The Cambridge Handbook of Language Learning 9781108333603, 9781108420433, 2018047193

Citation preview

The Cambridge Handbook of Language Learning Providing a comprehensive survey of cutting-edge work on second language learning, this handbook, written by a team of leading experts, surveys the nature of second language learning and its implications for teaching. Prominent theories and methods from linguistics, psycholinguistics, ­processing-based, and cognitive approaches are covered and organized thematically across sections dealing with skill development, individual differences, pedagogical interventions and approaches, and context and environment. This state-of-the-art handbook will interest researchers in second language studies and language education, and will also reach out to advanced undergraduate and graduate students in these and other related areas. j o h n w . s c h w i e t e r is an Associate Professor of Spanish and Linguistics and Faculty of Arts Teaching Scholar at Wilfrid Laurier University. He has written and edited numerous books and articles including The Handbook of the Neuroscience of Multilingualism (2019), The Handbook of Translation and Cognition (2017, with Aline Ferreira), and The Cambridge Handbook of Bilingual Processing (Cambridge, 2015). He is also Founding Co-Editor of Cambridge Elements in Second Language Acquisition (with Alessandro Benati).

is a Professor of English and Applied Linguistics and the Head of the Department of English at the American University of Sharjah, UAE. He is author and editor of several books and articles including The Handbook of Advanced Proficiency in Second Language Acquisition (2018, with Paul A. Malovrh). He is also Founding Co-Editor of Cambridge Elements in Second Language Acquisition (with John W. Schwieter).

alessandro benati

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

cambridge handbooks in language and linguistics

Genuinely broad in scope, each handbook in this series provides a complete state-of-the-field overview of a major sub-discipline within language study and research. Grouped into broad thematic areas, the chapters in each volume encompass the most important issues and topics within each subject, offering a coherent picture of the latest theories and findings. Together, the volumes will build into an integrated overview of the discipline in its entirety.

Published titles The Cambridge Handbook of Phonology, edited by Paul de Lacy The Cambridge Handbook of Linguistic Code-switching, edited by Barbara E. Bullock and Almeida Jacqueline Toribio The Cambridge Handbook of Child Language, Second Edition, edited by Edith L. Bavin and Letitia Naigles The Cambridge Handbook of Endangered Languages, edited by Peter K. Austin and Julia Sallabank The Cambridge Handbook of Sociolinguistics, edited by Rajend Mesthrie The Cambridge Handbook of Pragmatics, edited by Keith Allan and Kasia M. Jaszczolt The Cambridge Handbook of Language Policy, edited by Bernard Spolsky The Cambridge Handbook of Second Language Acquisition, edited by Julia Herschensohn and Martha Young-Scholten The Cambridge Handbook of Biolinguistics, edited by Cedric Boeckx and Kleanthes K. Grohmann The Cambridge Handbook of Generative Syntax, edited by Marcel den Dikken The Cambridge Handbook of Communication Disorders, edited by Louise Cummings The Cambridge Handbook of Stylistics, edited by Peter Stockwell and Sara Whiteley The Cambridge Handbook of Linguistic Anthropology, edited by N. J. Enfield, Paul Kockelman and Jack Sidnell The Cambridge Handbook of English Corpus Linguistics, edited by Douglas Biber and Randi Reppen The Cambridge Handbook of Bilingual Processing, edited by John W. Schwieter The Cambridge Handbook of Learner Corpus Research, edited by Sylviane Granger, Gaëtanelle Gilquin and Fanny Meunier The Cambridge Handbook of Linguistic Multicompetence, edited by Li Wei and Vivian Cook The Cambridge Handbook of English Historical Linguistics, edited by Merja Kytö and Päivi Pahta The Cambridge Handbook of Formal Semantics, edited by Maria Aloni and Paul Dekker The Cambridge Handbook of Morphology, edited by Andrew Hippisley and Greg Stump The Cambridge Handbook of Historical Syntax, edited by Adam Ledgeway and Ian Roberts The Cambridge Handbook of Linguistic Typology, edited by Alexandra Y. Aikhenvald and R. M. W. Dixon The Cambridge Handbook of Areal Linguistics, edited by Raymond Hickey The Cambridge Handbook of Cognitive Linguistics, edited by Barbara Dancygier

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

The Cambridge Handbook of Japanese Linguistics, edited by Yoko Hasegawa The Cambridge Handbook of Spanish Linguistics, edited by Kimberly L. Geeslin The Cambridge Handbook of Bilingualism, edited by Annick De Houwer and Lourdes Ortega The Cambridge Handbook of African Linguistics, edited by H. Ekkehard Wolff The Cambridge Handbook of Systemic Functional Linguistics, edited by Geoff Thompson, Wendy L. Bowcher, Lise Fontaine, and David Schönthal

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

The Cambridge Handbook of Language Learning Edited by John W. Schwieter Wilfrid Laurier University, Canada

Alesssandro Benati American University of Sharjah, UAE

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108420433 DOI: 10.1017/9781108333603 © Cambridge University Press 2019 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2019 Printed and bound in Great Britain by Clays Ltd, Elcograf S.p.A. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in–Publication Data Names: Schwieter, John W., 1979– editor. | Benati, Alessandro G., editor. Title: The Cambridge handbook of language learning / edited by John Schwieter, Wilfrid Laurier University, Ontario; Alesssandro Benati, American University of Sharjah, UAE. Description: Cambridge, United Kingdom; New York, NY: Cambridge University Press, 2019. | Series: Cambridge handbooks in language and linguistics | Includes bibliographical references and index. Identifiers: LCCN 2018047193 | ISBN 9781108420433 (hardback) Subjects: LCSH: Second language acquisition. | Language and languages – Study and teaching. | BISAC: LANGUAGE ARTS & DISCIPLINES / Linguistics / General. Classification: LCC P118.2 .C3554 2019 | DDC 401/.93–dc23 LC record available at https://lccn.loc.gov/2018047193 ISBN 978-1-108-42043-3 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this ­publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Downloaded from https://www.cambridge.org/core. Columbia University - Law Library, on 17 Aug 2019 at 01:53:04, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Contents

List of Figures List of Tables List of Contributors Acknowledgements

page x xi xiii xxv

Introduction  John W. Schwieter and Alessandro Benati 1 Part I Theories 11   1 Formal Linguistic Approaches to Adult L2 Acquisition and Processing  Jason Rothman, Fatih Bayram, Ian Cunnings, and Jorge González Alonso13   2 Cognitive Approaches to Second Language Acquisition  Nick C. Ellis and Stefanie Wulff41   3 The Qualitative Science of Vygotskian Sociocultural Psychology and L2 Development  Rémi A. van Compernolle62   4 Theoretical Frameworks in L2 Acquisition  John Truscott and Michael Sharwood Smith84 Part II Methods 109   5 Qualitative Classroom Methods  Peter I. De Costa, Wendy Li, and Hima Rawal111   6 Experimental Studies in L2 Classrooms  Charlene Polio and Jongbong Lee137   7 Action Research: Developments, Characteristics, and Future Directions  Anne Burns166   8 Classroom Observation Research  Nina Spada 186   9 Psycholinguistic and Neurolinguistic Methods  Leah Roberts 208

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:53:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

viii

Contents

Part III Skill Development 231 10 Interaction in L2 Learning  Jaemyung Goo 233 11 Speaking  Dustin Crowther and Susan M. Gass 258 12 Second Language Listening: Current Ideas, Current Issues  John Field283 13 Contemporary Perspectives on L2 Upper-Register Text Processing  Elizabeth B. Bernhardt and Cici Malik Leffell320 14 Language Learning Through Writing: Theoretical Perspectives and Empirical Evidence  Rosa M. Manchón and Olena Vasylets341 Part IV Individual Differences 363 15 Working Memory in L2 Learning and Processing  Zhisheng (Edward) Wen and Shaofeng Li365 16 Language Aptitudes in L2 Acquisition  Gisela Granena 390 17 Language Learner Motivation: What Motivates Motivation Researchers?  Stephen Ryan409 18 A New Look at “Age”: Young and Old L2 Learners  Carmen Muñoz430 19 Identity  Ron Darvin and Bonny Norton 451 Part V Pedagogical Interventions and Approaches 475 20 Pedagogical Interventions to L2 Grammar Instruction  Alessandro Benati and John W. Schwieter477 21 Task-Based Language Learning  Michael H. Long, Jiyong Lee, and Kyoko Kobayashi Hillman500 22 Task and Syllabus Design for Morphologically Complex Languages  Roger Gilabert and Joan Castellví527 23 Proficiency Guidelines and Frameworks  David Little 550 24 Technology-Mediated Language Learning  Carol A. Chapelle 575 25 Content-Based L2 Teaching  Hossein Nassaji and Eva Kartchava 597 26 Conceptions of L2 Learning in Critical Language Pedagogy  Graham Crookes621 Part VI Context and Environment 647 27 Bilingual Education and Policy  Christine Hélot and Ofelia García 649 28 Heritage Language Instruction  Kim Potowski and Sarah J. Shin 673 29 Minority Languages at Home and Abroad: Education and Acculturation  Aline Ferreira, Viola G. Miglio, and John W. Schwieter696 30 Study Abroad and Immersion  Jane Jackson and John W. Schwieter727 31 Teacher Education: Past, Present, and Future  Peter Swanson 751

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:53:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Contents

ix

Part VII Moving Forward 775 32 Future Directions in Language Learning and Teaching  Susan M. Gass777 Index 799

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:53:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Figures

  4.1. Simplified MOGUL architecture. page 91 11.1. ACTFL spring 2017 results for speaking, listening, and reading. 260 12.1. Simplified cognitive model of the listening process. 286 12.2. Topics favoured in L2 listening research. 295 14.1. Overview of the origin and development of research on the language learning potential of writing and WCF. 342 14.2. Main directions of research on the language learning potential of WCF. 353 19.1. Darvin and Norton’s (2015) Model of Investment. 457 28.1. Number of languages spoken in the fifteen largest metro areas in the United States. 678 28.2. The cultural iceberg model (adapted from Hall, 1976/1989).683 32.1. Article types appearing in Language Learning 1967–1972 and 1973–1979. 782

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:54:42, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Tables

  2.1. A contingency table showing the four possible combinations of events showing the presence or absence of a target Cue and an Outcome. page 50   5.1. CBQR paradigms. 116   6.1. Recent experimental classroom studies. 149 11.1. Number of students who took proficiency tests administered at Michigan State University from 2014–2017 in four languages. 259 17.1. Four stages in the development of L2 motivation theory and research. 421 21.1. Grammar-based and task-based language learning and teaching. 521 22.1. Terminological clarification of complexity and difficulty. 532 23.1. Summary of ACTFL Proficiency Guidelines for speaking. 555 23.2. CEFR self-assessment grid. 557 23.3. CEFR descriptors relevant to spoken interaction at level A2. 560 23.4. One-directional alignment of ACTFL assessments with the CEFR. 562 23.5. Length of time needed to achieve Novice, Intermediate, and pre-Advanced proficiency levels as specified in ACTFL Performance Guidelines for K–12.566 28.1. Factors related to heritage language (HL) proficiency. 682 28.2. Resources for heritage language educators. 687 29.1. The top ten languages, after English, spoken at home in California. 700 29.2. Linguistic competence of Basque in the Basque Autonomous Community in 1981 and 2011. 712

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xii

List of Tables

29.3. Linguistic competence of Basque in the Basque Autonomous Community by age group in 1981 and 2011. 29.4. Attitudes towards the promotion of Basque among 16+ year-olds across different historically Basque communities in 1991 and 2012. 29.5. Use of Basque in the BAC according to social context in 1991 and 2016. 32.1. SLRF conferences and themes from 1990–2018.

712 714 714 783

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:12, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Contributors

The C ontributors are international experts based at and/or affiliated with institutions and research centres in Australia, Canada, England, France, Hong Kong, Japan, Ireland, Macau, New Zealand, Norway, Scotland, South Korea, Spain, Taiwan, and the United States. These scholars include: Fatih Bayram is a Marie Sktodowska – Curie Postdoctoral Fellow at the UiT the Arctic University of Norway. His research covers various types of child and adult bilingualism with a primary focus on heritage language bilinguals, a specific subtype of bilingual first language learners. His research uses online and offline psycholinguistic methods to study language development and processing in later childhood and outcomes of development in adulthood. alessandro benati is a Professor of English and Applied Linguistics and the Head of the Department of English at the American University of Sharjah, UAE. He is author and editor of several books and articles including The Handbook of Advanced Proficiency in Second Language Acquisition (2018, with Paul A. Malovrh). He is also Founding Co-Editor of Cambridge Elements in Second Language Acquisition (with John W. Schwieter). Elizabeth B. Bernhardt is a Professor of German Studies and the John Roberts Hale Director of the Stanford Language Center at Stanford University. She has spoken and written on second-language reading, teacher education, and policy and planning for foreign- and second-­ language programmes. Her book, Reading Development in a Second Language (1991), earned the Modern Language Association’s Mildenburger Prize as well as the Edward Fry Award from the National Reading Conference as an outstanding contribution to literacy research. She is the author of Understanding Advanced Second Language Reading (2010). Anne Burns is a Professor of TESOL at the University of New South Wales, Professor Emerita at Aston University, and an Honorary Professor at the

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xiv

List of Contributors

University of Sydney and Education University in Hong Kong. She is a co-academic advisor for the Oxford Applied Linguistics book series (with Diane Larsen Freeman) and co-editor of the Research and Resources in Language Teaching book series (with Jill Hadfield). Her research interests span second language teacher education, discourse analysis, the teaching of speaking, and action and qualitative research in applied linguistics. Her latest book is The Cambridge Guide to Learning a Second Language (Cambridge, 2018, with Jack C. Richards). Joan Castellví is an Associate Professor in the Slavic Studies section of the Department of Modern Languages and Literatures and English Studies at the University of Barcelona. His teaching activity is centred on Russian language and linguistics and his current research focuses on second language acquisition, teaching Russian as a foreign language, task based language teaching, and task design. He is a member of the research group CLiC (the Language and Computation Centre). His most recent publications deal with task design for learning morphologically rich languages, the acquisition of vocabulary, and teaching Russian as a foreign language outside the language environment. Carol A. Chapelle is a Distinguished Professor of Liberal Arts and Sciences at Iowa State University. She is editor of The Encyclopedia of Applied Linguistics (2013) and co-editor of the Cambridge Applied Linguistics book series (with Susan Hunston). She is past president of the American Association for Applied Linguistics and former editor of TESOL Quarterly. Her recent books include The Handbook of Technology and Second Language Teaching and Learning (2017, with Shannon Sauro) and Teaching Culture in Introductory Foreign Language Textbooks (2016). She is currently working on projects that help to connect evaluation methods for learning materials and validation methods for assessment. Graham V. Crookes is a Professor (and at the time of writing, Chair) of the Department of Second Language Studies at the University of Hawai’i. His main research interests are critical language pedagogy and development of language teachers’ philosophies of teaching. His recent books are Critical ELT in Action (2013) and Values, Philosophies, and Beliefs in TESOL: Making a Statement (Cambridge, 2009). Dustin Crowther holds a PhD in Second Language Studies from Michigan State University. He previously completed his MA in Applied Linguistics at Concordia University in Montréal, Canada. His research interests include second language pronunciation, the promotion of mutual intelligibility in multilingual and multicultural contact, World Englishes, and research methodologies. His research has been published in a wide range of journals, including Studies in Second Language Acquisition, The Modern Language Journal, and TESOL Quarterly. Ian Cunnings is a Lecturer in Psycholinguistics at the University of Reading. Before arriving at Reading, he completed his PhD in Psycholinguistics

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

List of Contributors

xv

and Neurolinguistics at the University of Essex and a British Academy Postdoctoral Fellowship at the University of Edinburgh. His research interests are in psycholinguistics and multilingualism. In particular, he conducts research that investigates the similarities and differences between native and non-native sentence comprehension. His most recent research examines how the working memory operations that subserve successful sentence processing can help our understanding of native and non-native language comprehension. Ron Darvin is a Lecturer and Vanier Scholar in the Department of Language and Literacy Education at the University of British Columbia. His research interests include identity, social class, and digital literacy, and he has published articles in TESOL Quarterly, Annual Review of Applied Linguistics, and Journal of Language, Identity and Education. He is the recipient of the 2017 Language and Social Processes SIG Emerging Scholar Award of the American Educational Research Association and a corecipient of the 2016 TESOL Award for Distinguished Research. Peter I. De Costa is an Associate Professor in the Department of Linguistics, Germanic, Slavic, Asian and African Languages at Michigan State University, where he also holds a joint appointment in the department of Teacher Education in the College of Education. He is the author of The Power of Identity and Ideology in Language Learning: Designer Immigrants Learning English in Singapore (2016) and editor of Ethics in Applied Linguistics: Language Researcher Narratives (2016). His work has appeared in ELT Journal, Language Learning, Language Policy, Language Teaching, Research in the Teaching of English, System, TESOL Journal, TESOL Quarterly, and The Modern Language Journal. He is the co-editor of TESOL Quarterly (with Charlene Polio). Nick C. Ellis is a Professor of Psychology and Linguistics and Research Scientist in the English Language Institute at the University of Michigan. His research interests include language acquisition, cognition, emergentism, corpus linguistics, cognitive linguistics, applied linguistics, and psycholinguistics. His recent books include Usage-Based Approaches to Language Acquisition and Processing: Cognitive and Corpus Investigations of Construction Grammar (2016, with Ute Römer and Matthew O’Donnell), Language as a Complex Adaptive System (2009, with Diane Larsen-Freeman), and The Handbook of Cognitive Linguistics and Second Language Acquisition (2008, with Peter Robinson). He serves as General Editor of Language Learning. Aline Ferreira is an Assistant Professor of Linguistics in the Department of Spanish and Portuguese at the University of California, Santa Barbara where she is the Director of the Bilingualism, Translation, and Cognition Laboratory. She is the co-editor with John W. Schwieter of The Handbook of Translation and Cognition (2017), Psycholinguistic and Cognitive Inquiries into Translation and Interpreting (2015), and The Development of Translation Competence: Theories and Methodologies from Psycholinguistics and Cognitive

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xvi

List of Contributors

Science (2014). She has also published studies in journals such as Intercultural Education, International Journal of Bilingual Education and Bilingualism, Reading and Writing, Translation and Interpreting Studies, and Translation, Cognition and Behavior, among others. John Field is Reader in Cognitive Approaches to Language Learning in the Centre for Research in English Language Learning and Assessment at the University of Bedfordshire, where he conducts research and advises on the testing of second language listening. He formerly taught psycholinguistics at the University of Reading and has researched and written on second language listening for some thirty-five years. His Listening in the Language Classroom (Cambridge, 2008) is a standard text in the area. His thinking is informed by his background in psycholinguistics, which provides insights into the listening process and into the cognitive demands faced by second language listeners. An earlier career as an ELT teacher-trainer, schools inspector, and materials writer enables him to build bridges between theory and practice. His most recent publication, Rethinking the Second Language Listening Test (2018), challenges many received assumptions about how second language listening can be validly assessed. Ofelia García is a Professor in the PhD programmes in Urban Education and Hispanic and Luso Brazilian Literatures and Languages at the Graduate Center of the City University of New York. She is the General Editor of the International Journal of the Sociology of Language and the Co-Editor of Language Policy (with Helen Kelly-Holmes). Among her best-known books are Bilingual Education in the 21st Century (2009) and the BAAL Book Award recipient Translanguaging: Language, Bilingualism and Education (2014, with Li Wei, 2015). In 2017, she received the Charles Ferguson Award in Applied Linguistics from the Center for Applied Linguistics in Washington, DC, and the Lifetime Career Award from the Bilingual Education special interest group of the American Educational Research Association. Susan M. Gass is a University Distinguished Professor at Michigan State University, where she serves, at the time of writing, as Director of the English Language Center, Co-Director of the Center for Language Education and Research, and the Co-Director of the Center for Language Teaching Advancement. She has published more than thirty books and more than 150 articles in the field of second language acquisition, with works translated into Arabic, Russian, Korean, and Chinese. She is the co-author of Second Language Acquisition: An Introductory Course (1994, with Larry Selinker) and of Second Language Research: Methodology and Design (2005, with Alison Mackey). She is the winner of many local, national, and international awards. She has served as President of the American Association for Applied Linguistics and of the Association Internationale de Linguistique Appliquée (International Association of Applied Linguistics). She is currently Editor of Studies in Second Language Acquisition.

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

List of Contributors

xvii

Roger Gilabert is an Associate Professor and researcher at the University of Barcelona. His research interests include second and foreign language production and acquisition, task design and task complexity, and individual differences in second language production and acquisition. He is involved in a project within the language acquisition research group (GRAL) led and coordinated by Carmen Muñoz that investigates the effects of subtitling in series and film on second language learning. He is also involved in a reading and technology project called iRead. He has published extensively in the areas of task design, task complexity, and complexity, accuracy, and fluency. His most recent publications include research on writing and its potential for second language learning. Jorge González Alonso is a Postdoctoral Researcher within the Language Acquisition, Variation and Attrition (LAVA) research group at UiT the Arctic University of Norway. His main lines of research concern the processing of morphology and morphosyntax by native and non-native speakers of English and Spanish, as well as determining the source(s) of cross-linguistic influence in third language acquisition with the help of psycholinguistic methodologies. Jaemyung Goo is an Associate Professor in the Department of English Education at Gwangju National University of Education. He obtained his PhD in Applied Linguistics at Georgetown University, Washington, DC. Subsequently, he began his teaching career as a Visiting Assistant Professor teaching graduate courses in the Department of Second Language Studies at Indiana University. His research articles have appeared in various journals including Studies in Second Language Acquisition, Language Learning, and Language Teaching. He has also contributed book chapters and encyclopedia entries on such issues as interaction, corrective feedback, working memory, and implicit and explicit instruction. His research interests include, but are not limited to, task-based language teaching, cognitive individual differences and second language acquisition, age factor, second language research methods, and instructed second language acquisition. Gisela Granena is an Associate Professor in the School of Languages of the Universitat Oberta de Catalunya. She has published research on the role of cognitive aptitudes in both instructed and naturalistic contexts, aptitude-treatment interactions, task-based language teaching, measures of implicit and explicit language knowledge, and the effects of early and late bilingualism on long-term second language achievement. Recent co-edited books include Sensitive Periods, Language Aptitude, and Ultimate L2 Attainment (2013, with Michael Long) and Cognitive Individual Differences in Second Language Processing and Acquisition (2016, with Daniel O. Jackson and Yucel Yilmaz). Christine Hélot is an Emeritus Professor of English at the University of Strasbourg. She taught courses in sociolinguistics, bilingual education,

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xviii

List of Contributors

and didactics of plurilingualism at the Graduate School of Education of the University of Strasbourg from 1991. Previously she held a post of Lecturer in Applied Linguistics at the National University of Ireland (Maynooth College), where she was the Director of the Language Centre. As a sociolinguist, her research focuses on language in education policies in France and in Europe, bi-multilingual education, intercultural education, language awareness, early childhood education, and children’s literature and multiliteracy. Her most recent publications include: L’éducation bilingue en France: Politiques linguistiques, modèles et pratiques (2016, with Jürgen Erfurt) and Children’s Literature in Multilingual Classrooms (2014, with Raymonde Sneddon & Nicola Daly). Kyoko Kobayashi Hillman is a PhD candidate in the Second Language Acquisition programme at the University of Maryland. Her research interests include needs analysis and input elaboration in task-based language teaching, second language research methodology, such as reaction times in psycholinguistic experiments, and incidental vocabulary learning in instructed second language acquisition. Her co-authored article “Achieving Epistemic Alignment in a Psycholinguistic Experiment” recently appeared in Applied Linguistics Review (2017, with Steven Ross and Gabriele Kasper). Jane Jackson is a Professor of Applied Linguistics in the Department of English at the Chinese University of Hong Kong. Her research centres on study abroad, intercultural communication, language and identity, eLearning, and internationalization. She is the recipient of the University’s 2013 Education Award and the Hong Kong University Grant Council’s (2016– 2018) Prestigious Fellowship under the Humanities and Social Science Scheme. Her recent authored books include Online Intercultural Education and Study Abroad: Theory into Practice (2019), Interculturality in International Education (2018), and Introducing Language and Intercultural Communication (2014). Her recent edited and co-edited books include Intercultural Interventions in Study Abroad (2018, with Susan Oguro) and The Routledge Handbook of Language and Intercultural Communication (2012). Eva Kartchava is an Associate Professor of Applied Linguistics at Carleton University. Her main research interest is to explore the processes involved in the acquisition of a second language in the classroom setting. She has published research on the relationship between corrective feedback and second language learning, noticeability of feedback, and the role of individual differences in the language learning process. Her recent book is Corrective Feedback in Second Language Teaching and Learning: Research, Theory, Applications, Implications (2017, with Hossein Nassaji), and her forthcoming books are Noticing Oral Corrective Feedback in the Second-Language Classroom: Background and Evidence and The Cambridge Handbook of Corrective Feedback in Language Learning and Teaching (Cambridge, with Hossein Nassaji). Jiyong Lee is a PhD student in the Second Language Acquisition programme at the University of Maryland. Her research interests include the

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

List of Contributors

xix

validation of task complexity manipulations, relationships among task complexity, language aptitude, working memory, and second language performance, negative feedback, and age affects and maturational constraints on second language acquisition. Jongbong Lee is a PhD student in the Second Language Studies programme at Michigan State University. He has taught English and Korean in various locations in Korea and the United States. His research interests include second language writing, corpus linguistics, interactional feedback, and language assessment. Shaofeng Li is an Associate Professor of Second/Foreign Language Education at Florida State University and a Yunshan Chair Professor at Guangdong University of Foreign Studies. He received his PhD from Michigan State University and has worked in China, New Zealand, and the US. He has extensive teaching experience in various instructional and cultural settings. His main research interests include language aptitude, working memory, form-focused instruction, task-based language teaching, corrective feedback, and research methods. His publications have appeared in Applied Linguistics, International Review of Applied Linguistics, Language Learning, Language Teaching Research, The Modern Language Journal, and Studies in Second Language Acquisition, among others. Wendy Li is a PhD student in the Second Language Studies programme at Michigan State University. Before joining the programme in 2015, she taught English as a foreign language in different educational institutions in China for more than two years. Her research interests include instructed second language acquisition, language teacher identity, agency, emotions, and language socialization. She holds an MA in TESOL from Lancaster University. David Little is a Fellow Emeritus of Trinity College Dublin, where he established the Centre for Language and Communication Studies in 1979. His principal research interests are the theory and practice of learner autonomy in language education, the exploitation of linguistic diversity in schools and classrooms, and the application of the Common European Framework of Reference for Languages to the design of second language curricula, teaching, and assessment. He wrote one of the preparatory studies for the CEFR, on strategic competence, and between 2000 and 2010 he was in turn Member, Vice-Chair, and Chair of the European Language Portfolio Validation Committee. Michael H. Long is a Professor of Second Language Acquisition at the University of Maryland. He is the author of over 100 journal articles and book chapters; his recent books include The Handbook of Second Language Acquisition (2003, with Catherine J. Doughty), Second Language Needs Analysis (Cambridge, 2005), Problems in SLA (2007), The Handbook of Language Teaching (2009, with Catherine J. Doughty), Sensitive Periods, Language Aptitude, and

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xx

List of Contributors

Ultimate L2 Attainment (2013, with Gisela Granena), and Second Language Acquisition and Task-Based Language Teaching (2015). In 2009, he was awarded a Doctorate Honoris Causa by Stockholm University for his contributions to the field of second language acquisition. In 2017, he received a Lifetime Achievement Award from the International Association for Task-Based Language Teaching. Cici Malik Leffell is the Research Coordinator of the Language Center at Stanford University where she collaborates on second-language research projects with Elizabeth Bernhardt. She holds a certificate in language programme management from the Center and has also taught and published on Iberian language and literature. Rosa M. Manchón is a Professor of Applied Linguistics at the University of Murcia, where she teaches undergraduate courses in second language acquisition, as well as postgraduate courses in research methodology and second language learning and teaching. Her research interests and publications focus on cognitive aspects of instructed second language acquisition and, especially, second language writing. She has published extensively on the language learning potential of second language writing and written corrective feedback processing in the form of journal articles, book chapters, and edited books. Her recent books include Writing in Foreign Language Contexts: Learning, Teaching, and Research (2009), Learning to Write and Writing to Learn in an Additional Language (2011), L2 Writing Development: Multiple Perspectives (2012), Task-Based Language Learning—Insights from and for L2 Writing (2014, with Heidi Byrnes), Handbook of Second and Foreign Language Writing (2016, with Paul Matsuda) Writing and Language Learning: Advancing Research Agendas (forthcoming), and The Handbook of SLA and Writing (with Charlene Polio, forthcoming). She is the former Editor of the Journal of Second Language Writing. Viola G. Miglio is a Professor of Linguistics and Barandiaran Endowed Chair of Basque Studies at the University of California, Santa Barbara, and an affiliated faculty member at the University of Iceland. She trained in Germanic philology (University of Edinburgh) and linguistics (University of Maryland), and has taught at several institutions in Iceland, Italy, Mexico, and the US. She has published articles on linguistics, translation, and cultural and Basque studies and a book titled Interactions between Markedness and Faithfulness Constraints in Vowel Systems (2012). She has also co-edited Language Rights and Cultural Diversity (2013, with Xabier Irujo), The Protection of Cultural Diversity (2014), Basque Whaling in Iceland in the XVII Century (2015, with Xabier Irujo), and Approaches to Evidentiality in Romance (2015, with Josep Martines). Carmen Muñoz is a Professor of Applied English Linguistics at the University of Barcelona. Her research interests include the effects of age and context on second language acquisition, young learners in instructed

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

List of Contributors

xxi

settings, individual differences, and multimodality in language learning. She is the founder and coordinator of the GRAL research group at the University of Barcelona, where she has coordinated a dozen research projects with national and international funding, such as the BAF Project. She has edited books such as Intensive Exposure Experiences in Second Language Learning (2012) and Age and the Rate of Foreign Language Learning (2006). Some of her recent publications have appeared in journals such as Annual Review of Applied Linguistics, System, and Language Learning. Hossein Nassaji is a Professor of Applied Linguistics in the Department of Linguistics at the University of Victoria. He has authored and co-­authored numerous publications in the areas of second language acquisition, corrective feedback, form-focused instruction, grammar instruction, and task-based teaching. His recent books include Perspectives on Language as Action (2019, with Mari Haneda), Corrective Feedback in Second Language Teaching and Learning: Research, Theory, Applications, Implications (2017, with Eva Kartchava), and Interactional Feedback Dimension in Instructed Second Language Learning (2015). His forthcoming book is The Cambridge Handbook of Corrective Feedback in Language Learning and Teaching (Cambridge, with Eva Kartchava). Bonny Norton, FRSC, is a Professor and Distinguished University Scholar in the Department of Language and Literacy Education at the University of British Columbia. Her primary research interests are identity and language learning, critical literacy, and international development. Recent publications include a 2017 special issue on language teacher identity in The Modern Language Journal and a 2013 second edition of Identity and Language Learning. A Fellow of the Royal Society of Canada and the American Educational Research Association, she has a 2010 AERA Senior Research Leadership Award and was a 2016 corecipient of the TESOL Distinguished Research Award. Her work has been translated into Chinese, Portuguese, German, and French. Charlene Polio is a Professor in the Second Language Studies Program at Michigan State University. Her main area of research is second language writing, and she is particularly interested in the various research methods and measures used in studying writing, as well as the interface between the fields of second language writing and second language acquisition. She is the co-editor of TESOL Quarterly (with Peter De Costa), and her recent books include Understanding, Evaluating, and Conducting Second Language Writing Research (2016, with Debra Friedman) and Authentic Materials Myths: Applying Second Language Research to Classroom Teaching (2017, with Eve Zyzik). Kim Potowski is a Professor of Spanish Linguistics in the Department of Hispanic & Italian Studies at the University of Illinois at Chicago, where she also holds appointments in Latin American and Latino Studies, Curriculum and Instruction, and an affiliation with the Social Justice Initiative.

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xxii

List of Contributors

She is the Director of the Spanish for Heritage Speakers Programme and is the Founding Director of the heritage speaker study abroad programme in Oaxaca, Mexico. She has authored and edited books including IntraLatino Language and Identity: MexiRican Spanish (2016), El español de los Estados Unidos (2015, with Anna Maria Escobar), Heritage Language Teaching: Research and Practice (2014, with Sara Beaudrie and Cynthia Ducar), and Language Diversity in the USA (2011), as well as a heritage Spanish textbook Conversaciones escritas (2nd edn., 2017). Hima Rawal is a PhD student in the Second Language Studies programme at Michigan State University. She earned her MA in TESOL from Michigan State University and an MA in English Education from Tribhuvan University. She taught English as a foreign language for more than a decade at both K-12 and university levels (teacher education and teacher training) in Nepal. Her research interests are teacher identity and emotion, translanguaging, and heritage language acquisition in immigrant families. Leah Roberts is a Professor and Leader of the Centre for Language Learning and Use at the University of York. She works on real-time sentence comprehension in second language learners and bilinguals, using psycholinguistic and neurolinguistic methods, with a particular focus on cross-linguistic influences on grammatical processing. Jason Rothman is a Professor of Linguistics at UiT the Arctic University of Norway and an affiliate Professor in the Facul ad de Lenguas Education at Universidad Nebrija. He is the Executive Editor of the journal Linguistic Approaches to Bilingualism as well as Editor of the book series Studies in Bilingualism. Rothman’s research focuses on language acquisition and processing in children and adults in addition to the interface between domain general cognition and language. Some of his recent research has appeared in Bilingualism: Language and Cognition, Second Language Research, Studies in Second Language Acquisition, Language, Cognition and Neuroscience, and Frontiers in Psychology. Stephen Ryan is a Professor in the School of Culture, Media and Society at Waseda University, Tokyo. His research and publications cover various aspects of psychology in language learning, with his most recent books being The Psychology of the Language Learner Revisited (2015, with Zoltán Dörnyei) and Exploring Psychology in Language Learning and Teaching (2015, with Marion Williams and Sarah Mercer). He is currently Co-Editor of the Psychology in Language Learning and Teaching book series. john w. schwieter is an Associate Professor of Spanish and Linguistics and Faculty of Arts Teaching Scholar at Wilfrid Laurier University. He has written and edited numerous books and articles including The Handbook of the Neuroscience of Multilingualism (2019), The Handbook of Translation and Cognition (2017, with Aline Ferreira), and The Cambridge Handbook of Bilingual Processing (Cambridge, 2015). He is also Founding

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

List of Contributors

xxiii

Co-Editor of Cambridge Elements in Second Language Acquisition (with Alessandro Benati). Michael Sharwood Smith is an Emeritus Professor at Heriot-Watt University and Honorary Professorial Fellow at the University of Edinburgh. He is a Founding Editor of Second Language Research and has authored Introduction to Language and Cognition (2017), The Multilingual Mind: A Modular Processing Perspective (Cambridge, 2014, with John Truscott), and The Internal Context of Bilingual Processing (2019, with John Truscott). Although his current research interest is in language and cognition, he has also published widely on both theoretical and applied topics on various aspects of language learning and performance. Sarah J. Shin is a Professor of Education at the University of Maryland Baltimore County. She is an expert in bilingualism, heritage language education, and TESOL teacher preparation, and the author of Bilingualism in Schools and Society (2018), English Language Teaching as a Second Career (2017), and Developing in Two Languages (2005). She serves on the editorial boards of the Journal of Language, Identity, and Education and The International Multilingual Research Journal. Nina Spada is a Professor Emerita at the University of Toronto where she teaches courses in second language acquisition, classroom research in second language teaching and learning, and research methods. Her research investigates the effects of different types of form and meaning-based instruction on second language development. She is co-author of the award-winning book How Languages are Learned (1993, with Patsy Lightbown). She is the Co-Editor of the Language Learning and Teaching book series and the Oxford Key Concepts for the Language Classroom series. She is a past president of the American Association for Applied Linguistics. Peter Swanson is a Professor of Foreign Language Education at Georgia State University and is serving as Distinguished Visiting Professor at the US Air Force Academy in Colorado Springs. His research focuses on the characteristics of highly effective language teachers, recruitment and retention of language teachers, and the integration of technology into instruction. He is the author of five books and his research has been published in the US and abroad in journals such as the Canadian Modern Language Review, Hispania, and Foreign Language Annals. Currently, he serves as Past President of the American Council on the Teaching of Foreign Languages. John Truscott is a Professor in the Institute of Learning Sciences and Technologies and Adjunct Professor in the Center for Teacher Education at National Tsing Hua University in Taiwan. His primary research interest is the Modular Online Growth and Use of Language (MOGUL) framework, but he has also published extensively on the topic of error correction and in other areas of second language acquisition. He is the author

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

xxiv

List of Contributors

of Consciousness and Second Language Learning (2015) and co-author of The Multilingual Mind: A Modular Processing Perspective (Cambridge, 2014, with Mike Sharwood Smith) and The Internal Context of Bilingual Processing (2019, with Michael Sharwood Smith). Rémi A. van Compernolle is an Associate Professor and William S. Dietrich II Career Development Professor of Second Language Acquisition and French and Francophone Studies at the Carnegie Mellon University and is a Project Coordinator for the Center for Advanced Language Proficiency Education and Research at the Pennsylvania State University. His research extends cultural-historical psychology to second language development, with specific focus on second language pragmatics and sociolinguistics, concept-based instruction, dynamic assessment, and pedagogical interaction. In addition to numerous articles and book chapters, he has published two monographs, Sociocultural Theory and L2 Instructional Pragmatics (2014) and Interaction and Second Language Development: A Vygotskian Perspective (2015). Olena Vasylets is an Associate Professor of English in the Department of Modern Languages and Literatures and English Studies at the University of Barcelona. Her areas of research are cognitive processes in language learning, language-learning potential of second language writing, computer-assisted language learning, the role of mode (oral, written, multimodality) in second language acquisition, and language disorders. She has published articles in peer-reviewed journals such as Language Learning and Annual Review of Applied Linguistics. Zhisheng (Edward) Wen is an Associate Professor at Macao Polytechnic Institute, Macau SAR, China. He received his PhD from the Chinese University of Hong Kong. He has teaching and research interests in second language acquisition, cognitive sciences, and translation studies. His latest books include: Working Memory in Second Language Acquisition and Processing (2015, with Mailce Borges Mota and Arthur McNeill), Working Memory and Second Language Learning (2016), and Language Aptitude: Advancing Theory, Testing, Research and Practice (2019, with Peter Skehan, Adriana Biedron, Shaofeng Li, and Richard Sparks). Forthcoming is Researching L2 Task Performance and Pedagogy: In Honor of Peter Skehan (with Mohammad Ahmadian). Stefanie Wulff is an Assistant Professor in the Linguistics Department at the University of Florida. Her research interests are in quantitative corpus linguistics, second language acquisition, and student writing development. She is the author of Rethinking Idiomaticity (2008) and various journal articles and contributions to edited volumes. She is Editor-in-Chief of the journal Corpus Linguistics and Linguistic Theory and Co-Editor of the Cognitive Linguistics in Practice series (with Carita Paradis). She is the co-PI of the National Science Foundation-funded Technical Writing Project, a corpus-based research project seeking to improve technical student writing in the STEM disciplines.

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:55:43, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Acknowledgements

We are thankful to Helen Barton, Commissioning Editor (Linguistics) at Cambridge University Press for having suggested this timely handbook idea to us and for her efficient assistance throughout this project. We also acknowledge the excellent work, correspondance, and professionalism of Stephanie Taylor, Senior Editorial Assistant (Humanities & Social Sciences) at Cambridge University Press. We are also grateful to our editorial assistant, Rebecca Mueller, for her excellent work during the preparation of the manuscript. We gratefully acknowledge that financial support to hire an editorial assistant was received from a grant by Wilfrid Laurier University and a Social Sciences and Humanities Research Council of Canada Institutional Grant. Needless to say, The Cambridge Handbook of Language Learning would not exist had it not been for the dedication and hard work of individuals who contributed chapters. They have helped to build a comprehensive handbook which showcases prominent theories and findings on language learning, which we feel is a much-needed addition to the field of second language acquisition. Finally, we are extremely grateful to the more than sixty scholars— both internal and external to this project—who accepted our invitation to serve as anonymous peer-reviewers of the chapters. It is without a doubt that their knowledge and expertise have strengthened the content of this handbook and its implications for future research.

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:56:40, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 17 Aug 2019 at 01:56:40, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603

Introduction John W. Schwieter and Alessandro Benati

Second language learning is the study of how learners come to create a new language system with an often limited exposure to the second language. It is the study of how they can make use of that system during comprehension and speech production. In a general sense, a second language (L2) refers to a language that is acquired after the first language (L1) has been established in early childhood. Theory and research in second language acquisition has emphasized the complexity of acquisition processes. How learners process language, how they intake it and accommodate it into the new language system, and how they access the information for speech production are key areas of research in second language acquisition. Scholars in this field are mainly interested in exploring what the processes and key factors involved in language acquisition are. Research carried out in this context is often about learners and learning. However, the main findings of this research have implications for teachers and teaching. The Cambridge Handbook of Language Learning builds on this impetus and reports on topical areas and findings that have formulated and continue to define research trajectories in the field of L2 acquisition today. The authoritative, state-of-the-art works in this handbook are organized across the following seven parts: Part I: Theories Part II: Methods Part III: Skill Development Part IV: Individual Differences Part V: Pedagogical Interventions and Approaches Part VI: Context and Environment Part VII: Moving Forward

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

2

J o h n W. S c h wi e t e r a n d A l e ssa n d r o B e n at i

In Part I, the handbook opens with a set of chapters that are dedicated to discussing prominent theories and frameworks in L2 learning. Chapter 1 by Jason Rothman, Fatih Bayram, Ian Cunnings, and Jorge González Alonso opens the section with an exhaustive overview of the major contemporary theories, models, and hypotheses that drive formal linguistic and psycholinguistic research in adult L2 acquisition and processing. The authors focus on morphosyntax proper and its interfaces with semantics, pragmatics, and phonology (prosody). In addition to a historical overview of how the main questions of the paradigm have been refocused over time as formal linguistic and acquisition theories evolved, the chapter also presents how typically used methodologies have expanded, especially in the past decade or so. In Chapter 2, Nick C. Ellis and Stefanie Wulff present an overview of cognitive approaches to L2 learning which hypothesize that L2 learners acquire constructions (form–function mappings, conventionalized in a speech community) from language usage by means of general cognitive mechanisms (exemplar-based, rational, associative learning). The authors present a usage-based analysis of this in terms of fundamental principles of associative learning: low salience, low contingency, and redundancy all lead to form–function mappings being less well learned. From the review of experimental work of learned attention and blocking in L2 learning, the chapter presents educational interventions such as form-focused instruction which recruits learners’ explicit, conscious processing capacities and allows them to notice novel L2 constructions. Chapter 3 by Rémi A. van Compernolle presents a critical overview of the primary characteristics of Vygotsky’s sociocultural theory and how they relate to the qualitative science of L2 learning. The author makes extensions to these theories such that it can be invoked as an analytic lens for interpreting “naturalistic” data and can also be used as the basis for designing educational opportunities. Chapter 4 by John Truscott and Michael Sharwood Smith concludes Part I. In this chapter, the authors review four examples of narrow theoretical frameworks (interlanguage theory, creative construction approach, generative approach, and processability theory) that have served to guide L2 research. Following this, two broad frameworks are discussed: the complex adaptive systems approach, particularly as realized in the position paper of the Five Graces Group (2009), and the MOGUL framework of Sharwood Smith and Truscott (2014). The authors’ critical analysis of theoretical frameworks suggests that a broad framework to L2 learning should be emphasized in which assumptions about the nature of language, learning, and the mind must be made explicit and critically examined. Part II presents important methodological approaches in L2 learning research. Peter I. De Costa, Wendy Li, and Hima Rawal in Chapter 5 offer insight on several qualitative methodologies (e.g., case studies,

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

Introduction

3

ethnographies) that have been used in L2 classroom-based research along with methods (e.g., interviews, focus groups, observations) that are used in conjunction. The authors also review key theories in L2 acquisition, discourse analytic approaches, and discourse analytic tools that are generally adopted. The chapter demonstrates how some quantitative methods can effectively be paired with qualitative methods and contemporary constructs such as emotions and language ideologies in order to extend this research agenda, thereby creating a holistic understanding of the dynamics surrounding language acquisition. In Chapter 6, Charlene Polio and Jongbong Lee focus on experimental classroom research. They first present its historical context by explaining what early research investigated as well as some of the problems with the early research. Following this, the authors provide a summary of experimental classroom research published in the last ten years in which the researchers manipulated an independent variable within a classroom setting. The chapter concludes with a discussion of the drawbacks and challenges in experimental classroom research including the lack of random assignment or control groups and issues related to internal validity such as controlling for teacher and student variables. Chapter 7 by Anne Burns tracks the development of action research, a methodological approach used by practitioners in language educational contexts which takes a socio-constructivist approach and views teachers as agentive actors and investigators within their own social contexts. The author considers recent initiatives that have contributed to the spread of action research and discusses the findings about the impact on teachers who conduct action research. Nina Spada transitions to classroom observation research in Chapter 8. This method, often referred to as interaction analysis, typically involves the use of observation instruments that focus on broad-based, macro-level descriptions (e.g., the communicative orientation of instruction) as well as those that focus on micro-level analyses of specific instructional features (e.g., corrective feedback). Spada offers examples of both types and examines their structure and organization, the type of features included, and coding procedures. She also evaluates the strengths and weaknesses of the interaction analysis approach to L2 classroom observation in relation to other approaches such as discourse analysis and ethnographies. The last chapter of the section is Leah Roberts’ Chapter 9 which provides a state-of-the-art and critical overview of psycholinguistic and neurophysiological methods currently being used in L2 learning and processing. Some of the questions addressed by these methods include the extent to which L2 real-time processing is similar to or different from that of native speakers, how individual differences in factors such as proficiency and cognitive capacity affect L2 processing, and what the influences are of a

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

4

J o h n W. S c h wi e t e r a n d A l e ssa n d r o B e n at i

learner’s first language on the processes of production and comprehension. The methods reviewed by the author allow researchers to examine potentially distinct, qualitative changes in brain signatures that correlate with increased linguistic knowledge as well as learners’ error- and ­feedback-related responses to L2 linguistic stimuli. Part III reviews research on the development of various L2 skills. In Chapter 10, Jaemyung Goo illustrates the importance of interaction as a valuable opportunity for learners to refine their interlanguage through negotiation for meaning. The author discusses several factors that affect the benefits of interaction including noticing, feedback type, modified output opportunities, task type and complexity, cognitive individual differences, and language aptitude. Through these observations, Goo notes key issues that merit further research to gain a better understanding of L2 learning through interaction. Dustin Crowther and Susan M. Gass provide insight on the development of L2 speaking in Chapter 11. The authors begin by addressing the contract of L2 speaking from both a macro- and micro-perspective, drawing on a corpus of oral proficiency interviews across foreign languages (e.g., French, Spanish, Russian, Chinese). The authors also review phonology-, fluency-, and sociolinguistic-based factors that intersect with listener perceptions of L2 speech performance. The chapter finishes with a discussion on some of the variables that may influence the development of L2 speaking ability. John Field discusses the development of L2 listening in Chapter 12. The author first reviews the conventional comprehension approach to L2 listening and highlights recent arguments that demonstrate the shortcomings of these conventional approaches. Three recent developments are then discussed. The first is the notion of subdividing the skill of L2 listening into its component parts to enable them to be practised individually. A second development concerns the extent to which learners’ difficulties in L2 comprehension derive from the phonetic characteristics of connected speech. A third issue is the role of authentic speech in an L2 listening programme and the nature of the compensatory strategies in which L2 learners engage. In Chapter 13, Elizabeth B. Bernhardt and Cici Malik Leffell shift readers’ focus towards L2 reading. Their review of the state of the field shows that a significant amount of research in this area has focused on word- and sentence-level processing. The authors argue that this narrow focus diverts attention from two critical elements that have not been widely discussed: the implications of L2 extended text; and L2 comprehension. As such, Bernhardt and Malik Leffell explore the comprehension of complicated, extended L2 discourse, and their treatment of the interactive compensatory framework, as well as other research methodologies, suggests promising new insights into L2 upper-register text processing.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

Introduction

5

Chapter 14 by Rosa M. Manchón and Olena Vasylets closes the section by offering a review of research on how and why writing can be a site for L2 learning. The authors comment on empirical L2-oriented writing research looking at how writing and feedback processing can lead to language development, especially areas of development concerned with: (1) language processing in different conditions (individual and collaborative writing) and environments (pen-and-paper and computer-mediated writing); (2) task-related issues, including task complexity, task repetition, and task modality effects; and (3) attentional processes and potential learning outcomes during feedback processing. The authors conclude by ascertaining the implications of the theory and research reviewed and suggest ways of advancing research agendas in the area. Part IV synthesizes work on the role of individual learner differences in L2 learning. Zhisheng (Edward) Wen and Shaofeng Li report on working memory in Chapter 15. The authors discuss the nature and structure of the limited capacity of the human working memory (WM) system and elaborate on the implications of individual differences in WM has on L2 developmental domains including vocabulary, formulaic sequences and chunks, and morpho-syntactic constructions. Wen and Li also discuss the distinctive roles of executive WM components and functions in selective processes during L2 sub-skills learning and performance such as listening, speaking, reading, writing, and simultaneous interpreting. In Gisela Granena’s Chapter 16, the author explores the role of language aptitude in L2 learning and offers an approach to understanding and measuring aptitude components and functions in L2 learning. In addition to providing a review of past and present research on language aptitude, Granena addresses its predictive validity in naturalistic and instructed L2 contexts and analyses the implications for L2 teaching. Chapter 17 by Stephen Ryan examines some of the key shifts in recent thinking about the motivation to learn an L2, some of the methodological challenges these changes have posed, and some of the ways in which the field has responded to these challenges. The author situates motivation uniquely among learners’ individual differences and demonstrates that it has been enthusiastically embraced by both researchers and classroom practitioners, resulting in a fast-changing and expanding theoretical landscape. Ryan concludes the chapter by considering the implications of this period of high research activity and major upheaval for the future development of the field. In Chapter 18, Carmen Muñoz reviews research on age with respect to L2 learning. Recently, this body of work has acquired new perspectives, such as the concern with native-likeness, the view of age as a factor in isolation from other learner internal and external factors, or the generalization of age-related findings across different learning settings. Muñoz discusses these issues and argues for a revamping of

