 上 shàng to describe upward movements that should be specified by 起 qıˇ, potentially because 上 shàng has a wider range of use in encoding upward motions than 起 qıˇ in the language. These results suggest that, while advanced learners were more capable of supplying context-appropriate upward DCs, many of them were still challenged by the subtle distinctions between 上 shàng and

Inf luence of L1 Think ing for Speak ing on Use of an L2

19

Table 1.7 Appropriate use of the DCs for upward directions (上 shàng and 起 qıˇ ) Group

NSs Low HLLs FLLs High HLLs FLLs

N

Mean %

Part II Item (4) shàng

Part I Item (5) qıˇ

Part I Item (3) qıˇ rawa

%b

raw

%

raw

%

40

40

100

40

100

35

88

96

14 24

0 2

0 8

4 6

29 25

5 4

36 17

21 17

26 16

6 7

23 44

18 10

69 63

20 8

77 50

56 52

Notes: aThe column marked raw denotes raw frequency counts of appropriate use of shàng /qıˇ. bPercentage (%) indicates the rate of appropriate use of shàng/qı ˇ.

起 qıˇ. This holds true regardless of the learners’ language background or level of L2 proficiency. The fact that L2 learners continued to have difficulty telling when to use 上 shàng or 起 qıˇ for different upward-focused movements suggests that lack of correspondent spatial categories in their L1 TFS makes it more challenging for learners to comply with the new L2 TFS patterns. Finally, Table 1.8 presents the results for learners’ use of the DCs 过 guò and 下 xià. The path satellite 过 guò can denote the different spatial notions of Table 1.8 Appropriate use of the DCs 过 guò and 下 xià Group

NSs Low HLLs FLLs High HLLs FLLs

N

Part I Item (1) xià

Part II Item (7) guò rawa

%b

raw

%

40

33

83

37

93

14 24

1 0

7 0

12 17

86 71

26 16

9 3

25 19

24 15

92 94

Notes: aThe column marked raw denotes raw frequency counts of appropriate use of guò and xi. bPercentage (%) indicates the rate of appropriate use of guò and xià.

20

Studies in Second L anguage Acquisit ion of Chinese

‘moving past a point’, ‘moving across from one side to the other’ and ‘moving over to a point’. While Chinese NSs used 过 guò 83% of the time for Item 7 of Part I, the high proficiency HLLs made use of it only 25% of the time, which was the highest among all groups. Overall, L2 learners’ use of 过 guò was scarce. Instead, most learners used 搬来 ba ¯n lái ‘move-hither’ to encode the path in this item. The Chinese verb 搬 ba ¯n ‘move’ does not imply movement directed toward the speaker. By adding the deictic path 来 lái, the L2 learners created semantic content that was very similar to the English verb bring, which was adopted by the majority of the English NSs in their responses to this item. Thus, the L2 learners were still resorting to their L1 TFS while describing this motion scene. As noted earlier, the DC 过 guò encompasses different spatial notions defined by three English satellites, across, past and over. It is possible that the complex spatial abstraction conveyed by 过 guò presents a case in which the form–meaning relationships are less transparent than is the case for the other DCs. Faced with unfamiliar L2-specific spatial categorization, which does not exist in their L1 TFS about path, learning to use 过 guò thus poses more of a challenge for learners. Since many Chinese textbooks often give only the translation of ‘past; passing a point’ when introducing 过 guò, the complex abstraction denoted by 过 guò, as elicited in the present study, is likely to be obscure for learners, especially for FLLs, whose L2 exposure was limited to the classroom and who thus may not have experienced other meanings of 过 guò in natural contexts. Consequently, the learners were less ready to use 过 guò and therefore produced more non-target-like path encodings. Compared to 过 guò, the DC 下 xià does not encompass complex relations that correspond to more than one English satellite. The DC 下 xià is equivalent to the English satellite down. For Item 1 of Part I, the majority of the English NSs described the scenario as ‘B walked down the stairs’, and 90% of the English NSs used the particle down in their responses. The Chinese NSs also used xià 93% of the time, and the majority responded with: B zoˇu xià lóutı¯le., ‘B-walk-down-stair-PFV’, a sentence that has one-to-one correspondence with that of the English NSs’ response. Comparing learners’ use of 过 guò with that for 下 xià, one can see that the percentages of appropriate use of 下 xià across all groups were much higher. The high proficiency learners’ use of 下 xià, including both HLLs and FLLs, were very close to that of NSs (93%) at 92% and 94%, respectively. This suggests that because the minimal L1–L2 distance between the satellites 下 xià and down does not require learners to restructure their L1-based concepts in L2 learning, learners use it more accurately at an early stage of L2 acquisition.

Discussion and Conclusion This study investigated how English-speaking learners come to express path expressions in L2 Chinese. The results show that L2 learners’

Inf luence of L1 Think ing for Speak ing on Use of an L2

21

thinking-for-writing development in overall use of DCs was positively related to their global Chinese proficiency. Learners’ use of Chinese path expressions was more aligned with that of Chinese NSs in many ways as they advanced to a higher proficiency level, suggesting that learners’ L2 TFS about path is not static (cf. Cadierno & Ruiz, 2006; Stam, 2010) and can change during the course of L2 acquisition. Moreover, the results suggest that L2 use and acquisition of path expressions is influenced by their L1 TFS patterns. That is, predispositions for spatial organization or event construal established during L1 development play an important role during L2 acquisition of path expressions. As shown in this study, learners tend to use DCs that present minimal L1–L2 distance in spatial categorization at an early stage of L2 acquisition. Learners at the low proficiency level, for example, had already developed a good command and use of the DC 下 xià, which is directly equivalent to the English satellite down. However, they had more difficulty in using the L2-specific satellites that organize spatial relations in a manner different from those in their L1. Many of the advanced L2 learners were still challenged by the subtle distinctions between 上 shàng and 起 qıˇ, the two DCs used for an upward path, or the complex meanings of 过 guò ‘across, past and over’, even when they had reached a high L2 proficiency level. Additionally, the encoding of hither/ thither perspective, a characteristic of the L2 TFS system, requires learners to habitually attend to perception, proximity and ongoing interaction in context, which was found to be more challenging for learners. These findings show that, in cases where there is a greater degree of similarity between the TFS patterns about path in the L1 and the L2, the acquisition of L2 TFS is facilitated. In contrast, for L2 path expressions presenting larger differences from those of the L1, learning the L2-specific TFS patterns is more challenging. The effects of L1 TFS on learners’ L2 learning outcomes (cf. Cadierno, 2010; Han, 2010) reveal that the factors which influence the process of L2 learning of path expressions go beyond mere L2 linguistic ability. Although there was no significant difference between HLLs and FLLs in terms of their overall DC scores, it was observed that HLLs were more competent than FLLs at the same proficiency level in their ability to adopt L2-specific ways of path encodings, namely, the incorporation of deictic path and target-like use of the DC 过 guò. The HLLs’ better performance can perhaps be attributed to their greater access to participation in communicative practices that their background may have afforded them. That is, internalization of path expressions relies on embodied experiences (see Gibbs, 2003), where concrete actions or movements are coupled with linguistic input to illustrate how the L2 organizes and articulates spatial relations and how it directs its speakers to attend to dimensions of events or actions that are relevant and readily encodable in the L2 (Wu, 2011). By implication, pedagogical activities that provide abundant input that are presented along with physical motions or requiring learners to actively use the target language in

22

Studies in Second L anguage Acquisit ion of Chinese

meaningful tasks might be helpful in assisting them to adopt the new L2 TFS patterns. From an intra-typological point of view, the results of this study show that development of path expressions in L2 Chinese is influenced by a number of factors, among them L2 proficiency level, L1–L2 differences and similarities in spatial concepts, L2 exposure and, last but not least, L1 TFS. These factors go beyond a simple dichotomy of L1 or L2 thinking, and together they suggest that the influence of L1 TFS, like other kinds of L1 transfer phenomena, are sensitive to probabilistic tendencies. Although the existing literature on L2 acquisition of motion expressions generally points to the potential influence of L1 TFS patterns on L2, the extent to which L1 TFS plays a role in L2 performance seems to vary considerably across different studies. Research findings can be influenced by factors such as the subjects’ proficiency levels, modality of research operations, aspects of motion language studied, and the degree of typological differences and similarities between the target L1 and L2. It is important that variables besides L1 or L2 TFS be comprehensively taken into consideration when interpreting results. Findings from this cross-sectional study generally pointed out that L2 acquisition of path expressions at least partially involves the restructuring of L1-mediated concepts for spatial organization or event construal in order to acquire the subtle distinctions between the two upward DCs, 上 shàng and 起 qıˇ, the complex spatial relationships denoted by the DC 过 guò, as well as to habitually incorporate the deictic DCs. Nevertheless, to substantiate this claim, research adopting non-linguistic tasks gauging changes in the general domains of cognition as a result of L2 learning needs to be conducted. More L2 research looking into the underlying cognitive processes, which may not be obvious from linguistic data alone, is needed so as to better understand how learners come to categorize objects or events along the lines offered by the new language(s).

Appendix A: Background Information Questionnaire Q1. What was your first or strongest language before your age of 5? □ English □ Mandarin Chinese □ Chinese dialect (specify) _____ □ Other (specify) _____

Q2. What is your strongest language now? □ English □ Mandarin Chinese □ Chinese dialect (specify) _____ □ Other (specify) _____

Q3. Check if your parents, grandparents, or anyone else in your immediate/extended family is a native speaker of Mandarin Chinese or a Chinese dialect. □ Mother □ Father □ Maternal grandparent(s) □ Paternal grandparent(s) □ Other (specify) _____

Inf luence of L1 Think ing for Speak ing on Use of an L2

23

Q4. At what age did you start to hear or use Mandarin Chinese? _____ Q5. Mandarin learning inside classroom (1) How long (in years) in total have you studied Mandarin at school? _____ (2) List the following information for any previous Mandarin studies (e.g. college, high school, intermediate/elementary school, Chinese language school, private language institute, private tutor, etc.). Please also include the current study program. School 1: ______ (school name) in ______ (country name) Start year: ______ End year:______ Hours of Mandarin class per week ______ School 2: ______ (school name) in ______ (country name) Start year: ______ End year: ______ Hours of Mandarin class per week ______ School 3: ______ (school name) in ______ (country name) Start year: ______ End year: ______ Hours of Mandarin class per week ______ Q6. Have you visited/lived in a Chinese-speaking country? □ No □ Yes (if YES, see below) (At what age: ______________; For ______________ [length of the stay]; Location: ______________ ) (At what age: ______________; For ______________ [length of the stay]; Location: ______________ ) Q7. How much do you hear or use Chinese outside the classroom? 1: never 2: occasionally 3: sometimes 4: frequently 5: almost always – parents/grandparents speaking Chinese to you – relatives/friends speaking Chinese to you – self-study Chinese – others (specify):

N/A N/A N/A N/A

1 1 1 1

2 2 2 2

3 3 3 3

4 4 4 4

5 5 5 5

Appendix B: Mandarin Elicited Imitation Test Note: This test was originally developed by Ortega et al. (2002) for the measurement of L2 proficiency in cross-linguistic SLA research. The Mandarin items were developed by Ying Zhou and the author (2009) with minimal revisions for each item, when necessary to adjust the length of syllables or to more naturally reflect features of the Mandarin language in translation. Numbers in parentheses represent the total number of syllables included in each item.

24

Studies in Second L anguage Acquisit ion of Chinese

Sample Item 1.

我得去剪头发了。(7) I have to get a haircut. (7)

Item 2.

红色的书在桌子上。(8) The red book is on the table. (8)

Item 28.

可以麻烦您把桌子上的那本书递给我吗？(17) Would you be so kind as to hand me the book which is on the table? (17)

Item 29.

我不知道十点半的火车是不是已经开走了。(18) I don’t know if the 10:30 train has left the station yet. (18)

Item 30.

为什么还是有很多人早上什么东西都不吃呢？(19) Why are there so many people who don't eat anything at all in the morning? (19)

Appendix C: Picture-Cued Written Task Part I Instructions: You will see eight sets of sequential pictures describing different kinds of physical movement. Each set of pictures comes with a brief explanation. When you are responding to the questions, pretend you were A in each situation. Your job is to describe B’s movement from A’s perspective and complete the sentences in Chinese characters or Pinyin. You will start with B as the subject and use the designated verb. If there are location or object nouns specified in the pictures (shown in Chinese), remember to include the nouns in your sentences.

Sample

(1)

Picture A and B were sitting together in the classroom. Later the teacher called B’s name.

Picture Then, A saw B . . . Q: If you were A, what would you say to describe B’s movement? B_______________________________(use 站)。

Inf luence of L1 Think ing for Speak ing on Use of an L2

(2)

25

Picture A noticed his neighbor B passing by. B was finishing jogging and his house was nearby.

Picture Later, A saw B . . . Q: If you were A, what would you say to describe B’s movement? B_____________________________(use 跑)。

Part II Instructions: You will see eight sets of sequential pictures describing different kinds of requests. Each set of pictures comes with a brief explanation. When you are responding to the questions, pretend you were A in each situation. Your job is to describe B’s movement from A’s perspective and complete the sentences in Chinese. You will start with 请 (please . . .) and use the designated verb. If there are location or object nouns specified in the pictures (shown in Chinese), remember to include the nouns in your sentences.

Sample (1)

Picture A lived in an apartment and had just ordered a pizza. When B was at the building door, A was too lazy to get the pizza. Picture So, A told B . . . Q: If you were A, what would you say to ask for B’s help? 请 __________________________ (use 送)。

Notes (1) HLLs can be defined as learners who have acquired their cultural and linguistic competence in a non-dominant language through contact at home with foreignborn parents and/or other family members (Kondo-Brown & Brown, 2008; Valdés, 1995). (2) Pinyin is a Romanization system for transcribing the sound of Chinese characters.

26

Studies in Second L anguage Acquisit ion of Chinese

References Au, T.K. and Romo, L.F. (1997) Does childhood language experience help adult learners? In H.-C. Chen (ed.) The Cognitive Processing of Chinese and Related Asian Languages (pp. 417–441). Hong Kong: Chinese University Press. Beavers, J., Levin, B. and Tham, S.W. (2010) The typology of motion expressions revisited. Journal of Linguistics 46 (2), 331–377. Berman, R.A. and Slobin, D.I. (1994) Relating Events in Narrative: A Crosslinguistic Developmental Study. Hillsdale, NJ: Lawrence Erlbaum. Bowerman, M. (1996) The origins of children’s spatial semantic categories: Cognitive versus linguistic determinants. In J. Gumperz and S. Levinson (eds) Rethinking Linguistic Relativity (pp. 145–176). Cambridge: Cambridge University Press. Bowerman, M. and Choi, S. (2001) Shaping meanings for language: Universal and language-specific in the acquisition of spatial semantic categories. In M. Bowerman and S.C. Levinson (eds) Language Acquisition and Conceptual Development (pp. 475–511). Cambridge: Cambridge University Press. Cadierno, T. (2004) Expressing motion events in a second language: A cognitive typological perspective. In M. Achard and S. Niemeier (eds) Cognitive Linguistics, Second Language Acquisition, and Foreign Language Teaching (pp. 13–49). Berlin: Mouton de Gruyter. Cadierno, T. (2010) Motion in Danish as a second language: Does the learner’s L1 make a difference? In Z.-H. Han and T. Cadierno (eds) Linguistic Relativity in Second Language Acquisition: Thinking for Speaking (pp. 1–33). Clevedon: Multilingual Matters. Cadierno, T. and Ruiz, L. (2006) Motion events in Spanish L2 acquisition. Annual Review of Cognitive Linguistics 4, 183–216. Chen, L. (2007) The Acquisition and Use of Motion Event Expressions in Chinese. Munich: Lincom Europa. Choi, S. and Bowerman, M. (1991) Learning to express motion events in English and Korean: The influence of language-specific lexicalization patterns. Cognition 48, 83–121. Chu, C. (2004) Event conceptualization and grammatical realization: The case of motion in Mandarin Chinese. Unpublished doctoral dissertation, University of Hawai’i at Ma¯noa. Dai, J.-H.E. (2005) Conceptualization and cognitive relativism on result in Mandarin Chinese: A case study of the Mandarin Chinese ba construction using a cognitive and centering approach. Unpublished doctoral dissertation, Louisiana State University. Erlam, R. (2006) Elicited imitation as a measure of L2 implicit knowledge: An empirical validation study. Applied Linguistics 27, 464–491. Gathercole, V.M. (ed.) (2009) Routes to Language: Studies in Honor of Melissa Bowerman. New York: Psychology Press. Gibbs, R.W. (2003) Embodied experience and linguistic meaning. Brain and Language 84 (1), 1–15. Guo, J., Lieven, E., Ervin-Tripp, S., Budwig, N., Özçalişkan, S. and Nakamura, K. (eds) (2009) Crosslinguistic Approaches to the Psychology of Language: Research in the Tradition of Dan Isaac Slobin. Mahwah, NJ: Lawrence Erlbaum. Han, Z.-H. (2010) Grammatical morpheme inadequacy as a function of linguistic relativity: A longitudinal case study. In Z.-H. Han and T. Cadierno (eds) Linguistic Relativity in Second Language Acquisition: Thinking for Speaking (pp. 154–182). Clevedon: Multilingual Matters. Han, Z.-H. and Cadierno, T. (eds) (2010) Linguistic Relativity in Second Language Acquisition: Thinking for Speaking. Clevedon: Multilingual Matters. Hanks, W.F. (1996) Language form and communicative practices. In J.J. Gumperz and S.C. Levinson (eds) Rethinking Linguistic Relativity (pp. 232–270). Cambridge: Cambridge University Press.

Inf luence of L1 Think ing for Speak ing on Use of an L2

27

Hanks, W.F. (2009) Fieldwork on deixis. Journal of Pragmatics 41, 10–24. Hasko, V. (2010) The role of thinking for speaking in adult L2 speech: The case of (non) unidirectionality encoding by American learners of Russian. In Z.-H. Han and T. Cadierno (eds) Linguistic Relativity in Second Language Acquisition: Thinking for Speaking (pp. 34–58). Bristol: Multilingual Matters. Hickmann, M. and Robert, S. (eds) (2006) Space in Languages: Linguistic Systems and Cognitive Categories. Philadelphia, PA: John Benjamins. Jarvis, S. and Pavlenko, A. (2008) Crosslinguistic Influence in Language and Cognition. New York: Routledge. Jessop, L., Suzuki, W. and Tomita, Y. (2007) Elicited imitation in second language acquisition research. Canadian Modern Language Review 64, 215–220. Kim, E.-J. (2003) An analysis of particle errors by heritage and non-heritage learners of Korean. Korean Language in America 8, 37–49. Kim, H.-S.H. (2001) Issues of heritage learners in Korean language classroom. Korean Language in America 6, 257–273. Kondo-Brown, K. (2005) Differences in language skills: Heritage language learner subgroups and foreign language learners. Modern Language Journal 89 (4), 563–581. Kondo-Brown, K. and Brown, J.D. (2008) Introduction. In K. Kondo-Brown and J.D. Brown (eds) Teaching Chinese, Japanese, and Korean Heritage Language Students: Curriculum Needs, Materials, and Assessment (pp. 3–16). New York: Taylor & Francis. Langacker, R.W. (1987) Foundations of Cognitive Grammar, Vol. 1: Theoretical Prerequisites. Stanford, CA: Stanford University Press. Langacker, R.W. (1991) Foundations of Cognitive Grammar, Vol. 2: Descriptive Application. Stanford, CA: Stanford University Press. Lee, Y.-G., Kim, H.-S., Kong, D.-K., Hong, J.-M. and Long, M.H. (2005) Variation in the linguistic profiles of advanced English-speaking learners of Korean. Language Research 41 (2), 437–456. Liu, Y. (1988) Jıˇ zu ˇ yìyì xia¯nggua¯n de qu ¯xiàng bu ˇyu ˇ yu ˇyì fe¯nxi (Analysis of several semantic-related directional complements). Language Teaching and Linguistics Studies 14 (1), 1–17. Liu, Y., Pan, W. and Gu, W. (1983) Practical Modern Chinese Grammar. Beijing: Foreign Language Teaching and Research Press. McGinnis, S. (1996) Teaching Chinese to the Chinese: The development of an assessment and instructional model. In J.E. Liskin-Gasparro (ed.) Patterns and Policies: The Changing Demographics of Foreign Language Instruction (pp. 107–121). Boston, MA: Heinle & Heinle. McNeill, D. and Duncan, S.D. (2000) Growth points in thinking-for-speaking. In D. McNeill (ed.) Language and Gesture (pp. 141–161). Cambridge: Cambridge University Press. Narasimhan, B. and Brown, P. (2009) Getting the inside story: Learning to talk about containment in Tzeltal and Hindi. In V.C. Mueller-Gathercole (ed.) Routes to Language: Studies in Honor of Melissa Bowerman (pp. 97–132). New York: Psychology Press. Navarro, S. and Nicoladis, E. (2005) Describing motion events in adult L2 Spanish narratives. In D. Eddington (ed.) Selected Proceedings of the 6th Conference on the Acquisition of Spanish and Portuguese as First and Second Languages (pp. 102–107). Somerville, MA: Cascadilla Proceedings Project. Odlin, T. (2008) Conceptual transfer and meaning extensions. In P. Robinson and N.C. Ellis (eds) Handbook of Cognitive Linguistics and Second Language Acquisition (pp. 306– 340). London: Routledge. Oh, K.-J. (2003) Language, cognition, and development: Motion events in English and Korean. Unpublished doctoral dissertation, University of California. Ortega, L., Iwashita, N., Norris, J. and Rabie, S. (2002) An investigation of elicited imitation in crosslinguistic SLA research. Conference handout from paper presented at the October meeting of the Second Language Research Forum, Toronto, Canada.

28

Studies in Second L anguage Acquisit ion of Chinese

Peyraube, A. (2006) Motion events in Chinese: Diachronic study of directional complements. In M. Hickmann and S. Robert (eds) Space in Languages: Linguistic Systems and Cognitive Categories (pp. 121–135). Philadelphia, PA: John Benjamins. Robinson, P. and Ellis, N.C. (2008) Conclusion: Cognitive linguistics, second language acquisition and L2 instruction – issues for research. In P. Robinson and N.C. Ellis (eds) Handbook of Cognitive Linguistics and Second Language Acquisition (pp. 489–546). New York: Routledge. Shen, J. (2003) The resultative construction in Chinese: A typological perspective. Chinese Teaching in the World 65 (3), 17–23. Slobin, D.I. (1987) Thinking for speaking. In J. Aske, N. Beery, L. Michaelis and H. Filip (eds) Proceedings of the 13th Annual Meeting of the Berkeley Linguistics Society (pp. 435– 444). Berkeley, CA: Berkeley Linguistics Society. Slobin, D.I. (1993) Adult language acquisition: A view from child language study. In C. Perdue (ed.) Adult Language Acquisition: Cross-linguistic Perspectives (pp. 239–252). Cambridge: Cambridge University Press. Slobin, D.I. (1996a) From ‘thought and language’ to ‘thinking for speaking’. In J. Gumperz and S. Levinson (eds) Rethinking Linguistic Relativity (pp. 70–96). Cambridge: Cambridge University Press. Slobin, D.I. (1996b) Two ways to travel: Verbs of motion in English and Spanish. In M. Shibatani and S.A. Thompson (eds) Grammatical Constructions: Their Form and Meaning (pp. 195–220). Oxford: Clarendon Press. Slobin, D.I. (1998) A typological perspective on learning to talk about space. In H. Ragnarsdóttir and S. Strömqvist (eds) Learning to Talk about Time and Space. Proceedings of the 3rd NELAS Conference (pp. 1–30). Reykjavík and Göteborg: University College of Education and Department of Linguistics, University of Göteborg. Slobin, D.I. (2000) Verbalized events: A dynamic approach to linguistic relativity and determinism. In S. Neimeier and R. Dirven (eds) Evidence for Linguistic Relativity (pp. 107–138). Amsterdam: John Benjamins. Slobin, D.I. (2003) Language and thought online: Cognitive consequences of linguistic relativity. In D. Gentner and S. Goldin-Meadow (eds) Language in Mind: Advances in the Study of Language and Thought (pp. 157–192). Cambridge, MA: MIT Press. Slobin, D.I. (2004) The many ways to search for a frog: Linguistic typology and the expression of motion events. In S. Strömqvist and L. Verhoeven (eds) Relating Events in Narrative. Typological and Contextual Perspectives (pp. 219–257). Mahwah, NJ: Lawrence Erlbaum. Slobin, D.I. (2009) Relation between paths of motion and paths of vision. In V.C. MuellerGathercole (ed.) Routes to Language: Studies in Honor of Melissa Bowerman (pp. 197–222). New York: Psychology Press. Slobin, D.I., Bowerman, M., Brown, P., Eisenbeiss, S. and Narasimhan, B. (2010) Putting things in places: Developmental consequences of linguistic typology. In J. Bohnemeyer and E. Pederson (eds) Event Representation in Language and Cognition (pp. 134–165). Cambridge: Cambridge University Press. Stam, G. (2010) Can an L2 speaker’s patterns of thinking for speaking change? In Z.-H. Han and T. Cadierno (eds) Linguistic Relativity in Second Language Acquisition: Thinking for Speaking (pp. 59–83). Bristol: Multilingual Matters. Talmy, L. (1985) Lexicalization patterns: Semantic structure in lexical forms. In T. Shopen (ed.) Language Typology and Syntactic Description, Vol. 3: Grammatical Categories and the Lexicon (pp. 36–149). Cambridge: Cambridge University Press. Talmy, L. (1991) Path to realization: A typology of event conflation. In L.A. Sutton, C. Johnson and R. Shields (eds) Proceedings of the 17th Annual Meeting of the Berkeley Linguistics Society (pp. 480–519). Berkeley, CA: Berkeley Linguistics Society. Talmy, L. (2000) Toward a Cognitive Semantics, Vol. II: Typology and Process in Concept Structuring. Cambridge, MA: MIT Press.

Inf luence of L1 Think ing for Speak ing on Use of an L2

29

Valdés, G. (1995) The teaching of minority languages as academic subjects: Pedagogical and theoretical challenges. Modern Language Journal 79 (3), 299–328. Vinther, T. (2002) Elicited imitation: A brief review. International Journal of Applied Linguistics 12, 54–73. Wu, S.-L. (2011) Learning to express motion event in an L2: The case of Chinese directional complements. Language Learning 61 (2), 414–454. Wu, S.-L. and Ortega, L. (2013) Measuring global oral proficiency in SLA research: A new elicited imitation test of L2 Chinese. Foreign Language Annals 46 (4), 680–704. Yao, T.-C. and Liu, Y. (1997) Integrated Chinese, Level 1 & Level 2, Textbooks and Workbooks. Boston, MA: Cheng and Tsui Company. Zhou, Y. and Wu, S.-L. (2009) Can elicited imitation be used for the measurement of oral proficiency in L2 Chinese? A pilot study. Unpublished manuscript, University of Hawai’i at Ma¯noa.

2

Pragmatic Development: An Exploratory Study of Requests by Learners of Chinese Xiaohong Wen

Requests have been one of the most frequently examined speech acts in interlanguage pragmatics (ILP) research (Schauer, 2009). Early studies on the ILP of requests tended to focus on first language transfer, comparison between native speakers (NSs) and nonnative speakers (NNSs), and cross-cultural contrastive analysis (Blum-Kulka, 1982; Blum-Kulka & Levenston, 1987; Blum-Kulka & Olshtain, 1984; Faerch & Kasper, 1989; House, 1989; House & Kasper, 1987). This tradition has continued to the present day (e.g. Puetz & Neff van Aertselaer, 2008) and, as a result, only marginal attention has thus far been given to pragmatic development of the speech act of request in second language (L2) learners. Kasper and Schmidt (1996) underscored the need to investigate the pragmatic development of L2 learners and to strengthen connections between ILP and L2 acquisition research. Although recent years have seen more research on the acquisition of speech acts (e.g. Barron, 2003; Rose, 2000, 2009; Schauer, 2004), pragmatic development, as opposed to pragmatic performance and contrastive analysis, has in general remained under-researched, especially with Chinese as the target language (Hong, 1997; Sun & Zhang, 2008). The study reported in this chapter represents an attempt to fill the gap. We begin by briefly reviewing politeness theory in relation to the speech act of request and ILP development studies. We then present the study which aims at exploring developmental features and contextual variations in the written requests produced by English-speaking learners of Chinese at two proficiency levels, and the learners’ use of sociopragmatic and pragmalinguistic strategies.

Background According to Searle (1975) and House and Kasper (1987), requests are defined as directives: Speaker (S) wants Hearer (H) to do x at the expense of 30

Pragmat ic Development : An E xplorator y Study of Request s

31

H. Therefore, S attempts to persuade H to do x. Several variables may potentially make this speech act easy, with a low degree of imposition, or difficult, with a high degree of imposition. If S and H share equal social status, the request is considered socially small, and S is able to decide which linguistic forms to use in the context. This type of request is likely to be successful. However, if any of these conditions are not met, the request is likely to be a frustrating endeavor. According to Brown and Levinson (1987), requests are face-threatening by definition because hearers can interpret requests as intrusive impingements on freedom of action, or even as an exercise of power. Drawing on Brown and Levinson’s (1987) politeness theory, the present study examined Chinese-as-a-foreign-language (CFL) learners’ requests in various sociopragmatic situations. Politeness is a basic notion of pragmatic function and a fundamental principle that people abide by in social communication. Although this notion is shared across cultures, each culture may adopt its own politeness principles. Diverse politeness strategies in making requests can be observed in various languages. For example, imperatives and Want Statements are perceived to be direct and often impolite in English (Blum-Kulka, 1987). In Chinese, however, imperatives with a polite word such as 请 Qıˇng ‘please’ and Want Statement + a soft tag question such as 想 Xiaˇng ‘thinking of’, 行吗 Xíngma? ‘OK’ convey politeness and soft tones. Brown and Levinson (1987) categorize communication strategies based on their politeness theory, among which avoidance of face-threatening acts (FTAs) is one of the most frequently used. Social distance and power relations between interlocutors as well as the degree of imposition that the request represents may strongly affect the requester’s choice of strategies and pragmalinguistic expressions. Sociopragmatics is, therefore, the basis for expressing politeness.

Research on ILP development of requests Studies of pragmatic development include longitudinal studies (Achiba, 2003; Ellis, 1992; Schmidt, 1983) and cross-sectional studies (e.g. Hendriks, 2008; Hill, 1997; Rose, 2000, 2009; Scarcella, 1979; Trosborg, 1995). The longitudinal studies of Ellis (1992) and Achiba (2003) focused on the acquisition of ILP by younger learners. Participants in the Ellis (1992) study were two immigrant boys in England, aged 10 and 11. The study found a clear progression from direct to conventionally indirect request strategies over time. At the early stage, the learners used verbless requests, which were followed by imperatives with verbs. The two young learners were then observed to use formulaic permission such as ‘Can I have . . .?’, the ability question ‘Can you . . .?’, and the permission strategy with other verbs, such as ‘Can I take . . .?’ Both learners showed infrequent use of internal or external modifiers,1 although they frequently employed the politeness marker please.

32

Studies in Second L anguage Acquisit ion of Chinese

Achiba (2003) observed a seven-year-old girl who was visiting Australia from another country with her family. Achiba’s findings were consistent with those of Ellis (1992): the learner started with direct strategies, such as imperatives, moved to the suggestory formula ‘Let’s . . .’, and continued with conventionally indirect permission and ability questions. At a later stage, the learner further expanded her repertoire to include other strategies such as willingness: ‘Will you . . .?’ and past tense modals as in ‘Could you . . .?’ or ‘Would you . . .?’ Rose (2000, 2009) conducted cross-sectional studies with Chinesespeaking learners of English in Hong Kong. In his first study, Rose found a number of developmental patterns in the choice of request strategy, with conventional indirectness the most frequent strategy overall. Participants in the study were classified into three proficiency levels. The two higher proficiency groups showed the most frequent use of conventional indirectness (96.8% and 85.7% with the P-6 and P-4 groups, respectively). The lowest proficiency group also used conventional indirectness, although less frequently (35.4% with the P-2 group). The findings of Rose’s (2009) study are consistent with the previous and later studies (Achiba, 2003; Ellis, 1992; Hendriks, 2008; Hill, 1997; Rose, 2009; Trosborg, 1995). It should be noted, however, that the low proficiency group in Hill’s (1997) study relied heavily on direct strategies. Hill speculated that these learners were still struggling with the complexities of producing grammatically correct sentences. The use of direct strategies decreased as the learners’ proficiency level increased (see also Achiba, 2003; Ellis, 1992; Hendriks, 2008; Rose, 2009; Trosborg, 1995). Rose (2000, 2009) elicited data respectively through a cartoon oral production task (COPT) and a discourse completion test (DCT). In addition to yielding results consistent with those of his (2000) study, Rose (2009) revealed: (1) that the mean frequency of alerters2 increased with proficiency level; (2) that the occurrence of please decreased slightly with the increasing level of proficiency; and (3) that the frequency of external modifications more than doubled at each advancing level. In terms of sociopragmatics, learners exhibited some variation in their requests to reflect changes in dominance relations. For example, all learners evidenced some awareness of dominance relations in their use of alerters, with learners at the highest level of proficiency supplying more supportive moves in hearer-dominant situations. In summary, previous studies of ILP development of requests share some findings. First, the request strategies of NNSs gradually approximate those of the NSs. Second, as the requesters’ proficiency level advances, the frequency of their use of direct strategies decreases, while that of conventional indirectness increases (Achiba, 2003; Ellis, 1992; Hill, 1997; Rose, 2000, 2009; Schauer, 2004; Trosborg, 1995). Third, the frequency of using external modifications positively correlates with the proficiency level

Pragmat ic Development : An E xplorator y Study of Request s

33

(Hendriks, 2008; Hill, 1997; Rose, 2000, 2009; Trosborg, 1995). Fourth, the developmental patterns of internal modification seem to be more varied than those of supportive moves. Fifth, the occurrence of politeness markers (e.g. please) decreases as the requesters’ proficiency level increases (Hendriks, 2008; Hill, 1997; Rose, 2009). Last but not least, individual variations exist in the use of internal modification (Hill, 1997; Schauer, 2009; Trosborg, 1995).

Request strategies and modifications Drawing on the work of Brown and Levinson (1987) and Leech (1983), Blum-Kulka et al. (1989a) categorized request strategies in terms of degrees of directness from mood derivable (imperative) to mild hint (non-conventional indirectness). They concluded that requests in English usually adopt conventionally indirect strategies such as suggestory formulae and query preparatories, both of which are in question form. Blum-Kulka et al. (1989b) also developed schemes for requests in the Cross-cultural Speech Act Realization Project (CCSARP) that contain internal and external modifications. Internal modification refers to a set of linguistic devices that modify the illocutionary force of the head act, the request itself. Internal modifiers consist of upgraders and downgraders that increase or decrease the illocutionary force of requests. These modifiers are realized through syntactic and lexical/phrasal means. External modification (or supportive move) appears before or after the head act and helps minimize the face-threatening effect. Whether or not to include alerters and supportive moves depends on the desired imposition, and the social/psychological relationships of the interlocutors. The classification scheme used in the present study was based on BlumKulka et al.’s (1989b) CCSARP Coding Manual and on Zhang (1995), who used the same manual. Zhang’s data comprised requests produced by Chinese NSs. Her classification was adapted for use in the present data analysis. This study additionally drew on Thomas’ (1983) two types of judgment of a pragmatically successful utterance to determine the accessibility and acceptability of participants’ requests. Type 1, the pragmalinguistic assessment, means that the illocutionary intent should be clear and the sentence should be grammatical; type 2, the sociopragmatic judgment, refers to language use that should be acceptable in the context. The study focused on CFL learners’ pragmatic development in relation to requests, guided by the following questions: (1) Is there evidence of pragmalinguistic development in lower level versus advanced learners? (2) Is there evidence of sociopragmatic development in lower level versus advanced learners?

34

Studies in Second L anguage Acquisit ion of Chinese

Method Participants Eighty-nine CFL students enrolled in Chinese classes at a university in the US participated in the study. However, 41 of them, who were Chinese heritage learners, were excluded from data analysis on the assumption that their language background would influence their pragmatic performance. The remaining 48 participants, who were non-heritage language learners and whose data were analyzed and reported below, came from four classes at two levels, lower and advanced. Among them, 24 students (15 females, 9 males, average age 21.5) were at the lower level and 24 (13 females, 11 males, average age 23) at the advanced level. At the time of data collection, the lower level group had studied Chinese for 19 months, and the advanced group 31 months. Nine of the advanced students had been to China and stayed for varying lengths of time, from one month to 12 months. Their proficiency levels were based on their performance on an oral and a written Chinese proficiency test conducted by course instructors at the beginning of the semester. Twenty-four Chinese NSs (13 females, 11 males, average age 31), all from the People’s Republic of China, were recruited. Their data served as the baseline for the comparison of learners’ Chinese language use. Most of them were undergraduate and graduate students from the same university as the CFL learners.

Instrument A two-part questionnaire was used. The first part focused on biographic details such as age, gender, ethnic background, first language and exposure to the L2 in the home environment. The second part was a DCT. As Rose (2009: 2347) posits, the DCT provides researchers with information on respondents’ knowledge and attitudes, and ‘as such can be used as a measure of changes in knowledge and attitudes across groups that might be indicative of development’. Considering the likelihood, relevance and context variables of the participants’ requests (Bardovi-Harlig, 1999; Rose, 2009), the present study selected five scenarios: (1) borrowing a pen from a classmate, (2) asking to postpone an essay submission, (3) asking a professor for an appointment, (4) asking a roommate to return a library book and (5) asking a stranger for directions – four of them adapted from previous studies (Hong, 1997; Shauer, 2009). Each scenario came with a detailed description, explaining the relationship between the interlocutors. The interlocutors were assigned Chinese names such as 张明 Zha¯ng míng, 李小友 Li xia ˇoyou, 李小京 Li xia ˇojı¯ng, 王老师 Wáng la ˇoshı¯, 李老师 Li la ˇoshı¯, so that participants would use contextual information to avoid imagining their own contexts while completing the task. The questionnaire was written in English.