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

6

J o h n W. S c h wi e t e r a n d A l e ssa n d r o B e n at i

research on age and L2 learning to account for new priorities. Especially, the author notes that further research must investigate the two extremes of the age continuum, that of the very young school learners and that of older learners who may be learning an L2 because of immigration or as a hobby. Ron Darvin and Bonny Norton in Chapter 19 discuss L2 learner identity as a key factor in L2 acquisition. Through the lens of L2 learning as a social practice implicated in relations of power, the authors show how identity categories impact interactions in diverse learning contexts, positioning learners in different and sometimes unequal ways. Consequently, how L2 learners are recognized as legitimate speakers in these contexts shapes the extent to which they invest in their learning. On this backdrop, Darvin and Norton examine the notion that learners in the digital era need to navigate more complex, fluid spaces in which developing broader linguistic repertoires becomes increasingly crucial in negotiating their existing and imagined identities. The chapter demonstrates how an understanding of the complexities of learners’ identities can inform L2 teaching that extends the range of identities and opportunities available to L2 learners around the globe. Part V engages readers with a dialogue on the pedagogical interventions that facilitate L2 learning. Alessandro Benati and John W. Schwieter, in Chapter 20, examine how different pedagogical interventions to L2 grammar instruction have been addressed from a variety of theoretical frameworks (linguistic, cognitive, psycholinguistic, and neurolinguistic) and pedagogical perspectives (input, interaction, and output). The authors argue that although grammar instruction is limited and constrained due to several linguistic and processing factors, non-traditional pedagogical interventions might have some beneficial effects in terms of speeding up the rate of acquisition. After examining the characteristics, empirical evidence, and theoretical and pedagogical implications of different pedagogical interventions, the chapter outlines guiding principles which should be considered in language pedagogy. Michael H. Long, Jiyong Lee, and Kyoko Kobayashi Hillman discuss taskbased L2 learning in Chapter 21. The authors review a substantial body of research concerning several dimensions of tasks and task-based language learning, including task criticality, frequency, complexity, and difficulty, and the conditions under which tasks are performed. The chapter also synthesizes practical issues in the design and implementation of taskbased language teaching and with ways of dealing with problems that can arise in the field. A central issue in task-based L2 learning is how language may be attended to and processed in lessons in which the main focus is on meaning, and how such linguistic processing may interact with task design. Such is the topic of Chapter 22 by Roger Gilabert and Joan Castellví

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

Introduction

7

in which the authors address this issue with regard to morphologically rich languages such as Russian. The chapter provides an overview of communicative language teaching methods and then explores the issue of models of task design and task complexity and how they conceptualize the issue of linguistic complexity. The role of L2 proficiency and the use of tasks at lower levels of proficiency is addressed, and recommendations are made for researchers, syllabus designers, and educators who work with task-based programmes of morphologically-rich languages. In Chapter 23, David Little reviews the two most widely used proficiency guidelines and frameworks in L2 education worldwide: the American Council on the Teaching of Foreign Languages (ACTFL) Proficiency Guidelines and the Common European Framework of Reference for Languages (CEFR). The ACTFL guidelines comprise detailed descriptions of what learners can do at different proficiency levels, whereas the CEFR uses scaled descriptors to illustrate a taxonomic account of language use and the learner’s competences. The author explores the origin and objectives of both frameworks, considers their different approaches to language proficiency as language use, and gauges their impact on L2 learning, teaching, and assessment. Chapter 24 by Carol A. Chapelle reports on technology-mediated L2 learning and how technology has affected L2 pedagogical interventions. The author synthesizes work done on new pedagogies that motivate the creation of learning activities which operationalize theoretically-based learning principles. Within this synthesis is an overview of the technological affordances underlying new pedagogies for grammar, vocabulary, reading, writing, listening, and speaking, as well as pedagogical approaches that have been created or enriched through the use of technology (e.g., blended learning, technology-assisted taskbased language learning, and distance learning). In addition to reviewing research and evaluation methods used to investigate students’ use of technology for language learning, Chapelle highlights how research and development in this area has contributed to the study of L2 learning. In Chapter 25 by Hossein Nassaji and Eva Kartchava, the authors report on content-based language teaching (CBLT), an instructional method that combines the teaching of academic subjects (e.g., math, science, history) and L2 learning. The chapter outlines the main characteristics of CBLT in general and discusses its origins, various types, and contexts of use. The benefits and shortcomings of CBLT are assessed, as are pedagogical strategies that can address those shortcomings. The chapter concludes with implications for theory and practice and what the future landscape of CBLT research is likely to look like. The last topic of the section is Graham Crookes’ Chapter 26, in which the author discusses critical L2 pedagogy, specifically teaching approaches that draw on social justice. Crookes notes a considerable increase in research conducted in this area in recent

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

8

J o h n W. S c h wi e t e r a n d A l e ssa n d r o B e n at i

decades, with practical work first appearing in the 1980s and more theoretical work and increased examples of practice since the turn of the century. The chapter demonstrates that critical language pedagogy has taken for granted theories of L2 learning which are critical in nature, though little attention has been paid to characterizing them in detail. The author reports on what is implied by critical theories of L2 learning and offers possible alternatives. Within this discussion is the concept of praxis, the integration of theory and practice, which is considered to be of central importance throughout the wider area of critical pedagogy. Part VI considers the role of context, environment, and other factors in L2 learning. Bilingual education and policy is the first topic presented. In Chapter 27, Christine Hélot and Ofelia García argue that because learners, as well as ideologies about bilingualism vary, bilingual education implies many things in different contexts. As such, the chapter differentiates various types of learners and bilingualism, and describes how the main factor of differentiation has to do with power itself. These power differentials have a significant impact on which bilingual education programmes are implemented in different societies, as well as the pedagogical practices used and the resources that are available. This complicated landscape for bilingual education is discussed in the context of a globalizing world, and considerations are made for L2 learners, especially those who are minoritized. In Chapter 28, Kim Potowski and Sarah J. Shin report on heritage language instruction. The authors organize this discussion around four important questions. What is a useful working definition of heritage language learners? What are the major heritage languages around the world, and why are they important? What makes heritage language instruction different from teaching an L2? And what resources are available for educators working with heritage learners? The chapter concludes by advocating for a fundamental shift in societal attitudes towards heritage languages and speakers of those languages in order to accomplish a language competent society. Aline Ferreira, Viola G. Miglio, and John W. Schwieter in Chapter 29 discuss minority languages and learners in both home and abroad contexts. The authors specifically focus on the role that education and acculturation play in relation to language learning and maintenance. They also compare results in historical minorities in their own countries and heritage language speakers abroad. Examples include Spanish and Chinese as heritage languages in the United States and Canada, respectively, and minority languages in Europe. The chapter demonstrates that although speakers of a minority language constantly face issues like L1 attrition, some alternatives have helped immigrants to maintain their language abroad. In the case of minority speakers within their own country, new policies are decisive in maintaining languages alive in places where they

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

Introduction

9

have a secondary, inferior, legal status. In Chapter 30 by Jane Jackson and John W. Schwieter, the authors report on L2 learning in study abroad and immersion contexts. The chapter specifically focuses on L2 intercultural learning and the internal and external factors that can result in differing sojourn outcomes. As well as summarizing key findings, the chapter raises awareness of the theories and methodologies that characterize research in L2 study abroad settings and also calls on researchers to take a closer look at what actually happens during stays abroad so that study abroad educators and administrators can advocate for research-driven, t­ heory based pedagogical interventions that deepen language and intercultural learning at all stages of the study abroad cycle: pre-sojourn, sojourn, and post-sojourn. In Chapter 31, Peter Swanson discusses the past, present, and future of teacher education in the United States and juxtaposes this development with the highly successful system in Finland. His review, spanning over two centuries, advocates that teacher education holds many benefits and that care must be taken to advance the profession, continuously seek to improve learner achievement, and develop globally-minded citizens. Swanson’s depiction of L2 teaching and learning is a testament that it contributes to these goals. Part VII concludes the handbook with thought-provoking future trajectories in L2 learning and teaching. Chapter 32 by Susan M. Gass begins by first discussing early foundations of L2 learning by focusing on (1) the relationship between L2 teaching and learning; (2) the general growth of the field of L2 learning; and (3) changes in emphases suggesting a narrowing of the gap. Through her metaphoric crystal ball based on current research trends, Gass contemplates the future from three perspectives: the content of the field (based on past emphases); methodologies and statistical techniques; and research instruments. The author ties together the growth of L2 teaching and L2 learning with observations of ways in which the two continue to feed one another. The chapter concludes with comments about the overall development of the field of L2 learning, basing these observations on study quality, methodological rigour, and statistical sophistication. Throughout The Cambridge Handbook of Language Learning, findings from research have provided a shift in the way we understand L2 learning and the roles played by both teachers and students in the classroom. Current findings from research provide a much clearer picture on how learning another language happens. Understanding more about L2 learning pushes language teachers to question the prevailing methods and approaches in language teaching. Such is the aim of this handbook: to take readers on this journey in which they will learn about the main implications of research and theories for language teaching and the vast territory that is yet to be chartered.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:27:01, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.001

1 Formal Linguistic Approaches to Adult L2 Acquisition and Processing Jason Rothman, Fatih Bayram, Ian Cunnings, and Jorge González Alonso 1.1 Introduction The relative conformity with which (typically developing) children attain adult grammatical competence—ultimate attainment—and the similarity in developmental paths along which they progress is remarkable (e.g., Ambridge & Lieven, 2011; Clark, 2003; Guasti, 2002; Synder, 2007). This achievement is, however, so ubiquitous and mundane that we seldom marvel at it. Of course, monolingual adult grammars may also differ from one another, especially for some domains of grammar (e.g., Dąbrowska, 1997, 2012), but such variability pales in comparison to the variation in adult non-native second language (L2) grammars. Indeed, the path and outcomes of L2 acquisition can be highly variable from one individual to another, even under seemingly comparable contexts. Individual and group-level factors in adulthood that either do not apply or apply with much less consequence in young childhood conspire to explain at least some of the gamut of L2 variability. Adults are less likely to be exposed to high or sufficient quantities and qualities of (native) input, which provide the indispensable raw material for specific grammar building (e.g., Carroll, 2001; Rankin & Unsworth, 2016; Rothman & Guijarro Fuentes, 2010). It is possible that adults do not maintain the same linguistic and/or cognitive abilities to acquire (e.g., Bley-Vroman, 1989, 2009; DeKeyser, 2000; Hawkins & Casillas, 2008; Long, 2005; Tsimpli & Dimitrakopoulou, 2007) and/or process language (Clahsen & Felser, 2006a, 2006b, 2018; Grüter & Rohde, 2013; Grüter, Rohde, & Schafer, 2016) as children have. Whether or not the underlying mechanisms for language acquisition and processing are one and the same across the lifespan, one should expect factors such as motivation, influence from the first language (L1), cognitive differences (e.g., working memory),

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

14

J a s o n R ot h m a n , e t a l .

context of learning (classroom versus naturalistic immersion), and linguistic aptitude, among other variables, to contribute to the range of individual differences in adult L2 path and ultimate attainment (e.g., Dörnyei, 1998, 2003, 2014; Linck et al., 2014; Rankin, 2017; Robinson, 2001, 2005, 2013; Schwartz & Sprouse, 1996; White, 2003). We are all increasingly mindful of the inherent comparative fallacy implicated in most—but not all—monolingual to bilingual comparisons (e.g., Bley-Vroman, 1983; Ortega, 2013, 2016; Rothman & Iverson, 2010). It would, however, be inaccurate to pretend, even with contemporary calls to examine bilingualism in its own right, that the study of adult L2 acquisition is not motivated in large part by the goal of understanding what underlies typical L1 → L2 differences. No one denies that we need to understand the constitution of adult L2 grammars as the independent entities they are and in more ecologically valid terms. However, doing so completely devoid of the traditional comparison goes beyond simply abnegating the past research agendas that paved the way to modern second language acquisition (SLA) studies. Most calls for the bilingual turn in SLA challenging the monolingual as a golden standard benchmark also recognize the intense curiosity one has to understand L1 versus L2 differences, and the value of understanding this further. Such calls remind us, above all, that we are comparing various types of apples to each other, if not apples to oranges. Sometimes it is quite useful, if not necessary, to compare apples to oranges, as long as we are aware we are doing so. Thus, L1 to L2 comparisons are not dead, nor should they be. Alternatively, we need to be cautious of what we claim from L1–L2 difference comparisons and ask ourselves, juxtaposed against specific questions, if such a comparison is necessary and/or particularly insightful. We will be mindful of this discussion throughout this chapter; however, we will also be faithful to what has been done in the field within its proper (temporal) context, highlighting where some questions have been modified over time as advances in theory and otherwise shifting views have necessitated. Forty plus years into the modern study of adult SLA, various paradigms of linguistic inquiry (e.g., generative grammar, construction grammar, lexical functional grammar, cognitive grammar, etc.) have been adopted as a backdrop against which subfields of SLA have emerged to frame and research adult non-native language acquisition and processing. The goal of the present chapter is to focus on one such approach, that is, generative approaches to SLA (GenSLA). GenSLA research has always been focused on describing and explaining the implicit knowledge of L2 representations in the mind or brain of the learner (see Rothman & Slabakova, 2018; White, 2018, for recent state-of-the-science reviews). The role that Universal Grammar (UG) may continue to play—in part or in whole—in restricting the hypothesis space for L2 grammatical learning—distinguishing

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

15

between knowledge that is clearly acquired on the basis of exposure to L2 input and that which is more reflective of universal linguistic principles— has featured prominently in GenSLA for the entirety of the field’s tradition, starting in the early 1980s. By the late 1990s, as we will see, GenSLA had expanded considerably beyond the focus on whether or not there was UG-accessibility after puberty. Indeed, GenSLA applies the tenets of formal generative linguistic theory (e.g., Chomsky, 1965, 1981, 1995) to guide the questions asked, make predictions about L2 linguistic behaviour along the developmental continuum, inspire and define the selection of domains of grammar on which to empirically focus, and interpret the data and from them hypothesize about the constitution of non-native mental linguistic competence (i.e., I-language). In the remainder of this chapter, we contextualize what is meant by formal linguistics (past and present) as applied to non-native language studies, address what makes GenSLA different from other theories of SLA, and briefly review the history of GenSLA over the past three plus decades, with special consideration of how it has expanded since the early 2000s to include work on non-native processing using psycholinguistic and neurolinguistic methodologies.

1.2  The Main Tenets of Generative Linguistic Theory Universal Grammar (UG) is claimed to provide a biological blueprint of sorts for efficient and constrained language learning, at least in children. Indeed, over the past two decades in particular there has been much discussion among generative theorists as to what precisely the innate language faculty offers. To be sure, discussion, debate, and change at the level of formal theory have inevitable consequences for the evolution of and application in GenSLA theory; however, this is of less consequence than one might think. GenSLA—like all approaches to SLA—is an empirical field that generates tried and true data of its own. Data collected with sound methodology exist (somewhat) orthogonally to any specific theory used to try to explain them. As formal theories get modified on the basis of advancements that run in parallel to, or as a result of, applying generative theory to (non-native) language acquisition, GenSLA’s database stands the test of time, able to adjust to any then-current theory and/or serve as evidence to support it or reject it a priori (see also Schwartz & Sprouse, 2000). This said, the formation of theories, hypotheses, predictions, constructs, interpretations, and more within GenSLA finds its roots within particular instantiations of then-current generative linguistic theory. And so, in a handbook chapter such as this, it is useful to cursorily outline its main tenets as well as a few specific versions of generative theory along the past decades in which GenSLA has existed in parallel.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

16

J a s o n R ot h m a n , e t a l .

Perhaps the most well-known theory that GenSLA has drawn from is the Principles and Parameters (P&P) approach from the days of Government and Binding Theory (Chomsky, 1981). UG was argued to provide a blueprint of constraints, known as principles and parameters. Principles are constraints on the form that any potential human language can take for a given domain. Parameters are, essentially, principles with choices: a universal property that reflects limited variation in the world’s languages. For example, in all languages, a sentence must minimally have a subject and a verbal predicate (with optional object arguments depending on the verbal argument structure): SV(O). All verbs in context have an intended subject, whether expressed phonetically or not. This principle—a truism of grammar—was captured under the Extended Projection Principle (EPP; Chomsky, 1982). To the extent that the EPP is a proper universal principle, all utterances with a verb in all languages must always respect it. For example, sentences like *Robert thinks that __ like dogs more than cats is ungrammatical because it violates the EPP. The verb “like” in English is left without a specified subject that cannot be recovered alternatively, and so the universal rule that all verbs have a reliably interpretable subject is violated. As it turns out, not all subjects are overtly expressed, even in English, and so absence of an overt subject does not necessarily mean the EPP is violated. For example, in control structures like I want to go to the movies: [Ii want [CPPROi to go to the movies]], the EPP is satisfied by accessing the features of the higher subject. In other words, “I” serves as the subject of the matrix verb as well as the antecedent for co-­reference of the empty category subject (PRO) of the embedded verb. Equally with coordinated clause structures, as in I live in the UK and ___ teach at the University of Reading, the syntactic structure is such that the EPP is fulfilled by the presence of the first expressed pronoun “I”, accessible via coordination. With the few exceptions above, it is fair to say that English-type languages require overt pronouns to always be pronounced. This is clearly not true of all languages in the world. For languages like Spanish, Turkish, Arabic, and Korean, most subject pronoun positions in relation to verbs are phonetically unrealized. How could such languages exist if the EPP is truly a principle of grammar? The EPP merely states that all verbs have a readily accessible subject for interpretation; it does not stipulate how languages can vary in how this requirement is fulfilled. A related principle of grammar that provides a limited amount of choices for specific grammars to select is known as the Null-Subject Parameter (NSP). Whereas English is negatively valued for the NSP, thus requiring overt subjects ubiquitously, Spanish-type languages are positively valued and thus make use of other linguistic means to fulfill the EPP requirement. Consider (1) and (2) in Spanish and (3) and (4) in Turkish below: Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

17

(1) Yo creo que aquí nosotros vivimos bien. I believe-1PS that here we pro live-1PP well. “I believe that we live well here.” (2) ___ Creo que aquí ___ vivimos bien. pro believe-1PS that here pro live-1PP well. “I believe that we live well here.” (3) Ben inanıyorum ki biz burada iyi yaşıyoruz. I believe-1PS that we here well live-1PL “I believe that we live well here.” (4) ___ İnanıyorum ki ___ burada iyi yaşıyoruz. pro believe-1PS that pro here well live-1PL “I believe that we live well here.” In (1) and (3), the EPP is fulfilled ostensibly in the same way as in English— or in principle can be—via the use of overt pronouns in the canonical subject positions. In (2) and (4), the subjects are phonetically null (in main and embedded clauses alike), but the EPP is satisfied by the verbal morphology (in bold), which uniquely specifies the person and number of the only understandable subject, an impossibility in English given its weak verbal morphology. And so, while it can be said that all languages respect the EPP as a universal principle, parametric variation is captured under the NSP relating to how the EPP can be satisfied in various languages. With the advent of Minimalism (Chomsky 1995, 2001), the proliferation of parameters and thus their very ecological validity was questioned. Over the years, the list of proposed principles and parameters (P&P) had grown into the hundreds, reasonably making one wonder how the mind could encode so much information at a zero state and indeed why it would do so in the first place. Minimalism, as the name would suggest, called for a more streamlined, elegant approach to describing the same facts of language that P&P uncovered. It was argued that properties previously formalized under so-called principles and parameters were likely not specific mental constraints themselves, but rather consequences of core grammatical operations where functional features encoded on lexical items (including morphology) fulfilled their remit. And so, the EPP above was reduced to an AGREE(ment) operation, for example, between features associated with the subject position (Spec IP/TP) and the feature carried in languages like Spanish and Turkish on person and number morphology. The details are quite abstract, and their specificity is not really relevant to our purposes here. What is important, however, is the acknowledgement within Minimalism that a more reductionist approach was needed, but crucially one that could provide the same level of descriptive adequacy that decades of generative work had achieved. Indeed, as Ellis (1998) stated, “it is the assumptions of UG that are under attack [by connectionist and other competing cognitive approaches], not the generative grammar descriptions of the relations between the linguistic units” (p. 633). We respectfully Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

18

J a s o n R ot h m a n , e t a l .

disagree with his underlying reasoning and, like Bruhn de Garavito (2011), we would point out that the “generative grammar descriptions of the relations between linguistic units” that Ellis concedes are not problematic are in fact, often although not always, incompatible with theories that rely (virtually) solely on input and domain-general cognition. Nonetheless, it is prudent to point out that descriptions within generative theory at the level to which Ellis (1998, p. 633) refers are regarded even by competing theories as at least descriptively accurate, and as a result must be accounted for by all iterations of theory, generative and otherwise. The minimalist framework sets out to explain three basic properties claimed to be specific to human language syntax: (i) human language syntax is non-linear (except in production and parsing, including morphology and phonology/phonetics) as compared to non-human primates; (ii) hierarchical structure affects meaning and interpretation; and (iii) hierarchical structure is seemingly infinite (i.e., recursive; Berwick & Chomsky, 2017; Fitch, Hauser, & Chomsky, 2005; Fitch & Hauser, 2004; Hauser, Chomsky, & Fitch, 2002). It was proposed that these three properties together comprised the fewest explanatory, but necessary, variables of language and, therefore, the best candidate for a single operation for the building of syntactic structures, which later became formalized as MERGE (Chomsky, 2001). Merge, simplistically speaking, is responsible for forming and hierarchically ordering syntactic objects (i.e., sets). In this respect, UG then accounts for the Basic Property of Language (BPL)—very simply put—because it is proposed to be an internal computational system responsible for the hierarchical building of human language syntactic sets (i.e., Merge) that interfaces with both external components such as production (phonology/phonetics), parsing, and morphology (word building), as well as conceptual systems (i.e., cognition) for appropriate inference making and interpretation. Whatever UG’s actual composition turns out to be, its main functions will always be a gap-filler and an accelerator, providing an answer to how language can be acquired so effortlessly and rapidly despite the complexity of the task and in light of seemingly incomplete input. UG must also be the proverbial agent that reduces the hypothesis space the mind considers from all logical possibilities to those that are compositionally and computationally parsimonious with natural human language—why children make only a subset of all reasonably possible errors. Whether this turns out to be (a) a truly minimal set of core mental operations specific to language such as MERGE, (b) an inventory of formal (functional) features of grammar, or (c) more akin to an actual blueprint of principles of universal linguistic well-formedness with room for language specific (parametric) variation, the general abstraction of UG remains the same: UG will be the smallest amount of innate structure needed to explain outcome competence that ultimately cannot be reduced to deduction from the input and domain-general cognition alone.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

19

1.3 What Makes GenSLA Different (and Not So Different) from Other SLA Approaches? All SLA theories that focus on language as a cognitive phenomenon—as opposed to those in which learner-external factors take centre stage—are broadly concerned with describing the process by which child and adult non-native learners come to produce and understand utterances in the L2, to variable degrees of communicative efficiency and native-­likeness. In an argument that very much parallels debate in first language acquisition (FLA), the fundamental difference between GenSLA and other approaches is the extent to which the different theories consider some L2 knowledge to be acquirable solely from an interaction between the available input and domain-general cognitive systems. To be sure, any GenSLA scholar would agree that meaningful input and domain-general mechanisms are fundamental ingredients of L2 acquisition. The increasingly central position that input quantity and quality is assuming in generative SLA studies in recent times (see, e.g., Rankin & Unsworth, 2016; Slabakova, 2013; Yang & Montrul, 2017) represents perhaps the largest area of crossover between GenSLA and so-called data-driven approaches (see Rothman & Slabakova, 2018, for a detailed discussion of what generative and usage-based paradigms share beyond their differences in focus). Notwithstanding the above, a significant part of the knowledge that a typical L2 learner acquires poses a problem for theories that do not contemplate the involvement of domain-specific machinery. In the 1980s, a few scholars (e.g., Bley-Vroman, 1989; White, 1985) began considering the question of whether one could indeed speak of a logical problem of second language acquisition, very much in the same way that had been proposed for FLA. In the case of children, the logical problem relates to the baffling complexity and uniformity of language acquisition outcomes, in light of the quantity and quality of input to which they are typically exposed. In other words, how do typical human children (and not other species) come to acquire language in very similar ways, to very much the same degree (notwithstanding third-degree factors such as dialectal differences), without instruction, irrespective of differences in cognitive capacities (e.g., general intelligence, working memory)? How is it that children are able to produce utterances—and, most importantly, make errors—that do not (cannot) form part of the input they receive? The generative answer to these questions is that only by assuming some type of innate, domain-­ specific knowledge like UG can we begin to understand the discrepancies between the (never optimal) input children receive and the complex, rich adult grammars they target and successfully acquire. To be sure, much of the path of child acquisition can be accounted for on the basis

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

20

J a s o n R ot h m a n , e t a l .

of an interaction with environmental input and constraints traceable to domain-general cognition. However, the fact that not all of the complexity of linguistic computational grammars can be reduced to this interaction in light of the overwhelming success of virtually all children still leaves “little hope that much of the structure of the language can be learned by an organism initially uninformed as to its general character” (Chomsky, 1965, p. 58). While uniformity in ultimate attainment across learners is much less robust in L2 acquisition, it is still the case that the input available to the learner often underrepresents, or underdetermines, the kind of subtle knowledge that L2ers nevertheless come to possess. And so, there is also a logical problem of explanation for adult L2 acquisition, even if the typical path of development and ultimately attained grammars do not display the same conformities overall as FLA (Schwartz, 1998). The input underrepresentation is compounded by the fact that most input consists of positive evidence, that is, naturalistic input (both in L1 and L2 acquisition) is made up of grammatical sentences, which means that the only direct evidence learners get is about what can be done in a particular grammar, but they are rarely—and definitely never reliably—provided with evidence of what cannot be done (i.e., negative evidence). This does not mean that L2 learners, especially in a classroom setting, are not corrected in various ways, both directly and indirectly (e.g., recasting). The type of correction given, however, is not always likely to induce grammatical changes at the level of representation, especially because it is typically inconsistent and obtains on the basis of needing to correct truth-value or due to a lack of intelligibility. To understand this, one need look no further than the most ubiquitous errors of L2 English, and the fact that they are fairly consistently corrected, to begin to appreciate the limitations of overt corrections in adult L2 acquisition. Obligatory morphology such as third person –s or past tense –ed are often omitted or overused by L2 learners—interestingly irrespective of their L1, although there is a clear L1 effect in terms of how this distributes. These properties are taught explicitly in a repeated fashion and are often corrected, yet they persist to a high degree. Details aside on how and why this happens, the point to be made is that mental L2 grammars have their own time course of development—as in L1 acquisition—despite what explicit correction might hope to achieve. The argument that the linguistic input often underrepresents the actual complexity of the target grammar, also known as the poverty of the stimulus (PoS) problem (see Berwick, Chomsky, & Piattelli-Palmarini, 2012; Berwick et al., 2011; Kam & Fodor, 2012; Crain, 2012; Lasnik & Lidz, 2017; Schwartz & Sprouse, 2013, for updated reviews of PoS in general and its applicability to L2 acquisition), has been one of the guiding questions in SLA research from a generative perspective, although, as we will see, whether or not UG

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

21

remains accessible in adulthood to fill the gap between the input and the output of L2 acquisition has been, and continues to be, the source of much debate. Reasonably, much work in usage-based approaches to language has been dedicated to falsifying the basic claim of the PoS argument—i.e., that the input does indeed lack sufficient evidence of certain subtle contrasts that learners do eventually acquire. Most recent proposals argue that primary linguistic data might indeed contain indirect negative evidence of what have traditionally been defined as PoS properties, even if no utterance in the input explicitly exemplifies those properties (e.g., Lewis & Elman, 2001; Reali & Christiansen, 2003, 2005). However, simulations using sophisticated statistical learning have not yet succeeded in replicating the acquisition of this kind of fine-grained contrasts from arguably comparable input (see, e.g., Kam, Stoyneshka, Tornyova, Fodor, & Sakas 2008, in response to Reali & Christiansen, 2005). One of the issues that non-generative approaches have to address is that PoS is not merely articulated around the extremely low frequency of binary outcomes—i.e., the lack of direct evidence of whether something is allowed in the grammar or not—but also around very subtle conditions for optionality. In other words, some things are not just allowed or disallowed in the grammar, they are simply interpreted differently depending on tacit distinctions that vary across languages. Let us illustrate this with a well-known example of this type of property which relates in ways to the EPP and the NSP discussed above—in whatever form a then-current theory uses to explain their general observations—precisely because it embodies a highly specific constraint on interpretation of overt pronouns in null-subject languages: the Overt Pronoun Constraint (OPC; Montalbetti, 1984). Recall that in section 2, we ended our discussion of the NSP by stating that null-subject languages like Spanish or Turkish also respect the EPP, because even when subjects are phonetically null, they are recoverable from some combination of the verbal morphology and available discourse or context information. One might then be tempted to assume that null and overt subjects are equally acceptable in all contexts in, for example, Spanish, or that subjects are interpreted more or less in the same way whether or not they are phonetically realized. Neither of these assumptions would be correct. There are, indeed, well established patterns as to the general distribution of when overt and null subjects tend to be used pragmatically (e.g., overt pronouns in contrastive focus contexts), or when the grammar forces a null subject, such as in expletive contexts (e.g., subjects of existentials where there is no semantic meaning of the subject: ___ hay tres hombres en la sala. “(There) are three men in the room.” But, what about interpretation? Subject pronouns in English embedded clauses can be interpreted as referring back to a noun phrase (NP) in the main clause—whether this is a referential NP, a quantified expression

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

22

J a s o n R ot h m a n , e t a l .

like someone or everybody, or a wh- word or phrase—or, alternatively, to an extra-sentential antecedent that can be recovered from discourse without any grammatical restrictions, even if reference back to the matrix subject is preferred. Consider the examples in 6a–c and 7a–c (drawn from White, 2003): (6)

a. [Maryi thinks [that shei will win.]] b. [Everyonei thinks [that shei will win.]] c. [Whoi thinks [that shei will win?]]

(7)

a. Janeti is a great athlete. [Maryj thinks [that shei will win.]] b. Janeti is a great athlete. [Everyonej thinks [that shei will win.]] c. Janeti is a great athlete. [Whoj thinks [that shei will win?]]

In 6a–c, the pronoun she is coreferential with the subject of the main clause, which in 6b is a quantified expression and in 6c is a wh- phrase. In 7a–c, on the other hand, she refers to Janet, an antecedent which can be found earlier in the discourse but does not belong to the same sentence as the pronoun. Because each of the sentences in 6 is identical in form to the second sentence in the corresponding example from 7, the hearer/reader needs to rely on contextual information to assign the correct referent. Things, however, are not so straightforward in a null-subject language like Spanish. While both overt and null pronouns can have discourse antecedents (the type of referents we saw in 7), they behave differently when it comes to referring back to the main clause (matrix) subject. In short, null pronouns can have both referential and quantified NPs as their antecedents, whereas overt pronouns cannot refer back to a quantified expression (in theoretical jargon, they cannot have bound variable interpretations). Consider now the examples in 8 (for null pronouns) and 9 (for overt pronouns): (8)

a. [Martai considera [que __ i debería entrar la primera.]] Marta considers that (null) should go in the first. “Marta considers that [Marta] should go in first.” b. [Algunai considera [que __ i debería entrar la primera.]] Some(one) considers that (null) should go in the first. “Some of them consider that [they] should go in first.”

(9)

a. [Martai considera [que ellai debería entrar la primera.]] Marta considers that she should go in the first. “Marta considers that [Marta] should go in first.” b. *[Algunai considera [que ellai debería entrar la primera.]] Some(one) considers that she should go in the first. “Some of them consider that [they] should go in first.”



As we can see, null pronouns in 8a–b admit both a referential (8a) and a quantified (8b) NP as their antecedent, whereas this correferential

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

23

interpretation with a quantified expression is impossible for the overt pronoun in 9b. In fact, a sentence like 9b forces a discourse antecedent interpretation where the pronominal subject ella is interpreted to refer back to some antecedent in discourse, but crucially not to the quantifier alguna: (10) [Algunai considera [que ellaj debería entrar la primera.]] Some(one) considers that she should go in the first. “Some of them consider that [she, i.e., someone else] should go in first.”

In short, there is a highly specific restriction at play. Although Spanish has processing induced preferences in how this distributes, all null subjects can freely co-refer to all matrix and disjoint referent antecedents, especially in consideration of the discourse context. This, however, is not true of overt subjects. Embedded overt subjects cannot co-refer to the matrix clause only when the antecedent is a variable expression (quantified NP/wh- phrase). Of course, this is never taught. In fact, it is a fact about null subject languages that flies under the radar unless one is a linguist. What makes this property such a subtle contrast is the fact that there is no apparent reason why the string in 9b should be ungrammatical. However, the string is exactly the same as in 10, and yet no direct evidence in the input will suggest to the learner that some readings (coreferential with the main clause subject) are not possible in some sub-cases (quantified expression in the subject of the main clause) of a subtype (overt pronoun in the subject of the embedded clause) of embedded clauses in Spanish (and other null subject languages). And in fact, discourse context between two interlocutors would not necessarily help either, because taking a disjoint versus a matrix antecedent interpretation typically would not cause an obvious breakdown in communication, even if the two speakers unbeknownst to themselves understand distinctly the same utterance. Although many have examined this property since, Pérez-Leroux and Glass (1997, 1999) and Kanno (1997, 1998) were the first to focus on the acquisition of the OPC in L2 Spanish and L2 Japanese, respectively, by native speakers of English. These are particularly relevant groups because, as we have just seen, the OPC does not apply in English, and it is never taught in Spanish or Japanese lessons (see Kanno, 1997, and Pérez-Leroux & Glass, 1997, for discussion). In sum, the OPC does not receive support from direct evidence in the L2 input, is not explicitly taught in these learners’ lessons, and cannot be transferred from their L1, which makes it a perfect example of a PoS property and thus a good candidate to test whether UG guides, or constrains, the acquisition of non-native languages. Kanno (1997) tested the interpretation of sentences of the type 6–10 above in the L2 Japanese of twenty-eight native speakers of English, as compared to the judgments of twenty L1 Japanese speakers. She found

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

24

J a s o n R ot h m a n , e t a l .

that, like the natives, L2 learners majoritarily rejected a quantified NP as the antecedent of an overt pronoun in the embedded clause (87% of the time, as compared to 98% for the native speakers). Moreover, native and non-native speakers even displayed similar preferences when both interpretations were optionally available for the overt pronoun, that is, when the main clause subject was a referential NP. In those cases, both groups chose a coreferential interpretation only half of the time—the other half, they selected a discourse antecedent. Kanno’s (1997) results are particularly relevant because they not only point to the acquisition of a subtle, PoS property, but even to similar preferences in anaphora resolution where no constraints on distribution or interpretation seem to be in place in the native grammar. Similarly, Pérez-Leroux and Glass (1997; see also discussion in PérezLeroux & Glass, 1999) tested the same constraint in the L2 Spanish of three groups of non-native speakers (total n = 98) divided by proficiency, as compared to a control group of native speakers of different varieties of Latin American and European Spanish. The experiment, a translation task, elicited production data based on a fixed interpretation that was strongly suggested by the context provided before each sentence. Some contexts promoted a bound-variable interpretation (i.e., they suggested that the null or overt pronoun should be coreferential with the main clause subject, whether this was a referential or a quantified NP), whereas others fixated an interpretation in which the null or overt pronoun in the embedded clause referred back to a discourse antecedent. An example is provided in 11 (Pérez-Leroux & Glass, 1999): (11) a. Referential Story In the O. J. Simpson trial, it is clear that the press has a negative bias against the defendant in their reporting. Some journalist said that he was a wifebeater. To translate: “But no journalist said that he is guilty.” b. Bound-Variable Story The court charged that some journalists had been in contact with the jurors. Several of them were questioned by the judge. To translate: “No journalist admitted that he had talked to the jurors.”

The logic of the experiment is simple. If participants obey the OPC, they should never translate the pronominal subject of the embedded clause— bolded in our reproduction of the examples—as an overt pronoun in 11b, where the pronoun is forced to refer back to a quantified NP. However, they should be able to use both null and overt pronouns in 11a, where the referential NP is a possible antecedent for either. Pérez-Leroux and Glass found that all groups made a distinction between those contexts in which only null pronouns should be used (bound-variable stories) and

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

25

those in which either an overt or a null pronoun were acceptable (referential stories). While some of the learner groups displayed some violations of the OPC, this did not differ significantly from the native controls, who did in fact show some violations themselves—the authors attribute this to the effect of Caribbean Spanish, one of the varieties present in the sample, which overuses overt pronouns with respect to other varieties. Importantly, Pérez-Leroux and Glass’s (1997) results demonstrate that the OPC is operative from very early on in L2 development. Taken together, the results of these two studies strongly suggest that some type of domain-specific knowledge—not accessible via L1 transfer—actively constrains the acquisition process. The case is indeed compelling: from very early stages, L2 learners unfamiliar with any language where the OPC operates, and with no indication of its existence from their classroom instruction, manage to acquire (i) a restriction on the interpretation of a subset within a subset of grammatical sentences, for which there is no direct (negative) evidence in the input stream, and (ii) a similar distribution of anaphora resolution preferences in those contexts where both overt and null pronouns can be interpreted in similar ways. It is difficult to account for this type of evidence on the basis of L2 input and domain-general cognition alone, that is, without positing the existence of some innate knowledge that is able to fill in the gap between the notable scarcity of evidence, linguistic or otherwise, in the input and these learners’ successful acquisition of the OPC. Empirical evidence of this and other PoS properties being acquired by learners who are exposed to an L2 after puberty provides support for two different but interrelated positions. The first, relevant to cross-paradigm debates, is the one we have discussed as distinctive of GenSLA approaches: that domain-specific knowledge of a UG type must be invoked if we are to explain the relative ease with which L2 learners overcome this and other similar learnability problems. The second position, which applies internally to GenSLA, is that access to UG remains intact throughout the lifespan and across multiple iterations of language acquisition—i.e., L2/L3/L4/Ln. From very early on, differences in developmental sequence and ultimate attainment between L1 and L2 acquisition were ascribed by numerous authors to a complete or partial inaccessibility of the same underlying mechanisms that guide child L1 acquisition (e.g., Abrahamsson & Hyltenstam, 2008, 2009; Bylund, Abrahamsson, & Hyltenstam, 2012, 2013; Bley-Vroman, 1989, 2009; Clahsen & Muysken, 1989; Coppetiers, 1987; DeKeyser, 2000; Granena & Long, 2013; Hawkins & Casillas, 2008; Hawkins & Chan, 1997; Johnson, 1992; Johnson & Newport, 1989; Long, 2005; Meisel, 2011; Schachter, 1988; Tsimpli & Dimitrakopoulou, 2007). These arguments dovetail with claims that a critical period (Lenneberg, 1967) exists for language acquisition, direct evidence of which had become

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

26

J a s o n R ot h m a n , e t a l .

available in the late 1970s through the case of Genie, a feral child who had been abused and deprived of the most basic linguistic interaction (e.g., Curtiss, 1977). Essentially, the Critical Period Hypothesis (CPH) as applied to L2 acquisition argued that, due to a decrease in brain plasticity after a certain age (loosely identified with puberty), adults are unable to acquire new languages in qualitatively the same way as children do (e.g., DeKeyser, 2000; Franceschina, 2005; Long, 2005). Two main bodies of evidence problematize the above type of claims. The first is that extremely successful—i.e., native-like—adult L2 learners are indeed uncommon, but also not truly exceptional. However, a strong hypothesis like the CPH should not tolerate such exceptions, because its core argument is universally applicable: all brains mature, and therefore any maturational constraint one wishes to posit will inevitably encompass all adult learners. Under such a logic, that a majority of L2 learners do not achieve native-like proficiency must mean they cannot; any evidence to the contrary—i.e., any native-like L2 learner or the acquisition of specific properties that should not be acquirable by a much larger cohort of L2 learners (not just the super talented and highly advanced ones)—escapes the scope of such a position’s predictions and poses a serious challenge to it (see Rothman, 2008; Schwartz, 1986, for similar arguments). There of course exists the possibility that different neurocognitive mechanisms underlie seemingly identical behaviour. In other words, it might be that successful L2 learners manage to perform in very much the same way as natives making use of fundamentally different tools (e.g., Clahsen & Felser, 2006a, 2006b, 2018; Paradis, 2009; Prévost & White, 2000; Ullman, 2005, 2016). The second body of evidence is in fact relevant to this claim. First, research in cognitive neuropsychology has shown that the existence of a post-pubescent cutoff in neuroplasticity or even very sharp declines is simply not so (Fuchs & Flügge, 2014, for review). Second, asserting that different cognitive or processing mechanisms underlie the linguistic performance of native and non-native speakers generates very specific predictions that are amenable to testing directly through methodologies that are increasingly familiar to the field—electro- and magnetoencephalography (EEG, MEG), functional magnetic resonance imaging (fMRI), etc. Together with the improved sensitivity of implicit, real-time behavioural measures such as eye-tracking, online processing methodologies allow us to evaluate the claim that native and non-native language use are underlain by neuro-functionally different processes, despite occasional appearance of convergence. Similarly to the previous claim, such a theory is falsified if L2 learners are shown to (i) match native speakers in performance, while (ii) displaying comparable indices of online processing. A growing body of evidence suggests that this might indeed be the case (see, e.g., Roberts et al.,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

27

2018, for discussion). Indeed, the field of non-native processing research, to which we now turn, is a good friend to GenSLA and vice versa.

1.4  GenSLA and Language Processing Beginning in the late 1990s and early 2000s, L2 researchers began in earnest to utilize experimental methodologies from the fields of psycholinguistics and neurolinguistics to investigate not only L2 knowledge and representation but also how such knowledge is utilized during real-time language processing. This involved expanding the set of methodologies used in GenSLA research from “offline” measures, such as acceptability judgments, that index the final interpretation or judgment given to a sentence, to “online” measures, such as reaction times, that allow the researcher to investigate the moment-to-moment processes involved in language processing as a sentence unfolds in real-time (see Chapter 9 in the current volume by Roberts, for further discussion of psycholinguistic and neurolinguistic methods). Early research in this regard included studies on the processing of so-called wh- dependencies (e.g., Juffs & Harrington 1995, 1996). Consider, for example, (12a) and (12b). In (12a) the verb “read” and its direct object “the book” are adjacent to each other. In the wh- question (12b), “which book” is also interpreted as the direct object of “read”, but does not appear adjacent to it. Correct interpretation in this case requires a linguistic dependency to be formed between the verb and the displaced wh- constituent. Although wh- questions can span several clauses, as in (12c), subjacency or “island” constraints restrict wh- dependency formation (Ross, 1967). For example, a dependency cannot be formed between a wh- constituent and a verb inside a relative clause, as in (12d). Whether L2 learners can acquire knowledge of restrictions on grammatical and ungrammatical wh- dependencies has been widely studied in the GenSLA literature (e.g., Johnson & Newport, 1991; Martohardjono, 1993; Schachter, 1990; White, 1992). Juffs and Harrington (1995, 1996) utilized self-paced reading to investigate how processing influences wh- dependency formation in L2 learners. (12)

a. The boy read the book. b. Which book did the boy read? c. Which book did the boy say that the girl read? d. *Which book did the boy who read?

Two types of sentences that Juffs and Harrington examined include those in (13). In (13a), the wh- constituent “who” must be interpreted as the subject of “to fire”, while in (13b) it is interpreted as its object. Interestingly,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

28

J a s o n R ot h m a n , e t a l .

as first observed by Schachter and Yip (1990), L2 learners of English are less accurate in accepting sentences like (13b) compared to (13a), even though both are grammatical. This might be taken as evidence suggesting that L2 learners have difficulty acquiring knowledge of wh- dependencies in English, but Juffs and Harrington explored whether a processing-based explanation can account for the judgments given by L2 learners. (13)

a. Who does Tom expect to fire the manager? b. Who does Tom expect to fire?

In particular, although in (13a) “who” is ultimately interpreted as the subject of “to fire”, arriving at this interpretation during incremental sentence processing may involve several stages of reanalysis. For example, during processing “who” may initially be interpreted as the direct object of “expect” (“Who does Tom expect?”) and then the direct object of “to fire” (“Who does Tom expect to fire?”). It is only when “the manager” is reached that “who” can unambiguously be interpreted as the subject of “to fire”. Based on reaction time data, Juffs and Harrington argued that it is this reanalysis during processing that causes L2ers to judge such sentences as ungrammatical, rather than an inability to acquire the relevant knowledge of wh- dependencies. These results indicate that, like L1 speakers, L2 learners process wh- dependencies incrementally (see also Williams, Möbius, & Kims, 2001), and provide an example of how parsing considerations during real-time sentence processing can influence our understanding of the knowledge acquired by L2 learners. In their influential review of this and other related research on L2 sentence processing, Clahsen and Felser (2006a, 2006b) proposed the Shallow Structure Hypothesis (SSH) as an account of the similarities and differences in L1 and L2 processing. The SSH posits that L2ers have difficulty utilizing syntactic information during sentence processing, and instead rely more on semantic and discourse-level information. The hypothesis that syntactic processing might be the locus of difficulty during L2 sentence parsing sparked considerable interest in the SLA community, and motivated a great deal of research over the following decade that increasingly used online methodologies to investigate the role of syntactic and non-syntactic information during L2 acquisition and processing. For example, Felser et al. (2012) adopted a design previously used in L1 processing by Traxler and Pickering (1996) to investigate the application of island constraints during L2 processing. In (14a) and (14b), “the magazine” and “the shampoo” respectively are both ultimately the direct object of “about” but may initially be interpreted as the direct object of “read” during incremental processing. While both (14a) and (14b) are equally plausible at “about”, (14b) is implausible at “read”. In (14c), however, this initial dependency at “read” should be ruled out because it appears inside

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

29

of a relative clause. Felser et al. (2012) monitored the eye-movements of L1 English speakers and German L2 English learners as they read sentences like those in (14) and observed longer reading times at “read” in (14b) than (14a) in both groups, suggesting both L1 and L2 groups formed a dependency at the earliest available point. This contrast in plausibility at “read” was not seen in sentences like (14c), suggesting that both groups did not attempt to form a dependency when the verb appeared inside a relative clause. These results suggest L2 learners are able to utilize knowledge of island constraints to restrict dependency resolution during processing (see also Omaki & Schulz, 2011; but see Kim, Baek, & Tremblay, 2015, for a potential influence of transfer). (14)

a. Everyone liked the magazine that the hairdresser read extensively and with such enormous enthusiasm about before going to the salon. b. Everyone liked the shampoo that the hairdresser read extensively and with such enormous enthusiasm about before going to the salon. c. Everyone liked the magazine/the shampoo that the hairdresser who read extensively and with such enormous enthusiasm bought before going to the salon.

In investigating the role of syntactic information during L2 processing, other studies examined how L2ers utilize syntactic binding constraints on anaphora resolution during sentence comprehension (Felser & Cunnings, 2012; Felser, Sato, & Bertenshaw, 2009; Patterson, Trompelt, & Felser, 2014). Consider (15a), for example, where the reflexive “himself” must refer to “the boy”, and (15b), where the pronoun “him” must refer to “the man”. In the linguistics literature, constraints on the interpretation of reflexives and pronouns have typically been described in terms of syntactic binding constraints (Chomsky, 1981), such that reflexives, but not pronouns, must be bound by a local antecedent (i.e. “the boy”). (15)

a. The man said that the boy hurt himself. b. The man said that the boy hurt him.

In line with the long tradition of GenSLA research investigating whether L2ers can or cannot acquire such constraints (see Hawkins, 2001; White, 2003), Felser and Cunnings (2012) investigated the processing of binding constraints on reflexives in texts like (16). They found that L1 English speakers had longer reading times at the reflexive in (16b), when it mismatched with the stereotypical gender of the local antecedent “the soldier”, compared to (16a), when there was a stereotypical gender match (replicating Sturt, 2003). When they first encountered the reflexive during reading, however, German L2 learners had longer reading times when the non-local antecedent (She/He) mismatched the gender of the reflexive. The L2ers did indicate native-like knowledge of binding constraints in an offline task however, and their reading times after the reflexive were

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

30

J a s o n R ot h m a n , e t a l .

affected by the stereotypical gender manipulation. These results suggest the L2ers had acquired the binding constraints on reflexives in English, but temporarily violated them during processing. (16)



a. James/Helen has worked in the army hospital for years. He/She noticed that the soldier had wounded himself while on duty in the Far East. b. James/Helen has worked in the army hospital for years. He/She noticed that the soldier had wounded herself while on duty in the Far East.

In addition to studies that have manipulated gender congruence as a means of testing the application of syntactic binding constraints during processing, a number of studies have manipulated gender or number congruence to investigate L2 processing of morphosyntactic agreement. Much of this research has been motivated by long-standing questions in the GenSLA literature regarding whether L2 learners can acquire syntactic features not instantiated in their L1 and has adopted paradigms testing sentences like (17). Here, participants’ reading times at critical regions of text are recorded as an implicit measure of whether L2ers are sensitive to the (un)grammaticality of sentences like (17a) and (17b). In (17), longer reading times at “was” in ungrammatical (17b) than grammatical (17a) would be taken as evidence of L2 acquisition of the relevant agreement feature. (17)

a. The boy unsurprisingly was late for school again. b. *The boys unsurprisingly was late for school again.

The extent to which L2 learners show native-like sensitivity to such agreement violations has been widely debated. Jiang (2004, 2007), for example, claimed that Chinese L2 learners of English are insensitive to number agreement violations during processing, suggesting L2ers may have difficulty in acquiring agreement features (in this case number) not instantiated in the L1. A number of studies have since investigated which factors may influence L2 learners’ native-like sensitivity to agreement violations (e.g., Coughlin & Tremblay, 2013; Foote, 2011; Keating, 2009; Sagarra & Herschensohn, 2010, 2011, 2013). This has included work that has utilized neurolinguistic techniques such as event-related brain potentials (ERPs), where electrical activity in the brain is recorded as an implicit measure of language processing. Alemán Bañón, Fiorentino, and Gabriele (2014), for example, tested grammatical and ungrammatical gender agreement in Spanish, as in (18). (18)

a. El cuadro es auténtico y el grabado también. “The painting-MASC is authentic-MASC and the engraving too.” b. El cuadro es auténtica y el grabado también. “The painting-MASC is authentic-FEM and the engraving too.”