Pragmat ic Development : An E xplorator y Study of Request s

35

The five scenarios were subsequently presented to 10 CFL learners, who were not participants in the study, for their metapragmatic assessment focusing on estimating: (1) how often the requests would occur (very frequently, relatively frequently, not frequently); (2) how sociopragmatically difficult the request would be (easy, moderately difficult, difficult); and (3) justifying the ratings. The results showed that the raters agreed on all scenarios in terms of the degree of imposition, with scenarios 2 and 3 considered ‘difficult’ and ‘moderately difficult’, respectively, and scenarios 1, 4 and 5 ‘easy’. All scenarios were deemed ‘very frequent’ or ‘relatively frequent’ except for scenario 5 (asking a stranger for directions), which was rated as ‘not frequent’ by five student raters, ‘relatively frequent’ by four raters, and ‘very frequent’ by one rater. Due to the lack of consistency, scenario 5 was dropped. The DCT questionnaire was then piloted on five CFL learners and five NSs who were not otherwise involved in the study. Based on their responses, the wording of the questionnaire was adjusted. Table 2.1 shows the categorization of the four request scenarios according to two variables: social status and request imposition. The final questionnaire is presented in the Appendix.

Procedure The data were collected in two consecutive spring semesters. In spring 2010 students from two lower-level and one advanced Chinese classes participated in the data collection. However, after sorting students by heritage background, only a limited number of questionnaires were usable for the study. In spring 2011 data were again collected from participants in one lower-level and one advanced Chinese class. In both instances of data collection, copies of the questionnaire were handed out to the instructors, who asked their students to complete it in class. The instructors told students that this survey was a study on language use and that they could write their responses in characters or Pinyin. Copies of the questionnaire for NSs were sent via email to the Chinese Students Association at the same university where the CFL data were collected. Due to the limited number of responses from NSs, 10 additional copies were sent out to NSs originally from Mainland China who were similar in age to the NS participants. Table 2.1 Four scenarios with two levels of social status and request imposition Equal status (student to student) Low imposition

Higher status (student to professor) Relatively high imposition

Scenario 1: Borrowing a pen from a classmate

Scenario 2: Asking a professor to postpone an essay submission Scenario 3: Asking a professor for an appointment because you have lots of questions

Scenario 4: Asking a roommate to return a book since he/she is going to the library

36

Studies in Second L anguage Acquisit ion of Chinese

Coding First, six questionnaires, three from each level, were independently coded by the researcher and a colleague, with the interrater agreement 0.81. Then another six questionnaires, three from each level, were coded independently, with the interrater agreement 0.92. Any disagreements were resolved item by item through consensus coding. The remaining portions of the data were coded by the researcher alone. The requests were coded for the level of directness (direct, conventionally indirect and non-conventionally indirect) for the head act (e.g. the request of 能借一支笔吗? Néng jiè yı¯zhı¯bıˇ ma ‘Can (I) borrow a pen?’ was coded as conventionally indirect), for internal modification at the syntactic and lexical/ phrasal levels (e.g. 要是你去图书馆, 顺便把我的书也还了吧. Yàoshi nıˇ qù túshu ¯guaˇn, shùnbiàn baˇ woˇdeshu ¯ yeˇ huánle ba. ‘If you go to the library, (what about) returning my book while you are there?’), for use of alerters (e.g. 李小 京 Li xia ˇojı¯ng ‘a person’s name’) and supportive moves (e.g. ‘这个星期的作业 太多了, 中文作文我交不了了. Zhège xı¯ngqíde zuòyè tàiduo¯le, zho¯ngwén zuòwén woˇ jia ¯o bùliaˇole ‘I have too much homework this week. I am not able to submit the Chinese composition’).

Analysis and Results Request perspectives A request can be hearer or speaker dominant, or can assume an implicit perspective, as when the subject of the sentence is omitted in the context (Zhang, 1995). Examples 1–3 illustrate the hearer, speaker and implicit perspective, respectively. Table 2.2 shows that the request perspectives changed Table 2.2 Frequency of request perspectives by group (number and percentage) Scenarios 1 and 4 S–S: low imposition Perspectives Lower level Adv NSs

Hearer Domin 31 64.6 27 56.3 10 20.8

Speaker Domin 12 25 6 12.5 6 12.5

Scenarios 2 and 3 S–P: high imposition Implicit 5 17.9 15 31.3 32 66.7

Hearer Domin 17 35.4 24 50.0 15 31.3

Notes: S, Student; P, Professor; Domin, Dominant; Adv, Advanced.

Speaker Domin 26 54.2 17 35.4 16 33.3

Implicit 5 17.9 7 14.6 17 35.4

Pragmat ic Development : An E xplorator y Study of Request s

37

along with the sociopragmatic situation, and that the use of zero pronouns in the subject position increased with advancing proficiency level, aligning more with the usage of the NSs. Example 1 小友, 你可以借给我一只笔吗? (Adv #36, S1) Xiaˇoyou, nıˇ keˇyıˇ jiègeˇi woˇ yı¯zhıˇ bıˇ ma? ‘Xiaoyou, you may lend to me a pen QM3?’ Example 2 星期四我再给您, 好吧? (Adv #35, S3) Xı¯ngqísì woˇ zài geˇinín, haˇoba? ‘Thursday I again give you, OK?’ Example 3 小友, 可以借一下你的笔吗? (NS #4, S1) Xiaˇoyou, keˇyıˇ jièyı¯xià nıˇde bıˇ ma? ‘Xiaoyou, may borrow a little bit your pen QM?’

Request strategies The requests produced by the participants were classified into two major categories: the direct and the conventionally indirect. The direct category included ‘imperative/plain statements’, ‘hedged performatives’ and ‘want statements’. The most direct forms in the data were imperatives (Example 4) and plain statements (Example 5). The plain statement, as shown in Example 5, presented no requestive intent but called the interlocutor’s attention, in which case it appeared that the speaker did not have to persuade the professor. Example 4 小友, 请（借）给我一支/个bi/笔. (Lower level #74, 78, Adv #45, S1) Xiaˇoyou, qıˇng (jiè) geˇi woˇ yı¯zhı¯/gè bıˇ. ‘Xiaoyou, please (lend) give me a pen.’ Example 5 星期二我不可能交作文, 因为别的课都有大大的功课. (Lower level #68, S2) Xı¯ngqíèr woˇ bùkeˇnéng jia ¯ozuòwén, yı¯nwèi biédekè do¯uyoˇu dàdàde go¯ngkè. ‘Tuesday I not possible submit composition, because other classes all have lots of homework.’ The data showed frequent occurrences of hedged performatives; that is, ‘the illocutionary verb denoting the requestive intent is modified, e.g. by modal verbs or verbs expressing intention’ (Blum-Kulka et al., 1989b: 279). This is illustrated in Example 6, where the illocutionary intent is explicitly stated and the illocutionary force softened by hedging expressions. Examples 7–9

38

Studies in Second L anguage Acquisit ion of Chinese

show that the illocutionary force is modified by downgraders. In Example 9, the learner used a promise statement. Example 6 李小京, 你可以(也)把我的书还给图书馆. (Adv #36, 45, S4) Li xia ˇojı¯ng, nıˇ keˇyıˇ (yeˇ) baˇ woˇdeshu ¯ huángeˇi túshu ¯guaˇn. ‘Li Xiaojing, you may (also) BA my book return to library.’ Example 7 即然你要到图书馆去还书, 请你也拿我借的书还给他们, 好吗? (Adv #33, S4) Jírán nıˇ yào dàotúshu ¯gua ˇn qù háishu ¯, qıˇngnıˇ yeˇ ná woˇjièdeshu ¯ huángeˇi ta ¯men, ha ˇoma? ‘Since you will go library return book, please you also take I borrowed book return to them, OK?’ Example 8 小友, 给我jie 一 zhibi吧. (Adv #35, S1) Xiaˇoyou, geˇi woˇ jiè yı¯zhibıˇ ba. ‘Xiaoyou, give me lend a pen SM4.’ Example 9 老师, 我有另一作业在别课, 我能不能星期四我一定得交给你. (Lower level #56, S2) Laˇoshı¯, woˇ yoˇu lìng yı¯ zuòyè zài biékè, woˇ néngbùnéng xı¯ngqísì woˇ yı¯dìng dei jia ¯ogeˇi nıˇ. ‘Teacher, I have other homework in other course. I can not can Thursday I definitely must submit to you.’ Yet another category at the level of directness is the want statement, which was mainly found in the NSs’ data for scenario 3 involving requesting an appointment with a professor. Although such a statement is perceived to be direct and impolite in English, it is quite downtoned in Chinese with a soft intonation, as illustrated in Example 10. On the other hand, a ‘want statement’ in Chinese can be impolite if a particular verb, such as 要 Yào ‘want’ instead of 想 Xiaˇng ‘want/intend’, is used with little modification. The two want statements produced by the lower level group both used a statement form, as seen in Example 11. Example 10 李老师, 我想看看你什么时候有空? 我想问问您书本里的几个问题。(NS #12, S3) Li laˇoshı¯, woˇ xiaˇng kànkàn nıˇ shénme shíhou yoˇukòng? Woˇ xiaˇng wènwènnín shu ¯beˇnlıˇde jıˇgè wèntí. ‘Teacher Li, I intend find out a bit you what time have free time? I intend ask a bit you (honorary you) book a few questions.’

Pragmat ic Development : An E xplorator y Study of Request s

39

Example 11 老师, 我有课文的问题所以我要和你定时间。(Lower level #62, S3) La ˇoshı¯, woˇ yoˇu kèwénde wèntí suoˇyıˇ woˇ yào hé nıˇ dìng shíjia ¯n. ‘Teacher, I have lessons question therefore I want with you make appointment.’ A number of utterances contained grammatical errors that interfered with communication (cf. Thomas, 1983), and were therefore pragmalinguistically inaccessible. As discussed previously, a basic condition for an utterance to be communicatively effective is that it is semantically clear and grammatically correct. Utterances such as Example 12 were not amenable to categorization. They were treated as uninterruptable in this study. They frequently had an interrogatory form. Table 2.3 shows the distribution of the request strategies in all four scenarios. Example 12 小友, 请问, 你有一zhi bi我借吗? (Lower level #55) Xiaˇoyou, qıˇngwèn, nıˇ yoˇu yı¯zhibi woˇ jiè ma? ‘Xiaoyou, please ask, you have a pen I borrow QM?’ Conventionally indirect requests in the data were classified into two categories: (1) query preparatory with modal auxiliaries such as 能 Néng indicating ability/possibility, or 可以 Keˇyıˇ indicating possibility/permission; and (2) time availability, as used in scenario 3: 您下星期什么时候有时间? Nín xiàxı¯ngqí shénme shíhou yoˇushíjia ¯n? ‘When will you have time next week?’. To the extent that a reason (i.e. lack of time) was given here for rejecting the request, this strategy shows consideration towards the hearer (Schauer, 2009). Although the conventionally indirect strategy is a polite means of realizing a request in English, it may not necessarily be so in Chinese unless downgraders are used. Conventional indirectness in Chinese does not have as many syntactic modifications as in English (e.g. a variety of tenses and modalities). In the present study, conventional indirectness was predominantly used by both groups of NSs and CFL learners (Table 2.3) although the degree of politeness varied to a large extent due to the different use of internal and external modifications. Examples 13–15 illustrate utterances containing few internal and external downgraders, thereby sounding direct and impolite. Example 13 现在我那么忙, 没有时间做作业。星期四可以给你我的作文吗? (Adv #32, S2) Xiànzài woˇ nàmemáng, méiyoˇu shíjia¯n zuòzuoyè. Xı¯ngqísì keˇyıˇ geˇinıˇ woˇde zuòwén ma? ‘Now I that busy, not have time do homework. Thursday may give you my composition QM?’

40

Studies in Second L anguage Acquisit ion of Chinese

Table 2.3 Frequency of request strategy by group (number and percentage) Strategies

Direct Strategies

Examples

Imper/Plain S. Want S. 12 12.5 6 6.3 2 2.1

Lower level Adv NSs

Convent. Indirect Hedged perform. Want S. 18 18.8 19 19.8 23 24.0

Query preparatory 50 52.1 69 71.9 71 74.0

Uninterruptable

16 16.7 2 2.1 0

Notes: Convent., Conventionally; Imper., Imperative; S., Statement; Perform., Performative; Adv, Advanced.

Example 14 我有一些问题。你有没有时间帮助我跟我学习说中文? (Adv# 32, S3) Woˇ yoˇu yı¯xie¯ wèntí. Nıˇ yoˇuméiyoˇu shíjia¯n ba¯ngzhù woˇ ge¯nwoˇ xuéxí shuo¯ zho¯ngwén? ‘I have some questions. You have not have time help me with me learn speak Chinese?’ Example 15 老师, 你什么时候有时间（见面）? (Lower level #73, 70, S3) La ˇoshı¯, nıˇ shénme shíhou yoˇu shíjia ¯n (jiànmiàn)? ‘Teacher, you when have time (meet)?’ The distribution of the strategies as shown in Table 2.3 indicates that conventional indirectness was most preferred among participants, both the NSs and the CFL learners. The advanced and native groups used conventional indirectness with approximately the same frequency. The second most frequently used strategy across the proficiency levels was the hedged performative. Such requests employed hedging expressions such as the sentence-final particle for suggestion 吧 ba, adverbial downgrader 一定 Yı¯dìng, and/or a tag-question form which mitigated the illocutionary force by asking for permission 行吗 Xíngma, possibility 可以吗 Keˇyıˇ ma, and opinion 好吗 Ha ˇoma. An independent-samples t-test was used to determine if the lower level group differed significantly from the advanced group in terms of the conventionally indirect strategies they produced. The results revealed that t (45) = −2.28, p = 0.014 (lower level group, M = 2.08, SD = 1.65; advanced group, M = 2.88, SD = 1.25). Therefore, a clear pattern of development was observed, with use of the request strategies approximating the NSs’ norms as learners’ level of proficiency increases.

Pragmat ic Development : An E xplorator y Study of Request s

41

It should be noted that the lower level group produced two want statements, whereas the advanced group produced none. A close examination revealed that the NSs’ group exclusively used the verb 想 Xia ˇng ‘want/intend’, whereas the lower level group used 要 Yào ‘want’ in one instance that made the request impolite, and 想 Xiaˇng in another. It is possible that the production data used in this study may not adequately reflect what the learners at a given level were capable of, and that the advanced group may have used the avoidance strategy, since the advanced group did use 想 Xiaˇng and 要 Yào in their supportive moves. It may be that the advanced learners were aware that the want statement in Chinese could be polite if a correct verb was used. For them, choosing the correct form was not easy due to the fact that both verbs, 想 Xiaˇng and 要 Yào, often share similar meanings, but function differently depending on the sociopragmatic context.

Internal modification Internal modification is classified at the syntactic and lexical/phrasal levels. The syntactic modifiers produced by the participants included (1) adverbial clauses for condition, time and reason, (2) tag questions and (3) modal auxiliaries. An adverbial clause provides an opt-out option for the hearer, thus allowing him or her to keep a distance from the speaker. In this way, both interlocutors can comfortably avoid FTAs, as in Example 16. Example 16 你去图书馆的时候可不可以也拿我的书还去? (Adv #50, S4) Nıˇ qù túshu ¯guaˇnde shíhou keˇbùkeˇyıˇ yeˇ ná woˇdeshu ¯ huán qù? ‘You go library time may not may also take my book return?’ When a question-tag is added to a declarative sentence, the tone of the utterance is softened and encodes politeness. As Zhang (1995) pointed out, if request intention is specified in the main clause, maximum clarity and a degree of politeness are achieved all at once in a tag-question. The choice of appealer, however, can mitigate or aggravate the illocutionary force. For example, . . . 好不好 Haˇobù haˇo? ‘. . . OK?’ would be appropriate if used by an interlocutor of a higher social status addressing someone of a lower social status, but would otherwise be impolite if it were the reverse (Examples 17–18). None of the NSs used it in their utterances (Table 2.4). The data revealed that three CFL learners at the advanced level and five at the lower level used the tag-question appealer 好不好 Haˇobù haˇo? ‘OK?’, although five of them also used alternative forms such as 可以吗 Keˇyıˇma? ‘OK?’ or 好吗 Haˇoma? ‘OK?’. The data therefore suggested that the participants may have treated appealers as interchangeable across different contexts, without realizing that the question-tag appealers in Chinese may perform different functions, and thus are sociopragmatically sensitive. For example, in the

42

Studies in Second L anguage Acquisit ion of Chinese

Table 2.4 Frequency of syntactic modification by group (number and percentage) Levels

Adverbial clause Tag question ok?

Examples

要是..., If...; ... 的时侯, when ...,... 3

..., 行吗/行不 ..., 好不好? 行? ..., 好/可以 Suggestion 吗? Possibility 9 5

可以 May

3.1 8 8.3 9 9.4

9.4 7 7.3 10 10.5

40.6 46 47.9 28 29.2

Lower level Adv. NSs

Modal auxiliary

5.2 6 6.3 0

39

能 Can 3 3.1 7 7.3 25 26.0

会 Would 2 2.1 0 0

Notes: Synt., Syntactic; Modif., Modification; Adv, Advanced.

situation where a request was proposed from a student to a teacher (Examples 17 and 18), the tag-question appealer that entails approval 行吗 Xíng ma? ‘OK?’ or possibility 可以吗 Keˇyıˇ ma? ‘OK?’ is more appropriate than 好不好 Haˇobùhaˇo? ‘OK?’. Example 17 王老师, 这个星期我忙极了。我给您我的功课这个星期四, 好不好? (Adv #40, S2) Wáng laˇoshı¯, zhège xı¯ngqí woˇ máng jíle. Woˇ geˇi nín woˇde go¯ngkè zhège xı¯ngqísì, haˇobùhaˇo? ‘Wang teacher, this week I busy extremely. I give you (honorary you) my homework this Thursday, OK?’ Example 18 王老师, 现在我有非常多的功课, 所以我不能交我的作文。在星期四交, 好不 好? (Adv #44, S2) Wáng laˇoshı¯, xiànzài woˇ yoˇu fe¯ichángduo¯de go¯ngkè, suoˇyıˇ woˇ bùnéng jia ¯o woˇde zuòwén. Zài xı¯ngqísì jia ¯o, haˇobùhaˇo? ‘Wang teacher, now I have a lot of homework. Therefore I cannot submit my composition. On Thursday submit, OK?’ It is interesting to note that both the lower level and advanced groups mostly used the modal auxiliary 可以 Keˇyıˇ ‘may’ instead of 能 Néng ‘can’, although both were acceptable in the context. The NSs’ group used approximately 50% of each in their conventionally indirect requests (see Table 2.4). A close examination of the data revealed that 87.5% of the CFL learners relied almost exclusively on 可以 Keˇyıˇ. Only six learners used 能 Néng, one of whom

Pragmat ic Development : An E xplorator y Study of Request s

43

at the advanced level consistently used 能 Néng in the formula 麻烦你能不能 V Máfannıˇ néng bùnéng V. . . ‘bother you if you can V . . .’. This student reported having studied in China for a year, which may explain his formulaic usage. In a study abroad context (see Chapters 4 and 5, this volume), learners tend to receive large amounts of target language input, including formulaic expressions, and have opportunities to use their L2 in a variety of contexts involving different interlocutors. This can have an impact on their communicative and cross-cultural competence (Barron, 2003; Schauer, 2009; Taguchi et al., 2013). In addition to syntactic modifiers, lexical/phrasal expressions can also modify the head act. Participants produced three types of lexical/phrasal modifiers: (1) politeness markers that included downgraders such as 麻烦帮我 Máfan ba ¯ngwoˇ ‘bother you to help me’ or upgraders such as 给我/为我 Geˇiwoˇ/ wèiwoˇ ‘for me’; (2) downtoners, i.e. sentential adverbials used to mitigate the request; and (3) understaters, i.e. adverbial modifiers used to decrease the imposition by under-representing the proposition of the request. Moreover, participants’ use of the honorific form of the second person pronoun 您 Nín increased with proficiency. In contrast, their use of the politeness marker 请 Qıˇng ‘please’ decreased with proficiency. In general, 请 Qıˇng ‘please’ is not a preferred marker of politeness in Chinese. If a request is pragmalinguistically or sociopragmatically inappropriate, 请 Qıˇng helps little, as illustrated in Examples 19 and 20. Example 19 你去图书馆吗? 请把我的书拿去。(Lower level #57, S4) Nıˇ qù túshu ¯guaˇn ma? Qıˇng baˇ woˇde shu ¯ ná qù. ‘You go library QM? Please BA my book take to (there).’ Example 20 请你让我星期四交作文。 (Lower level #60, S2) Qıˇng nıˇ ràng woˇ xı¯ngqísì jia ¯o zuòwén. ‘Please you let me Thursday turn in composition.’ The formulaic expression with a politeness marker and a downtoner 麻 烦/帮我顺便VO5 Máfan/ba ¯ngwoˇ shùnbiàn VO ‘bother you/help me VO’ was absent in the learners’ data, but most of the NSs (87.5%) used it in scenario 4 when asking a roommate to return a book. The learners mostly used 给我 Geˇiwoˇ ‘for me’ or 为我 Wèiwoˇ ‘for me’ to realize the request (see Examples 21–22), when 帮我 Ba ¯ngwoˇ ‘help me’ should have been the pragmatically appropriate expression. Although semantically it was true that the hearer was the sole agent who would return the book, pragmatically 给我 Geˇiwoˇ ‘for me’ or 为我 Wèiwoˇ ‘for me’ sounded aggressive and impolite (Example 21). The previously mentioned advanced learner consistently used the polite formulaic expression 麻烦你能不能. . . Máfan nıˇ néngbùnéng. . . ‘to bother you if

44

Studies in Second L anguage Acquisit ion of Chinese

you can. . .’ (Example 22), although the sentence still sounded slightly impolite because of the use of 为我 Wèiwoˇ ‘for me’. Example 21 你可不可以huan这本书给我ma? (Lower level #59, S4) Nıˇ keˇbùkeˇyıˇ huán zhèbeˇnshu ¯ geˇiwoˇ ma? ‘You can not can return this book for me QM?’ Example 22 小京, mafan你能不能为我把这本书还给图书馆? (Adv II #48, S4) Xiaˇojı¯ng,máfán nıˇ néngbùnéng wèiwoˇ baˇ zhèbeˇnshu ¯ huángeˇi túshu ¯ guaˇn? ‘Xiaojing, bother you can not can for me BA this book return to library?’ Table 2.5 shows a small discrepancy in the production of the upgraders 给我 Geˇiwoˇ, 为我 Wèiwoˇ ‘for me’. The lower level group produced one fewer upgrader than the advanced group, while the NSs group produced none. Yet, the advanced group’s overuse of the upgrader may not necessarily be interpreted as their acquisition of polite expressions being behind that of the lower level group, because it is possible that the advanced group may have attempted to elaborate on their meaning more than the lower level group, and that the lower level group may have resorted to avoidance upon finding the expressions difficult. Notice also that none of the learners used the downtoner 顺便 Shùnbiàn ‘simultaneously’. A number of them instead used the adverb 也 Yeˇ ‘also’ (Example 23). 也 Yeˇ has a similar meaning to 顺便 Shùnbiàn ‘simultaneously’, but the latter makes the utterance more polite and face-saving for both interlocutors (Example 24). Example 23 你可以也还我的书/这个书吗? (Lower level #64, Adv #41, S4) Nıˇ keˇyıˇ yeˇ huán woˇdeshu ¯/zhègeshu ¯ ma? ‘You can also return my book/this book QM?’ Example 24 你可以顺便帮我把这本书还了么? (NS #2, S4) Nıˇ keˇyıˇ shùnbiàn ba ¯ngwoˇ baˇ zhèbeˇnshu ¯ huánle me? ‘You can simultaneously help me BA this book return QM?’ The downtoners used by learners were limited. In comparison, the NSs used a large variety of them, as shown in Table 2.5. The advanced group produced a total of three utterances containing the understaters 用一用 Yòngyı¯yòng ‘use a bit’ and 说一说 Shuo¯yı¯shuo¯ ‘speak a bit’, while the lower level group used none. In comparison, the NSs’ group predominantly used the phrase V一下 such as 用一下 Yòngyı¯xia ‘use a bit’, 还一下 Huányı¯xia ‘return (the book) a bit’, or 请教一下 Qıˇngjiào yı¯xia ‘consult a bit’ (Example 3).

6 6.3 23 24.1 34 35.4

Lower level

8 8.3 2 2.1 1 1.0

请 Please

3 3.1 3 3.1

0

麻烦 To bother

2 2.1 19 19.8

0

帮我/忙 To help me

Notes: Lexic, Lexical/phrasal; Modif., Modification; Adv, Advanced; Ptcl, Particle.

NSs

Adv

您 Honorary you

Politeness marker

Examples

Lexic. Modif

2 2.1 3 3.1 0

1一定 1.0 1再 1.0 11 11.5

1 1.0 5 5.2

0

3 3.1 22 22.9

0

V一下 V一V V. a bit

吧 Ptcl

顺便, 一起, 也 With no effort Together/Also 3也 3.1 9也 9.4 21 21.9

给我/为我 For me

一定 再 Definitely Again

Understater

Downtoner

Upgrader

Table 2.5 Frequency of lexical/phrasal modification by group (number and percentage) Pragmat ic Development : An E xplorator y Study of Request s 45

46

Studies in Second L anguage Acquisit ion of Chinese

These downgraders used by the NSs’ group share at least one distinctive feature. The relationships between the downgraders’ form and function are opaque, and their meaning is frequently lost in translation. Additional downgraders include the particle 吧 ba as a suggestion marker, the VP 帮我 ba ¯ngwoˇ ‘help me’ meaning ‘for me’, and the adverb 顺便 shùnbiàn ‘simultaneously’. Semantically, these expressions are not transparent, and pragmatically they are highly functional, making utterances less coercive and consequently more polite. The only modifier used by both learner groups is the downtoner 也 yeˇ ‘also’, whose meaning is concrete and whose form and meaning mapping is straightforward. Tables 2.4 and 2.5 display internal modifications at the syntactic and lexical/phrasal levels, respectively. The data provided evidence of the CFL learners’ pragmalinguistic development. First, as the proficiency level increased, they produced more syntactic modifiers, such as adverbial clauses and the modal auxiliary 能 Néng ‘can’, and lexical/phrasal modifiers, such as politeness markers and the downtoner 也 yeˇ ‘also’. Second, as the proficiency level increased, the use of 请 Qıˇng ‘please’ decreased. Nonetheless, there was a persistent underuse of a number of highly functional phrases. This may have to do with the fact that the meaning of these phrases is not literal. For instance, the formulaic expression 顺便帮我V shùnbiàn ba ¯ngwoˇ V ‘help me simultaneously V’ is merely a polite way of saying ‘do it for me’. Thus, the complexity of meaning and function may pose mapping difficulties for CFL learners. Tables 2.6 and 2.7 compare internal modifications across two different sociopragmatic scenarios. Both lower level and advanced groups used more lexical downgraders such as the politeness marker 您 Nín in scenarios 2 and 3, suggesting that they were aware that politeness markers could be strategically used to help make a successful request. Table 2.6 Internal modifications by group in scenarios 1 and 4 (number and percentage) Internal modifier Syntactic DG

Syntactic UPG

Lexical/phrasal DG

Examples

Tag 好不好? OK?

PM DT US 3 6.3 16 33.3 65 135.4

Lower level Adv NSs

Adv C Tag Q 7 14.6 8 16.7 11 22.9

4 8.3 3 6.3 0

PM 请 Please 7 14.6 1 2.1 1 2.1

Phrasal UPG 给我 为我 For me 1 2.1 3 6.3 0

Notes: DG, Downgrader; UPG, Upgrader; Adv C, Adverbial clause; Tag Q, Tag-question; PM, Politeness marker; DT, Downtoner; US, Understater; Adv, Advanced.

Pragmat ic Development : An E xplorator y Study of Request s

47

Table 2.7 Internal modifications by group in scenarios 2 and 3 (number and percentage) Internal modifier

Syntactic DG

Syntactic UPG

Examples

Adv C Tag Q

Tag 好不好 OK?

Lower level

5 10.4 7 14.6 8 16.7

1 2.1 3 6.3 0

Adv NSs

Lexical/phrasal DG PM DT US 7 14.6 26 54.2 50 104.2

PM 请 Please 1 2.1 1 2.1 0

Phrasal UPG 给我 为我 For me 1 2.1 0 0

Notes: DG, Downgrader; UPG, Upgrader; Adv C, Adverbial clause; Tag Q, Tag-question; PM, Politeness marker; DT, Downtoner; US, Understater; Adv, Advanced.

Alerter Alerters precede the head act to attract the interlocutor’s attention. Alerters produced by the participants were categorized into two types. The first type addresses the interlocutor by name, or generally as 朋友 Péngyoˇu ‘friend’ or 兄弟 Xio¯ngdì ‘brother’. The second type involves the use of 请问 Qıˇngwèn ‘may I ask/excuse me’. The frequency of the first type steadily increased with proficiency. In addition, there was greater use of alerters in scenarios 2 and 3 than in 1 and 4, suggesting that learners at both the lower level and advanced level of proficiency were aware of the sociopragmatic differences between the situations (Table 2.8). A paired samples t-test was conducted to analyze if there were significant differences between low imposition (in scenarios 1 and 4) and relatively high imposition (in scenarios Table 2.8 Comparison of frequency of alerters across scenarios (number and percentage) Scenarios

Scenarios 1, 4

Examples

名字/朋友/兄弟 Name/friend/brother 8 16.7 19 39.6 40 83.3

Lower level Adv NSs

Notes: Adv, Advanced; N, Name; T, teacher.

Scenarios 2, 3 请问 May I ask 3 6.3 1 2.1 0

老师姓 + 老师 Teacher Last N + T 19 39.6 35 72.9 47 97.9

请问 May I ask 0 2 4.2 7 14.6

48

Studies in Second L anguage Acquisit ion of Chinese

2 and 3) across proficiency levels. The lower level group did not show a significant difference between the two sociopragmatic situations, t (23) = −1.86, p = 0.076. In contrast, the advanced group revealed a significant difference between the two sociopragmatic situations, t (23) = −3.24, p = 0.00 (Scenarios 1 and 4, M = 0.79, SD = 0.833; Scenarios 2 and 3, M = 1.46, SD = 0.833). Therefore, sociopragmatic development in terms of interlocutor addresses was observed with increasing proficiency level. Although the function of the phrase 请问 Qıˇngwèn ‘may I ask/excuse me’ is to convey politeness, learners’ usage did not seem to be sociopragmatically appropriate in four out of six occurrences produced by both groups. For example, they used the phrase in a context between friends, making their requests over-polite and thus inappropriate (Example 25). Although the utterance shown in Example 26 happened between a student and a professor, it still sounded unnatural due to a lack of supportive moves in the discourse. Example 27 illustrates the use by an advanced learner and several NSs of 请问 Qıˇngwèn ‘may I ask’ in the middle of the discourse after giving reasons or explanations for the request and before the head act to politely highlight the request. Such strategies make the illocutionary intent clear and the utterances sufficiently polite. As Examples 25–27 show, language forms such as ‘may I ask/excuse me’ function differently according to their discourse context. Example 25 请问, 你可以给我一个bizi? (Lower level #63, S1) Qıˇngwèn, nıˇ keˇyıˇ geˇiwoˇ yı¯gè bıˇzi? ‘Please ask, you may give me a pen?’ Example 26 请问, 可以不可以给你我的作文星期四? (Adv #53, S2) Qıˇngwèn, keˇyıˇ bùkeˇyıˇ geˇinıˇ woˇde zuòwén xı¯ngqísì? ‘Please ask, may not may give you my composition Thursday?’ Example 27 王老师, 这个星期我有很多的功课, 考试。请问我可以星期四给你我的作文 吗? (Adv #41, S2) Wáng laˇoshı¯, zhège xı¯ngqí woˇ yoˇu heˇnduo¯de go¯ngkè, kaˇoshì. Qıˇngwèn woˇ keˇyıˇ xı¯ngqísì geˇinıˇ woˇde zuòwén ma? ‘Wang teacher, this week I have lots of homework, tests. Please ask I may Thursday give you my composition?’

Supportive moves Supportive moves are used by a speaker to persuade a hearer to perform a desired act. The supportive moves produced by the participants included grounders, which provide explanations or justifications for the request,

Pragmat ic Development : An E xplorator y Study of Request s

49

Table 2.9 Frequency of supportive moves by group (number and percentage) Supportive moves Category

Grounder

Preparatory

Apology

Promise

Thanking

Effort

Lower level

36 37.5 62 64.6 71 74.0

4 4.2 8 8.3 17 17.7

1 1.0 3 3.1 8 8.3

0

0

0

1 1.0 3 3.1

1 1.0 6 6.3

0

Adv NSs

1 1.0

Notes: Adv, Advanced.

preparatories, which prepare the hearer for a request, apologizing, promises and thanking. Table 2.9 shows the frequency of supportive moves by group. An independent-samples t-test was used to detect if the lower level group differed significantly from the advanced group in terms of the supportive moves they produced. The results revealed that t (46) = 3.381, p = 0.001 (lower level group, M = 1.67, SD = 1.31; advanced group, M = 3.17, SD = 1.74). Therefore, a significant difference was observed with the advanced group more approximating the NSs’ norms. Among the six identified supportive moves, the grounder was the most used strategy by all participants, suggesting that in human communication, reasons and justifications for a request are cognitively primary. The preparatory supportive move is the second most frequently used strategy. A close examination of the data revealed that 66.7% of the learners used both grounder and preparatory strategies, and 33.3% used the preparatory move only. Some supportive moves are sociopragmatically specific, closely tied to certain scenarios. Two sets of data are presented in Tables 2.10 and 2.11 to Table 2.10 Frequency of supportive moves in scenarios 2 and 3 (number and percentage) Supportive moves

Grounder

Preparatory

Apology

Promise

Thanks

Effort

Lower level

29 60.4 49 102.1 59 122.9

1 2.1 2 4.2 7 14.6

1 2.1 3 6.3 8 16.7

0

0

0

0

0

0

3 6.3

2 4.2

1 2.1

Adv NSs

Notes: Adv, Advanced.

50

Studies in Second L anguage Acquisit ion of Chinese

Table 2.11 Frequency of supportive moves in scenarios 1 and 4 (number and percentage) Levels

Grounder

Preparatory

Promise

Thanking

Lower level

7 14.6 13 27.1 12 25.0

3 6.3 6 12.5 10 20.8

0

0

1 2.1 0

1 2.1 4 8.3

Adv NSs

Notes: Adv, Advanced.

illustrate differences between two types of sociopragmatic scenarios. As shown, both learner groups produced more and varied supportive moves in the situations where the degree of imposition was high than when the situation was otherwise. Moreover, their production of supportive moves and alerters showed that, as their proficiency increased, they were more aware of various sociopragmatic situations, and were generally able to alter their means of external modification accordingly, such as providing more reasons and apologies for requests of high imposition.

Discussion With regard to the first research question, ‘Is there evidence of pragmalinguistic development in lower level versus advanced learners?’, the present study consistently found evidence of pragmalinguistic development in the CFL learners (see Tables 2.2–2.7 and the independent-samples t-tests). Like the NSs, the learners at both proficiency levels predominantly relied on the query preparatory form. Their use of the conventionally indirect and hedged performative strategies increased, and their use of the bare imperative with little modification decreased with increasing proficiency. It is important to point out, however, that the use of a query preparatory form such as ‘Can you/I . . .?’ in Chinese does not guarantee that the request will come across as polite. Internal downgraders and supportive moves are still essential to softening the illocutionary force. Although limited, learners’ use of downgraders increased; these included adverbial clauses, politeness markers such as 您 Nín ‘honorific you’ and 想 Xiaˇng ‘want/intend’, and the downtoner 也 Yeˇ ‘also’. The overuse of the politeness marker 请 Qıˇng ‘please’ decreased with the advanced group. This result is consistent with previous findings (Barron, 2003; Hendriks, 2008; Hill, 1997; Rose, 2009). With regard to the second research question, ‘Is there evidence of sociopragmatic development in lower level versus advanced learners?’, we also

Pragmat ic Development : An E xplorator y Study of Request s

51

found consistent evidence of sociopragmatic development throughout the data (see Tables 2.2, 2.6–2.8, 2.10–2.11, and the paired-samples t-tests). First, the use of request perspectives increasingly varied with the sociopragmatic situation. The learners’ use of an implicit perspective generally progressed toward that of the NSs. Second, the advanced group outperformed the lower level group, using more lexical/phrasal downgraders in scenarios 2 and 3 than in 1 and 4. Third, the advanced group employed more alerters and supportive moves than the lower level group in scenarios 2 and 3 than in 1 and 4, exhibiting increasing variation corresponding to differing sociopragmatic situations. In summary, the data revealed that the CFL learners seemed to be not only increasingly aware of the sociopragmatic differences embodied in various request situations, but also able to use a progressively greater variety of strategies appropriate to the context. The ample evidence of learners’ pragmalinguistic development included increasing use of conventionally indirect strategies, internal modifications at syntactic and lexical/phrasal levels, alerters, supportive moves, and zero pronouns in the subject position. The findings also provided evidence of learners’ sociopragmatic development across different proficiency levels. This development was seen in particular in their use of a variety of downgraders, alerters and supportive moves to accommodate the degree of imposition and the social status between the interlocutors. The overextension of 请 Qıˇng ‘please’ and upgraders decreased in different contexts as the proficiency level increased. The findings suggest that the increase in proficiency from lower level to advanced correlates with more accurate use of NS-like strategies. This study, however, revealed a large discrepancy between the CFL learners and the NSs, particularly in the use of pragmatically functional particles and phrases such as: 吧 ba, a sentential particle for suggestion; 行 吗 Xíngma, a tag-question appealer for permission; a downtoner, 顺便 Shùnbiàn; a politeness marker, 麻烦 Máfan; a polite expression for acknowledging help, 帮我/忙V Ba¯ng woˇ/máng V; an understater for softening an imposition, V一下/V V yı¯xià/V; and a polite expression for requestive intention, 想 Xiaˇng (Tables 2.4–2.7). It is unlikely that these modifiers were lacking in the learners’ instructional input, since all the expressions appeared in their lower level course materials and had been practiced both in class and in their home assignments. Interestingly (but not surprisingly), the learners’ underuse of the expressions seems to have been selectively confined to those that do not provide clear form-meaning/function connections. The meanings of some of the expressions are opaque. For example, 吧 ba is devoid of referential meaning, but pragmatically has a suggestive function. Similarly, V一下 yı¯xià/VV is a mild pragmatic understater, and 顺便 Shùnbiàn has no concrete meaning, but can function as a pragmatic downtoner. Furthermore, the meanings of some expressions are non-literal. For example, 帮我/忙 Ba¯ng woˇ/máng literally translates as ‘help

52

Studies in Second L anguage Acquisit ion of Chinese

me’ but figuratively means ‘for me’. Similarly, the literal translation of 麻烦 Máfan is ‘to bother’, but the expression serves as a politeness marker. Still other expressions, such as 行吗 Xíngma? ‘OK?’ and 想 Xiaˇng ‘want/ intend’, are not always interchangeable with 好不好 Haˇobù haˇo? ‘OK?’ and 要 Yào ‘want’, respectively, each conveying a different level of politeness. The use of these expressions is context-sensitive, depending on sociopragmatic factors. They may be optional or necessary in a particular situation. For example, 一下 yı¯xià ‘a bit’ in V一下 V yı¯xià can be optional. The verb itself has inherent semantic value, but the complement 一下 yı¯xià is semantically redundant, though pragmatically preferred sometimes (cf. VanPatten, 2002). The learners may not have realized that these opaque and non-literal expressions were not just ‘regular words’ needed for correct grammar usage, but could be essential and functional devices for producing polite requests. DeKeyser (2005) posits that the redundancy and optionality of a formmeaning relation may result in its lack of transparency and pose learning difficulties. Findings from the present study lend support to this claim. Among other things, the learners appear to have acquired the expressions that convey concrete meanings and are transparent in their form-meaning mapping, e.g. 也 Yeˇ ‘also’, while the more abstract expressions such as 用一 用 Yòngyı¯yòng ‘use a bit’, 说一说 Shuo¯yı¯shuo¯ ‘speak a bit’, 麻烦 Máfan ‘bother’, and 帮我 Ba¯ngwoˇ ‘help me’ or 替我 Tìwoˇ (polite form of) ‘for me’ remained a persistent challenge.

Further Inquiry The results of this study should be regarded as tentative due to the small number of scenarios and participants employed. These limitations may affect conclusions to a certain extent. In addition, since the focus of the study was on the pragmatic development of CFL learners, it would have been desirable that a longitudinal perspective be incorporated in its design, which would have led to a more complete picture of CFL ILP. In spite of these limitations, the study did yield several important findings as reported and discussed in the previous sections. The few extant studies of requests in CFL learners (Hong, 1997; Sun & Zhang, 2008) have focused almost exclusively on analyzing request strategies and supportive moves, with little attention paid to the internal modification and pragmatic function of particles, such as syntactic and lexical/ phrasal downgraders. However, it is in this latter area that the present study revealed large discrepancies across the learner groups and between the CFL learners and the NSs. Further research is warranted on these discrepancies, particularly in relation to those linguistic elements that are inherently low in semantic value (see VanPatten, 2002) but high in sociopragmatic value.

Pragmat ic Development : An E xplorator y Study of Request s

53

Appendix: Discourse Completion Test Thank you for participating in this ‘test’! The questionnaire is designed to identify your Chinese language ability to communicate properly in predesigned contexts. Proper communication means that you are able to use linguistically and pragmatically appropriate language and devices in various social situations. Please respond to the four items in the second part of the questionnaire in the manner of natural conversation. Your name: (1) Your gender: M; F (2) Your age: (3) Check your ethnic background: Caucasian; African-American; Hispanic; Chinese; Asian (please specify your nationality); Other (Specify) (4) What is your first language? (5) Are you somewhat bilingual with Chinese as one of the languages?_____ Yes; No (6) What is/are the languages that you speak or hear from your parents or relatives? (You may check more than one) English; Chinese; Asian language; Spanish; Others, specify (7) Your birth place: (8) At what age did you move to the US if you were not born in the US? __________ (9) The Chinese language course you are currently enrolled on: First year; Second year; Third year; Fourth year (10) Number of years of Chinese language study in high school: 0; ____ 1; 2; 3; 3+ (11) Number of years of Chinese language study in Chinese School: 0; ____ 1; 2; 3. 3+ (12) Have you studied in Chinese-speaking countries after high school? ___ Yes; No. If ‘yes’, how long ?

I. Complete the dialogue in Chinese characters or Pinyin (1) (Between two classmates, 张明 and 李小友.) 张明 was about to write a Chinese dictation and realized that he forgot to bring pens. 张明 wants to borrow a pen from 李小友, his classmate sitting next to him. 张明: ________________________________________________________ 李小友: 好啊, 在这儿, 你拿去用吧。

54

Studies in Second L anguage Acquisit ion of Chinese

(2) (Between 张明 and his Chinese literature professor, 王老师.) 张明 was supposed to turn in his composition on Tuesday. But he cannot do so because of massive assignments from other classes. 张明 explains the situation and makes a request to 王老师 to see if he can turn it in on Thursday. 张明: ________________________________________________________ 王老师: 好吧, 星期四你一定得交啊！ (3) (Between 张明 and his Chinese language teacher, 李老师.) 张明 wants to make an appointment with 李老师 to consult with her about a few questions in the textbook. He also wants to speak Chinese with 李老师. He wonders when 李老师 has time for the appointment. 张明: ________________________________________________________ 李老师: 下星期四一天我都有时间。 (4) (Between 张明 and his roommate, 李小京.) 李小京 is going to the school library to return a book. 张明 also has a book that needs to be returned. 张明 wants to ask 李小京 to return the book for him, too. 张明: ________________________________________________________ 李小京: 好, 没问题。

Notes (1) Internal modifiers consist of linguistic devices within the request itself that modify the illocutionary force, whereas external modifiers are external to the request, occurring either before or after it. (2) Alerters are used to get the interlocutor’s attention. (3) QM, question marker. (4) SM, suggestion marker. (5) VO, verb and object.

References Achiba, M. (2003) Learning to Request in a Second Language: Child Interlanguage Pragmatics. Clevedon: Multilingual Matters. Bardovi-Harlig, K. (1999) Researching method. In L.F. Bouton (ed.) Pragmatics and Language Learning, Vol. 8 (pp. 237–264). Urbana-Champaign: University of Illinois at Urbana-Champaign, Division of English as an International Language. Barron, A. (2003) Acquisition in Interlanguage Pragmatics: Learning How to Do Things with Words in a Study Abroad Context. Pragmatics & Beyond Series No. 108. Amsterdam: John Benjamins. Blum-Kulka, S. (1982) Learning how to say what you mean in a second language: A study of the speech act performance of learners of Hebrew as a second language. Applied Linguistics 3, 29–59. Blum-Kulka, S. (1987) Indirectness and politeness in requests: Same or different? Journal of Pragmatics 11, 131–146.

Pragmat ic Development : An E xplorator y Study of Request s

55

Blum-Kulka, S. and Levenston, W.A. (1987) Lexical-grammatical pragmatic indicators. Studies in Second Language Acquisition 9 (2), 155–170. Blum-Kulka, S. and Olshtain, E. (1984) Requests and apologies: A cross-cultural study of speech act realization patterns (CCSARP). Applied Linguistics 5 (3), 196–213. Blum-Kulka, S., House, J. and Kasper, G. (1989a) Investigating cross-cultural pragmatics: An introductory overview. In S. Blum-Kulka, J. House and G. Kasper (eds) CrossCultural Pragmatics: Requests and Apologies (pp. 1–36). Norwood, NJ: Ablex. Blum-Kulka, S., House, J. and Kasper, G. (1989b) The CCSARP coding manual. In S. Blum-Kulka, J. House and G. Kasper (eds) Cross-Cultural Pragmatics: Requests and Apologies (pp. 273–294). Norwood, NJ: Ablex. Brown, P. and Levinson, S.D. (1987) Politeness: Some Universals in Language Use. Cambridge: Cambridge University Press. DeKeyser, R.M. (2005) What makes learning second language grammar difficult? A review of issues. Language Learning 55 (1), 1–25. Ellis, R. (1992) Learning to communicate in the classroom. Studies in Second Language Acquisition 14, 1–23. Faerch, C. and Kasper, G. (1989) Internal and external modification in interlanguage request realization. In S. Blum-Kulka, J. House and G. Kasper (eds) Cross-Cultural Pragmatics: Requests and Apologies (pp. 221–247). Norwood, NJ: Ablex. Hendriks, B. (2008) Dutch English requests: A study of request performance by Dutch learners of English. In M. Puetz and J. Neff van Aertselaer (eds) Developing Contrastive Pragmatics: Interlanguage and Cross-Cultural Perspectives (pp. 335–354). Berlin: Mouton de Gruyter. Hill, T. (1997) The development of pragmatic competence in an EFL context. Unpublished doctoral dissertation, Temple University, Tokyo. Hong, W. (1997) Socio-pragmatics in language teaching: With examples of Chinese request. Journal of Chinese Language Teachers Association 32 (1), 95–107. House, J. (1989) Politeness in English and German: The functions of please and bitte. In S. Blum-Kulka, J. House and G. Kasper (eds) Cross-Cultural Pragmatics: Requests and Apologies (pp. 96–122). Norwood, NJ: Ablex. House, J. and Kasper, G. (1987) Interlanguage pragmatics: Requesting in a foreign language. In W. Lörscher and R. Schultze (eds) Perspectives on Language in Performance. Festschrift für Werner Hüllen (pp. 1250–1288). Tübingen: Narr. Kasper, G. and Schmidt, R. (1996) Developmental issues in interlanguage pragmatics. Studies in Second Language Acquisition 18, 149–169. Leech, G.N. (1983) Principles of Pragmatics. New York: Longman. Puetz, M. and Neff van Aertselaer, J. (eds) (2008) Developing Contrastive Pragmatics: Interlanguage and Cross-Cultural Perspective. Berlin: Mouton de Gruyter. Rose, K.R. (2000) An exploratory cross-sectional study of interlanguage pragmatic development. Studies in Second Language Acquisition 22, 27–67. Rose, K.R. (2009) Interlanguage pragmatic development in Hong Kong, phase 2. Journal of Pragmatics 41, 2345–2364. Scarcella, R. (1979) On speaking politely in a second language. In C.A. Yorio, K. Peters and J. Schachter (eds) On TESOL ‘79: The Learner in Focus (pp. 275–287). Washington, DC: TESOL. Schauer, G.A. (2004) May you speak louder maybe? Interlanguage pragmatic development in requests. In S.H. Foster-Cohen, M. Sharwood Smith, A. Sorace and M. Ota (eds) Interlanguage Pragmatic Development in Requests. EUROSLA Yearbook, Vol. 4 (pp. 253–273). Amsterdam: John Benjamins. Schauer, G.A. (2009) Interlanguage Pragmatic Development: The Study Abroad Context. London: Continuum. Schmidt, R. (1983) Interaction, acculturation and the acquisition of communicative competence: A case study of an adult. In E. Judd and N. Wolfson (eds) Sociolinguistics and Language Acquisition (pp. 137–174). Rowley, MA: Newbury House.

56

Studies in Second L anguage Acquisit ion of Chinese

Searle, J.R. (1975) Indirect speech acts. In P. Cole and J.L. Morgan (eds) Syntax and Semantics: Vol. 3. Speech Acts (pp. 59–82). New York: Academic Press. Sun, X. and Zhang, D. (2008) American college students’ requesting competence in Chinese as a foreign language. Chinese Teaching in the World 3, 105–113. Taguchi, N., Li, S. and Xiao, F. (2013) Production of formulaic expressions in L2 Chinese: A developmental investigation in a study abroad context. Chinese as a Second Language Research 2 (1), 23–58. Thomas, J. (1983) Cross-cultural pragmatic failure. Applied Linguistics 4 (2), 91–112. Trosborg, A. (1995) Interlanguage Pragmatics. Berlin: Mouton de Gruyter. VanPatten, B. (2002) Processing instruction: An update. Language Learning 52, 755–803. Zhang, Y. (1995) Strategies in Chinese requesting. In G. Kasper (ed.) Pragmatics of Chinese as Native and Target Language. Technical Report No. 5 (pp. 23–68). Honolulu: University of Hawai’i at Manoa, Second Language Teaching & Curriculum Center.

3

Peer/Group Interaction in a Mandarin Chinese Study Abroad Context Li Jin

It has been assumed that study abroad provides an ideal environment for foreign language learners to receive various types of language input, which leads to effortless and osmotic linguistic and cultural development. Many educators and researchers believe students should be completely immersed in the target language context, interacting with native speakers in the target language as much as possible from the moment they set foot in the host country, regardless of their prior language learning experience (Lafford, 1995; Lapkin et al., 1995). However, there are serious inconsistencies among research studies as to what results in students’ successful linguistic and cultural development in study abroad. One line of research that has gained much attention in recent years focuses on the formal and informal social interaction students have within local communities (e.g. Isabelli-Garcia, 2006; Levin, 2001; Magnan & Back, 2007; Wilkinson, 1998). Some second language (L2) researchers (e.g. Kinginger, 2008, 2009; Ochs, 2002) propose viewing L2 development in study abroad as a language socialization process, similar to how children engage in their first language (L1) environment. In other words, students in study abroad contexts are assumed to strive to be socialized into the target culture by learning the socially acceptable language behaviors through interactions with local residents. Questions such as whether these students are ready to be socialized into the target culture and whether native speakers of the target language are willing to provide appropriate linguistic and cultural assistance to help socialize the learners remain open to debate. In recent years, a growing number of American students have been studying abroad in China. Many of them go with the assumption that their language proficiency will inevitably improve once they live in China and are surrounded by Chinese-speaking people. On the other hand, it is still unclear whether the historically monocultural Chinese people, particularly those with whom American study abroad students interact on a daily basis, are 57

58

Studies in Second L anguage Acquisit ion of Chinese

ready or willing to provide a learning-conducive social environment. There is an urgent need to provide educators and students involved in study abroad programs with a better understanding of how to take advantage of the formal and informal learning resources available in study abroad programs. Peer interaction, despite the undisputed learning conduciveness discovered in foreign language learning settings (e.g. Donato, 1994; Ohta, 1995; Swain et al., 2002; Watanabe & Swain, 2007), is still a largely neglected topic in study abroad research. This chapter reports on an empirical study of how college-level American students used and perceived peer/group interaction as a way of learning in a summer intensive language study abroad program. Peer/group interaction here refers to one-on-one or group interaction in either the L1 or L2 between American learners in and outside the classroom.

Background Researchers caution that language learning is a rather complex process, and that language educators and researchers should avoid conflating foreign language learning with mere linguistic development (Kramsch, 2000; van Lier, 2004). Compared to traditional foreign language classrooms, the context of study abroad is inherently much more complex and unpredictable. This section reviews previous second language acquisition (SLA) research on peer interaction and relevant theoretical propositions such as language socialization, language (L1 and L2) and identity, which will be drawn on to help understand how informal interactions with L1 peers may shape the language development of American learners of Mandarin Chinese in a study abroad context and how these learners view the process.

Peer interaction in SLA In cognitivist SLA theories (e.g. Doughty & Long, 2003), L2 knowledge is perceived as being transmitted directly from an expert of the language, preferably a native speaker, to a novice learner through meaningful interaction so that comprehensible input is absorbed and eventually imitated by the learner. From the perspective of sociocultural theory (SCT) (Vygotsky, 1978), learning originates and is co-constructed in dynamic and mediated social interactions in which the novice learner is an active rather than passive agent who actively controls how to engage in learning activities. Although early SCT researchers emphasized asymmetrical learning in expert–novice collaboration, more recent studies in SLA (e.g. Donato, 1994) have discovered that both lower and higher level learners in symmetrical interaction can make language gains through collaboration. In other words, learners who work with lower level peers can also benefit from collaborative learning activities.

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

59

One line of such research is found in Storch’s studies (2000, 2001a, 2001b, 2002), which distinguish two dimensions of dyadic interaction: equality (i.e. authority over the task) and mutuality (i.e. level of engagement in the task). Four distinct peer interaction patterns are uncovered: collaborative, dominant/dominant, dominant/passive and expert/novice. Storch (2002) asserts that peer interaction among collaborative and expert/novice dyads is more conducive to language learning. Reviewing empirical studies on the impact of peer interaction on L2 learners’ development in writing, speaking, listening and reading skills, Swain et al. (2002) conclude that peer–peer collaborative dialogue has a positive impact on L2 learners’ language learning, regardless of their respective proficiency level; that is because learners who adopt a collaborative orientation co-construct the language and provide scaffolding support for each other. Specifically, during collaborative dialogues, the more proficient learner can improve his or her phonology, syntax and lexis by repeating or recycling the performance, even if the less proficient learner does not provide particular assistance. A recent study (Watanabe & Swain, 2007) further confirms that learners’ proficiency levels do not necessarily affect the patterns of peer collaboration or the results of language learning. In other words, the degree of collaborative peer interaction and its impact on learning are not significantly related to whether learners obtain support from a high-level performer or a native speaker of the target language. These findings indicate that peer interaction can provide as much support as interaction with native speakers of the target language for language development.

Identity and L1 use in L2 learning settings Another line of SLA research analyzes second language learning as a process of learner identity formation, contestation and transitioning (e.g. Belz, 2002; Kramsch, 2000; Pavlenko & Lantolf, 2000). From this viewpoint, language is used as both a communicative and a cognitive tool that provides an additional semiotic repertoire and, consequently, a different lens through which to view the world. L2 learners are perceived as ‘agents whose actions are situated in particular contexts and influenced by their dynamic ethnic, national, gender, class, and social identities’ (Lantolf & Pavlenko, 2001: 155). Acquiring an L2 is equal to acquiring a new tool that may reframe the learner’s views and provide a new voice for the learner to express him or herself through the selection of certain words and meanings over others. Thus, through the language learning process, a new self – hopefully a bilingual and bicultural one – emerges. From this point of view, the goal of foreign or second language learning is to become a ‘multicompetent language user’ (Cook, 1995). In other words, L2 learners should not be viewed as deficient L2 language users who will always be inferior to native speakers of the target language. Rather, they are acquiring a new set of competences that include different cognitive processes

60

Studies in Second L anguage Acquisit ion of Chinese

in a different language in addition to their native one, which renders them completely different from monolingual individuals. Aligning with this line of thought, Kramsch (2004) emphasizes that L2 learners are constructing a ‘third space’ at the intersection of several identities and subjectivities through language rather than seeking to be completely assimilated into the target culture. Belz (2002) argues that L1 use in a L2 learning environment should not be viewed as language deficiency. Instead, the use of multiple languages should be the representations of multiple speaker identities. In certain situations, systematic and well-thought-out L1 use is even considered to be an effective and creative tool that facilitates rather than hinders L2 development (Cook, 2001; van Lier, 2004). For example, learners may feel more comfortable with certain functions or topics in their L1 than in their L2. Thus, it is suggested that foreign or second language educators view L2 learning as an identity (re)configuration process (Pavlenko & Lantolf, 2000), and L1 use should not be avoided at all costs.

L2 socialization and study abroad The surging number of students participating in study abroad programs has attracted increasing attention to L2 learning in various study abroad contexts. One intriguing theory is language socialization (Ochs, 2002), which is derived from theories on L1 children’s language socialization in their native culture. Analogizing L2 learners to L1 children who strive to be socialized in their native community through acquiring appropriate language use, L2 researchers (e.g. Iino, 2006; Kinginger, 2008; Siegal, 1996) are interested in understanding how L2 learners experience various opportunities when interacting with the host community. Unfortunately, existing studies yield more controversial than concerted findings in terms of what contributes to successful language learning in a study abroad context. However, one widely accepted consensus (e.g. Kinginger, 2008) is that L2 learners’ language gains from studying abroad depend not only on how they are accepted in the host community, but also on how they opt to position themselves in these communities, or on their agency (van Lier, 2004). In other words, whether both the host community members and the learners are willing or ready to socialize the learners into one or more L2 communities may affect the quality of social interactions, which is consequential to the success of language learning in study abroad. Although the theory of language socialization is mainly used to explain home-stay L2 learners’ language learning in a sojourn, it provides an instrumental approach to understanding how native speakers’ efforts to socialize L2 learners and L2 learners’ stance toward being socialized into the host community intertwiningly influence this identity reconfiguration process. It is widely acknowledged that study abroad entails a very complex L2 learning environment comprised of a myriad of social, cultural and individual

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

61

factors that shape the learning process. A focus on either the social cultural factors or on learners’ psychological states alone would yield only a one-sided understanding of this complex process. Currently, due to logistical constraints and programming concerns, many American students studying abroad in China are enrolled in summer intensive programs that require students to share rooms with another American peer, and they are therefore not fully immersed in the local Chinese language environment. Despite the increasing recognition of the learning conduciveness of peer interaction in foreign language learning contexts, maximum elimination of L1 peer interaction is still a widely preferred pedagogical strategy. Inspired by relevant SCT theoretical constructs such as identity, agency and language socialization, this study aims to understand how American learners of Mandarin Chinese in a study abroad program interact with their American peers in their L1 or L2, and how they perceive these interactions in relation to their language development. Considering the situation of students inevitably having both one-on-one and group interactions in a study abroad program, the current study investigated peer interaction behaviors in both dyadic and group interactive activities. Two specific questions guided the research: (1) How do American learners of Mandarin Chinese interact with their peers throughout the study abroad program? (2) How do American learners of Mandarin Chinese perceive the role of peer/group interaction in their Mandarin Chinese learning during the study abroad program?

The Study Study context and participants The current study, which is part of a larger scale study, was conducted in the summer of 2010 in an intensive language program hosted by a prestigious university in Shanghai. The program spanned eight weeks from late June to early August (seven weeks of intensive formal language instruction in Shanghai and one week of tour in Beijing). A total of 12 students participated in this program. Each student was required to have at least one year of Chinese language learning experience before going abroad. Before departure, they were required to take a cultural preparation course (a total of 21 hours of class contact time) to develop a basic understanding of modern Chinese society, such as the diversity of social and linguistic norms in various regions of the country and how modern Chinese people view issues like gender. After arriving in Shanghai, students were divided into two classes: the 2ndyear class (five students), was for students who had finished one year of Chinese learning in the US, and the 3rd-year class (six students) was for

62

Studies in Second L anguage Acquisit ion of Chinese

those who had finished two years. One student who had finished three years of Chinese at the home university was placed in the 3rd-year class but was provided with a supplemental textbook and additional instruction outside of the normal class time. Thus, the 3rd-year class ended up with seven students. From Monday through Thursday, each class met for three grammar sessions (45 minutes per session) in which a professor from the host institution would go through the texts in the textbook and explain important grammatical structures. In the afternoon, each class participated in two language practicum sessions (45 minutes per session) during which a graduate student from the host institution organized various speaking and writing activities and helped students practice what they had learned in the morning. All classroom-based sessions ended at 15:10 every weekday. All the professors and teaching assistants were native speakers of Mandarin Chinese. Each student in the program was assigned a native Chinese tutor. Every weekday after all classroom-based sessions, students met their tutors individually for one hour to practice freely speaking Chinese or to discuss cultural issues of personal interest. Every Friday morning there was a review and test session in which students were assessed on their weekly syntactical and lexical gains in spoken and written Chinese. The group also made three one- or two-day trips to nearby attractions at the weekends. All the students in this program stayed in an off-campus hotel within walking distance of the university campus. All except for one male and one female shared the room with another same-gender student. Because of the class structure and the hotel room arrangements, students in this program had ample time to interact with each other on weekday evenings and on weekends. Participants were recruited once they were accepted into the program in the spring of 2010. Eleven out of the 12 students agreed to participate in the study. They did so voluntarily and were allowed to withdraw from the study any time during the eight-week period. The researcher was also the director of the summer program. During that time, she was handling logistic issues in addition to conducting the study. The entire group met with the program director for a group dinner every Friday evening. As shown in Appendix A, five of the participants were female, including two Asian Americans, two Caucasians and one half-Asian-and-half-Caucasian. Six were male, including one Asian, four Caucasians and one halfAsian-and-half-Caucasian. All participants were between 18 and 21 years old at the time of the study: five were enrolled in the 2nd-year class and six in the 3rd-year class. Among the three Asian American participants, two were Cantonese heritage learners whose proficiency in literary Chinese was similar to the non-Asian participants, although their speaking skills were slightly better than those of their non-Asian peers. They both reported that they had visited relatives or family in Hong Kong and Guangzhou a few times before this trip. The other Asian American participant was a Mandarin heritage

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

63

learner whose conversational and literary proficiency was better than the entire group as she had finished three years of classes before the trip and had visited family in Beijing during summer breaks. One Caucasian student had visited China on a two-week tour three years earlier, but had minimum interaction with local people on the tour. All other participants had never been to China before. Both half-Asian-and-half-Caucasian participants had minimum exposure to Chinese language and culture at home.

Data collection and analysis A case study approach was adopted for this study. Qualitative data were collected from a pre-study survey, weekly interviews, informal observations and participants’ weekly reflective journals. A pre-study survey (see Appendix B) was administered at the beginning of the study to understand each participant’s demographic background, prior experience with Chinese language learning, and their goals in the study abroad program. A weekly 15–20-minute semi-structured interview (see Appendix C for sample interview questions), the primary data source in this study, was conducted with each participant every Tuesday or Wednesday evening at a time and location convenient for the participants. The interview was intended to help each participant recollect their weekly peer or group interactions and their perception about those interactions for the purpose of language learning. Informal observations were also conducted during group activities such as meals and outings to provide triangulating data about each participant’s behavior in peer or group interaction. The frequency and length of each observation varied depending on the group meeting format. As students in the study abroad program were required to write weekly reflective journals, these journals were also collected to glean information pertaining to their reflections on peer/group interaction. The researcher also kept a reflective journal to record thoughts stimulated by observations and other relevant information. Once the pre-study survey results were collected, all information from the survey was compiled to assist with the data analysis. Data from qualitative data sources (interview transcripts, observations and documents) were analyzed in two phases: within-case analysis and cross-case analysis (Yin, 2003). Each participant was considered as one case. Two qualitative methods, constant comparison method (Lincoln & Guba, 1985) and matrix display (Miles & Huberman, 1994), were employed to organize and decipher data from each case. In the within-case analysis phase, data were preliminarily analyzed each week as they became available. Within-case analysis was conducted generally in five steps. (1) One participant’s interview transcripts as well as the researcher’s reflective journal entries were inspected for a general view of how participants behaved in peer interaction.

64

Studies in Second L anguage Acquisit ion of Chinese

(2) Relevant raw data were grouped under recurring themes, such as the types of peer interaction and the circumstances under which the interactions occurred. Constant comparison method was used to avoid redundant or mislabeled themes. (3) Each participant’s weekly blog entries were reviewed for triangulating or additional information pertinent to peer/group interaction. (4) A scheme was created for each recurring theme. The scheme in the first within-case analysis was used to guide the data analysis in the other 10 cases, which yielded a matrix for peer/group interaction behaviors (see Appendix D). The same within-case analysis procedures were followed each week with each new set of data. All data were recorded in the matrices. (5) Finally, a diachronic analysis was conducted to investigate how each participant interacted with their peers throughout the program. In the cross-case analysis phase, a synchronic analysis was conducted to compare and contrast each participant’s peer/group interaction behaviors and identify the differences and similarities.

Results and Discussion How did American learners of Mandarin Chinese interact with their peers throughout the study abroad program? The results revealed interesting grouping and interaction patterns among the participants throughout the program.1 Based on the diachronic analysis, the 2nd-year participants (Janie, Jay, Dave, Matthew and Brett) exhibited three distinct and sequential phases (see Appendix D for a detailed chart) of grouping and interaction patterns, while the 3rd-year participants (Amy, Andrew, Annie, Cathy, Emma and Mike) displayed two phases. Both the 2nd-year and 3rd-year participants demonstrated similar grouping and interaction patterns in phase 1 (weeks 1–2 of the study abroad program). In this phase, they had frequent group outings including shopping and eating out after class. Participants reported more frequent group as opposed to one-on-one interactions, which were characterized as: (1) collective conversations with local Chinese; (2) lower level participants’ conscious observations and imitation of higher level peers’ language use; and (3) constant knowledge sharing of living and language-learning tips. In terms of collective conversations, two types of group behaviors were discerned. The first was that everyone in the group contributed, such as filling in missing vocabulary or correcting ungrammatical sentence structures whenever a conversation took place with a local Chinese. The second type was that one conferred with peers before conversing with native speakers. For example, Mike, a 3rd-year participant, explained that ‘When we interact with people outside of our group and we have to

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

65

speak in Chinese, we will help each other if one person understands what someone is saying and someone else does . . .’2 (interview with Mike on 1 July 2010). The following excerpt illustrates how lower level learners used higher level learners as a language resource to check the appropriate tone for niú (beef): (Scenario: At the Noodle Bowl the group frequented. Andrew wanted to order a bowl of beef noodle.) Andrew: 我要. . . (woˇ yào, ‘I want’) (looking at the Chinese menu, and turning to a peer standing next to him) Andrew: niu ¯ ròu (‘beef’)? Mike: niú ròu (with a clearly pronounced and deliberately prolonged tone for niu) Andrew (turning to the cashier): 对，我要牛肉面 (duì, woˇ yào niú ròu miàn. ‘Yes, I want beef noodle’). Lower level participants also tended to observe and imitate higher level peers’ language use. For example, Emma, who lacked confidence in her Chinese, reported that ‘I was not sure of many things. So most time I just let them speak to the native speakers. It helped me a lot by just listening to their conversations. I learned many vocabulary and expressions I didn’t know before’ (interview with Emma on 7 July 2010). Participants constantly shared living and languagelearning tips in this phase. Emma said, ‘Mike told us where to find a clean and affordable massage shop. I told them where to find fresh vegetables. It is nice.’ (interview with Emma on 7 July 2010). Janie recalled, ‘when we were together, we always talked about how to learn Chinese, like how to memorize characters. I told them the stories I created to help me remember characters’ (interview with Janie on 1 July 2012). Phase 2 (Weeks 3–5) for the 2nd-year participants was characterized by frequent smaller group activities while whole group outings subsided. Participants’ interaction patterns included (1) a larger amount and higher frequency of L2 use in group conversations, and (2) using each other as learning resource in and out of class. In this phase, the 2nd-year participants seldom went out with their 3rd-year peers or as a whole class. Instead, they split into groups of two or three. In terms of more L2 use, Jay noted that ‘Janie and I always used some simple Chinese when we had lunch together or studied. I feel like we talk to each other more in Chinese and we practice more’ (interview with Jay on 14 July 2011). Dave reported that ‘We’ve increased the amount that we’ve been speaking and I’ve been speaking and I’ve been trying to speak Chinese even to people in our group. For example, Andrew and I went to dinner Monday and we tried to speak Chinese most of it until we got . . . because we started out . . . Chinese vocabulary is good and we were sitting in talking about our days. We tried to put some of it Chinese and some of it worked and some of it didn’t’ (interview with Dave on 21 July 2010).

66

Studies in Second L anguage Acquisit ion of Chinese

They also frequently met in small groups to prepare for classes and tests, focusing on the grammatical structures from the textbook: ‘As we get further into the chapters, we talk to each other more because the lessons and words are more difficult and we help each other with the grammar a lot’ (interview with Jay on 14 July 2010). Dave, the self-reported slowest learner in the class, in particular benefited from the group help. According to him, ‘we’ve been doing some studying together as a group – with Janie and Jay and Amy has been a big help with the grammar and stuff that I don’t understand. I find that I use Amy a lot for some of the things that I cannot communicate with my tutor about. If one of us has a question, we always put it out there and have a kind of think-tank together’ (interview with Dave on 21 July 2010). The participants often turned to each other for help, even in the presence of their Chinese instructor. During one class observation, the researcher noticed that Dave was confused about the phrases 回去 huí qù ‘go back’ and 回来 huí lái ‘come back’ presented in Chinese by the professor. He turned to Jay who was sitting next to him for help. Jay explained the concepts in English. Dave later reflected, ‘until that moment, I didn’t understand why Chinese use one verb followed with different suffixes to mean two actions’ (interview with Dave on 21 July 2010). Phase 3 (weeks 6–7) featured much less group interaction and more individual activities. In this phase, most 2nd-year participants started to explore the city individually or with their respective tutors. For example, Dave commented that ‘I’ve been trying to do a little more studying on my own and I’ve been kind of separating myself a little bit more to be able to use my Chinese . . . the idea was – it gave me a hour practicing my Chinese and then it gave me about two hours of time where I kind of practiced with people’ (interview with Dave on 27 July 2010). Both Jay and Janie reported that they spent more time with their own tutors during the weekdays. However, despite more outings with native tutors, the 2nd-year participants still sought help from their peers occasionally, usually in the classroom. Brett explained that ‘a lot of it was Janie or Jay. They were able to describe that we didn’t understand if Zhou Laoshi said something that we didn’t understand or he couldn’t explain any better . . . or sometimes he would try to say something and he wouldn’t know the English word and he would look to Jay hoping Jay to understand . . .’ (interview with Brett on 13 August 2010). Although the participants became more immersed in the target language environment, they continued to rely on their peers, particularly those with higher proficiency, to overcome communication breakdown in the classroom. In contrast to the 2nd-year participants, the 3rd-year participants entered their phase 2 (weeks 3–7) soon after they became acquainted with the new environment. Their phase 2 was characterized by more one-on-one activities, either with a classmate or with their respective tutor, and more immersion in the local community. Mike repeatedly mentioned in the interviews that ‘Pretty much I just go to school and come back – like my tutor helps me with my interactions but not really anyone in the group’ (interview with Mike on 21 July 2010)

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

67

and ‘I don’t really spend much time with them during the week because I am usually in my room studying or doing homework or going out with my tutor. I occasionally talk to Jack (his roommate) if I have questions about our homework’ (interview with Mike on 28 July 2010). Emma said in week 3, ‘I think we are all fairly independent more – we are becoming more independent and we don’t really need as much’ (interview with Emma on 13 July 2010). The 3rd-year participants also offered more voluntary language help to lower level peers when needed. For example, Amy, who was a heritage speaker and had the highest language proficiency in the group, frequently remarked that ‘I’ve been helping a bunch of people because like when Janie and Jay and I hang out or like if we are with a bunch of people studying together as a group, I know I like to correct a lot of people’s pronunciation or like . . . I am correcting tone or pronunciation – sometimes I correct their sentences every now and then. I know Dave always asks me for like meanings of words’ (interview with Amy on 13 July 2010). In sum, the results showed that language proficiency and familiarity with the environment affected the participants’ grouping and interaction patterns. At the beginning of the program, all participants were unfamiliar with the linguistic and cultural environment. They relied on one another to learn about the new environment and to interact with native speakers. As soon as they became acquainted with the environment, the 3rd-year participants tended to immerse themselves in the local community. The 2nd-year participants split into smaller groups and gradually immersed themselves in the local community by interacting more with their tutors. However, peers continued to be frequently consulted as language resources in and outside the classroom. Interestingly, the results also indicated patterns of either collaborative or expert/novice interaction rather than dominant/dominant or dominant/passive interaction (Storch, 2000, 2001a, 2001b, 2002). In other words, the peer interactions seemed more conducive to each participant’s language learning. In particular, it was evidenced that lower level participants offered each other affective, linguistic and metalinguistic help throughout the program, which resonates with earlier researchers’ observations in regular foreign language classrooms (Cook, 2001; Van Lier, 2004).