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

31

In (18a), the form of the adjective “auténtico” matches the gender of the noun “El cuadro”, and as such the sentence is grammatical, while (18b) is ungrammatical as the adjective is feminine while the noun is masculine. In recording participants’ brain responses to such sentences, Alemán Bañón, Fiorentino, and Gabriele (2014) were interested in testing whether both L1 and L2 speakers exhibit the so-called P600 effect (e.g., Friederici, Pfeiffer, & Hahne, 1993; Osterhout & Holcomb, 1992; Osterhout et al., 1996; Kaan et al., 2000) as an index of sensitivity to the (un)grammaticality of (18 a/b) at the critical adjective. Alemán Bañón, Fiorentino, and Gabriele indeed reported P600 effects for ungrammatical sentences for both L1 and L2 Spanish speakers. These results were interpreted as indicating that L2 learners can indeed acquire morphosyntactic features (in this case gender) that are not instantiated in the L1 (see also Alemán Bañón, Miller, & Rothman, 2017). In sum, the past decade of research in L2 sentence processing has brought considerable new insight into the role of processing in L2 acquisition and has led some researchers to reconsider the source of potential differences between L1 and L2 processing. In a recent review of research on L2 comprehension, Clahsen and Felser (2018) maintained the SSH as a viable account of L2 sentence processing, but alternative accounts have also been proposed. Cunnings (2017) proposed that L2ers construct similarly well-specified syntactic structures during processing as L1ers, but argued that L1–L2 differences that persist at high levels of L2 proficiency are best described in how information held in memory during sentence processing is accessed rather than shallow syntactic parsing. Other GenSLA (friendly) accounts have claimed that the primary source of difficulty during L2 processing relates to how lexical information is accessed during processing (Hopp, 2013) or a reduced ability to predict and anticipate upcoming information (Grüter, Rohde & Schafer, 2017). These different accounts provide different perspectives on the role of sentence processing in SLA, and will no doubt motivate further research that will provide new insight into L2 acquisition and processing in the future.

1.5 Conclusions As can be appreciated in this chapter via a rather reduced (by necessity) review of a subset of available GenSLA work in acquisition and processing, the remit of GenSLA is mainly focused on understanding the constraints on and the shape of mental L2 linguistic competence and how such knowledge interacts with L2 language processing. It is probably fair to say that while the paradigm is not focused on L2 language learning in the traditional sense, there is no question that the two are inherently related. Earnest and promising attempts at bridging GenSLA with

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

32

J a s o n R ot h m a n , e t a l .

language learning and teaching in the past few years in particular have been made (e.g., Marsden & Slabakova, 2017; Marsden, Whong, & Gil, 2018; Whong, Gil, & Marsden, 2013). The links, nevertheless, are inherent and have thus always existed. Uncovering how L2 acquisition and processing unfold in a cognitive sense is important for L2 language learning and classroom practice because they inform what is potentially more difficult and challenging in L2 learning, the relative role of certain variables that will either need to be compensated for or can be excluded from further consideration in pedagogical construction and intervention, what needs to be explicitly taught, and potentially what, over time and good quality exposure, comes for free, and so much more. Modelling language learning and application of what cognitive research reveals can increase L2 learning success if properly applied; after all, time is a rich and a scarce commodity, and so maximizing its use most effectively in adult L2 language teaching is not to be understated. One variable that will contribute to the best use of time and better construction of more effective pedagogical materials is the increased understanding of how the mind actually acquires and processes language in adulthood. Understanding if it is truly fundamentally different, and if so how, will reveal better paths for applied language learning and teaching. In this chapter, we have seen that there are two observable facts one should attempt to reconcile in order to provide an explanatorily adequate theory of L2 acquisition and processing in a cognitive sense, which can thus provide the grounds for an informed applied linguistics agenda for language teaching. First, that despite the ubiquity in L2 differences as compared to FLA, L2 grammars much more widely instantiate gems of successful acquisition and indications of fundamentally similar processing, which should not be expected under critical and sensitive period approaches. As we do not want to miss the forest for the trees by focusing only on global differences in L2 acquisition, it is important to herald all the predictions of claiming there is a critical or sensitive period for L2 syntax when evidence abounds that problematizes it. Such a claim does not simply intend that L2 syntax will look different from L1 syntax and/or be processed differently, it also claims that new properties such as the OPC, detailed in section 1.3, should not obtain in L2 grammars if not transferable from the L1—yet, as we have seen, they reliably do. Second, that despite this testament to the possibility of native-like proficiency, the majority of L2 learners reach a stable state at a level that is not comparable to native speakers. Since tapping underlying processing mechanisms is the only reliable way to determine if these differences are indeed representational or simply operational (or neither, or a combination of the two), GenSLA will continue, as it has recently, to direct considerable attention to processing methodologies. Doing so also helps build stronger bridges with applied linguistics. Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

33

References Abrahamsson, N., & Hyltenstam, K. (2008). The robustness of aptitude effects in near-native second language acquisition. Studies in Second Language Acquisition, 30, 481–509. Abrahamsson, N., & Hyltenstam, K. (2009). Age of onset and nativelikeness in a second language: Listener perception versus linguistic scrutiny. Language Learning, 59, 249–306. Alemán Bañón, J., Fiorentino, R., & Gabriele, A. (2014). Morphosyntactic processing in advanced second language (L2) learners: An event-related potential investigation of the effects of L1–L2 similarity and structural distance. Second Language Research, 30, 275–306. Alemán Bañón, J., Miller, D., & Rothman, J. (2017). Morphological variability in second language learners: An examination of electrophysiological and production data. Journal of Experimental Psychology: Learning, Memory, and Cognition, 43(10), 1509–1536. Ambridge B., & Lieven, E. (2011). Child language acquisition: Contrasting theoretical approaches. Cambridge: Cambridge University Press. Berwick, R. C., & Chomsky, N. (2017). Why only us: Recent questions and answers. Journal of Neurolinguistics, 43, 166–177. Berwick, R. C., Chomsky, N., & Piattelli-Palmarini, M. (2012). Poverty of the stimulus stands: Why recent challenges fail. In M. Piattelli-Palmarini & R. C. Berwick (eds.), Rich languages from poor inputs (pp. 19–42). Oxford: Oxford University Press. Berwick, R. C., Pietroski, P., Yankama, B., & Chomsky, N. (2011). Poverty of the stimulus revisited. Cognitive Science, 35(7), 1207–1242. Bley-Vroman, R. (1983). The comparative fallacy in interlanguage studies: The case of systematicity. Language Learning, 33(1), 1–17. Bley-Vroman, R. (1989). What is the logical problem of foreign language learning? In S. M. Gass & J. Schachter (eds.), Linguistic perspectives on second language acquisition (pp. 1–68). Cambridge: Cambridge University Press. Bley-Vroman, R. (2009). The evolving context of the fundamental difference hypothesis. Studies in Second Language Acquisition, 31, 175–198. Bruhn de Garavito, J. (2011). Subject/object asymmetries in the grammar of bilingual and monolingual Spanish speakers: Evidence against connectionism. Linguistic Approaches to Bilingualism, 1(2), 111–148. Bylund, E., Abrahamsson, N., & Hyltenstam, K. (2012). Does first language maintenance hamper nativelikeness in a second language? A study of ultimate attainment in early bilinguals. Studies in Second Language Acquisition, 34, 215–241. Bylund, E., Hyltenstam, K., & Abrahamsson, N. (2013). Age of acquisition effects or effects of bilingualism in second language ultimate attainment? In G. Granena & M. Long (eds.), Sensitive periods, language aptitude, and ultimate L2 attainment (pp. 69–102). Amsterdam: John Benjamins.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

34

J a s o n R ot h m a n , e t a l .

Carroll, S. E. (2001). Input and evidence: The raw material of second language acquisition. Amsterdam: John Benjamins Publishing. Chomsky, N. (1965). Aspects of the theory of syntax. Boston, MA: MIT Press. Chomsky, N. (1981). A note on non-control PRO. Journal of Linguistic Research, 1(4), 1–11. Chomsky, N. (1982). Some concepts and consequences of the theory of government and binding. Boston, MA: MIT Press. Chomsky, N. (1995). The minimalist program. Cambridge, MA: MIT Press. Chomsky, N. (2001). Derivation by phase. In M. Kenstowicz (ed.), Ken Hale: A life in language (pp. 1–52). Cambridge, MA: MIT Press. Clahsen, H., & Felser, C. (2006a). Grammatical processing in language learners. Applied Psycholinguistics, 27, 3–42. Clahsen, H., & Felser, C. (2006b). How native-like is non-native language processing? TRENDS in Cognitive Sciences, 10, 564–570. Clahsen, H., & Felser, C. (2018). Some notes on the Shallow Structure Hypothesis. Studies in Second Language Acquisition, 40(3), 693–706. Clahsen, H., & Muysken, P. (1989). The UG paradox in L2 acquisition. Second Language Research, 5, 1–29. Clark, E. (2003). Languages and representations. In D. Gentner & S. GoldinMeadow (eds.), Language in mind (pp. 17–23). Cambridge, MA: MIT Press. Crain, S. (2012). The emergence of meaning. Cambridge: Cambridge University Press. Coppieters, R. (1987). Competence differences between native and near-­ native speakers. Language, 63, 544–573. Coughlin, C. E., & Tremblay, A. (2013). Proficiency and working memory based explanations for nonnative speakers’ sensitivity to agreement in sentence processing. Applied Psycholinguistics, 34(3), 615–646. Cunnings, I. (2017). Parsing and working memory in bilingual sentence processing. Bilingualism: Language and Cognition, 20(4), 659–678. Curtiss, S. (1977). Genie: A psycholinguistic study of a modern day “wild child”. New York: Academic Press. Dąbrowska, E. (1997). The LAD goes to school: A cautionary tale for nativists. Linguistics, 35, 735–766. Dąbrowska, E. (2012). Different speakers, different grammars: Individual differences in native language attainment. Linguistic Approaches to Bilingualism, 2(3), 219–253. DeKeyser, R. M. (2000). The robustness of critical period effects in second language acquisition. Studies in Second Language Acquisition, 22, 499–533. Dörnyei, Z. (1998). Motivation in second and foreign language learning. Language Teaching, 31(3), 117–135. Dörnyei, Z. (2003). Attitudes, orientations, and motivations in language learning: Advances in theory, research, and applications. Language Learning, 53(S1), 3–32.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

35

Dörnyei, Z. (2014). The psychology of the language learner: Individual differences in second language acquisition. London: Routledge. Ellis, N. C. (1998). Emergentism, connectionism and language learning. Language Learning, 48, 631–664. Felser, C., & Cunnings, I. (2012). Processing reflexives in a second language: The timing of structural and discourse-level constraints. Applied Psycholinguistics, 33(3), 571–603. Felser, C., Cunnings, I., Batterham, C., & Clahsen, H. (2012). The timing of island effects in nonnative sentence processing. Studies in Second Language Acquisition, 34(1), 67–98. Felser, C., Sato, M., & Bertenshaw, N. (2009). The on-line application of Binding Principle A in English as a second language. Bilingualism: Language and Cognition, 12(4), 485–502. Fitch, W. T., & Hauser, M. D. (2004). Computational constraints on syntactic processing in a nonhuman primate. Science, 303(5656), 377–380. Fitch, W. T., Hauser, M. D., & Chomsky, N. (2005). The evolution of the language faculty: Clarifications and implications. Cognition, 97(2), 179–210. Foote, R. (2011). Integrated knowledge of agreement in early and late English–Spanish bilinguals. Applied Psycholinguistics, 32(1), 187–220. Franceschina, F. (2005). Fossilized second language grammars: The acquisition of grammatical gender. Amsterdam: John Benjamins. Friederici, A. D., Pfeifer, E., & Hahne, A. (1993). Event-related brain potentials during natural speech processing: Effects of semantic, morphological and syntactic violations. Cognitive Brain Research, 1(3), 183–192. Fuchs, E., & Flügge, G. (2014). Adult Neuroplasticity: More Than 40 Years of Research. Neural Plasticity, 2014: 541870, 1–10. Granena, G., & Long, M. (2013). Age of onset, length of residence, language aptitude, and ultimate L2 attainment in three linguistic domains. Second Language Research, 29(3), 311–343. Grüter, T., & Rohde, H. (2013). L2 processing is affected by RAGE: Evidence from reference resolution. Paper presented at The 12th Conference on Generative Approaches to Second Language Acquisition (GASLA). Grüter, T., Rohde, H., & Schafer, A. J. (2017). Coreference and discourse coherence in L2. Linguistic Approaches to Bilingualism, 7(2), 199–229. Guasti, M. (2002). Language development: The growth of grammar. Cambridge, MA: MIT Press. Hauser, M. D., Chomsky, N., & Fitch, W. T. (2002). The faculty of language: What is it, who has it, and how did it evolve? Science, 298(5598), 1569–1579. Hawkins, R. (2001). The theoretical significance of Universal Grammar in second language acquisition. Second Language Research, 17(4), 345–367.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

36

J a s o n R ot h m a n , e t a l .

Hawkins, R., & Casillas, G. (2008). Explaining frequency of verb morphology in early L2 speech. Lingua, 118, 595–612. Hawkins, R., & Chan, C. (1997). The partial availability of UG in second language acquisition: The Failed Functional Features Hypothesis. Second Language Research, 13, 187–226. Hopp, H. (2013). Grammatical gender in adult L2 acquisition: Relations between lexical and syntactic variability. Second Language Research, 29, 33–56. Jiang, N. (2004). Morphological insensitivity in second language processing. Applied Psycholinguistics, 25(4), 603–634. Jiang, N. (2007). Selective integration of linguistic knowledge in adult second language learning. Language Learning, 57, 1–33. Johnson, J. S. (1992). Critical period effects in second language acquisition: The effect of written versus auditory materials on the assessment of grammatical competence. Language Learning, 42(2), 217–248. Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60–99. Johnson, J. S., & Newport, E. L. (1991). Critical period effects on universal properties of language: The status of subjacency in the acquisition of a second language. Cognition, 39(3), 215–258. Juffs, A., & Harrington, M. (1995). Parsing effects in second language sentence processing: Subject and object asymmetries in wh- extraction. Studies in second language acquisition, 17(4), 483–516. Juffs, A., & Harrington, M. (1996). Garden path sentences and error data in second language sentence processing. Language learning, 46(2), 283–323. Kaan, E., Harris, A., Gibson, E., & Holcomb, P. (2000). The P600 as an index of syntactic integration difficulty. Language and Cognitive Processes, 15(2), 159–201. Kam, X. N. C., & Fodor, J. D. (2012). Children’s acquisition of syntax: Simple models are too simple. In M. Piattelli-Palmarini & R. C. Berwick (eds.), Rich languages from poor inputs (pp. 43–60). Oxford: Oxford University Press. Kam, X. N. C., Stoyneshka, I., Tornyova, L., Fodor, J. D., & Sakas, W. G. (2008). Bigrams and the richness of the stimulus. Cognitive Science, 32, 771–787. Kanno, K. (1997). The acquisition of null and overt pronominals in Japanese by English speakers. Second Language Research, 13, 265–87. Kanno, K. (1998). The stability of UG principles in second language acquisition. Linguistics, 36, 1125–1146. Keating, G. D. (2009). Sensitivity to violations of gender agreement in native and nonnative Spanish: An eye-movement investigation. Language Learning, 59(3), 503–535. Kim, E., Baek, S., & Tremblay, A. (2015). The role of island constraints in second language sentence processing. Language Acquisition, 22(4), 384–416.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

37

Lasnik, H., & Lidz, J. (2017). The argument from the poverty of the stimulus. In I. Roberts (ed.), Oxford Handbook of Universal Grammar (pp. 221–249). Oxford: Oxford University Press. Lenneberg, E. (1967). Biological foundations of language. New York: Wiley. Lewis, J. D., & Elman, J. L. (2001). Learnability and the statistical structure of language: Poverty of stimulus arguments revisited. In B. Skarabela, S. Fish, & A. H. J. Do (eds.), Proceedings of the twenty-sixth annual Boston University Conference on Language Development (pp. 359–370). Somerville, MA: Cascadilla. Linck, J. A., Osthus, P., Koeth, J. T., & Bunting, M. F. (2014). Working memory and second language comprehension and production: A meta-­ analysis. Psychonomic Bulletin & Review, 21(4), 861–883. Long, M. (2005). Problems with supposed counter-evidence to the critical period hypothesis. International Review of Applied Linguistics, 43(4), 287–317. Marsden, H. L., & Slabakova, R. (2017). Grammatical meaning and the second language classroom: introduction. Language Teaching Research, 1–11. Marsden, H. L., Whong, M., & Gil, K. H. (2018). What’s in the textbook and what’s in the mind: Polarity item “any” in learner English. Studies in Second Language Acquisition, 40(1), 1–28. Martohardjono, G. (1993). Wh-movement in the acquisition of a second language: A cross-linguistic study of 3 languages with and without overt movement. Unpublished PhD dissertation, Cornell University. Meisel, J. M. (2011). First and second language acquisition. Cambridge: Cambridge University Press. Montalbetti, M. (1984). After binding: On the interpretation of pronouns. Unpublished PhD dissertation, Massachusets Institute of Technology. Omaki, A., & Schulz, B. (2011). Filler-gap dependencies and island constraints in second-language sentence processing. Studies in Second Language Acquisition, 33(4), 563–588. Ortega, L. (2013). SLA for the 21st century: Disciplinary progress, transdisciplinary relevance, and the bi/multilingual turn. Language Learning, 63(S1), 1–24. Ortega, L. (2016). Multi-competence in second language acquisition: Inroads into the mainstream. In V. Cook and L. Wei (eds.), The Cambridge Handbook of Linguistic Multi-Competence (pp. 50–76). Cambridge: Cambridge University Press. Osterhout, L., & Holcomb, P. J. (1992). Event-related brain potentials elicited by syntactic anomaly. Journal of Memory and Language, 31(6), 785–806. Osterhout, L., McKinnon, R., Bersick, M., & Corey, V. (1996). On the language specificity of the brain response to syntactic anomalies: Is the syntactic positive shift a member of the P300 family? Journal of Cognitive Neuroscience, 8(6), 507–526. Paradis, M. (2009). Declarative and procedural determinants of second languages. Amsterdam: John Benjamins.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

38

J a s o n R ot h m a n , e t a l .

Patterson, C., Trompelt, H., & Felser, C. (2014). The online application of binding condition B in native and non-native pronoun resolution. Frontiers in Psychology, 5, 1–16. Pérez-Leroux, A. T., & Glass, W. (1997). OPC effects in the L2 acquisition of Spanish. In A. T. Pérez-Leroux & W. Glass (eds.), Contemporary perspectives on the acquisition of Spanish. Vol. 1: Developing grammars (pp. 149–165). Somerville, MA: Cascadilla Press. Pérez-Leroux, A. T., & Glass, W. (1999). Null anaphora in Spanish second language acquisition: Probabilistic versus generative approaches. Second Language Research, 15, 220–249. Prévost, P., & White, L. (2000). Missing surface inflection or impairment in second language acquisition? Evidence from tense and agreement. Second Language Research, 16, 103–133. Rankin, T. (2017). Working memory and L2 acquisition and processing. Second Language Research, 33(3), 389–399. Rankin, T., & Unsworth, S. (2016). Beyond poverty: Engaging with input in generative SLA. Second Language Research, 32, 563–572. Reali, F., & Christiansen, M. H. (2003). Reappraising poverty of stimulus argument: A corpus analysis approach. In A. Brugos, L. Micciulla, & C. E. Smith (eds.), Proceedings supplement of the 28th annual Boston University conference on language development (Epub, access at: http://www.bu.edu/bucld/ proceedings/supplement/vol28/). Somerville, MA: Cascadilla. Reali, F., & Christiansen, M. H. (2005). Uncovering the richness of the stimulus: Structure dependence and indirect statistical evidence. Cognitive Science, 29, 1007–1028. Roberts, L., González Alonso, J., Pliatsikas, C., & Rothman, J. (2018). Evidence from neurolinguistic methodologies: Can it actually inform linguistic/language acquisition theories and translate to evidence-based applications? Second Language Research, 34(1), 125–143. Robinson, P. (2001). Individual differences, cognitive abilities, aptitude complexes and learning conditions in second language acquisition. Second Language Research, 17(4), 368–392. Robinson, P. (2005). Aptitude and second language acquisition. Annual Review of Applied Linguistics, 25, 46–73. Robinson, P. (2013). Aptitude in second language acquisition. In C. Chapelle (ed.), The encyclopedia of applied linguistics (pp. 198–202). Malden, MA: Wiley–Blackwell. Ross, J. R. (1967). Constraints on variables in syntax. PhD dissertation, Massachusets Institute of Technology [later published in 1986 as Infinite syntax! (Norwood, NJ: Ablex).] Rothman, J. (2008). Why all counter-evidence to the critical period hypothesis in second language acquisition is not equal or problematic. Language and Linguistics Compass, 2(6), 1063–1088.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

Formal Linguistic Approaches to Adult L2 Acquisition and Processing

39

Rothman, J., & Guijarro-Fuentes, P. (2010). Input quality matters: Some comments on input type and age-effects in adult SLA. Applied Linguistics, 31(2), 301–306. Rothman, J., & Iverson, M. (2010). Independent normative assessments for bi/multilingualism, Where art thou. In M. Cruz-Ferrera (ed.), Multilingual norms (pp. 33–51). Frankfurt am Main: Peter Lang. Rothman, J., & Slabakova, R. (2018). State of the scholarship: The generative approach to SLA and its place in modern second language studies. Studies in Second Language Acquisition, 40(2), 417–442. Sagarra, N., & Herschensohn, J. (2010). The role of proficiency and working memory in gender and number agreement processing in L1 and L2 Spanish. Lingua, 120(8), 2022–2039. Sagarra, N., & Herschensohn, J. (2011). Proficiency and animacy effects on L2 gender agreement processes during comprehension. Language Learning, 61(1), 80–116. Sagarra, N., & Herschensohn, J. (2013). Processing of gender and number agreement in late Spanish bilinguals. International Journal of Bilingualism, 17(5), 607–627. Schachter, J. (1988). Second language acquisition and its relation to Universal Grammar. Applied Linguistics, 9, 219–235. Schachter, J. (1990). On the issue of completeness in second language acquisition. Interlanguage Studies Bulletin (Utrecht), 6(2), 93–124. Schachter, J., & Yip, V. (1990). Grammaticality judgements: Why does anyone object to subject extraction. Studies in Second Language Acquisition, 12, 379–392. Schwartz, B. D. (1986). The epistemological status of second language acquisition. Second Language Research, 2, 120–159. Schwartz, B. D. (1998). The second language instinct. Lingua, 106, 133–160. Schwartz, B. D., & Sprouse, R. A. (1996). L2 cognitive states and the full transfer/full access model. Second Language Research, 12, 40–72. Schwartz, B. D., & Sprouse, R. A. (2000). When syntactic theories evolve: Consequences for L2 acquisition research. In J. Archibald (ed.), Second language acquisition and linguistic theory (pp. 156–186). Malden, MA: Blackwell. Schwartz, B. D., & Sprouse, R. A. (2013). Generative approaches and the poverty of the stimulus. In J. Herschensohn & M. Young-Scholten (eds.), The Cambridge handbook of second language acquisition (pp. 137–158). Cambridge: Cambridge University Press. Slabakova, R. (2013). Adult second language acquisition: A selective overview with a focus on the learner linguistic system. Linguistic Approaches to Bilingualism, 3, 48–72. Sturt, P. (2003). The time-course of the application of binding constraints in reference resolution. Journal of Memory and Language, 48(3), 542–562.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

40

J a s o n R ot h m a n , e t a l .

Snyder, W. (2007). Child language: The parametric approach. Oxford: Oxford University Press. Traxler, M. J., & Pickering, M. J. (1996). Plausibility and the processing of unbounded dependencies: An eye-tracking study. Journal of Memory and Language, 35(3), 454–475. Tsimpli, I. M., & Dimitrakopoulou, M. (2007). The interpretability hypothesis: Evidence from wh- interrogatives in second language acquisition. Second Language Research, 23(2), 215–242. Ullman, M. T. (2005). A cognitive neuroscience perspective on second language acquisition: The declarative/procedural model. In C. Sanz (ed.), Mind and context in adult second language acquisition: Methods, theory and practice (pp. 141–178). Washington, DC: Georgetown University Press. Ullman, M. T. (2016). The declarative/procedural model: A neurobiological model of language learning, knowledge, and use. In S. L. Small (ed.), Neurobiology of language (pp. 953–968). San Diego, CA: Academic Press. White, L. (1985). Is there a “logical problem” of second language acquisition? TESL Canada Journal, 2(2), 29–41. White, L. (1992). Subjacency violations and empty categories in second language acquisition. In H. Goodluck & M. Rochemont (eds.), Island constraints (pp. 445–464). Dordrecht: Springer. White, L. (2003). Second language acquisition and Universal Grammar. Cambridge: Cambridge University Press. White, L. (2018). Formal linguistics and second language acquisition. In D. Miller, F. Bayram, J. Rothman, & L. Serratrice (eds.), Bilingualism: The state of the science across its subfields (pp. 57–77). Amsterdam and Philadelphia, PA: John Benjamins. Williams, J. N., Möbius, P., & Kim, C. (2001). Native and non-native processing of English wh- questions: Parsing strategies and plausibility constraints. Applied Psycholinguistics, 22(4), 509–540. Whong, M., Gil, K. H., & Marsden, H. (eds.) (2013). Universal Grammar and the second language classroom. Dordrecht: Springer. Yang, C., & Montrul, S. (2017). Learning datives: The Tolerance Principle in monolingual and bilingual acquisition. Second Language Research, 33, 119–144.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:21:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.002

2 Cognitive Approaches to Second Language Acquisition Nick C. Ellis and Stefanie Wulff 2.1 Introduction Cognitive approaches to L2 acquisition minimally share these two assumptions: • The primary source for both first (L1) and second language (L2) learning is the learner’s participative, contextualized experience of language. Language learning is largely usage-based. Humans use language in order to communicate and make meaning. • The cognitive mechanisms that learners employ in language learning are not exclusive to language learning, but are general cognitive mechanisms associated with learning of any kind. In this chapter, we describe the constructs and working assumptions that characterize such approaches to language learning, with a particular focus on their cognitive underpinnings and how these explain differences between the linguistic forms that distinguish L1 and L2 speakers. We first define constructions as the targets of language learning and then describe the processes of construction learning in terms of ­ exemplar-­ based, rational, associative learning. Not all constructions are equally learnable by all learners: naturalistic second language learners process open-class words more efficiently than grammatical cues even though the grammatical cues may be more frequent. We outline a usage-based account of this phenomenon in terms of salience, contingency, and redundancy, and explain how effects of learned attention and blocking further limit learning in adult L2 learners. We describe educational interventions taking these findings into consideration and conclude with further readings on usage-based approaches to L2 acquisition.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

42

N i c k C . E ll i s an d S t e f an i e W u l f f

There are other relevant and interesting aspects of usage-based second language acquisition (SLA) that we simply cannot deal with here. We have made the conscious choice to focus in this chapter upon L1–L2 differences in morphosyntax. In the final section, we provide pointers to more social, interactional, and meaning-based investigations of L2 cognition.

2.2  Constructions as the Targets of L2 Acquisition Learning a language involves the learning of constructions, which are the conventionalized form–meaning mappings used in a speech community. Constructions include morphemes—the smallest pairing of form and meaning in language—as well as words, phrases, and syntactic frames (Goldberg, 2006; Trousdale & Hoffmann, 2013). Simple morphemes such as –aholic (meaning “being addicted to something”) are constructions in the same way as simple words like nut (meaning “a fruit consisting of a hard or tough shell around an edible kernel”), idioms like It is driving me nuts (meaning “It is greatly frustrating me”), and abstract syntactic frames like Subject–Verb–Object–Object (meaning that something is being transferred to someone, as realized in sentences as diverse as Max gave the squirrel a nut, Nick gave Max a hug, or Steffi baked Max a cake, where nuts, hugs, and cakes are being transferred, respectively). Including abstract syntactic frames means that not all constructions carry meaning in the traditional sense, but rather serve a functional or meaningful purpose. For example, the passive construction serves the function of shifting the attentional focus in an utterance from the agent of the action to the patient undergoing the action (compare the passive A cake was baked for Max with its active counterpart Steffi baked Max a cake). From this, it follows that constructions have to be stored in multiple forms simultaneously that differ in their level of complexity and abstraction. For instance, the word nut and the plural –s morpheme are both simple constructions that are also stored as constituent parts of the more complex construction nuts (“more than one nut”). Different levels of constructional abstraction (also referred to as schematization) are evident in the fully lexicalized formula You’re welcome versus the partially schematized slot-and-frame greeting pattern [Good + (time of day)], which renders lexicalized phrases like Good afternoon and Good evening, and the completely schematic [Adjective + Noun Phrase] construction, which in turn could be lexically specified as happy baby, delicious cake, or grand opening, to give just three examples. This wide definition of constructions entails a blurred dividing line between the lexicon and the grammar, or what traditional approaches have labelled words and rules: from a construction grammar perspective,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

43

a sentence is not the product of applying a rule that strings a number of words into a particular order, but the product of combining a number of constructions—some simple, some complex, some lexically specific, some abstract—in a particular way. A sentence like What did Max give the squirrel, for instance, combines the following constructions: • • • •

Max, squirrel, give, what, do constructions VP, NP constructions Subject–Verb–Object–Object construction Subject–Auxiliary inversion question construction.

We can therefore equate an adult speaker’s knowledge of their language(s) to a huge warehouse of constructions that vary in terms of complexity and abstraction. Constructions have properties that specify if and how they can combine with other constructions; these properties are mostly semantically and/or functionally motivated such that constructions can only be combined if their meanings or functions are compatible or can at least temporarily attain compatibility in a specific context or discourse situation (Goldberg, 2006). Constructional compatibility is crucially solidified by the frequency with which they are used (and therefore, heard) together: the more often they co-occur, the more entrenched that particular constructional arrangement becomes. Likewise, L2 learners will acquire constructions first in the contexts of the constructions with which they most often co-occur in the input before they gradually expand the repertoire of combinations to less frequent combinations and even acceptable novel combinations.

2.3 The Processes of L2 Acquisition: Exemplar-Based Rational Contingency Analysis In other words, language learning means learning the associations within and between constructions. Constructionist accounts of language acquisition involve the distributional analysis of the language stream and the parallel analysis of meaning in terms of contingent perceptual experience, with abstract constructions being learned from the conspiracy of concrete exemplars of usage following statistical learning mechanisms relating input and learner cognition (see Rebuschat & Williams, 2012, on statistical learning mechanisms). Psychological analyses of this learning of constructions is informed by the literature on the associative learning of cue–outcome contingencies that hinge on both construction-related and learner-related factors. For constructions, their frequency of experience, salience of form, significance of meaning, prototypicality, redundancy versus surprise value, and the contingency of form and function seem to

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

44

N i c k C . E ll i s an d S t e f an i e W u l f f

be relevant factors; for learners, cognitive factors like learned attention, automaticity, transfer, overshadowing, and blocking each play important roles (Ellis, 2008b). These various psycholinguistic factors conspire in the acquisition and use of any linguistic construction (see Ellis & Wulff, 2015a, 2015b, for a detailed discussion of each factor). Psycholinguistic research has demonstrated that, generally, the more frequently a construction (or combination of constructions) is experienced, the earlier it is acquired and the more fluently it is processed (Ellis, 2002). Words such as one or give occur more frequently than sixteen or syndicate, and the learner’s perceptual system accordingly attunes to the probabilities of these constructions in the input. When a learner notices a construction for the first time, this can result in a unitary representation in memory that binds all its properties (i.e., phonological make-up, spelling, etc.) together. This representation is subsequently activated whenever the construction’s properties are noticed in the language environment, so it serves as a form of detector or pattern-­ recognition unit. Whenever the detector unit’s activation threshold is met, it will fire. With each firing, the resting level of activation of the detector unit increases (and correspondingly, the threshold for firing decreases)—in other words, it is readied, or primed, for re-activation. This priming effect accrues over a speaker’s lifespan such that frequently occurring constructions and the properties associated with them obtain habitually high resting activation levels. In the same fashion, the form–function mappings between a phonological form and its interpretation are strengthened through continued use: every encounter of /wʌn/ as one strengthens the association between the two; every encounter of /wʌn/ signalling won is tallied as well; as is the association between /wʌn/ when it is the initial part of wonderland. After a first memory representation is formed, the language system compares each subsequent exemplar that the learner encounters in their language environment against that representation, and gradually modifies it to fit the accumulating experience of that construction, its properties, and its contexts. Since repeated encounters with exemplars of a construction manifest similar or identical properties time and again, prototypes emerge that then serve as the basis of comparison for future encounters. Prototypes are knowledge representations of the most typical properties of a construction. They are mental constructs in the sense that they are abstractions of a learner’s accumulated encounters of sufficiently similar exemplars. Prototypes are the defining centrepieces of categories: they are maximally similar to other members of that category and maximally dissimilar to non-members of that category. For example, people are quicker to confirm that sparrows are birds than they are with other kinds of birds, like geese or albatrosses. This is because sparrows are more prototypical

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

45

birds: they unite the most typical characteristics of birds in terms of size, beak shape, wing length, etc. Importantly, this adaptive fine-tuning of a learner’s language representations is not conscious and explicit in nature, but happens unconsciously and implicitly. As far as properties of categories are concerned (whether it is a conceptual category like bird or a linguistic category like noun phrase), learners do not consciously inventory frequencies in the cognitive and linguistic environment; instead, statistical learning happens unconsciously (Ellis, 1994; Rebuschat, 2015). Another important tenet of usage-based theories in this context is that no principled distinction is drawn between linguistic and other cognitive categories. Psycholinguistic research has demonstrated prototypicality, neighbourhood, and other categorization effects in learning quasi-regular patterns of construction form. For instance, people are fastest when asked to produce regular forms (like, for example, plural sparrow + s), slower and less accurate at generating more marked forms (like finch + es), and slowest still to produce irregular forms (such as geese; Chater & Manning, 2006; Seidenberg & Plaut, 2014).

2.4  Usage Leads to an Emerging Language System Through usage experience, form–function mappings are woven into a network of construction forms and their meanings. This language system is sometimes referred to as the “constructicon”. Through this network, activation spreads as a function of the learned probabilities of the different form–meaning associations that a speaker has formed over his or her lifespan. The resulting mental model is, at any time in language development, a custom-tailored, adaptively fine-tuned reflection of the learner’s summed language experience (Ellis, 2006a). In that sense, language learning is rational as defined in the field of rational cognition: a major impetus for human psychology is to adapt behaviour as best as possible to its environmental conditions (Anderson, 1989). Language learning is also emergent in the sense that the mechanisms that learners employ are few and simple, yet the knowledge networks that arise from employing these mechanisms over time are complex, dynamic, and adaptive (Ellis, 1998; Beckner et al., 2009; Ellis & Larsen-Freeman, 2009). Furthermore, language is a complex-adaptive system in the sense that it involves many agents (people communicating with each other) in many different configurations (individuals, groups, networks, and cultures), and it operates across many different levels of the system architecture (neurons, brains, and bodies; phonemes, constructions, interactions, and discourses), as well as on multiple time scales (evolution, epigenesis, ontogenesis, interactional,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

46

N i c k C . E ll i s an d S t e f an i e W u l f f

neuro-synchronic, and diachronic; Ellis, Römer, & O’Donnell, 2016; MacWhinney & O’Grady, 2015).

2.5 Lexical and Grammatical Constructions in L1 and L2 Acquisition As stated above, the frequency of use of a construction drives its learnability. That said, not all frequent constructions are equally learnable to all learners. In early stages of acquisition (and for many learners, even after years of language immersion), learners process open-class words (nouns, verbs, adjectives, and adverbs) more efficiently than grammatical cues. The limited ultimate attainment of L2 learners who remain at that stage stabilizes at a “Basic Variety” of interlanguage that is less grammatically sophisticated than that of native-like L1 ability (Klein & Perdue, 1992; Bardovi-Harlig, 1992). Although naturalistic L2 learners are exposed to rich target language input, only a subset of that input turns into intake that promotes further language development (Corder, 1967). A classic case study demonstrating the limitations of intake is that of the naturalistic language learner Wes, who was described as being very fluent, with high levels of strategic competence, but low levels of grammatical accuracy: “using 90% correct in obligatory contexts as the criterion for acquisition, none of the grammatical morphemes counted have changed from unacquired to acquired status over a five year period” (Schmidt, 1984, p. 5). The Basic Variety might be sufficient for everyday communicative purposes, but grammatical morphemes tend not to be put to full use (e.g., Bardovi-Harlig, 1992; Clahsen & Felser, 2006; Schmidt, 1984; VanPatten, 1996, 2007). For example, learners initially reference time by use of temporal adverbs, prepositional phrases, serialization, and calendric reference, with the grammatical expression of tense and aspect emerging only slowly thereafter, if at all (Bardovi-Harlig, 1992, 2000; Klein, 1998; Lee, 2002; Meisel, 1987; Noyau, Klein, & Dietrich, 1995). L2 learners have been found to prefer adverbial over inflectional cues to tense in naturalistic L2 acquisition (e.g., Bardovi-Harlig, 2000; Noyau, Klein, & Dietrich, 1995), training experiments (e.g., Cintrón-Valentín & Ellis, 2015; Ellis et al., 2014), and studies of L2 language processing alike (e.g., Sagarra & Ellis, 2013; VanPatten, 2007). A key challenge for L2 acquisition research is therefore to explain why closed-class constructions are more difficult to learn than open-class constructions, for some learners throughout their L2 career. Usage-based theories attribute this to three tenets of the psychology of learning: the learnability of a construction is affected by (i) salience, (ii) contingency of form–function association, and (iii) learned attention.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

47

2.6  Salience and Learning Less salient cues are less readily learned than highly salient ones (Ellis, 2006c, 2017; Rescorla & Wagner, 1972). Salience refers to the property of a stimulus which makes it stand out from the rest and consequently more likely to be perceived, attended to, and entered into subsequent cognitive processing and learning. Salience can be determined independently by physics and the environment, and by our knowledge of the world: • The physical world, our embodiment, and our sensory systems jointly render certain sensations to be more intense (louder, brighter, heavier, etc.) than others. • As we experience the world, we learn from it, and our resultant knowledge values some associations higher than others (James, 1890, p. 82). A favoured stimulus stands out, either because of weighty associations ($500000.0 vs. $0.000005, however similar the number of pixels, characters, or ink in their sensation) or because it matches a motivational state (a meal when hungry but not when full). Psychological salience is experience-dependent: hotdog, sushi, and 寿司 mean different things to people of different cultural and linguistic experience. This is why, unlike sensation, the units of perception are subjective and therefore cannot simply be measured in physical terms. This is reflected in Miller’s definition of the units of short-term memory as “chunks”: “We are dealing here with a process of organizing or grouping the input into familiar units or chunks, and a great deal of learning has gone into the formation of these familiar units” (Miller, 1956, p. 91). Rescorla and Wagner (1972) presented a formal model of conditioning that describes the capacity of any cue (Conditioned Stimulus, CS; for example a bell in Pavlovian conditioning) to become associated with an outcome (Unconditioned Stimulus, US; for example food in Pavlovian conditioning) on any given encounter of their pairing. The formula below is arguably the most influential formula in the history of learning theory, encapsulating over eighty years of research. It elegantly unites psychophysical salience, psychological salience, and surprisal: dV = ab(L – V). The associative strength of the US to the CS is referred to by the letter V, and the change in this strength which occurs on each trial of conditioning is called dV. On the right-hand side, a is the salience of the US, b is the salience of the CS, and L is the amount of processing given to a completely unpredicted, surprising, US. Thus, both the salience of the cue (a) and the psychological importance of the outcome (b) are essential factors in any associative learning. As for (L – V), the more a CS is associated with a US,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

48

N i c k C . E ll i s an d S t e f an i e W u l f f

the less additional association the US can induce. As Beckett (1954) put it: “habit is a great deadener.” Alternatively, with novel associations where V is close to zero, there is much surprisal, and consequently much learning: first impressions, first love, first time … One factor determining the learning of construction form is psychophysical salience. In his landmark study of first language acquisition, Brown (1973) breaks down the measurement of perceptual salience, or “clarity of acoustical marking” (p. 343), into “such variables as amount of phonetic substance, stress level, usual serial position in a sentence, and so on” (p. 463). Temporal phrases (such as on my birthday, at Christmas, etc.), temporal adverbs (yesterday, tomorrow, later, etc.), and other lexical temporal cues (morning, winter, etc.) are salient and stressed in the speech stream, while verb inflections usually are not. Many grammatical form–function relationships in English, such as grammatical particles and inflections like the third person singular –s, are of low salience in the language stream. This is a consequence of the well-documented effect of frequency and automatization in the evolution of language. The basic principles of automatization apply to all kinds of motor activities and skills (like playing a sport or a musical instrument) as well as to languages across the world: through repetition, previously independent sequences of units come to be processed as a single unit or chunk (Ellis, 1996). The more frequently they use a form, the more speakers reduce it. Zipf (1949) summarized this in the principle of least effort— speakers want to minimize articulatory effort, and this leads to brevity and phonological reduction. They tend to choose the most frequent words, and the more they use them, automatization of production causes their shortening. Grammatical functors are the most frequent words of a language, and so they lose their emphasis and tend to become abbreviated and phonologically fused with surrounding material (Bybee, 2008; Jurafsky et al., 2001; Zuraw, 2003). In a corpus study by Cutler and Carter (1987), 86% of strong syllables occurred in open-class words and only 14% in closed-class words; for weak syllables, 72% occurred in closed-class words and 28% in open-class words. Since grammatical function words and bound inflections tend to be short and unstressed, they are difficult to perceive from the input. When words are clipped out of connected speech and presented in isolation, adult native speakers perceive open-class words (buy, four, know, ewe, etc.) 90 to 100% correctly, but grammatical function words (by, for, no, you, etc.) only 40% to 50% of the time (Herron & Bates, 1997). Clitics (accentless words or particles that depend accentually on an adjacent accented word and form a prosodic unit together with it) are the most extreme examples of this: the /s/ of “he’s”, /l/ of “I’ll” and /v/ of “I’ve” can never be

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

49

pronounced in isolation. If native speakers have trouble correctly perceiving grammatical function words in isolation, the task for the L2 learners is even greater. In sum, grammatical functors are extremely difficult to perceive based on bottom-up auditory evidence alone. Fluent language processors provide top-down schematic support to perceive these elements in continuous speech. However, this top-down knowledge is exactly what learners lack: they haven’t had sufficient experience in the L2 and corresponding retuning of their L1 system to develop a sufficiently schematized knowledge system (or constructicon) that would afford them the same levels of L2 top-down support as in fluent L1 processing. Thus, the low psychophysical salience of grammatical functors contributes to L2 learners’ difficulty in learning them (Ellis, 2006c; Goldschneider & DeKeyser, 2001). Salience effects are compounded by redundancy. Grammatical morphemes often appear in redundant contexts where their interpretation is not essential for correct interpretation of the sentence (Schmidt, 2001; Terrell, 1991; VanPatten, 1996). For instance, tense markers often appear in contexts where other cues have already established the temporal reference (e.g. yesterday he walked), plural markers are accompanied by quantifiers or numerals (10 nuts), etc. Since their neglect does not result in communicative breakdown, they carry little psychological importance of the outcome (term b in the Rescorla–Wagner equation).

2.7  Contingency and Learning The degree to which animals (including humans) learn associations between cues and outcomes also depends upon the contingency of the relationship. In classical conditioning, it is the reliability of the bell as a predictor of food that determines how easily this association is acquired (Rescorla, 1968). In language learning, it is the reliability of the form as a predictor of an interpretation that determines its acquisition and processing (Ellis, 2006b; Gries & Ellis, 2015; Gries & Stefanowitsch, 2004; MacWhinney, 1987). The last thirty years of psychological research into humans’ sensitivity to cue–outcome contingencies (Shanks, 1995) demonstrates that when given sufficient exposure to a relationship, people’s judgments match the contingency specified by ΔP (the one-way dependency statistic; Allan, 1980). ΔP measures the directional association between a cue and an outcome, as illustrated in Table 2.1. a, b, c, and d represent frequencies of occurrence. For example, a is the frequency of conjunctions of the cue and the outcome, and c is the number of times the outcome occurred without the cue.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

50

N i c k C . E ll i s an d S t e f an i e W u l f f

Table 2.1.  A contingency table showing the four possible combinations of events showing the presence or absence of a target Cue and an Outcome. Outcome

No Outcome

Cue

a

b

No cue

c

d

ΔP is the probability of the outcome given the cue P(O|C) minus the probability of the outcome in the absence of the cue P(O|¬C), which can be calculated using this formula: a

c

ΔP = P(O|C) – P(O|¬C) = a + b – c + d When the outcome is just as likely when the cue is present as when it is not, there is no covariation between the two events, and ΔP amounts to 0. As the presence of the cue increases the likelihood of the outcome, ΔP approaches 1.0. A learnable cue is one where the outcome is there whenever the cue is there, and where the outcome is not there when the cue is not there either, i.e., where a and d are large and b and c are small. The less reliably a form is associated with a particular meaning or functional interpretation, the more difficult learning becomes (Ellis, 2006b; Shanks, 1995). Cues with multiple interpretations are ambiguous and thus hard to resolve; cue–outcome associations of high contingency are reliable and readily processed. Consider how, in the learning of the category of birds, while eyes and wings are equally frequently experienced features in the exemplars, it is wings which are distinctive in differentiating birds from other animals. Wings are important features to learning the category of birds because they are reliably associated with class membership while being absent from other categories of animals. Raw frequency of occurrence is therefore less important for construction learning than the contingency between cue and interpretation. In language, there are rarely 1:1 mappings between forms and their interpretations. Cue–outcome reliability can be reduced in two directions: either forms have multiple interpretations (polysemy and homophony) or interpretations are realized by more than one form (synonymy). The same usage-phenomenon that promotes the reduction of frequently used words also drives grammatical functors towards homophony: different functions associated with forms that were originally distinct eventually merge into the same shortened form. An example is the –s suffix in English: in modern English, it has come to encode a plural form (squirrels), it indicates

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

51

possession (Max’s toy), and it marks third person singular present (Nick sleeps). The –s form is abundantly frequent in learners’ input, but not reliably associated with any or just one of these meanings and functions (increasing b in Table 2.1). Conversely, the plural, possessive, and third person singular constructions are all realized by more than one form: they are all variably expressed by the allomorphs [s], [z], and [ɨz]. Thus, if we assess just one of these, say [ɨz], as a cue for one particular outcome, say plurality, then it is clear that there are many instances of that outcome in the absence of the cue (c in Table 2.1). In short, the low cue-interpretation contingency of plurals and many other highly frequent grammatical constructions (see Gries, 2015) makes them difficult to learn (DeKeyser, 2005; Ellis, 2008a; Goldschneider & DeKeyser, 2001).

2.8  Learned Attention L2 acquisition is vulnerable to attentional biases that stem from the L2 learners’ knowledge of a prior language. For example, Ellis (2006a, 2006c) attributes L2 difficulties in acquiring inflectional morphology to an effect of learned attention known as “blocking” (Kamin, 1969; Kruschke, 2006; Kruschke & Blair, 2000; Mackintosh, 1975). Blocking is an associative learning phenomenon that occurs in animals and humans alike. It shifts the learner’s attention to selective aspects of the input as a result of prior experience (Rescorla & Wagner, 1972; Shanks, 1995; Wills, 2005). Knowing that a particular stimulus is associated with a particular outcome makes it harder to learn that another cue paired with that same outcome is also a good predictor of it, and so the prior association effectively “blocks” further associations. For example, all languages have lexical and phrasal means of expressing temporality, so anyone with knowledge of any first language is aware that there are reliable and frequently used lexical cues to temporal reference (words like German gestern, French hier, Spanish ayer, English yesterday). Such are cues to look out for in an L2 because of their frequency, their reliability of interpretation, and their salience. Learned attention theory holds that, once known, such cues block the acquisition of less salient and less reliable verb tense morphology from analysis of redundant utterances such as Yesterday I walked. Benati (2013) reviews a series of studies showing learners are better able to identify temporal reference when presented with temporal adverbs rather than verbal morphology. A number of theories of L2 acquisition incorporate related notions of transfer and learned attention. The Competition Model (MacWhinney, 2001; MacWhinney & Bates, 1989), for one, focuses on dealing with competition between multiple linguistic cues to interpretation. Similarly,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

52

N i c k C . E ll i s an d S t e f an i e W u l f f

Input Processing (IP) theory (VanPatten, 1996) includes a Lexical Preference Principle: “Learners will process lexical items for meaning before grammatical forms when both encode the same semantic information” (VanPatten, 2007, p. 118) as well as a Preference for Nonredundancy Principle: “Learners are more likely to process nonredundant meaningful grammatical markers before they process redundant meaningful markers” (VanPatten, 2007, p. 119). The basic mechanisms of learned attention in SLA have been examined through a series of experiments in which participants learned a small number of Latin expressions and their English translations. Ellis and Sagarra (2011) included three groups: an Adverb Pretraining, Verb Pretraining, and Control group. In Phase 1, the Adverb Pretraining group learned two adverbs and their temporal reference—hodie “today” and heri “yesterday”; the Verb Pretraining group learned verbs (shown in either first, second, or third person) and their temporal reference—e.g., cogito present or cogitavisti past; and the Control group had no pretraining. In Phase 2, all groups were shown sentences which appropriately combined an adverb and a verb (e.g. heri cogitavi, hodie cogitas, cras cogitabis) and learned whether these sentences referred to the past, the present, or the future. In Phase 3, the Reception test, all combinations of adverb and verb tense marking were presented individually and participants were asked to judge whether each sentence referred to the past, present, or future. The logic of the design was that in Phase 2, every utterance contained two temporal references—an adverb and a verb inflection. If participants paid equal attention to both cues, their judgments in Phase 3 should be equally affected by them. If, in contrast, they paid more attention to adverb (or verb) cues, then their judgments would be biased towards them in Phase 3. The results showed that the three groups reacted to the cues in very different ways—the Adverb Pretraining group followed the adverb cue, the Verb Pretraining group tended to follow the verb cue, and the Control group fell in between. Multiple regression analyses, one for each group, with the group mean temporal interpretation for each of the Phase 3 strings as the dependent variable and the information conveyed by the adverbial and verbal inflection cues as independent variables, showed in standardized ß coefficients, Adverb Group Time = 0.99Adverb – 0.01Verb; Verb Group Time = 0.76Adverb + 0.60Verb; Control Group Time = 0.93Adverb + 0.17Verb. This experiment demonstrated how short-term instructional manipulations affect attention to language. Ellis and Sagarra (2010, 2011) furthermore illustrated long-term language transfer effects from the L1. They found that the nature of learners’ first language (+/– verb tense morphology) biased the acquisition of morphological versus lexical cues to temporal reference in the same subset of Latin. First language speakers of Chinese (no tense morphology) were less able than first language

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

53

speakers of Spanish or Russian (rich morphology) to acquire inflectional cues from the same language experience where adverbial and verbal cues were equally available, with learned attention to tense morphology being in standardized ß coefficients: Chinese (–0.02) < English (0.17) < Russian (0.22) < Spanish (0.41) (Ellis & Sagarra, 2011, Table 4). Ellis et al. (2014) replicated Ellis and Sagarra (2010), extending the investigation using eye-tracking measures to determine the extent to which short-term learned attention biases in the acquisition of temporal reference in L2 Latin in English as a foreign language (EFL) learners are overt or covert biases. The results indicated that prior experience of particular cue dimensions affected upon what participants overtly focused during subsequent language processing, which in turn resulted in covert attentional biases in comprehension and in productive knowledge. These learned attention effects include elements of both positive and negative transfer. Prior use of adverbial cues causes participants to pay more attention to adverbs, i.e., a positive effect of entrenchment of the practised cue. At the same time, however, increased sensitivity to adverb cues is accompanied by a reduced sensitivity to morphological cues—blocking. A meta-analysis of the combined results of Ellis and Sagarra (2010, 2011) demonstrated that the average effect size of entrenchment was large (+1.23) while that of blocking was moderate (–0.52). Sagarra and Ellis (2013) showed the results of blocking over years of learning in intermediate and advanced learners of Spanish (as opposed to one hour of learning Latin). 120 English (a morphologically simple language) and Romanian (a morphologically rich language) learners of Spanish (also a morphologically rich language) and 98 English, Romanian, and Spanish monolinguals read sentences in L2 Spanish (or their L1 for the monolinguals) containing adverb–verb or verb–adverb congruencies or incongruencies. Eye-tracking data revealed significant effects of incongruency (all participants were sensitive to tense incongruencies), cue location in the sentence (participants spent more time at their preferred cue), L1 experience (learners and monolinguals with morphologically rich L1s looked longer at verbs than learners and monolinguals with morphologically simple L1s), and L2 experience (intermediate learners read more slowly and regressed longer than advanced learners). Experience with the second language is shaded by attentional biases and other types of interference from the first language (Flege, 2002; Jarvis & Pavlenko, 2008; Lado, 1957; MacWhinney, 1997; Odlin, 1989). As a result of this interference, second language learning is rarely entirely native-like in outcome, even if the learner is surrounded by dense input. Since everything is filtered through the lens of the L1, not all of the relevant input can be taken advantage of (hence Corder’s distinction between input and intake; Corder, 1967).