How did American learners of Mandarin Chinese perceive peer/ group interaction while studying abroad? The results revealed the participants’ perception about peer/group interaction during the study abroad program was shaped by their language proficiency and shifted according to their familiarity with the environment. The diachronic analysis yielded two sequential phases that the vast majority of the 2nd-year and the 3rd-year participants experienced throughout the study abroad program, although the time span of each phase differed slightly for

68

Studies in Second L anguage Acquisit ion of Chinese

each language group. It is noteworthy that Matthew, the half-Asian-andhalf-Caucasian student in the 2nd-year class, again demonstrated a different perspective from the rest of the group. He maintained the perspective that ‘we are just equally clueless and in the same boat so I think it’s kind of most like anything culturally you are just kind of finding out independently’ (interview with Matthew on 7 July 2010) and ‘The best interaction that should be focused on is to try to interact with the people as much as you can. I mean that’s the most important thing because that is where you are going to hear true Chinese’ (interview with Matthew on 14 July 2010). In phase 1 (weeks 1–3), all five participants except for Matthew as mentioned earlier made predominantly positive comments on peer/group interaction in relation to their language learning and daily life during the study abroad program. For example, Dave described his peers as a ‘comfort blanket’ or ‘safety net’ (interview with Dave on 6 July 2010) because ‘it is less embarrassing to ask them for help – also if none of us get it, it is kind of comforting that we are not the only one that is struggling with it’ (interview with Janie on 7 July 2010). Brett elaborated on the importance of interacting in L1 with his peers after class: ‘If we were to study for six hours and then come back and be with a host family and have to speak it even – I think it would be way more frustrating. Especially for only being a 2nd-year and it would really put a lot of pressure on people because I know some people are already struggling with missing family and friends and stuff and stressing out about the test every Friday. So if you were to come back and not be able to release and hang out and talk in English, then it would be way too stressful and these seven weeks of studying way too intensive’ (interview with Brett on 1 July 2010). After three weeks of experience in China, all 2nd-year participants entered phase 2 (weeks 4–7), in which they weighted L1 or L2 peer/group interaction less significantly and started to value direct interactions in Chinese with their teachers, tutors and local residents. The data showed two salient views about peer/group interaction: (1) peers provided better grammatical explanation, and thus complemented the teacher’s grammar instruction; (2) peers should speak more L2. In terms of peers’ helpfulness in grammar learning, Dave explained that ‘grammar-wise I am very weak and so it is nice to be able to talk to people and have them help out with what I am missing and I find it much easier learning grammar from people who speak English because I can learn formula of it. When I learn it from a person who is Chinese or I read it in a book, sometimes it is not written in a way that I understand’ (interview with Dave on 14 July 2010). Even Janie, the top student in the 2nd-year class, agreed with this view, ‘Because my classmates are also English speakers, they understand if I get hung up on something or . . . they were much more likely to understand why I didn’t get something and to help out for myself’ (interview with Janie on 13 August 2010). The 2nd-year participants also noticed the pitfalls of spending too much time with peers and speaking L1. Jay admitted that ‘I feel like at this point we could use a little more Chinese practice with each other during class and

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

69

outside of class because we still primarily speak English to each other’ (interview with Jay on 14 July 2010) and in week 5, he realized that ‘it is slightly better to talk with the tutor because it forces me to use more Chinese – because when I talk with them I can occasionally slip in English and then with the tutor – they can correct me with a lot of things and when we are going outside somewhere with the tutor, I try to tell them my thoughts in Chinese . . .’ (interview with Jay on 27 July 2010). In phase 1 (weeks 1–2), all 3rd-year participants also felt mostly positive about their peer/group interaction, particularly in their L1, because of two reasons: (1) peers helped with language when communicating with local residents as shown in the previous section; and (2) L1 interactions with peers help to relax the mind, or as Mike put it rather bluntly, ‘I’d say it is a good thing to have a group here because if it was just me and every day I was to take Chinese and only talk in Chinese, I would want to kill myself. I would go crazy because I wouldn’t have anyone to just relax and talk to’ (interview with Mike on 1 July 2010). Compared to the 2nd-year participants, all 3rd-year participants entered their phase 2 (weeks 3–7) sooner. In phase 2, they expressed three salient views about peer/group interactions: (1) peers were too distractive and unhelpful with language learning; (2) peer interaction was necessary when a break was needed at weekends; and (3) helping lower level peers was fun and helpful for their own language development. There existed a consensus among all six 3rd-year participants that peer/group interaction was not helpful, and sometimes even hindered language learning in this phase. For example, Emma disclosed that ‘I actually enjoy going out by myself a little bit more than with the group because it forces me to speak more than when I’m with everyone else in the group’ (interview with Emma on 13 July 2010). Mike felt he gained more help from native speakers as he became more confident in his language proficiency, explaining that ‘most help . . . I’d say just from the teachers – mostly the teachers because practice – that is pretty much all you need . . . my language has gotten better so I don’t need any more help from my peers/group members’ (interview with Mike on 28 July 2010). However, the majority of 3rd-year participants (Mike, Amy, Andrew and Cathy) did not object to the idea of occasionally hanging out with their peers on weekends because ‘it relaxed me when we went to bars on weekends. Speaking Chinese all the time would make me feel like brain fried’ (interview with Mike on 13 August 2010). On the other hand, Amy, who constantly helped the 2nd-year peers, felt offering language help was interesting and useful for her own language learning: ‘I’ve found that it is fun speaking with them in Chinese’ (interview with Amy on 27 July 2010). She also acknowledged that ‘helping them helped me a lot because I’ve always heard the saying “You know one thing, but teaching it is another”. I always found that because they started questioning me, I found myself questioning the things that I learned not so much as if it was right but more why it was right or like why it was this way and I found that it was kind of neat. I mean like it made me strive to learn more about the language and excel more in the language so that I could

70

Studies in Second L anguage Acquisit ion of Chinese

help them in turn . . . and yeah it did help review a lot because I knew there were some questions that I should have been able to answer and then sometimes I would be like, “wait a minute . . . is this correct?”’ (interview with Amy on 14 August 2010). There was one perception that was frequently mentioned by all of the participants, although they didn’t relate it to their language learning, namely their view of the study abroad group. Since the day they arrived in Shanghai, all of the participants realized that they had formed a unique support group that was different from the Chinese people living around them and that would not have existed if they had stayed at their home university in the US. In week 1, Amy recognized that ‘When we combine as a group together, it is kind of like a nice taste of America again’ (interview with Amy on 1 July 2010), an observation upon which Jay had his own take: ‘I like how we are all one big community’ (interview with Jay on 1 July 2010). The sense of a unique group emerged and was recognized among the majority of the participants, regardless of their language proficiency and heritage. For example, in week 7, Andrew, a 3rd-year participant with a Semitic background, reflected that ‘Because we were in China, we were our own little community and in our own little niche. At DePaul, all we did was go to class and then after we all left and did our own things and I didn’t really know any of them. And now we are all really good friends and we feel really close to each other so it’s just like helping a buddy out now. Because we only have ourselves to rely on’ (interview with Andrew on 13 August 2010). Dave, a 2nd-year participant with a Caucasian background, also wrote that ‘the thing I am going to miss most are the people who have cared the most for me and spent the most time with me over this trip: my new family. I say miss with a caveat because I sincerely hope that our group stays in touch over the coming years. We might not have felt like the most close-knit group when we first came together. But friendships have developed immensely since the start of the program and I find myself close to people that I might have otherwise never talked to. At first I thought it would be no big deal leaving the hotel and not seeing everybody as often. But the farther and farther I get from A8 lüˇ gua ˇn, the more I realize how much my family meant to me.’ (Dave’s reflective journal #7). This group identity appeared even more meaningful to Cathy and Jay, two American-born Chinese, as they did not feel they were fully accepted by the local Chinese community due to their appearance coupled with relatively low L2 proficiency, a feeling that Cathy expressed in her journal: ‘Whenever I try to communicate with the people here and when they do not understand me because of either my tones or simply because I do not know how to answer them they would give me a weird look as if they are questioning why can I not communicate with them when I clearly look like I should be able to.’ (Cathy’s reflective journal #1). At one point, Cathy even worried that they were losing part of their collective American identity because the group was speaking too much Chinese. She wrote that ‘maybe a couple of days ago in class – someone was trying to say something in English or spell something and kind of spelled it wrong. It seems we are losing our English skills. It seems like it.’ (interview with Cathy on 13 July 2010).

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

71

However, this new group identity was not necessarily embraced by all participants in this program. For example, in week 1 Matthew outlined his plan to immerse himself in the local culture: ‘I mean, we come to China, we want to learn the language and culture. I want to be able to blend it. The only way I can do it is to talk to more Chinese people and learn how to behave like a native speaker. I usually just go to talk to those street vendors and ask about their life.’ (interview with Matthew on 1 July 2010). This belief in the importance of immersion language learning may explain why Matthew consciously and constantly distanced himself from other group members during the program. On the other hand, Emma seemed to have grown out of this new group because of her familiarity with the environment, her increased confidence in her Chinese abilities, and her tutor’s acceptance of her as a friend. The researcher observed that at the beginning of the program Emma relied heavily on the other group members when going out, trying to observe and learn from her peers when they interacted with native speakers. By the middle of the program, she spent more time with her tutor. She noted that ‘Now I am comfortable with her. Today she called me xiao meimei (little sister). I feel pretty comfortable when I am alone by myself, away from my group members. Everybody is so busy now.’ (interview with Emma on 13 July 2010) and ‘my tutor is definitely my close Chinese friend and because of the nature of the program, we have learned a lot about each other . . . Occasionally she will bring me ice cream and slowly I have started to notice her soft side. When we went as a group to qipu lu, I felt she took the role of a sisterly figure. Everywhere we went, she tried to hold my hand or put her arm around my shoulders to make sure I wouldn’t get lost in the sea of people’ (Emma’s reflective journal #4). It thus appears that recognition of the new group identity also depended on the learners’ personal goals in the study abroad program (also known as agency), as well as on their social acceptance in the host culture. The findings presented above suggest that participants’ dependence on peer/group interaction evolved with their language development and increased familiarity with the study abroad environment. So did their perceptions regarding the importance of peer/group interaction in language learning and daily life. In general, the more confident they felt about their language proficiency, the less help they sought via peer/group interaction, and the less they valued the interactions with their peers. Lower level participants felt peers with the same or higher level proficiency provided important affective, linguistic and metalinguistic support throughout the program, although the dependence receded gradually. While higher level participants soon moved out of the comfort zone that was enabled through L1 and L2 peer/group interaction, they still felt L1-based peer/group interaction was indispensable, either because it provided a relaxing break from intensive L2 study or because it helped refresh their language memory. This finding resonates with those in previous studies (e.g. Swain et al., 2002; Watanabe & Swain, 2007). Moreover, the peer/group interaction patterns reported in the

72

Studies in Second L anguage Acquisit ion of Chinese

study indicate that more collaborative and expert/novice interactions occurred as part of peer/group interactions, which is believed to be conducive to language learning (Storch, 2000, 2001a, 2001b, 2002). On the other hand, the findings illustrate the complexity inherent in a study abroad context due to the intertwining influence of various social cultural factors such as identity, agency and L2 socialization. It is unrealistic to expect all learners to forget about their American identity when they are in the host country. It seems that two situations kept them from becoming immediately and fully immersed in the local community: (1) their lack of confidence in their Chinese language abilities and understanding of the culture; and (2) local residents’ hesitance to socialize with them due to their physical features and lack of language proficiency (Kinginger, 2008). This made it necessary for the learners to form their own community within the group before they felt confident enough to reach out to their tutors and the local residents. At that point, the tutors’ willingness to socialize with the learners seemed particularly helpful, as in the case of Emma. Interestingly, despite their increased language proficiency and greater comfort when communicating with their tutors and other local residents toward the end of the program, many participants still preferred peer help with certain aspects of language learning and planned to maintain connections with the group even after the study abroad program had ended. It may be far-reaching to claim an emergence of a new bilingual and bicultural identity (Kramch, 2004; Pavlenko & Lantolf, 2000), but the participants forged a special bond with one another that would otherwise never have occurred, had it not been for the shared experience. This special group identity may mark the beginning of a transitional phase where learners shift from a monocultural identity to a full-fledged bicultural identity, if the immersion experience continues on. In sum, the findings of this study show that peer/group interactions in both L1 and L2 played a viable role in the study abroad program, particularly for lower level learners at the beginning of the program. Learners’ language proficiency, identity, the local residents’ willingness to socialize with learners, as well as learners’ agency, all seemed to have affected the participants’ interactions with peers and, in turn, their perception of the helpfulness of peer/group interactions.

Conclusion The ‘sink or swim’ strategy has long been taken for granted in study abroad programs. Educators and students tend to believe that learners should be fully immersed in the target language environment from the moment they land in the host country. In this way, they will not waste the resources and opportunities putatively accessible to them. Although the controversial findings from study abroad research (e.g. Kinginger, 2008) have underscored

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

73

the complexity of study abroad contexts, very few empirical studies have been conducted to investigate the interaction among learners themselves. The present study analyzed 11 study abroad students’ peer/group interaction patterns as well as their perceptions of peer/group interactions during a study abroad program. Participants’ shifting peer/group interaction patterns and evolving perceptions of the importance of peer/group interaction suggest that L2 learning in a study abroad context is a very complex process shaped by a myriad of intertwining factors. Further research is warranted to examine the impact of these factors alone or in combination on students both as language learners and social beings. Recent years have seen an increasing number of American students studying abroad in China. For those who are enrolled in school-run intensive programs and live with their American peers, peer interaction is inevitable. Whether and how much they should speak Chinese during the program has been an acute concern for administrators and educators. It is, therefore, hoped that the present study will provide a spin-off for more research on this topic. To that end, two major limitations of the study must be noted. First, the data were only collected from 11 participants, and the program model under investigation was only eight weeks long. Thus, the findings of this study may not be generalized to other program models that last longer and have more students, which could significantly change the group interaction dynamics. Second, the primary data sources of the study were confined to interview transcripts and journal entries, which indirectly rather than directly shed light on peer interaction. Research on L2 acquisition of Chinese in a study abroad context is still very much an uncharted area (see Chapter 5, this volume). As a rapidly emerging study abroad site, China, compared to other Western countries, has a long history of being a monocultural society. Many Chinese citizens are still not used to seeing and living with people who have different physical features. Thus, aside from the need for more studies to investigate peer/group interaction and its potential language gains, it would also be interesting to examine how study abroad students and native Chinese people handle their interactions. In addition, the unique socialization issues encountered by heritage and mixed-race learners in a study abroad context merit systematic investigation.

74

Studies in Second L anguage Acquisit ion of Chinese

Appendix A: Demographic Information About the Participants Language proficiency

Gender

Ethnicity

Prior experience in China

Janie

2nd-year

F

Caucasian

Jay

2nd-year

M

Asian

Dave Matthew

2nd-year 2nd-year

M M

Caucasian Asian/ Caucasian

Brett Mike Emma

2nd-year 3rd-year 3rd-year

M M F

Caucasian Caucasian Asian/ Caucasian

Annie

3rd-year

F

Caucasian

Andrew Cathy

3rd-year 3rd-year

M F

Caucasian Asian

Amy

3rd-year

F

Asian

Toured in Taiwan for two weeks years ago. Heritage learner, Cantonese speaker, has family in HK, never been to mainland China. Never been to mainland China. Minimum exposure to Chinese language and culture at home, never been to mainland China. Never been to mainland China. Never been to mainland China. Minimum exposure to Chinese language and culture at home, Never been to mainland China. Visited China on a two-week tour in high school. Minimum contact with native speakers of Chinese on the tour. Never been to mainland China. Heritage learner, Cantonese speaker, has been to HK and Guangzhou for occasional family visits. Heritage learner, Mandarin speaker, has been to Beijing for family visits in summers.

Appendix B: Pre-Program Survey Dear students, thank you for participating in this study. The information you provide here will help the researcher understand your educational background and your expectation for enrolling in the study abroad program. (1) Your major(s): _________________________ (2) Your minor(s): _________________________

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

75

(3) Are you a heritage speaker of Chinese? □ Yes. If so, which dialect do you speak at home? □ Mandarin □ Cantonese □ Other dialect □ No. If no, what is your race? ________________ (4) In which year are you (in this upcoming Autumn quarter) at DePaul University? □ Freshman □ Sophomore □ Junior □ Senior □ Other, please explain: ___________________________ (5) For how many months have you been learning Mandarin? □ more than 36 months □ 24–36 months □ 12–24 months □ less than 12 months (6) Have you been to mainland China before? □ Yes If yes, for how long? ____________________ If yes, for what purpose? _____________________ □ No (7) What do you expect to learn during the study abroad program? _____________________________________________________________ _____________________________________________________________ _____________________________________________________________

Appendix C: Sample Interview Questions (1) Did you go out with one or more of your group members? When? (2) Can you describe what you and your peers did and said when you were together? (3) Did you use Chinese or English when you talked to your peers? (4) Do you think your interaction with your peers in the past one week helps your Chinese learning? If yes, in what ways? If no, why do you think so?

76

Studies in Second L anguage Acquisit ion of Chinese

Appendix D: Peer/Group Interaction Behaviors 2nd-year learners

3rd-year learners/Confident learners

Phase 1: Weeks 1–2 • Mostly group outings (shopping and meals) – Group efforts when conversing with native speakers of Chinese: fill in vocabulary and sentence structures – Confirming with peers ‘how to say this?’ before conversing with native speakers – Lower level students observing higher level peers • Sharing language learning and living tips (grammar learning, memorizing characters, restaurants, body massage studios) Phase 2: Week 3–7 Phase 2: Week 3–5 • More L2 use with the local • Smaller group outings with residents such as their respective occasional L1 use tutors – Exchange cultural observations • More voluntary language help to • Study together lower level peers when needed – Prepare for tests • Help lower level students with – Study grammar, vocabulary homework, give language examples – Speaking Chinese with each other Phase 3: Week 6–7 • Individual or w/tutor outings • More L2 use with peers • Help grammar in class or when needed

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

77

Appendix E: Perceptions About Peer/Group Interaction 2nd-year learners

3rd-year learners/Confident learners

Phase 1: Weeks 1–3 • ‘Family’, ‘comfort blanket’, ‘more affective support’ • ‘Less embarrassed by oneself’ in group interaction with native speakers of Chinese Phase 2: Weeks 4–7 • Native speakers and peers complement each other – Teachers more knowledgeable and native speakers for speaking – Peers explain grammar better and understand the mistakes • Convenient and supportive

Phase 1: Weeks 1–2 • Peers helpful when communicating with native speakers in L2 • ‘Relaxation’ with peers in L1

Phase 2: Week 3–7 • L1 with peers are distracting, prefer self-exploration of the environment or with tutors • Peers for hanging out with at weekends or night time when in need of a break

Notes (1) Among the participants, Matthew, a half-Caucasian-and-half-Asian student in the 2nd-year class, was the only one who constantly avoided group activities since the day he arrived to Shanghai and chose to go out individually or with his tutor. Thus, the grouping and peer/group interaction patterns presented in this section apply to the other 10 participants in the program. (2) All quotations are direct interview transcriptions or excerpts from participants’ blogs. To maintain authenticity, grammar mistakes were not corrected.

References Belz, J. (2002) Second language play as a representation of the multicompetent self in foreign language study. Journal of Language, Identity, and Education 1 (1), 13–39. Cook, V. (1995) Multi-competence and the learning of many languages. Language, Culture and Curriculum 8 (2), 93–98. Cook, V. (2001) Use first language in the classroom. The Canadian Modern Language Review 57, 402–423. Donato, R. (1994) Collective scaffolding in second language learning. In J.P. Lantolf and C. Appel (eds) Vygotskian Approach to Second Language Research (pp. 33–56). Norwood, NJ: Ablex Publishing. Doughty, C.J. and Long, M.H. (eds) (2003) The Handbook of Second Language Acquisition. Malden, MA: Blackwell Publishing. Iino, M. (2006) Norms of interaction in a Japanese homestay setting – toward two-way flow of linguistic and cultural resources. In M.A. DuFon and E. Churchill (eds) Language Learners in Study Abroad Contexts (pp. 151–173). Clevedon: Multilingual Matters.

78

Studies in Second L anguage Acquisit ion of Chinese

Isabelli-Garcia, C.L. (2006) Study abroad social networks, motivation, and attitudes: Implications for SLA. In M.A. DuFon and E. Churchill (eds) Language Learners in Study Abroad Contexts (pp. 231–258). Clevedon: Multilingual Matters. Kinginger, C. (2008) Language learning in study abroad: Case studies of Americans in France. Modern Language Journal 92, Monograph. Kinginger, C. (2009) Language Learning and Study Abroad: A Critical Reading of Research. New York: Palgrave Macmillan. Kramsch, C. (2000) Social discursive constructions of self in L2 learning. In J.P. Lantolf (ed.) Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press. Kramsch, C. (2004) The multilingual experience: Insights from language memoirs. TRANSIT 1 (1), 1–13. Lafford, B. (1995) Getting into, through, and out of a survival situation: A comparison of communicative strategies used by students studying Spanish abroad and ‘at home’. In B. Freed (ed.) Second Language Acquisition in a Study Abroad Context (pp. 97–121). Philadelphia, PA: John Benjamins. Lantolf, J.P. and Pavlenko, A. (2001) (S)econd (L)anguage (A)ctivity: Understanding learners as people. In M. Breen (ed.) Learner Contributions to Language Learning: New Directions in Research (pp. 141–158). London: Pearson. Lapkin, S., Hart, D. and Swain, M. (1995) A Canadian interprovincial exchange: Evaluating the linguistic impact of a three-month stay in Quebec. In B. Freed (ed.) Second Language Acquisition in a Study Abroad Context (pp. 67–94). Philadelphia, PA: John Benjamins. Levin, D.M. (2001) Language learners’ sociocultural interaction in a study abroad context. Published PhD dissertation, Indiana University. Lincoln, Y. and Guba, E. (1985) Naturalistic Inquiry. New York: Sage. Magnan, S. and Back, M. (2007) Social interaction and linguistic gain during study abroad. Foreign Language Annals 40 (1), 43–61. Miles, M.B. and Huberman, A.M. (1994) Qualitative Data Analysis (2nd edn). Thousand Oaks, CA: Sage. Ochs, E. (2002) Becoming a speaker of culture. In C. Kramsch (ed.) Language Acquisition and Language Socialization: Ecological Perspectives (pp. 99–120). London: Continuum. Ohta, A.S. (1995) Applying sociocultural theory to an analysis of learner discourse: Learner–learner collaborative interaction in the zone of proximal development. Issues in Applied Linguistics 6 (2), 93–121. Pavlenko, A. and Lantolf, J.P. (2000) Second language learning as participation and the (re)construction of selves. In J.P. Lantolf (ed.) Sociocultural Theory and Second Language Learning (pp. 155–177). Oxford: Oxford University Press. Siegal, M. (1996) The role of learner subjectivity in second language sociolinguistic competency: Western women learning Japanese. Applied Linguistics 17 (3), 356–382. Storch, N. (2000) Is pair work conducive to language learning? The nature of assistance in adult ESL pair work and its effect on language development. Paper presented at the conference on Scaffolding and Language Learning in Educational Contexts: Sociocultural Approach to Theories and Approaches. Center for Language and Literacy, University of Technology, Sydney. Storch, N. (2001a) How collaborative is pair work? ESL tertiary students composing in pairs. Language Teaching Research 5, 29–53. Storch, N. (2001b) An investigation into the nature of pair work in an ESL classroom and its effect on grammatical development. Unpublished dissertation, University of Melbourne. Storch, N. (2002) Patterns of interaction in ESL pair work. Language Learning 52 (1), 119–158. Swain, M., Brooks, L. and Tocalli-Beller, A. (2002) Peer-peer dialogue as a means of second language learning. Annual Review of Applied Linguistics 22, 171–185.

Peer/Group Interac t ion in a Mandar in Chinese Study Abroad Conte x t

79

Van Lier, L. (2004) The Ecology and Semiotics of Language Learning: A Sociocultural Perspective. Boston, MA: Kluwer. Vygotsky, L.S. (1978) Mind in Society. Cambridge, MA: MIT Press. Watanabe, Y. and Swain, M. (2007) Effect of proficiency differences and patterns of pair interaction on second language learning: Collaborative dialogues between adult ESL learners. Language Teaching Research 11 (2), 121–142. Wilkinson, S. (1998) On the nature of immersion during study abroad: Some participants’ perspectives. Frontiers: The Interdisciplinary Journal of Study Abroad 4, 121–138. Yin, R. (2003) Case Study Research: Design and Methods (2nd edn). Thousand Oaks, CA: Sage Publications.

4

Task-based Language Teaching of Chinese in a Study Abroad Context: A Learner Perspective ZhaoHong Han and Joo-oeck Maeng

Task-based language learning (and task-based teaching for that matter) has been a hot topic in second language acquisition (SLA) research over the past two decades, garnering interest from psycholinguistic, sociolinguistic and sociocultural perspectives (Ellis, 2000; Tarone & Swierzbin, 2009). Focusing on the construct of ‘task’, researchers have amply investigated its various uses – as a pedagogical vehicle, a tool of assessment and/or an instrument for data collection (Bygate et al., 2001; Ellis, 2003; Van den Branden et al., 2009a). Nevertheless, there still remain some fundamental neglected issues. One such issue concerns learners as experiencers of tasks – be they pedagogic tasks, measurement tasks or research tasks (Albert, 2011; Breen, 1987; Dörnyei, 2002; Ellis, 2012; Ortega, 2005; Robinson, 2011a, 2011b). Research on task as a pedagogical vehicle, in particular, within the framework of task-based language teaching (Long & Crookes, 1992; Robinson, 2001; Skehan, 1996), has devoted the bulk of its attention to investigating whether or not (and if so, what) task characteristics can be externally manipulated and what fallouts there may be for learning (Robinson, 2011c). In contrast, there has as yet been scant attention given to the task-doer in the process. As a result, the general understanding has been inadequate of the relationship between task, task implementation and task outcome. This chapter attempts to make inroads on the topic of learners as taskdoers. To that end, we will report on a piece of action research1 that surveys learners’ views of a two-week Chinese study abroad program of task-based language teaching (TBLT) and, more importantly, use that as a springboard for us to contemplate further questions for future research. The chapter is hereafter organized as follows. We will first provide the 80

Task-based L anguage Teaching in a Study Abroad Conte x t

81

background of the study. Given that TBLT is relatively new to the field of teaching Chinese as a foreign language, the discussion is purposely somewhat extensive. Then, we will describe the method of the study. After that, the results will be reported and discussed, and a set of questions posed for further investigation.

Background SLA research aimed at improving the efficacy of instructed learning has over the last two decades turned its attention to pedagogic tasks as a basic unit of organization for teaching, learning and assessment, spurring on a new pedagogical paradigm widely known today as task-based language teaching (Van den Branden et al., 2009b). This shift of attention signified a dramatic departure from a view of language as a system of structural elements and, in turn, from a view of language acquisition as quintessentially a process of learning those elements one by one in a piecemeal fashion. Arguments for TBLT have been built around practical observations of failures of discrete teaching, on the one hand, and theoretical and empirical research favoring integrative teaching, on the other (see, for example, Candlin, 1987; Keck et al., 2006; Long & Crookes, 1992; Skehan, 1996, 1998). As such, TBLT has been hailed as the first ever research-based pedagogy.

What is TBLT? TBLT enacts three broad principles that are in line with the current understanding both of the goal of language teaching – to enable learners to use the target language for real communicative purposes – and of the means to achieve it (Long & Crookes, 1992; Van den Branden et al., 2009a). First, teaching should be holistic rather than discrete. In other words, both teachers and learners should be concerned with language as discourse rather than language as discrete structural elements. Second, teaching should be learner centered rather than teacher centered. That is, teachers’ online and offline decisions about what, when and how to teach should be based on the learners’ functional and cognitive needs. Third, teaching should be communication based rather than form focused. In other words, the focus of teaching and learning should be on meaning making in context rather than on manipulating linguistic forms in isolation. The three principles have translated into various concrete proposals for syllabus development and implementation (Ellis, 2003; Long, 2000; Long & Crookes, 1992; Samuda, 2001; Skehan, 1996; Willis & Willis, 2007). What these proposals have in common is the realization that teaching is fundamentally about providing conditions for learning to naturally unfold, since

82

Studies in Second L anguage Acquisit ion of Chinese

learning is ultimately controlled by the learner, not by the teacher (Corder, 1967). Long (2009) states: Learners, not teachers, have most control over their language development. Students do not – in fact, cannot – learn (as opposed to learn about) target forms and structures on demand, when and how a teacher or a textbook decree that they should, but only when they are developmentally ready to do so. Instruction can facilitate development, but needs to be provided with respect for, and in harmony with, the learner’s powerful cognitive contribution to the acquisition process. (Long, 2009: 378, emphasis in original) Clearly, when the relationship between teaching and learning is thus reversed, the role of the teacher changes accordingly: rather than playing the singular role of a lecturer as in a traditional classroom, a teacher in a TBLT classroom has multiple roles to play – an advisor, a chairperson, a monitor, a language guide, a facilitator, a co-communicator and, even, a bystander (Samuda, 2001; Willis, 1996). But if teaching is about creating conditions for learning to occur, what conditions, then, should be created? Thanks to four decades of SLA research, we now have been able to state these conditions with certainty (see, for example, Ellis, 2005; Long, 2009), which boil down to four – input, negotiation, feedback and output – conveniently dubbed INFO (Han, 2007). First and foremost, there needs to be ample input in the learning environment for learners to selectively attend to and process within the limits of their cognitive and linguistic capacity (Gass, 1997; Long, 1996; Sharwood Smith, 1993; VanPatten, 1996). Second, opportunities for negotiation must be present to aid learners’ processing and assimilation of input, and this can occur both between the learner and another interlocutor and within the learner as s/he interacts with (written) input solo (Kenning, 2006). Next in line is feedback: learners, adult learners, in particular, require corrective feedback to help make up for deficiencies in their input processing and output production as they strive to gain a command of the target language (Bley-Vroman, 1989; Doughty & Williams, 1998; Long, 2000; see also Chapter 6, this volume). The last required condition is output. Just as there is a need for rich exposure to input, there is a need for abundant opportunities to produce output in the classroom through which to hone learners’ fluency, noticing and metalinguistic awareness (Swain, 1985, 1995). Differences, nevertheless, remain among researchers concerning certain of the macro and micro aspects of TBLT. At the macro level, there is, for example, not yet a full consensus on the role of planning: Should teaching be planned ahead of the game, via a pre-determined syllabus, or should it all be about improvisation, that is, performed on the spur of the moment? This difference mirrors a larger, underlying conceptual divide between a psycholinguistic approach and a sociocultural approach to task-based teaching (Ellis, 2000), the former emphasizing planning, individuals’ cognitive

Task-based L anguage Teaching in a Study Abroad Conte x t

83

processing, the need for input and output, and acquisition, and the latter putting a premium on improvisation, dialogic mediation, scaffolding, intersubjectivity and language use. At the micro level, a conspicuous lack of conceptual unity pertains to the role of focus on form in the TBLT classroom. Even within the psycholinguistic camp, where the general sentiments are pro-intervention, views differ as to whether an external focus on form (i.e. as carried out by a teacher to direct learners’ attention to grammatical features of the target language) should be done explicitly or implicitly, and/or whether or not it should be done immediately or delayed after the task is completed. Skehan (1996), for example, was concerned that, due to its primary focus on meaning, TBLT may risk undermining interlanguage development, and hence SLA, the essence of which is construction of a linguistic system that can be effectively exploited for communicative purposes. Skehan writes: Tasks themselves, given their defining properties of meaning primacy, outcome evaluation, and realism, may well predispose those engaged in task completion to engage in a mode of communication which does not prioritize a focus on form, either in terms of using linguistic elements to achieve precision or to achieve accuracy. As a result, it may not be possible to rely on a task-based approach to automatically drive interlanguage forward. (Skehan, 1996: 87) Driven by his conviction that attention to form is necessary for interlanguage development (cf. Schmidt, 1990, 1995, 2001), Skehan recommends a pedagogical sequence for implementing tasks in a TBLT classroom that begins with a ‘preemptive attention to form’ through consciousness-raising and planning, the goal being to establish target language and reduce the cognitive load. The next phase of the sequence involves subjecting learners to the task itself, where meaning is front and center, and where the goal is to mediate fluency – and, to a lesser extent, accuracy – achieved through strategies such as task choice and manipulation of pressure. Once the task is completed, learners’ attention is then shifted away from meaning or fluency back to form or language, with a view to honing in on their accuracy and propelling further restructuring of their linguistic system (see, however, Samuda, 2001). This objective would be achievable through such activities as public performance, analysis and tests. In Skehan’s words, ‘the knowledge while the task is being done that a task may have to be re-done publicly will cause learners to allocate attention to the goals of restructuring and accuracy where otherwise they would not’ (Skehan, 1996: 101). The pedagogical sequence ends with a return to similar but not identical tasks that provide opportunities for learners to revise, learn and develop control of aspects of their linguistic system (i.e. interlanguage). Skehan’s sequence (cf. Willis, 1996) implies that learner attention is limited, selective and yet amenable to external manipulation, and it would play

84

Studies in Second L anguage Acquisit ion of Chinese

in favor of deliberate, planned and guided attention to form. In contrast, other researchers have promulgated implicit, integrative and intrinsic attention to form in TBLT (see, for example, Norris et al., 2009). Of note, Robinson (e.g. Robinson, 1996, 2001, 2005, 2011b) posits that learner attention is regulated by the processing demands of a task to the extent that the more cognitively complex tasks would elicit greater attention to form from the task-doer (i.e. the learner). The putative correlate between task complexity and learner attention to form underpins Robinson’s (2001, 2011b) Cognition Hypothesis, and its five ancillary predictions: (i) that cognitively more complex tasks should push learners to greater accuracy and complexity of L2 output than cognitively less complex tasks; (ii) that cognitively more complex tasks should engender more interaction and negotiation of meaning and, in turn, better noticing of and memory for input; (iii) that cognitively more complex tasks should engage greater depth of processing, leading to longer term retention of input; (iv) that gradual approximation of cognitively more complex tasks should increase automaticity of production; and (v) that individual differences should be more visible on cognitively more complex tasks than their simpler counterparts. Both the hypothesis and the predictions have, to varying degrees, been subject to empirical validation, yet with mixed results (see, for example, Robinson, 2011c). While more empirical research is certain to contribute to a clearer picture of the validity of these theoretical assumptions, Robinson’s (2001, 2007, 2011b) triadic componential framework for task design has already emerged as a clear winner among the competing proposals, having duly gained much traction among TBLT researchers and practitioners. Central to this framework is the distinction made between task complexity (i.e. the inherent processing demands of a task) and task difficulty (i.e. individual differences in terms of cognitive ability and affective capacity). Robinson maintains that task complexity should determine the task sequence in a TBLT syllabus, whereas task difficulty should modulate task outcome, affecting decisions on task implementation. Both must, therefore, be taken into account when designing a task, along with task condition (i.e. participation and participant variables), hence triadic. The framework thus distinguishes itself by the weight it gives to task-doers, who in most accounts have largely been ignored.

Learners as task-doers The general neglect of learners in the task process may in large part explain why extant empirical studies have failed to yield consistent findings. But awareness is growing about the need to examine learners’ role in task implementation. As Breen (1987: 333) aptly notes, ‘Learners are capable of playing havoc with even the most carefully designed and much-used task’. Hence, it is conceivable that a task outcome is the function of the

Task-based L anguage Teaching in a Study Abroad Conte x t

85

actual contributions of the individual learner, the particular task type, and the situational conditions within which the task was completed. Breen writes: Any learning outcome will be significantly shaped by the learner’s own perceptions of all three: their assumptions about what they themselves should contribute; their view of the nature and demands of the task itself; and their personal definitions of the task situation. In other words, any language learning tasks will be reinterpreted by a learner in his or her own terms. This implies that a pre-designed task – the task-as-workplan – will be changed the moment the learner acts upon it. The task-asworkplan will be redrawn so that the learner can relate to it in the first place and, thereby, make it manageable. (Breen, 1987: 334). Given the instrumental role of a learner in shaping the task process and/ or outcome (Coghlan & Duff, 1994; Dörnyei, 2002; Mori, 2002; Ortega, 2005), it would only seem critical that the learner perspective be investigated when evaluating the efficacy of a task, a task-based lesson or a TBLT program (cf. Ellis, 2012). The latter is what we set out to do in the present action research study. With its wealth of potentials and possibilities, a TBLT program can be developed for a variety of purposes, all in line with the overarching goal of enabling language acquisition, including improving students’ dispositions towards the target language, the focus of the present study.

The Study In our study, informed by the program goals (discussed below), we sought to survey learners’ reactions to their experience with a study abroad TBLT program. The guiding questions were as follows: (1) Does attending the study abroad TBLT program boost students’ confidence in speaking Chinese, the target language? (2) Does attending the program reduce their fear in speaking Chinese? (3) Does attending the program enable them to spontaneously speak Chinese? (4) Do students attending the program find it helpful to their acquisition of Chinese? As is typical of action research, these questions were not posited a priori; rather, they were generated impromptu while one of the authors observed some of the program activities. As such, the study had a very modest scope and depth and may even seem deficient, in the orthodox view. Nevertheless,

86

Studies in Second L anguage Acquisit ion of Chinese

the results from the study helped to validate – importantly, from a learner’s perspective – the program goals.

Context The TBLT program in which the study was conducted was a summer study abroad program of Chinese designed by and for a foreign languages university in the Republic of Korea. Held in Beijing over two weeks in summer 2009, the program is better characterized as task-supported than taskbased, the two main variants of TBLT. Ellis (2003) likens task-based instruction – whereby task is the sole unit of analysis for syllabus development and implementation – as the strong version of TBLT, and task-supported instruction – whereby both tasks and form-focused instruction are part and parcel of the curriculum – as the weak version of TBLT. The program was driven by the perceived limitations of foreign language instruction2 in providing opportunities for students to use the target language, in this case, Chinese. It was also motivated by the aspirations that by immersing students in a target language environment and, most importantly, through engaging them in real-world tasks, students would be induced to speak the target language, thereby, (a) overcoming their fear of communicating in the target language – a problem plaguing foreign language learners including the participants in this study, (b) boosting their confidence, (c) increasing their interest in crosscultural interaction, and (d) cultivating a desire to actively use the target language and an ability for autonomous learning (cf. Chapter 3, this volume). Previous research on task-based instruction has indeed shown that ‘learners’ confidence grows when they realize they can do something without the teacher’s direct support’ (Willis, 1996: 56). In a nutshell, taking advantage of the study abroad context – an inputrich environment and using the leverage of real-world tasks, the main goal of the present TBLT program was to improve the affect of its attendees so that they would be more confident in, and less fearful of, speaking the target language, and would, in turn, be more willing to seek out opportunities to improve their language skills. The curriculum of the program consisted of two modules, one focusing on the language and one on real-world tasks, delivered in parallel – the language module in the morning and the task module in the afternoon. Each morning session, three hours long, had students work mostly on formulaic expressions loosely relating to the theme of the tasks that the students would be carrying out in the afternoon. In other words, the language work in the morning was intended to support students’ task performance in the afternoon, as schematized in Figure 4.1. The instruction was traditional, featuring teacher centeredness, exclusive attention to form accuracy and limited opportunity for free practice. The task module, rendered in the afternoon, consisted of 16 real-world (otherwise known as target) tasks grouped into two

Task-based L anguage Teaching in a Study Abroad Conte x t

87

Figure 4.1 The summer TBLT program at a glance

categories, common tasks that everyone had to perform and group-specific tasks that were different for each group, as shown in Table 4.1. Overall, the arrangement of the two modules is reminiscent of Skehan’s (1996) recommended sequence discussed earlier, insofar as it began with form-focused instruction and was then ensued by communicative tasks. An example of a common task is drawing a so-called life map, which entails mapping out the location of the students’ dorm and its surroundings on A4-size paper, and marking out specific locations such as the classroom, the cafeteria, the clinic, the supermarket and the post office. An example of a group-specific task is visiting a supermarket, which involves finding the location and directions to the supermarket and its opening hours, and finding out about what it offers, and so on. Table 4.1 Module 2 tasks Common tasks

Group-specific tasks

• •

•

Drawing a life map Interviewing a salesperson at the Xiushui marketplace

• • •

Visiting: – a publishing house – a supermarket – a theater – a bank – a hospital – a technology zone – the National Library – the railway station – the History Museum – the Korean General Consulate in Beijing – the Korean Culture Society – a police station – a glass factory Buying a bicycle Renting an apartment Finding out scenic spots for one-day sightseeing in Beijing

88

Studies in Second L anguage Acquisit ion of Chinese

Regardless of its category, each task required students to work in groups, thereby prompting much interaction among the students and collaborative work, especially during the pre-task and post-task phases. Goal oriented, each task was implemented in three steps (cf. Willis, 1996): (i) pre-task planning whereby each group drew up a task-implementation plan; (ii) task execution whereby the group carried out the task; and (iii) post-task reporting, which had three components to it – preparation, rehearsal and presentation. According to Maeng (2009), the tasks were designed on the basis of students’ immediate survival needs and interests in the study abroad context and were therefore highly motivating. Maeng vividly recalled the extraordinary level of enthusiasm with which students approached their very first common task, drawing a life map, 3 upon their check-in at their dorm in Beijing. In spite of their fatigue from the journey, and because of the immediate relevance of the task, students completed it promptly and accurately, without any help from the instructor. Each group subsequently posted their map on a designated bulletin board for other groups to view, and they looked at each other’s map with curiosity and an earnest desire to learn. Table 4.2 shows the schedule of tasks.

Participants Participants in the study abroad program and, hence, in the present study were 40 college students, some of whom were Chinese majors and some nonmajors studying at a foreign languages university in the Republic of Korea. They were between 20 and 33 years old (mean = 22.45), and had studied Chinese for varying amounts of time, from one month to five years (mean = 22.07 months). They had all made between one and six trips to a Chinese-speaking region (mean = 1.97). By virtue of their voluntary participation in the program, all students aspired to improvement of sort in relation to their Chinese proficiency.4 Table 4.3 gives the descriptive statistics of the bio data.

Instrument Given the goal of the program and the research questions, a survey is probably the best method of data collection. There are various ways to conduct surveys, among them tests, observations and meetings (Brown, 2001). Table 4.2 Schedule of tasks Task Common Task 1 Time 1st week Monday, Tuesday

Group-specific Task 1 Common Task 2 1st week 1st week Thursday Saturday

Group-specific Task 2 2nd week Monday

Task-based L anguage Teaching in a Study Abroad Conte x t

89

Table 4.3 Descriptive statistics of bio data (N = 40)

Age Learning history Trips to China

Minimum

Maximum

Mean

Std.

Skewness

Kurtosis

20.00 1.00 1.00

33.00 60.00 6.00

22.45 22.07 1.97

3.02 15.85 1.27

2.117 0.719 1.55

4.81 −0.44 2.15

The most common ways are, however, interviews and questionnaires, especially for program evaluation. For the present study, we chose the latter as the tool for gathering data. Compared to survey interviews, questionnaires putatively have the advantage of allowing researchers to pursue issues on a broader scale. They are also easier to develop and more efficient to use, the attributes we sought at the time, compelled by logistic and temporal constraints. Thus, a questionnaire was developed for the present study, guided by five sets of considerations: (a) the program goals; (b) the curriculum; (c) students’ gains; (d) the program as a whole; and (e) students’ biographic information. 5 The questionnaire contained a total of 14 questions (see Appendix): three on dispositions towards using the target language (#6, #7, #8); four about the curriculum (#9, #10, #11, #12); one about gains (#5); four about the program as a whole (#3, #4, #13, #14); and two on the biographic background of the students (#1, #2). Each question was phrased in simplest terms to ensure that students would be able to understand it, and all questions were in the open-response format – requiring the respondents to produce a written answer.

Procedure Considering that the students were at varying levels of proficiency, some of them beginners, the questionnaire, originally written in the target language, was translated into Korean, the students’ L1, for use. Students were given the choice of answering the questions either in Korean or Chinese. The questionnaire was administered in one sitting in one place (Brown, 2001)6 on the last day of the program, and 20 minutes were allowed for its completion. The responses were collected on the spot, the response rate being 100%.

Analysis and Results Responses to the biographic questions were coded by recording raw numbers from which descriptive statistics were derived. Responses relating to the research questions were coded in terms of ‘yes’, ‘no’ and, in some cases, ‘perhaps’ types. Tokens of each type were then tallied and an overall percentage calculated. Responses to the remaining questions – referred to as ‘auxiliary

90

Studies in Second L anguage Acquisit ion of Chinese

questions’ below – were coded into percentages (or otherwise) summarizing the students’ expressed preferences. In this section, only the results directly speaking to students’ perspective on the program were reported, leaving out, that is, the biographic information, which had already been reported in a previous section. Moreover, the results directly addressing the research questions are taken up first,7 followed by the results on the additional concerns such as the curriculum and the program as a whole.

Does attending the program boost students’ confidence in speaking Chinese? Results revealed that 87.5% of the students responded positively to the question, 10% negatively and 2.5% with uncertainty. A student wrote, ‘刚开 始持有只有完整的文章才能去说的观念说的不是很多, 现在即使知道一点也要 去说有了这样的自信心’ (In the beginning I would speak only after I had a complete idea of what to say and how to say it and I therefore spoke very little, but now even if I knew little, I would speak, and I have such confidence).

Does attending the program reduce students’ fear in speaking Chinese? 67.5% of the participants felt that attending the program had reduced their fear in speaking Chinese, 30% did not share the sentiment and 2.5% reported no change. The same student wrote, ‘虽然不能完美的去说, 但是基本 的对话和去问路的程度能够自信地去说了。虽然感觉中文很难, 但是不再害怕了’ (Though I don’t speak perfectly, I can carry out basic conversations and ask for directions with enough self-confidence. Although Chinese is difficult, I’m no longer afraid of it).

Does attending the program enable students to spontaneously8 speak Chinese? 85% of the participants provided an affirmative answer to the question, while 10% answered negatively and 5% were uncertain. Again, the same student noted that ‘我能够进行一般的对话, 但是, 听和翻译还有不足’ (I can carry out basic conversations, but my listening and translation are still inadequate).

Do students attending the program find it helpful to their acquisition of Chinese? Students’ responses showed that overall, 95% of the students found the program helpful to their acquisition of Chinese, but 5% were unsure. The same student wrote, ‘一天到晚接触中文机会在韩国没有, 通过这次机会接受了 很多中文知识, 也渐渐有了说的自信心’ (There was never any day-long opportunity to experience Chinese in Korea, but through this program I have received a lot of Chinese knowledge, and have gradually developed confidence in speaking).

Task-based L anguage Teaching in a Study Abroad Conte x t

91

When probed about skill areas that they thought had benefited from the program: 45% of the students reported gains in one skill – speaking; while 32% reported gains in two skills – speaking and listening; 2% in three skills – speaking, listening, and reading; 15% in all four skills – speaking, listening, reading and writing; and 5% no gain in any skill whatsoever. These main results are visually displayed in Figure 4.2. Turning now to the ‘auxiliary’ questions (i.e. not directly related to the research questions)9 in the questionnaire, the results are summarized below:

Did you like the morning class or the afternoon class? 5% of the students indicated that they liked both, 32% liked the afternoon class, 60% liked the morning class and 3% were not sure. Figure 4.3 provides a visual display of the results. The same student whose responses were cited above commented that ‘早上上课更好, 虽然有点累一早开始一新的 开始新的一天, 感觉可以节约很多时间’ (The morning classes were better. Though a bit of tiring, I began a new day early, feeling that I could save much time).

Did the morning class facilitate your completion of tasks in the afternoon? 72.5% of the students responded positively and 15% negatively. In addition, 12.5% indicated that the morning class was somewhat helpful. Figure 4.4 visually displays the results. The same student noted that ‘因为日常生活 中用的词语学习了很多, 任务时可以实地去运用’ (Because I learned many everyday words in the morning, I could use them when I performed the tasks [in the afternoon]).

Of all the tasks, which one did you like the most? Of all the tasks (see Table 4.1), the one that was mentioned the most was visiting Xiushuijie, a marketplace frequented predominantly by foreign 100 90 80 70 60 50 40 30 20 10 0

Figure 4.2 Summary of main results

YES NO UNSURE

92

Studies in Second L anguage Acquisit ion of Chinese 100 90 80 70 60 50 40 30 20 10 0 Both morning and aernoon

Morning

Aernoon

Not sure

Figure 4.3 Preference for morning or afternoon classes

tourists. The task required students to conduct interviews with salespersons, focusing on their use of a foreign language, including the foreign language they spoke, learning strategies, learning environment, learning background and proficiency. According to Maeng (2009), the task was both fun and educational. The same student wrote, ‘去秀水街, 也可以买东西, 也可以完成任 务,一石二鸟’ (Going to Xiushuijie, I could do shopping and carry out tasks, thus killing two birds with one stone). Other students commented that ‘在 秀水街市场讨价还价记忆深刻’ (bargaining at the Xiushuijie market was memorable), ‘去秀水街, 因为可以知道中国生活方式和可以说汉语。很有意思’ (going to Xiushuijie, because there one could get to know the Chinese way of life and could speak Chinese), ‘秀水街, 亲身去感受了中国’ (Xiushuijie, where one could experience China first-hand), and so on.

Of all the tasks, which one did you like the least? The responses to this question were rather varied, in part because the tasks mentioned happened to be from the category of group-specific tasks (see 100 80 60 40 20 0 Helpful

Not helpful

Figure 4.4 Helpfulness of morning class

Somewhat

Task-based L anguage Teaching in a Study Abroad Conte x t

93

Table 4.1). Nonetheless, two tasks relatively stood out as the least liked, visiting the Korean Embassy in Beijing and visiting the History Museum. The first task required students to find out about current exchange programs between Korea and China, and the types of assistance available to Korean residents in China. Students noted that ‘最不喜欢去领事馆, 因为太远非常累’ (I didn’t like going to the Consulate, because it was far and exhausting), ‘太难’ (too difficult), ‘太远了’ (too far), and so forth. The task of visiting the History Museum had students find out the transportation route, the location, and the purposes of and the articles displayed in the museum. The same student noted that ‘博物馆关门活动改变’ (the museum was closed and therefore there was a change to the original task). Others elaborated that ‘人太多所以很累’ (too many people and exhausting), ‘两个活动都有所变动有点累, 国家博物馆因为 关门有点混乱’ (Two tasks had to be adjusted, so it was a bit tiring. The national museum was a bit chaotic because of closing), etc.

While in Beijing, how many hours a day did you spend on Chinese?10 17.5% of the participants reported 6 hours, 45% 5.6 hours, 20% 3.4 hours and 17.5% 1.2 hours. The same student wrote ‘三到四个小时’ (three to four hours).

What do you think was your biggest gain from the program? 32.5% of the participants mentioned ‘improving the level of Chinese’, 25% ‘increasing self-confidence in using Chinese’, 22.5% ‘building friendship’, 7% ‘getting experience’ and 2% ‘understanding the Chinese culture’, the rest providing miscellaneous answers. The same student wrote, ‘听和说 跟之前比起来, 恐惧心减少了, 看到中文好的人, 增加了学习中文的动机’ (compared to before the trip, I had less fear of listening and speaking, met people who could speak good Chinese, and increased my motivation for studying Chinese).

If you had a chance to attend the program again, would you do it? 72.5% of the participants indicated ‘yes’, 20% ‘no’ and 7.5% ‘maybe’. The same student responded positively, noting that ‘可以在实地使用中文’ (one can use Chinese in the real world).

Discussion The present study surveyed learners’ experience with a two-week TBLT of Chinese summer immersion program, using a questionnaire as the instrument for data collection. The primary purpose of the study was to see whether or not the program had achieved its goals, importantly, from the learners’ point of view. Results appeared to indicate that the goals were amply met. Students overwhelmingly felt that attending the program boosted their confidence, lessened their fear of speaking Chinese and enabled

94

Studies in Second L anguage Acquisit ion of Chinese

them to spontaneously speak Chinese (see Figure 4.2). During the two-week program, students’ interaction with the target language was substantial – 62.5% of them spending at least five hours a day on the target language. Taken together, the results showed that the program did improve the students’ dispositions towards using the target language. However, because there were no baseline data from the same participants (i.e. data collected from the same participants before the onset of the program), the results did not provide for an interpretation of the exact degree of improvement, especially on an individual level, a notable weakness to be mindful about in designing similar types of research in the future. The survey additionally shed light on the program curriculum. Students generally were in favor of the morning class. This finding, albeit unexpected, appears to suggest that they were more used to the traditional than the nontraditional mode of teaching and learning (see Figures 4.3 and 4.4). The fact that 72.5% of them found it helpful to the afternoon activities points to a dual possibility: (a) that the teaching was generally in line with the task demands, or (b) that students depended on explicit teaching to help forge an ability to use the target language. Further research may confirm either possibility. As for the task module of the curriculum, students seemed to favor the tasks (a) involving active interaction rather than passive listening or viewing (e.g. visiting the market place vs. visiting the consulate) and (b) providing input and output opportunities that were more attuned to their interest and linguistic readiness. It is worth noting that nothing along these lines was mentioned by the participants, who, instead, gave reasons pointing to the logistics of the tasks as mentioned earlier. Further research is warranted to shed light on the interaction between task conditions and students’ interest and readiness, and between task design and the circumstances under which tasks are implemented, from a learner’s perspective. Overall, the overwhelming majority of the participants found the program helpful to their Chinese – in particular, to their listening and speaking skills. Students’ spontaneous comments,11 a selection of which are presented below, were the best testimony: • • • • •

‘得到了很多帮助, 与在韩国只从书本上学习相比亲身去体验到的更有意思。 听和说进步最多’ (got a lot of help, experiencing it was more interesting than learning from books in Korea); ‘有帮助, 在一定程度上习惯了说和听’ (helpful; got used to speaking and listening to some extent); ‘在中文基本对话方面得到了很多帮助’ (got lots of help in basic Chinese conversations); ‘听, 说, 写得到了帮助’ (listening, speaking, and writing were helped); ‘一起小组活动时有帮助, 听和说’ (listening and speaking benefitted from group work);

Task-based L anguage Teaching in a Study Abroad Conte x t

• • • • •

• •

95

‘口语得到了很大的提高’ (speaking had much improvement); ‘对听和说有很多帮助, 因为有很多时间与中国人说话’ (very helpful to listening and speaking, because there was a lot of time speaking with Chinese people); ‘对说话练习, 有很多帮助’ (very helpful to practicing speaking); ‘有很多帮助, 听、说、读, 各方面都有帮助’ (very helpful, helpful to listening, speaking, reading, every aspect); ‘我认为自己出去体验后感觉到汉语, 比听课好些。为了深刻了解中国, 实行 更多的外面体验会有帮助’ (I feel that once I have gone out to experience Chinese, it’s much better than attending classes. In order to deeply understand China, it would help to experience it more outside the classroom.); ‘很帮助, 说和听’ (very helpful, speaking and listening); ‘很帮助了。每天可以用汉语所以自然提高我的汉语水平’ (It helped a lot. I could use Chinese every day, so naturally it improved my Chinese proficiency.);

Still, it must be recognized that, as a piece of action research, the present study took only a minuscule step towards understanding learner agency in TBLT, limited both in scope and methodology. A full-blown evaluation of the efficacy of the program should, according to Brown (1995: 218), involve ‘systematic collection and analysis of all relevant information necessary to promote the improvement of the [program]’ and, clearly, the relevant information should come not just from the students but from the teachers and the administrators as well. The limitations notwithstanding, the survey we conducted could serve – as had been intended as well – as a starting point, providing a working basis for further investigations. As expected, the results of our survey revealed issues requiring further investigation as much as they provided answers to our initial questions. One obvious issue is the number of dissenting voices – the negative and lukewarm responses – combined at 5%–27% – on questions directly related to the program goals. This population deserves closer scrutiny. What might have given rise to their lack of enthusiasm for the program? Was it the curriculum, the tasks, or was it their own lack of linguistic readiness to respond to the curriculum, or was it their personality that defied their involvement in collaborative work, or was it the teacher who failed to provide sufficient language support, or was it the fact that they lost their self-esteem as a result of being placed in a class or group with mixed abilities, and so on?12 Another visible issue was the differential preferences for the tasks. Why were the participants more receptive to some tasks than they were to other tasks? What was it about the tasks that might have put them down – was it task demands, task conditions, individuals performing the tasks, or a combination thereof (Robinson, 2001)? These issues and speculations are sufficient to feed off a productive

96

Studies in Second L anguage Acquisit ion of Chinese

research agenda, one that might require the following actions to be taken, among others: (1) conduct an analysis of the tasks used in the task module of the curriculum to compare their design characteristics both in light of Robinson’s triadic framework and in terms of INFO (Han, 2007), as noted earlier in this chapter, as such analyses may yield insights about why some tasks were better received than others by the students; (2) examine the relationship between the morning class and the afternoon task(s) to ascertain the extent to which one supports the other; (3) examine learners’ task-based output to see how it is constrained by task demands and conditions in relation to accuracy, fluency and complexity – the standard indicators of linguistic gains from task-based learning (Robinson, 2011d; Skehan, 1996; Van den Branden et al., 2009a); (4) explore individual difference variables such as proficiency, aptitude, memory, learning style, motivation, anxiety, personality, willingness to communicate, learner beliefs and learning strategies; and last but not least (5) Simultaneously examine learner task-based output, learning conditions afforded by the tasks, learners’ self-perception of gains, and other individual differences. A systematic pursuit of this research agenda would likely yield a deeper and fuller understanding of the extent to which, and why, learners have benefitted from the TBLT program – aside from vastly improving their dispositions towards the target language. Methodologically, this research agenda could only be realized via an analytic approach integrative of qualitative, survey and statistical methods.

Conclusion Factoring learners into the equation of TBLT has only been a recent endeavor, although the idea has long been around (Breen, 1987). As the literature has attested, current research on TBLT is pervaded by concerns with the question of how task design characteristics (e.g. planning time, number and nature of interlocutors, direction and type of information exchange, task complexity, task structure) may induce learners to focus on form in particular ways while engaging in the performance of tasks (see, for example, Tavakoli & Foster, 2011), while attention to learners’ agency in the TBLT process has been a ‘second emphasis’ in TBLT research (Norris et al., 2009: 243). To the extent that learning is ultimately regulated by the learner (Corder, 1967; Long, 2009) and that learners are mediators of task outcomes, TBLT research must balance its attention to task design characteristics with

Task-based L anguage Teaching in a Study Abroad Conte x t

97

attention to task implementation in order to achieve an adequate understanding of the efficacy of TBLT. Breen’s (1987) differentiation between taskas-workplan and task-in-process is particularly instructive, as the field ponders its future directions (cf. Ellis, 2012). Task-as-workplan refers to predesigned tasks, whereas task-in-process refers to tasks as reality. While both are part and parcel of TBLT, it is the latter, Breen rightfully stresses, which generates diverse learning outcomes and the quality and efficacy of any task. Therefore, task designers need to anticipate the reasons why, and the ways in which, learners reinterpret a workplan during the task-in-process (cf. Engeström & Sannino, 2012; Larsen-Freeman, in press). Learners may reinterpret a task in terms of its objective, content, procedure, context and conditions, and they may do it for a variety of reasons, not the least as a function of their internal constraints. These constraints can be cognitive, linguistic and/or affective. Recent research on TBLT has, indeed, begun to examine some of these factors (see, for example, Albert, 2011; Kormos & Trebits, 2011) and their interaction with task characteristics. These studies are, however, almost exclusively behavior oriented, experimental in nature and inferential. On the other hand, studies directly eliciting learners’ accounts of their own psychological states during task performance are as yet few and far between (e.g. Ortega, 2005), and surveys of learners’ feelings, attitudes, reactions and the like are all but non-existent (cf. Ellis, 2012). In that sense, the study reported in this chapter begins to help fill a gap in TBLT research. More and varied research is clearly needed to push for a greater understanding of learner agency in TBLT. On that note, we would like to close with a list of 12 questions to serve as a point of departure: (1) (2) (3) (4)

What is the learner’s attitude towards task-based learning? What are the learners’ definitions of a task? What is their view of the nature and demands of a given task? What are their assumptions about what they themselves should contribute when asked to perform a task? (5) How do their perceptions interact with their linguistic capacity when performing a task? (6) How does a learner manage a task when s/he is working on it? (7) What resources are drawn on by the learner while working on a task and towards its completion? (8) How do learners’ cognitive and affective capacities interact with their linguistic capacity while working on a task and towards its completion? (9) What does a learner do when asked to perform a task that is cognitively above, below or within his/her current cognitive and linguistic capacity? (10) What do learners plan when given the planning time and, relatedly, how does it vary with different tasks?

98

Studies in Second L anguage Acquisit ion of Chinese

(11) What opportunities do learners have for focus on form while engaging in tasks where meaning is primary and, relatedly, how do learners allocate their attention to meaning and form during task performance? (12) How do learners feel about their attention being directed to form before, during and after task performance? As an analytic approach to teaching (Long & Crookes, 1992), TBLT distinguishes itself by its intended focus on the learner, seeking to create conditions that would engage, and facilitate the operation of, learners’ built-in syllabus (Corder, 1967). It hence follows that learners, more than anyone or anything else, should be the touchstone of its efficacy and, for that matter, that learners’ voices must have a central place in TBLT research.

Appendix: The Questionnaire 暑期赴华学习结业问卷 请在二十分钟内回答以下问题。你可以用中文或者韩文。 名字: 日期: 年龄: 在韩国大学所学专业: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)

你学汉语有多久了? 来中国几次了? 每次住多久? 这次在北京期间, 除上课外, 还做了些什么事? 在北京期间, 平均每天听说中文的时间大概有多少? 这次暑期在北京的学习对你学汉语有帮助吗? 在哪些方面-听, 说, 读, 还 是写? 你现在对自己是不是有更大的信心了? 你现在还怕不怕说汉语? 你现在敢在北京街上与中国人对话吗? 你是喜欢上午的课还是下午的课? 为什么? 上午上的课对你下午做任务有帮助吗? 如果有帮助的话,有什么样的帮助? 在你与同学们所做过的任务当中, 你最喜欢哪些任务? 为什么? 在你与同学们所做过的任务当中, 你最不喜欢哪些任务? 为什么? 你觉得这次在北京学习的最大收获是什么? 如果今后还有来华学习的机会你还会来吗? 为什么?

谢谢! 祝你学习进步, 生活愉快!

Task-based L anguage Teaching in a Study Abroad Conte x t

99

Acknowledgments We would like to thank the reviewers for their helpful suggestions on an earlier version of this chapter. Any errors that remain are our sole responsibility.

Notes (1) Action research is a type of research which attempts to understand and improve an existing practice. It is situation based and context specific, and involves data collection which can be used for reflection, decision-making and the development of more effective classroom strategies (Parsons & Brown, 2002). (2) Students attending the program had been learning Chinese in a foreign language environment (Korea – where the target language Chinese is not spoken outside the classroom), not in a second language environment (e.g. China, where the target language is spoken both in and out of the classroom). The two environments afford qualitatively different learning conditions. For example, the foreign language environment is typically input-poor, while the second language environment is input-rich. It is generally believed that the outcome of learning in a foreign language environment is inferior to that of learning in a second language environment. (3) A reviewer questioned the nature of this task as communicative, arguing that the task requires a sense of direction, memory and discrete words, and that it can be completed without much linguistic skill. We agree that the tasks are different in terms of the amount of linguistic skills and knowledge needed for their completion, but nevertheless maintain that ‘drawing a life map’ is a task in that it fits squarely with the general definition of ‘task’ in the realm of TBLT. For example, Bygate et al. (2001) define a task as an activity that requires learners to use language, with emphasis on meaning, to attain an objective. Similarly, Willis and Willis (2007) identify three essential ingredients of a task: meaning, discourse and activity. By all these accounts, ‘drawing a life map’ is a task. The language use that comes with it lies not only in using vocabulary, but also in the process of negotiation leading up to the completion of the task. (4) According to Maeng (2009), the participants were of low, mid- or high proficiency in Chinese. (5) Given that it only has a small number of questions, each targeting a different variable, the questionnaire defies a statistical measure of its reliability. In fact, in a case like this, the evidence of ‘reliability’ should be that learners’ responses to the different questions show heterogeneity rather than homogeneity. As Brown (2001: 175) has noted, ‘if a survey is designed with subsections that measure distinctly different things, the fact that high reliability is found for the whole survey may be bad news’. (6) Brown (2001) recommends that questionnaires be group-administered rather than self-administered so as to avoid the pitfalls often associated with the latter: that there is often a low return rate; that on-the-spot clarification is not possible; and that the conditions under which the questionnaire was filled out are unknown. (7) The wording of the research questions is somewhat different from that of the relevant questions in the questionnaire. (8) ‘Spontaneously’ here means ‘unscripted’ or ‘unplanned’. (9) The wording of the auxiliary questions presented here matches that of the questions in the questionnaire.

100

Studies in Second L anguage Acquisit ion of Chinese

(10) The question, which was made clear to the students, addresses the total amount of time learning or using Chinese per day. It was intended to help us get at a spectrum of scenarios including ones in which a student attended both morning and afternoon sessions and yet ‘tuned out’ most of the time. (11) It may be recalled that the questions were open-ended. Respondents were free to write anything additional as they wished (see Appendix). (12) By design, each group had mixed abilities, with members at low, mid-, or high levels of proficiency.

References Albert, A. (2011) When individual differences come into play: The effect of learner creativity on simple and complex task performance. In P. Robinson (ed.) Second Language Task Complexity (pp. 239–266). Amsterdam: John Benjamins. Bley-Vroman, R. (1989) What is the logical problem of second language learning? In S. Gass and J. Schachter (eds) Linguistic Perspectives on Second Language Acquisition (pp. 41–68). Cambridge: Cambridge University Press. Breen, M. (1987) Learner contributions to task design. In C. Candlin and D. Murphy (eds) Language Learning Tasks (pp. 23–46). London: Prentice Hall. Brown, J.D. (1995) The Elements of Language Curriculum: A Systematic Approach to Program Development. Boston, MA: Heinle & Heinle. Brown, J.D. (2001) Using Surveys in Language Programs. Cambridge: Cambridge University Press. Bygate, M., Skehan, P. and Swain, M. (eds) (2001) Researching Pedagogic Tasks: Second Language Learning, Teaching and Testing. Harlow: Longman. Candlin, C. (1987) Towards task-based language learning. In C. Candlin and D. Murphy (eds) Language Learning Tasks (pp. 5–22). London: Prentice Hall. Coghlan, P. and Duff, P. (1994) Same task, different activities: Analysis of a SLA task from an activity theory perspective. In J. Lantolf and G. Appel (eds) Vygotskian Approaches to Second Language Research (pp. 173–194). Norwood: Praeger. Corder, S.P. (1967) The significance of learners’ errors. International Review of Applied Linguistics 5, 161–170. Dörnyei, Z. (2002) The motivational basis of language learning tasks. In P. Robinson (ed.) Individual Differences and Instructed Language Learning (pp. 137–158). Philadelphia, PA: John Benjamins. Doughty, C. and Williams, J. (1998) Focus-on-Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press. Ellis, R. (2000) Task-based research and language pedagogy. Language Teaching Research 4, 193–220. Ellis, R. (2003) Task-Based Language Learning and Teaching. Oxford: Oxford University Press. Ellis, R. (2005) Principles of instructed language learning. System 33 (2), 209–224. Ellis, R. (2012) Language Teaching Research and Language Pedagogy. Oxford: Blackwell. Engeström, Y. and Sannino, A. (2012) Whatever happened to process theories of learning? Language, Culture and Social Interaction 1, 45–56. Gass, S.M. (1997) Input, Interaction, and the Second Language Learner. Mahwah, NJ: Lawrence Erlbaum. Han, Z.-H. (2007) INFO: Optimizing conditions for instructed learning of Chinese as a second/foreign language. Paper presented at the the 2nd International Forum on the Teaching and Learning of Chinese, December, Guangzhou, China. Keck, C., Iberri-Shea, G., Tracy, N. and Wa-Mbaleka, S. (2006) Investigating the empirical link between task-based interaction and acquisition: A meta-analysis. In J.M. Norris

Task-based L anguage Teaching in a Study Abroad Conte x t

101

and L. Ortega (eds) Synthesizing Research on Language Learning and Teaching (pp. 91–131). Amsterdam: John Benjamins. Kenning, M. (2006) Evolving concepts and moving targets: Communicative competence and the mediation of communication. International Journal of Applied Linguistics 16 (3), 363–388. Kormos, J. and Trebits, A. (2011) Working memory capacity and narrative task performance. In P. Robinson (ed.) Second Language Task Complexity: Researching the Cognition Hypothesis of Language Learning and Performance (pp. 267–286). Amsterdam: John Benjamins. Larsen-Freeman (in press) Another step to be taken – rethinking the end point of the interlanguage continuum. In Z.-H. Han and E. Tarone (eds) Interlanguage: Forty Years Later. Amsterdam: John Benjamins. Long, M.H. (1996) The role of the linguistic environment in second language acquisition. In W.C. Ritchie and T.K. Bhatia (eds) Handbook of Second Language Acquisition (pp. 413–468). New York: Academic Press. Long, M.H. (2000) Focus on form in task-based language teaching. In R. Lambert and E. Shohamy (eds) Language Policy and Pedagogy: Essays in Honor of A. Ronald Walton (pp. 179–192). Philadelphia, PA: John Benjamins. Long, M. (2009) Methodological principles for language teaching. In M.H. Long and C. Doughty (eds) The Handbook of Second Language Teaching (pp. 373–394). Oxford: Blackwell. Long, M. and Crookes, G. (1992) Three approaches to task-based syllabus design. TESOL Quarterly 26, 27–56. Maeng, J. (2009) An exploration of TBLT in a target language environment. Paper presented at the 10th Annual Conference of the International Society of Chinese Language Teaching, August, Shenyang, China. Mori, J. (2002) Task design, plan, and development of talk-in-interaction: An analysis of a small group activity in a Japanese language classroom. Applied Linguistics 23, 323–347. Norris, J., Bygate, M. and Van den Branden, K. (2009) Variables affecting task-based learning and performance. In K. Van den Branden, M. Bygate and J. Norris (eds) Task-Based Language Teaching: A Reader (pp. 242–247). Amsterdam: John Benjamins. Ortega, L. (2005) What do learners plan? Learner-driven attention to form during pre-task planning. In R. Ellis (ed.) Planning and Task Performance in a Second Language (pp. 77–109). Amsterdam: John Benjamins. Parsons, R. and Brown, K. (2002) Teacher as Reflective Practitioner and Action Researcher. Belmont, CA: Wadsworth/Thomson Learning. Robinson, P. (1996) Introduction: Connecting tasks, cognition and syllabus design. University of Queensland Working Papers in AppliedLinguistics 1 (1), 1–15. Robinson, P. (2001) Task complexity, task difficulty, and task production: Exploring interactions in a componential framework. Applied Linguistics 22 (1), 27–57. Robinson, P. (2005) Cognitive complexity and task sequencing: Studies in a componential framework for second language task design. IRAL 43 (1), 1–32. Robinson, P. (2007) Criteria for classifying and sequencing tasks. In M.D. Garcia-Mayo (ed.) Investigating Tasks in Formal Language Settings (pp. 1–27). Clevedon: Multilingual Matters. Robinson, P. (2011a) Task-based language learning: A review of issues. Language Learning 61, 1–36. Robinson, P. (2011b) Second language task complexity, the Cognition Hypothesis, language learning, and performance. In P. Robinson (ed.) Second Language Task Complexity: Researching the Cognition Hypothesis of Language Learning and Performance (pp. 3–38). Amsterdam: John Benjamins. Robinson, P. (2011c) Task-based language learning: A review of issues. Language Learning 61, 1–36.

102

Studies in Second L anguage Acquisit ion of Chinese

Robinson, P. (ed.) (2011d) Second Language Task Complexity: Researching the Cognition Hypothesis of Language Learning and Performance. Amsterdam: John Benjamins. Samuda, V. (2001) Guiding relationships between form and meaning during task performance: The role of the teacher. In M. Bygate, P. Skehan and M. Swain (eds) Researching Pedagogic Tasks, Second Languge Learning, Teaching, and Testing (pp. 119–140). Harlow: Logman. Schmidt, R. (1990) The role of consciousness in second language learning. Applied Linguistics 11 (2), 129–158. Schmidt, R. (1995) Consciousness and foreign language learning. In R. Schmidt (ed.) Attention and Awareness in Foreign Language Learning (pp. 1–64). Honolulu, HI: University of Hawai’i at Manoa, Second Language Teaching and Curriculum Center. Schmidt, R. (2001) Attention. In P. Robinson (ed.) Cognition and Second Language Instruction (pp. 3–32). New York: Cambridge University Press. Sharwood Smith, M. (1993) Input enhancement in instructed SLA. Studies in Second Language Acquisition 15, 165–179. Skehan, P. (1996) A framework for the implementation of task-based instruction. Applied Linguistics 17, 38–62. Skehan, P. (1998) A Cognitive Approach to Language Learning. Oxford: Oxford University Press. Swain, M. (1985) Communicative competence: Some roles of comprehensive input and comprehensible output in its development. In S. Gass and C. Madden (eds) Input in Second Language Acquisition (pp. 235–256). Cambridge, MA: Newbury House. Swain, M. (1995) Three functions of output in second language learning. In G. Cook and B. Seidelhofer (eds) Principles and Practice in Applied Linguistics: Studies in Honor of H.G. Widdowson (pp. 125–144). Oxford: Oxford University Press. Tarone, E. and Swierzbin, B. (2009) Exploring Learner Language. Oxford: Oxford University Press. Tavakoli, P. and Foster, P. (2011) Task design and second language performance: The effect of narrative type on learner output. Language Learning 61, 37–72. Van den Branden, K., Bygate, M. and Norris, J. (2009a) Task-based language teaching: Introducing the reader. In K. Van den Branden, M. Bygate and J. Norris (eds) TaskBased Language Teaching: A Reader (pp. 1–14). Amsterdam: John Benjamins. Van den Branden, K., Bygate, M. and Norris, J. (eds) (2009b) Task-Based Language Teaching: A Reader. Amsterdam: John Benjamins. VanPatten, B. (1996) Input Processing and Grammar Instruction in Second Language Acquisition. Norwood, NJ: Ablex. Willis, D. and Willis, J. (2007) Doing Task-Based Teaching. Oxford: Oxford University Press. Willis, J. (1996) A Framework for Task-Based Language Teaching. New York: Longman.