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

54

N i c k C . E ll i s an d S t e f an i e W u l f f

It is important to emphasize here that the limitations of L2 learning do not license the conclusion that L2 learning is qualitatively different from L1 learning. Second language learners employ the same statistical learning mechanisms that they employed when they acquired their first language. Rather, first language learning is (nearly always) so incredibly successful that it—somewhat ironically—hampers second language learning.

2.9  Implications for Language Teaching The fact that L2 learners have to learn to adjust the attention biases they obtained through their L1 has consequences for L2 instruction. Children acquire their first language primarily in an implicit manner. Implicit learning is the learning of complex information without conscious attention to what is being learned. In contrast, L2 acquisition is largely characterized by explicit learning. For reviews on implicit and explicit language learning, see Ellis (1994) and Rebuschat (2015). Schmidt’s (2001) Noticing Hypothesis holds that “people learn about the things they attend to and do not learn much about the things they do not attend to” (Schmidt, 2001, p. 30). That is, in order to successfully acquire specific aspects of their L2, learners must pay conscious and selective (i.e., focused) attention to the target structures. Given the bottleneck effect of input versus intake discussed above, even in dense-input, immersive environments, explicit learning and teaching gain even more relevance for the second language learners in foreign language environments with only limited L2 input. This holds in particular for aspects of form in the L2 that are redundant or polysemous and/or lack perceptual salience as discussed above. Form-focused instruction (FFI) thus attempts to encourage noticing, drawing learners’ attention to linguistic forms that might otherwise be ignored (Ellis, 2012). Different variants of FFI vary in the degree and manner in which they recruit learner consciousness and in the role of the learner’s metalinguistic awareness of the target forms (Ellis, 2005). Norris and Ortega (2000) compared the outcomes from studies that employed differing levels of explicitness of L2 input in a meta-analysis. Their results suggest that FFI instruction results in substantial target-oriented L2 gains, that explicit types of instruction are more effective than implicit types, and that the L2 instruction has durable effects. Similarly, more recent meta-analyses of effects of type of instruction by Spada and Tomita (2010) and Goo et al. (2015) report large advantages of explicit instruction in L2 acquisition. However, the studies gathered in these meta-analyses used a wide variety of types of instruction, learner, targeted feature, and method of assessment that future research should control to determine how robust effects of FFI are.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

55

Cintrón-Valentín and Ellis (2015, 2016) used eye-tracking to investigate whether different types of FFI can aid learners in overcoming learned attention and blocking biases. English and Chinese native speakers viewed Latin utterances combining lexical and morphological cues to temporality under control conditions (CC) and three types of explicit focus on form (FonF): verb grammar instruction (VG); verb salience with textual enhancement (VS); and verb pretraining (VP). All participants completed an exposure phase, comprehension test, and production test. VG participants viewed a short lesson on Latin tense morphology prior to exposure; VS participants saw the verb inflections highlighted in bold and red during exposure; and VP participants had an additional introductory phase where they were trained on solitary verb forms and their English translations (the rationale being that when the verb is presented in isolation rather than in potentially redundant combination with adverbial cues, there is less scope for blocking). CC participants were significantly more sensitive to the adverbs than verb morphology, while instructed participants showed greater sensitivity to morphological cues in comprehension and production. Such results demonstrate that form-focused instruction recruits learners’ explicit, conscious processing capacities and allows them to consolidate unitized form–function bindings of novel L2 constructions (Ellis, 2005). When a construction is learned this way, its use in subsequent implicit processing can update the statistical tallying of its frequency of usage and probabilities of form–function mapping.

2.10  Conclusion and Further Reading This chapter has focused upon salience, contingency, and learned-attention in usage-based accounts of the L2 acquisition of morphology. In so doing, we have covered in detail just one facet of usage-based approaches to the L2 acquisition of linguistic form. Ellis and Wulff (2015a, 2015b) and Ortega et al. (2016) provide broader descriptions. Ellis et al. (2016) describe a large body of complementary work showing the joint effects of typetoken frequency, contingency, and prototypicality in the usage-based L1 and L2 acquisition and processing of verb-argument constructions. There are many other relevant and interesting aspects of usage-based SLA relating to social and cultural motivations: The target for many second language learners is not just “to speak another language”, but to become part of the social and cultural environment in which the language is used. This entails frequent and rich participation in the second-language life worlds into which the learner “bricolages” his or her way. (Wagner, 2015, p. 75)

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

56

N i c k C . E ll i s an d S t e f an i e W u l f f

Digital technologies give increasing opportunities for rich “rewilding” of education (Thorne, 2018), and such embodied, environmentally embedded, enacted, socially encultured, and situated environments support rich language learning (Eskildsen & Wagner, 2015). Beckner et al. (2009), Cadierno and Eskildsen (2015), Douglas Fir Group (2016), Hulstijn et al. (2014), and Ellis (2019) outline how the cognitive underpinnings of usage-based approaches can be integrated with a social perspective on SLA. Robinson and Ellis (2008), Littlemore (2009), and Tyler (2012) give broader overviews of applied cognitive-linguistic research in L2 learning and teaching.

References Anderson, J. R. (1989). A rational analysis of human memory. In H. L. I. Roediger & F. I. M. Craik (eds.), Varieties of memory and consciousness: Essays in honour of Endel Tulving (pp. 195–210). Hillsdale, NJ: Lawrence Erlbaum Associates. Allan, L. G. (1980). A note on measurement of contingency between two binary variables in judgment tasks. Bulletin of the Psychonomic Society, 15, 147–149. Bardovi-Harlig, K. (1992). The use of adverbials and natural order in the development of temporal expression. IRAL, 30, 299–320. Bardovi-Harlig, K. (2000). Tense and aspect in second language acquisition: Form, meaning, and use. Oxford: Blackwell. Beckett, S. (1954). Waiting for Godot. New York: Grove Press. Beckner, C., Blythe, R. A., Bybee, J., Christiansen, M. H., Croft, W., Ellis, N. C., & Schoenemann, T. (2009). Language is a complex adaptive system. Position paper. Language Learning, 59(S1), 1–26. Benati, A. (2013). The input processing theory in second language acquisition. In M. P. García Mayo, M. J. Gutierrez Mangado, M. Martínez Adrián, F. Myles, J. Rothman, & B. VanPatten (eds.), Contemporary approaches to second language acquisition (pp. 93–110). Amsterdam: John Benjamins. Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press. Bybee, J. (2008). Mechanisms of change in grammaticization: the role of frequency. In B. D. Joseph & R. D. Janda (eds.), The Handbook of Historical Linguistics (pp. 602–623). Malden, MA: Wiley. Cadierno, T., & Eskildsen, S. W. (eds.) (2015). Usage-based perspectives on second language learning. Berlin: De Gruyter Mouton. Chater, N., & Manning, C. (2006). Probabilistic models of language processing and acquisition. Trends in Cognitive Science, 10, 335–344. Cintrón-Valentín, M., & Ellis, N. C. (2015). Exploring the interface: Explicit focus-on-form instruction and learned attentional biases in L2 Latin. Studies in Second Language Acquisition, 37, 197–235.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

57

Cintrón-Valentín, M., & Ellis, N. C. (2016). Salience in second language acquisition: Physical form, learner attention, and instructional focus. Frontiers in Psychology, 7, 1284. Clahsen, H., & Felser, C. (2006). Grammatical processing in language learners. Applied Psycholinguistics, 27(1), 3–42. Corder, S. P. (1967). The significance of learners’ errors. International Review of Applied Linguistics, 5, 161–169. Cutler, A., & Carter, D. M. (1987). The predominance of strong initial syllables in the English vocabulary. Computer Speech & Language, 2(3), 133–142. DeKeyser, R. (2005). What makes learning second-language grammar difficult? A review of issues. Language Learning, 55(S1), 1–25. Douglas Fir Group. (2016). A transdisciplinary framework for SLA in a multilingual world. The Modern Language Journal, 100, 19–47. Ellis, N. C. (ed.). (1994). Implicit and explicit learning of languages. San Diego, CA: Academic Press. Ellis, N. C. (1996). Sequencing in SLA: Phonological memory, chunking, and points of order. Studies in Second Language Acquisition, 18(1), 91–126. Ellis, N. C. (1998). Emergentism, connectionism and language learning. Language Learning, 48(4), 631–664. Ellis, N. C. (2002). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24(2), 143–188. Ellis, N. C. (2005). At the interface: Dynamic interactions of explicit and implicit language knowledge. Studies in Second Language Acquisition, 27, 305–352. Ellis, N. C. (2006a). The Associative-Cognitive CREED. In B. V. Patten & J. Williams (eds.), Theories in second language acquisition: An introduction (pp. 77–96). Mahwah, NJ: Lawrence Erlbaum Associates. Ellis, N. C. (2006b). Language acquisition as rational contingency learning. Applied Linguistics, 27(1), 1–24. Ellis, N. C. (2006c). Selective attention and transfer phenomena in SLA: Contingency, cue competition, salience, interference, overshadowing, blocking, and perceptual learning. Applied Linguistics, 27(2), 1–31. Ellis, N. C. (2008a). The dynamics of second language emergence: Cycles of language use, language change, and language acquisition. The Modern Language Journal, 41(3), 232–249. Ellis, N. C. (2008b). Usage-based and form-focused language acquisition: The associative learning of constructions, learned-attention, and the limited L2 endstate. In P. Robinson & N. C. Ellis (eds.), Handbook of cognitive linguistics and second language acquisition (pp. 372–405). London: Routledge. Ellis, N. C. (2017). Salience. In M. Hundt, S. Pfenninger, & S. Mollin (eds.), The changing english language (pp. 71–92). Cambridge: Cambridge University Press.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

58

N i c k C . E ll i s an d S t e f an i e W u l f f

Ellis, N. C. (2019). Essentials of a theory of second language cognition. The Modern Language Journal, 103, 39–60. Ellis, N. C., & Larsen-Freeman, D. (eds.). (2009). Language as a complex adaptive system. Mahwah, NJ: Wiley. Ellis, N. C., Hafeez, K., Martin, K. I., Chen, L., Boland, J., & Sagarra, N. (2014). An eye-tracking study of learned attention in Second Language Acquisition. Applied Psycholinguistics, 35, 547–579. Ellis, N. C., Römer, U., & O’Donnell, M. B. (2016). Language usage, acquisition, and processing: Cognitive and corpus investigations of construction grammar. Malden, MA: Wiley-Blackwell. Ellis, N. C., & Sagarra, N. (2010). The bounds of adult language acquisition: Blocking and learned attention. Studies in Second Language Acquisition, 32(4), 553–580. Ellis, N. C., & Sagarra, N. (2011). Learned attention in adult language acquisition: A replication and generalization study and meta-analysis. Studies in Second Language Acquisition, 33(4), 589–624. Ellis, N. C., & Wulff, S. (2015a). Second language acquisition. In E. Dąbrowska & D. Divjak (eds.), Handbook of cognitive linguistics (pp. 409– 431). Berlin: DeGruyter Mouton. Ellis, N. C., and Wulff, S. (2015b). Usage-based approaches in second language acquisition. In B. VanPatten & J. Williams (eds.), Theories in second language acquisition: An introduction (pp. 75–93). London and New York: NY Routledge. Ellis, R. (2012). Form-Focused Instruction and Second Language Learning. In R. Ellis (ed.) Language Teaching Research and Pedagogy (pp. 217–306). Hoboken, NJ: Wiley-Blackwell. Eskildsen, S. W., & Wagner, J. (2015). Embodied L2 construction learning. Language Learning, 65, 268–297. Flege, J. (2002). Interactions between the native and second-language phonetic systems. In P. Burmeister, T. Piske, & A. Rohde (eds.), An integrated view of language development: Papers in honor of Henning Wode (pp. 217–244). Trier: Wissenschaftlicher Verlag Trier. Goldberg, A. E. (2006). Constructions at work: The nature of generalization in language. Oxford: Oxford University Press. Goldschneider, J. M., & DeKeyser, R. (2001). Explaining the “natural order of L2 morpheme acquisition” in English: A meta-analysis of multiple determinants. Language Learning, 51(1), 1–50. Goo, J., Granena, G., Yilmaz, Y., & Novella, M. (2015). Implicit and explicit instruction in L2 learning. In P. Rebuschat (ed.), Implicit and explicit learning of languages (pp. 443–482). Amsterdam: John Benjamins. Gries, S. Th. (2015). Polysemy. In E. Dąbrowska & D. S. Divjak (eds.), Handbook of Cognitive Linguistics (pp. 472–490). Berlin: De Gruyter Mouton. Gries, S. Th., & Ellis, N. C. (2015). Statistical measures for usage-based linguistics. Currents in Language Learning, 2, 228–255.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

59

Gries, S. Th., & Stefanowitsch, A. (2004). Extending collostructional analysis: A corpus-based perspective on alternations. International Journal of Corpus Linguistics, 9, 97–129. Han, Z.-H., & Odlin, T. (eds.) (2006). Studies of fossilization in second language acquisition. Clevedon: Multilingual Matters. Herron, D., & Bates, E. (1997). Sentential and acoustic factors in the recognition of open- and closed-class words. Journal of Memory and Language, 37, 217–239. Hulstijn, J. H., Young, R. F., Ortega, L., Bigelow, M. H., DeKeyser, R., Ellis, N. C., & Talmy, S. (2014). Bridging the gap: Cognitive and social approaches to research in second language learning and teaching. Studies in Second Language Acquisition, 33(3), 361–421. James, W. (1890). The principles of psychology (Vol. 2). New York: Holt. Jarvis, S., & Pavlenko, A. (2008). Crosslinguistic influence in language and cognition. New York: Routledge. Jurafsky, D., Bell, A., Gregory, M., & Raymond, W. D. (2001). Probabilistic relations between words: Evidence from reduction in lexical production. In J. Bybee & P. Hopper (eds.), Frequency and the emergence of linguistic structure (pp. 229–254). Amsterdam: John Benjamins. Kamin, L. J. (1969). Predictability, surprise, attention, and conditioning. In B. A. Campbell & R. M. Church (eds.), Punishment and aversive behavior (pp. 276–296). New York: Appleton-Century-Crofts. Klein, W. (1998). The contribution of second language acquisition research. Language Learning, 48, 527–550. Klein, W., & Perdue, C. (1992). Utterance structure: Developing grammars again. Amsterdam: John Benjamins. Kruschke, J. K. (2006). Learned Attention. Paper presented at the Fifth International Conference on Development and Learning, Indiana University. Kruschke, J. K., & Blair, N. J. (2000). Blocking and backward blocking involve learned inattention. Psychonomic Bulletin & Review, 7, 636–645. Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers. Ann Arbor, MI: University of Michigan Press. Lee, J. F. (2002). The incidental acquisition of Spanish future morphology through reading in a second language. Studies in Second Language Acquisition, 24, 55–80. Littlemore, J. (2009). Applying cognitive linguistics to second language learning and teaching. Basingstoke and New York: Palgrave Macmillan. Mackintosh, N. J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276–298. MacWhinney, B. (1987). The Competition Model. In B. MacWhinney (ed.), Mechanisms of language acquisition (pp. 249–308). Hillsdale, NJ: Erlbaum. MacWhinney, B. (1997). Second language acquisition and the Competition Model. In A. M. B. De Groot & J. F. Kroll (eds.), Tutorials in bilingualism:

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

60

N i c k C . E ll i s an d S t e f an i e W u l f f

Psycholinguistic perspectives (pp. 113–142). Mahwah, NJ: Lawrence Erlbaum Associates. MacWhinney, B. (2001). The Competition Model: The input, the context, and the brain. In P. Robinson (ed.), Cognition and second language instruction (pp. 69–90). New York: Cambridge University Press. MacWhinney, B., & Bates, E. (1989). The crosslinguistic study of sentence processing. Cambridge: Cambridge University Press. MacWhinney, B., & O’Grady, W. (eds.) (2015). The handbook of language emergence. Oxford: Wiley-Blackwell. Meisel, J. (1987). Reference to past events and actions in the development of natural second language acquisition. In C. Pfaff (ed.), First and second language acquisition (pp. 206–224). New York: Newbury House. Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81–97. Norris, J., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and quantitative meta-analysis. Language Learning, 50, 417–528. Noyau, C., Klein, W., & Dietrich, R. (1995). Acquisition of temporality in a second language. Amsterdam: John Benjamins. Odlin, T. (1989). Language transfer. New York: Cambridge University Press. Ortega, L., Tyler, A. E. In Park, H., & Uno, M. (eds.) (2016). The usagebased study of language learning and multilingualism. Washington, DC: Georgetown University Press. Rebuschat, P. (ed.). (2015). Implicit and explicit learning of language. Amsterdam: John Benjamins. Rebuschat, P., & Williams, J. N. (eds.) (2012). Statistical learning and language acquisition. Berlin: Mouton de Gruyter. Rescorla, R. A. (1968). Probability of shock in the presence and absence of CS in fear conditioning. Journal of Comparative and Physiological Psychology, 66, 1–5. Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (eds.), Classical conditioning II: Current theory and research (pp. 64–99). New York: Appleton-Century-Crofts. Robinson, P., & Ellis, N. C. (eds.) (2008). Handbook of cognitive linguistics and second language acquisition. London: Routledge. Sagarra, N., & Ellis, N. C. (2013). From seeing adverbs to seeing morphology. Language experience and adult acquisition of L2 tense. Studies in Second Language Acquisition, 35, 261–290. Schmidt, R. (1984). The strengths and limitations of acquisition: A case study of an untutored language learner. Language, Learning, and Communication, 3, 1–16. Schmidt, R. (2001). Attention. In P. Robinson (ed.), Cognition and second language instruction (pp. 3–32). Cambridge: Cambridge University Press.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

Cognitive Approaches to Second Language Acquisition

61

Seidenberg, M. S., & Plaut, D. C. (2014). Quasiregularity and its discontents: The legacy of the past tense debate. Cognitive Science, 38(6), 1190–1228. Shanks, D. R. (1995). The psychology of associative learning. New York: Cambridge University Press. Simon, H. A. (1957). Models of man: Social and rational. New York: Wiley and Sons. Slobin, D. (1996). From “thought and language” to “thinking for speaking”. In J. J. Gumperz & S. Levinson (eds.), Rethinking linguistic relativity (pp. 70–96). Cambridge: Cambridge University Press. Spada, N., & Tomita, Y. (2010). Interactions between type of instruction and type of language feature: A meta-analysis. Language Learning, 60, 263–308. Terrell, T. (1991). The role of grammar instruction in a communicative approach. The Modern Language Journal, 75, 52–63. Thorne, S. L. (2018). Technologies, morphologies of communicative action, and the rewilding of language education. Paper presented at the American Association for Applied Linguistics, Chicago, IL. Trousdale, G., & Hoffmann, T. (eds.) (2013). Oxford handbook of construction grammar. Oxford: Oxford University Press. Tyler, A. (2012). Cognitive linguistics and second language learning: Theoretical basics and experimental evidence. Abingdon: Routledge. VanPatten, B. (1996). Input processing and grammar instruction in second language acquisition. New York: Ablex. VanPatten, B. (2007). Input processing in adult second language acquisition. In B. VanPatten & J. Williams (eds.), Theories in second language acquisition: An introduction (pp. 115–135). Mahwah, NJ: Lawrence Erlbaum Associates. Wagner, J. (2015). Designing for language learning in the wild: Creating social infrastructures for second language learning. In T. Cadierno & S. W. Eskildsen (eds.), Usage-based perspectives on second language learning (pp. 75–101). Berlin: de Gruyter Mouton. Wills, A. J. (2005). New directions in human associative learning. Mahwah, NJ: Erlbaum. Zipf, G. K. (1949). Human behaviour and the principle of least effort: An introduction to human ecology. Cambridge, MA: Addison-Wesley. Zuraw, K. (2003). Probability in language change. In R. Bod, J. Hay, & S. Jannedy (eds.), Probabilistic linguistics (pp. 139–176). Cambridge, MA: MIT Press.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 02:42:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.003

3 The Qualitative Science of Vygotskian Sociocultural Psychology and L2 Development Rémi A. van Compernolle

3.1 Introduction Since the early papers of Frawley and Lantolf (1985; Lantolf & Frawley, 1984), Vygotskian psychology, often referred to as sociocultural theory (SCT), has gained prominence as one of several “mainstream” (Swain & Deters, 2007) or “alternative” (Atkinson, 2011) second language acquisition (SLA) theories. A central concern of the general theory is how sign systems (e.g., language, literacy, numeracy) are internalized to reorganize basic, or biologically endowed, psychological functions into higher, or culturally mediated, ones, which give rise to consciousness. In particular, Vygotsky (1986) focused on the role of language in constituting higher mental functions: language does not simply facilitate cognition, but it is part and parcel of it. In this way, Vygotsky’s semiotic analysis of consciousness and word meaning connects him not only to Marxian dialectical materialism, but also to the German psycholinguistic tradition inspired by such scholars as Hegel and Herder, as Leitch (2011) has pointed out: “Human consciousness is formed through linguistic interactions, and the language which constitutes consciousness is therefore always part of it, unable to be separated” (p. 306). This perspective on the constitutive role of language in the formation of consciousness has a number of important implications for how we operationalize and study SLA processes and outcomes, and why SCT research in general favours situated qualitative inquiry over the more common quantitative experimentalist model of science that derives from the logical positivism developed by the Vienna Circle in the early twentieth century (Packer, 2011).

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

63

This chapter examines the extension of Vygotskian sociocultural psychology to second language (L2) research as a qualitative science of L2 development. To do so, I draw on Packer’s (2011) critique of social science that aims to turn contextualized human subjects into abstract objects through common qualitative research procedures such as the coding and thematic analysis of L2 data, interviews, and journals and diaries. I argue that Vygotsky’s theory compels us to situate rather than abstract L2 phenomena in relation to their histories and contexts of occurrence, focusing on the recontextualizability of principles rather than the generalizability of empirical observations. To make the case, I present an overview of the main tenets of Vygotsky’s theory and how they relate to qualitative science. I then present exemplar studies that highlight what Vygotsky (1986) referred to as the “individually specific” (p. 5) in terms of the unity of the social environment of L2 development and cognition. In concluding, I synthesize the discussion and make recommendations for future research in this domain of L2 studies. It should be noted from the outset that the current chapter has the aim of examining and illustrating the ontological and epistemological underpinnings of SCT research in its own right. The chapter does not set out to compare and contrast SCT with so-called mainstream SLA theories, nor does it propose to sketch out an SCT “perspective on” or “account of” mainstream SLA constructs such as input, implicit and explicit knowledge, feedback, predictable paths of development, and so on. Readers interested in such comparaisons are referred to the wellknown publications of Lantolf and Thorne (2006), as they will not be repeated here.

3.2  Qualitative Science and Vygotskian Psychology Packer (2011) opens his wide-ranging treatise on the scientific underpinnings of qualitative research with the following observation about research priorities in countries such as the United States: We are told repeatedly that there is a “gold standard” for research in the social sciences, the randomized clinical trial. Other kinds of research—typically cast as naturalistic, observational, and descriptive—are viewed as mere dross in comparison, good only for generating hypotheses, not for testing them. They are seen as lacking the rigor necessary for truly scientific research and as failing to offer practical solutions to pressing problems … Any suggestion that there might be inquiry that follows a logic … different from that of traditional experimental research is dismissed. The possibility that complex human phenomena might require a kind of investigation that traces them in time and space and explores how they are constituted is not considered. (p. 1)

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

64

R é m i A . va n Co m p e r n o l l e

He goes on to observe that there exists a clichéd view of experimentalist quantitative research, such as the clinical trial, and qualitative research. While the former is seen to be objective, focused on causes and capable of testing hypotheses and providing explanations of phenomena, the latter is considered to be subjective, focused on experience (rather than facts) and only capable of describing phenomena and generating hypotheses to be tested in quantitative research (Packer, 2011, p. 19). Packer’s point, however, is not to pit quantitative against qualitative methods of data analysis; indeed, he argues throughout the book that qualitative science often involves quantification, especially in the descriptive phase of research. The problem, as he sees it, is that so-called “qualitative researchers” have for too long attempted to carry out their work within an incommensurable scientific paradigm: logical positivism, later modified by Popper (1959) to follow a “‘hypothetico-deductive’ logic” (Packer, 2011, p. 27).1 In other words, investigations are driven by research questions designed to test the null hypothesis (i.e., falsification), an epistemological orientation that requires researchers to make the ontological assumption that the component parts of an object of study (e.g., human behaviour) can be separated and investigated independently of one another. We see this play out especially in quantitative experimental research where a dependent variable (e.g., test scores) is studied in relation to such abstracted independent variables as treatment condition (e.g., control versus intervention) and individual characteristics (e.g., age, gender, socioeconomic status). Researchers who attempt to do the same by using qualitative materials (e.g., interviews, recorded interactions, field notes) run the risk of obscuring their object of study by removing it from its context of occurrence. Conducting thematic analyses of interviews and other qualitative materials (e.g., journals, natural interaction) so that the data become comparable (i.e., abstract, generalizable) across a defined population is a common example. Qualitative science, Packer argues, must follow an alternative logic, one that assumes the constituent pieces of a phenomenon have to be studied in situ, not as separable parts, but as a system of dynamic relations. Such relations are established in particular lived experiences and vary from one invidual to the next. This is why qualitative inquiry, and especially case study, is the key to understanding our phenomena of interest. Packer’s (2011) view of the science of qualitative research connects with Marxian dialectical logic, which expands “our notion of anything to include, as aspects of what it is, both the process by which it has become that and the broader interactive context in which it is found” (Ollman, 2003,  Discussion is beyond the scope of this chapter, but the interested reader is referred to Packer’s (2011, pp. 28–39)

1

discussion of the project of the logical positivists of the Vienna Circle, as well as his interpretation of Kuhn’s (1962) oft-cited book, The Structure of Scientific Revolutions, in relation to social scientific research today.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

65

p. 18).2 This is what Packer (pp. 167–207) means when he describes the concept of constitution. Any given phenomenon has a history of formation; it is constituted in and by the conditions in which it comes into being. At the same time, the phenomenon in question shapes the conditions in which it occurs and in which it will recur, change, or disappear in the future. The process of mutual constitution continues indefinitely, in principle. Packer’s sentiments are reflected in the Vygotskian, or cultural-­historical, psychology literature as well. Commenting on the field of developmental psychology in general, Valsiner and van der Veer (2014) opine that: Developmental psychology has failed to put the premises of the developmental science into its theoretical and empirical practice. As the discipline continues toward accumulation of evidence across persons (large samples) and dismisses the historical linkages within a sample (Valsiner and Sato, 2006) it necessarily passes by the phenomena of development and makes methodological decisions that render the discovery of developmental principles conceptually impossible. Psychology’s efforts to “measure” characteristics that are created by the very act of such “measurement” have backfired in the construction of a vast variety of “variables” (Toomela and Valsiner, 2010). Such constructed entities are correlated with one another ad nauseam in contemporary psychology, using theoretical systems as superficial covers. Statistics has become a via regia for the road of psychology to be a “true Science”—resulting in an illustrious hell of pseudo-empiricist discourses—as Jan Smedslund (1979, 1997) has claimed over the past thirty years. (pp. 148–149)

It is important to highlight that this is not an argument against any and all uses of measurement and quantification; there are certainly appropriate times to measure, count, score, conduct statistical tests, and so on.3 But the problem to which Valsiner and van der Veer point is that the science of development, which is necessarily longitudinal (i.e., involving change over time) and qualitative (i.e., involves the formation of new functions), has been left by the wayside in favour of producing statistical findings, typically representing a single snapshot in time, for the sake of producing statistical findings. The solution to this problem, according to Valsiner and van der Veer (2014), may be found in two of Vygotsky’s most important concepts. The first, the zone of proximal development (ZPD), represents a transitional stage between two developmental states. It is the actual locus of development, when a particular function is no longer what it was in the past, but it has not yet matured into a new, qualitatively different form. Vygotsky’s  In the L2 field, Lantolf and Poehner (2014) have recently argued that dialectics is the metatheory that underpins

2

Vygotskian SCT.  Indeed, Vygotsky and his colleagues frequently quantified aspects of the development of psychological functions

3

(e.g., Vygotsky, 1978, 1986). However, quantification was used to describe phenomena at a general level; they then provided qualitative accounts of underlying processes.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

66

R é m i A . va n Co m p e r n o l l e

(1978) oft-referenced definition of the ZPD as the difference between what one is capable of achieving alone and what becomes possible under guidance from another person underscores an important aspect of the concept: with assistance, teaching, modelling, and so on, psychological functions that are not yet matured, but are in an embryonic state, can be brought into being. In so doing, performance under guidance shows that the person has advanced past a previous, more stable developmental state, but that he or she has not yet reached the next, or proximal, state of functioning. The ZPD is in a very real sense the zone in which a person is transitioning from one psychological state to a new, qualitatively different one. Valsiner and van der Veer (2014) argue that psychology tends to dismiss phenomena occurring in this transitional zone as “error” (p. 152) because they are messy, in constant flux, and hard—if not impossible—to quantify, preferring instead to measure the more stable states that precede and follow it. The result is that psychological research documents the end products of development in various states, but “is blind to the study of development [because it eliminates] developmental phenomena at the outset” (Valsiner & van der Veer, 2014, p. 152). By contrast, a qualitative science of L2 development is fundamentally interested in this messiness. The second concept, perezhivanie in Russian (in English, “personal experience”), is less well known than the ZPD, but it underpinned Vygotsky’s enterprise from his earliest work, The Psychology of Art (1971). Vygotsky used the term perezhivanie to refer to “the process of experiencing and the state of ‘living-through’” (Valsiner & van der Veer, 2014, p. 160), and this assumed a unity of both intellectual and affective processes occurring in concrete social contexts (Vygotsky, 1994; see also González Rey & Martínez, 2016; Mok, 2015; and Smagorinsky, 2011). For Vygotsky, developmental phenomena (i.e., the ZPD) were necessarily formed in, shaped by, and in turn shaped, perezhivanie—people think and feel as they experience the unique and ever-changing social situations in which they are developing, as they internalize the cultural tools (e.g., language and other sign systems) necessary for regulating their psychological functioning. Consequently, it becomes impossible to separate intellect from affect, or psychological functioning from the environment in which it is situated, because intellect, affect, and environment together form a unified whole (Chaiklin, 2002; Davydov, 1995; Zinchenko, 2009). It follows that, in contrast to psychology’s emphasis on accumulating statistical evidence across persons (Valsiner & van der Veer, 2014) and studying phenomena by breaking them into so-called component parts, what is necessary is to particularize the developmental trajectory of individuals through situated, qualitative investigations. This is because no two individuals could be expected to develop in the same way, even if the social situation of their development appears similar (or even the same, e.g., siblings in the

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

67

same home, students in the same classroom), for the simple reason that no two individuals experience development in the same way. The concept of perezhivanie therefore compels researchers to investigate the constitution (Packer, 2011) of developmental phenomena as they occur in the ZPD, to examine how a person’s process of experiencing and livingthrough a particular moment in time (e.g., a minute, hour, day, week, year, decade) shapes his or her psychological development across their lifespan and in concrete practical activity. In other words, it is a holistic, or organic (Zinchenko, 2002, 2009), approach to studying and explaining the dynamics of development. Mok (2015) reminds us that Vygotsky (1994) distinguished psychology’s focus on an elemental approach to analysis (i.e., focusing on constituent parts, or elements) from his proposal of a unitary approach (i.e., focusing on holistic units), as embodied in the concept of perezhivanie. As Mok writes: Units and elements are both empirically identified actual parts (i.e., not merely analytic constructs) of the larger phenomenon under investigation. Where they differ is that units retain the dynamic relations (e.g., those characterizing its growth, development, and on-going operation) of the whole by maintaining a unity of what would otherwise be reduced to separate elements. (p. 150)

Vygotsky (1986) provided a well-known illustration of the difference between an analysis of elements and an analysis of units. Elemental analysis, writes Vygotsky, “may be compared to the chemical analysis of water into hydrogen and oxygen, neither of which possesses the properties of the whole and each of which possesses properties not present in the whole” (p. 4). Thus, studying the constituent parts of water—the elements of hydrogen and oxygen—may lead to generalization about each one but leaves unaddressed the question of how, for example, the unit of liquid water extinguishes fire since “hydrogen burns and oxygen sustains fire” (Vygotsky, 1986, p. 4). A unitary analysis, by contrast, would focus not simply on the unit of water (H20) as a chemical formula but the behaviour of specific units of water in concrete circumstances and in relation to the larger whole of which they are a part. In fact, this is precisely what climate scientists are doing as they study the interconnections and behaviours of melting polar ice and tundra, the Gulf Stream, and extreme weather events such as tropical storms and flooding. The point, as Vygotsky wrote, is not to identify “[the phenomenon’s] most general characteristics” (pp. 4–5) but to understand “the individually specific” (p. 5).4 So it is in the study  It is important to recognize that identifying general characteristics can be important in many fields, including

4

psychology. However, as Vygotsky (1986, p. 4) noted, this does not equate to explanatory analysis of the phenomenon under study, only a description of some basic features of the phenomenon that can be identified in most—if not all—circumstances. Unitary analysis seeks to explain a specific, concrete instantiation of a phenomenon rather than describe an idealized, abstracted version of it.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

68

R é m i A . va n Co m p e r n o l l e

of human cognition, including in L2 development. Indeed, as Levinson (2012) has pointed out, the “original sin” of cognitive science was to deny “variation and diversity in human cognition” (p. 397) by presuming a single universal and uniform mental capacity. Our interest, therefore, must be focused on generating an understanding of the specifics of an individual’s or a group’s developmental trajectory, and the dynamic relations that exist between its parts, rather than attempting to identify and correlate the most general characteristics of it.

3.3  Researching the Constitution of L2 Development Understanding the constitution of L2 development in general and in instructed contexts in particular, as we will see, creates challenges for researchers. Reasons for this are many, but a central one is that our field has a long history of attempting to produce generalizable findings that can be applied in, or at least inform, practice—namely, in the form of pedagogical implications. This is certainly an important goal, and one that has for a long time been embraced, and recently amplified, by SCT scholars as an “imperative” for the field (Lantolf & Poehner, 2014). However, as in most social sciences, the dominant paradigm of “‘hypothetico-deductive’ logic” (Packer, 2011, p. 27) has pushed the field to privilege (quasi)experimental quantitative research that engages in the analysis of elements over most other forms of investigation. Research syntheses in instructed SLA showing a statistical advantage of explicit instruction over implicit instruction (e.g., Norris & Ortega, 2000; Taguchi, 2015) are prime examples. Such research tells us that, at a general level, the provision of metalinguistic information tends to help most learners more than withholding such information, but the specific dynamic relations between how pedagogy is carried out on the ground and such elements as attention, perception, memory, learner histories, and social relations as they occur in the process of development are largely ignored. In other words, what we get is a description of decontextualized general characteristics correlated with measures of development (e.g., test scores, performance accuracy statistics) rather than an explanation of “the actual causal-dynamic relations that underlie” the phenomenon under study (Vygotsky, 1978, p. 62). The reader will note, of course, that quantitative research syntheses (e.g., meta-analysis) will exclude nonquantitative, nonexperimental research because the method of synthesis—statistics— cannot handle qualitative evidence. The result is that what is purported to be the collective knowledge of our field, which forms the basis for language education policy as well as teacher education, is severely limited in scope, not because a wider array of evidence is unavailable but because

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

69

such evidence is inconvenient to include or even considered to be “mere dross” (Packer, 2011, p. 1) next to the numbers, p values, effect sizes, and so on that appear in experimental research reports. I want to reiterate here that quantification itself is not the issue. What qualitative science in general argues for is a different logic, one that understands “that human beings are in the world in the sense not of spatial inclusion but of practical involvement” in forms of life (Packer, 2011, p. 179). As noted, this logic is rooted in a focus on understanding the unity of constituent parts as they operate together as a whole in specific contexts, a position aligned well with Vygotskian psychology. This kind of research may very well involve quantification and statistical procedures, but the goal is to “replace object analysis [i.e., elements] by process analysis [i.e., units]” (Vygotsky, 1978, p. 62). The role of quantification in qualitative L2 science is in describing some parts of a bigger picture; but it is not presumed to be the gold standard when it comes to providing explanations of the constitution of L2 development. Indeed, qualitative science opens up the possibility not only of including alternative forms of evidence for our existing questions about L2 development but, more importantly, it offers new possibilities for the kinds of research questions we can ask. For example, rather than asking what effect a given pedagogical intervention has on a group of learners, we might ask how the pedagogical intervention is experienced and lived through as a form of practical involvement in the world. Learning outcomes would certainly be relevant, as in hypothetic-deductive research, but the difference is that a qualitative inquiry would seek to understand the ways in which learning outcomes (or a lack thereof) are embedded in, and constituted by, the lived experiences of research participants, including accounts of the heterogeneity of outcomes, rather than correlating outcomes with some set of decontextualized and abstracted independent variables (e.g., treatment type, proficiency level, time on task, quantitative individual differences). In what follows, I present a selective review of recent L2 SCT work that exemplifies the principles of qualitative science. I have limited my review to studies published in the ten years preceding the writing of this chapter (2007–2017). This represents a surge of SCT scholarship inspired, at least in part, by Lantolf and Thorne’s (2006) landmark exegesis of the theory as it has been extended to L2 development scholarship. I have also limited my review to include only studies that include analytic focus on the situated experiences of participants in relation to their development over time. This is not meant to minimize the contribution of many other studies that have been more centrally focused on learning or developmental outcomes (e.g., communicative or test performance). However, since my concern here is on illustrating qualitative unitary analysis, using the concept of perezhivanie, studies that do not include sufficient information

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

70

R é m i A . va n Co m p e r n o l l e

about participant experience cannot be included. Finally, in order to show the potential breadth of qualitative L2 science but at the same time include some depth in the discussion of methods and findings, I selected a single representative study to describe in detail for each of three contexts of research: study abroad, classroom praxis, and teacher education.

3.3.1  Study Abroad Kinginger (2008) reported on the language learning outcomes and individual experiences of a cohort of twenty-four US university students who participated in a semester-long study abroad (SA) programme in France during the spring of 2003. Building on research showing the heterogeneity of SA as a context of language learning, the study had the aim of “‘digging deeper’ into the nature of the study abroad experience and of students’ contributions to language learning abroad” (p. 12). Digging deeper, according to Kinginger, means going beyond the documentation of general characteristics of the SA experience that may be generalized across participants and correlated with learning outcomes: In addition to knowing what students do, for how long, and in what language, it is also helpful to see why they choose to study abroad and, in this context, what language learning means to them. It is useful to scrutinize the accessibility of language learning resources within host communities and the students’ stances toward them. (Kinginger, 2008, p. 12)

A clear indication of a break from hypothetic-deductive logic and a move towards the analysis of units such as perezhivanie is in Kinginger’s statement about the twenty-four students involved in the study. Noting that Vygotskian theory sees human activity as historically, socially, and institutionally situated, she writes, “these participants are not bundles of variables or mosaics of individual differences, but intentional human agents who invest in language competence to varying degrees and for a host of diverse reasons” (Kinginger, 2008, p. 12). Kinginger’s study involved a range of data sources that allowed her to analyse the SA experience in relation to language learning outcomes. Pre-SA and post-SA data for the entire cohort included a standardized language competence test (Test de français international), language awareness interviews, and communicative performances elicited in role-play tasks. In addition, Kinginger traced individual case histories of six students, through the narrative analysis of journal entries and audiorecorded interviews that were collected throughout the SA programme. This design allowed Kinginger to document general characteristics of language learning in SA in a first phase of research (i.e., pre-post comparisons), and then to delve into the sources of variation found in the data in a second phase of research focused on explaining “the individually specific”, as Vygotsky (1986, p. 5) wrote.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

71

Overall, the cohort made significant gains in terms of reading and listening abilities, pragmatics (i.e., awareness and use of second-person pronouns, awareness of leave-taking speech acts), and colloquial French (i.e., informal or everyday vocabulary and expressions not typically taught in a classroom). However, the data also showed a great deal of individual variation in the cohort. Through qualitative analysis of case histories, Kinginger was able to situate learning outcomes within the lived experiences of her participants. For instance, a recurring theme in the students’ experiences was political discourse and actions in France against the US invasion of Iraq, which coincided with their sojourn in France. Hannah, for instance, reported becoming estranged from her host mother after a particular dinner-time incident in which the mother was criticizing US energy and foreign policy (i.e., “war for oil”). She wrote the following in a journal entry from March 2003: She made me lose my appetite. I realise [sic] that everyone has their own opinions, but this was personal. All of a sudden, her husband told her to stop talking about it in front of me because it was really rude. He started defending me saying that, “here we have an american [sic] that is a guest in our house, and she hears criticism of her own country all of the time, do you think she really wants to hear more?” The whole time I was holding back the tears, but it just took a load off of my shoulders because here was a french [sic] person defending exactly what I was thinking! I felt this huge release that someone actually understood what it felt like, and then I started to cry. (Kinginger, 2008, p. 65)

Clearly, this was a salient, emotionally laden experience for Hannah, especially because she felt that she did not have the linguistic resources in French to stand up for herself in the face of perceived criticism. However, it is more than an external context for her SA experience and language learning. Indeed, as Kinginger points out, estrangement from the host mother reduced Hannah’s access to French at home, which shaped her eventual learning outcomes. In addition, the broader socio-political landscape, including Franco-American tensions stemming from the war, compelled Hannah and other students to seek out information in Englishlanguage media through the internet, and to “stick together … [and] keep their own company rather than to strike out on their own” (p. 65). To be sure, not all students in the study isolated themselves away from perceived negative experiences. For example, although at first hesitant about the perception of Americans in France following the invasion of Iraq, Beatrice reportedly established a positive rapport with her host sisters through her telling of stories about her ex-boyfriend, which led to opportunities to learn everyday French vocabulary and pragmatics (p. 70). She also provided evidence of the emotional dimension of language learning and use in a journal entry she recorded about two-and-a-half months

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

72

R é m i A . va n Co m p e r n o l l e

into her sojourn, where she expressed joy and pride in relation to her progress, and even charted out a future for her professional involvement with French (i.e., as a teacher). I just love speaking French. (…) I just really enjoy learning it and being able to hear the progress and the difference on my accent and my rapidity of speech. I’ve pretty much decided that I want to be a French teacher at some point in my life. I know I still have so much progress to make before I do that but I really think that that will be my ultimate career goal in the end. (Kinginger, 2008, p. 70)

The point is that the qualities of Beatrice’s L2 French development, as well as those of the other learners, were embedded in a broader nexus of socio-political discourse, particular emotional experiences, and access to French-language resources controlled by, and negotiated with, native speakers (e.g., host families in study abroad; see also Fernández, 2016; Shively, 2016). Note also, as Coleman (2013) contends, that language learning itself is rarely the process or experience that is most salient to or transformative for participants: Ask any applied linguist confidentially, in the corner of a bar, about their own time abroad as a student, and the emphasis will never be on enhanced TL [target language] lexis and mean length of utterance, but rather on romance, on discovery of self and others, on people and places.  (p. 29)

Thus, understanding L2 development in study abroad requires an understanding of the particular systems of experience that participants live through (i.e., perezhivanie).

3.3.2  Classroom Praxis The relevance of lived experience for L2 development is of course not limited to study abroad contexts. In classrooms, students think, feel, and relate to one another and their teachers in ways that are shaped by their histories and in turn shape their futures, including the qualities of their L2 development. Motives for learning languages (i.e., the goals in which one is investing) can be dynamic and change in response to positive and negative experiences in the classroom and the way in which classroom praxis is perceived by students to enhance or limit L2 resources deemed to be valuable for learning to occur (Lantolf & Pavlenko, 2001). Consequently, a qualitative scientific inquiry into the outcomes of instructed L2 learning requires researchers to examine the ways in which language learning and teaching are integral parts of individuals’ lives, past, present, and future. Van Compernolle and Williams (2012) reported two case studies of intermediate-level US university learners of French who were enrolled in a class in which the teaching of sociolinguistic and pragmatic variation

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

73

had been integrated into the curriculum in the hopes of developing learners’ abilities to vary their level of discourse (e.g., formality, politeness) in socially appropriate ways. Overall results from the study indicated a high degree of individual variation in learning outcomes measuring conventionally appropriate use of sociolinguistic and pragmatic variants in performance tasks. Because the researchers had included metalinguistic tasks and interviews in the study, however, they were able to probe “the individually specific” (Vygotsky, 1986, p. 5) experiences, attitudes, and goals of the learners and how these dimensions of learners’ engagement with pedagogy shaped the observed outcomes. The two young women focused upon in the report, Casey and Melanie, represented two dramatically different orientations to the learning of sociolinguistic and pragmatic variation. Consequently, rather than posing the question of whether such variation is teachable and learnable in the classroom, based on a standard or universal outcome measure, van Compernolle and Williams aimed to account for the ways in which learners might appropriate, negotiate, modify, resist, or reject what is taught in relation to their histories as language learners and the imagined role of the L2 in their futures. Casey, for instance, was described by van Compernolle and Williams (2012) as a bit of a loner (no negative connotation intended), who had begun studying French in junior high school after becoming dissatisfied with her Spanish teacher. She oriented to the learning of French as an individual academic pursuit, and her principal goal was to be able to access literary works rather than to engage in social interaction with other French speakers.5 Casey had no plans for using French in the future beyond her individual interests (e.g., reading books). During her time in class, she was grade-focused: she earned As on virtually all coursework, including tasks centred on sociolinguistics and pragmatics. On end-of-term metalinguistic tasks and in interviews, she demonstrated a good understanding of how variation works and the possible social meanings of target forms (e.g., address pronouns). However, in her performances, she opted to maintain a formal stance: she used virtually no informal or everyday discourse features that had been taught in the class, even in tasks (e.g., role plays) where the use of more formal language would typically be considered socio-contextually inappropriate. What was gleaned from interviews with Casey was that while she understood how variation worked and what it could mean, she simply preferred the more formal variants because, on one hand, she was more used to them and, on the other, they fit better with her own history as a language learner and goals for the future (e.g., reading French literature).  Van Compernolle and Williams (2012) note that Casey only mentioned using French with other people in the

5

context of practising her French to become more grammatically correct. In other words, the social aspect of communication in French was about language learning rather than developing social relationships.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

74

R é m i A . va n Co m p e r n o l l e

Melanie, by contrast, oriented to language learning as a social-relational endeavour and imagined herself as a developing multilingual. She was an early bilingual (born in Mexico, and Spanish and English were spoken by her parents), and she was studying Japanese in addition to French at the time of the study. Because her father had been in the military, she had lived in Australia, the UK, and Germany before moving to the US. These experiences resulted in Melanie having a keen sense for the relationship between language use and people in a social context. She saw language learning as a “key to another door” and was interested exclusively in learning “the way people do things in real life” (p. 243). Van Compernolle and Williams argued that Melanie’s previous experiences in intercultural and diverse linguistic environments led her to appropriate the new, more informal L2 resources taught in the class at the expense of the more formal textbook-style forms she had previously learned: as the term progressed, Melanie began to use the informal variants across most communicative contexts, including in an interview with her teacher. While this could be seen as overgeneralizing the appropriateness of the new forms, Melanie’s sociolinguistic and pragmatic behaviour aligned with her stated goals for learning French, which included making French-speaking friends, travelling abroad and conversing with everyday people, and so on. In other words, she was less concerned with being grammatically correct, and more concerned with using French in a way that would help her develop intimate personal relationships. What van Compernolle and Williams demonstrated was that L2 development outcomes were only properly understood by situating them within the lived experiences and imagined futures of the learners under study, not by measuring the frequency, accuracy, or appropriateness of language form use, metalinguistic knowledge, and so on against a predetermined normative baseline (e.g., native speaker conventions). This is because L2 learning is a form of life that connects with, shapes, and is shaped by what it means to be a person, one’s sense of self, and the identities one performs across the lifespan (Blommaert & Varnis, 2013). It bears pointing out that van Compernolle and Williams also included a psycholinguistic explanation for the performance data, namely, following variable competence models introduced in the 1980s (e.g., Tarone, 1988, 2007) and later modified and expanded (e.g., Adamson, 2009; Fasold & Preston, 2007), they explained their findings in relation to attention paid to speech. They posited that conscious attention to form underpinned both Casey’s and Melanie’s development. The difference was that because the two women had different goals, their motives for attending to the new forms introduced in the course were different. Casey was comfortable continuing to use more standard, formal French, about which she did not need to think much, since this was the register that fit her future goals (she only

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

75

attended to her speech and use of informal features when she needed to do so for a graded task). By contrast, Melanie was consciously attempting to use the new informal forms as much as she could in hopes that they would become more automatic in her speech. Thus, although on the surface outcomes were very different, the underlying explanatory principals related to motives, identities, learner histories, and conscious attention were the same for both participants. The concept of perezhivanie is apt here because, as Mok (2015) contends, it is in lived experience that the social environment of development connects with lower, biologically endowed psychological processes, such as attention. These findings have since been corroborated and expanded in a series of related studies involving different learners in different instructional contexts (van Compernolle, 2014, 2016; van Compernolle & Henery, 2014; van Compernolle, Gomez-Laich, & Weber, 2016).

3.3.3  Teacher Education The development of teachers, particular during their early formation in educational settings, is another important domain of L2 qualitative science. Although the focus of such work is not typically on learning processes and outcomes of language learners, teacher cognition, development, and practice determine the qualities of language instruction that learners encounter in the classroom. Donato and Davin (in press) make an important contribution to this line of inquiry by tracing teacher development in relation to history-in-person processes, defined by Holland et  al. (1998) as “the sediment from past experiences upon which one improvises, using the cultural resources available, in response to the subject positions afforded one in the present” (p. 18). This perspective holds that an individual’s history (i.e., the diverse perezhivanie that account for one’s development; my interpretation) is dialectically united with one’s way of interacting with the world in the present as one looks towards the future: history shapes, but it is also shaped by, the present, inasmuch as it is remembered and made relevant by one’s present circumstances and future goals. Importantly, the concept of history-in-person connects micro- and macro-phenomena ranging from the moment-to-moment discursive practices one engages in day to day to the larger societal, political, and institutional affordances and constraints within which one operates. Donato and Davin present two case studies of preservice teachers in a certification programme at a Midwestern US university: Brian, a twentyeight-year-old master’s student seeking his initial French teaching certification, and Sue, a nineteen-year-old undergraduate student seeking her certification for teaching Spanish. The two were selected in order to illustrate two different kinds of history-in-person processes. As the authors note,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

76

R é m i A . va n Co m p e r n o l l e

While the case of Brian illustrates how history-in-person may influence present discursive practice [i.e., teaching in the classroom], the case of Sue offers a paradigm example of how both history-in-person and history in institutionalized struggles can result in local contentious practice. (p. 7)

“Thought-based” and “practice-based” data were collected in order to examine teacher beliefs and histories (e.g., through written reflections, autobiographical narratives) in addition to what the two participants did in the classroom (e.g., through field notes from classroom observation, analysis of transcripts of classroom interaction). In the case of Brian, thought-based findings included expressions of “frustration and humiliation while learning French in France” (p. 7) after six years of study in the US. For example, he reported receiving 0 out of 20 points on an assignment because he had made twenty mistakes, and that, in his words, “[teachers] humiliate students … they call them dumb and stuff” (as quoted by Donato and Davin, in press, p. 7). Brian’s experience in France challenged his perceived identity as a proficient user of French, especially in one case where he was unable to distinguish between words on a dictation exercise. Because so much of his negative experience in France had focused on pronunciation, he developed an orientation to language learning and teaching that emphasized the teaching of correct pronunciation and speaking before introducing the written language, but was based on “naïve beliefs about phonological systems” (p. 8). As Donato and Davin note, in Brian’s view, “a good deal of the teacher’s instructional responsibilities was reductively defined as producing and hearing French sounds” (p. 8). His beliefs about what constituted good teaching were borne out of the practice-based data. Analysis of classroom interactions revealed that he created a safe, comforting environment in order to avoid potentially embarrassing or humiliating students, as had been his experience in France, and that the main focus of his teaching was on perceived correct pronunciation. For example, in a lesson on irregular subjunctives, Brian provided feedback on and explicit evaluation of the pronunciation of target forms; his stated goal, as revealed in a written lesson reflection, was “to say [the verb] as many times as possible to kind of stick it in there” (p. 8). It is important to point out that, in contrast to Sue’s case, described below, Brian faced few institutional challenges or constraints: the cooperating teacher allowed him to develop and carry out his lessons with relative independence, and since the learners were serious and dedicated advanced placement students, he faced few classroom disruptions. Together, these circumstances afforded Brian the freedom to enact his history-in-person without contention. By contrast, Sue faced considerable practical and institutional challenges that forced her to negotiate the relationship between her ­history-in-person and the circumstances in which she found herself teaching. Prior to the

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

77

beginning of her student teaching experience, Sue wrote that “a foreign language is learned primarily through speaking and listening” (quoted by Donato and Davin, in press, p. 12). She also claimed to adhere to one interpretation of communicative language teaching, focusing on contextualized language use (nearly) exclusive target language use in the classroom, and an emphasis on spoken language. Donato and Davin traced Sue’s beliefs about language learning and teaching to her high school experience as a language learner. Sue’s Spanish II and IV teacher spoke exclusively in Spanish and encouraged her students to do the same, using “enthusiasm and dramatic mannerisms” (p. 12) to support learning of new vocabulary. This was an overwhelmingly positive experience for Sue. It bears highlighting that Sue actually identified an instance of L2 learning in a written reflection: In the beginning of my Spanish II class, one student said something in English and my teacher freaked out. She said, “Pones un cuchillo en mi corazón. Hay un fuente de sangre! ¿Por qué?” [“You put a knife in my heart. There is a fountain of blood. Why?”] and she acted it out. It was hilarious and we didn’t speak English again. At that point, I did not know how to say knife or blood, but after that, I will never forget.  (quoted by Donato and Davin, in press, p. 12).

Note the emotional vocabulary (e.g., freaked out, hilarious) that shades this memory, as well as Sue’s claim that through this experience (i.e., perezhivanie) she learned the words cuchillo “knife” and corazón “heart”. Sue’s Spanish I and III teacher, however, used English as a mode of instruction, focusing on reading, writing, and listening rather than spoken communicative interaction. Sue reported this as a generally negative experience. As Donato and Davin argue, this contrast between Sue’s two teachers, and her experiences in their classes, formed the basis of her history-in-person process vis-à-vis her emerging philosophy of teaching. As noted, however, Sue faced challenges in enacting her history-in-­ person in the context of her student teaching. In one Spanish III class, which included about 50% heritage speakers, Sue struggled to maintain 100% target language use, despite her attempts to emulate the practices of her favourite teacher. In fact, and somewhat ironically, she was forced to accept English responses from highly proficient heritage speakers who, according to researcher field notes, frequently chatted in Spanish to each other about topics unrelated to the lesson. Sue therefore found herself having to (re)negotiate her beliefs about teaching in light of the “conflict between two institutionalized educational settings: what she had learned in her methods course and knew to be empirically-based practice (use of the L2) and what her cooperating teacher was asking her to do (translate to English)” (Donato and Davin, in press, p. 16). The authors also reveal that, at the end of the term, Sue failed to achieve the requisite level of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

78

R é m i A . va n Co m p e r n o l l e

proficiency, as measured by ACTFL’s Oral Proficiency Interview, for teaching certification. In hindsight, Sue’s cooperating teacher opined that her “inadequate level of Spanish proficiency lowered her credibility with the heritage language learners and explained their use of English with Sue” (Donato and Davin, in press, p. 16). Sue’s development as a teacher of Spanish was therefore intertwined with her history as a language learner, including both her fond memories of her favourite high school teacher but also her failure to attain an adequate proficiency level, as well as external, institutional pressures. Donato and Davin’s (in press) study reveals the deep interconnections between cognition, learning, socio-emotional experience, and development—that is, perezhivanie—in the context of teacher education. Like the van Compernolle and Williams (2012) study discussed above, the individuals’ pasts were brought to bear on their engagement with the world in the present, including what they were learning to do and in relation to their imagined futures; but the present and future also shaped the qualities of how Brian and Sue remembered and re-experienced their pasts. In order to understand the outcomes of their development, Donato and Davin had to engage in the analysis of what was “individually specific” (Vygotsky, 1986, p. 5) to Brian and Sue, a type of qualitative inquiry that aimed to identify points of congruence and incongruence between thoughts and beliefs with discursive practice in the classroom and situated within the social-material circumstances of their past and ongoing development as people.

3.4 Conclusion This chapter has made the case for a qualitative science of L2 development, drawing on Vygotskian sociocultural psychology, particularly the concept of perezhivanie “lived experience”, and Packer’s (2011) emphasis on studying the constitution of forms of life. I have argued that such an approach adopts a different logic from hypothetico-deductive research, one that takes as axiomatic the idea that phenomena cannot be understood by reducing them into separable elements but only by studying them as unitary wholes. In the exemplar studies presented above, this was borne out in the focus on participant histories as they became relevant to their ongoing development. The past informed the present, but the present, as well as future goals, desires, emotions, and so on shaped how aspects of individuals’ pasts became relevant from moment to moment. In addition, broader social, political, and institutional affordances and constraints shaped the way individuals were able to enact, modify, or even abandon their goals. What we gain from such research is an understanding

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

79

of that messy, fuzzy space between more stable states of development— the ZPD—where the real action takes place during development. To be sure, researching the constitution of any form of life does not lead to the presumption of generalizability. Instead, the veracity of qualitative science is in its ability to particularize the dynamic relations of the components of situated phenomena. Since dynamic relations are constituted in lived experience, research must aim to account for the individually specific relevance of components and their qualitative differences from one person to another. In other words, there is no presumuption that any single component will play the same developmental role for different learners. As Packer (2011) writes, “A ‘true account’ is one that provides a way of seeing what is relevant to our current situation, our problem” (p. 392). It should be noted that methods create objects of study. This is because epistemology constructs ontology as much as ontological assumptions drive methodological choices. This is just as true for qualitative science as it is for hypothetico-deductive research. In other words, we make different assumptions about the nature of reality (ontology), which leads to different methods for investigating that reality and constructing the entities we study (epistemology) that are in turn analysed to confirm, modify, or reject our ontological assumptions. But we also need to recognize that ontology is our assumption about and understanding of reality, not objective reality itself, something Vygotsky (1987) forcefully argued. As Backhurst (2007) notes, “Although Vygotsky appreciated that our research methods, language, and other conceptual tools influence our conception of the objects of our inquiries, he never lost confidence in the idea that those objects are independent of our forms of understanding them” (p. 67). In other words, we certainly construct knowledge (i.e., how we understand reality) that is shaped by our methods, but we do not construct reality itself. A qualitative science of L2 development makes the ontological assumption that psycholinguistic phenomena involved in L2 learning, use, development, etc. are embedded in the lived experiences of individually specific people, and this is not simply an assertion of an “alternative perspective” on L2. It is in fact an argument about which way of knowing, which ontology, “most accurately reflects [the] nature” of material, objective reality (Ratner, 2006, p. 232). Thus, the knowledge we construct through research may be wrong, but it can be challenged, revised, or rejected if it is found to inaccurately reflect reality. Qualitative science challenges the findings of other forms of researching L2 development that engage in elemental analysis because they tend to reveal only the most general characteristics of phenomena, to borrow Vygotsky’s (1986) turn of phrase, which do not provide insight into the reality of individually specific instances of a given phenomenon. Qualitative science is capable of providing a true account of and a way of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

80

R é m i A . va n Co m p e r n o l l e

seeing what is relevant to individuals’ L2 development experiences that goes unnoticed in hypothetico-deductive research relying on quantitative experimentalist methods. Here, it is useful to reiterate the point made earlier in this chapter that qualitative science is not just about different approaches to data collection and analysis; rather, it opens up different kinds of questions we can ask, lines of inquiry that seek to investigate holistic units of forms of life like the state of “living through” learning an additional language (perezhivanie). I would like to echo Mok’s (2015) suggestion that an exciting avenue for future research lies in reconceptualizing standard L2 acquisition concepts such as attention, memory, and noticing, implicit and explicit knowledge, and so on, within a qualitative scientific paradigm that focuses on perezhivanie as the unit of analysis. This would involve flipping the relationship between quantitative and qualitative research with regard to hypothesis generating and testing: namely, quantification in a descriptive phase of research, especially relevant for identifying important elements of a phenomenon, can generate hypotheses about the parts that form a unit to be tested, confirmed, modified, or falsified through qualitative investigations of their dynamic relations. Indeed, we might see this as a paradigm shift in L2 research, one that implies a rejection of hypothetico-­ deductive logic in favour of research designs that, instead of abstracting and objectifying our participants, their behaviours, and their characteristics in the belief that such an approach makes our field more scientific, are actually capable of accounting for individual variability and change across time and space by situating developmental phenomena in the lives—the perezhivanie—of real people.

References Adamson, H. D. (2009). Interlanguage variation in theoretical and pedagogical perspective. New York: Routledge. Atkinson, D. (ed.). (2011). Alternative approaches to second language acquisition. New York: Routledge. Backhurst, D. (2007). Vygotsky’s demons. In M. Cole, H. Daniels, & J. Wertsch (eds.), The Cambridge companion to Vygotsky (pp. 50–76). Cambridge: Cambridge University Press. Blommaert, J., & Varis, P. (2013). Enough is enough: The heuristics of authenticity in superdiversity. In J. Duarte & I. Gogolin (eds.), Linguistic superdiversity in urban areas: Research approaches (pp. 143–160). Amsterdam: John Benjamins. Chaiklin, S. (2002). A developmental teaching approach to schooling. In G. Wells & G. Claxton (eds.), Learning for life in the 21st century: Sociocultural perspectives on the future of education (pp. 167–180). London: Blackwell.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

81

Coleman, J. A. (2013). Researching whole people and whole lives. In C. Kinginger (ed.), Social and cultural aspects of language learning in study abroad (pp. 17–44). Amsterdam: John Benjamins. Davydov, V. V. (1995). The influence of L. S. Vygotsky on education theory, research, and practice (trans. S. T. Kerr). Educational Researcher, 24(3), 12–21. Donato, R., & Davin, K. (in press). The genesis of classroom discursive practices as history-in-person processes. Language Teaching Research. Fasold, R., & Preston, D. (2007). The psycholinguistic unity of inherent variability: Old Occam whips out his razor. In R. Bayley & C. Lucas (eds.), Sociolinguistic variation: Theory, methods, and applications (pp. 45–69). Cambridge: Cambridge University Press. Fernández, J. (2016). Problematizing authentic out-of-class interactions in study abroad. In R. A. van Compernolle & J. McGregor (eds.), Authenticity, language, and interaction in second language contexts (pp. 131–150). Bristol: Multilingual Matters. Frawley, W., & Lantolf, J. P. (1985). Second language discourse: A Vygotskian perspective. Applied Linguistics, 6, 19–44. González Rey, F. L., & Martínez, A. M. (2016). Perezhivanie: Advancing on its implications for the cultural-historical approach. International Research in Early Childhood Education, 7, 142–160. Holland, D., Lachicotte Jr., W., Skinner, D., & Cain, C. (1998). Identity and agency in cultural worlds. Cambridge, MA: Harvard University Press. Kinginger, C. (2008). Language learning in study abroad: Case studies of Americans in France. The Modern Language Journal, 92(S1), 1–124. Kuhn, T. S. (1962). The structure of scientific revolutions. Chicago, IL: University of Chicago Press. Lantolf, J. P., & Frawley, W. (1984). Second language performance and Vygotskyan psycholinguistics: Implications for L2 instruction. In A. Manning, P. Martin, & K. McCalla (eds.), The tenth LACUS forum 1983 (pp. 425–440). Columbia, SC: Hornbeam Press. Lantolf, J. P., & Pavlenko, A. (2001). (S)econd (L)anguage (A)ctivity theory: Understanding learners as people. In M. Breen (ed.), Learner contributions to language learning: New directions in research (pp. 141–158). London: Pearson. Lantolf, J. P., & Poehner, M. E. (2014). Sociocultural theory and the pedagogical imperative in L2 education. New York: Routledge. Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural theory and the genesis of second language development. Oxford: Oxford University Press. Leitch, D. G. (2011). Vygotsky, consciousness, and the German psycholinguistic tradition. Mind, Culture, and Activity, 18, 305–318. Levinson, S. C. (2012). The original sin of cognitive science. Topics in Cognitive Science, 4, 396–403. Mok, N. (2015). Toward an understanding of perezhivanie for sociocultural SLA research. Language and Sociocultural Theory, 2, 139–159.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

82

R é m i A . va n Co m p e r n o l l e

Norris, J. M., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and quantitative meta-analysis. Language Learning, 50, 417–528. Ollman, B. (2003). Dance of the dialectic: Steps in Marx’s method. Urbana, IL: University of Illinois Press. Packer, M. (2011). The science of qualitative research. Cambridge, England: Cambridge University Press. Popper, K. (1959). The logic of scientific discovery. New York: Routledge. Ratner, C. (2006). Cultural psychology: A perspective on psychological functioning and social reform. New York: Psychology Press. Shively, R. L. (2016). An activity theoretical approach to social interaction during study abroad. L2 Journal, 8, 51–75. Smagorinsky, P. (2011). Vygotsky’s stage theory: The psychology of art and the actor under the direction of perezhivanie. Mind, Culture, and Activity, 18, 319–341. Swain, M., & Deters, P. (2007). “New” mainstream SLA theory: Expanded and enriched. The Modern Language Journal, 91(S1), 820–836. Smedslund, J. (1979). Between the analytic and the arbitrary: A case study of psychological research. Scandinavian Journal of Psychology, 20, 129–140. Smedslund, J. (1997). The structure of psychological common sense. Mahwah, NJ. Lawrence Erlbaum. Taguchi, N. (2015). Instructed pragmatics at a glance: Where instructional studies were, are, and should be going in interlanguage pragmatics. Language Teaching, 48, 1–5. Tarone, E. (1988). Variation in interlanguage. London: Edward Arnold. Tarone, E. (2007). Sociolinguistic approaches to second language acquisition research—1997–2007. The Modern Language Journal, 91(S1), 837–848. Toomela, A., & Valsiner, J. (eds.) (2010). Methodological thinking in psychology: 60 years gone astray? Charlotte, NC. Information Age Publishing. Valsiner, J., & Sato, T. (2006). Whom to study in cultural psychology: From random to historically structured sampling. In J. Straub, C. Kolbl, D. Weidemann, & B. Zielke (eds.), Pursuit of meaning. Theoretical and methodological advances in cultural and cross-cultural psychology. Bielefeld: Transcript. Valsiner, J., & van der Veer, R. (2014). Encountering the border: Vygotsky’s zona blizhaishego razvitia and its implications for theories of development. In A. Yasnitsky, R. van der Veer, & M. Ferrari (eds.), The Cambridge handbook of cultural-historical psychology (pp. 148–173). Cambridge: Cambridge University Press. van Compernolle, R. A. (2014). Sociocultural theory and L2 instructional pragmatics. Bristol: Multilingual Matters. van Compernolle, R. A. (2016). Sociolinguistic authenticity and classroom L2 learners: Production, perception, and metapragmatics. In R. A. van Compernolle & J. McGregor (eds.), Authenticity, language, and interaction in second language contexts (pp. 61–81). Bristol: Multilingual Matters.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

The Qualitative Science of Vygotskian Sociocultural Psychology

83

van Compernolle, R. A., Gomez-Laich, M. P., & Weber, A. (2016). Teaching L2 Spanish sociopragmatics through concepts: A classroom-based study. The Modern Language Journal, 100(1), 341–361. van Compernolle, R. A., & Henery, A. (2014). Instructed concept appropriation and L2 pragmatic development in the classroom. Language Learning, 64, 549–578. van Compernolle, R. A., & Williams, L. (2012). Reconceptualizing sociolinguistic competence as mediated action: Identity, meaning-making, agency. The Modern Language Journal, 96, 234–250. Vygotsky, L. S. (1971). The psychology of art. Boston, MA: MIT Press. Vygotsky, L. S. (1978) Mind in society: The development of higher mental processes. Cambridge, MA: Harvard University Press. Vygotsky, L. S. (1986). Thought and language. Cambridge, MA: MIT Press. Vygotsky, L. S. (1987). The historical meaning of the crisis in psychology: A methodological investigation. New York: Plenum Press. Retrieved from https:// www.marxists.org/archive/vygotsky/works/crisis/index.htm. Vygotsky, L. S. (1994). The problem of the environment. In R. van der Veer and J. Valsiner (eds.), The Vygotsky reader (pp. 338–354). Oxford: Blackwell. Zinchenko, V. P. (2002). From classical to organic psychology. Journal of Russian and East European Psychology, 39, 32–77. Zinchenko, V. P. (2009). Consciousness as the subject matter and task of psychology. Journal of Russian and East European Psychology, 47, 44–75.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:25:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.004

4 Theoretical Frameworks in L2 Acquisition John Truscott and Michael Sharwood Smith

4.1 Introduction It is important to distinguish between a theory and the hypotheses that are derived from it, on the one hand, and a theoretical framework on the other. The crucial criterion for judging a theory, in addition to its explanatory value, is its ability to make useful predictions. A framework functions at a higher level of abstraction. The level can vary between a narrow, “local” framework restricted to a well-specified research domain and one that has a wider coverage, ending up with frameworks that also facilitate interdisciplinary research projects. What all frameworks have in common is that they contain some commitments that cannot be significantly altered: this is the framework’s metatheory. Then there are the areas where variation is perfectly possible, allowing researchers to carry out many alternative applications of the metatheory that defines that particular framework. In general, researchers in second language acquisition research, and indeed in other related research domains, are both used to and comfortable with working at a less abstract level, i.e., with theories and hypotheses, precisely because experimental predictions can be easily derived from them and form the basis of a continuous series of research projects. These practical benefits are accentuated by the demands that institutional academic life often imposes on individual investigators for financial reasons and because of the pressure to publish regularly. These factors do not make background assumptions and an overall model of the mind that broad frameworks provide any less important, but they do encourage an overreliance on unquestioned background assumptions. The general theory versus framework distinction needs some qualification, because many researchers do work with and benefit from

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

85

frameworks. This means that relative breadth is also important in this discussion. As just mentioned, frameworks can cover more or less territory. SLA investigators working within a generative linguistic framework, for example, can point to the considerable number of studies that this particular approach to language cognition has produced. It has been the most successful of frameworks in this respect, but there have been others that do not share its basic assumption of a special and unique human language ability distinct from a more general type of cognition. These alternative approaches include the basic behaviouristic principles that defined many people’s thinking in the years that immediately preceded the development of SLA as an independent field of investigation in the 1960s (Lado, 1957). They also include basic principles of cognitive psychology which assumed an anti-behaviourist stance by claiming that the mind was a complex information processing system. It should be noted that behaviourism was a real theory of learning, in other words eminently applicable to the development of L2 ability (Fries, 1945; Lado, 1957; Skinner, 1957). Generative linguistics, on the other hand, which has had a long track record in SLA, does not incorporate a theory of learning as such, or indeed a theory of linguistic processing: it just establishes the bases on which such theories can be developed. It has therefore allowed an analysis of learner systems at particular moments in time from which hypotheses can be formed about what processes led to their creation. As a framework guiding research, this still leaves the door open for theories of L2 learning (acquisition) and theories of online L2 processing to be developed separately: this would be in tune with generative principles, or principles of general cognitive psychology that exclude any notion of a special language faculty, or indeed other sets of principles such as those espoused by behaviourist psychologists. In what follows, we will first consider four examples of relatively narrow frameworks which have served to guide research in SLA in the past. We then consider in more detail two broad frameworks: MOGUL (Sharwood Smith & Truscott, 2014) and the Five Graces Group (2009). As the most relevant aspects of the latter are described in some detail elsewhere in this volume, our description will be less extensive than would otherwise be necessary. After these introductions, we explore the similarities and differences between the two broad frameworks and then draw some implications.

4.2  Big Frameworks, Little Frameworks Various approaches guiding SLA research have marked some interim point in the gradient between the two poles of a broad framework at one end and a single hypothesis at the other. Without any attempt at a proper Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

86

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

history of the field, we will pick out four examples of relatively narrow frameworks whose origins date back to the 1970s and 1980s.

4.2.1  Interlanguage Theory Selinker (1972) was the first to provide, for SLA, what for the time was something that could be called a theoretical framework, something that could and did guide early research in this particular field. It was based on his notion of interlanguage, a concept that had as its immediate forebears Corder’s (1967) transitional competence and Nemser’s (1971) approximative system (see Sharwood Smith, 1994, pp. 22–42). It was Selinker’s particular contribution to assemble within one theoretical perspective the different ideas that emanated from the basic idea of a learner’s mental representation of the target language at any given point in their learning career being treatable as a grammar in its own right rather than simply as a source of error. His 1972 paper proposed different kinds of learner strategies that could explain the systematic properties of an interlanguage system and especially the notion of fossilization, i.e., the cessation of interlanguage development despite repeated exposure and practice in the language. Fossilization was a particular characteristic that distinguished, for Selinker, the learner of a mother tongue and the learner of a second language after a certain age, basing his view on Lenneberg’s (1967) notion of a critical period for first language acquisition after which native-like acquisition would no longer be possible. Lenneberg’s brief speculation that the existence of a critical period might also explain the difficulty older learners have in fully acquiring a second language was in this way adopted as a basic principle by Selinker. It also justified, for him, the creation of a separate theory, more or less along lines familiar from cognitive psychology, to explain the processes and hypotheses (or “strategies”) that guide second language acquisition after puberty (Selinker, 1972; Selinker & Lamendella, 1978).

4.2.2  The Creative Construction Approach Selinker’s assertion of the need for an entirely separate theoretical approach to adult and adolescent language acquisition as opposed to child language acquisition soon provoked a counter-reaction. Some researchers denied the validity of appropriating Lenneberg’s critical period to definitely exclude complete1 second language (L2) acquisition as well. Rather they, notably Dulay and Burt (1973), claimed that once an L1 had been acquired, the acquisition of further languages by adults and adolescents was in principle perfectly possible: the same grammar building instinct of young children, which they referred to as triggering creative construction 1

 We will bypass any discussion of what “full”, “complete”, or “native-like” might imply or what they actually mean.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

87

of the target grammar, remained intact for later re-use, and hence the older learner did not and should not have to fall back on general cognitive learning strategies. It was, in fact, Krashen (1982, 1985) who elaborated on these creative construction claims, together with some of his own, to create an alternative theoretical framework for researchers to follow, investigating the presumed innately determined sequence of stages that learners go through on their way, in principle at least, to full, native-like mastery of their target language. As things worked out it was investigations using this research framework, rather than Selinker’s interlanguage model, that for a time dominated the SLA literature and provided the basis for controlled experimentation built on the techniques and ideas used by researchers into L1 acquisition. Krashen’s theoretical approach was centred on five hypotheses and provided researchers both with a mental model of how L2 acquisition proceeds and guidelines for how a typical investigation should be set up using specific types of experimental techniques and specific targets. For example, the basic acquisition versus learning distinction meant that any investigation into “acquisition”, in his terms a purely subconscious process, should be conducted without appealing to or eliciting conscious reflections about the L2 grammar. A typical target would be to reveal or provide further confirmation of some hypothesized sequence in which particular grammatical constructions were acquired provided there had been sufficient exposure to the language being acquired. The Krashen approach provided plenty to occupy researchers at the time, and this resulted in a small tradition of experimental investigations that contributed to the field at this early stage in its history. Some of the ideas still resonate today, notably Krashen’s insistence that subconscious and consciously driven grammatical development should be treated as qualitatively different both in terms of the underlying processes and the typical outcomes. This idea is not completely uncontroversial even today, but it already suggested at the time that theoretical models that are designed to explain second language acquisition phenomena have to be grounded in the theoretical debates that take place about cognition and cognitive development in general, as well as those concerning the nature of language cognition.

4.2.3  The Generative Approach It was not long before the latter type of debate began to impact SLA research, leading to the tradition that was based on generative linguistic theory, one which has continued up to the present day. If the theoretical framework provided by the generative approach has not led to much refinement in the more strictly psychological aspects of development, it has substantially refined the quality of the linguistic questions being

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

88

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

posed about the processes and systems underlying learner grammars. It is still a relatively narrow framework; neither its metatheory nor its theoretical linguistic applications cover all aspects of development and performance. Nevertheless, it has provided a platform for a great deal of theoretically interesting studies of learner grammars at different stages of development from the early 1980s onwards (Hawkins, 2005). In one way, it was a development of the interlanguage approach as it focused on learners’ grammars, whereas the creative construction approach was target-oriented and concentrated on the points at which native-like mastery of given constructions was achieved. In another way, it was a continuation of the creative construction approach in that it focused on the extent to which learner grammars conformed to innate principles specific to language that determined or rather constrained the shape these grammars would take even if the approach did not deal directly with the mechanisms of transition that moved learners on from one stage to another.

4.2.4  Processability Theory A different line of enquiry was inspired in the 1980s by the ZISA project, which looked at migrant workers from various language backgrounds acquiring German in Germany (Meisel, 2013). Pienemann (1998), one of the researchers who worked on ZISA, developed his own research framework in which he provided an L2 processing-oriented theory, which came to be called the Processability Theory, to predict and explain grammatical stages of development as manifested in learner production. Whereas the Burt–Dulay–Krashen approach had no real explanation for the developmental sequences they claimed to show, Pienemann did have an explanation, one which was based on the inherent processing ease or difficulty with which a given grammatical construction could be produced. This particular explanation was a development of an idea that Clahsen (1980) had proposed to explain the sequences manifested in migrant worker data. For Krashen and associates, the very existence of predictable developmental sequences given sufficient naturalistic exposure to the language was already an important finding in itself since it seemed to confirm the continued existence of the same developmental mechanisms that drive child language acquisition and disconfirm Selinker’s views that the types of acquisition were fundamentally different. Exactly why the developmental sequences that they found were as they were remained a mystery that still had to be unravelled. In one sense Pienemann’s theoretical framework, equipped as it was with an underlying explanation based on current notions of processing complexity, was less ambitious. His approach makes no claims about covert grammatical development, that is to say the series of new hypotheses and insights learners might be entertaining

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

89

about the nature of the target grammar, whether they actually show up in performance or not. Pienemann’s explanation, using lexical-functional grammar to define linguistic structures, is restricted to how given types of construction appear in production, in other words in the form of overt, easily measurable phenomena. In a similar way to the Krashen framework, Pienemann’s model provided guidance for how research should be carried out, namely on natural spontaneous performance. It also provided a criterion for when a given structure was deemed to be acquired. Whereas the creative construction approach required that a form should appear in spontaneous production at least eighty per cent of the time, for Pienemann a few appearances, around five, were sufficient to say that a new stage had “emerged” (Pienemann, 1998, p. 146). In this way, both approaches provided a set of hypotheses and criteria for researchers to be able to design experiments and contribute to the advance of our understanding of how L2 acquisition proceeds.

4.2.5 Summary Since we wish to focus on two more recent, broader frameworks, we will not elaborate further on these examples of what we call narrow frameworks. Space precludes a full description of either of our two main examples, but it should be enough to show how they differ and in what sense they are similar.

4.3  The MOGUL Framework (MCF) We will now provide the main features of the Modular Online Growth and Use of Language (MOGUL) framework. This was first introduced in Truscott and Sharwood Smith (2004) and has been the subject of various publications since that time including three books (Sharwood Smith, 2017; Sharwood Smith & Truscott, 2014; Truscott, 2015; also, Truscott & Sharwood Smith, in preparation). It now has a website which describes it more accurately as a particular instantiation of the Modular Cognition Framework (MCF). This is because this framework, in order to explain language within the mind, essentially provides a basic model of the mind as a whole. For present purposes, we shall naturally continue to talk of MOGUL, which focuses on the place of language in the mind. However, by doing so, we will also by implication be talking about MCF.

4.3.1  The Availability of a Big Picture MOGUL, being broad-based, does not replace narrower research models currently in use. Rather those approaches that are compatible with its principles can be incorporated within and benefit from the overarching

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

90

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

explanatory framework. This also goes for the Five Graces approach to be discussed later. Narrow frameworks that are not related to wider explanations will force busy researchers to rely on their background assumptions about the mind. For something like a Big Picture, this often means a combination of personal notions of how the mind works and a small selection of theoretical and experimental work outside their field of interest which appears relevant and which they have found the time to become at least a little acquainted with. These include major issues such as the nature of knowledge (cognition), how memory works, and where conscious processes fit in. One important characteristic of MOGUL (MCF) is that it is based in relevant strands of contemporary research drawn from different areas of cognitive science. In this sense, the service it aims to provide may be described as, to a greater or lesser extent, derivative. Its distinctiveness and its value must be judged by assessing exactly how it integrates these various strands into a coherent and systematic representation of the mind’s organization and how it functions in real time. The sections below will give some account of this.

4.3.2  Modular Cognition As the names suggest, MCF and, hence, MOGUL assume a modular mind. The idea that the mind is composed of different systems that each have their own organizational principles is scarcely controversial in modern cognitive science. Controversy still abounds, though, on what type of modularity describes the mind. This means differences in how many modular systems there are in the mind, how they are internally organized and in what manner they interact and collaborate when dealing with the many tasks the mind has to perform every millisecond, including those specifically to do with language. Figure 4.1 illustrates the architecture of the mind assumed by MOGUL, showing just six of the systems. The figure shows a network of modules, each of which has a basic design, i.e., each composed of a processor and a store, and each of which has a unique function. Visual processing and storage is carried out by the visual system; the visual processor has its unique way of building and managing representations which makes such representations (stored only in the visual store) different from auditory representations or indeed any other type of representation, i.e., visual representations are encoded differently. This should not be surprising for a neuroscientist because brain imaging and other types of technique show that visual processing patterns are distinct from auditory ones even where their pathways route them through the same modular brain systems, such as the amygdala or the thalamus (Sharwood Smith, 2017, p. 23). Since MOGUL espouses the view that some aspects of linguistic processing involve modular systems

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Audition processor

Figure 4.1  Simplified MOGUL architecture.

Hearing

Speech structure

AUDITORY SYSTEM

PHONOLOGICAL SYSTEM

Phonology processor

SYNTACTIC SYSTEM

Affect processor

CONCEPTUAL SYSTEM

Values & Emotion

AFFECTIVE SYSTEM

Syntactic structure

Syntax processor

Meaning

VISUAL SYSTEM

Seeing

DOUBLE ARROWS = INTERFACES BETWEEN STORES

BOXES = SPECIALIZED STORES

OVAL SHAPES = SPECIALIZED PROCESSORS

Conceptual processor

Vision processor

92

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

specifically and uniquely used for language functions, these have to be included in the overall model of the mind. In the case of MOGUL, and following lines proposed by Jackendoff (1987, 2002), the linguistic modules are two in number, namely the phonological system and the syntactic system. Anything to do with meaning (semantics, pragmatics, etc.) is handled by the conceptual system which deals with any kind of meaning, even meanings not easily expressible by the individual in language. In MOGUL, the phonological module takes as its input generic auditory input which it processes to see, putting it figuratively, what active generic sound representations can be further processed as speech structures, i.e., phonological representations2 written in phonological code. More specifically, speech events in the environment trigger activity in the auditory system which processes all sound and encodes it in auditory code, and this provides the input to the phonological system. At this point the mind has made sense of incoming speech both as generic sound and as speech, even though it may not yet be able to associate a syntactic structure with it or indeed a conceptual (meaning) representation. The conceptual system, as just indicated, is responsible for encoding meaning representations, and these can be associated with language-related structures but also most other kinds of structures (sounds, visual images, touch sensations, and so forth). Auditory, phonological, and syntactic structures acquire meanings by association with conceptual representations.

4.3.3  Processing Versus Representation Language learning seen from a MOGUL perspective means both acquiring new knowledge representations and the ability to process utterances in comprehension and production based on the new representations. In other words, a MOGUL interpretation of given language learner data will have something to say about both. A standard way of studying these two basic aspects of cognition, representation, and processing, is to treat them as separate and to think of them as the objects of two distinct subfields of research. To take an example of the first type of research, investigations based on the generative linguistic approach have typically focused on the abstract representational properties of different learner grammars (for an overview, see Rothman & Slabakova, 2018). Generative research has sought to reveal the degree to which learners’ mental grammars have stayed within the limits of possible grammars given the constraint of Universal Grammar. Other questions that have been investigated include the degree to which linguistic properties of a previously acquired language have influenced the new developing L2 grammar. On that basis, hypotheses have been proposed to explain why certain L2 grammatical 2

 The words “structure” and “representation” will be used here interchangeably.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

93

properties cause difficulties for some or all L2 acquirers, sometimes even at very advanced levels of proficiency, and others do not (see White, 2003). By way of contrast, researchers in psycholinguistics have focused on the way learners process L2 utterances in real time. In doing this, they make assumptions about what representations are being processed based on research in the first type of investigation (see Hopp, 2010). As an overarching explanatory framework, MOGUL integrates processing and representational questions so that asking questions about representational properties immediately invokes questions about processing and vice versa. This it does without collapsing the distinction between representation and processing. Compare this with a behaviouristic view where abstract concepts like “symbolic representations” and “knowledge” simply do not make any sense, except perhaps as convenient ways of describing what are actually a set of stimulus-response connections or, in the case of modern connectionism, networks of nodes, none of which individually have any symbolic value (see, for example, Elman, 1993).

4.3.4  Memory, Acquisition, and Activation Briefly, the way the system works is as follows. To solve various tasks it is faced with, the mind (like the brain) recruits, figuratively speaking, a selection of its specialized systems, each of which activates representations from its store or seeks to form new ones to fit any input they are receiving. The outcome is a network of representations activated in parallel. Because some regularly occurring types of task are identical or very similar, the mind will have ready-made representations to deal with them, which can be called schemas, but due to a constantly changing environment even these will need frequent tweaking. Language is a case in point. Some stretches of language are highly frequent in conversation; others have never been encountered but require little more than a ready-made schema to understand or produce. Importantly, representations from different modular systems that have been activated in parallel cannot be merged because they are encoded differently: like a jpg (visual) file and an mp3 (sound) file they are mutually incompatible but can still be activated simultaneously. Representations are associated via an interface (shown as double lines in Figure 4.1). The function of an interface is twofold. Firstly, it forms new associations,3 as when a noun representation is activated at the same time as a phonological representation creating for the first time an association between, say /pen/ and N(oun), an event that may also see an association formed with a conceptual representation, namely one for the meaning of a writing instrument that we know as “pen”. The second function is co-activation so 3

 This association takes the form of a shared index.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

94

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

that on a later encounter with the sound of the word “pen”, these three representations are activated together from their respective stores. MOGUL has adopted a principle called acquisition by processing (APT), already illustrated in the previous paragraph where learning is taking place across modules. What we know as the word “pen” is learned as a result of a set of structures from different modules being activated together. In this way, an association is formed. Note in passing that we are talking of different types of representation and not contentless nodes in an associative network familiar from other theoretical frameworks. Association also takes place within one module in essentially the same way. Different properties, i.e., small-scale representations, have to be associated together in different ways according to, for example, the design of particular languages. In building a noun representation for many languages, you need to associate with other nominal properties the property of grammatical gender, itself composed of smaller features like masculine, feminine, etc. Gender features are not required for nouns in languages like English or Chinese because in these cases nouns have no inherent gender. This means that English noun structure will not work properly where the target language is Spanish, for instance, because necessary gender features are missing. Thus far it seems that we are talking of instant learning, and, in a sense, we are, but there is a big difference between a learned word or larger construction being made available in principle as a result of an association and being readily available in comprehension and production. To explain this, MOGUL adopts a view of memory as a store in which representations are retained at a particular resting level of activation. New associations are formed and later co-activated in working memory (WM). WM is not a separate system as in Baddeley’s (2007, 2012) modularized view of memory: it simply means that structures in a given store are in an activated state (see Cowan, 1999, Cowan et al., 2012). This is conceptualized in MOGUL as a process of leaving a structure’s current resting level and rising upwards to an imaginary top level of the store, i.e., WM. This rising or falling idea is a convenient metaphor for activation. The main point is that newly acquired representations typically start at a low resting level, which means they are relatively hard to access being “far” from the top level. The more frequently they are activated the higher their resting level becomes, making them more accessible. Given that language learners have a great many readily available representations, i.e., ones with a relatively high resting level, in the stores that are needed for language use, attempts to understand or produce utterances in a target language will activate many structures that are stronger than ones that have been recently learned, and, for a time, these more established and stronger representations will frequently outcompete the weaker ones because of their relative inaccessibility.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

95

4.3.5  Learning Versus Losing Another illustration of where MOGUL as a broad framework integrates two separate concepts and areas of investigation into one without eliminating the original distinction is the separate treatment of, on the one hand, acquisition, that is learning new languages, and attrition, often, somewhat misleadingly, thought of as losing or forgetting. Both phenomena are treated as a form of development and subject to precisely the same principles of representation and processing that form the core of the MOGUL approach. When processing L2 utterances, one or another of the mind’s processors forms a new association between two representations, either within a module or across modules, that is acquisition. However, following acquisition by processing theory (APT), a new representation or association will typically have a very low resting level, making it relatively unlikely at first to survive in the inevitable competition with more established rivals. If there is no further stimulation resulting from subsequent processing occasions, it will effectively never see the light of day. Whether this means attrition in the sense of “complete extinction” or just continuing to exist but with permanently very low resting levels is immaterial. It will not figure in overt linguistic performance, so it is indeed as good as “lost” or “forgotten”. The main point here is that some degree of attrition is going on all the time as a learner acquires a new language, making it (attrition) both a very short-term and long-term phenomenon.

4.3.6  Categories of Knowledge “Knowledge” in the sense of all representations in the mind (as opposed to beliefs that are true, not false) can thus be applied in principle to anything in the memory stores. However, it is useful to think of knowledge as having the conceptual system as its hub. Hence knowledge of the visual world is composed of meanings associated with visual representations. Clearly there will be multiple connections, not just this one. For example, much of an individual’s knowledge will be expressible in the form of language. The meaning of an orange will be more than a connection between a visual image and a meaning in the conceptual system. It will be associated with a particular set of interconnected representations involving syntactic structures and phonological structures and, if the subject is literate, other visual representations, namely those representing the way “orange” is written in a given language. A relevant question is how, in MOGUL, we should conceive of implicit knowledge, things we know but of which we are not conscious, and explicit knowledge, things we know consciously and about which we can reflect and talk. In MOGUL, conscious processing is largely a perceptual experience (see Truscott, 2015, for an extended discussion of consciousness in

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

96

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

L2 acquisition). In the case of explicit knowledge, we are talking about conceptual (i.e., meaning) representations whose content is projected into the perceptual system. For this to happen, the conceptual representations must have relatively high resting levels of activation. Many conceptual structures do not have this, and these form part of implicit knowledge. The same is true of syntactic and phonological representations, due to the relatively isolated state of their modules. This is, then, another example of how in this broad framework, talking about representation also means talking about processing.

4.4  The Five Graces Framework The Five Graces Group (2009; henceforth “Graces”) offer a very broad-­ ranging proposal for understanding and studying language and language-related phenomena, notably including second language learning. Taking as their leading idea the view that language is a complex adaptive system (CAS), they incorporate in their proposal psycholinguistics, historical linguistics, language evolution, computer modelling, first language learning, and, of course, second language learning. This breadth, plus the usefulness of the proposal for interpreting research and suggesting new research, qualify it as a broad framework. In presenting this framework, our emphasis will be on its view of second language acquisition.

4.4.1  Complex Adaptive Systems On Holland’s (2006, p. 1) definition, complex adaptive systems are “systems that involve many components that adapt or learn as they interact”. Their dynamic, constantly changing character is a standard feature of CAS descriptions, as is the adaptive character of the change, the latter meaning, in Graces’ description, that past interaction is the basis for current behaviour and that past and current interactions shape future behaviour. Because CAS is a broad and rather loose concept, those who use it have many choices about the specific form of the CAS they are proposing. These include its scope and various features of its nature. The universe is an interconnected system which can potentially be analysed into countless complex adaptive systems, all interacting with one another. The same can be said for the human, social/psychological world. Lines have to be drawn between what is in the target system (or is the system) and what is outside but interacting with it. For the case of language, and more specifically SLA, important issues also exist regarding the nature of language and the nature of language learning. The Graces framework is thus much more than a proposal that “language is a complex adaptive system.” It is a relatively specific instantiation of that proposal.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

97

4.4.2  Usage-Based Grammar Graces present usage-based as one feature of their CAS framework, but it takes on a dominant role in the discussion of language acquisition and might well be seen as a broad framework in itself, particularly in regard to acquisition. Elsewhere it clearly has this character (e.g., Bybee, 2010). The leading idea is that language learning occurs through experience of the learner with language. This view does not follow from or imply that language is a complex adaptive system, but it does fit comfortably with the idea, especially given the added assumption that development of the grammar is not mediated by any specifically linguistic principles (see section 4.6.3 below). Because usagebased approaches are presented in detail elsewhere in this volume, we will not go into great detail in describing them, focusing instead on how they resemble and differ from the MOGUL framework.

4.4.3  The Nature of Language Graces define their CAS (i.e., language) at two levels, the individual speaker and the speech community, treating individuals as interacting agents in the large CAS. Other communities are presumably outside the system but interacting with it. While a strong social emphasis exists, the framework is also very much cognitive, especially in its treatment of (second) language acquisition. This merging of the social and the cognitive is potentially a great strength of the approach, as an adequate understanding of language and SLA requires recognition of both and an adequate reconciliation of them. At the level of the individual, a crucial scope issue for a CAS proposal involves the question of modularity: is language a distinct component of the mind? In the Graces perspective the answer is a clear no: language is so intertwined with general cognitive mechanisms that drawing lines around it would be counterproductive, if not simply impossible. In terms of scope, the interactions of elements of the system with outside factors could be greater than internal interaction, undermining the idea of a system. The approach is often defined by its contrasts with generative approaches, including the absence of a distinct language faculty. It emphasizes, for example, the dynamic nature of language, rejecting any notion of a static grammar or of a distinction between competence and performance. The characteristics of language emerge from the interactions among the agents and are constantly changing as a result of continuing interaction. In terms of the nature of language, perhaps the most important concept for this approach is construction, which Bybee (2010, p. 9) defines as “a direct form–meaning pairing that has sequential structure and may include positions that are fixed as well as positions that are open”. The contrast with rules and principles of traditional generative grammar is again stressed.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

98

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

4.4.4  The Nature of Language Learning In regard to language learning—both first and second—the key idea of the framework is, again, usage-based (e.g., Bybee, 2010; Ellis, 2002, 2008). Language learning occurs through interaction among the agents of the system, i.e., the individual speakers, consistent with the social focus of the framework as a whole. But usage-based approaches are very much cognitive as well, so this is a cognitive approach to learning placed within a framework that as a whole emphasizes the social. From the cognitive perspective, the social setting provides the input for individual perception—the material for the general cognitive mechanisms that produce the grammar of the target language. The first requirement for acquisition is the storage of a large number of categorized instances (exemplars) of language use. This is a largely explicit process, relying on noticing (Schmidt, 1990). The explicit nature of the process suggests a role for instruction of language form: encouraging the noticing of relevant forms, particularly those that might not be noticed otherwise. Stored instances are subject to a variety of processes, the commonly proposed mechanisms including categorization, chunking, generalization, analysis into constituents, and analogy with existing constructions. The process of chunking stored instances has received considerable attention. It involves the development and strengthening of associations between stored items, based on repeated co-occurrence in the input. This process is implicit and statistical, frequency playing a crucial role. Regarding the relation between L1 and L2 learning, usage-based research is interested in both similarities and differences. The mechanisms behind L1 acquisition, being the general mechanisms underlying cognition, are not lost at puberty (or any other point), and so the ability to learn language in a natural manner is in principle still present in L2 acquisition. In this respect, the approach is largely consistent with creative construction and contrasts with Selinker’s narrow framework. But various features of learners and the context of learning result in significant differences in the details of how second language acquisition occurs and in how much success it ultimately achieves. These factors represent an important focus of research within the framework.

4.5  Similarities Between the Frameworks The two broad frameworks we have considered appear quite different, and important differences do exist, as we will show in the following section. But they also have much in common. First, and most fundamentally, neither is a theory, making clear empirical predictions, but rather a framework within which specific

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

99

testable theories and hypotheses can be formulated. Both take ideas and findings from a range of disciplines related to SLA and seek to bring them together in a coherent manner, placing SLA in the context of much more general ideas about the social and psychological world. They are thus broad frameworks. Breadth is in fact a major goal for both. Another common feature of the frameworks is their shared interest in cognitive explanations of language phenomena, particularly in SLA. In the primary sense of the word cognitive, both frameworks are clearly cognitive. They fall under the general heading of cognitive science; they are interested in understanding the human mind as such; they exploit the computer metaphor. Both frameworks view language as firmly embedded in the mind as a whole, with the important implication that it is impossible, difficult, or at least ultimately very unenlightening to explain language systems, and how they develop in the individual and operate during online processing, without seriously considering the role of factors that are not specific to language. But the word cognitive has an additional, more restricted meaning, seen in the phrase “cognitive psychology”, which usually refers not simply to cognitively-oriented psychology but rather to a particular type of cognitive approach, a type that includes the Graces framework but probably excludes MOGUL, due to its emphasis on modularity and innate constraints. The word is often used in another restrictive sense as well, as a contrast with neural. We can speak of explanations being at the neural level or at the cognitive level, and this distinction is important because drawing connections across the levels has to be an ultimate goal for research and theory. Both frameworks operate mainly at the cognitive level, but both are also concerned with neural plausibility—their proposals should at least be compatible with what is known about the brain. In terms of learning, important similarities also emerge. Both take cognitive approaches, in the broad sense of the term. Both treat unconscious (implicit) learning as primary while also granting a significant place to more conscious (explicit) learning. Both stress the importance of perception for learning: the way that sensory input is processed is, to a considerable extent, what is learned. Both treat association of items during processing as a crucial part of learning. Both hold that frequency plays an important role. Both treat constructions as important components of linguistic knowledge. Both stress the importance of experience for development. While the full connotations of usage-based apply specifically to the Graces framework, the core idea behind it is shared: the system grows through use. In all these respects, the two frameworks show clear similarities. Crucially, though, significant differences exist in the specifics, differences we will consider in the following section.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

100

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

A difference that is more apparent than real is the relative concern with social factors in learning. A genuine difference does exist in the emphasis each framework places on the social, but this does not reflect any difference of principle. For MOGUL, context is a crucial part of processing and the representations that result from it (see especially Truscott & Sharwood Smith, in preparation). The concern is with internal context—the individual’s cognitive representations of contextual factors—as it is these rather than the objective nature of the context that shape learning and subsequent use. MOGUL, like Graces, recognizes the importance of the social and can readily accommodate it.

4.6  Genuine Differences Between the Frameworks The two broad frameworks thus resemble one another in various respects. But important differences exist, especially in background assumptions drawn from source fields. The central differences relate to innateness and modularity, particularly as they apply to language. An important note on this discussion, though, is that the differences may not be as dramatic as they appear.