5

The Relationship Between the Effectiveness of Recasts and Working Memory in the Learning of Different Linguistic Structures Shaofeng Li

Recasting refers to the reformulation of an erroneous second language (L2) utterance. Since the late 1990s, there has been a plethora of research in second language acquisition (SLA) on different aspects of recasts. Research has shown that recasting is the most frequent feedback type in both teacher–student interaction (Lyster & Ranta, 1997; Sheen, 2004) and NS– NNS dyadic interaction (Braidi, 2002), that it facilitates SLA (Li, 2010; Lyster & Saito, 2010), and that it leads to less learner repair (Lyster, 1998) and is less effective compared with prompts (metalinguistic feedback, clarification, repetition and elicitation) in classroom settings (Ammar & Spada, 2006). Nevertheless, it is still premature and misleading to draw definitive conclusions about the effectiveness of recasts because there is a lack of research on the factors constraining the effects of this corrective strategy (Ellis & Sheen, 2006; Li, 2010). This study investigates the relationship between recasts and two potential mediating factors: the choice of target structure and working memory capacity. These two variables are included because they represent two different sources of potential influence on the effects of recasts. The former is linguistic and learner external and the latter is cognitive and learner internal. There has been insufficient research into the two potential mediating factors and, to the best of my knowledge, there is no research on whether or not they may interact in affecting the effects of recasts. The present study is an attempt to take a dialectic approach to the effectiveness of the feedback type in question. 103

104

Studies in Second L anguage Acquisit ion of Chinese

Background Previous research on recasts Recasts constitute the most studied among all identified feedback types, according to both qualitative and quantitative research syntheses on corrective feedback (Ellis & Sheen, 2006; Li, 2010; Lyster & Saito, 2010; Mackey & Goo, 2007; Russell & Spada, 2006). There are likely multiple explanations for the popularity of recasts among SLA researchers, but in essence it is motivated by two overlapping factors. The first relates to the strengths of recasts as compared with other feedback types. Recasts contain both positive evidence and negative evidence, juxtapose the correct form with the erroneous form (thereby providing a cognitive contrast), do not interrupt the flow of communication, and afford opportunities for a brief time-out from the main orientation of communication (Doughty, 2001; Ellis & Sheen, 2006; Long, 1996, 2007). These characteristics are ideal for the current mainstream form-focused instruction where linguistic forms are attended to in meaningful communication. The second reason, which relates to the first, is that it is the most favored feedback type in second language classes. Classroom teachers may find recasts an exceptionally useful strategy because it is not intrusive or imposing. Recasts are especially favored by teachers in immersion programs because this type of feedback, by virtue of its being a dual source of positive and negative evidence, helps students ‘communicate about subject matter that requires communicative abilities that exceed their current developmental level’ (Lyster & Mori, 2006: 290). Research on recasts has centered around three major themes: occurrence, noticing and effectiveness. With respect to the occurrence of recasts, one consistent finding is that it is the most frequent type of feedback in all instructional settings (e.g. Lyster & Mori, 2006; Lyster & Ranta, 1997; Sheen, 2004). Studies investigating the noticing of recasts revealed that recasts were more facilitative of SLA when interpreted as positive evidence or a combination of positive and negative evidence than as negative evidence alone (Egi, 2007; Leeman, 2003). It was also found that the noticing of recasts was affected by the characteristics of recasts. For instance, the corrective force of short and simple recasts was more easily recognized than that of long and complex recasts (Kim & Han, 2007; Loewen & Philp, 2006; Philp, 2003; Sheen, 2006). A second striking finding is that the noticing of recasts as indicated by uptake (i.e. a learner’s response immediately following a recast) varied according to the orientation and dynamics of instructional contexts – recasts were more easily perceived as corrective in settings that had a stronger focus on linguistic form than meaning (Lyster & Mori, 2006; Sheen, 2004). A third factor constraining the noticing of recasts is the nature of the linguistic target (Carpenter et al., 2006; Mackey et al., 2000): Recasts pertaining to morphosyntax were less likely to be noticed than those targeting other linguistic domains such as phonology.

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

105

As to the extent to which recasts contribute to L2 development, previous studies found that L2 learners indeed benefited from this type of feedback (e.g. Han, 2002; Mackey & Philp, 1998). Quantitative research syntheses (Li, 2010; Lyster & Saito, 2010; Mackey & Goo, 2007) showed that recasts led to medium to large effects (Cohen’s d ranges from 0.5 to 0.9). These research syntheses also found that the effects of recasts increased over time, in that the delayed effects were larger than the immediate effects. However, recasts have been criticized for being implicit and input-providing rather than output-prompting, and have been found to be less effective than more explicit feedback types and prompts, at least in classroom settings (Ellis et al., 2006; Lyster, 2004). Notwithstanding the relatively large amount of research on recasts, further research is warranted because, as Ellis and Sheen (2006) have pointed out and as the research syntheses have demonstrated, more often than not, the effects of recasts are likely mediated by various factors. One of such factors is the choice of the linguistic target.

Recasts and the choice of target structure Different linguistic targets may react differently to recasts.1 For instance, Long et al. (1998) found that recasts worked for Spanish adverb placement but not for object topicalization; Iwashita (2003) found that recasts were effective in the learning of Japanese te-form verbs but not of the locative-initial targets; Ishida (2004) demonstrated that recasts worked better for resultative use of the -te i-(ru) structure than the progressive use of the same structure. These studies were all conducted in the laboratory. However, there is also evidence for the differential effects of recasts on different linguistic targets from classroom research. For instance, recasts were not effective in the learning of English article use in Sheen (2007), but they were in the learning of French gender agreement in Lyster (2004) and English possessive determiners in Ammar and Spada (2006). It should be noted that in these studies, the choice of target structure was not an independent variable. To date, there have been only two studies that have investigated the choice of target structure as a variable: one by Ellis (2007) and the other by Yang and Lyster (2010). Ellis investigated whether recasts and metalinguistic feedback have differential effects on the learning of the English past tense –ed and comparative –er. Thirty-four adult ESL learners were randomly assigned to three conditions: recasts (n = 12), metalinguistic feedback (n = 12) and control (n = 10). Results showed that recasts did not promote the learning of either structure, but metalinguistic feedback did. Yang and Lyster (2010) investigated the effects of prompts and recasts with 72 Chinese EFL learners. The target structures were regular and irregular English past tenses. It was found that prompts showed an advantage over recasts on eight measures and that, while prompts worked better than recasts in assisting the

106

Studies in Second L anguage Acquisit ion of Chinese

acquisition of regular past-tense forms, both feedback types worked equally well for the learning of irregular past-tense forms. Recasts were more effective for the learning of irregular past forms than that of regular past forms because of the greater saliency the former had. This study investigates the effects of recasts on two very different linguistic structures: Chinese classifiers and the Chinese perfective -le. A classifier is a word that is used between a determiner (which is typically a number but can also be a demonstrative or quantifier) and a noun. The classifier is one of the most striking features of the Chinese language (Li & Thompson, 1981). The Chinese people started to use classifiers as early as 1400 BC (Erbaugh, 1986), and there are over 900 classifiers in the language (Zhang, 2007). There are three syntactic permutations with respect to classifier use in the Chinese language and in each permutation the use of a noun is optional if the referent is inferable from the discourse context (Li & Thompson, 1981): (1) Number + Classifier + Noun e.g. yı¯ gè rén. 一 个 人。 One-CL person ‘One person.’ (2) Demonstrative + Classifier + Noun e.g. zhè pıˇ maˇ. 这 匹 马。 This-CL horse ‘This horse.’ (3) Quantifier + Classifier + Noun e.g. meˇi liàng che¯ 每 辆 车。 Every-CL vehicle ‘Every vehicle.’ The Chinese perfective -le is an aspect marker that encodes completion (termination) or inchoativity (change of state of affairs). It is generally agreed that there is a distinction between a verbal -le (encoding completion) and a sentence final -le (encoding inchoativity or change of state of affairs) (Li & Thompson, 1981; Liu, 2001; Van den Berg, 1989; Xiao & McEnery, 2004), although linguists are at times divided on the distinction. This study endorses the two -le view, and the target structure selected to receive

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

107

feedback treatment is the verbal -le, which is named the perfective -le in this case. The following examples show how the verbal -le differs from the sentence-final -le. Example (a) contains a sentence-final -le and means that it did not rain at first, but now it has started to rain; or the situation changes from not raining to raining. In (b), the post-verbal -le is used to indicate that the event of raining lasted for three days and it was completed sometime in the past. (a) xià yu ˇ le. 下 雨 了。 Fall rain-PFV ‘It has started to rain.’ (b) xià le sa ¯ntia ¯n yu ˇ. 下 了 三天 雨。 Fall-PFV three day rain. ‘It rained for three days.’ The perfective -le occurs with situations that are [+bounded] or [+telic]. As to the interaction between lexical aspect and grammatical aspect, -le is naturally compatible with accomplishment and achievement verbs. For verbs (state, activity, etc.) that encode atelic situations, that is, situations without an endpoint, to be used with -le, an external device (usually a quantifier) needs to be added to set a beginning and end point or a boundary for the event. Sentences (c) and (d) below illustrate how -le is used to indicate perfectivity. In sentence (c), the verb shua ¯ida ˇo (fall) is an achievement verb and has a natural endpoint. In sentence (d), the verb pa ˇo (run) is an activity verb without a natural endpoint, but the time duration shíwu ˇfe¯nzho¯ng (15 minutes) delimits the situation to license the use of -le. (c) ta ¯ shua ¯ida ˇo le. 他 摔倒 了。 He fall-PFV ‘He fell.’ (d) ta ¯ pa ˇo le shíwu ˇ fe¯nzho¯ng. 他 跑 了 十五 分钟。 He run-PFV fifteen minutes. ‘He ran for fifteen minutes.’ The Chinese classifier and the Chinese perfective -le were selected as the target structures for feedback treatment because they are different in nature. Drawing upon the related findings and discussion in previous studies (Ellis, 2007; Goldschneider & DeKeyser, 2005), the two selected structures are compared on the following two dimensions:

108

Studies in Second L anguage Acquisit ion of Chinese

(1) Form-meaning mapping. Form-meaning mapping can be transparent or opaque. In the case of the perfective -le, the form-meaning mapping is opaque because it has two variants that are semantically different. The verbal -le encodes completion and boundedness, and the sentence-final -le indicates current relevance or change of situation. Syntactically, this same form may occur in different positions in a sentence. The formmeaning mapping of classifiers, on the other hand, is transparent in that a certain classifier is prototypically associated with one object or a cluster of objects that fall into the same category because of the physical properties they have in common. (2) Explicit knowledge. The rule explanation for the use of the perfective -le is complex because it involves at least two components: (a) the event is completed, and (b) the situation must be bounded or have an endpoint. Thus, the rule is difficult to understand and learn as explicit knowledge. The rule of classifiers, conversely, is relatively easy: It only states that a certain classifier must be used with a particular noun or a certain category of objects.

Recasts and working memory Working memory is a cognitive construct that is responsible for storing and processing incoming stimuli. As a concept developed to replace shortterm memory, working memory catches the dynamic nature of this type of memory: It performs the dual function of information maintenance and manipulation. Theorists are divided over the structure of working memory: some hold that it is a unitary construct that is responsible for both information storage and processing (Daneman & Carpenter, 1980), while others (e.g. Baddeley, 2007) argue that it is componential, including a central executive and several slave systems. Working memory has been measured using digit span or non-word repetition tasks, in which case it taps into only the storage function, or using reading or listening span tests, where both the storage and processing components are drawn upon. A few studies have investigated the relationship between working memory and the effectiveness of recasts. Mackey et al. (2002) examined the relationship between the effects of recasts, noticing and working memory in learning English question formation by Japanese learners. Learners with higher working memory reported more noticing. It was also found that learners with lower working memory capacity showed larger immediate effects, whereas those with higher working memory showed better performance on the delayed posttest. Trofimovich et al. (2007) conducted a study on how memory (including working memory and phonological memory in this case), attention and aptitude affected the effects of recasts in a computer mode. It was found that working memory was the only non-significant predictor among the three independent variables. In

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

109

a similar vein, Sagarra (2007) looked at the relationship between working memory and the effectiveness of computerized recasts. Unlike Trofimovich et al. (2007), Sagarra found a positive correlation between working memory and the effects of recasts. (The discrepancy will be interpreted in the Discussion section.) These three studies obtained conflicting findings about the relationship between recasts and working memory, and they all pointed to the need for further research into the topic, given their idiosyncratic methodological features. Mackey et al.’s study involved a small sample, and both Tromfimovich et al.’s and Sagarra’s studies used computerized recasts. For one thing, it is not known how working memory relates to recasts in other instructional settings. For another, it remains to be seen whether different relationships exist between recasts and working memory in learning different linguistic targets. This study is undertaken to answer these questions.

Research questions The review of the literature demonstrates that, although there has been much research into different aspects of recasts, there is a need for more research into the factors affecting the effects of such type of feedback, such as the choice of target structure and learners’ working memory capacity. To date, there has been no research into the interaction between recasts, the linguistic target and working memory. Hence, the following research questions are formulated: (1) Do recasts have differential effects on the learning of Chinese classifiers and the Chinese perfective -le? (2) What is the relationship between the effects of recasts and learners’ working memory capacity in the learning of the two different structures?

Method Participants Forty-six learners of Chinese as a foreign language from two large US universities participated in the study. They were aged between 18 and 38, with a mean of 20.67 years. Among the 46 participants, 45 were native speakers of English and one was a native speaker of Korean. These learners were assigned to two groups: recast (n = 28) and control (n = 18). During the study, the recast group received corrective feedback in the form of recasts on their wrong use of the target structures, and the control group engaged in some placebo treatment by answering some questions, responses to which did not require the use of the target structures. To ensure that the

110

Studies in Second L anguage Acquisit ion of Chinese

Table 5.1 Group statistics Group

n

Mean

SD

Recast Control

28 18

29.86 29.83

7.50 8.93

two groups were comparable in terms of their L2 proficiency, a revised version of a standardized Chinese proficiency test called HSK was administered (detailed information on the test is provided in the section on Testing). Table 5.1 shows the descriptive statistics of the two groups. An independent sample t-test was performed to ensure that the two groups of learners did not differ significantly in their proficiency prior to the interventional treatment. The result showed that this was indeed the case, t (44) = 0.01, p = 0.99.

Treatment tasks Two treatment tasks were used for each of the target structures. In all four tasks, the learners engaged in dyadic interaction with the native-speaking researcher. The two tasks for classifiers are called spot the difference and picture description, respectively. In the spot-the-difference task, there were three sets of pictures, and in each set there were two similar pictures, the content of which differed in various aspects including the number of certain objects in them. The learner asked questions to the native-speaker interlocutor to identify the differences. In the second task, picture description, the learner was asked to describe seven pictures. In both tasks, the interlocutor prompted the learner to use classifiers in obligatory contexts so as to provide opportunities for recasting non-target-like use of the target structure. Phonetic (Chinese Pinyin), orthographic (Chinese characters) and semantic (English translations) annotations are provided on the pictures for what is believed to be the key words (mostly nouns) for the successful completion of the tasks. The annotations served to facilitate smooth communication – especially for less advanced learners – as well as to orient the learner’s attention toward the available linguistic resources (nouns in this case), the use of which necessitates or at least increases the likelihood of the use of classifiers. The availability of these linguistic resources might also alleviate the learner’s processing load, thereby making it possible for the learner to allocate more cognitive resources for the target structure. It must be noted that there is a general classifier (gè) in Chinese that can in many cases replace special classifiers. The extent to which the general classifier can replace a special classifier is an issue of controversy, difficult to determine, and subject to multiple factors such as the genre of speech, the speaker’s educational background, etc. It is beyond the scope of

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

111

the study to resolve the controversy. However, in order to establish the obligatory contexts for classifier use, that is, selecting the classifier–noun combinations where the general and special classifiers do not stand in free variation, a survey was administered to 45 native speakers of Chinese in the local community, who were provided with 40 sentences and were asked to determine what classifier should be used in each and if the classifier can be replaced by the general classifier. A classifier use must reach an 80% agreement rate to be included in the tasks. Consequently, 15 contexts for classifier use were selected. The two tasks used for the perfective -le are called video narrative and interview. In the video narrative, each learner watched a seven-minute video clip called the Pear Film (Chafe, 1980) and was asked to retell the story. The clip had sound effects, and the characters in it mimed the story without using any verbal expressions. The story is full of background and foreground actions and has been used in numerous previous studies to elicit Chinese narratives and investigate Chinese aspectual marking (Christensen, 1994; Duff & Li, 2002; Erbaugh, 2011; Yang, 2002). The learner was provided with a list of clues to follow when retelling the story. The clues were some discrete linguistic items and sentence fragments both in Chinese and English, and they served similar purposes to the linguistic annotations in the classifier tasks – to facilitate task implementation and to free up cognitive resources for the target structure. Another function of the clues is to supply the obligatory contexts for the use of -le, thereby preventing the possibility of the learner’s avoidance of the target structure as a result of a lack of competence or confidence in using it. To establish the obligatory contexts for the use of the perfective -le, the scripts of the oral narratives of the Pear Story from 40 native speakers of Chinese from Erbaugh (1986) and Christensen (1994) were examined. Identification of obligatory contexts was also based on Li and Duff’s detailed description of the use of -le by nine non-native speakers and nine native speakers of Chinese in their narratives of the Pear Story (2002). After the obligatory contexts were established, they were matched with the corresponding contexts in the English scripts of the oral narratives of 20 native speakers of English from Erbaugh’s study. The English clues are therefore from the speech data of native speakers of English. The second task for the perfective -le was an interview, where the learner answered some questions on flash cards. The questions (n = 16) were asked in English to prevent the modeling of the target structure as would happen if the questions were asked in the target language. The main purpose of the second task was to increase the number of tokens and types of verbs accompanying -le. A tally of the verbs used with -le from the narrative scripts of native speakers in previous studies showed that there were more achievement verbs than activity and accomplishment verbs in the Pear Story. The second task was created to address this asymmetry.

112

Studies in Second L anguage Acquisit ion of Chinese

In all four tasks, the learners received recasts on their non-target-like use of classifiers or the perfective -le. Although the bulk of feedback concerned the two target structures, incorrect use of other linguistic structures by the learners also received feedback. This served to distract the learner so that the purpose of the study was not easily perceived. Also, the learners were allowed to ask any questions related to vocabulary, but not to grammar.

Feedback operationalization In the present study, no distinction was made between different kinds of recasts such as simple versus complex, interrogative versus affirmative, and so on (cf. Loewen & Philp, 2006; Sheen, 2006). Instead, a broad definition was adopted, which defines recasts as the reformulation of erroneous production of the target structures while maintaining the central meaning of the utterance. Examples of how a recast was provided are given below. As shown, in Episode 1, the learner (non-native speaker) used the general classifier gè for pigs, which was inappropriate. The researcher (native speaker) reformulated the utterance by replacing gè with tóu, the correct classifier. In Episode 2, the learner failed to use the perfective -le in a bounded situation where the activity verb shuì (sleep) was used with an expression of duration. The native speaker responded by adding the aspect marker, hence providing a recast. Episode 1 *NNS: zhàopiàn shì lia a ˘ng gè zhu ¯ 照片 是 两 个 猪。 Photo be two-CL [wrong] pig ‘This photo is two pigs.’ NS: lia ˘ng tóu zhu ¯ 两 头 猪。 Two-CL pig ‘Two pigs.’ Episode 2 NNS: *wo˘ zuótia ¯n wa ˘nshang zhı˘ shuì wu ˘ gè xia ˘oshí 我 昨天 晚上 只 睡 五 个 小时。 *I yesterday night only sleep-[missing PFV] five-CL hour. ‘I only slept for five hours last night.’ NS: zhı˘ shuì le wu ˘ gè xia ˘oshí 只 睡 了 五 个 小时。 Only sleep-PFV five-CL hour ‘Only slept five hours’ NNS: shuì le wu ˘ gè xia ˘oshí 睡 了 五 个 小时。 Sleep-Asp five-CL hour. ‘Slept five hours’

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

113

Testing and scoring In this study, three tests were used for different purposes. A revised (shortened) version of the HSK was used to measure the learners’ general L2 proficiency. The test had 60 items, of which 30 relate to listening comprehension, 20 to grammar knowledge and 10 to reading comprehension. The possible score for this test was 60, with one point assigned to each item. More weight was given to listening and grammar because the study involved the learners’ understanding of input that contained feedback on the use of two linguistic structures. Grammaticality judgment tests (GJTs) were used to measure treatment effects. The GJT for each target structure had three versions: pretest, posttest 1 and posttest 2. Each test had 15 target items, which were the same across the three versions, and some distracter items. Among the 15 target items, eight were ungrammatical and seven grammatical. For each test item, the learner was asked to determine whether it was grammatical, ungrammatical or whether he/she was not sure. If a sentence was judged ungrammatical, the learner was asked to correct it. The pretest related to both of the target structures and had 40 items, 30 of which were target items (15 for each target structure) and 10 were distracters. In each of the posttests, there were 15 target items and eight distracters. To ensure that the learners’ performance on the tests reflected their knowledge about the target structures and was not affected by the lack of vocabulary knowledge, annotations were provided for words that were likely to impede the comprehension of the related items. Also, the learners were allowed to ask vocabulary-related questions during the test. The possible score for each GJT was 15, and the tests were scored based on the following two criteria: (1) For a grammatical item, a response received one point if the item was judged grammatical. If a grammatical item was considered ungrammatical but a correction was made on a non-target item, the response also received one point. (2) For an ungrammatical item, a response received one point only if the sentence was judged ungrammatical and a correction was made on the wrong use of the target structure. No credit was given if only a correct judgment was made without making the correction. A listening span test was used to measure the learners’ working memory capacity. The test consisted of 72 sentences, which were divided into four sets at span sizes 3, 4, 5 and 6. The sentences were syntactically varied, and half of them were semantically reasonable and half were not. All were created and validated in Waters and Caplan (1996). The test was developed using DMDX, free software used in psycholinguistic studies to measure

114

Studies in Second L anguage Acquisit ion of Chinese

reaction time when visual and auditory stimuli are responded to. During the test, the learner was asked to listen to each item, to decide whether the sentence made sense or not, and at the end of each set (when the program paused), to recall the final word of each sentence in that set. All three components of the test were scored – reaction time, veracity judgment and sentence-final word recall. The three scores generated by the test were transformed into z scores, and the average of the z scores was used for statistical analysis (Pearson’s correlation).

Procedure Each participant in the recast group partook in four sessions. In session 1, the learner took the HSK proficiency test and the GJT pretest. The HSK test lasted about 60 minutes and the GJT took 20–30 minutes. In sessions 2 and 3, respectively, each learner received recasts on their wrong use of Chinese classifiers and the Chinese perfective -le during dyadic interaction with the native-speaking researcher, followed by the immediate posttest. The sequence in which each learner participated in the two experimental sessions was randomized. The two treatment sessions happened on two consecutive days, and each lasted approximately 60 minutes. One week after the third session, the learner participated in the fourth session (about 60 minutes), where the working memory test and the delayed posttest were administered. There were three sessions for the control group. As with the recast group, the control group took the proficiency test and the GJT pretest in the first session, and they took the delayed posttest and the working memory test in session 3. In session 2, the control group only engaged in some tasks that were irrelevant to the target structures and did not receive feedback on their oral production. After completing some placebo treatment, they took the immediate posttest. Figure 5.1 illustrates the procedure of the study.

Results Classifiers Descriptive statistics regarding the participants’ classifier use as measured on the pretest, immediate posttest and delayed posttest are presented in Table 5.2, including the means and standard deviations of the recast and control groups. As demonstrated, both groups improved in their performance on classifier use as reflected by their higher scores on both posttests in comparison with their pretest scores. Also, for both groups, the scores from the delayed posttest are lower than their scores on the immediate posttest. A mixed design repeated measures ANOVA was performed with time as the within-group variable and group (recast versus control) as the between-group variable. The results appear in Table 5.3. It can be seen that

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

115

Control

Recast

HSK Proficiency Test + GJT Pretest

Day 1

Day 2

Treatment + Posttest 1 (CL or -le)

Day 3

Treatment + Posttest 1 (CL or -le)

Placebo Treatment + Posttest 1

Working Memory Test + Posttest 2

Day 4

Figure 5.1 Procedure of the study

there is a significant effect for time, F (2, 88) = 27.79, p < 0.05, for time × feedback interaction, F (2, 88) = 16.27, p < 0.05, and for feedback, F (2, 44) = 15.81, p < 0.05. Independent sample t-tests were conducted on the pretest scores and showed that there was no significant difference between the two groups, t (44) = 0.24, p = 0.81, indicating that any difference in their posttest scores resulted largely from the provision or absence of the instructional treatment. T-tests on the posttest scores showed that the recast group performed significantly better than the control group on posttest 1, t (44) = 4.01, p < 0.01, and posttest 2, t (44) = 4.47, p < 0.01.

The perfective -le The descriptive statistics for the perfective -le are presented in Table 5.4. Similar to the results related with classifiers, both groups showed higher scores on both posttests than on the pretest. Dissimilar to the results with classifiers, however, the experimental group’s learners’ performance scores on the delayed posttest were higher than their scores on the immediate posttest. Table 5.2 Classifiers: Descriptive statistics Condition

Recast Control

n

28 18

Pretest

Posttest 1

Posttest 2

M

SD

M

SD

M

SD

6.00 5.92

1.15 1.18

9.23 6.58

2.39 1.81

9.20 6.11

2.51 1.88

116

Studies in Second L anguage Acquisit ion of Chinese

Table 5.3 ANOVA results for classifiers Source Within-group results Time Time × feedback Error Between-group results Feedback Error

Sum of squares

df

98.45 57.65 155.89

2 2 88

123.60 344.05

1 44

Mean square

F

p

49.23 28.82 1.77

27.79 16.27

0.00 0.00

123.60 7.82

15.81

0.00

The data were subjected to a 2 × 3 mixed design repeated measure ANOVA (see Table 5.5), which showed that there was a main effect for time, F (2, 88) = 29.81, p < 0.05, and for feedback, F (2, 44) = 6.51, p < 0.05. There is no interaction effect between time and feedback, F (2, 88) = 1.50, p > 0.05. T-test results showed that there was no significant difference between the recast group and the control group prior to the feedback treatment, t (44) = 1.60, p = 0.12, so it is reasonable to claim that any between-group differences in posttest scores were largely attributable to whether feedback was available. Post hoc comparisons were conducted and showed that the recast group outperformed the control group immediately after the treatment, t (44) = 2.24, p < 0.05, and the same pattern was obtained at the time of the delayed posttest, t (44) = 2.54, p < 0.05.

Classifiers versus perfective -le One goal of this study was to ascertain whether recasts would have differential effects on the two different target structures. To achieve the goal, effect sizes associated with both target structures were computed. Effect size (Cohen’s d) indicates the magnitude of the effect of a certain treatment or independent variable. The utility of effect size makes it possible to examine the effectiveness of a certain treatment across different conditions. An effect size below 0.2 is considered small, 0.5 is considered as a medium effect, and above 0.8 is large. The confidence interval of an effect size represents the Table 5.4 The perfective -le: Descriptive statistics Condition

Recast Control

n

28 18

Pretest

Posttest 1

Posttest 2

M

SD

M

SD

M

SD

6.52 5.53

1.67 2.53

8.75 7.03

2.27 2.94

9.75 7.56

2.91 2.79

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

117

Table 5.5 ANOVA results for the perfective –le Source Within-group results Time Time × feedback Error Between-group results Feedback Error

Sum of squares

df

Mean square

F

p

160.44 8.07 236.82

2 2 88

80.22 4.04 2.69

29.71 1.50

0.00 0.23

29.31 198.11

1 44

29.31 4.50

6.51

0.01

level of confidence we have about the range the effect falls into. Table 5.6 displays the obtained effect sizes and confidence intervals. As shown, with respect to classifiers, the effect sizes (d = 1.18 and 1.32) associated with both posttests are large. Also, the lower and upper limits of the confidence intervals concerning both effect sizes are all positive, indicating that the effects of recasts in learning classifiers are significant. In contrast, recasts seemed to have smaller-than-medium effects on the learning of the perfective -le (d = 0.29, 0.42 for immediate and delayed effects, respectively). Furthermore, the confidence intervals of both effect sizes cross zero, which suggests that if the same treatment is repeated many times, there exists a possibility that recasts would show no effects on the learning of the perfective -le. However, this should not be taken as evidence that there is no effect (Borenstein et al., 2009); null effect is only a possibility. Combined with the significant p-values for the effects in question that are previously shown, it would seem justified to claim a small, non-robust effect for recasts on the learning of -le.

Working memory and recasts Pearson’s correlation analyses were performed to determine the relationship between working memory and the effects of recasts in the learning of the two different linguistic structures. Table 5.7 shows the correlation Table 5.6 Effect sizes associated with classifiers and the perfective –le Posttest 1 d

Classifiers -le

1.18 0.29

Posttest 2 95% Confidence interval Lower

Upper

0.54 −0.31

1.82 0.88

d

1.32 0.42

95% Confidence interval Lower

Upper

0.67 −0.18

1.97 1.02

118

Studies in Second L anguage Acquisit ion of Chinese

Table 5.7 Correlation coefficients for the relationships between working memory scores and test scores for classifiers and perfective -le Classifiers (r)

Recast Control

Perfective -le (r)

Pretest

Posttest 1

Posttest 2

Pretest

Posttest 1

Posttest 2

0.27 −0.43

0.34 0.01

0.48** −0.01

0.12 0.14

0.44* 0.27

0.27 −0.14

Notes: *p < 0.05; **p < 0.01.

coefficients for each bivariate analysis. To have a clear, holistic picture of the relationship, analyses were performed for both pretest and posttest scores, and for both the recast group and the control group. The results showed that the learners’ working memory scores were significantly correlated with the delayed effects of recasts in the learning of classifiers, r = 0.48, p < 0.01, and with the immediate effects of such feedback type in the learning of the perfective -le, r = 0.44, p < 0.05; there was no significant correlation otherwise.

Discussion This study aimed to investigate (1) whether recasts have differential effects on the learning of different grammatical structures, and (2) whether the effects of recasts correlate differently with L2 learners’ working memory capacity in the learning of different grammatical structures. The answers to both questions are positive. With respect to the first research question, it was found that while the recast group outperformed the control group on both posttests in learning both Chinese classifiers and the perfective -le, the magnitude of effect indexed by Cohen’s d was substantially different for the two structures. More specifically, as far as classifier learning is concerned, recasts showed large immediate and delayed effects. The effect sizes associated with the learning of -le, however, were small and inconsistent as the confidence intervals crossed zero. The finding that recasts had differential effects on different linguistic structures constitutes a significant contribution to feedback research. Some researchers (e.g. Long, 2007) consider recasts an ideal strategy in form-focused instruction because of its non-obtrusiveness, contingency, and the cognitive comparison it affords between the target form and the flawed L2 production. Others have discouraged this type of feedback because it led to less learner repair and therefore did not fare well in classroom settings where it tended to go unnoticed, such as in a content-based classroom (Lyster, 2004; Sheen, 2007; Yang & Lyster, 2010). These researchers also noted that recasts almost always showed large effects in the laboratory where its corrective force was easily perceived. It follows that the research setting or instructional context

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

119

alone is responsible for the disparity between previous studies in terms of the effectiveness of recasts. This study has shown that recasts worked differently after the research setting was controlled for and that the nature of the linguistic target mediated the effects of this corrective move. The fact that recasts demonstrated large effects on classifier learning but small, inconsistent effects on the learning of the perfective -le has to do with the characteristics of the two structures. Classifiers are salient, and transparent in form-meaning mapping. The classifier is situated in the DP (determiner phrase) and does not involve complex movement. The interpretation of a classifier is made locally in relation to the noun phrase it is attached to and, in most cases, there is a one-to-one correspondence between a classifier and a category of objects. The perfective -le, however, is relatively redundant, non-salient and opaque in form-meaning mapping. The interpretation of -le has to be made at the sentential and discourse level. A comparison of the attributes between the two structures leads us to a reasonable speculation: Recasts on classifiers were more likely to be noticed than those on the perfective -le. Previous studies on learners’ perceptions about corrective feedback showed that learners were indeed more successful in noticing lexical recasts than morphosyntactic recasts (Carpenter et al., 2006; Mackey et al., 2000). The distinction between classifiers and the perfective -le parallels that between lexicon and morphosyntax. While it stands to reason to attribute the small effects of recasts on the non-salient -le to the failure on the learners’ part to notice the corrective force of the feedback, it can also be argued that the complexity or learnability of a linguistic structure also affects the effectiveness of feedback. As previously discussed, the perfective -le involves complex rule explanation. The learners may need more external support or assistance, such as that in the form of explicit rule explanation, to master this linguistic target (see, however, Hulstijn & de Graaff, 1994). It follows that even if the corrective force of recasts was noticed, the scaffolding information contained in the feedback was likely insufficient to lead to a satisfactory learning outcome. Conversely, classifiers, a semantically transparent structure that does not require much rule explanation, may only require minimal instruction (in the form of recasts in this case) to be mastered.2 Corroboration for this argument can be found from previous studies. For instance, Sheen (2007) found that whereas recasts did not benefit the learning of English articles (a/the), metalinguistic correction, a more explicit type of feedback that contains rule explanation as well as the correct form, did. English articles are without question a complex, difficult linguistic structure, the internalization of which requires noticing at the level of understanding. The second research question asks whether the effects of recasts relate to learners’ working memory capacity. It was found that working memory correlated with the delayed effects of recasts in the learning of classifiers and the immediate effects in the learning of the perfective -le. Several questions

120

Studies in Second L anguage Acquisit ion of Chinese

arise related to the findings. The first concerns how working memory interacts with the recasts. To answer this question accurately, it is essential to know that a central component of working memory is conscious processing (Baddeley & Logie, 1999; Cowan, 1999; Dehn, 2008; Paradis, 2009) which, it is commonly believed, the central executive is responsible for. Consciousness necessarily relates to noticing, which is crucial to the intake of input, an integral step to acquisition. It then becomes easy to understand that under the (relatively) implicit condition where explicit instruction is absent, those with high working memory are better at noticing the positive and negative evidence contained in the feedback and therefore achieve more. As Sawyer and Ranta pointed out, ‘Assuming that noticing is crucial to learning, and attention is required for noticing, and attention at any moment is limited by WM capacity, then there must logically be a close relationship between amount of learning and size of WM’ (Sawyer & Ranta, 2001: 342). Recall that there have been two studies that examined the interaction between working memory and recasts. In both studies recasts were delivered via the computer. The two studies showed different results: Trofimovich et al. (2007) did not find any association between the effectiveness of recasts and measures of working and phonological memory. Sagarra (2007), on the other hand, found such an association. It is not clear what caused such a discrepancy, but some speculations can be made based on the research methods of these two studies. Lack of correlation between working memory and recasts in Trofimovich’s study might be due to the fact that computerized recasts were salient, that picture cues were provided and that the treated items and test items were identical. These external resources alleviated the learners’ cognitive load and minimized the role of working memory. The recasts in Sagarra’s study were also salient, but since the target structure has ‘low perceptual salience [and] limited communicative value’ (Sagarra, 2007: 234), the learners in the study might have had to make more use of their working memory resources to notice, store and process the structure in question. While this study showed that working memory was correlated significantly with the learning of both linguistic features, the magnitude and timing of the correlations differed. The correlation with classifiers was stronger; the correlation with the perfective -le was immediate but that with classifiers was delayed. The stronger link of working memory with the effects of recasts in classifier learning may have to do with the nature of this structure. Classifier learning is more exemplar-based and draws more on the learners’ ability to temporarily hold and manipulate information than their ability to conduct the syntactic processing that is required in learning the perfective -le. This is not to say that working memory is drawn upon in item learning as an information store and is not implicated in rule-based learning, as classifier learning also involves syntactic processing. The point to be made is that rule-based learning draws less on working memory than more exemplar-based learning, hence the stronger correlation with the latter.

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

121

It is not clear why working memory was correlated with the immediate effects of recasts in the learning of -le but with the delayed effects of the feedback in the learning of classifiers. Further research is needed to examine the interface between working memory and the timing of treatment effects.

Conclusion As the most frequent and probably the most favored feedback type in the classroom, recasts have been much studied. However, previous research has mostly looked at the effects of recasts in isolation, and what is lacking is research that examines and identifies the factors that moderate the effects of this corrective strategy (Ellis & Sheen, 2006). As is true with feedback research in general, it is misleading to make precocious, absolute claims about the superiority or inferiority of a certain type of feedback before the impact of its moderators has been adequately explored. This study attempts to address the problem by investigating the interface between recasts and two potential moderators – the choice of the target structure and learners’ working memory capacity. It was found that recasts showed a larger effect on the structure that involves transparent form-meaning mapping and simple linguistic projection (classifiers) than the structure that is semantically opaque and syntactically complex (-le). It was argued that this might be due to the difficulty involved in noticing -le vis-à-vis classifiers in corrective recasts. It was further contended that development in the learners’ interlanguage regarding the aspect marker -le may require a type of instructional intervention that not only contributes to its noticing at the level of awareness but also that at the level of understanding. As a central cognitive construct involved in academic learning, working memory has recently engaged an increasing amount of attention in SLA research. However, the mechanism regarding its role in SLA is far from clear and empirical research is without question warranted. This study revealed that working memory was related to the effects of recasts, a relatively implicit instructional strategy, in the learning of two linguistic structures, although the relationships varied between the two structures. Tentative hypotheses were attempted, but further investigation is in order to confirm or disconfirm the speculations. This study has several limitations. One concerns the relatively short duration of the experimental treatments. Learners may need a larger dose of feedback in order to acquire a complex structure like the perfective -le. A second drawback has to do with the heterogeneity of the learners with regard to the amount of prior instruction in the target language: They were from different levels of Chinese classes. Despite the fact that the two participant groups were comparable in their general proficiency (based on their HSK scores) and their prior knowledge about the target structures (based on their pretest

122

Studies in Second L anguage Acquisit ion of Chinese

scores), the results would be more convincing had the amount of prior instruction been controlled. A third weakness relates to the unequal number of learners in the two groups, which may pose a problem for statistical robustness. To conclude, this study represents an attempt to take an interactional approach to the effects of corrective feedback. It is hoped that more research of similar nature is undertaken, to the cumulative effect that comprehensive research syntheses are made possible to reach conclusions about the mediating factors of recasts/feedback. That, in turn, will prevent arbitrary, absolute claims about the strengths or weaknesses of any type of feedback.