4.6.1  What is Innate? Both frameworks logically must assume innate elements, implicitly or explicitly. What members of a species can do is indisputably an interaction of their genes and the environment in which they develop. This is particularly true of language. A human growing up in isolation does not develop language, regardless of his or her genetic endowment. Chimpanzees (and, a fortiori, all other non-human species) do not acquire language, regardless of the environment in which they grow up. The issue is exactly how this interaction of human genes with human environment results in the development of language. The two frameworks illustrate the two different ways to approach this issue. MOGUL falls within a rationalist tradition hypothesizing that in human evolution the process of natural selection produced a genetic basis for a language faculty distinct from other components (modules) of the mind. This genetically based faculty serves as the basis for development of language in the individual. Processing of input, in particular, is guided by its principles, which provide the basic categories to be used in the analysis and constrain the possible relations among elements of the input. Without such guidance the learner would be faced with an essentially infinite number of possible analyses for any given input and its relation to already stored information. The assumption of innateness thus provides an answer for what is perhaps the most fundamental issue in language

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

101

learning: how do the mechanisms that underlie learning know what to look for? This approach ultimately raises the question of exactly how the evolutionary process took place, which has not been addressed within the framework. In the Graces framework, language evolved not as a distinct entity but rather as an integral part of general cognitive mechanisms. The innate basis for learning thus has to be found in these mechanisms rather than in any specifically linguistic principles. The answer to how they know what to look for and what sorts of associations to make lies in the inherent—i.e., innate—nature of the mechanisms. This of course raises the questions of exactly what their innate components are, how they evolved, and how they yield the kinds of constraints on perception and categorization that usage-based learning requires. Such questions have not, to our knowledge, been a significant area of theoretical development. We will conclude this discussion with some doubts about how different the frameworks really are at the genetic level. If language is inseparable from general cognitive mechanisms, as assumed by Graces, then evolutionary changes in the genes occurred because they facilitated development of a complex of abilities of which language is an integral part. While language on this view is not a physically distinct part of the genes, it is nonetheless part of our genetic endowment. This intermingling of language with other things in the genes might appear to separate this view from the “innatist” view of language, but, in fact, it does not. The hypothesis of a genetically based language faculty does not require a distinct set of language genes. It simply requires that whatever genes exist should be such that the faculty will result from their interactions with one another, with the chemical environment in which they are expressed, and with the social environment in which development occurs. This allows for the possibility (likelihood) that genes responsible for the language faculty are involved in other functions as well. These points have been developed in some detail by Barrett and Kurzban (2006), Carruthers (2006), and especially Marcus (2004). Thus, the seemingly dramatic contrasts between “innatist” and “non-­ innatist” views of language may turn out to be not so much substantive differences as preferred ways of looking at things. The practical significance of different ways of looking at things should not be dismissed, though, as such contrasts can lead to important differences in the way research is carried out and interpreted.

4.6.2  The Modularity of Language The hypothesis of distinct modules, two of which are specifically linguistic, distinguishes the MOGUL framework from Graces. It is to some extent a consequence of assumptions about language and innateness. On the

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

102

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

standard view of modules, incorporated in the MOGUL framework, they are a product of our genetic endowment. To say that language is innate is then to say that linguistic modules are genetically based. Knowledge of language is centred in these modules and their development constitutes the heart of language learning. In the Graces framework, no such modules exist. Modularity also has consequences for the treatment of different knowledge types. While both frameworks hypothesize two distinct types of linguistic knowledge, fundamental differences exist. Graces distinguish implicit and explicit knowledge, while hypothesizing that linguistic ability is a systematic intertwining of the two. For MOGUL, the contrast is between representations in the linguistic modules and conceptual representations of language-related information, in conceptual structures. The explicit–implicit contrast is closely related to but cannot be equated with this distinction, as conceptual representations can be either explicit or implicit, as described above (section 4.3.6). The frameworks also differ regarding the interaction between the knowledge types, a complex issue that we will not go into here. As with the issue of innateness, we can ask to what extent the modular versus non-modular distinction blurs when closely examined. In terms of neural systems, an ideal linguistic module would be one that occupies one particular spot in the brain and serves no function other than language. But a module defined at the cognitive level does not require such ideal neural underpinnings. Physically separate brain regions that are richly interconnected could well act as a cognitive module (for review of evidence on neural localization of language, see Friederici, 2011, 2012). Nor is complete separation of function essential. Some low-level neural or cognitive processes might well be shared by different modules. Thus, the modularity issue is also more subtle than it might seem.

4.6.3 Learning Above, in section 4.5, we noted important similarities between the frameworks regarding the character of learning. Here we will consider the “but” side of that discussion. The central differences are consequences of assumptions about what is innate, including modularity. While both frameworks stress that learning is a consequence of experience with the language, in one of them development is based on innate linguistic knowledge, including an established architecture; in the other the innate base is expressed in general cognitive mechanisms. Both frameworks emphasize unconscious learning, but for Graces, it is a generic implicit learning applied to language phenomena, while for MOGUL, its core is a specifically linguistic process following in-built linguistic principles. Both also give conscious learning a significant place,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

103

but again major differences occur in how it works and how it is related to implicit learning. One significant contrast is the extensive application in usage-based work of the concept of noticing, rejected by MOGUL authors as too confused to be of value (see especially Truscott & Sharwood Smith, 2011). Both frameworks emphasize perception as the foundation of learning, but the specific treatments differ considerably, the use of noticing being one example. Both frameworks treat association of items during processing experience as crucial, but for MOGUL the associations are constrained by specifically linguistic principles within a particular architecture, while in the Graces approach it is a matter of general principles of association. Both treat frequency as an important factor driving learning, but for MOGUL, it is how frequently a given representation is activated (internal frequency), not a count of external events.

4.6.4  Dynamic Versus Stable Systems One difference that is real but easily exaggerated lies in the contrast between dynamic and stable. Consistent with familiar ideas of CAS, Graces emphasize the dynamic. The framework has nothing analogous to the stable cognitive architecture of MOGUL, which provides a basis for everything done in the framework. The presence or absence of an explicit architecture is an important contrast, inevitably influencing the use of a framework. But this should not obscure the fact that language in both approaches is both dynamic and stable. An individual’s linguistic behaviour is subject to change and is characterized by considerable stability and continuity. In the usage-based view, constructions and associations among them are far from ephemeral. It is doubtful, in fact, that the details of linguistic representation change more quickly or frequently in the usage-based approach than in the MOGUL approach. Thus, while differences in this regard are meaningful, the issue is not whether the system is dynamic or stable; it is whether the constant, extensive change that occurs takes place within a relatively stable architecture.

4.7  Further Research The most obvious suggestion for further research is that it be conducted within one of the broad frameworks currently available. Working within a narrow framework with all the convenient guidance that option provides is no substitute; nor is it something that should be given up in favour of a broad framework. The whole idea is that a broad crossdisciplinary

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

104

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

framework extends, supplements, and enriches research, not that it necessarily replaces a narrow domain-defined one. This might perhaps make it sound like an optional extra. If it does so, then that idea should be firmly set aside. Broad frameworks are a real necessity if substantial progress is to be made in an efficient manner, taking advantage of research elsewhere to inform theory-building and the analysis of results, and avoiding reinventing the wheel in each subdiscipline. Apart from that, it should simply make research more interesting if not necessarily easier. In regard to what kinds of research can be done within the frameworks, the answer is largely trivial: essentially everything that is now being done in SLA. This is the nature of a broad framework. We will not attempt a general survey of potential research topics here but instead note just two illustrative examples of significant general issues. One is to what extent and exactly how consciousness is involved in learning. The rigorous scientific study of consciousness has progressed considerably in recent years, providing rich material for SLA work. Broad frameworks these days must necessarily take account of this research and provide a clear perspective on the challenges that have been with us since the early SLA work in this area, notably those involving the dual-knowledge ideas of Krashen (1982, 1985) described in section 4.2.2. As detailed above, both broad frameworks discussed in this chapter do this, and each can therefore provide a basis for further research in the area. Another significant issue is the role of emotion in learning. Research in this area can benefit from a general framework that provides an account of how emotion fits into the cognitive system and how it interacts with other components of the system, based on the extensive relevant literatures in other fields. The clear importance of emotion in language learning and language use makes it an especially significant research topic.

4.8 Conclusion We have devoted considerable space to the two broad frameworks, MOGUL and Graces, because we believe this provides insights into the kinds of choices with which researchers are faced in studying language learning. The two have important similarities, a natural consequence of their shared concern with understanding language learning in its cognitive and social context. Differences revolve around issues of innateness and modular structure. At the most fundamental levels these contrasts may have less substance than commonly believed, but in practice, they make a great deal of difference. In studying language learning, should we take linguistic principles, perhaps those specified in generative grammar, as the foundation for our explanations, or should this role be filled by

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

105

the characteristics of domain-general mechanisms? Should we work with a stable cognitive architecture, based on distinct, functionally specified units, or focus on the dynamic processes involved in language and its development? Whatever the answers to such questions, the value of a broad framework should be emphasized. Assumptions about the nature of language, learning, and the mind are an inevitable part of SLA research. Intelligent development of the field requires that those assumptions be made explicit and critically examined. Perhaps even more importantly, the ultimate goal for a field of research has to be a filled-in version of the kind of broad framework we are discussing. Without it, there can be no genuine understanding.

References Baddeley, A. (2007). Working memory, thought, and action. Oxford: Oxford University Press. Baddeley, A. (2012). Working memory: Theories, models, and controversies. Annual Review of Psychology, 63, 1–29. Barrett, H. C., & Kurzban, R. (2006). Modularity in cognition: Framing the debate. Psychological Review, 113, 628–647. Bybee, J. (2010). Language, usage and cognition. Cambridge: Cambridge University Press. Carruthers, P. (2006). The architecture of the mind: Massive modularity and the flexibility of thought. Oxford: Clarendon. Clahsen, H. (1980). Psycholinguistic aspects of L2 acquisition: Word-order phenomena in foreign workers’ interlanguage. In S. Felix (ed.), Second language development: Trends and issues (pp. 7–79). Tübingen: Narr. Corder, S. P. (1967). The significance of learners’ errors. International Review of Applied Linguistics, 5, 160–170. Cowan N. (1999). An embedded-processes model of working memory. In A. Miyake & P. Shah (eds.), Models of working memory: Mechanisms of active maintenance and executive control (pp. 62–101). Cambridge: Cambridge University Press. Cowan, N., Rouder, J. N., Blume, C. L., & Saults, J. S. (2012). Models of verbal working memory capacity: What does it take to make them work? Psychological Review, 119, 480–499. Dulay, H., & Burt, M. (1973). Should we teach children syntax? Language Learning, 23, 245–258. Ellis, N. C. (2002). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24, 143–188.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

106

J o h n T r u s c o tt a n d M i c h a e l S h a r w o o d S m i t h

Ellis, N. C. (2008). The dynamics of second language emergence: Cycles of language use, language change, and language acquisition. The Modern Language Journal, 92, 232–249. Elman, J. (1993). Learning and development in neural networks: The importance of starting small. Cognition, 48, 71–99. Five Graces Group (2009). Language is a complex adaptive system: Position paper. Language Learning, 59(S1), 1–26. Friederici, A. D. (2011). The brain basis of language processing: From structure to function. Physiological Reviews, 91, 1357–1392. Friederici, A. D. (2012). The cortical language circuit: From auditory perception to sentence comprehension. Trends in Cognitive Sciences, 16, 262–268. Fries, C. (1945). Teaching and learning English as a foreign language. Ann Arbor, MI: University of Michigan Press. Hawkins, R. (2005). Second language acquisition and Universal Grammar (review). Language, 81, 754–757. Holland, J. H. (2006). Studying complex adaptive systems. Journal of Systems Science and Complexity, 19, 1–8. Hopp, H. C. (2010). Ultimate attainment in L2 inflection: Performance similarities between non-native and native speakers. Lingua, 120, 901–931. Jackendoff, R. (1987). Consciousness and the computational mind. Cambridge, MA: MIT Press. Jackendoff, R. (2002). Foundations of language. Oxford: Oxford University Press. Krashen, S D. (1982). Principles and practice in second language acquisition. Oxford: Pergamon. Krashen, S. D. (1985). The Input Hypothesis: Issues and implications. New York: Longman. Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers. Ann Arbor, MI: University of Michigan Press. Lenneberg, E. (1967). Biological foundations of language. New York: Wiley. Marcus, G. (2004). The birth of the mind: How a tiny number of genes creates the complexities of human thought. New York: Basic Books. Meisel, J. M. (2013). The ZISA Project. In P. Robinson (ed.), The encyclopedia of second language acquisition (pp. 708–709). New York: Taylor & Francis/ Routledge. Nemser, W. (1971). Approximative systems of foreign language learners. International Review of Applied Linguistics in Language Teaching (IRAL), 9, 115–124. Pienemann, M. (1998). Language processing and second language development: Processability theory. Amsterdam: John Benjamins. Rothman, J., & Slabakova, R. (2018). The generative approach to SLA and its place in modern second language studies. Studies in Second Language Acquisition, 40, 417–442.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Theoretical Frameworks in L2 Acquisition

107

Schmidt, R. W. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129–158. Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics in Language Teaching (IRAL), 10, 209–231. Selinker, L., & Lamendella, J. (1978). Two perspectives on fossilization in interlanguage learning. Interlanguage Studies Bulletin, 3, 143–191. Sharwood Smith, M. (1994). Second language acquisition: Theoretical foundations. London: Longman. Sharwood Smith, M. (2017). Introducing language and cognition: A map of the mind. Cambridge: Cambridge University Press. Sharwood Smith, M., & Truscott, J. (2014). The multilingual mind: A modular processing perspective. Cambridge: Cambridge University Press. Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts. Truscott, J. (2015). Consciousness and second language learning. Bristol: Multilingual Matters. Truscott, J., & Sharwood Smith, M. (2004). Acquisition by processing: A modular approach to language development. Bilingualism: Language and Cognition, 7, 1–20. Truscott, J., & Sharwood Smith, M. (2011). Input, intake, and consciousness: The quest for a theoretical foundation. Studies in Second Language Acquisition, 33, 497–528. Truscott, J., & Sharwood Smith, M. (in preparation). The internal context of bilingual processing. Amsterdam: Benjamins. White, L. (2003). Fossilization in steady state L2 grammars: Persistent problems with inflectional morphology. Bilingualism: Language and Cognition, 6, 129–143.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 11:21:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.005

5 Qualitative Classroom Methods Peter I. De Costa, Wendy Li, and Hima Rawal

5.1 Introduction While earlier research on second language learning may have distinguished language acquisition from language use (Krashen, 1982), more recent conceptualizations of second language learning have come to refer to the field of second language acquisition (SLA) as second language studies (e.g., Gass, Lee, & Roots, 2007) and second language development (e.g., Larsen-Freeman, 2018). In light of this conceptual shift, the focus has also expanded beyond the individual language learner, with the research lens turning towards the language teacher and the wider social ecology in which the learner is embedded. In tandem with the broadening of the field, there has been a marked increase in the volume of qualitative research published since Lazaraton’s (1995) earlier lament over the paucity of qualitative studies published in applied linguistics journals. While the distribution of qualitative research across international journals remains uneven (Benson et al., 2009), with some journals such as Applied Linguistics, The Modern Language Journal, System, and TESOL Quarterly being more open to the publication of qualitative research than others, the field of SLA has also witnessed a proliferation of handbook chapters on qualitative research (e.g., Benson, 2012; De Costa, Valmori, & Choi, 2017; Harklau, 2011; Lew, Yang, & Harklau, 2018) and books (e.g., Holliday, 2007; Mirhosseini, 2017b; Richards, 2003) that focus exclusively on qualitative research. Importantly, and in line with the qualitative turn in research on second language learning, the classroom has become the centre of attention, as evidenced in books such as Allwright and Bailey’s (1999) Focus on the Language Classroom, Wallace’s (1998) Action Research for Language Teachers, and McKay’s (2006) Researching Second Language Classrooms. In short, as complex and messy as classroom research may be, a qualitative interest in

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

112

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

what goes on in a second or foreign language classroom shows no signs of abating.1 However, as new research methods and methodologies emerge (e.g., Choi & Richards, 2016; De Costa, Rawal, & Zaykovskaya, 2017), it is vitally important that we take stock of what constitutes a contemporary language classroom, unpack what constitutes qualitative research, and explore the underpinnings for this upward trajectory in classroom-based research before examining some well-established methodologies, with accompanying exemplar studies, that have been adopted to investigate language-related classroom dynamics qualitatively. The chapter ends with a brief consideration of how to advance the classroom-based qualitative research (CBQR hereafter) agenda.

5.1.1  (Re)defining Language and the Language Classroom In a recent reconceptualization of the foreign language classroom, Collins and Muñoz (2016) questioned the future of the “bricks and mortar” language classroom in the twenty-first century, given the emergence of technological tools that afford new learning platforms such as Massive Open Online Courses (MOOCs) and the upward trend of self-directed and self-regulated learning. Indeed, the educational landscape has changed because even physical learning sites have expanded to include complementary weekend schools, community centres, religious institutions, and study abroad satellite campuses—sites where learning takes place formally and informally. Such an expanded understanding of what constitutes the language classroom is timely for two reasons. First, it is in line with the Douglas Fir Group’s (2016) reconceptualization of language learning and teaching along ecological lines, that is, one that reminds us of the needs to take into consideration how learning and teaching are shaped by forces that exist at the micro-level (e.g., individual, classroom), meso-level (e.g., school, family), and macro-level (e.g., community, nation) of society.2 Second, and as rightly observed by Harklau (2011), we need to look across a diverse set of populations and not just focus on US-based university English as a second language (ESL) classrooms. Thus, in line with Collins and Muñoz (2016), the Douglas Fir Group (2016), and Harklau (2011), we extend the notion of the language classroom to include the physical classroom and its digital corollary. We also conceive of the classroom as being nested within a school, community, and nation. In addition, we also take on an embodied and semiotic-inflected understanding of language (Atkinson, 2014), that is, one which includes non-linguistic  A second language refers to a language that is learned in the context where it is spoken; whereas a foreign

1

language is learned in the context where it is not generally spoken by the community (VanPatten & Benati, 2015).  To view the multifaceted nature of language learning and teaching see Douglas Fir Group (2016, Figure 1, p. 25) at

2

https://onlinelibrary.wiley.com/doi/full/10.1111/modl.12301.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

113

elements such as prosodic, interactional, and non-verbal resources that are mobilized by learners and teachers in order to enhance the language learning experience.

5.2 Characteristics of Classroom-Based Qualitative Research Building on insights from Benson (2012) and Friedman (2012), we posit that CBQR is: • inductive, descriptive, and interpretive, with aims to provide detailed explanations of a phenomenon such as classroom behaviours. And in the spirit of open inquiry, pre-set coding schemes are eschewed in favour of coding categories that emerge from the data at hand. In addition, there is generally no attempt to control or manipulate variables because the goal is often to describe classroom routines. • examined from multiple perspectives, with several methods adopted to ensure data trustworthiness. For example, when exploring classroom processes, both emic (insider) perspectives, such as those of the teacher and students, and etic (outsider) perspectives, such as those of school authorities and parents, are sought to corroborate findings. • context dependent, in that the micro-level features of a classroom, such as individual student–teacher relationships, are examined in relation to meso-level features such as school-wide language curricula and macro-level national language policy.

5.2.1  Why is CBQR Becoming Increasingly Popular? Several reasons can be attributed to the increasing adoption of qualitative methods and methodologies in contemporary language classroom research. These reasons need to be considered in relation to an increasingly research-friendly social infrastructure available to teachers. Such a supportive system includes positive developments such as (a) greater involvement of teachers in collaborative research projects as teachers are engaged as coinvestigators, from planning stages to the interpretation of results; (b) the increasing availability of resources (e.g., methodology books and special issues of journals as described earlier) and courses that provide methodological guidance; and (c) wider acceptance and recognition that detailed longitudinal small-scale studies can be just as insightful as larger-scale experimental studies (Duff, 2007).

5.2.1.1  Clearer Guidelines As noted, there has been a proliferation of literature on how to conduct qualitative research in classrooms (e.g., Friedman, 2012; Holliday, 2007; Richards, 2003). The book- and chapter-length overviews have been

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

114

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

accompanied by journal articles that provide succinct methodology-­ specific descriptions. The guidelines put out by TESOL Quarterly (Chapelle & Duff, 2003; Mahboob et al. 2016), in particular, have been instrumental in making both emerging and established classroom researchers think about best practices associated with carrying out qualitative research in an ethical manner.

5.2.2  Emergence of Social SLA Theories Given the strong emphasis on the social context of learning, CBQR has also received a boost from a stream of social theories that have been coopted to investigate the complex social dynamics of classrooms (for a greater discussion of alternative approaches to SLA, see Atkinson, 2011) as part of the broader social turn in SLA (Block, 2003; Firth & Wagner, 1997, 1998, 2006). This turn has witnessed the emergence of adjacent constructs such as (a) translanguaging (e.g., García, Ibarra Johnson, & Seltzer, 2016; Li, 2018), as researchers focus on the flexible instructional arrangements and communicative repertoires of multilingual students and teachers; (b) scales (De Costa & Canagarajah, 2016; Maloney & De Costa, 2017); and (c) emotions (De Costa, 2016a; De Costa, Rawal, & Li, 2018), as we attempt to account for the multiple spatio-temporal dimensions of language learning and teaching. In addition, a large body of classroom-based qualitative research is generally set against a backdrop that is characterized by contemporary concerns such as transnationalism (Duff, 2015) and neoliberalism (Li & De Costa, 2017) that intrude upon and shape the language learning and teaching processes. In the interest of space, and in spite of a growing adoption of social theories in SLA, we have identified three widely used social SLA theories—identity, language socialization and sociocultural theory—to which we now turn.

5.2.2.1 Identity Identity has emerged as a key concept in SLA research, and a whole body of classroom-based investigative work over the last two decades (De Costa & Norton, 2017; Li & De Costa, 2018; Norton & De Costa, 2018) has been undertaken through qualitative means. Underpinning much of the post-structurally-oriented identity research has been the central understanding that identity is a site of struggle and that the identities of both learners and teachers are fluid and multiple. As a consequence, a key concern has been to examine how teachers can create opportunities for learners to speak from a position of power in ways that improve their language learning outcomes (Luk, 2017). In line with digital developments, there has also been an expanding interest in the construction and performance of digitally mediated identities (Thorne, Sauro, & Smith, 2015), a line of research whose analytic rigour has been made possible through corresponding developments in multi-modal analysis (see section 5.7 below).

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

115

5.2.2.2  Language Socialization According to Garrett (2017), language socialization researchers seek to understand how interactions shape the developmental trajectories of individuals as they negotiate larger systems of cultural meaning and practice. Generally longitudinal in nature, some CBQR is ethnographic in nature and takes into consideration the sociocultural and historical contexts within which learners are embedded as researchers examine how learner and teacher identities and linguistic practices are reproduced and transformed over the course of time (Mangual Figueroa & Baquedano-López, 2017). Particular attention is paid to social and institutional structures that intertwine with the everyday experiences of learners in the classroom (e.g., Duff, 2014; Zappa-Holman & Duff, 2015). Zappa-Holman and Duff (2015), for example, examine how the study abroad experience of three international Mexican students is inextricably linked to the complex social networks they developed both within and outside a Canadian university classroom.

5.2.2.3  Sociocultural Theory Originating primarily from the work of the cultural psychologist Lev Vygotsky, CBQR inspired by sociocultural theory also views learners as engaging in embodied interactions that are influenced by their broader cultural and interactional histories (Ohta, 2017). While earlier micro-level work looked at how teachers, peers, and learning materials create affordances to scaffold learning (e.g., Gibbons, 2003; Walqui & van Lier, 2010) and attendants’ zones of proximal development, more recent work has examined dynamic assessment (Lantolf & Poehner, 2014), concept-based instruction (van Compernolle, 2014), and the affective dimensions of teaching (Johnson & Golombek, 2016). Van Compernolle (2014), for example, explored the sociopragmatic development of French within a foreign language classroom.

5.2.3  Discourse Analysis in Classrooms The development of CBQR has moved in congruence with the emergence of increasingly sophisticated discourse analytic tools that allow for in-depth analyses of spoken and written texts (Benson et al., 2009) and “the construction of social realities through discourse” (Harklau, 2011, p. 178). In their review of discourse analysis in educational research, Warriner and Anderson (2017) underscore that micro-ethnographic analysis allows researchers to conduct fine-grained analysis of how interaction unfolds at the micro-­ interactional level of classroom events in ways that relate to macro-social educational issues. One study that adopts micro-ethnographic analysis is De Costa (2015a), who examined how the social imaginary of high performing scholarship students ultimately affected their foreign language anxiety. Equally influential in shaping the classroom interaction agenda has been the role of ethnomethodologically-inflected interaction analysis

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

116

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

such as Conversation Analysis (Waring, 2017) that investigates how language, meaning, and learning are co-constructed in and as interaction in classrooms. Conversation Analysis provides effective tools to help us investigate how learners and teachers manage interactive processes in an orderly manner while negotiating social reality and social relationships that contribute to language development (for an overview of interactional approaches to classroom discourse and student learning, see Hellermann & Jakonen, 2017 and Miller, 2018). Crucially, discourse analytic methods in general are often used in conjunction with the social SLA theories in CBQR. Smotrova and Lantolf (2013), for example, paired Conversation Analysis with sociocultural theory in order to investigate speech–gesture synchronization in an EFL class in Ukraine.

5.3  Paradigm Matters As observed by De Costa, Valmori, and Choi (2017; see also King & Mackey, 2016; Lew, Yang, & Harklau, 2018; Mirhosseini, 2017a; Paltridge & Phakiti, 2015), we need to recognize that there are fundamentally different types of knowledge and different ways of knowing. Relatedly, and given the commitment of CBQR researchers to conduct contextually-situated, in-depth, and practically-relevant research, it is vitally important that we consider the distinction between ontology (a researcher’s views on the nature of reality) and epistemology (a researcher’s views on the nature of knowledge and how it can be acquired). To date, much of the SLA classroom-based research can be grouped broadly within two primary paradigms—­postpositivism and postmodernism—which are summarized in Table 5.1. In light of the significance of a researcher’s paradigm (i.e., her epistemology and ontology), there needs to be an alignment between how the classroom-based research is conducted (i.e., methodology) and the theory that guides a research study. Thus, an SLA researcher who conducts an investigation of language learner identity development would probably subscribe to a postmodernist paradigm that views identities as multiple.

Table 5.1.  CBQR paradigms (adapted from De Costa, Valmori, & Choi, 2017, p. 526). SLA Paradigms

Epistemology

Ontology

Postpositivist

Test hypotheses or look for cause– effect relationships in language learning

Seek an objective reality and a single truth

Postmodernist

Try to understand the experiences, abilities, perceptions, and performances of language learners

Seek subjective realities and multiple truths

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

117

Methodologically, such an identity researcher would have a range of options to select from (see sections 5.4.1–5.4.6), resulting in paradigmatic (e.g., postmodernist) or theoretical (e.g., identity) or methodological (e.g., case study, ethnography, conversation analysis, grounded theory, narrative inquiry, action research) alignment. According to Benson et al. (2009), while some applied linguists expli­ citly identify their research as “qualitative”, many more simply use qualitative methods without using this term. This observation is noteworthy because it underlines the need to distinguish qualitative methodologies from qualitative methods. Qualitative methodologies, which are linked to a researcher’s theoretical framework and paradigm, generally comprise several qualitative methods, such as interviews and focus groups (Prior, 2018; Winke, 2017), as well as observations and fieldnotes (Copland, 2018; for an overview of individual qualitative methods and methodologies, see Friedman, 2012; Mirhosseini, 2017b). These methods are often used in conjunction with data sources such as classroom interactions (audio- and/ or video-recorded) and classroom artefacts (e.g., samples of written work). Given the methodology–method distinction, it is highly possible to conduct a postpositivist experimental study (a quantitative methodology) based in a classroom by adding a qualitative method, such as interviews, to investigate a phenomenon (for a comprehensive discussion of mixed methods research in language learning and teaching see Riazi, 2017).

5.4 Six Classroom-Based Qualitative Research Methodologies In this section, we review six qualitative methodologies—conversation analysis; narrative inquiry; case study; action research; ethnography; and grounded theory—that have been used to better understand language learning and teaching processes in second- and foreign-language classrooms. To illustrate each methodology, we also provide an exemplar study.

5.4.1  Conversation Analysis Conversational analysis, or CA as it is commonly referred, originates from the work of ethnomethodologists and examines the social organization of talk-in-interaction (Tsui, 2017). Because the analysis is done from an insider perspective of how participants understand and manage talk with their interlocutors, CA takes on an emic dimension. Through finegrained analyses of verbal and nonverbal interactions in conversations, classroom-based CA work forces us to reconsider some of the fundamental questions related to learning and teaching. These questions address concerns about what needs to be taught, how teaching is conducted, and how learning takes place (Waring, 2017). The ways that learners manage Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

118

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

repairs,3 navigate tasks, and negotiate identities are but some of the many topics that have been studied in CA-oriented CBQR (see Kasper & Wagner, 2011, for a description of a conversation-analytic approach to SLA). It is important to note, as Kasper and Wagner (2011) observe, that CA is both a methodology and a theoretical approach to SLA, hence the term CA-for-SLA. More recent classroom-based CA research has focused on content and language integrated learning (CLIL) classrooms by investigating how teachers and students use language when they negotiate content knowledge (Hellermann & Jakonen, 2017). In line with this growing interest in CLIL classrooms, our exemplar study (Jakonen & Morton, 2015) examines problem-solving sequences during the interactions with peers in a history class. Exemplar Study 1 Jakonen, T., & Morton, T. (2015). Epistemic search sequences in peer interaction in a content-based language classroom. Applied Linguistics, 36(1), 73–94. Research Concerns The study sought to investigate three main aspects of interaction in terms of how students: • display their own and orient to others’ epistemic stance and status in relation to particular knowledge objects (whether “content” or “language” or other matters); • manage rights, responsibilities, and obligations towards what is known or unknown about the knowledge objects on which they focus; and • resolve knowledge gaps by using ordinary conversational resources of sequence organization. Theoretical Framework • A CA-based study of epistemics in content-based instruction. Methods • Video-recordings of history lessons taught through English to students. Participants • Students aged between fourteen and fifteen years at a school in Finland. Results • Students discovered their own knowledge gaps and deployed resources in resolving the gaps. • Their engagement in collaborative tasks enabled them to mobilize their linguistic, embodied, and artefactual resources in order to negotiate what they knew and did not know about the task.  Repair in CA occurs as part of the turn-taking system for talk-in-interaction. It is seen as one of the mechanisms

3

employed by language users to “maintain the course of action in their talk when some source of interactional trouble arises” (Hellermann, 2011, p. 147).

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

119

5.4.2  Narrative Inquiry According to Barkhuizen (2014), SLA studies that use narrative inquiry may be referred in the published literature as life histories, language learning histories, language learning experience, diary studies, memoirs, language biographies, autobiographies, and autoethnography. This methodology, which is generally used to collect data outside the classroom, can illuminate our understanding of the complex learning and teaching dynamics of the classroom. Specifically, Lew, Yang, and Harklau (2018) maintain that narrative inquiry is most commonly used to document language learners’ and teachers’ development, practices, identity, agency, beliefs, emotion, positionality, and motivation. Benson (2018) posits two reasons for the strong uptake of narrative inquiry among qualitative SLA researchers. First, he credits a growing interest in examining language learning in its social contexts and environments. Second, he contends that the ways in which research findings are presented through narrative inquiry make it both accessible and convincing to consumers of research. Given that narrative inquiry draws from a rich intellectual history that sees stories as a fundamental way for people to make sense of the world, there are three broad ways that narrative data can be analysed. They include: (a) content (i.e., thematic) analysis of narratives (e.g., Xu, 2014); (b) discourse analysis of narratives that focuses on how narratives are discursively constructed by the story teller and the listener (e.g., Kasper & Prior, 2015); and (c) narrative analysis, which involves identifying narratives within collected data (e.g., from interviews) and representing findings as narratives in published work. It is on this third approach (i.e., narrative analysis) that our second exemplar study (Barkhuizen, 2017) is based. Barkhuizen thematically analysed tutors’ stories using a short story framework that focused on both content and context. The content aspect of the stories was analysed according to who (the characters), where (the place), and when (the time). Context was analysed in accordance with three scales: story (e.g., emotions and ideas); Story (e.g., assumptions of members of the communities); and STORY (e.g., language ideologies). Exemplar Study 2 Barkhuizen, G. (2017). Investigating language tutor social inclusion identities. The Modern Language Journal, 101(S1), 61–75. Research Concern • This narrative study sought to investigate how the identities of the tutors in the study were co-constructed during their narratives of their experiences of tutoring immigrant and refugee English language learners. Theoretical Framework • An identity framework guided this study. Tutor social inclusion was examined through how they narrativized their tutor identities.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

120

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

Methods • Interviews to elicit short stories (narratives) of their tutoring experiences. Participants • Forty-two English tutors in New Zealand participated in the larger study. The published study, however, focused on the stories of five tutors. Results • The tutors engaged in cognitive, sociocultural, and emotional work when helping their immigrant and refugee background learners achieve English learning and social inclusion goals. • Tutors’ identities were also constructed and reflected in the stories they told about their tutoring social inclusion experiences.

5.4.3  Case Study Case studies, as noted by Casanave (2015) and Duff (2012), are generally constituted by multiple data sources (e.g., interviews, documents, observations) and seek to describe and analyse particular bounded phenomena such as individual learners and teachers or a given classroom or school. The aim is, therefore, to provide in-depth portrayals of the chosen phenomena that are situated in particular contexts. Such a methodological approach provides “an understanding of individuals’ experiences, issues, insights, developmental pathways, or performance within a particular linguistic, social, or educational context” (Duff, 2014, p. 233). Applying this methodological principle to instructed second and foreign language contexts, case studies have been conducted on study abroad (e.g., Zaykovskaya, Rawal, & De Costa, 2017), linguistic minority (e.g., Kanno & Harklau, 2012), heritage language (Manosuthikit & De Costa, 2016), and international (De Costa, Tigchelaar, & Cui, 2016) students as well as novice teachers (e.g., Wolff & De Costa, 2017) in order to explicate how their identities and emotions are transformed over time and mediated by factors within their social contexts. Identity and accent development is the focus of our third exemplar study (Diao, 2017), which traces the language learning experiences of three US case participants studying abroad in China. Exemplar Study 3 Diao, W. (2017). Between the standard and non-standard: Accent and identity among transnational Mandarin speakers studying abroad in China. System, 71, 87–101. Research Questions • How do speakers of transnational Mandarin become aware of the culturally embedded concept of standard Mandarin?

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

121

• How do they negotiate their existing non-standard accent and (re)interpret the meanings associated with the accent while in China? Theoretical Framework • Language Socialization Methods • Diao’s participants were instructed on how to audio record their conversations with their Chinese hosts. • Interviews were conducted and language awareness questionnaires administered during the pre- and post-study abroad period. • Field data were collected via participant observations. • A detailed discourse-orientated approach was used to delve into the participants’ language use during their interactions with their hosts. Participants • The three focal participants went to Shanghai, China for their study abroad programme. Results • Participants’ use or rejection of the nonstandard Mandarin feature, the retroflex or dental merger, was affected by how they constructed their identities when studying abroad in China.

5.4.4  Action Research Given the growing interest in teacher agency (Li & De Costa, 2017) and teacher identity (De Costa & Norton, 2017), it is not surprising that an increasing number of teachers has chosen to undertake action research by investigating issues closely aligned with their own questions or concerns about their daily pedagogical practices (Allwright & Bailey, 1991). As noted by Burns and McPherson (2017), action research is often characterized by a cycle of planning, action, observation, reflection, and further action in order to enact pedagogical change and enhance a practitioner’s understanding of her classroom practices. Recent published work examined teachers’ experience with their documented implementation of new teaching methods and curricula (e.g., Calvert & Sheen, 2015; Lee & Wallace, 2018) and explored the importance of support factors at the macro-, meso- and microlevel (e.g., Burns & Edwards, 2016). Another key mounting concern is how action researchers manage the asymmetrical power relationship between teacher-researcher and their students. In line with these recent developments in action research, our fourth exemplar study (Galloway, 2017) addresses the challenges associated with taking on the dual role of teacher and researcher as she attempts to balance her workload and negotiate ethical issues related to the blurred line between teacher and researcher.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

122

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

Exemplar Study 4 Galloway, N. (2017). Researching your own students: Negotiating the dual practitioner-researcher role. In J. McKinley & H. Rose (eds.), Doing research in applied linguistics: Realities, dilemmas, and solutions (pp. 146–156). New York: Routledge. Research Questions • How do my students use English outside the classroom? • To what extent is the native English-speaking model, being taught by a native English speaker, meeting their needs? • What are their attitudes of the role of English as a global language? Theoretical Framework • Global Englishes (Galloway & Rose, 2015). Methods • A mixed methods approach was adopted: a questionnaire was administered and personal and focus group interviews were conducted. Participants • Third- and fourth-year university students in Japan who were enrolled in two content-based courses that Galloway taught for thirteen weeks. Because of her tenuous role as a teacher-researcher who was conducting research on her own students, Galloway’s extensive discussion of her researcher positionality points to her own key role as a participant in this study. Results • Managing the asymmetrical power relationship between teacher-researcher and the students proved problematic. • “Teacher research is research” that can help us better understand and improve teaching. It can also lead to better learning, and it has the potential to influence theory.

5.4.5 Ethnography A classroom ethnography, according to Bloome (2012), seeks to understand how social boundaries are defined and how meanings are constructed in relation to learning, achievement, and social identities. A core characteristic of a classroom ethnography is the sustained and prolonged engagement of the classroom by the researcher as she attempts to obtain an emic (insider) perspective of classroom interactions and subsequently triangulates a range of data sources that include participant observation, field notes, open-ended interviews, audio- or video-taped classroom interactions, and artefacts (Davis, 2012; Starfield, 2015). A classroom ethnography is not bound to just spoken interactions, however. Ethnographies related to L2 writing (e.g., Paltridge, Starfield, & Tardy, 2016) have also been conducted. What is more central to the ethnographic enterprise is the examination of

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

123

classroom processes that reflect and refract wider social and historical practices and values (Lau, 2017). Adopting an ethnography of communication approach in her classroom-based study, Duff (2002), for example, investigated how broader social issues such as immigration influenced the ways in which student–teacher interactions developed in a Canadian high school. In a similar vein, our next exemplary study (Sayer, 2013), which is situated against a larger backdrop of globalization and immigration, examines how the use of a hybridized Spanish-English variety called TexMex served as a valuable mediating tool to help bilingual elementary school students in San Antonio, Texas. Sayer describes in detail the ethnographic, linguistic, and demographic contexts of the school, as the focal teacher and students used TexMex to navigate an academic context and standard languages. Exemplar Study 5 Sayer, P. (2013). Translanguaging, TexMex, and bilingual pedagogy: Emergent bilinguals learning through the vernacular. TESOL Quarterly, 47(1), 63–88. Research Questions • What role does language mixing have in the bilingual classroom? • Does it help emergent bilingual students learn standard English or Spanish? • Does it mediate other sorts of learning? If so, how? Theoretical Framework • Translanguaging Methods • Participant observation of the classroom over two years. The field notes and interviews were also recursively analysed. Participants • The focus of this ethnographic study was a second-grade classroom with a teacher and fifteen children. Results • The teacher’s use of translanguaging helped both the teacher and students to create discursive spaces by allowing the latter to exercise their bilingual Latinos status. • Translanguaging as a pedagogy also helped them to make sense of content and language learning as well as perform their desired identities.

5.4.6  Grounded Theory Grounded theory, according to Dillon (2012), seeks to generate theory inductively from data collected by creating categories that describe patterns in the data. This theory helps explain a process associated with particular phenomena. As the analytical process progresses, researchers

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

124

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

become more familiar with the data and generate insights that can subsequently be used to facilitate further data collection and analysis, resulting in the emergence of a theory (see Hadley, 2017, for a comprehensive review of grounded theory in applied linguistics). Thus, more often than not, the whole process involves stepping back from the data after each cycle of coding, and writing about how a theory emerges from the analysed data (Baralt, 2012). Making a clear distinction between grounded theory (GT) as a general method and GT as methodology (GTM), Dillon (2012) points out that the latter warrants attention to the researcher’s philosophical and theoretical perspectives that undergird the study. With the notable exception of Hadley (2015), who adopts a grounded theory methodology to examine the strategies used by English for academic purposes professionals to survive in neoliberal higher education institutions, most SLA classroom researchers are, however, more likely to adopt a grounded theory as a general method (versus grounded theory as a methodology) approach. Hence, rather than developing a new theory, their objective is generally to explain the processes underlying a classroom phenomenon guided by an established theory such as possible selves theory (see Exemplar Study 6). For example, in her study of an EFL teacher in Slovakia, Kubanyiova (2015) adopted a language teacher cognition theoretical framework to study how her focal teacher used a mental image of her teacher self to shape her teacher-led discussions. Using a similar framework as Kubanyiova, our exemplar study (Valmori & De Costa, 2016) adopted a grounded theory method—in this case interview data that were subjected to a multi-stage coding process developed from Strauss and Corbin (1998)—to explore how foreign language teachers maintained their proficiency. Exemplar Study 6 Valmori, L., & De Costa, P. I. (2016). How do foreign language teachers maintain their proficiency? A grounded theory approach. System, 57, 98–108. Research Questions • How do foreign language teachers think their foreign language proficiency changes during the course of their teaching career? • How can foreign language teachers’ perceived proficiency shape their teaching? Theoretical Framework • Markus and Nurius’ (1986) possible selves theory and Kubanyiova’s (2009) possible language selves construct. Methods • In-person and Skype interviews with nine Italian foreign language (FL) high school teachers.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

125

Participants • Nine Italian FL high school teachers who taught in either a college preparation or vocational school. Results • The foreign language teachers saw their foreign language proficiency as being a foundation for their teaching and central to their students’ learning. • The teachers thought that their foreign language proficiency can take different developmental trajectories, and they perceived proficiency changes as contributing to their self-efficacy.

5.4.7  Ethical Considerations Given the wider ethical turn in applied linguistics (e.g., De Costa, 2014, 2016b) and the growing calls for social justice in language education (e.g., Hawkins, 2011), SLA researchers have come to embrace the core ethical principles of maintaining respect for persons and yielding optimal benefits while minimizing harm and justice. However, in view of the emphasis on the social learning context and, correspondingly, the close social contact that classroom-based qualitative researchers have with their research participants, the former needs to make an added effort to ensure that the integrity and privacy of their participants is preserved at all costs. Beyond getting formal participant consent from an institutional ethical review board, qualitative researchers are increasingly expected to engage in researcher reflexivity and researcher positioning (Anderson, 2017; De Costa, 2015b). In other words, researchers need to see themselves as being a crucial part of the investigative process—a part that recognizes that they are individuals who have to be “explicitly and systematically accounted for and placed in [their] historical, political, and symbolic context” (Kramsch & Whiteside, 2007, p. 918), and thus be subjected to as much scrutiny as their participants.

5.5  Looking Ahead Undeniably, CBQR has made deep inroads in SLA over the past two decades since (a) Firth and Wagner’s (1997) exhortation to broaden the SLA base, which traditionally had been primarily cognitively- and quantitatively-oriented, and (b) more recent calls to bridge the gap between cognitive and social approaches to research in second language learning and teaching (Hulstijn et al., 2014). In a follow up to Richards’ (2009) review of three major journals—Applied Linguistics (AL), The Modern Language Journal (MLJ), and International Review of Applied Linguistics (IRAL)—which noted that qualitative research constituted less than twenty per cent of work in applied

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

126

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

linguistics, Lew, Yang, & Harklau, 2018 observed that significant differences and patterns within journals still exists across journals, with some (e.g., IRAL) still publishing very little, while the percentage had gradually increased in others (e.g., MLJ, AL, and TESOL Quarterly).

5.6  Methodological Rigour Throughout this chapter we have reiterated that the methodologies chosen by classroom-based researchers are influenced by their respective research paradigms and the theoretical frameworks they adopt to guide their research. And while these primary concerns need to be and are taken into consideration in the research design process, one way to generate further interest in CBQR is to ensure that steps are taken to provide a transparent explanation of (a) how the research participants and sites were chosen and how the data collection tools were created, and (b) how the design plan was executed and how the data were analysed. As noted, qualitative researchers are also expected to discuss their researcher positionality and describe how they negotiated ethical issues that may have emerged as they go about making appropriate claims based on their findings. Together, a sound description of these measures would contribute to the credibility and dependability (Abbuhl & Mackey, 2017) of CBQR.

5.7  Methodological Innovation As the field of SLA evolves, CBQR needs to move in sync with a changing educational landscape. For example, in view of a greater digital presence in and across language classrooms, new data collection and analytic tools need to be created. Working within an SLA study abroad context, the special issue of System edited by De Costa, Rawal, and Zaykovskaya (2017), for example, describes how the study abroad research agenda can be advanced through methodological innovation (see Choi & Richards, 2016, for a discussion on innovation in applied linguistics research). This innovation could include exploring new forms of ethnography, such as connection ethnography (Wargo & De Costa, 2017), which allows researchers to trace the relationship between in-class literacy development and out-of-class digital literacy development, and new ways of analysing multimodal data as a result of shuttling by learners between different learning domains. Thus, a greater focus on the digital aspects of language learning would require more sophisticated analytic tools, such as those recommended by Norris (2014), who used Multimodal

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

127

(Inter)action Analysis to investigate literacy development in a school setting (see Pirini, Matelau-Doherty, & Norris, 2018, for an overview of multimodal analysis). At the same time, and in order to manage “big” qualitative data that might emerge from classroom research (e.g., through a comparative analysis of several classrooms), computer-assisted qualitative data analysis software (CAQDAS) packages such as ATLAS.ti, MAXQDA, and NVivo, would need to be used to effectively manage the data (AndersonGough, Edgeley, & Sharma, 2017; Paulus, Jackson, & Davidson, 2017). Methodological innovation can also include the use of methodologies such as autoethnography (e.g., Canagarajah, 2012) and poetic inquiry (e.g., Hanauer, 2012) to trace language learner and teacher identity development. In light of the growing transdisciplinary nature of SLA, some SLA researchers have started to move beyond a quantitative/qualitative binary and opted for an eclectic mixed methods approach (Riazi, 2017). Thus, rather than engaging in protracted conceptual debates, such researchers prefer a layered approach (King & Mackey, 2016; Subtirelu, Mackey, & Plonsky, 2017) in order to investigate a phenomenon from varied epistemological stances. However, such a methodological sea change will take time and will require SLA researchers to continually update their methodological knowledge and skills, as astutely observed by Gonulal, Loewen, and Plonsky (2017). In relation to carrying out qualitative research, academic upgrades could take the form of learning how to ensure that ethical research practices are in place (e.g., Sterling & Gass, 2017) and learning about how to respond to reviewer criticism when attempting to publish qualitative research (Gao, 2017). Fortunately, qualitative SLA researchers now have the option to participate in webinars, summer and winter school workshops, and pre-conference workshops. In 2018, for example, the American Association for Applied Linguistics (AAAL) organized two workshops for individuals interested in CBQR: Exploring Digital Tools for Qualitative Research and Integrative Mixed Methods Research Design and Analysis.

5.8 Conclusion In a recent forum article published in Applied Linguistics, Rose and McKinley (2017) warned of a possible two-tier system emerging in the field, one where language pedagogy-related research, once the cornerstone of applied linguistics, is in jeopardy of being consigned to education-focused journals that place practice at their core. Specifically, they cite Kramsch (2015), who claimed that applied linguistics has become highly intellectualized,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

128

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

resulting in a widened gap between researchers and practitioners. As a consequence, journals such as Applied Linguistics have become relatively inaccessible or irrelevant to the daily professional lives of language teaching professionals. Building on this lament, we like to close this chapter with a call to SLA researchers to heed this warning. While it would be safe to claim that SLA researchers are generally committed to studying the processes underlying language learning and teaching, we also need to remind ourselves about the need to consider the social utility (Ortega, 2005) of our research. One way, as we have argued throughout this chapter, is to engage in CBQR because it puts the language classroom front and centre. The onus then is on us to make our findings accessible and relevant to consumers of our work, be it fellow researchers, teachers, or policymakers. We must not forget that language learning and teaching is an endeavour that cuts across multiple scales.

References Abbuhl, R., & Mackey, A. (2017). Second language acquisition research methods. In K. A. King, Y. Lai, & S. May (eds.), Research methods in language and education (pp. 183–193). New York: Springer. Allwright, D., & Bailey, K. M. (1991). Focus on the language classroom: An introduction to classroom research for language teachers. New York: Cambridge University Press. Anderson, C. (2017). Ethics in qualitative language education research. In S.-A. Mirhosseini (ed.), Reflections on qualitative research in language and literacy education (pp. 59–73). New York: Springer. Anderson-Gough, F., Edgeley, C., & Sharma, N. (2017). Qualitative data management and analysis software. In Z. Hoque, L. D. Parker, & M. A. Covaleski (eds.), The Routledge companion to qualitative accounting research methods (pp. 405–431). London, England: Routledge. Atkinson, D. (ed.) (2011). Alternative approaches to second language acquisition. New York: Routledge. Atkinson, D. (2014). Language learning in mindbodyworld: A sociocognitive approach to second language acquisition. Language Teaching, 47, 467–483. Baralt, M. (2012). Coding qualitative data. In A. Mackey & S. M. Gass (eds.), Research methods in second language acquisition (pp. 222–244). Malden, MA: Wiley-Blackwell. Barkhuizen, G. (2014). Research timeline: Narrative research in language teaching and learning. Language Teaching, 47(4), 450–466. Barkhuizen, G. (2017). Investigating language tutor social inclusion identities. The Modern Language Journal, 101(S1), 61–75.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

129

Benson, P. (2012). Qualitative methods: Overview. In C. A. Chapelle (ed.), The encyclopedia of applied linguistics (pp. 4773–4783). Oxford: Blackwell. Benson, P. (2018). Narrative analysis. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 595–613). London: Palgrave Macmillan. Benson, P., Chik, A., Gao, X., Huang, J., & Wang, W. (2009). Qualitative research in 10 language teaching and learning journals, 1997–2006. The Modern Language Journal, 93(1),79–90. Block, D. (2003). The social turn in second language acquisition. Edinburgh: Edinburgh University Press. Bloome, D. (2012). Classroom ethnography. In M. Grenfell, D. Bloome, C. Hardy, K. Pahl, J. Roswell, & B. Street (eds.), Language, ethnography, and education: Bridging New Literacy Studies and Bourdieu (pp. 8–26). New York: Routledge. Burns, A., & Edwards, E. (2016). Language teacher action research: Achieving sustainability. English Language Teaching Journal, 70 (1), 6–15. Burns, A., & McPherson, P. (2017). Action research as iterative design: Implications for English language education research. In S.-A. Mirhosseini (ed.), Reflections on qualitative research in language and literacy education (pp. 105–120). New York: Springer. Calvert, M., & Sheen, Y. (2015). Task-based language learning and teaching: An action-research study. Language Teaching Research, 19(2), 226–244. Canagarajah, S. (2012). Teacher development in a global profession: An autoethnography. TESOL Quarterly, 46, 258–279. Casanave, C. P. (2015). Case studies. In B. Paltridge & A. Phakiti (eds.), Research methods in applied linguistics: A practical resource (pp. 119–136). London: Bloomsbury. Chapelle, C. A. & Duff, P. A. (eds.) (2003). Some guidelines for conducting quantitative and qualitative research in TESOL. TESOL Quarterly, 37, 172–178. Choi, S., & Richards, K. (guest eds.) (2016). Innovation in research methods in applied linguistics [Special issue]. Applied Linguistics, 37(1). Collins, L., & Muñoz, C. (2016). The foreign language classroom: Current perspectives and future considerations. The Modern Language Journal, 100(S1), 133–147. Copland, F. (2018). Observation and fieldnotes. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 249–268). London: Palgrave Macmillan. Davis, K. A. (2012). Ethnographic approaches to second language acquisition research. In C. A. Chapelle (ed.), The encyclopedia of applied linguistics (pp. 1–8) (Oxford: Blackwell). De Costa, P. I. (2014). Making ethical decisions in an ethnographic study. TESOL Quarterly, 48, 413–422.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

130

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

De Costa, P. I. (2015a). Re-envisioning language anxiety in the globalized classroom through a social imaginary lens. Language Learning, 65(3), 504–532. De Costa, P. I. (2015b). Tracing reflexivity through a narrative and identity lens. In Y. L. Cheung, S. B. Said & K. Park (eds.), Advances and current trends in language teacher identity research (pp. 135–147). New York: Routledge. De Costa, P. I. (2016a). Scaling emotions and identification: Insights from a scholarship student. Linguistics and Education, 34, 22–32. De Costa, P. I. (ed.) (2016b). Ethics in applied linguistics research: Language researcher narratives. New York: Routledge. De Costa, P. I., & Canagarajah, S. (guest eds.) (2016). Scalar approaches to language teaching and learning [Special issue]. Linguistics and Education, 34. De Costa, P. I., & Norton, B. (guest eds.) (2017). Transdisciplinarity and language teacher identity [Special issue]. The Modern Language Journal, 101(S). De Costa, P. I., Rawal, H., & Li, W. (2018). L2 teachers’ emotions: A sociopolitical and ideological perspective. In J. de Dios Martínez Agudo (ed.), Emotions in second language teaching (pp. 91–108). New York: Springer. De Costa, P. I., Rawal, H., & Zaykovskaya, I. (guest eds.) (2017). Study abroad in contemporary times: Toward greater methodological diversity and innovation [Special issue]. System, 71. De Costa, P. I., Tigchelaar, M., & Cui, Y. (2016). Reflexivity, emotions and transnational habitus: The case of a “poor” cosmopolitan Chinese international student. AILA Review, 29, 173–198. De Costa, P. I., Valmori, L., & Choi, I. (2017). Qualitative research methods. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 522–540). New York: Routledge. Diao, W. (2017). Between the standard and non-standard: Accent and identity among transnational Mandarin speakers studying abroad in China. System, 71, 87–101. Dillon, D. R. (2012). Grounded theory and qualitative research. In C. A. Chapelle (ed.), The encyclopedia of applied linguistics. Oxford: Blackwell. Douglas Fir Group (2016). A transdisciplinary framework for SLA in a multilingual world. The Modern Language Journal, 100(S1), 19–47. Duff, P. A. (2002). The discursive co-construction of knowledge, identity, and difference. An ethnography of communication in the high school mainstream. Applied Linguistics, 23, 289–322. Duff, P. A. (2007). Qualitative approaches to second language classroom research. In J. Cummins & C. Davison (eds.), Handbook of English language teaching (pp. 973–986). New York: Springer. Duff, P. A. (2012). How to carry out case study research. In A. Mackey & S. M. Gass (eds.), Research methods in second language acquisition (pp. 95–116). Malden, MA: Wiley-Blackwell.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

131

Duff, P. A. (2014). Case study research on language learning and use. Annual Review of Applied Linguistics, 34, 233–255. Duff, P. A. (2015). Transnationalism, multilingualism, and identity. Annual Review of Applied Linguistics, 35, 57–80. Firth, A., & Wagner, J. (1997). On discourse, communication, and (some) fundamental concepts in SLA research. The Modern Language Journal, 81(3), 285–300. Firth, A., & Wagner, J. (1998). SLA property: No trespassing! The Modern Language Journal, 82(1), 91–94. Firth, A., & Wagner, J. (2006). Second/foreign language learning as a social accomplishment: Elaborations on a reconceptualized SLA. The Modern Language Journal, 91(S1), 800–819. Friedman, D. (2012). How to collect and analyze qualitative data. In A. Mackey & S. M. Gass (eds.), Research methods in second language acquisition (pp. 180–200). Malden, MA: Wiley-Blackwell. Galloway, N. (2017). Researching your own students: Negotiating the dual practitioner-researcher role. In J. McKinley & H. Rose (eds.), Doing research in applied linguistics: Realities, dilemmas, and solutions (pp. 146–156). New York: Routledge. Galloway, N., & Rose, H. (2015). Introducing global Englishes. New York: Routledge. Gao, X. (2017). Dealing with criticism when publishing qualitative research. In J. McKinley & H. Rose (eds.), Doing research in applied linguistics: Realities, dilemmas, and solutions (pp. 225–234). New York: Routledge. García, O., Ibarra Johnson, S., & Seltzer, K. (2016). The translanguaging classroom: Leveraging student bilingualism for learning. Philadelphia, PA: Caslon. Garrett, P. B. (2017). Researching language socialization. In K. A. King, Y.-J. Lai, & S. May (eds.), Research methods in language and education (pp. 283– 295). New York: Springer International Publishing. Gass, S. M., Lee, J., & Roots, R. (2007). Firth and Wagner: New ideas or a new articulation? The Modern Language Journal, 91, 788–799. Gibbons, P. (2003). Mediating language learning: Teacher interactions with ESL students in a content-based classroom. TESOL Quarterly, 37, 247–271. Gonulal, T., Loewen, S., & Plonsky, L. (2017). The development of statistical literacy in applied linguistics graduate students. ITL: International Journal of Applied Linguistics, 168(1), 4–32. Hadley, G. (2015). English for academic purposes in neoliberal universities: A critical grounded theory. New York: Springer. Hadley, G. (2017). Grounded theory in applied linguistics research. New York: Routledge. Hanauer, D. I. (2012). Meaningful literacy: Writing poetry in the language classroom. Language Teaching, 45(1), 105–115. Harklau, L. (2011). Approaches and methods in recent qualitative research. In E. Hinkel (ed.), Handbook of research in second language teaching and learning (Vol. 2, pp. 175–189). New York: Routledge.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

132

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

Hawkins, M. R. (ed.) (2011). Social justice language teacher education. Tonawanda, NY: Multilingual Matters. Hellermann, J. (2011). Members’ methods, members’ competencies: Looking for evidence of language learning in longitudinal investigations of other-initiated repair. In S. P. Doehler, J. H. Hellermann, & J. K. Hall (eds.), L2 interactional competence and development (pp. 147–172). Bristol: Multilingual Matters. Hellermann, J., & Jakonen, T. (2017). Interactional approaches to the study of classroom discourse and student learning. In K. A. King, Y. Lai, & S. May (eds.), Research methods in language and education (pp. 297–309). New York: Springer. Holliday, A. (2007). Doing and writing qualitative research (2nd edn.). Thousand Oaks, CA: Sage. Hulstijn, J. H., Young, R. F., Ortega, L., Bigelow, M., DeKeyser, R., Ellis, N. C., & Talmy, S. (2014). Bridging the gap: Cognitive and social approaches to research in second language learning and teaching. Studies in Second Language Acquisition, 36(3), 361–421. Jakonen, T., & Morton, T. (2015). Epistemic search sequences in peer interaction in a content-based language classroom. Applied Linguistics, 36(1), 73–94. Johnson, K. E., & Golombek, P. R. (2016). Mindful L2 teacher education: A sociocultural perspective on cultivating teachers’ professional development. New York: Routledge. Kanno, Y., & Harklau, L. (eds.) (2012). Linguistic minority students go to college: Preparation, access, and persistence. New York: Routledge. Kasper, G., & Prior, M. (2015). Analyzing storytelling in TESOL interview research. TESOL Quarterly, 49(2), 226–255. Kasper, G., & Wagner, J. (2011). A conversation-analytic approach to second language acquisition. In D. Atkinson (ed.), Alternative approaches to second language acquisition (pp. 117–142). New York: Routledge. King, K. A., & Mackey, A. (2016). Research methodology in second language studies: Trends, concerns, and new directions. The Modern Language Journal, 100, 209–227. Kramsch, C. (2015). Applied linguistics: A theory of the practice. Applied Linguistics, 36(4), 454–465. Kramsch, C., & Whiteside, A. (2007). Three fundamental concepts in second language acquisition and their relevance in multilingual contexts. The Modern Language Journal, 91(1), 907–922. Krashen, S. D. (1982). Principles and practice in second language acquisition. New York: Pergamon Press. Kubanyiova, M. (2009). Possible selves in language teacher development. In Z. Dörnyei & E. Ushioda (eds.), Motivation, language identity and the L2 self (pp. 314–332). Bristol: Multilingual Matters. Kubanyiova, M. (2015). The role of teachers’ future self-guides in creating L2 development opportunities in teacher-led classroom discourse:

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

133

Reclaiming the relevance of language teacher cognition. The Modern Language Journal, 99(3), 565–584. Lantolf, J. P., & Poehner, M. E. (2014). Sociocultural theory and the pedagogical imperative in L2 education: Vygotskian praxis and the research/practice divide. New York: Routledge. Larsen-Freeman, D. (2018). Second language acquisition, WE and language as a complex adaptive system (CAS). World Englishes, 37(1), 80–92. Lau, S. M. C. (2017). Classroom ethnography on critical language and literacy teaching and learning. In S.-A. Mirhosseini (ed.), Reflections on qualitative research in language and literacy education (pp. 77–90). New York: Springer. Lazaraton, A. (1995). Qualitative research in applied linguistics: A progress report. TESOL Quarterly, 29(3), 455–472. Lee, G., & Wallace, A. (2018). Flipped learning in the English as a foreign language classroom: Outcomes and perceptions. TESOL Quarterly, 52(1), 62–84. Lew, S., Yang, A. H., & Harklau, L. (2018). Qualitative methodology. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 79–101). London: Palgrave Macmillan. Li, W. (2018). Translanguaging as a practical theory of language. Applied Linguistics, 39(1), 9–30. Li, W., & De Costa, P. I. (2017). Professional survival in a neoliberal age: A case study of an EFL teacher in China. Asia TEFL Journal, 14(2), 277–291. Li, W., & De Costa, P. I. (2018). Exploring novice EFL teachers’ identity development: A case study of two EFL teachers in China. In S. Mercer & A. Koustoulas (eds.), Teacher psychology in SLA (pp. 86–104). Bristol: Multilingual Matters. Luk, J. C. M. (2017). Classroom discourse and the construction of learner and teacher identities. In S. Wortham, D. Kim, & S. May (eds.), Discourse and education (pp. 173–184). New York: Springer. Mahboob, A., Paltridge, B., Phakiti, A., Wagner, E., Starfield, S., Burns, A., Jones, R. H., & De Costa, P. I. (2016). TESOL Quarterly research guidelines. TESOL Quarterly, 50(1), 42–65. Maloney, J., & De Costa, P. I. (2017). Imagining the Japanese heritage learner: A scalar perspective. Language, Discourse, & Society, 9(1), 35–52. Mangual Figueroa, A., & Baquedano-López, P. (2017). Language socialization and schooling. In P. A. Duff & S. May (eds.), Language socialization (pp. 1–13). New York: Springer. Manosuthikit, A., & De Costa, P. I. (2016). Ideologizing age in an era of superdiversity: A heritage language practice perspective. Applied Linguistics Review, 7(1), 1–25. Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41(9), 954–969.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

134

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

McKay, S. L. (2006). Researching second language classrooms. Mahwah, NJ: Lawrence Erlbaum. Miller, E. R. (2018). Interaction analysis. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 615–639). London: Palgrave Macmillan. Mirhosseini, S.-A. (2017a). Introduction: Qualitative research in language and literacy education. In S.-A. Mirhosseini (ed.), Reflections on qualitative research in language and literacy education (Vol. 29, pp. 1–13). New York: Springer. Mirhosseini, S.-A. (ed.) (2017b). Reflections on qualitative research in language and literacy education. New York: Springer. Norris, S. (2014). The impact of literacy-based schooling on learning a creative practice: Modal configurations, practices and discourses. Multimodal Communication, 3(2), 181–195. Norton, B., & De Costa, P. I. (2018). Research tasks on identity and language education. Language Teaching, 51(1), 90–112. Ohta, A. S. (2017). Sociocultural theory and second/foreign language education. In N. Van Deusen-Scholl & S. May (eds.), Second and foreign language education (pp. 57–68). New York: Springer. Ortega, L. (2005). For what and for whom is our research? The ethical as transformative lens in instructed SLA. The Modern Language Journal, 89(3), 427–443. Paltridge, B., & Phakiti, A. (2015). Approaches and methods in applied linguistics research. In B. Paltridge & A. Phakiti (eds.), Research methods in applied linguistics: A practical resource (pp. 5–25). London: Bloomsbury. Paltridge, B., Starfield, S., & Tardy, C. M. (2016). Ethnographic perspectives on academic writing. Oxford: Oxford University Press. Paulus, T. M., Jackson, K., & Davidson, J. (2017). Digital tools for qualitative research: Disruptions and entanglements. Qualitative Inquiry, 23, 751–756. Pirini, J., Matelau-Doherty, T., & Norris, S. (2018). Multimodal analysis. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 639–658). London: Palgrave Macmillan. Prior, M. T. (2018). Interview and focus group research. In A. Phakiti, P. I. De Costa, L. Plonsky, & S. Starfield (eds.), The Palgrave handbook of applied linguistics research (pp. 225–248). London: Palgrave Macmillan. Riazi, A. M. (2017). Mixed methods research in language teaching and learning. Sheffield: Equinox. Richards, K. (2003). Qualitative inquiry in TESOL. New York: Palgrave MacMillan. Richards, K. (2009). Trends in qualitative research in language teaching since 2000. Language Teaching, 42(2), 147–180. Rose, H., & McKinley, J. (2017). The prevalence of pedagogy-related research in applied linguistics: Extending the debate. Applied Linguistics, 38(4), 599–604.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

Qualitative Classroom Methods

135

Sayer, P. (2013). Translanguaging, TexMex, and bilingual pedagogy: Emergent bilinguals learning through the vernacular. TESOL Quarterly, 47(1), 63–88. Smotrova, T., & Lantolf, J. (2013). The function of gesture in lexically focused L2 instructional conversations. The Modern Language Journal, 97(2), 397–416. Starfield, S. (2015). Ethnographic research. In B. Paltridge & A. Phakiti (eds.), Research methods in applied linguistics: A practical resource (pp. 137– 152). London: Bloomsbury. Sterling, S., & Gass, S. (2017). Exploring the boundaries of research ethics: Perceptions of ethics and ethical behaviors in applied linguistics research. System, 70, 50–62. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage. Subtirelu, N. C., Mackey, A., & Plonsky, L. (2017). Working across methodological paradigms: Mixed methods research in applied linguistics. Workshop presented at Second Language Research Forum, Columbus, OH. Thorne, S. L., Sauro, S., & Smith, B. (2015). Technologies, identities and expressive activity. Annual Review of Applied Linguistics, 35, 215–33. Tsui, A. B. M. (2017). Classroom discourse: Theoretical orientations and research approaches. In J. Cenoz, D. Gorter, & S. May (eds.), Language awareness and multilingualism (pp. 187–203). New York: Springer. Valmori, L., & De Costa, P. I. (2016). How do foreign language teachers maintain their proficiency? A grounded theory approach. System, 57, 98–108. Van Compernolle, R. A. (2014). Sociocultural theory and L2 instructional pragmatics. Bristol: Multilingual Matters. VanPatten, B., & Benati, A. G. (2015). Key terms in second language acquisition. New York: Bloomsbury Publishing. Wallace, M. J. (1998). Action research for language teachers. New York: Cambridge University Press. Walqui, A., & van Lier, L. (2010). Scaffolding the academic success of adolescent English language learners: A pedagogy of promise. San Francisco, CA: WestEd. Wargo, J. M., & De Costa, P. I. (2017). Tracing academic literacies across contemporary literacy sponsorscapes: Mobilities, ideologies, identities, and technologies. London Review of Education, 15(1), 101–114. Waring, H. Z. (2017). Conversation analytic approaches to language and education. In K. A. King, Y.-J. Lai, & S. May (eds.), Research methods in language and education (pp. 463–474). New York: Springer. Warriner, D., & Anderson, K. T. (2017). Discourse analysis in educational research. In K. A. King, Y.-J. Lai, & S. May (eds.), Research methods in language and education (pp. 297–309). New York: Springer. Winke, P. (2017). Using focus groups to investigate study abroad theories and practice. System, 71, 73–83.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

136

P e t e r I . D e C o s ta , W e n d y L i , a n d H im a R a w a l

Wolff, D., & De Costa, P. I. (2017). Expanding the language teacher identity landscape: An investigation of the emotions and strategies of a NNEST. The Modern Language Journal, 101(S1), 76–90. Xu, Y. (2014). Becoming researchers: A narrative study of Chinese university EFL teachers’ research practice and their professional identity. Language Teaching Research, 18(2), 242–259. Zappa-Holman, S., & Duff, P. A. (2015). Academic English socialization through individual networks of practice. TESOL Quarterly, 49, 339–368. Zaykovskaya, I., Rawal, H., & De Costa, P. I. (2017). Learner beliefs for successful study abroad experience: A case study. System, 71, 113–121.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.006

6 Experimental Studies in L2 Classrooms Charlene Polio and Jongbong Lee

6.1 Introduction 6.1.1  Defining Experimental Research The nature of experimental research is not controversial, but we begin here with Gass’ (2015, p. 101) definition in relation to applied linguistics: Experimental research is a way of determining the effect of something on something else. In other words, a researcher begins with an idea of why something happens and manipulates at least one variable, and controls others, to determine the effect on some other variable.

A simple example of a research question related to classroom instruction that might be answered by experimental research is: does the use of authentic listening materials instead of simplified listening materials result in improved listening comprehension for second language (L2) learners? In order to answer this question through a true experimental study, the researcher would have to manipulate the hypothesized cause (listening materials) on some effect (listening comprehension). Of course, many will immediately realize that designing a study to answer such a question is not without problems, but the hypothesized nature of the relationship between manipulated or independent variable (listening materials) and the outcome or dependent variable (listening comprehension) in an experimental study is one of cause and effect. Interestingly, Phakiti’s definition of experimental research in the TESOL Quarterly research guidelines (Mahboob et al., 2016, p. 46) of experimental research is as follows: Experimental research aims to determine causal-like relationships between one or more independent variables (e.g., types of instruction, feedback) and one or more dependent variables (e.g., language acquisition, learning behaviors). [emphasis added]

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

138

C h a r le n e P o l i o a n d J o n g b o n g L ee

Given the challenges of designing true experimental studies, which will be discussed here, Phakiti’s use of the term causal-like may be significant in that he may see true causal relationships as elusive, but he does not directly address the matter. In the past, experimental research was somewhat at odds with qualitative research, both in terms of ontology (one’s view of reality) and epistemology (what knowledge is). Experimental research falls within a postpositivist paradigm, which suggests that an objective and discoverable truth exists (its ontology) and that we can find out that truth by generating and testing hypotheses (its epistemology). Using the example of authentic listening materials, this means that under ideal circumstances, researchers should be able to determine the effect of authentic listening materials on listening comprehension. Qualitative research, on the other hand, falls within a postmodernist paradigm that rests on the ideas that there are multiple truths (its ontology) and that research should try to uncover the diversity of experiences and perspectives on a topic of inquiry (its epistemology). Despite seemingly irreconcilable differences between experimental and qualitative research, the field of applied linguistics has begun to accept mixed methods research, which in its true form integrates quantitative and qualitative data. Although not all studies that claim to be mixed methods truly integrate both types of data (Polio & Friedman, 2017), many researchers have recently promoted mixed methods research as a viable option for applied linguistics (e.g., Brown, 2014; Dörnyei, 2007; Hashemi, 2012; Hashemi & Babaii, 2013; Jang, Wagner, & Park, 2014; Riazi & Candlin, 2014). Indeed, we will later argue that mixed methods research that uses qualitative data in conjunction with an experimental, or quasiexperimental, study is ideal for examining language classrooms, and a few such studies are mentioned in this chapter. Not all quantitative research is experimental; much of it is causal-­ comparative. This means the study may have the same research question or hypothesis as an experimental study, but that the researcher does not manipulate the independent variable and instead measures or categorizes the independent variable as it exists in the research context. For example, to examine the effect of authentic listening materials, instead of manipulating the materials, a researcher might try to measure the amount of authentic materials used in various classes and try to find the relationship to students’ listening comprehension. In other words, if researchers want to study the relationship between an instructional variable and a learning outcome, they can design an experimental study where different groups receive different treatments, or they can look at the relationship between variables as they exist a priori. Causal-comparative studies often examine variables that can never be manipulated by the researcher, such as learner

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

139

variables including gender or first language, but they can also include classroom variables that are impractical for the researcher to manipulate, such as student level of language anxiety. These types of studies will not be discussed further, but it is important to remember that they are less effective at showing cause–effect relationships because the researcher does not control the variables. We note, however, that some experimental studies often include a causal-comparative aspect such as Rassaei (2015), who studied the effect of recasts but at the same time included students’ cognitive style as a variable. In this chapter, we will review research that attempts to isolate the effect of some treatment or intervention in the classroom. This includes true experimental studies in which students are randomly assigned to different conditions. Random assignment is a hallmark of experimental research, but because random assignment is often impossible in classroom research, we have chosen to include quasi-experimental research in this review. Specifically, we have included studies that use intact classes as opposed to random assignment and studies that have no control group.

6.1.2  Defining Classroom Research Defining the scope of classroom research is not as easy as one might expect. For example, many studies of students who are learning in classrooms are designed so that the intervention actually occurs outside of the classroom. In the example of the hypothetical authentic listening materials study, students could complete all their listening activities online outside of the classroom. Written corrective feedback studies are often, but not always, designed so that the feedback is independent of what happens in the classrooms: teachers give feedback on papers and students revise the papers at home. In this chapter, we address only studies in which some or all of the intervention is done in the classroom or in which the intervention appears to be an integral part of the curriculum as opposed to an intervention that was added for research purposes. Other recent reviews (Collins & Muñoz, 2016; Mackey, 2017) define the scope of classroom research in similar, but not identical ways. In addition, there are many laboratory studies of classroom activities, and by laboratory, we mean studies done outside of the classroom independent of the curriculum. These laboratory studies sometimes examine activities or tasks that could easily be done in the classroom, but were not (e.g, the vocabulary study of Sonbul & Schmitt, 2013). Researchers often choose to conduct such studies in the laboratory, often a small room, for logistic reasons such as noise level or to control for issues such as time on task. These studies do not fall under what we consider to be classroom research even though they would fall under the scope of research on

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

140

C h a r le n e P o l i o a n d J o n g b o n g L ee

instructed second language acquisition (ISLA), which is defined by Loewen (2015) as “the systematic manipulation of the mechanisms of learning and/or the conditions under which they occur” (p. 2). In other words, most classroom experimental research is a subset of instructed second language acquisition, but it is possible that a study might have something other than student language learning as the dependent variable (e.g., anxiety level in Dolean, 2016). Finally, we want to point out that qualitative research, in contrast, is less likely to narrowly define the classroom. For example, in De Costa, Valmori, and Choi’s (2017) review of qualitative research methods for ISLA, they refer to the multifaceted nature of language learning and teaching as detailed in the Douglas Fir Group (2016) piece, which views the classroom as only a small part of the learning and teaching process situated within a larger community. Indeed, De Costa, Li, and Rawal (this volume, Chapter 5) problematize the notion of an intact class. Qualitative research often tries to examine the bigger, and more holistic, picture of a phenomenon, so limiting the scope of inquiry may not be desirable.

6.2 Historical Perspectives on Experimental Research in Language Classes In an early review of classroom research, Chaudron (1988) characterized classroom research as falling into four categories: psychometric, interaction analysis, discourse analysis, and ethnographic approaches. The latter three refer to various ways to describe classroom interaction and will not be discussed further, but what Chaudron referred to as the psychometric approach seems to refer to experimental research. He described studies that compare methods of teaching such as grammar-translation and audio­ lingual approaches (as in Scherer & Wertheimer’s, 1964, study of German foreign language instruction). Chaudron echoed Long’s (1980) earlier concern that many studies did not document what was actually happening in the classroom in these comparative method studies and instead discussed only the outcomes. The result was that there was no check on what intervention was really being implemented or how the teachers may have adapted the treatment. Beretta (1986) also addressed a problem with studies of classrooms that compared teaching methods, namely, that the tests of the outcome variables can be biased towards one of the methods or treatments. Beretta refered to a study comparing a Total Physical Response (TPR) class with a control class (Asher, 1972). One of the outcome measures used in the study was the same as one of the treatments used in the TPR class.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

141

In the area of interaction, because much experimental research has been conducted in the laboratory, Gass, Mackey, and Ross-Feldman (2005) stated that researchers needed to consider the applicability of interaction research to classroom settings. They thus conducted a study in which students of Spanish completed tasks either in a classroom with a teacher giving directions, or in a laboratory with the researcher meeting individually with pairs of students. They found differences in interaction across task types, regardless of whether the tasks took place in the laboratory or in the classroom, but little difference across the two contexts. Gass, Mackey, and Ross-Feldman are quick to point out that their results cannot be extended to all classrooms, but their study was an important step in considering how experimental findings might be replicated in a classroom setting.

6.3  Issues Involved in Experimental Classroom Research In order to explain key concepts related to experimental classroom research, we use an example of a quasi-experimental study on the use of authentic materials (Gilmore, 2011). Although it is challenging to design good quasi- or true experimental studies, Gilmore was able to deal with many of the problems that are not addressed in other studies. We first summarize the study and then point to the central issues in designing an experimental classroom study using details from Gilmore’s study as an example. Any limitations of his design are in no way a criticism of the study given that we see this study as one of the more rigorous experimental classroom studies available. Gilmore (2011) wanted to study the effect of input from textbooks versus input from authentic materials on learners’ communicative competence. He conducted a quasi-experimental study with sixty-two second-year English majors from four intact English classes in Japan. Two intact classes were randomly assigned to an experimental group, which received input mostly from authentic sources such as films, television shows, novels, and newspaper articles. The control group used their textbooks, which appeared to contain no authentic materials, and what Gilmore called contrived texts. The classes, both taught by the same teacher, followed these two different syllabi for ten months. The students were tested before and after instruction with eight tests that were purported to test their communicative competence. Gilmore found that the experimental group outperformed the control group on five of the eight tests. In any kind of experimental research, the issues that need to be addressed can be grouped into matters related to internal validity, appropriate measurement of the outcome variables, and external validity, each addressed in the next three sections. Regarding classroom research specifically,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

142

C h a r le n e P o l i o a n d J o n g b o n g L ee

ensuring internal validity can be more challenging than in other types of language learning and teaching experimental studies. External validity, on the other hand, is often less of a concern. We address these three broad issues using Gilmore’s (2011) study as an example.

6.3.1 Operationalizing the Independent Variable and Ensuring Internal Validity In experimental research, the goal is to isolate the effect of the independent variable. The researcher has to first operationalize the variable. In Gilmore (2011), the author had to define what authentic materials are, explain how they are used in the experimental group, and detail how the intervention in the experimental group differs from that in the control group. After a discussion of how he defines authentic materials, Gilmore provides a two-page table of the themes of the sixteen units, including exactly which authentic materials were used for each unit. The control group used only materials from the textbook, which, the author explains, used contrived texts. Thus, he clearly articulates what is meant by authentic materials. What is less detailed is how the materials in the two classes were used. In other words, when one refers to the use of authentic materials, it would be helpful to operationalize what is meant by use. It is not clear that use could be explained in detail in a journal article, but in the dissertation version of the study (Gilmore, 2008) sample lessons that show how the materials were used are included. If a study is internally valid, it means that the outcome is due strictly to the manipulation of the independent variable. In laboratory studies, the researcher can control more of the variables than in a classroom study. Gilmore’s independent variable was the materials. To ensure that any outcomes were due to the materials, Gilmore randomly assigned the classes (not the students) to the experimental or control group. Although random assignment of individuals is preferable, it is often not possible in classroom research. When intact classes are used, there could be factors such as time of day that affect the outcome or, more likely, student placement into classes may be related to some factor that might affect achievement, such as the other courses they are taking. To deal with initial differences in students’ proficiency, Gilmore pre-tested the students and then used their pre-test scores in his statistical analysis (an analysis of covariance that takes into account initial group differences). Another common source of variability not related to the independent variable in classroom research is the teacher. Failing to consider the effect a teacher has on learning shows a misconception of the complexities of classroom learning. In Gilmore’s study, he taught both the control and experimental groups. Of course, this introduces a new source of

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

143

variability: the teacher’s attitude towards the treatment. This variability is unavoidable in any study in which the teacher knows what is being studied. Gilmore (2011) acknowledges this problem and states that he attempted to teach all classes “with equal commitment and enthusiasm” (p. 813). What’s more, in a longitudinal classroom study, factors such as teacher commitment can change over time or create other sources of variability. For example, a teacher’s attitude might affect the students’ attitudes, and then that too becomes a source of variability. One way to address internal validity in experimental classroom research is to include observations that detail how the intervention was administered. (Recall Chaudron’s, 1988, and Long’s, 1980, early calls for documenting what happens in the classroom during experimental research.) This should be done in classroom research where a teacher other than the researcher administers the instruction. Gilmore did not include observation data because he was the instructor, but it might have been helpful to have another person periodically observe the classes and note any differences in instruction not related to the materials. Classroom observation data should also be used as a qualitative data source to triangulate the results of the experimental study or to explain them. For example, Gilmore found that the experimental group outperformed the control group on receptive vocabulary. He gives several reasons why this may have happened, but classroom observations could have provided more details on what the instructor did or did not do to focus the learners’ attention on the new vocabulary.

6.3.2 Measuring the Dependent Variables: Practicality, Validity, and Reliability Determining precisely what is an appropriate dependent variable in a classroom study is not always easy. In laboratory studies, the focus is often very narrow and thus easier to measure. In a classroom study, the researcher may want to study an outcome such as overall language learning. The researcher must then decide whether to focus on general proficiency or specific areas. Gilmore chose to examine the general construct of communicative competence as his dependent variable. Communicative competence is multifaceted and challenging to measure. Gilmore discusses the construct at length and ultimately chose to examine a large number—thirteen—of dependent variables that came from eight different tests as a way to assess communicative competence. For example, a version of the IELTS oral interview was used to collect some of the data, and the samples were rated on five different factors related to communicative competence. When choosing any measures of the outcome variable, the researcher has to consider practicality, reliability, and validity of the measures.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

144

C h a r le n e P o l i o a n d J o n g b o n g L ee

Gilmore (2011) chose a range of tests that could be practically administered to the sixty-two participants. Of course, the tests and measures need to be valid. In other words, the tests need to measure what they say they are measuring. How to best measure communicative competence is beyond the scope of this chapter, but Gilmore makes an excellent case for why he chose the tests and measures that he did. Many studies do not clearly justify their measures, but Gilmore is very thorough in his discussion of how the tests assess the components of communicative competence including linguistic competence, pragmalinguistic competence, sociopragmatic competence, strategic competence, and discourse competence. The measures also need to be reliable. This means that the tests must be consistent and provide the same results each time. To assess vocabulary, for example, Schmitt’s Vocabulary Levels Test (Schmitt, 2000) was used. This is an established test used in other studies that, Gilmore reports, has a Cronbach alpha coefficient of over .9, which means the test is internally consistent. Other measures that involve rating responses need to be rated by more than one rater, and inter-rater reliability needs to be calculated. For example, the student role plays used to assess strategic and pragmatic competence had to be rated by more than one rater to ensure the reliability of the scores. Gilmore used the Pearson product-moment correlation coefficient to check the inter-rater reliability. The coefficient values (conversational behaviour: r = .70; conversational management: r = .85) indicate that the scale was reliable.

6.3.3  External Validity, Ecological Validity, and Reductionism External validity refers to the extent to which the results of a study can be extended to other contexts. In Gilmore’s (2011) study, the issue is whether or not similar results would have been found in contexts outside of Japan. Gilmore explicitly addresses this issue saying: The same experimental treatment would probably lead to very different results with a group of learners from another culture; for example, it is unlikely that statistically significant differences in sociopragmatic measures would have been observed in European students, for which behavioural norms are closer to those of NSs [native speakers] of English. Similarly, the lack of statistically significant results for the grammar test may have been due to the already well-developed grammatical competence exhibited by the Japanese subjects.  (pp. 810–811)

Ecological validity is similar but refers to the extent that a study can be extended to real-life contexts. Slightly different definitions of this term exist, but one from social psychology is as follows:

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

145

Ecological validity is one aspect of external validity in which researchers ask whether research results represent what happens in everyday life. More specifically, ecological validity addresses whether an effect has been shown to operate in conditions that occur often for people in the population of interest. (Wegener & Blankenship, 2007, p. 275)

In other words, if the results of a laboratory study cannot be extended to the classroom, the study would not have ecological validity. We can also consider the extent to which the experimental conditions are similar to classroom, or real-life, conditions. If the conditions match, the study can be said to have an ecologically valid design (as discussed in Spinner, Gass, & Behney, 2013, in relation to eye-tracking studies). Ecological validity is generally an issue for laboratory studies because sometimes the intervention cannot be implemented in a classroom. Experimental classroom studies usually have ecological validity because they are conducted in real classrooms. Nevertheless, the intervention may not be practical for a wide range of classes. For example, a study such as Gilmore’s (2011) may have not been possible in a context with very large classes. Gilmore addresses ecological validity as well by explaining the difficulties related to implementing his syllabus. He states that using authentic materials can be very time consuming and that non-native speaker teachers may have difficulty assessing the pragmalinguistic or sociolinguistic appropriateness of the materials. One final concern of experimental research is reductionism. Reductionism is a concept that has been addressed throughout the social sciences, particularly by qualitative researchers, who often problematize the examination of specific phenomena in isolation. In applied linguistics, researchers working in a complex systems framework have argued against trying to understand a concept as complex as language learning by examining individual parts (e.g., Larsen-Freeman & Cameron, 2008). For example, had Gilmore (2011) measured only listening comprehension, we might not have seen what else was affected by the use of authentic materials. He chose, however, to use a broad construct as his outcome measure, but not all studies choose to do so.

6.3.4  Reporting and Replicability Experimental studies need not only to be well-designed, but they also need to be well-detailed so that they can be replicated and evaluated. Replication has been written about extensively over the last ten years in the field of applied linguistics (e.g., Porte, 2012). While most agree that true replication is not possible in the social sciences, there is consensus that researchers need to report their design, measures, instruments, and coding procedures so that they can be replicated and evaluated. Thus, a

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

146

C h a r le n e P o l i o a n d J o n g b o n g L ee

clear description of the intervention and the instruments used to assess outcomes is essential. Gilmore (2011) describes the experimental class in detail. Readers do not have access to the materials, but this is not surprising given copyright issues. We note that the dissertation on which the journal article is based includes some sample materials (Gilmore, 2008). Whether the intervention is described in enough detail to use in a replication study is not clear, but given probable space limitations, Gilmore’s description is quite thorough. Of concern also are the control group activities. Some studies do not describe the control group well, but Gilmore provides a relatively clear discussion of its activities. Because of a concern for clear reporting in experimental studies, a repository for materials was created. The digital repository of Instruments and Materials for Research into Second Language (https://www.iris-database .org/iris/app/home/index), usually referred to as IRIS, was created so that researchers could share materials. Gilmore (2011) uploaded his assessment materials to IRIS so they can be easily found by other researchers. In addition, given the importance of replication, many journals now have space on their websites for researchers to include materials used in their studies. Gilmore also included his assessment materials on the Language Learning website.

6.3.5  Ethical Concerns In order to address ethical issues in research, some journals, such as TESOL Quarterly, include ethical guidelines (Mahboob et al., 2016). However, we know of no particular guidelines for classroom-based L2 research, though there are numerous ethical concerns, and Gass and Sterling’s (2017) discussion of ethics in instructed SLA are quite relevant here. For example, the control treatments need to be ethical. The experimental group should not have more class time and more experienced teachers than the control group. Furthermore, the experimental group might have what is considered a better treatment. Gilmore (2011) proposes that authentic materials are better, so it may, at first, seem unethical to withhold the materials from the control group. However, we cannot be sure of the authentic materials’ benefits unless a study is done. In the same vein, it is possible that the experimental group was exposed to a risky or unattested teaching practice. Another way to think about this situation is to consider with what one is challenging the status quo. If we can justify the hypothesis that authentic materials are better, we can argue that without the study, the control group would use the current materials and thus would not be disadvantaged. Other related concerns for classroom research are confidentiality and whether the experimental test outcomes affect students’ grades. Students

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

147

need to consent to share the results of any part of the study with their instructor. Conversely, they have the right to decide whether or not to share any class grades with the researcher. Gass and Sterling (2017) also mention the possibility of researcher bias. In some studies, such as (Gilmore, 2011), the researcher may be the teacher and treat students differently, even if unconsciously. Gilmore specifically addresses this concern, as mentioned earlier. Finally, researchers should work with teachers before proposing any study. It is unethical to force an instructor to take on a treatment that he or she thinks will not be effective or that will take away time from the current curriculum that focuses on goals related to how the students are assessed.

6.4  Review of Recent Studies 6.4.1  Search Process and Resulting Table In order to present an overview of experimental classroom research conducted over the last five years, we conducted a manual search of three journals: Language Teaching Research, the Modern Language Journal, and TESOL Quarterly. Searches of Linguistics and Language Behavior Abstracts using key words did not result in capturing the range of published experimental studies. The three journals were chosen because they publish empirical research related to language learning and teaching. Some journals that might include classroom research, such as Language Learning, where Gilmore’s (2011) study was published, focus more on laboratory research, while others, such as ELT Journal, which do publish some research studies, focus more on pedagogy. We felt that the three chosen journals would have the greatest numbers of experimental classroom studies. In addition, TESOL Quarterly publishes exclusively on the learning and teaching of English, while The Modern Language Journal strives to include a range of languages, and Language Teaching Research says that they welcome articles on languages other than English. The goal of this section is to provide novice researchers with an understanding of which variables have been studied in the classroom, and to provide all researchers with a list of studies that they can consult when doing similar research. Although we do try to generalize about the nature of experimental classroom research, the list of studies is not at all intended to be exhaustive and is most certainly biased by the journals that we chose to examine. As stated earlier, defining experimental classroom research is not straightforward, but we chose certain criteria when creating the list, and these criteria were related to the nature of the independent variable or intervention. The intervention had to be done in a real classroom, but could

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

148

C h a r le n e P o l i o a n d J o n g b o n g L ee

include a school computer lab. Furthermore, the focus had to include the teacher (or guest instructor) administering the intervention. For example, in Gilmore (2011), the focus was on the materials but involved the teacher teaching from those materials. We excluded studies such as Boers, Dang, and Strong (2017) in which teachers were given vocabulary materials to distribute but did not have to do anything with the materials. We also excluded studies such as Kim and Taguchi’s (2015) pragmatics study in which the independent variable was related to different task features as opposed to the implementation of the tasks. The end result of our search was a list of thirty studies shown in Table 6.1. We organized the studies by the independent variable and not the dependent variable. For example, the focus of a study might fall under general instruction technique, such as content and language integrated learning (CLIL) versus non-CLIL, but only one narrow dependent variable, such as formulaic sequences (Serrano, Stengers, & Housen, 2015), was measured. (Of course, this raises the issue of reductionism that we addressed earlier, namely whether or not examining only one dependent variable is sufficient to understand a complex phenomenon.) In addition to the independent and dependent variables, we included information about the context of the study and the design, as well as a column with additional comments.

6.4.2  Contexts for Classroom Research Not surprisingly, most of the research was conducted in English classes with nineteen studies done in EFL classes and six in ESL. The remaining five studies were conducted in Chinese, Japanese, French, and Spanish language classes. Not surprisingly, twenty of the thirty studies were conducted in a university setting, while seven were done in a secondary school setting and three in some other adult-learning context. We thus need to be careful about claims related to classroom teaching when so few studies were completed in K-12 settings or non-traditional adult language teaching contexts. The reason for most studies addressing English can be explained by the fact that we examined only English language journals, and researchers from non-English contexts who publish in English are likely to teach English. Similarly, most researchers teach at universities, and this probably accounts for the fact that so many of the studies were done in a university setting. In addition, university instructors tend to have more control over their own curriculum and can generally accommodate researchers or conduct research themselves. Furthermore, getting institutional review board (IRB) approval for research with minors can be challenging and might discourage some researchers. But we also note that we did not

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Context

Private language school in Brazil

Hung (2017) TESOL Quarterly

CLIL

Drama-based English instruction

Taiwanese university Flipped versus EFL class conventional classroom

Heras & Lasagabaster Spanish-Basque (2015) speaking Language Teaching secondary school Research English as a foreign language (EFL) students in Spain

Galante & Thomson (2017) TESOL Quarterly

Overall instructional approach

Authors and journals

Independent variable (Treatment)

Table 6.1.  Recent experimental classroom studies.

Achievement and attitude

Affective factors and vocabulary learning

Fluency, comprehensibility, and accentedness

Dependent variable (Construct) Design

End-of-term exam that included vocabulary, listening, and speaking Survey on attitudes towards class and learning

Questionnaire and receptive and productive vocabulary tests

Each class had a different teacher to minimize teacher effect

Comments

Quasi-experimental: Researcher taught both intact classes classes. Longitudinal: eighteen Data was triangulated weeks of instruction with research reflections and observations

Quasi-experimental: Delayed post-test was intact CLIL and nonused. CLIL classes Researchers noted the Longitudinal: twelve problem of selfone-hour sessions selection into CLIL of physical classes education taught in English

Untrained raters Quasi-experimental: assessed speech intact classes samples on a Longitudinal: variety of tasks four months of using a Likert scale instruction

Measure of dependent variable

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Taiwanese university Data-driven EFL students in learning grammar classes approach versus a traditional classroom

University-level Flipped versus students studying traditional Spanish in the US classroom

Beginning Japanese Online versus language students face-to-face at a US university instruction

University Spanish/ Catalan-speaking EFL students in Spain

Lin (2016) TESOL Quarterly

Moranski & Kim (2016) The Modern Language Journal

Sato, Chen, & Jourdain (2017) The Modern Language Journal

Serrano, Stengers, & Housen (2015) Language Teaching Research

Oral proficiency and learners’ perceptions of oral proficiency

Learning of Spanish se and student attitude

Student learning motivation and selfefficacy regarding grammar

Dependent variable (Construct)

Time allocation Acquisition of formulaic Proficiency level sequences

Context

Independent variable (Treatment)

Authors and journals

Table 6.1.  (cont.)

Quasi-experimental: intact classes Longitudinal: three sessions over ten weeks

Design

Quasi-experimental: self-selection into sections Longitudinal: fifteen weeks

The same teacher taught both sections The treatment conditions are well described

Follow-up interview with instructor to discuss Hawthorne effect possibilities Time on task was checked for the two conditions

The teachers were rotated throughout the treatment to avoid a teacher affect

Comments

Coding of formulaic Quasi-experimental: Both classes had similar structures in an oral intact classes curricula, but the narrative Longitudinal: the two intensive class have sections both had more audio-visual a total of 110 hours activities to make distributed over 4.5 them more engaging weeks or 7 months during the extended classes

Oral interviews and questionnaires

Test of metalinguistic Quasi-experimental: knowledge, a intact classes grammaticality Treatment was judgment test, and administered over production test three days

Questionnaire

Measure of dependent variable

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

University English as a second language (ESL) students at a US university

Secondary school The use of Romanian songs students studying French

Bardovi-Harlig, Mossman, & Vellenga (2015) Language Teaching Research

Dolean (2016) Language Teaching Research

Instruction of pragmatic routines

University EFL students in Japan

Aubrey (2017b) Language Teaching Research

Intercultural contact

US middle and high Sheltered school content instruction classes observation protocol (SIOP).

Specific instructional technique: oral skills

Short, Fidelman, & Louguit (2012) TESOL Quarterly

Anxiety

Use of pragmatic routines

Flow and engagement

Reading, writing, and oral proficiency

Quasi-experimental: two school districts participated in the experimental condition and one in the control Longitudinal: two years

The authors described in detail how the three districts compared on various features Classroom observations were included

Quasi-experimental: intact classes Short term: four fiftyminute lessons

An adapted version of Quasi-experimental: Foreign Language intact classes Classroom Anxiety Short term: five Scale lessons over five weeks

Use of targeted expressions in a simulated conversation

The same teacher taught the four classes The experimental and control groups were from different schools

Flow: a questionnaire Quasi-experimental: The author compared Engagement: number intact classes the two groups on a of words and turns Short term: treatment number of variables administered over to show that they five weeks were similar Author addresses ecological validity

State-mandated English test

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Japanese university EFL students

Secondary Chinese EFL students

Japanese university EFL students

Hamada (2016) Language Teaching Research

Li (2012) TESOL Quarterly

Saito & Saito (2017) Language Teaching Research

Tavakoli, Campbell, & UK university ESL McCormack (2016) students TESOL Quarterly

Context

Authors and journals

Table 6.1.  (cont.)

Dependent variable (Construct)

Acquisition of request supportive moves

Awarenessraising and fluency strategy training

Fluency, accuracy, and complexity of oral speech

Design

A discoursecompletion test

A large number of measures

The researcher taught the lessons

The same test was used for the pre- and post-test

Comments

Quasi-experimental: intact classes with nationalities balanced Longitudinal: treatment administered over four weeks

Quasi-experimental: Both classes had the intact classes same teacher Short term: three hours of instruction over six weeks

Quasi-experimental: intact classes Short term: three forty-five-minute lessons over three weeks

Dictation-cloze test Quasi-experimental: Multiple choice no control group listening Short-term: comprehension test shadowing lessons were twice a week for a month

Measure of dependent variable

Suprasegmental Comprehensibility, word A reading aloud instruction stress, rhythm, and task evaluated on intonation comprehensibility and an acoustic analysis of the speech samples.

Pragmatic instruction

Shadowing Phoneme perception Proficiency level Listening comprehension

Independent variable (Treatment)

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

University EFL students in Taiwan

Sun (2017) Language Teaching Research

Listening skills and strategy use

Amount of verbal communication Learner investment

Variations of a picture book reading task

Vocabulary learning

Learning of twenty target words

Using visual Learning of vocabulary imagery to teach abstract vocabulary

Tian & Macaro (2012) University EFL Lexical focus Language Teaching students in China on form, Research language of focus on form, proficiency level

US university students in Spanish classes

Farley, Ramonda, & Liu (2012) Language Teaching Research

Specific instructional technique: vocabulary

Taiwanese university Interactive EFL students versus strategies approach to teaching listening

Yeldham (2016) TESOL Quarterly

Form-focused versus meaningfocused instruction

Japanese high school EFL students

Tomita & Spada (2013) The Modern Language Journal

A receptive vocabulary test

Vocabulary Knowledge Scale (Paribakht & Wesche, 1997)

Receptive vocabulary test

A general listening proficiency test. Test of bottom-up listening skills. Questionnaire about strategies.

Number of turns

Experimental: randomization by proficiency level Short term: 1.5 hours a week for six weeks

Quasi-experimental: intact classes Short term: three 100-minute sessions over three weeks

Quasi-experimental: intact classes One brief treatment session

Quasi-experimental: intact classes Longitudinal: twentytwo hours of instruction

A delayed post-test was used The same teacher taught in all conditions

A delayed post-test was included

Delayed post-test was included

Experimental: random The study also included assignment a qualitative analysis Four fifty-minute of the students’ treatment sessions interaction and of interviews

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Context

Canadian university ESL classes

Korean middle Pre-task school EFL classes modelling

Kartchava & Ammar (2014b) Language Teaching Research

Kim (2013) TESOL Quarterly

Grammar learning and noticing

Dependent variable (Construct)

Attention to question forms and development of question forms

Type of oral Grammar learning and corrective noticing feedback and grammatical structure

Canadian university ESL classes

Type of oral corrective feedback and learner beliefs

Independent variable (Treatment)

Kartchava & Ammar (2014a) TESOL Quarterly

Specific instructional technique: grammar

Authors and journals

Table 6.1.  (cont.)

Design

Focus on questions during a thinkaloud planning session Question use on an oral production task

Quasi-experimental: intact classes Longitudinal: treatment administered over five weeks

Learners’ use of past Quasi-experimental: tense and question intact classes formation on a Short-term: two variety of oral tasks 120-minute Reflection sheets treatment sessions

Learners’ use of past Quasi-experimental: tense and question intact classes formation on a Short-term: two variety of oral tasks 120-minute Reflection sheets treatment sessions

Measure of dependent variable

Included a delayed post-test

There is no detailed analysis of what actually happened in the classroom No delayed post-test

There is no detailed analysis of what actually happened in the classroom No delayed post-test

Comments

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

US university Grouping of students studying Chinese Chinese characters based on radicals

Xu, Chang, & Perfetti (2014) The Modern Language Journal

Reading strategy instruction

EFL Iranian adult learners at a private language school

Karimi (2015) The Modern Language Journal

Specific instructional technique: reading

Community-based ESL classes for adults in Canada

Spada et al. (2014) Language Teaching Research

Learning of Chinese characters

Comprehension of multiple documents Self-reported use of strategies

Integrated Learning of passive versus construction isolated form-focused instruction

Adult Iranian Corrective The use of articles students studying feedback in EFL as part the form of of their job recasts and requirements cognitive style

Rassaei (2015) Language Teaching Research

Lexical decision, sound matching, and meaning matching tasks and a written production task

A true–false test that involved inference across texts A questionnaire

Experimental: appears Included a delayed to be random post-test assignment Each treatment consisted of four learning sessions

Experimental: random Data triangulated using assignment think-aloud sessions Longitudinal: ten with some of the ninety-minute participants sessions

Independent coders evaluated materials for features of the independent variable, but no videoing was allowed. Teachers kept logs on how and if they deviated from the lesson. A delayed post-test was included

Quasi-experimental: Classes were observed intact classes for the presence of One treatment session recasts

Error-correction task Quasi-experimental: and oral production intact classes task Short term: twelve hours of class over three days

Article use on an oral and written production task

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Context

University EFL students in Sri Lanka

Vietnamese university EFL students

Iranian university EFL students

De Silva (2015) Language Teaching Research

Nguyen & Gu (2013) Language Teaching Research

Rahimi (2013) Language Teaching Research

Specific instructional technique: writing

Authors and journals

Table 6.1.  (cont.)