Appendix: Items in the Grammaticality Judgment Test (Items 1–15 are for classifiers and 16–30 for perfective -le) (1) 李友喜欢看赛马, 她自己也有两马。 (2) 今天我的邮箱里有三封信。 (3) 要学好中文, 就必须有一个好的中文词典。 (4) 今天要下雨, 你最好带一个伞。 (5) 她男朋友送给她一条裙子, 所以她很高兴。 (6) 他的农场里有50 个猪。 (7) 你去邮局的时候能帮我买 5个邮票吗? (8) 长 江和黄河是中国的两河。 (9) 她的书包很大, 但是里 面只有一 本书。 (10) 李老师的家里有两个树, 很漂亮。 (11) 歹徒手里拿着一 把刀威胁公安人员。 (12) 如果你需要钱, 我就给你寄一张支票。 (13) 他昨天特别郁闷, 一连抽了五支烟。 (14) 我弟弟喜欢车, 他打算明年买两车。 (15) 那个女孩手里拿着两件黄色衬衫。 (16) 他今天早饭吃三个鸡 蛋和一块蛋糕。 (17) 那个模特在舞台上表演时摔倒了。 (18) 我昨天晚上学了三个小时中文。 (19) 我在路上走的时候, 看到有一个人的钱 (20) 上周我在商店买一双鞋。 (21) 上星期五, 我们在一家农场摘 了五磅草莓。 (22) 地震期间, 他在帐篷里住了一 个月。 (23) 他找到了他的自行车, 然 后骑走了。 (24) 去年我面试的时候, 老板问了我20个问 题。 (25) 在 “老北京” 饭店, 我和朋友点三个菜, 结果没有吃完。

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

123

(26) 他今天开车撞了一座房子。 (27) 你昨天睡10个小时, 不应该感到困。 (28) 去年夏天, 他在中国呆两周。 (29) 他先是在麦当劳工作三个月, 然后开始上学。 (30) 我的咖啡撒, 所以我还得再买一杯。

Notes (1) That linguistic structure is a variable has long been noted (see, for example, Hulstijn, 1995). (2) It is recognized, however, that the Chinese classifier system can be more complex than addressed here. The treatment provided in this study focused on the rather straightforward and frequent pairings of classifiers and nouns; more complex cases, such as one noun versus multiple classifiers, were purposely left out of the study.

References Ammar, A. and Spada, N. (2006) One size fits all? Recasts, prompts, and L2 learning. Studies in Second Language Acquisition 28, 543–574. Baddeley, A. (2007) Working Memory, Thought, and Action. Oxford: Oxford University Press. Baddeley, A. and Logie, R. (1999) Working memory: The multiple-component model. In A. Miyake and P. Shah (eds) Models of Working Memory: Mechanisms of Active Maintenance and Executive Control (pp. 28–61). Cambridge: Cambridge University Press. Borenstein, M., Hedges, L.V., Higgins, J.P.T. and Rothstein, H.R. (2009) Introduction to Meta-analysis. Chichester: John Wiley & Sons. Braidi, S. (2002) Reexamining the role of recasts in native-speaker/non-native-speaker interactions. Language Learning 52, 1–42. Carpenter, H., Jeon, K.S., MacGregor, D. and Mackey, A. (2006) Learners’ interpretations of recasts. Studies in Second Language Acquisition 28, 209–236. Chafe, W. (1980) The Pear Stories: Cognitive, Cultural and Linguistic Aspects of Narrative Production. Norwood, NJ: Ablex. Christensen, M. (1994) Variation in spoken and written Mandarin narrative discourse. Unpublished doctoral dissertation, Ohio State University. Cowan, N. (1999) An embedded-process model of working memory. In A. Miyake and P. Shah (eds) Models of Working Memory (pp. 62–101). Cambridge: Cambridge University Press. Daneman, M. and Carpenter, P. (1980) Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior 19, 450–466. Dehn, M. (2008) Working Memory and Academic Learning: Assessment and Intervention. Hoboken, NJ: Wiley. Doughty, C. (2001) The cognitive underpinnings of focus on form. In P. Robinson (ed.) Cognition and Second Language Instruction (pp. 206–257). Cambridge: Cambridge University Press. Duff, P. and Li, D. (2002) The acquisition and use of perfective aspect in Mandarin. In R. Salaberry and Y. Shirai (eds) The L2 Acquisition of Tense-Aspect Morphology (pp. 417– 452). Philadelphia, PA: John Benjamins. Egi, T. ( 2007) Recasts, learners’ interpretations, and L2 development. In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition: A Collection of Empirical Studies (pp. 249–267). Oxford: Oxford University Press. Ellis, R. (2007) The differential effects of corrective feedback on two grammatical structures. In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition: A Series of Empirical Studies (pp. 339–360). New York: Oxford University Press.

124

Studies in Second L anguage Acquisit ion of Chinese

Ellis, R. and Sheen, Y. (2006) Reexamining the role of recasts in second language acquisition. Studies in Second Language Acquisition 28, 575–600. Ellis, R., Loewen, S. and Erlam, R. (2006) Implicit and explicit corrective feedback and the acquisition of L2 grammar. Studies in Second Language Acquisition 28, 339–368. Erbaugh, M. (1986) Taking stock: The development of Chinese noun classifiers historically and in young children. In C. Craig (ed.) Noun Classes and Categorization (pp. 399–436). Philadelphia, PA: John Benjamins. Erbaugh, M. (2001) The Chinese Pear Stories: Narratives Across Seven Dialects. See http:// www.pearstories.org (accessed in September 2009). Goldschneider, J. and DeKeyser, R. (2005) Explaining the ‘natural order of L2 morpheme acquisition’ in English: A meta-analysis of multiple determinants. Language Learning 55 (Suppl.), 27–77. Han, Z.-H. (2002) A study of the impact of recasts on tense consistency in L2 output. TESOL Quarterly 36, 543–572. Hulstijn, J. (1995) Not all grammatical rules are equal: Giving grammar instruction its proper place in foreign language teaching. In R. Schmidt (ed.) Attention and Awareness in Foreign Language Learning (pp. 359–386). Honolulu, HI: University of Hawai’i at Manoa. Hulstijn, J. and de Graaff, R. (1994) Under what conditions does explicit knowledge of a second language facilitate the acquisition of implicit knowledge? A research proposal. AILA Review 11, 97–112. Ishida, M. (2004) Effects of recasts on the acquisition of the aspectual form -te i-(ru) by learners of Japanese as a foreign language. Language Learning 54, 311–394. Iwashita, N. (2003) Positive and negative input in task-based interaction: Differential affects on L2 development. Studies in Second Language Acquisition 25, 1–36. Kim, J. and Han, Z. (2007) Recasts in communicative eEFL classes: Do teacher intent and learner interpretation overlap? In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition (pp. 269–297). Oxford: Oxford University Press. Leeman, J. (2003) Recasts and second language development: Beyond negative evidence. Studies in Second Language Acquisition 25, 37–63. Li, C. and Thompson, S. (1981) Mandarin Chinese: A Functional Reference Grammar. Los Angeles, CA: University of California Press. Li, S. (2010) The effectiveness of corrective feedback in SLA: A meta-analysis. Language Learning 60, 309–365. Liu, X. (2001) Explaining the grammatical meaning of the sentence-final le in modern Chinese. Paper presented at the 10th International Conference of Chinese Linguistics, in conjunction with the 13th North American Conference on Chinese Linguistics, University of California. Loewen, S. and Philp, J. (2006) Recasts in the adult English L2 classroom: Characteristics, explicitness, and effectiveness. Modern Language Journal 90, 536–556. Long, M.H. (1996) The role of the linguistic environment in second language acquisition. In W.C. Ritchie and T.K. Bhatia (eds) Handbook of Language Acquisition (pp. 413–468). New York: Academic Press. Long, M.H. (2007) Problems in SLA. Mahwah, NJ: Lawrence Erlbaum. Long, M.H., Inagaki, S. and Ortega, L. (1998) The role of implicit negative feedback in SLA: Models and recasts in Japanese and Spanish. Modern Language Journal 82, 357–371. Lyster, R. (1998) Negotiation of form, recasts, and explicit correction in relation to error types and learner repair in immersion classrooms. Language Learning 48, 183–218. Lyster, R. (2004) Different effects of prompts and effects in form-focused instruction. Studies in Second Language Acquisition 26, 399–432. Lyster, R. and Mori, H. (2006) Interactional feedback and instructional counterbalance. Studies in Second Language Acquisition 28, 269–300. Lyster, R. and Ranta, L. (1997) Corrective feedback and learner uptake. Studies in Second Language Acquisition 19, 37–66.

Rel at ionship Bet ween the Ef fec t iveness of Recast s and Work ing Memor y

125

Lyster, R. and Saito, K. (2010) Oral feedback in classroom SLA: A meta-analysis. Studies in Second Language Acquisition 32, 265–302. Mackey, A. and Goo, J. (2007) Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (ed.) Conversational Interaction in SLA: A Collection of Empirical Studies (pp. 408–452). New York: Oxford University Press. Mackey, A. and Philp, J. (1998) Conversational interaction and second language development: Recasts, responses, and red herrings? Modern Language Journal 82, 338–356. Mackey, A., Gass, S. and McDonough, K. (2000) How do learners perceive international feedback? Studies in Second Language Acquisition 22, 471–497. Mackey, A., Philp, J., Egi, T., Fujii, A. and Tatsumi, T. (2002) Individual differences in working memory, noticing of interactional feedback, and L2 development. In P. Robinson (ed.) Individual Differences and Instructed Language Learning (pp. 181–209). Philadelphia, PA: John Benjamins. Paradis, M. (2009) Declarative and Procedural Determinants of Second Languages. Philadelphia, PA: John Benjamins. Philp, J. (2003) Constraints on ‘noticing the gap’: Nonnative speakers’ noticing of recasts in NS–NNS interaction. Studies in Second Language Acquisition 25, 99–126. Russell, J. and Spada, N. (2006) The effectiveness of corrective feedback for second language acquisition: A meta-analysis of the research. In J. Norris and L. Ortega (eds) Synthesizing Research on Language Learning and Teaching (pp. 131–164). Amsterdam: John Benjamins. Sagarra, N. (2007) From CALL to face-to-face interaction: The effect of computerdelivered recasts and working memory on L2 development. In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition (pp. 229–248). New York: Oxford University Press. Sawyer, M. and Ranta, L. (2001) Aptitude, individual differences, and instructional design. In P. Robinson (ed.) Cognition and Second Language Instruction (pp. 319–353). Cambridge: Cambridge University Press. Sheen, Y. (2004) Corrective feedback and learner uptake in communicative classrooms across instructional settings. Language Teaching Research 8, 263–300. Sheen, Y. (2006) Exploring the relationship between characteristics of recasts and learner uptake. Language Teaching Research 10, 361–392. Sheen, Y. (2007) The effects of corrective feedback, language aptitude, and learner attitudes on the acquisition of English articles. In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition (pp. 301–322). New York: Oxford University Press. Trofimovich, P., Ammar, A. and Gatbonton, E. (2007) How effective are recasts? The role of attention, memory, and analytical ability. In A. Mackey (ed.) Conversational Interaction in Second Language Acquisition (pp. 171–195). New York: Oxford University Press. Van Den Berg, M. (1989) Modern Standard Chinese: Een Functionele Grammatical. Muiderberg: Coutinho. Waters, G. and Caplan, D. (1996) The measurement of verbal working memory capacity and its relation to reading comprehension. Quarterly Journal of Experimental Psychology 49A, 51–79. Xiao, R. and McEnery, T. (2004) Aspect in Mandarin Chinese: A Corpus-Based Study. Philadelphia, PA: John Benjamins. Yang, J. (2002) The acquisition of temporality by adult second language learners of Chinese. Unpublished doctoral dissertation, University of Arizona. Yang, Y. and Lyster, R. (2010) Effects of form-focused practice and feedback on Chinese EFL learners’ acquisition of regular and irregular past tense forms. Studies in Second Language Acquisition 32, 235–263. Zhang, H. (2007) Numeral classifiers in Mandarin Chinese. Journal of East Asian Linguistics 16, 43–59.

6

The Effect of Mixed-sensory Mode Presentation on Retaining Graphic Features of Chinese Characters Yongan Wu

From the earliest stages to the most advanced levels, teaching a foreign language usually comprises the task of vocabulary instruction to allow the comprehension or construction of meaning in a new semiotic code (Shu & Richard, 1999; Wang et al., 2003). The traditional method of presenting Chinese characters to Chinese as a foreign language (CFL) learners is essentially textual and flattened (Chung, 2003). Three values of a character, its meaning, writing and pronunciation, are simultaneously presented in the text form. In the context of teaching CFL in the US, characters are typically accompanied by their English equivalents and Hanyu Pinyin, a widely used Romanized system serving as a phonetic alphabet for Chinese characters (see, for example, Liu & Yao, 2008; Wu, 2007). This method, however, has been criticized for its ineffectiveness in classroom teaching for the following reasons: (1) the interference of orthographic features of Pinyin and the learner’s primary language (Bassetti, 2006); (2) ignorance of the relationship between a character and its components in order to help retain and infer meaning (Shen & Ke, 2007; Taft & Chung, (1999); and (3) the split of attention and overload on working memory as learners search and match characters with corresponding meanings and pronunciations (Chung, 2007). Thus far, several studies have explored using the mixed-sensory mode (MSM) presentation to deliver the values of characters through more than one sensory channel. Researchers believe this method can considerably reduce the cognitive load of processing all values of a character at once (e.g. Chuang & Ku, 2011; Chung, 2007; Jin, 2006). The results of these studies demonstrate that, when characters are presented visually as symbols and acoustically as sounds, learners are better able to retain the meaning and 126

Ret aining Graphic Features of Chinese Charac ters

127

pronunciation, as well as better able to distinguish the target character from its distracters. Similarly, this chapter intends to further study the issue by examining the effect of the MSM presentation on retaining the graphic features of characters with a consideration of two factors: (1) character density, i.e. characters with a low number of strokes versus a high number of strokes; and (2) the effect of time, i.e. a participant’s performance differs between immediate and delayed posttests. The study reported herein also examines how the MSM presentation may influence a learner’s frequency of using phonological retrieval cues to recall characters. In an age where technology is increasingly available in the classroom, the potential and practicality of using multimedia to facilitate CFL vocabulary instruction is unquestionably a promising method which should receive further attention in its development (Xie & Yao, 2009; Yao, 2009).

Background The distance between Chinese and English has a considerable impact on the nature and process of how characters can be presented, making the issue a persistent focus of research for many decades (Chu, 1974; DeFrancis, 1968; Li & Lee, 2006; Wang, 1989). For CFL learners whose knowledge of the Chinese language is in its developing stage, a character appears to be either an amalgam of different distinct graphic units or a visualization of a single, solitary, graphic concept (Li & Lee, 2006). Unlike alphabets, these graphic units exist in a large quantity, yet cannot be easily associated with a given sound or meaning (Shen & Ke, 2007). It then becomes a particularly challenging task for CFL learners to squeeze an extensive knowledge map that arbitrarily connects the graphic features and semantic values of numerous characters into their working memory. To help remember and transcribe the sounds of characters, learners often rely heavily on the Pinyin script, which adds another layer of information to process and retain (Bassetti, 2007; Ke, 1998). In order to lower the load on a learner’s working memory (see also Chapter 6, this volume), scholars have been actively searching for effective methods through which faster cognition and longer retention of information can be achieved in the task of learning novel characters. One noticeable line of research stems from the theory that the human mind processes and stores information above two independent slave cognitive systems, namely, the verbal system for words and the visual system for images and analog representations (Baddeley & Hitch, 1974; Paivio, 1986). There are three major forms of representations from which information can be ingested: (1) symbolic representation as text; (2) auditory representation as sound; and (3) analog representation as image. The first two are considered closely entwined because they both involve ‘similar cognitive

128

Studies in Second L anguage Acquisit ion of Chinese

processes of text comprehension’ (Chun & Plass, 1997: 65). This feasibility of dividing the verbal form into two modes, i.e. written words (text) and auditory signals (vocalization of text), provides the key rationale for employing more than one presentation mode as an instructional means to enhance the velocity and reliability of the entire encoding process in case a singular mode will lead to detrimental overflows in the cognitive system (Andres & Petersen, 2002; Tabbers et al., 2004). The proliferation of multimedia learning environments has provided incentives for scholars to transfer the MSM from psychology labs into general classrooms (Mayer, 2001). During the last decade, a few studies have been conducted to adapt the modality effect to the domain of CFL. By presenting characters visually on screen and their pronunciation and meaning aurally through headphones, researchers can successfully isolate, compare and confirm the superiority of the modality effect on character learning. Jin (2006) administrated a tutorial to university-level CFL students from different orthographic backgrounds, presenting 36 characters in three modes to the treatment groups via multimedia technology in computer labs, each of which highlighted one aspect of character instruction: (1) the meaning and formation of sub-character components (in textual form); (2) the sequence of strokes (with animation); and (3) the pronunciation (delivered through headphones). The control group received printouts composed in the traditional format in which a character was presented next to its English meaning and Pinyin script. The results suggested the pronunciation mode was better than the traditional method in terms of producing a significantly higher correct rate in tasks involving distinguishing semantic, orthographic and phonological distractors, regardless of the participants’ language backgrounds. The study conducted by Chung (2008) compared the effect of the MSM presentation in character instruction among beginning- and intermediate-level CFL learners. In his first experiment, Chung presented 20 characters in random order and their prompts (pronunciations and meanings in English) to a group of beginning CFL learners who were then asked to pronounce these characters and translate them into English at the end of the presentation. Two weeks later, in a surprise, delayed test, learners were asked to perform the same task so Chung could measure their long-term memory. Compared to the control group that received prompts only in visual forms such as Pinyin scripts or texts, the treatment group who heard pronunciations and the accompanying English meanings through headphones scored higher in meaning recall tests in both rounds. Chung duplicated his research method on the same group of students two years later and again found the superiority of the MSM presentation on meaning recall among intermediate-level CFL learners. Chuang and Ku (2011) examined beginning-level CFL learners’ ability to retain Chinese characters’ graphic features under two conditions: a text group, which saw the

Ret aining Graphic Features of Chinese Charac ters

129

character and read an introduction in English about the character’s etymological formation, and a narration group, which saw the character and heard the same introduction in English delivered as an auditory input. An immediate and a delayed posttest were given to measure how well students distinguished and recognized the writing of target characters among graphic distractors. There was not a significant difference between the text group and narration group, but all participants tested higher in the immediate posttest than in the delayed test. In order to better evaluate the significance of the MSM presentation, one aspect that deserves particular attention from the positive results in the studies reviewed above is the number of strokes, or the density of target characters. Sergent and Everson (1992) discovered that CFL learners, regardless of their proficiency levels, felt increasingly challenged to accurately and promptly recognize characters that have a large number of strokes. Ke (1996) confirmed the common speculation that characters, in general, are easier to recognize than to produce. Those with low density are less challenging to produce than those with high density. Yet Ke’s study disagreed with that of Sergent and Everson in character recognition, suggesting that the factor of density did not play a role in character recognition. Ke’s study also confirmed a correlation that character recognition and character production decreased in the delayed test, indicating that the retention of detailed graphic features weakens at a different rate from that of general features. These findings were replicated and further refined by later studies (Ping, 2006; Xiao, 2002). Although at present there is no consensus mandating a definite stroke number as the dividing line between low- and high-density characters, CFL learners’ ability to retain graphic features of characters varies considerably at different density levels (Liu, 2008). Therefore, the advantage of the MSM presentation cannot be generalized without weighing this factor. Studies have shown, under the MSM condition, that CFL learners could retain more information about a character for a longer period of time and demonstrated an enhanced ability to distinguish target characters from their distractors, yet the relationship between retrieving graphic features and using phonological cues was left unexplored. According to Chun and Plass (1996), even when English native speakers learned words in German, both of which are alphabet languages, the percentage using phonological retrieval cues to correctly recall the English meaning of a German word was quite low and therefore ‘sound has very limited importance as a retrieval cue for the words learned’ (Chun & Plass, 1996: 190). For novice CFL learners it would be more unlikely to choose sound as the retrieval cue because, by nature, Chinese phonology is disconnected from its graphic features. As a result, when CFL learners try to connect their mental lexicon to the prompts of the given task, either to recognize or produce characters, the only feasible option they have is to rely on remnants of graphic units in their long-term memory. This leads to a reasonable extrapolation

130

Studies in Second L anguage Acquisit ion of Chinese

that auditory input may not enhance the retention of graphic features, per se. Instead, it allows CFL learners to allocate more time and attention to look at a character so they can achieve better memorization by removing the burden of scanning its Pinyin and English, and by improving the overall quality of the encoding process. If the former mechanism plays a role, then extending the exposure time to characters should be able to achieve a similar result; otherwise the latter mechanism must be true and recognized as a contributing factor as well. Previous studies have proven the effectiveness of using the MSM presentation to facilitate character acquisition. The role of character density in recognizing and producing characters has also been studied. These combined studies indicate that the next step would focus on the issue of using the MSM presentation to enhance graphic features retention under different character densities, a factor that has not yet been identified for measurement. Such a lack of differentiation raises questions concerning a possible variation of CFL learners’ performance between low-density and high-density character tasks. Also, the issue of how the MSM presentation enhances character learning needs to be further explained by examining retrieval cues used by CFL learners when recalling characters. Intrigued by these uncertainties, this study intends to examine beginning-level CFL learners’ ability to retain graphic features of characters with different densities and the impact of auditory input on the retrieval cues learners used. The issues examined concern: (1) the direction of performance differences, if any, between low- and highdensity characters in recognition; (2) the direction of performance differences, if any, between low- and high-density characters in production; and (3) the effects of the MSM presentation on the frequency of using phonological retrieval cues to recognize and produce characters.

Method Participants A total of 104 students participated in this study. They were either native English speakers who had no previous knowledge of Chinese or came from a Chinese or East Asian background. The latter group was allowed to participate, but their test results were removed in data analysis, thereby allowing the study results to better reflect the situation experienced by CFL learners. This adjustment lowered the total to 86 (50 males and 36 females, mean age = 21.2 years). At the beginning of this study, participants had learned approximately 450 characters in six months, eight strokes per character on average, using the same textbook as the curriculum. They were familiar with Hanyu Pinyin and followed the traditional stroke-order method when learning how to write characters.

Ret aining Graphic Features of Chinese Charac ters

131

Instruments Two instruments, a character recognition task and a character production task, were prepared to measure the retention of graphic features of lowand high-density characters. In the character recognition task, participants first saw English annotations and were asked to distinguish the corresponding characters from their orthographic distractors, characters that resemble target characters in structure and appearance. Questions from both groups were mixed up in random order to avoid a possible impact made by the sequence of questions received. Every target character was accompanied by three distractors. In the character production task, participants needed to produce characters by looking at the English equivalent. The target characters were chosen from the character list of the HSK (Hanyu Shuiping Kaoshi); a well-developed Chinese proficiency test that identified approximately 2900 commonly used characters. The researcher only selected characters that were not found in the participants’ textbook or used in class to avoid incidental study effects. Based on the findings and research designs of relevant studies done in the past (cf. Jin, 2006; Ke, 1996; Ping, 2006; Xiao, 2002), the threshold of the character density in this study was set at 11 strokes. The low-density group consisted of characters with 6–7 strokes; the high-density group had 13–14 strokes. A total number of 60 characters were selected, 30 low-density and 30 high-density, 6.6 and 13.5 strokes on average, respectively. Characters in each density group were randomly assigned to either a recognition or production task. This arrangement resulted in 15 low- and 15 high-density characters per recognition and production task (see Appendix A and Appendix B for the list of target characters and their distractors). In both tasks, participants were asked to indicate how frequently they evoked sounds in the recalling process by checking one of the following options: ‘very rarely’, ‘infrequently’, ‘frequently’ and ‘always’, at the end of the task. Their answers were then coded into categories 1–4 for data analysis. By comparing this information with test scores, the researcher examined how MSM presentation affected the participants’ choice of retrieval cues in both tasks.

Scoring In the recognition task, one point was given when participants chose the right answer to a question as they eliminated the interference of distractors. In the production task, one point was given when a character was written correctly and easily identifiable. Writings that resembled the target character, but deviated in stroke numbers and structures, regardless of the degree of resemblance, received no points. Both instruments were graded by the researcher and a native Chinese speaker. The interrater agreement was 100%.

132

Studies in Second L anguage Acquisit ion of Chinese

Procedure Participants were randomly assigned into the treatment and control groups. Both groups received characters and later took the tests in a computer lab. The order of presentation was random. The treatment group saw characters on screen and heard the corresponding pronunciation and English meaning via headphones. By clicking a link next to the character they could hear the auditory input multiple times. One character was displayed at a time for 10 seconds each and was automatically replaced by another. The control group saw the character, along with its Pinyin and English annotation on the same screen but heard no sound. The display time (per character) was twice as long (20 seconds) following the same, automatic replacement mechanism. The reason why the control group received 10 additional seconds in order to learn the characters is due to the known effects of MSM presentation, which has proven to be capable of yielding a positive impact on information processing and retention. Since no previous studies have addressed both MSM presentation and character density at the same time, it is reasonable to give the group without the advantages of MSM presentation, i.e. the control group, more time to study characters so the author can better understand the interplay between these two major variables. Participants received the same presentation for three consecutive days, one round per day, and were told to try their best to remember all three values of the characters: the written form, the meaning and the sound. On day three, right after the presentation, both groups completed the first round of tests on character recognition and production. No feedback or results of the tests were given. Five days later the second round of tests was administered, using the same instruments, again with no prior notice. Questions remained unchanged, but were in a different order to avoid text effects. The duration of both rounds of tests was 25 minutes. Oral instructions were given during the test to avoid confusion and anxiety.

Results What effect did the MSM presentation have on the participants’ ability to recognize characters and the direction of the performance difference, if any, between recognizing low- and high-density characters? Means and standard deviations of test scores for recognition of lowdensity characters in both rounds are presented in Table 6.1. A significant group difference was found, F (1, 84) = 4.33, p < 0.05 with a near medium effect size (partial η2 = 0.05), indicating an advantage of MSM presentation

Ret aining Graphic Features of Chinese Charac ters

133

Table 6.1 Means and standard deviations and N for test scores of low- and high-density characters as a function of presentation mode and time in recognition test Time

Presentation mode

Density

M

SD

N

Round 1

Mixed

Low High Low High Low High Low High

9.95 8.58 8.58 8.60 6.93 7.19 6.60 6.09

2.00 2.18 1.87 1.95 2.07 2.40 2.32 1.95

43 43 43 43 43 43 43 43

Traditional Round 2

Mixed Traditional

for recognizing low-density characters. The average score of the treatment group was significantly higher than that of the control group in the immediate posttest (t = 3.29, p < 0.05) with a medium-to-large effect size (d = 0.71). The difference was greatly reduced as time went by, i.e. the average score of the treatment group was 0.33 points higher than that of the control group in the delayed posttest, but insignificant (t = 0.69, p > 0.49). Compared to the mean difference of 1.37 in Round 1, it was evident that the advantage of MSM for low-density characters can hardly be retained for an extended period of time without reinforcement. For high-density characters, the data analysis revealed a different pattern of performance from that of the low-density characters. Group difference was insignificant, F (1, 84) = 1.51, p = 0.22. Both groups demonstrated an equal degree of competency when recognizing high-density characters in the immediate posttest (t = −0.05, p > 0.95), yet the treatment group performed significantly better in the delayed posttest (t = 2.32, p < 0.05) with a medium effect size (d = 0.50). This suggested that the MSM presentation, compared to the traditional method, did not enable the participants to recognize highdensity characters in a short period of time. However, the general orthographic features of more complex characters could be retained longer in the treatment group. The scores of those who received the MSM presentation were then analyzed by a Wilcoxon Signed Ranks test to compare participants’ performance under two conditions, low-density versus high-density characters. The difference was significant in Round 1 (z = −4.33, p < 0.05) with a large effect size (r = 0.66), but not in Round 2 (z = −0.56, p > 0.57). This proved that MSM helped participants better remember details about characters that are relatively low density in short-term memory. As time went by, however, annotations of both types of characters were eventually forgotten at such a speed that the difference became insignificant in the end.

134

Studies in Second L anguage Acquisit ion of Chinese

What effect did the MSM presentation have on the participants’ ability to produce characters and the direction of the performance difference, if any, between producing low- and high-density characters? Table 6.2 contains means and standard deviations of test scores for participants’ production of low-density and high-density characters in both rounds. For low-density characters, a significant group difference was found since the treatment group produced more correct target characters, F (1, 84) = 5.25, p < 0.05 with a medium effect size (partial η2 = 0.06). An insignificant group difference was found in Round 1 (t = −0.17, p > 0.87), but became significant in Round 2, (t = 4.52, p < 0.05) with a very large effect size (d = 0.93). This suggested the MSM presentation could not make participants remember more characters in the immediate posttest, but it did help them better remember the details of low-density characters. The participants’ performance with high-density characters was not significant between the two groups, F (1, 84) = 0.05, p = 0.82, neither in the immediate (t = 1.85, p > 0.06) nor the delayed posttest (t = −1.44, p > 0.15). The main effect of time was significant, F (1, 84) = 124.41, p < 0.05 with a large effect size (partial η2 = 0.60). Compared to the task of recognition, writing complex characters from scratch was so difficult that the MSM presentation did not produce any difference in Round 1 and the participants were unable to hold onto the comprehensive orthographic features of the various characters. For the treatment group, the low-density characters were significantly easier to produce than the higher density characters in both Round 1 (z = −5.65, p < 0.05) and Round 2 (z = −5.15, p < 0.05); the effect sizes were 0.86 and 0.79, respectively. The large effect size not only confirmed the common experience that characters with more strokes required a longer time Table 6.2 Means and standard deviations and N for test scores of low- and high-density characters as a function of presentation mode and time in production test Time

Presentation mode

Density

M

SD

N

Round 1

Mixed

Low High Low High Low High Low High

9.16 5.77 9.23 5.28 6.16 3.44 4.53 3.84

1.72 1.39 2.10 1.03 2.01 1.18 1.24 1.36

43 43 43 43 43 43 43 43

Traditional Round 2

Mixed Traditional

Ret aining Graphic Features of Chinese Charac ters

135

to learn, but also indicated the inability of the MSM presentation to lead to the retention of minute, subtle graphic features.

How did the MSM presentation affect the participants’ frequency in using phonological retrieval cues when recognizing and producing characters? Since participants did not receive any type of reviews on target characters between the two rounds of testing, their memory of the sound attributes seemed to have faded rapidly and therefore was not relevant to the scope of this study. The data analysis of retrieval cues only focused on the result of the first round. A chi-square test of independence was performed and found an insignificant difference between the control and treatment groups regarding their frequency of evoking sounds in the task of recognizing characters, χ² (4, N = 86) = 1.43, p > 0.69. Within each group, η was used to investigate the strength of the association between the participants’ frequency of using sounds to assist recall and the scores they received. A large effect size was found: η was 0.45 for the treatment group and 0.50 for the control group. A Mann–Whitney U-test was performed to find the significant contrast. It revealed that in the treatment group, those who frequently recalled the sounds scored significantly higher than those who rarely did so, z = −2.65, p < 0.05 with a large effect size (r = 0.49). The same data analysis procedure was applied to the scores of character production. The group difference was insignificant; i.e. participants from neither group demonstrated a higher tendency of using sounds to help their memory, χ² (4, N = 86) = 2.78, p > 0.42. Within each group, a Kruskal–Wallis H-test was conducted to examine the score difference among the participants who rarely, infrequently, frequently or always recalled sounds while producing characters. The results were insignificant for the treatment group as well as for the control group, p > 0.47 and p > 0.44, respectively. The lack of performance difference revealed the MSM presentation had little impact on character writing, in terms of influencing the participants’ ability to recall a character and produce characters.

Discussion With almost no exceptions, CFL teaching involves a challenging yet crucial task of presenting words and characters for learners to remember. Vocabulary learning creates considerable challenges for CFL learners in the US, by and large due to the difficulty in retaining orthographic features of Chinese characters and remembering their pronunciation and English annotations (Hamada & Koda, 2008; Zhang, 2009). By taking advantage of the modality effect, which has been proven effective in many areas (cf. Neath & Surprenant, 2003),

136

Studies in Second L anguage Acquisit ion of Chinese

CFL teachers have hoped to create a multimedia environment where students would be able to allocate more time to closely observing and encoding the orthographic features of characters while receiving their pronunciation and meaning aurally. One important finding from the present study is the positive effect of the MSM presentation for the learning of characters, both in terms of recognition and production. For example, the treatment condition promoted shortterm retention of general features as well as longer term retention of specific orthographic details of low-density characters. Such effects varied considerably from one condition to another and often fell short of retention. The variable and short-term effects, to a certain extent, were manifest across density levels, and therefore deserve attention. An in-depth examination of studies on character density reveals some, albeit partial and indirect, clues. The average number of strokes in Jin’s (2006) experiment was approximately 10. Characters associated with beginning-level participants in Chung’s (2008) study had seven strokes on average. Compared to the current study, this conspicuous difference in character density may help explain the absence of a general, dominant modality effect. According to You (2003), seven strokes constitutes a difficulty threshold for CFL learners, such that characters containing more than seven strokes are considerably more difficult to recall and produce than those containing fewer strokes. This finding is in accordance with the results of Feng’s (2002) study, which proposed a bell-shaped learning curve that changes direction at the point of six strokes; i.e. characters containing six strokes are the easiest to recall and those with more than 14 strokes are the hardest. The difficulty imposed by the high-density characters in the current study undoubtedly overloaded the encoding process even after Pinyin and English annotation as possible interferences were removed to make room for the visual channel. The results suggest that character density is an impacting factor to the outcome of the MSM presentation and therefore must be taken into full consideration in instructional design. Without such qualified knowledge, the claimed advantage of the MSM presentation is likely to fail as a result of oversimplification in manipulating the cognitive mechanism to retain orthographic features. Time is another factor which both teachers and students should pay close attention to for the purpose of achieving better and longer retention. As the present study has shown, without reinforcement between the immediate and delayed posttest, the auxiliary function of the modality effect in mnemonics quickly dissipated and lapsed into insignificance, particularly in character recognition. Studies have proven that externally supplied mnemonic aids, such as visual illustrations of characters (Kuo & Hooper, 2004) or instructor-provided explanations (Shen, 2004), do help participants score higher with a better short-term retention rate for general orthographic features and meaning, but can hardly resist the work of time.

Ret aining Graphic Features of Chinese Charac ters

137

Meanwhile, self-generated mnemonics allow participants to actively infer and construct meaning based upon their existing knowledge and personal experiences. This very act of pressing oneself to interpret and digest the new information into meaningful expressions may enable deeper processing (Phan, 2008; Wittrock, 1990). The MSM presentation appears to hold little potential to stimulate deep procession and therefore must be either integrated into broader teaching strategies or repetitively implemented throughout the entire learning process. The findings of the present study revealed that the possible improvement of the encoding quality promised by the MSM presentation was not found in a number of situations where character density and time lapse played a substantial role. Unlike the research design of Chung (2008), under which participants in both groups were exposed to target characters for the same amount of time, this study granted longer time for the control group, which resulted in some noticeable changes. The insignificant group difference proved that simply exposing participants to target characters for a longer period of time could achieve a similar effect to that of the MSM presentation for the purpose of learning orthographic features. Mayer (1984), in his classic study on reading aids and text comprehension, identified three types that are most relevant and valuable: (1) aids for selecting information, (2) aids for building internal connections, and (3) aids for building external connections. This concept was then adapted into the multimedia environment, where the target information can be delivered in a larger volume (cf. Chun & Plass, 1997). Following this line of thinking, researchers are looking for new methods through which the MSM presentation can be connected with all three types of aids to foster text comprehension. Svenconis and Kerst (1995) coupled audio input with the semantic mapping technique, i.e. using illustrative maps to depict the semantic relationship between target Spanish words. Compared to the traditional word listing method, semantic mapping alone did not lead to higher scores on vocabulary tests, yet one hybrid method (semantic mapping with sound) produced the highest average scores and another (word list with sound) impeded learning. This interaction between the modality effect and other techniques was duplicated in CFL by Jin’s (2006) study, in which the group that received both radical presentations with the option to hear the sound performed the best. It follows that audio annotation must be appropriately combined with other variables for optimal vocabulary learning in a hypertext environment. In addition to the direct effect of the MSM presentation, one important aspect associated with the ramifications of this practice must not be ignored, namely the role of phonological cues in learning and reading characters. CFL learners not only use phonological information embedded in orthographic features to learn characters and make word decisions (Hue, 1992;

138

Studies in Second L anguage Acquisit ion of Chinese

Shen, 2005), but also consciously add a layer of phonological prompts to assist reading comprehension (Everson & Ke, 1997; Lee-Thompson, 2008). The results of this study are in line with those findings. Three tutorials obviously were too ephemeral to impact the participants’ overall character learning strategies. Not surprisingly, there was no group difference in the frequency of evoking sounds when participants performed either task. Yet, those who made use of sounds ended up scoring significantly higher than those who infrequently or rarely did so. In conclusion, by incorporating auditory input in character instruction to stimulate the modality effect, this study examined the density factor that had been overlooked by previous studies and found the strengths as well as the weaknesses of the MSM presentation method in developing character recognition and production. The relatively small sample size and the possible homogeneity among participants from the same institute, along with other methodological inadequacies, suggests that the results of this study are tentative in nature and must be interpreted with caution. Researchers need to experiment with different ways of integrating the modality effect along with other methods in order to foster better retention of information presented.

Ret aining Graphic Features of Chinese Charac ters

139

Appendix A: List of Characters and their Distractors Used in the Recognition Task Target

D

D

D

SN

Target

D

D

D

SN

颖 翠 暮 慷 谱 锯 蔼 频 箩 碌 慈 腹 蜻 酶 誉

颍 粹 莫 慵 潽 钾 萌 颅 管 禄 意 腥 蜡 醉 举

顷 羿 墓 惟 湴 踞 荡 颁 曼 碍 恳 腰 睛 鲤 誓

款 翌 菒 廉 潽 错 霭 颍 簧 硕 悲 暖 靖 酣 詧

13 14 14 14 14 13 14 13 14 13 13 13 14 14 13

杠 伶 苏 呕 汪 亦 忌 州 沛 贞 岂 妄 巫 驱 忧

朴 玲 芍 叹 洰 戒 忘 训 沶 占 出 妾 夹 驼 优

杜 矜 芬 抠 纴 示 息 洲 污 负 邑 至 卒 眍 鱿

朽 仱 艻 讴 沣 亢 态 圳 沞 贡 岜 妟 亚 驻 犹

7 7 7 7 7 6 7 6 7 6 6 6 7 7 7

Notes: D, distractor; SN, stroke numbers.

Appendix B: List of Characters Used in the Production Task SN SN

鹊

锹

辐

碧

睦

膜

滨

蛾

熔

瑰

槐

蝉

碳

鄙

塘

13 匈 6

14 沃 7

13 讼 6

14 劫 7

13 舟 6

14 饪 7

13 伐 6

13 寺 6

14 孝 7

13 伪 6

13 刑 6

14 君 7

14 罕 7

13 驰 6

13 辰 7

Notes: SN, stroke numbers.

References Andres, H.P. and Petersen, C. (2002) Presentation media, information complexity, and learning outcomes. Journal of Educational Technology Systems 30 (3), 225–246. Baddeley, A.D. and Hitch, G.J. (1974) Working memory. In G.A. Bower (ed.) The Psychology of Learning and Motivation: Advances in Research and Theory (pp. 47–89). New York: Academic Press. Bassetti, B. (2006) Orthographic input and phonological representations in learners of Chinese as a foreign language. Written Language & Literacy 9 (1), 95–114.

140

Studies in Second L anguage Acquisit ion of Chinese

Bassetti, B. (2007) Effects of hanyu Pinyin on the pronunciation of learners of Chinese as a foreign language. In A. Guder, X. Jiang and Y. Wan (eds) The Cognition, Learning and Teaching of Chinese Characters (pp. 156–179). Beijing: Beijing Language and Culture University Press. Chu, Y.-K. (1974) Perception of Chinese characters: An experimental study. Journal of Chinese Language Teachers Association 9 (2), 57–65. Chuang, H.-Y. and Ku, H.-Y. (2011) The effect of computer-based multimedia instruction with Chinese character recognition. Educational Media International 48 (1), 27–41. Chun, D.M. and Plass, J.L. (1996) Effects of multimedia annotations on vocabulary acquisition. The Modern Language Journal 80 (2), 183–198. Chun, D.M. and Plass, J.L. (1997) Research on text comprehension in multimedia environments. Language Learning & Technology 1 (1), 60–81. Chung, K.K.H. (2003) Effects of Pinyin and first language words in learning of Chinese characters as a second language. Journal of Behavioral Education 12 (3), 207–223. Chung, K.K.H. (2007) Presentation factors in the learning of Chinese characters: The order and position of Hanyu Pinyin and English translations. Educational Psychology Review 27 (1), 1–20. Chung, K.K.H. (2008) What effect do mixed sensory mode instructional formats have on both novice and experienced learners of Chinese characters? Learning and Instruction 18 (1), 96–108. DeFrancis, J. (1968) The measurement of Chinese character study. Journal of Chinese Language Teachers Association 3 (3), 121–123. Everson, M.E. and Ke, C. (1997) An inquiry into the reading strategies of intermediate and advanced learners of Chinese as a foreign language. Journal of Chinese Language Teachers Association 32 (1), 1–20. Feng, L. (2002) 冯丽萍. 非汉字背景留学生汉字形音识别的影响因素. 汉字文化 3, 47–49. Hamada, M. and Koda, K. (2008) Influence of first language orthographic experience on second language decoding and word learning Language Learning 58 (1), 1–31. Hue, C.W. (1992) Recognition processes in character naming. In H.C. Chen and D. Tzeng (eds.) Language Processing in Chinese (pp. 93–107). Amsterdam: North Holland. Jin, H. (2006) Multimedia effects and Chinese character processing: An empirical study of CFL learners from three different orthographic backgrounds. Journal of Chinese Language Teachers Association 41 (3), 35–56. Ke, C. (1996) An empirical study on the relationship between Chinese character recognition and production. Modern Language Journal 80 (3), 340–349. Ke, C. (1998) Effects of strategies on the learning of Chinese characters among foreign language students. Journal of the Chinese Language Teachers Association 33 (2), 93–112. Kuo, M.-L.A. and Hooper, S. (2004) The effects of visual and verbal coding mnemonics on learning Chinese characters in computer-based instruction. Educational Technology Research and Development 52 (3), 23–38. Lee-Thompson, L.C. (2008) An investigation of reading strategies applied by American learners of Chinese as a foreign language. Foreign Language Annals 41 (4), 702–721. Li, J. and Lee, K. (2006) The graphic factor in the teaching and learning of Chinese characters. Journal of Chinese Language Teachers Association 41 (1), 79–92. Liu, L. (2008) 刘丽萍. 笔画数与结构方式对留学生汉字学习的影响. 语言教学与研究 1, 89–96. Liu, Y. and Yao, D. (2008) Integrated Chinese: Level 1, Part 1 (3rd edn). Boston, MA: Cheng & Tsui. Mayer, R.E. (1984) Aids to text comprehension. Educational Psychologist 19 (1), 30–42. Mayer, R.E. (2001) Multimedia Learning. Cambridge: Cambridge University Press. Neath, I. and Surprenant, A.M. (2003) Human Memory: An Introduction to Research, Data, and Theory (2nd edn). Belmont, CA: Wadsworth.

Ret aining Graphic Features of Chinese Charac ters

141

Paivio, A. (1986) Mental Representation: A Dual-Coding Approach. New York: Oxford University Press. Phan, H.P. (2008) Unifying different theories of learning: Theoretical framework and empirical evidence. Educational Psychology Review 28 (3), 325–340. Ping, M. (2006) Vocabulary acquisition in CFL (Chinese as a foreign language) contexts: A correlation of performance and strategy use. Unpublished doctoral dissertation, Brigham Young University. Sergent, W.K. and Everson, M.E. (1992) The effects of frequency and density on character recognition speed and accuracy by elementary and advanced L2 readers of Chinese. Journal of the Chinese Language Teachers Association 27 (1–2), 29–44. Shen, H.H. (2004) Level of cognitive processing: Effects on character learning among non-native learners of Chinese as a foreign language. Language and Education 18 (2), 167–182. Shen, H.H. (2005) An investigation of Chinese-character learning strategies among nonnative speakers of Chinese. System 33 (1), 49–68. Shen, H. and Ke, C. (2007) Radical awareness and word acquisition among nonnative learners of Chinese. Modern Language Journal 91 (1), 97–111. Shu, H. and Richard, C.A. (1999) Learning to read Chinese: The development of metalinguistic awareness. In J. Wang, A.W. Inhoff and H.-C. Chen (eds) Reading Chinese Script: A Cognitive Analysis (pp. 1–18). Mahwah, NJ: Lawrence Erlbaum. Svenconis, D.J. and Kerst, S. (1995) Investigating the teaching of second-language vocabulary through semantic mapping in a hypertext environment. CALICO Journal 12 (2–3), 33–57. Tabbers, H.K., Martens, R.L. and van Merrienboer, J.J.G. (2004) Multimedia instructions and cognitive load theory: Effects of modality and cueing. British Journal of Educational Psychology 74 (1), 71–81. Taft, M. and Chung, K. (1999) Using radicals in teaching Chinese characters to second language learners. Psychologia 42, 234–251. Wang, G.C. (1989) Research on teaching Chinese in forty-five universities: Analysis of survey results. Journal of Chinese Language Teachers Association 24 (3), 101–114. Wang, M., Perfetti, C. and Liu, Y. (2003) Alphabetic readers quickly acquire orthographic structure in learning to read Chinese. Scientific Studies of Reading 7 (2), 183–208. Wittrock, M.C. (1990) Generative processes of comprehension. Educational Psychologist 24 (4), 345–376. Wu, S. (2007) Chinese Link: Elementary Chinese: Level 1, Part 1. Upper Saddle River, NJ: Prentice Hall. Xiao, Y. (2002) The effect of character density on learning Chinese as a foreign language. Journal of Chinese Language Teachers Association 37 (3), 71–84. Xie, T. and Yao, T.-C. (2009) Technology in Chinese Language Teaching and Learning. In M.E. Everson and Y. Xiao (ed.) Teaching Chinese as a Foreign Language (pp. 151–172). Boston, MA: Cheng & Tsui. Yao, T. (2009) The current status of Chinese CALL in the United States. Journal of Chinese Language Teachers Association 44 (1), 1–23. You, H. (2003) 尤浩杰. 笔画数、部件数和拓扑结构类型对非汉字文化圈学习者汉字掌握的 影响. 世界汉语教学 2, 72–81. Zhang, W. (2009) Semantic prosody and ESL/EFL vocabulary pedagogy. TESL Canada Journal 26 (2), 1–12.

Epilogue Bringing It All Together: Where Are We? ZhaoHong Han

The research reported in this volume should best be taken as work in progress. No study has come to a firm conclusion about anything and, as a result, no generalizations are made nor intended. The issues addressed in and across the studies vary in the extent of attention they have received in the research into Chinese as a second language specifically or second language acquisition (SLA) more generally. Nonetheless, this book is useful in a number of ways: first, it serves as an interim benchmark as to where we are in terms of understanding what it means and takes to acquire Chinese as an additional language. Second, it offers some useful lenses through which to examine the process and outcome of L2 acquisition of Chinese. Third, it shows what methodological options are available (or not) to the researchers who intend to conduct research into the learning of Chinese in various settings. Fourth, it reveals the extent to which current research on the SLA of Chinese joins forces with the general field of study (see, e.g. Bhatia & Ritchie, 2009; Doughty & Long, 2003; Ellis, 2008; Gass & Selinker, 2008; Gass & Mackey, 2012; Robinson, 2012). Last but not least, it provokes thoughts as to what further questions to ask and investigate. By virtue of this being the closing piece of this book, it would only seem fitting that we address this last aspect, even if briefly. Chapter 1 looks into the L2 acquisition of Chinese path expressions and, to the extent that it deals with an intra-typological scenario where English and Chinese are both satellite-framed languages but are different from each other in subtle ways when it comes to path expression, the study fills a gap in the literature about typological influence on L2 acquisition, in general, and cross-linguistic influence of L1-based thinking for speaking, in particular. A useful, and even necessary, extension of the current research would be to carry out longitudinal case studies to track down the acquisition process and major factors influencing it. As it is, the study provides a developmental snapshot and, as such, any claims on development or acquisition, a quintessentially 142

Epilogue

143

temporal and diachronic process, can only be speculative. As the author has duly recognized: ‘the questions of whether L2 learners can shift toward the new L2 patterns and how fast and thorough such shift may occur remain to be determined.’ Another direction that one can take in substantiating the current inquiry is to empirically unpack the linguistic and cognitive relationship encapsulated in much of the rhetoric about the differences between native speakers and non-native speakers. Is it true, for example, that the use of deictic expressions by native speakers is a cognitive (albeit automatic) process and, relatedly, is it true that L2 learners need to change their conceptions of motion events in order to learn the Chinese expressions? In other words, will learners ‘need to reallocate attentional resources and reclassify the related spatial concepts in their L2 TFS in order to promptly describe motion events’, as the author has claimed? No studies to date have, independently, taken up these questions. A general question confronting the thinking-for-speaking line of SLA research is: How much of it is tied to habitual linguistic expressions and how much to underlying cognitive differences? Similar to the study reported in Chapter 1, the study presented in Chapter 2 also adopted a cross-sectional design, sampling participants from different proficiency levels. As such, it provided a ‘pseudo-longitudinal’ perspective on English-speaking learners’ use of sociopragmatic and pragmalinguistic strategies for producing the speech act of request. Given that the data for the study were elicited via a discourse completion questionnaire, further research could consider collecting naturalistic data. Such data may be difficult to come by, but are necessary to yield a trustworthy picture of learners’ pragmatic competence. SLA variationist research on learner language has consistently shown, inter alia, that learners’ L2 production varies systematically by context1 (see, e.g. Bayley & Tarone, 2011; Beebe & Zuengler, 1983; Tarone, 1983), but most importantly, that IL competence is best revealed when the learner attempts meaningful production in a natural context (Selinker, 1972; see also Bardovi-Harlig, in press; Han, in press; Tarone, in press). The understanding derived from the present study may also be substantiated through case studies to literally track individuals’ pragmatic development. As is becoming increasingly clear, when it comes to investigating development (or lack thereof), there is really no substitute for longitudinal research (Long, 2003). What would be even more interesting, and beneficial to instruction seeking to facilitate CFL learners’ pragmatic development, would be to naturally or experimentally explore the relation between the input available to the learner and their use of speech acts. Such kind of research would make a valuable contribution to the general field of SLA, since the connection between input and output has largely fallen out of the general purview. The next two chapters (Chapters 3 and 4) take an obvious turn away from learner language to learner perception. Both studies were situated in a study abroad program, a learning environment increasingly recognized as

144

Studies in Second L anguage Acquisit ion of Chinese

essential to improving foreign language proficiency, Jin’s study (Chapter 3) focusing on peer interaction as both a factual and a perceived source of learning, and Han and Maeng’s (Chapter 4) on the impact of task-based learning on learners’ affect. Using multiple types of qualitative data ranging from semi-structured interviews and journals to informal observations, Jin’s study sought to reveal how peers interacted among themselves and what their perceptions were of such interaction in an intensive study abroad program, resulting in interesting findings including, but not limited to, evolving patterns of peer interaction as a seeming function of increasing proficiency and familiarity with the local environment. However, as is typical of studies relying on qualitative data and subjective interpretation, the picture obtained was suggestive rather than tangible, and hence needs to be taken with a grain of salt. Still, the heuristic value of the findings cannot be overstated. If followed up on properly, they may potentially lead to a tangible understanding of the issue(s) in question. One direction that an ensuing study can take is to pick up on the possible connection between the changing status of peer interaction and the increasing ease with which individual learners interact directly with native speakers of Chinese in the local community, to ascertain, first of all, the extent to which the putative connection applies to a given group of learners.2 This is achievable through conducting a series of focused surveys and/or interviews to directly probe individuals’ participation in peer interaction, on the one hand, and their relative ease when interacting with locals, on the other, importantly over time. Next, one can look, separately or concurrently, into the contents of peer interaction over time and factors contributing to the changing ease with which learners interact with native speakers. If proficiency is one of the factors, as claimed in the current study, then a more objective, repeated measure of proficiency – listening and speaking in particular – would be in order, so as to keep track of the ongoing linguistic progress. These types of data could eventually feed into a correlation analysis to derive a more objective than subjective basis for understanding peer interaction in a study abroad context, and its contribution to language learning. Compared to Jin’s study (Chapter 3), the Han and Maeng study (Chapter 4) employed fewer types of data in probing learner perceptions. Using a focused questionnaire, the study aimed at, and provided for, an initial assessment of whether or not a study abroad task-based language teaching (TBLT) program had met its goal of boosting learners’ confidence in, and reducing their fear of, using the target language. With the learner perspective as its chosen lens, the study constituted one of the few existing attempts to begin investigating learners as task-doers in TBLT, a topic of growing interest in task-based SLA research (Ellis, 2012), opening up a host of avenues for future research. Its novelty and ecological validity (i.e. the study was improvised to satisfy a real-world need to understand if the goal of a TBLT program was met) notwithstanding, the study was simplistic in design, its findings

Epilogue

145

awaiting further, more objective verification. In particular, there is a need to conduct task analysis, examine the relationship between form-focused instruction and task performance, analyze learners’ task-based output, explore individual differences in interpreting and executing tasks and, lastly, cross-examine these various types of data to yield a fuller understanding of the extent to which, and why, a study abroad TBLT program of this nature may aid students in their L2 development. Chapters 5 and 6 both examine the efficacy of instruction on learning. Chapter 5 explores the contribution of working memory, an individual difference variable that has garnered increasing attention from SLA researchers, to the effectiveness of recasts in the learning of different linguistic structures. The focus of the study not only represents an emerging strand of research in the general study of SLA which looks at the extent to which the impact of instruction is modulated by individual difference variables, but also adds to it a new spin by including a linguistic variable (i.e. simple versus complex structures). As such, the study investigated a two-way interaction between recasts, on the one hand, and working memory and type of structure, respectively, on the other. Further research in this spirit could revisit the linguistic variable, in the light of learner readiness. Just as it is possible that linguistic structures can be differentially complex, posing in turn differential challenges to the learner and thereby affecting the efficacy of instruction, so it is likely that learners have a differential grasp of the structures to begin with, regardless of their complexity. When the latter scenario occurs, instruction may fail to have a desired impact due to the learner’s lack of readiness, even when provided on a supposedly simple structure; conversely, even when instruction is targeted at a putatively complex structure, it could be readily discernible to the learner (cf. Han et al., 2008). Learner readiness vis-à-vis the target of acquisition has been a much neglected construct in SLA research, beyond the realms of the morpheme order (e.g. Cazden et al., 1975) and processability (e.g. Pienemann, 1998) studies. Moreover, the role of working memory warrants further scrutiny, especially in an instructed learning environment, where its function is likely to be confounded with effects of instruction and proficiency. The fact that extant SLA studies involving working memory as a modulating variable have turned up conflicting, and even random, findings suffices to justify the need to re-evaluate the construct and the means by which it is measured (cf. Han & Liu, 2013). Chapter 6 focuses on the learning of Chinese characters in beginning learners. The study manipulated the mode of presentation (auditory or auditory and visual) and investigated its effects on short-term and long-term retention of graphic features of different types of characters (i.e. low versus high density of strokes), resulting in the important finding that auditory exposure to low-density – as opposed to high-density – characters led to notable short-term, but not long-term, retention. The study underscored the benefit of phonological aid in character learning. To further the understanding

146

Studies in Second L anguage Acquisit ion of Chinese

of the effects of multimodal exposure on character learning, this study can be branched out in at least two ways. Firstly, learners’ retention of characters as a function of low versus high density could be further investigated by providing multimodal contextual input. Thus, the learning of characters would not occur in isolation, but in context. Natural, contextual input may provide more cues to the learners which might facilitate their learning and retention of the characters. The fact that externally provided support did not have a lasting impact on learning and retention may have to do with a lack of sufficient cues (formal and semantic) available to assist the encoding. Secondly, future research can experiment with frequency as a variable. Repeated exposure to contextual input might provide the frequency that is needed for accurate encoding and reinforcement. Both the availability of cues and frequency are deemed critical to input processing and intake in SLA (see, e.g. Carroll, 1999; Ellis, 2002). In closing, L2 acquisition of Chinese provides a fertile ground for examining a wide array of issues, as partially exemplified in this volume. The amount of interest and the actual work that has been conducted so far bodes well for a bright future for SLA research of Chinese. The coming years are bound to witness a steady aggregate of studies, leading eventually to an empirical basis that allows generalizations to be made, and to be harnessed by practitioners for the improvement of instructional procedures – so that teaching will no longer be conducted by hunch, and teachers need no longer ‘摸着石头过河’ (wade through the river by touching the stones). As the research on L2 acquisition of Chinese progresses, it will show a greater alignment with the general field of study, both in terms of theoretical perspectives and methodological approaches. Such increasing alignment is both desirable and necessary, as it would, for one thing, prevent researchers from reinventing the wheel. Unique as it is, the learning of Chinese follows the same general psycholinguistic processes governing the learning of all other natural languages (cf. Long, 2009). Clearly, much has yet to happen. As has in part been attested in this book, current research on L2 acquisition of Chinese exhibits a notable dearth of (a) longitudinal research, (b) individual-oriented case studies, and (c) research on heritage speakers. As any experienced researcher knows, case studies, specially real longitudinal studies – spanning at least a year – are time consuming and can even be unfavorably judged, due to their alleged lack of generalizability, but these studies are nevertheless of utmost importance to a sophisticated understanding of SLA (see, e.g. Duff, 2007; Hakuta, 1976; Huebener, 1983; Lardiere, 2007; Schmidt & Frota, 1986; Schumann, 1978), SLA of Chinese included. One of the thorny issues facing current L2-Chinese researchers concerns the two distinct learning populations, speakers of other languages and heritage speakers. While much has been written elsewhere about heritage speakers, particularly in relation to Spanish as the L2 (e.g. Cabo & Rothman, 2012;

Epilogue

147

Montrul, 2010; Valdés, 2001, 2005), there is only scant and incoherent knowledge about Chinese heritage speakers (He & Xiao, 2008). What seems lacking is comparative research investigating heritage speakers alongside adult or child second language learners (Gass & Lewis, 2007; Montrul, 2008). SLA research focusing on Spanish as the L2 has shown striking parallels between L2 learners and heritage speakers in terms of morphosyntactic development (e.g. Montrul, 2010). This volume features a bias towards English-speaking learners of Chinese, as is true as well of the research published in prominent journals. Thus, while the understanding of this population is growing, attention must also be given to the learners who are speakers of other languages. This is not to suggest, however, that learners with different L1s will necessarily approach the same target language – in the case at hand, Chinese – in different ways; what it does suggest is that different relationships between the L1 and the L2 – for example, whether the two languages are typologically close or distant – may augur a different prospect for the learning process and outcome. Four decades of SLA research have taught us that, while the fundamental psycholinguistic processes of learning a second language are universal (Long, 2009), learners’ prior as well as their current language and experience may significantly modulate, if not alter, their operation, leading to changes in the rate, route and the ultimate outcome of acquisition (passim the SLA literature). All these issues and more must be taken on board en route to a robust understanding of the SLA of Chinese. To that end, it is hoped that this book will serve as a useful springboard.

Notes (1) ‘Context’ can be defined in terms of the discourse domain, the relationship between interlocutors, and the setting. (2) Generalizability is probably the biggest threat to qualitative studies where data analysis is interpretative, the results more or less pieced together from shreds of evidence here and there.

References Bardovi-Harlig, K. (in press) Documenting interlanguage development. In Z.-H. Han and E. Tarone (eds) Interlanguage: 40 Years Later. Amsterdam: John Benjamins. Bayley, R. and Tarone, E. (2011) Variationist perspectives. In S. Gass and A. Mackey (eds) The Routledge Handbook of Second Language Acquisition (pp. 41–56). New York: Routledge. Beebe, L. and Zuengler, J. (1983) Accommodation theory: An explanation for style shifting in second language dialects. In N. Wolfson and E. Judd (eds) Sociolinguistics and Language Acquisition (pp. 195–213). Rowley, MA: Newbury House. Bhatia, T.K. and Ritchie, W.C. (eds) (2009) The New Handbook of Second Language Acquisition (2nd edn). Bingley: Emerald Group Publishing. Cabo, D.P.Y. and Rothman, J. (2012) The (il)logical problem of heritage speaker bilingualism and incomplete acquisition. Applied Linguistics 33 (4), 450–455.

148

Studies in Second L anguage Acquisit ion of Chinese

Carroll, S. (1999) Input and SLA: Adults’ sensitivity to different sorts of cues to French gender. Language Learning 49, 37–92. Cazden, C., Cancino, H., Rosansky, E. and Schumann, J. (1975) Second Language Acquisition Sequences in Children, Adolescents, and Adults. Cambridge, MA: Harvard University Press. Doughty, C. and Long, M. (eds) (2003) The Handbook of Second Language Acquisition. Oxford: Blackwell. Duff, P. (2007) Case Study Research in Applied Linguistics. Mahwah, NJ: Lawrence Erlbaum. Ellis, N. (2002) Frequency effects in language processing. Studies in Second Language Acquisition 24 (2), 141–142. Ellis, R. (2008) The Study of Second Language Acquisition. Oxford: Oxford University Press. Ellis, R. (2012) Language Teaching and Language Pedagogy. Oxford: Blackwell. Gass, S. and Lewis, K. (2007) Perceptions about interactional feedback: Differences between heritage language learners and non-heritage language learners. In A. Mackey (ed.) Conversational Analysis in Second Language Acquisition (pp. 79–99). Oxford: Oxford University Press. Gass, S. and Mackey, A. (eds) (2012) The Routledge Handbook of Second Language Acquisition. New York: Routledge. Gass, S. and Selinker, L. (2008) Second Language Acquisition: An Introductory Course (3rd edn). New York: Routledge. Hakuta, K. (1976) A case study of a Japanese child learning ESL. Language Learning 26, 321–352. Han, Z.-H. (in press) From Julie to Wes to Alberto: Revising the construct of fossilization. In Z.-H. Han and E. Tarone (eds) Interlanguage: 40 Years Later. Amsterdam: John Benjamins. Han, Z.-H. and Liu, Z.H. (2013) Input processing of Chinese by ab initio learners. Second Language Research 29 (2), 145–164. Han, Z.-H., Park, E. and Combs, C. (2008) Textual input enhancement: Issues and possibilities. Applied Linguistics 29 (4), 597–618. He, A. and Xiao, Y. (eds) (2008) Chinese as a Heritage Language: Fostering Rooted World Citizenry. Honolulu, HI: University of Hawai’i Press. Huebner, T. (1983) A Longitudinal Analysis of the Acquisition of English. Ann Arbor, MI: Karoma. Lardiere, D. (2007) Ultimate Attainment in Second Language Acquisition. Mahwah, NJ: Lawrence Erlbaum. Long, M.H. (2003) Stabilization and fossilization in interlanguage development. In C. Doughty and M.H. Long (eds) The Handbook of Second Language Acquisition (pp. 487–535). Malden, MA: Blackwell. Long, M. (2009) Methodological principles for language teaching. In M. Long and C. Doughty (eds) The Handbook of Second Language Teaching (pp. 373–394). Oxford: Blackwell. Montrul, S. (2008) Review article: Second language acquisition welcomes the heritage language learner: Opportunities of a new field. Second Language Research 24 (4), 487–506. Montrul, S. (2010) How similar are adult second language learners and Spanish heritage speakers? Spanish clitics and word order. Applied Psycholinguistics 31 (1), 167–207. Pienemann, M. (1998) Language Processing and Second Language Acquisition: Processability Theory. Amsterdam: John Benjamins. Robinson, P. (ed.) (2012) The Routledge Encyclopedia of Second Language Acquisition. New York: Routledge. Schmidt, R. and Frota, S. (1986) Developing basic conversational ability in a second language: A case study of an adult learner of Portuguese. In R. Day (ed.) Talking to Learn: Conversation in Second Language Acquisition (pp. 237–326). Rowley, MA: Newbury House.

Epilogue

149

Schumann, J. (1978) The Pidginization Process: A Model for Second Language Acquisition. Rowley, MA: Newbury House. Selinker, L. (1972) Interlanguage. International Review of Applied Linguistics 10 (2), 209–231. Tarone, E. (1983) On the variability of interlanguage systems. Applied Linguistics 4, 142–163. Tarone, E. (in press) Enduring themes from the interlanguage hypothesis. In Z.-H. Han and E. Tarone (eds) Interlanguage: 40 Years Later. Amsterdam: John Benjamins. Valdés, G. (2001) Heritage language students: Profiles and possibilities. In J. Peyton, D. Ranard and S. McGinnis (eds) Heritage Languages in America: Preserving a National Resource (pp. 37–77). McHenry, IL: Center for Applied Linguistics and Delta Systems. Valdés, G. (2005) Bilingualism, heritage language learners, and SLA research: Opportunities lost or seized? Modern Language Journal 89 (3), 410–426.

Index

Achiba, M., 31, 32 Agency, 60, 61, 71, 72, 95, 96, 97 Albert, A., 80, 97 Ammar, A., 103, 105 Andres, H.P., 128 Au, T., 9

Chun, D.M., 128, 129, 137 Chung, K., 126 Chung, K.K.H., 126, 128, 137 Classifier, 106, 107, 111, 120 Coghlan, P., 85 Cognition hypothesis, 84 Conceptual restructuring, 4, 22, 83 Conceptual transfer, 4 Contrastive analysis, 30 Cook, V., 59, 60, 67 Corder, S. P., 82, 96, 98 Cowan, N., 120 Cross-case analysis, 63, 64 Cross-cultural speech act realization project (CCSARP), 33

Back, M., 57 Baddeley, A., 108, 120, 127 Bardovi-Harlig, K., 34, 143 Barron, A., 30, 43, 50 Bassetti, B., 126, 127 Beavers, J., 5 Belz, J., 59, 60 Berman, R., 13 Bley-Vroman, R., 82 Blum-Kulka, S., 30, 31, 32, 33, 37 Bowerman, M., 1, 2, 9 Breen, M., 80, 84, 85, 96, 97 Brooks, L., 58, 59 Brown, J.D., 1, 31 Brown, P., 31, 33 Budwig, N., 1 Built-in syllabus, 98 Bygate, M., 80

Dai, J-H., 7 Daneman, M., 108 DeFrancis, J., 127 de Graaff, R., 119 DeKeyser, R., 52 Demonstrative, 106 Diachronic analysis, 64, 67 Discourse completion test (DCT), 32, 53 Donato, R., 58 Dörnyei, Z., 80, 85 Doughty, C., 104 Duff, P., 85, 111, 146 Duncan, S.D., 1

Cadierno, T., 4, 5, 21 Candlin, C., 81 Caplan, D., 113 Carpenter, H., 104, 108, 119 Cartoon oral production task (COPT), 32 Characters, 127, 136, 139 Chinese as a foreign language (CFL), 31, 81, 109, 126 Choi, S., 1, 2, 9 Christensen, M., 111 Chu, C., 7 Chu, Y.-K., 127 Chuang, H.-Y., 126, 128

Egi, T., 104 Ellis, N.C., 4, 9 Ellis, R., 107, 142, 144 Erbaugh, M., 106, 111 Erlam, R., 11 Ervin-Tripp, S., 1 Everson, M. E., 129, 138 150

Inde x

Face-threatening acts (FACs), 31 Faerch, C., 30 Feng, L., 136 Form, 20, 108 Form-focused instruction, 86, 87, 104, 118, 145 Form-meaning mapping, 8, 52, 108, 119, 121 Foster, P., 96 Fujii, A., 108 Gass, S., 142, 147 Gatbonton, E., 108, 109, 120 Gathercole, V., 1 Gibbs, R., 21 Goldschneider, J., 107 Goo, J., 104, 105 Grammaticality judgment test (GJT), 113, 122 Grammaticalization, 4 Guba, E.G., 63 Guo, J., 1 Gu, W., 7 Hamada, M., 135 Han, Z-H., 4, 105, 142 Hanks, W., 6 Hanyu Pinyin, 126, 130 Hanyu Shuiping Kaoshi (HSK), 131 Hart, D., 57 Hasko, V., 5 Hendriks, B., 31, 32, 33, 50 Hickman, M., 1 Hill, T., 31, 32, 33, 50 Hitch, G. J., 127 Hong, W., 34, 52 Hooper, S., 136 Host community, 60 House, J., 30 Huberman, A. M., 63 Hue, C. W., 137 Hulstijn, J., 119 Iberri-Shea , G., 81 Identity, 59–60, 70, 71, 72 INFO Input, 82 Negotiation, 82 Feedback, 82 Output, 82 Iino, M., 60 Interlanguage, 5, 9, 83, 121

Isabelli-Garcia, C.L., 57 Ishida, M., 105 Iwashita, N., 105 Jarvis, S., 4 Jeon, K.S., 104, 119 Jin, H., 126, 128, 131 Jin, L., 57 Kasper, G., 30 Ke, C., 126, 127, 129, 131, 138 Keck, C., 81 Kenning, M., 82 Kerst, S., 137 Kim, E-J., 9 Kim, H-S., 9 Kinginger, C., 57, 60, 72 Koda, K., 135 Kondo-Brown, K., 9 Kong, D-K., 9 Kormos, J., 97 Kramsch, C., 58, 59, 60 Ku, H.-Y., 126, 128 Kuo, M.-L.A., 136 Lafford, B., 57 Language socialization, 57, 58, 60, 61 Lantolf, J.P., 59, 60, 72 Lapkin, S., 57 Lee, Y-G., 9 Leech, G.N., 33 Leeman, J., 104 Lee-Thompson, L. C., 138 Levenston, W.A., 30 Levin, B., 5 Levin, D. M., 57 Levinson, S.D., 31, 33 Lexicalization., 9 Li, D., 111 Li, J., 127 Li, S., 103, 104, 105 Lincoln, Y. S., 63 Liu, L., 129 Liu, X., 106 Liu, Y., 7 Liven, E., Loewen, S., 104, 112 Logie, R., 120 Long, M. H., 58, 80, 81, 82, 104, 142 Lyster, R., 103, 105

151

152

Studies in Second L anguage Acquisit ion of Chinese

MacGregor, D., 104, 119 Mackey, A., 104, 105, 108, 109, 119, 142 MacNeill, D., 1 Maeng, J., 88, 92 Magnan, S.S., 57 Manner of motion verbs, 4 Martens, R.L., 128 Matrix display, 63 Mayer, R.E., 128, 137 McDonough, K., 104, 119 McEnery, T., 106 McGinnis, S., 9 Miles, M. B., 63 Mixed-Sensory Mode (MSM), 126 Mori, H., 104 Mori, J., 85 Morphosyntax, 9, 104, 119 Motion events, 2, 4, 6, 18, 143 Nakamura, K., 1 Narasimhan, B., 1 Navarro, S., 5 Neath, I., 135 Neff van Aertselaer, J., 30 Negative evidence, 104, 120 Nicoladis, E., 5 Norris, J., 84, 96 Noticing, 84, 104, 108, 119, 120, 121 Ochs, E., 57, 60 Odlin, T., 4 Oh, K-J., 4 Olshtain, E., 30 Ortega, L., 11, 23, 80, 85, 97 Orthographic features, 126, 133–137 Paivio, A., 127 Pan, W., 7 Paradis, M., 120 Path, 1, 3, 7, 9, 16, 21 Pavlenko, A., 4, 59, 60 Peer interaction, 10, 58–59, 67, 144 Perfective, 106–107, 111, 115–120, 122 Perfetti, C., 126 Petersen, C., 128 Peyraube, A., 5 Phan, H.P., 137 Philp, J., 104, 105, 112 Ping, M., 129, 131 Pinyin, 24, 25, 35, 126, 127, 130, 136 Plass, J.L., 128, 129, 137 Politeness theory, 30, 31

Positive evidence, 104 Pragmatic development, 10, 30, 31, 52, 143 Production, 4, 13–17, 84, 129, 131, 139 Prompt, 12, 103, 105, 128, 129, 138 Puetz, M., 30 Rabie, S., 11, 23 Ranta, L., 103, 104, 120 Recall, 114, 120, 127, 129, 135, 136 Recasts, 104, 105–109, 119 Richard, C. A., 126 Recognition, 61, 71, 129, 132, 133, 136, 139 Robert, S., 1 Robinson, P., 4, 9, 84, 95, 142 Rose, K. R., 30, 31, 32, 33, 34 Romo, L., 9 Ruiz, L., 5, 21 Russell, J., 104 Sagarra, N., 109, 120 Saito, K., 103, 104, 105 Samuda, V., 81, 82, 83 Satellite-framed language (S-language), 2, 3, 142 Sawyer, M., 120 Scaffolding, 59, 83, 119 Scarcella, R., 31 Schauer, G. A., 30, 32, 33, 39, 43 Schmidt, R., 30, 31, 83, 146 Second language acquisition (SLA), 5, 58, 80, 103, 142 Searle, J. R., 30 Sergent, W. K., 129 Sheen, Y., 103, 104, 112, 118, 119, 121 Shen, H., 126, 127 Shen, J., 5 Siegal, M., 60 Skehan, P., 80, 81, 83, 87 Slobin, D.I, 1–4, 6, 9 Sociocultural theory (SCT), 58 Sociopragmatics, 31 Spada, N., 103, 104, 105 Spatial relations, 1, 9, 21, 22 Speech acts, 30, 143 Stam, G., 5, 21 Storch, N., 59, 67, 72 Study abroad, 10, 57, 58, 60, 67, 72, 81, 85, 95, 143, 145 Sun, X., 11, 30, 52 Surprenant, A. M., 135 Svenconis, D. J., 137

Inde x

Swain, M., 58, 59, 71, 82 Swierzbin, B., 80 Synchronic analysis, 64 Tabbers, H.K., 128 Taft, M., 126 Talmy, L., 1, 3, 5 Target culture, 57, 60 Tarone, E., 80, 143 Task-as-workplan, 85 Task-based language teaching (TBLT), 10, 80, 144 Task complexity, 84, 96 Task difficulty, 84 Task-in-process, 97 Tatsumi, T., 108 Tavakoli, P., 96 Tham, S., 5 Thinking-for-speaking hypothesis, 4, 9, 143 Thomas, J., 33, 39 Thompson, S., 106 Tocalli-Beller, A., 58, 59, 71 Tracy, N., 81 Transfer, 4, 22, 30, 128 Trebits, A., 97 Trofimovich, P., 108, 109, 120 Trosborg, A., 31, 32, 33 Uptake, 104 Valdés, G., 25, 147 Van Den Berg, M., 106 Van den Branden, K., 80, 81, 96 Van Lier, L., 58, 60, 67 van Merrienboer, J. J. G., 128 VanPatten, B., 52, 82

153

Verb-framed language (V-language), 2, 3 Vygotsky, L.S., 58 Wa-Mbaleka., S., 81 Wang, G.C., 127 Wang, M., 126 Watanabe, Y., 58, 59, 71 Waters, G., 113 Wen, X-H., 9 Wittrock, M. C., 137 Williams, J., 82 Willis, D., 81, 82, 83, 86, 88 Wilkinson, S., 57 Within-case analysis, 63, 64 Working memory, 10, 103, 108–109, 113, 117, 120, 126, 127, 145 Wu, S., 126 Wu, S-L., 12, 18, 21 Wu, Y., 126 Xiao, R., 106 Xiao, Y., 129, 131, 147 Xie, T., 127 Yang, J., 111 Yang, Y., 105, 118 Yao,T., 127 Yao, T-C., 7, 127 Yin, R. K., 63 You, H., 136 Zhang, D., 30, 52 Zhang, H., 106 Zhang, W., 135 Zhang, Y., 33, 36, 41 Zhou, Y., 11, 23, 66