Dependent variable (Construct)

Feedback training

Strategy instruction

Types of feedback, writing quality, and incorporation of feedback

Writing quality and selfregulation strategies

Writing strategy Strategy use. instruction Writing performance.

Independent variable (Treatment)

Coding of student comments and performance on an essay

A written production test and a questionnaire

Questionnaire Two written tasks

Measure of dependent variable Comments

One of the researchers co-taught the experimental group and did three observations of the two control groups A delayed writing test was included Experimental: random The treatment was assignment added, and there was Two training sessions no activity for the control group

Quasi-experimental: intact classes Longitudinal: nine sessions over eight weeks

Experimental: random It is not clear assignment that classroom Longitudinal: observations were instruction was over done twenty-four weeks

Design

Experimental Studies in L2 Classrooms

157

search education or bilingual education journals that tend to focus on young or adolescent learners. It was interesting to see that the majority of studies were conducted in EFL classes. In general, many EFL classes are larger than ESL classes, so researchers can have a greater number of participants. Interestingly, one could argue that research in an EFL setting may have fewer threats to internal validity because students are less likely to use English outside of class on a regular basis. While it may or may not be true, Hamada (2016) explicitly stated that she assumed that the students in her study had no exposure to English outside the classroom.

6.4.3 Focus of Classroom Research: Independent and Dependent Variables The second column in Table 6.1 lists the independent variable for the study. The first eight studies test some kind of broad instructional approach, including drama-based instruction, CLIL, flipped or online instruction, data-driven learning, and sheltered instruction. In each case, the teaching approach under investigation in the experimental group was compared to a control group that received instruction that was the status quo in the particular institution. This group also includes a study (Serrano, Stengers, & Housen, 2015) that compares two groups based on the distribution of instruction. In other words, both groups received the same type and amount of instruction, but in one case, the instruction was distributed over 4.5 weeks and in the other over seven months. The next ten studies focus on the teaching of listening and speaking by testing a variety of specific interventions such as intercultural contact (i.e., contact with native speakers of the L2), suprasegmental instruction, the teaching of pragmatic routines, and even shadowing (i.e., repeating language while listening). In all cases except the shadowing study (Hamada, 2016), a control group who received regular instruction was used for comparison. Tomita and Spada’s (2013) was a repeated measures study, discussed further below, where all students participated in both conditions. They compared students in focus on meaning and focus on form conditions. Despite a large number of studies in the field that examine vocabulary learning and instruction, we found only three studies that we considered to be classroom research. Many studies of vocabulary learning study materials or activities that can be completed by students without their teachers (e.g., Laufer & Rozovski-Roitblat, 2015). We found four that focused on grammar. There were so few studies most likely because studies of learning specific grammar points are often done in a lab setting. Finally, we found two studies about reading instruction and three about writing. Of these five studies, three examined some type of strategy instruction.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

158

C h a r le n e P o l i o a n d J o n g b o n g L ee

We also included a column showing what the dependent variable and outcome measure were for each study. Twenty-six of the thirty studies measured some type of learning as evidenced by a general proficiency test or the learning of specific vocabulary or grammatical items. The focus of the learning measures varies considerably and includes constructs such as general oral or written proficiency, fluency, comprehensibility, the acquisition of formulaic sequences or specific grammatical items, phoneme perception, the complexity of oral speech, and so on. Four studies measured only some other type of outcome, including flow and engagement (Aubrey, 2017a), anxiety (Dolean, 2016), self-efficacy (Lin, 2016), and amount of verbal communication (Tomita & Spada, 2013). Five of the twenty-six studies that examined learning also included some other type of dependent variable such as learner attitude (Hung, 2017) or strategy use (Nguyen & Gu, 2013). For these non-learning related outcomes, usually questionnaires were used, but Aubrey and Tomita and Spada also used the number of turns in oral interaction as a measure. Of the twenty-six studies that measured a learning outcome, eight used a delayed post-test. Delayed post-tests are used to see if any effects of the independent variable last for any length of time after the treatment is complete. They are often used in vocabulary research as the goal of vocabulary teaching is the retention of new words, and, indeed, all three vocabulary studies in Table 6.1 included delayed post-tests. Heras and Lasagabaster (2015) studied the effects of CLIL and also included a delayed post-test. The other studies that used delayed post-tests measured grammar learning outcomes (Spada et al., 2014), Chinese character learning (Xu, Chang, & Perfetti, 2014), and writing performance (Nguyen & Gu, 2013). In general, including a delayed post-test when studying any learning outcome is good practice.

6.4.4 Design Only six of the thirty studies randomly assigned participants to groups and can thus be considered true experimental studies. The lack of random assignment in classroom research is quite common because of the logistic problems in randomly assigning students to different classes. Only one of the studies (Tomita & Spada, 2013) was a repeated measures study. In a repeated measures study, students first participate in one intervention and then another, and the effects of the intervention are compared for each student. For this type of design to be possible, the first intervention should have no lasting effect, otherwise the research would not be able to determine the source of the cause. Tomita and Spada were able to use a repeated measures design because their study was one of the few that did not examine learning. Instead, they were interested in how formversus meaning-focused activities affected learner engagement, and they

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

159

examined the number of student turns in an interaction as a measure of engagement. The presumption is that learners will start with a clean slate during the second activity. The order of activities needs to be counterbalanced in any repeated measures study because there is always a possibility of the first activity affecting performance on the second, even though that effect should not be strong or long-lasting. The twenty-four remaining studies can be considered quasi-experimental because intact classes were used. Only one of these studies, Hamada’s (2016), used no control group. Both types of quasi-experimental studies can be rigorously designed. As discussed earlier, when there is no random assignment, threats to internal validity increase because outside variables related to how students are placed in a class, for example, can influence the outcome. The use of pre-test in these cases can help show that groups from intact classes are similar, and, in fact, most of the studies used a pretest to analyse gains. Other ways that the studies addressed the problem of intact classes included using the same teacher for each group (Sato, Chen, & Jourdain, 2017) or rotating the teachers (Lin, 2016). Aubrey (2017a) compared his groups on a number of variables to show how they were similar. Quasi-experimental studies without a control group should include qualitative data to help triangulate the results of the pre-test–post-test design. The Hamada (2016) study was particularly problematic because not only was there no control, but no qualitative data were reported and the same test was used for the pre- and post-tests, so there could have been a practice effect from students having taken the same test a month earlier. The median sample size for the studies was sixty-seven across all groups. One study, that of Short, Fidelman, and Louguit (2012) was an outlier with their cross-school-district study of 1,444 students. Only one of the thirty studies used no inferential statistics. Most of the studies used parametric tests, such as ANOVA or t-tests, to test for group differences, but a few used non-parametric tests. What was surprising was that only two studies used ANCOVA, which can account for initial group differences. Moranski and Kim (2016) added a multiple regression to better understand what might predict student attitudes and Spada et al. (2014, pp. 460–461) used multi-level modelling (MLM). They give a clear explanation as to why they used this technique: MLM has several advantages over other statistical procedures traditionally used in instructed SLA research. First, it allows for the analysis of individual learner development by examining nested data (e.g., pre, post and delayed tests within one participant), rather than analysing aggregated mean points. Second, MLM takes into consideration pretest differences as a covariate. Finally, MLM tolerates missing data because treating multiple measurements as nested data makes it possible to analyse all the data regardless of the number of measurement points (Barkaoui, 2010;

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

160

C h a r le n e P o l i o a n d J o n g b o n g L ee

Luke, 2004; Tabachnick & Fidell, 2007). This is particularly advantageous in classroom research where attrition rates can be high as experienced in this study.

MLM is certainly something that future longitudinal studies should consider. In terms of length, the treatment periods ran from one single session up to two years with most being what we might call short-term studies with periods of treatment running from two to eight weeks, but seven of the studies ran for four months or more. It is not necessarily the case that longer is better in experimental research, but it was good to see that most studies went beyond one class session.

6.5 Conclusion We have presented a summary of the methodological issues related to conducting experimental classroom research. In contrast to De Costa, Li, and Rawal (this volume, Chapter 5), we did not discuss the range of theoretical frameworks. This was because the range of topics covered in the studies did not suggest any type of unified framework. Generally speaking, the frameworks tended to be more cognitively oriented than socially, but see Tomita and Spada (2013) for an exception. Therefore, we will not discuss where the field should go in terms of topics or research questions, but rather end with a list of suggestions about how to move forward methodologically. First, replication is challenging, if not impossible, when it comes to classroom research. Nevertheless, researchers need detailed reporting of their interventions and assessments. By uploading materials to IRIS, researchers can encourage replication. Given the dearth of research on languages other than English, replication in non-English classrooms should be done. Replication would also expand the generalizability of the studies to ensure greater external validity. Replication of laboratory studies in the classroom context, such as those done by Gass, Mackey, and Ross-Feldman (2005), would increase the ecological validity of those lab-based studies to show that the interventions are practical and usable. Second, researchers might want to consider more sophisticated statistical techniques that can be used with longitudinal data. Most studies reviewed here included both a pre- and post-test, yet only one used MLM. Barkaoui (2014) reviews analyses that can be used with longitudinal data, including MLM, and this review is a useful starting point. We also encourage random assignment of participants where possible, but given that it may not be possible, more sophisticated statistics can help account for initial group differences.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

161

Finally, in future research, more qualitative data should be used in conjunction with quantitative data. At the very least, observations of the interventions should be included as was done in Rassaei (2015) to ensure that the students were receiving recasts. Most studies did not include observations, but some included other qualitative data. Hung (2017), for example, taught both the experimental and control groups in her study and collected reflection data on her teaching. Although this was a good start, she did not systematically report the data. These studies used qualitative data as a check, but other studies, not included in the scope of our search, can serve as examples of mixed methods studies that collect and integrate a range of quantitative and qualitative data. Yasuda (2011), for example, studied genre-based instruction and included both text analyses and interview data in her study. Although well-designed mixed methods studies can be challenging (Bryman, 2007), future studies might better integrate quantitative and qualitative data. Classroom research is without a doubt an area that would benefit from the increased attention in applied linguistics to mixed methods research (e.g., Brown, 2014; Dörnyei, 2007; Polio & Friedman, 2017).

References Asher, J. J. (1972). Children’s first language as a model for second language learning. The Modern Language Journal, 56, 133–139. Aubrey, S. (2017a). Measuring flow in the EFL classroom: Learners’ perceptions of inter- and intra-cultural task-based interactions. TESOL Quarterly, 51(3), 661–692. Aubrey, S. (2017b). Inter-cultural contact and flow in a task-based Japanese EFL classroom. Language Teaching Research, 21(6), 717–734. Bardovi-Harlig, K., Mossman, S., & Vellenga, H. E. (2015). The effect of instruction on pragmatic routines in academic discussion. Language Teaching Research, 19(3), 324–350. Barkaoui, K. (2010). Explaining ESL essay holistic scores: A multilevel modeling approach. Language Testing, 27, 515–535. Barkaoui, K. (2014). Quantitative approaches for analyzing longitudinal data in second language research. Annual Review of Applied Linguistics, 34, 65–101. Beretta, A. (1986). Program-fair language teaching evaluation. TESOL Quarterly, 20(3), 431–444. Boers, F., Dang, T. C. T., & Strong, B. (2017). Comparing the effectiveness of phrase-focused exercises: A partial replication of Boers, Demecheleer, Coxhead, and Webb (2014). Language Teaching Research, 21(3), 362–380. Brown, J. (2014). Mixed methods research for TESOL. Edinburgh: Edinburgh University Press.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

162

C h a r le n e P o l i o a n d J o n g b o n g L ee

Bryman, A. (2007). Barriers to integrating quantitative and qualitative research. Journal of Mixed Methods Research, 1(1), 8–22. Chaudron, C. (1988). Second language classrooms: Research on teaching and learning. Cambridge: Cambridge University Press. Collins, L., & Muñoz, C. (2016). The foreign language classroom: Current perspectives and future considerations. The Modern Language Journal, 100(S1), 133–147. De Costa, P., Valmori, L., & Choi, I. (2017). Qualitative research methods. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 541–561). New York: Routledge. De Silva, R. (2015). Writing strategy instruction: Its impact on writing in a second language for academic purposes. Language Teaching Research, 19(3), 301–323. Dolean, D. D. (2016). The effects of teaching songs during foreign language classes on students’ foreign language anxiety. Language Teaching Research, 20(5), 638–653. Dörnyei, Z. (2007). Research methods in applied linguistics: Quantitative, qualitative, and mixed methodologies. Oxford: Oxford University Press. Douglas Fir Group (2016). A transdisciplinary framework for SLA in a multilingual world. The Modern Language Journal, 100, 19–47. Farley, A. P., Ramonda, K., & Liu, X. (2012). The concreteness effect and the bilingual lexicon: The impact of visual stimuli attachment on meaning recall of abstract L2 words. Language Teaching Research, 16(4), 449–466. Galante, A., & Thomson, R. I. (2017). The effectiveness of drama as an instructional approach for the development of second language oral fluency, comprehensibility, and accentedness. TESOL Quarterly, 51(1), 115–142. Gass, S. (2015). Experimental research. In B. Paltridge & A. Phakiti (eds.), Research methods in applied linguistics: A practical resource (pp. 101–118). New York: Bloomsbury Publishing. Gass, S., Mackey, A., & Ross-Feldman, L. (2005). Task-based interactions in classroom and laboratory settings. Language Learning, 55, 575–611. Gass, S., & Sterling, S. (2017). Ethics in ISLA. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 577– 595). New York: Routledge. Gilmore, A. (2008). Getting real in the language classroom: Developing Japanese students’ communicative competence with authentic materials. Unpublished PhD dissertation, Nottingham University. Gilmore, A. (2011). “I prefer not text”: Developing Japanese learners’ communicative competence with authentic materials. Language Learning, 61(3), 786–819. Hamada, Y. (2016). Shadowing: Who benefits and how? Uncovering a booming EFL teaching technique for listening comprehension. Language Teaching Research, 20(1), 35–52.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

163

Hashemi, M. R. (2012). Reflections on mixing methods in applied linguistics research. Applied Linguistics, 33, 206–212. Hashemi, M. R., & Babaii, E. (2013). Mixed methods research: Toward new research designs in applied linguistics. The Modern Language Journal, 97, 828–852. Heras, A., & Lasagabaster, D. (2015). The impact of CLIL on affective factors and vocabulary learning. Language Teaching Research, 19(1), 70–88. Hung, H. T. (2017). Design-based research: Redesign of an English language course using a flipped classroom approach. TESOL Quarterly, 51(1), 180–192. Jang, E. E., Wagner, M., & Park, G. (2014). Mixed methods research in language testing and assessment. Annual Review of Applied Linguistics, 34, 123–153. Karimi, M. N. (2015). EFL learners’ multiple documents literacy: Effects of a strategy-directed intervention program. The Modern Language Journal, 99(1), 40–56. Kartchava, E., & Ammar, A. (2014a). Learners’ beliefs as mediators of what is noticed and learned in the language classroom. TESOL Quarterly, 48(1), 86–109. Kartchava, E., & Ammar, A. (2014b). The noticeability and effectiveness of corrective feedback in relation to target type. Language Teaching Research, 18(4), 428–452. Kim, Y. (2013). Effects of pretask modeling on attention to form and question development. TESOL Quarterly, 47(1), 8–35. Kim, Y., & Taguchi, N. (2015). Promoting task-based pragmatics instruction in EFL classroom contexts: The role of task complexity. The Modern Language Journal, 99(4), 656–677. Larsen-Freeman, D., & Cameron, L. (2008). Research methodology on language development from a complex systems perspective. The Modern Language Journal, 92(2), 200–213. Laufer, B., & Rozovski-Roitblat, B. (2015). Retention of new words: Quantity of encounters, quality of task, and degree of knowledge. Language Teaching Research, 19(6), 687–711. Li, Q. (2012). Effects of instruction on adolescent beginners’ acquisition of request modification. TESOL Quarterly, 46(1), 30–55. Lin, M. H. (2016). Effects of corpus-aided language learning in the EFL grammar classroom: A case study of students’ learning attitudes and teachers’ perceptions in Taiwan. TESOL Quarterly, 50(4), 871–893. Loewen, S. (2015). Introduction to instructed second language acquisition. New York: Routledge. Long, M. H. (1980). Inside the “black box”: Methodological issues in classroom research on language learning. Language Learning, 30(1), 1–42. Luke, D. (2004). Multilevel modeling. Thousand Oaks, CA: Sage.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

164

C h a r le n e P o l i o a n d J o n g b o n g L ee

Mackey, A. (2017). Classroom-based research. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 541– 561). New York: Routledge. Mahboob, A., Paltridge, B., Phakiti, A., Wagner, E., Starfield, S., Burns, A., …, De Costa, P. I. (2016). TESOL Quarterly research guidelines. TESOL Quarterly, 50(1), 42–65. Moranski, K., & Kim, F. (2016). “Flipping” lessons in a multi-section Spanish course: Implications for assigning explicit grammar instruction outside of the classroom. The Modern Language Journal, 100(4), 830–852. Nguyen, L. T. C., & Gu, Y. (2013). Strategy-based instruction: A learnerfocused approach to developing learner autonomy. Language Teaching Research, 17(1), 9–30. Paribakht, T. S., & Wesche, M. (1997). Vocabulary enhancement activities and reading for meaning in second language vocabulary development. In J. Coady & T. Huckin (eds.), Second language vocabulary acquisition: A rationale for pedagogy (pp. 174–200). New York: Cambridge University Press. Polio, C., & Friedman, D. A. (2017). Understanding, evaluating, and conducting second language writing research. New York, NY: Routledge. Porte, G. (2012). Replication research in applied linguistics. Cambridge: Cambridge University Press. Rahimi, M. (2013). Is training student reviewers worth its while? A study of how training influences the quality of students’ feedback and writing. Language Teaching Research, 17(1), 67–89. Rassaei, E. (2015). Recasts, field dependence/independence cognitive style, and L2 development. Language Teaching Research, 19(4), 499–518. Riazi, A. M., & Candlin, C. N. (2014). Mixed-methods research in language teaching and learning: Opportunities, issues and challenges. Language Teaching, 47, 135–173. Saito, Y., & Saito, K. (2017). Differential effects of instruction on the development of second language comprehensibility, word stress, rhythm, and intonation: The case of inexperienced Japanese EFL learners. Language Teaching Research, 21(5), 589–608. Sato, E., Chen, J. C. C., & Jourdain, S. (2017). Integrating digital technology in an intensive, fully online college course for Japanese beginning learners: A standards-based, performance-driven approach. The Modern Language Journal, 101(4), 756–775. Scherer, G. A. C., & Wertheimer, M. (1964). A psycholinguistic experiment in foreign language teaching. New York: McGraw-Hill. Schmitt, N. (2000). Vocabulary in language teaching. Cambridge: Cambridge University Press. Serrano, R., Stengers, H., & Housen, A. (2015). Acquisition of formulaic sequences in intensive and regular EFL programmes. Language Teaching Research, 19(1), 89–106.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

Experimental Studies in L2 Classrooms

165

Short, D. J., Fidelman, C. G., & Louguit, M. (2012). Developing academic language in English language learners through sheltered instruction. TESOL Quarterly, 46(2), 334–361. Sonbul, S., & Schmitt, N. (2013). Explicit and implicit lexical knowledge: Acquisition of collocations under different input conditions. Language Learning, 63(1), 121–159. Spada, N., Jessop, L., Tomita, Y., Suzuki, W., & Valeo, A. (2014). Isolated and integrated form-focused instruction: Effects on different types of L2 knowledge. Language Teaching Research, 18(4), 453–473. Spinner, P., Gass, S. M., & Behney, J. (2013). Ecological validity in eye-tracking: An empirical study. Studies in Second Language Acquisition, 35(2), 389–415. Sun, C. H. (2017). The value of picture-book reading-based collaborative output activities for vocabulary retention. Language Teaching Research, 21(1), 96–117. Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. Boston, MA: Pearson Education. Tavakoli, P., Campbell, C., & McCormack, J. (2016). Development of speech fluency over a short period of time: Effects of pedagogic intervention. TESOL Quarterly, 50(2), 447–471. Tian, L., & Macaro, E. (2012). Comparing the effect of teacher codeswitching with English-only explanations on the vocabulary acquisition of Chinese university students: A lexical focus-on-form study. Language Teaching Research, 16(3), 367–391. Tomita, Y., & Spada, N. (2013). Form-focused instruction and learner investment in L2 communication. The Modern Language Journal, 97(3), 591–610. Wegener, D., & Blankenship, K. (2007). Ecological validity. In R. F. Baumeister & K. D. Vohs (eds.), Encyclopedia of social psychology (Vol. 1, pp. 275–276). Thousand Oaks, CA: SAGE Publications. Xu, Y., Chang, L. Y., & Perfetti, C. A. (2014). The effect of radical-based grouping in character learning in Chinese as a foreign language. The Modern Language Journal, 98(3), 773–793. Yasuda, S. (2011). Genre-based tasks in foreign language writing: Developing writers’ genre awareness, linguistic knowledge, and writing competence. Journal of Second Language Writing, 20(2), 111–133. Yeldham, M. (2016). Second language listening instruction: Comparing a strategies-based approach with an interactive, strategies/bottom-up skills approach. TESOL Quarterly, 50(2), 394–420.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:50:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.007

7 Action Research: Developments, Characteristics, and Future Directions Anne Burns 7.1 Introduction Action research has become increasingly prevalent in the field of English language teaching over the last two decades. It can be considered as a form of professional learning for language teachers which takes a socio-constructivist approach in which teachers are seen as agentive actors and investigators within their own social contexts. This chapter begins by providing a brief overview of the origins of action research and its development in the field of English language teaching, considering also some of the more recent initiatives that have contributed to the spread of this form of research in the field. It then discusses what various recent studies have shown about the impact on teachers of conducting action research. The discussion ends with a brief consideration of future directions for this research. In this chapter, I give preference to the term action research to refer to research carried out by practitioners in language educational contexts, and I focus primarily on research in this area. Overall, I eschew the term teacher research, which is sometimes used to describe such work in educational sites (e.g., Borg, 2013), since in my view it excludes the range of others (principals/administrators/school leaders, pre-service candidates and teacher educators, teaching colleagues/assistants/aides, technical staff, community-based supporters and parents) who might also be involved in the research and unnecessarily narrows the perspective on the potential for the scope of educational action research to lead to transformation and innovation. It also tends to suggest an individualistic, cognitively-based, and self-referential orientation towards the teacher that underplays the embeddedness of teaching within wider social, relational, and interpersonal networks.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

167

7.2  The Development of Educational Action Research The roots of action research go back to Aristotelian traditions of practical philosophy, in particular the concept of praxis, but it found prominence in more recent times in the work of the educator John Dewey, social psychologist, Kurt Lewin, and sociologist and philosopher, Jürgen Habermas. I have described the origins of action research in some detail elsewhere (Burns, 2011); therefore, in this section and the next, I provide just a brief account of some of the key figures in action research, particularly educational action research, and more recent developments in action research in the field of language teaching. Kurt Lewin is widely credited with the introduction of the term action research. Lewin, a refugee to the US from Nazi Germany, founded the Research Centre for Group Dynamics at the Michigan Institute of Technology (MIT), where his work on the social organization and operation of groups gave rise to new directions in social psychological research through “involving participants in a cyclical process of fact finding, planning, exploratory action and evaluation (Lewin, 1948, 202–6)” (Somekh & Zeichner, 2009, p. 7). By including those most directly concerned in the social action, Lewin contested “the separation of the investigation from the action needed to solve the problem” (McFarland & Stansell, 1993, p. 14). An early champion of educational action research in the US was Stephen Corey at Colombia University. Corey advocated “cooperative action research”, believing that: One of the psychological values in action research is that the people who must, by the very nature of their professional responsibilities, learn to improve their practices are the ones who engage in the research to learn what represents improvement. (Corey, 1954, p. 375)

By the mid-1950s, action research was under attack in the US, seen as unscientific and lacking in scholarly rigour. Its re-emergence in educational contexts came primarily through Lawrence Stenhouse, a historian whose work in the Humanities Curriculum Project in the UK was concerned with the “enacted curriculum” (i.e., both what was planned and what students learned). Stenhouse believed that curriculum development should emanate from researchers and teachers working together in joint exploration. He wanted teachers’ accounts of their classroom investigations to contribute to building educational curriculum theory: Each classroom should not be an island … teachers working to such a tradition should communicate with one another … they should report their work … a common vocabulary of concepts and a syntax of theory need to be developed … If teachers report their own work in such a tradition,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

168

Anne Burns

case studies will accumulate, just as they do in medicine … Professional research workers will have to master this material and scrutinize it for general trends. It is out of this synthetic task that general propositional theory can be developed. (Stenhouse, 1975, p. 157)

New influences on concepts of action research, that took it beyond its predominantly rationalistic and practical epistemologies, came from Carr and Kemmis (1986) in Australia, who repositioned action research within Habermasian critical theory, and incorporated ideological dimensions of emancipation, democracy, social justice, and resistance to oppresssion. These concepts echoed the liberationist and participatory educational goals of Paolo Freire in Brazil and Orlando Fals Borda in Colombia, and placed emphasis on social and dialectic ways of thinking and knowledge production. Carr and Kemmis’ exegesis of participatory and critical educational action research led to their often-cited definition: Action research is simply a form of self-reflective inquiry undertaken by participants in social situations in order to improve the rationality and justice of their own practices, their understanding of these practices, and the situations in which the practices are carried out.  (Carr and Kemmis, 1986, p. 162)

They argued that action research has two essential aims: to improve and to involve. Improvement relates to improvements in practice, understanding of the practice, and of the situation in which the action takes place. Involvement means that the actors in the situation should be involved in all the cyclical phases of planning, acting, observing, and reflecting.

7.3 The Development of Action Research in Language Teaching The concepts of action research in English language teaching (ELT) began to appear from around the mid-1980s. Here I trace its development over two time periods.

7.3.1  Early Developments: Late 1980s–2000 Although by the late 1980s, action research had still “not so far received much serious attention as a distinct style of research in language teaching” (van Lier, 1988, p. 67), the idea of teacher education based on teacher inquiry and problem solving was beginning to enter the literature (e.g., Breen et al. 1989; Brumfit & Mitchell, 1989; Fanselow, 1987; Nunan, 1987; Ramani, 1987; Roberts, 1986). Nunan (1989) was the first to provide language teachers with explicit guidance about researching their teaching contexts. Drawing mainly on traditional methods for conducting

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

169

classroom research, Nunan spoke directly to an audience of teachers and teacher educators, and introduced them to the key idea that “in order for language teachers to understand the classroom in which they work, they need to systematically observe and investigate these classrooms” (p. 15). Other publications foregrounding teacher reflective practice and inquiry as a part of teacher education quickly followed throughout the 1990s, including those oriented to teacher classroom research and “exploratory teaching” (Allwright & Bailey, 1991), “reflective” teaching (Richards & Lockhart, 1994; Roberts, 1998; Wallace, 1991), teacher research (Edge & Richards, 1994; Freeman, 1998; McDonough & McDonough, 1997), teacher learning (Freeman & Richards, 1996), action research (Wallace, 1998), and collaborative action research (Burns, 1999). However, despite the upsurge in volumes for teachers and graduate students conducting research in the field of ELT and applied linguistics since 2000 (e.g., Holliday, 2001; McKay, 2006; Richards, 2003; Richards, Ross, & Seedhouse, 2012; Paltridge & Phatiki, 2015), surprisingly there are still very few volumes that focus on guiding language teachers through the processes of conducting action research specifically. Exceptions are Wallace (1998), Burns (1999, 2010), and more recently Dikilitaş & Griffiths (2017). Various players and waves of activity by agencies in different geographical locations have served to propel the action research movement in language teaching. Burns (1999), for example, drew on extensive professional development initiatives employing action research in Australia’s Adult Migrant English Program (AMEP) from the early 1990s. This work emanated from earlier research by Nunan (e.g., 1987) and an agenda for research (Brindley, 1990) developed within the Australian Government’s National Centre for English Language Teaching and Research, based at Macquarie University in Sydney, which envisaged “practitioner research” as an important component of the centre’s work with the AMEP. Burns and her colleagues (respectively Hammond, Hood, and de Silva Joyce) conducted ten national action research projects with AMEP teachers from around the country over an almost twenty-year period from 1990 to 2008. The focus for these projects was derived from national feedback from AMEP service providers about curriculum development and teaching issues that were seen as widely relevant to teacher development and classroom practice (see Burns, 2016, for a more extended discussion of this work). In the UK, Edge and Richards at Aston University took a leading role in promoting research by teachers, a concept which had gained increasing currency within the Teacher Development Special Interest Group (TD SIG) of the International Association of Teachers of English as a Foreign Language (IATEFL). Based on their growing perceptions of the “overlap between teacher development and teacher research” (Edge & Richards, 1994, p. 6), they initiated a collaborative venture between the TD SIG

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

170

Anne Burns

and the more recently established Research SIG to organize a conference at Aston that “would attract some attention” (p. 6) to these linkages. Adopting the title “Teachers Develop Teachers’ Research” (TDTR) that aptly united the two concepts, this event initiated a series of other TDTR conferences held at first in the UK (e.g., Cambridge in 1995), but then also overseas through IATEFL affiliates (e.g., in Israel in 1997 and Chile in 1999). The conferences provided a singular opportunity for those conducting action research and other related types of teacher inquiry in the field of ELT to present and publish their work (e.g. Edge & Richards, 1994; Field et al., 1997). In India, Mathew, Eapen, and Tharu (2000), from the then Central Institute of English and Foreign Languages (CIEFL), were involved in a national Curriculum Implementation Study from 1993 to 1997, which eventually involved 250 teachers and 800 secondary schools. They saw the concept of the teacher-as-researcher as central to successful implementation of the project. They introduced the teachers to research techniques and strategies in classroom investigation and data-gathering and used the teachers’ input in developing and implementing the curriculum. They also supported action research mini-studies by fifty of these teachers. Meanwhile in Colombia, in the late 1990s, C árdenas from the PROFILE Research Group in the Departamento de Lenguas Extranjeras, at the Universidad Nacional de Colombia, was promoting the concept of action research by teachers through the establishment of a journal. Launched in 2000, PROFILE Issues in Teachers’ Professional Development provided (and still does) a publication forum, initially for teachers who had been introduced through the university’s professional development programmes to concepts of action research, but later for teacher researchers and teacher educators from a wide range of international backgrounds. The journal was unique in its concern to “share results of classroom research projects, reflections and innovations undertaken by teachers of English as a second or foreign language … teacher educators and novice teacher-researchers” and to address itself mainly “to an international readership of pre- and in-service teachers” (see PROFILE homepage: https://revistas.unal.edu.co/ index.php/profile).

7.3.2  More Recent Developments: 2000 to the Present A further impetus to practitioner action research was provided by the International TESOL organization. Although the TESOL Research Interest section had been disbanded in the mid-2000s, to the consternation of several of its members, the body that replaced it, the TESOL Standing Committee for Research (now the TESOL Research Professional Council), which was seen as an umbrella committee that crossed the range of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

171

Interest sections in TESOL, began to offer pre-convention workshops for teacher researchers (e.g., Borg & Burns, 2008), which have continued to this time. TESOL was also promoting research by teachers through publications such as The Language Teacher Research Series, edited by Thomas Farrell (2006–2009), which highlighted projects conducted in Asia, the Americas, Europe, Australasia, the Middle East, and Africa. The series was described as “a relatively new stance toward educational research”, aimed at encouraging teachers to take “an inquiry stance toward language teaching” (TESOL, 2004, p. 14). While the number of other projects involving action research initiated both by individual teachers and educational organizations slowly increased in various parts of the world across the decade, from the late 2000s there seems to have been an upsurge in action research initiatives funded by major international bodies. In Australia, the Action Research in ELICOS (English Language Courses for Overseas Students) Program instigated by the professional body, English Australia, was funded by Cambridge English Language Assessment (now Cambridge Assessment English). Starting as a pilot programme in 2010, it is now approaching its ninth year (see Burns & Khalifa, 2017 for an account of its development and examples of participating teachers’ action research; see also Research Notes, volumes 44, 48, 53, 56, 60, 64, 67, at http://www .cambridgeenglish.org/research-notes/). From 2014, this programme was replicated in the UK by Cambridge English in collaboration with English UK as an Action Research Award Scheme (see Borg, 2015). Both these programmes provide opportunities for teachers of international students in each country to undertake a programme of action research across a year. Since around 2010, other initiatives have been funded and supported by The British Council. To give some examples, from 2012 to 2016 The British Council sponsored the English Language Teaching Research Partnerships (ELTReP) in India, which resulted in thirty-three publications published in the series Explorations: Teaching and Learning English (see https://www.britishcouncil.in/programmes/english-partnerships/ research-policy-dialogues/eltrep-papers). In conjunction with the AllIndia Network of English Teachers (AINET), The British Council continues to host the Aptis Action Research Mentor Scheme (AARMS) in India. From 2013, The British Council and Ministry of Education in Chile instigated the Champion Teachers programme which provides support for secondary school teachers to engage in exploratory action research projects (see Rebolledo, Smith, & Bullock, n.d.). The Action Research Award Scheme (ARAS), which is sponsored by The British Council Aptis, also aims to support the development of teacher-research in English language education in Latin-America more generally. In Singapore The

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

172

Anne Burns

British Council’s Teacher Development Centre (TDC) has recently begun developing and supporting Action Research projects with education partners there, for example a project on critical literacy with a local primary school (see https://www.britishcouncil.sg/about/press/newsletter-articles/ teacher-development/action-research-singapore). Cambridge University Press has also entered the scene of practitioner research, launching its own Teacher Research Programme in 2014 to further “the development of teachers and teaching around the world, and … to encourage teacher research projects by offering financial and professional support to a number of projects every year” (see http://languageresearch.cambridge.org/). In addition, IATEFL, through its Research SIG, has been very active over the last few years in promoting research by teachers both through its pre-conference events, its newsletter ELT Research, and its web discussions (e.g., Wyatt, Burns, & Hanks, 2016). Employing the title Teachers Research!, the 2014 pre-conference event aimed to “revive the spirit of the TDTR [Teachers Develop Teachers Research] conferences started off in 1992 at Aston University” (http://resig.weebly.com/background-to-teachersresearch.html; see also Bullock & Smith, 2015). From 2015, IATEFL also began to support a series of Teacher Research! conferences in Turkey, building on the work of Kenan Dikilitaş and his colleagues in promoting action research and other forms of research by teachers in that country (see, for example, Dikilitaş, Smith, & Trotman, 2015, Dikilitaş et al., 2016; Burns et al., 2017). A more recent initiative that is aiming to unite and promote research by teachers globally is the International Festival of Teacher-Research in ELT. Initiated in 2017 by Richard Smith from Warwick University, former Chair of the IATEFL Research SIG, it aims to “co-brand” international research initiatives involving teachers in order to promote such research and to bring together what were previously “disparate local initiatives” (see https://trfestival.wordpress.com/about/). The Festival has a number of major supporters including the IATEFL Research SIG, the IATEFL Teacher Development SIG, the All-India Network of English Teachers (AINET), the Argentine Federation of Associations of Teachers of English (FAAPI), TESOL International, the Cameroon English Language and Literature Teachers Association (CAMELTA), the Association of English Language Teachers of Buenos Aires (APIBA), the British Council English Language Agenda/Teaching English, and English Australia/Cambridge Assessment English. The website hosts video-recordings of teacher presentations in different locations with the aim of building a global community for research by teachers and producing published work by teacher researchers and research facilitators.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

173

7.4  Action Research in Language Teaching In this section, I provide a brief overview of some of the main trends and themes emerging from the literature on teachers as action researchers in the field of language teaching. I consider the literature from this specific perspective rather than the field of education more broadly.

7.4.1  Language Teachers’ Engagement with Action Research A general concensus in the literature is that under favourable conditions (see section 7.5.1) many language teachers derive considerable benefits from engaging in action research. Nunan’s (1989) early work (citing Beasley & Riordan, 1981, from the education literature) set out nine benefits, ranging from its relevance to the concerns of teachers and its capacity to build on their accumulated knowledge and provide them with better information about their practices, to the rather more general observation that “it matches the subtle, organic process of classroom life” (p. 3). More recently, Borg (2013, p. 16) summarizes other benefits (presented here with reference to works other than those cited in Borg): • develops teachers’ capacity for autonomous professional judgments (e.g., Viera, 2003; Vilches, 2007; Wang & Zhang, 2014) • reduces teachers’ feelings of frustration and isolation (van Lier, 1994; Tinker Sachs, 2002) • allows teachers to move out of a submissive position and be curriculum innovators (Edwards & Burns, 2016a; Pickering & Gunashekar, 2015) • allows teachers to become more reflective, critical, and analytical about their teaching behaviours in the classroom (Hine, 2013; Dajani, 2015) • makes teachers less vulnerable to and less dependent on external answers to the challenges they face (Allison & Carey, 2007; Wyatt, 2010) • fosters connections between teachers and researchers (O’Brien, Newnham, & Tinker, 2000; Yuan & Lee, 2015) • boosts teachers’ sense of status (Edwards & Burns, 2016b; Mangion & Stokes, 2017). Various studies have highlighted yet other benefits. Burns (1999) emphasizes the sense of professional community that can emerge from collaborative approaches to action research, also reported in a number of other recent studies (e.g. Atay, 2008; Banegas et al., 2013; Burns & KurtogluHooton, 2015; Castro Garcés & Martínez Granada, 2016; McDonough, 2006; Rajuan, 2015). Teachers report gaining skills in conducting research (e.g., Dikilitaş & Griffiths, 2017; Perrett, 2003; Wyatt, 2010), which potentially provide them with the means to continue to reflect on and problematize

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

174

Anne Burns

their practices. There has also been substantial evidence that teachers who conduct action research find it to be a powerful and empowering form of professional development (e.g., Esposito & Smith, 2006; Hei & David, 2017; Mok, 1997). Perhaps most significantly in relation to the potential impact of action research on teaching, many teachers report a much stronger sense of engaging with and understanding the needs and interests of their students (e.g., Alexander & Onslow-Mato, 2015; Burns & McPherson, 2017; Castellano, Mynard, & Rubesch, 2011). Despite the reported benefits, challenges encountered by teachers in doing research are also well documented in the literature. Two in particular are foregrounded as the primary barriers. Time to do research is a particularly pressing issue, particularly in the light of the heavy teaching loads encountered by most teachers (e.g., Borg, 2013; Burns, 1999; Crookes & Chandler, 2001; McKay, 2009; Rainey, 2000). A further major obstacle is the extent of teachers’ skills and abilities to conduct research as well as their self-confidence to become researchers (e.g., Çelik & Dikilitaş, 2015; Mehrani, 2016; Negi, 2016). Other challenges relate to access to resources to support research, such as current literature and research mentoring, lack of understanding or encouragement of research, and institutional constraints.

7.5  Emerging Research Areas for Language Teacher Action Research So far in this chapter, I have outlined the development of action research as well as what the literature reveals about the engagement of language teachers in action research. In this section, I highlight four themes in particular that are emerging from this literature as valuable directions for future empirical research.

7.5.1  Conditions for Action Research In several parts of the world, policy demands for teachers to become active in reflective practice and research as part of their professional development have been steadily increasing (see for example NCTE, 2009, for India; NFL2020, 2014, for Vietnam; MEST, 2015, for Kenya). However, concomitant discussion at a systems level of how teachers can best be supported within a particular educational context to enable them to conduct research is rare (e.g., Borg & Lui, 2013; and see also some of the teacher narratives in Dikilitaş & Griffiths, 2017, pp. 237–262). Rainey (2000) states that the dominant form of action research is small-scale studies carried out for individual teacher self-reflection (see also Crookes, 1993). However, recent literature

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

175

is beginning to indicate that this kind research, ostensibly self-initiated by teachers, is both hard (if not impossible) to sustain and likely to remain private and have limited impact (e.g., Kutlay, 2013; Negi, 2016). Empirical studies documenting the kind of support that is facilitative of successful outcomes from research by teachers are still emergent in the language teaching field. While, as Borg (2013) points out, action research conducted by teachers as part of a formal academic programme has increased, this kind of research “may not generate realistic understandings of teacher research engagement” (p. 185) and may be driven more by instrumental motives related to obtaining qualifications. If action research by teachers as part of continuing teacher development is to be understood more generally, the conditions that support it need to be documented outside formal academic requirements. A few publications have begun to identify the kinds of factors needed to facilitate such research (e.g. Al-Maskari, 2015; Borg, 2013; Burns, 2000; Burns & Brandon, 2017; Yucel & Gűndoğdu, 2017). Various essential conditions identified by these studies include: • teachers’ voluntary participation in research • guidance from an experienced facilitator, mentor, or academic partner • a programme of input about the theory, practices, procedures, and processes of action research • support in accessing and using appropriate literature • timely and ongoing feedback on the design and progress of the research • guidance on dissemination of the research (e.g., through presentation and publication) • peer support through communities of practice (e.g., research conducted in pairs or teams, access to a collaborative peer group) • institutional support (e.g. financial, time allocation) • institutional recognition (e.g. acknowledgement of research efforts, use of teachers’ research for further curriculum development) • positive institutional research climate (e.g., valuing or promotion of action research by teachers, professional development research opportunities). While the findings emerging from these studies serve to illuminate the key facilitative factors, there are few studies that provide empirical data illustrating the way these processes unfold gradually in practice over time at an institutional level in response to local needs and teacher reactions (although see Burns, Westmacott, & Hidalgo Ferrer, 2016, for one recent example). In relation to these factors, more empirical studies are also needed on ways that teacher–academic mentoring partnerships could operate effectively in support of action research by teachers (see Yuan & Lee, 2015; Chan, 2016, for recent examples).

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

176

Anne Burns

7.5.2  Genres for Teacher Action Research Freeman (1998) noted that research by teachers, including the way it was conducted and reported, constitutes a new research genre. He suggests four kinds of possible forms of presentation: interactive presentation (e.g., poster sessions, teacher-research conferences; see Bullock & Smith, 2015; Rebolledo, Smith, & Bullock, n.d.); virtual presentation (e.g., video or multimedia; see https://trfestival.wordpress.com/past-events/); performed presentation (e.g., readers’ theatre, narrative; e.g., Yoshida & Kambara, 2014); and written presentation (e.g., codified and new forms of writing, taking purpose and audience into account; see Burns & Khalifa, 2017 for examples of teacher accounts of action research). While illustrations of some of these ways of presenting action research have started to emerge in the literature, as suggested by these references, there is little research on how teachers manage the processes of planning, preparing, visualizing, or activating such presentations or what forms of support are needed to assist novice teacher researchers. Perhaps the clearest glimpses so far are the few studies conducted on mentoring teacher writing. Drawing on teacher reports, field-notes, and retrospective surveys, Cárdenas (2003) suggests five “criteria” that supported the writing process for teachers in Colombia: using a style sheet (report structure and standards); sustaining “a steady gaze” (narrative style, teacher audience); integrating own and others’ views (personal perspectives integrated with references to the literature); supporting with concrete examples (data samples and illustration); and coping with editor feedback (acceptance of feedback and redrafting). The main difficulties experienced were: language proficiency; lack of writing practice and experience in writing articles; synthesizing content; short times for writing; and lack of peer critical feedback. More recently, Dikilitaş and Mumford’s (2016) interview data with Turkish teachers reveal the importance of collaboration through expert and peer support. While expert mentors were seen to provide moral and pedagogic or technical support, peers provided assistance in language improvement, through “democratic” dialogue and sharing the writing load. These studies also suggest that, since professionally oriented writing is “not a common practice among teachers” (Cárdenas, 2003, p. 52) and given the increasing pressure on language teachers in different parts of the world to begin conducting and publishing research (see Tran, Burns, & Ollerhead, 2017), much more empirical work is desirable to understand the processes and support needs involved.

7.5.3  Teacher Identity and Action Research Identity research related to learners (e.g., Norton, 2000), and more recently teachers (e.g., Barkhuizen, 2017), has expanded considerably in the language teaching field over the last two decades. However, while research

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

177

on the benefits for language teachers of doing action research are now quite well documented, research on how a teacher researcher identity emerges, and what new roles and agencies are taken on, is very limited. Only a handful of studies to date has tracked such identity dimensions. Trent’s (2010) study, conducted within a pre-service teacher course in Hong Kong, showed that, as researchers, participants developed a more critical stance towards previously held perceptions of teacher roles and their images of teaching, as well as their alignment with current educational discourses. Banegas (2012) traced his own development as a researcher and facilitator during a collaborative action research project conducted with four other colleagues in Argentina. He found that there were complex tensions between his various roles as a teacher, classroom researcher, and facilitator relating to his own concern to develop professionally, but also his sense of responsibility to contribute to his fellow teacher-researchers’ professional development and to the curriculum development of the institution. Xu (2014), in China, found that teacher-researcher identities became fragmented at different times as they grappled with the processes of undertaking research and with the external and internal drives surrounding their research participation. More recently, case studies by Edwards and Burns (2016b) of two teachers of international students in Australia showed that the teachers gained a strong sense of themselves as emerging researchers, increased their professionalism and workplace power, and developed greater confidence, but that their developing teacher-researcher “conceptual selves” (p. 742) were constrained by external institutional and sectoral perspectives (see section 7.4.1). Burns (2017) recommends that research on language teacher-researcher identity construction needs to broaden beyond its reliance on the individualist and cognitivist obtained through written or verbal data in order to include socioconstructivist, ethnographic, ecological, and sociohistorical perspectives that would illuminate its emergent, multiple, and shifting nature. Overall, the initial insights from this line of research suggest that developing a teacher-researcher identity is highly complex and dynamic, and characterized by numerous tensions arising from the sociocultural context, as well as from teachers’ cognitive, psychological, and emotional responses (see also Yuan and Burns, 2016). Much more research is necessary, however, to understand these factors more fully.

7.5.4  Sustainability of Action Research The final emergent theme to be discussed here is that of the sustainability of action research. Around twenty-five years ago van Lier (1994) remarked that “if action research is going to make us even more exhausted than we already are, then it will not be a popular or successful activity … It has to enrich our professional life” (p. 33). Allwright (1997) argues that it is

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

178

Anne Burns

unlikely that teachers “will maintain the required level of effort indefinitely, and much more likely that they will soon abandon the research entreprise altogether”; rather what is needed is a “research perspective” (p. 368). Although for many teachers worldwide, conducting action research may indeed be challenging or impossible, recent studies such as those highlighted in this chapter show that for others it is highly rewarding (see, for example, Yucel & Bos, 2015). The question therefore arises of how action research and its impact can be sustained by teachers over time. There is little research at present that documents the longer-term sustainability effects of language teachers conducting action research; however, a study by Edwards and Burns (2016a) suggests that the sustainability of the impact of action research, as perceived by the teachers they interviewed during and after their research involvement, consists of more than teachers adopting and maintaining “a research perspective”: they argue that the impacts are multifaceted, interconnected, and evidenced at the microand macro-levels. At the micro-level, for some teachers, action research reinforced but also enriched their position as classroom teachers, with the added gains of tools for reflection and practice. For others, action research acted as a catalyst to pursue further academic goals and to envisage themselves as researchers. At the macro-level, the teachers’ research impacted on the teachers’ institutions and influenced curriculum development and sustainability. Edwards and Burns propose that a combination of bottom-up individual teacher motivation and experimentation (e.g., also Xie, 2015) and top-down institutional support is crucial in ensuring the sustainability of the impact of action research over time. This line of enquiry is however, at present in its infancy.

7.6 Conclusion Just over a decade ago, Dörnyei (2007) wrote that there was too little action research and, “I am still to meet a teacher who has been voluntarily involved in it” (p. 191). Setting aside larger arguments about the dysfunction of the theory–practice divide in the field of language teaching (e.g., Clark, 1994), one might argue that, these days, there is ample opportunity for Dörnyei to encounter teacher action researchers who have voluntarily conducted research on their practice and to read their publications if he so wished. This chapter has aimed to briefly review the development of action research in the language teaching field, but more importantly to document recent initiatives as well as to suggest areas that merit further research. On the basis of this review, it seems that action research in ELT is a movement that continues to expand as part of an overall trend towards opening up the possibilities of conducting research to language teachers.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

179

References Alexander, L., & Onslow-Mato, W. (2015). Students as active protagonists in the reading process: Developing tasks for critical reading in the EAP classroom. Research Notes, 60, 30–37. Allison, D., & Carey, J. (2007). What do university language teachers say about language teaching research? Canada Journal/Revue TESL du Canada, 24(2), 61–81. Allwright, D. (1997). Quality and sustainability in teacher-research. TESOL Quarterly, 31(2), 368–370. Allwright, D., & Bailey, K. (1991). Focus on the language classroom. Cambridge: Cambridge University Press. Al-Maskari, A. S. (2015). Suppporting teacher research through a practical in-service course. In S. Borg & H. Santiago-Sanchez (eds.), International perspectives on teacher research (pp. 113–124). Basingstoke: Palgrave Macmillan. Atay, D. (2008). Teacher research for professional development. ELT Journal, 62(2), 139–147. Banegas, D. L. (2012). Identity of the teacher-researcher in collaborative action research: Concerns reflected in a research journal. PROFILE, Issues in Teachers’ Professional Development, 14(2), 29–43. Banegas, D. L., Pavese, A., Velázquez, A., & Vélez, S. M. (2013). Teacher professional development through collaborative action research: Impact on foreign English-language teaching and learning. Educational Action Research, 21(2), 185–201. Barkhuizen, G. (2017). Reflections on language teacher identity research. New York: Routledge. Beasley, B., & Riordan, L. (1981). The classroom teacher as researcher. English in Australia, 55(1), 36–42. Borg, S. (2013). Teacher research in language teaching. Cambridge: Cambridge University Press. Borg, S. (2015). Professional development through the Cambridge English/ English UK Action Research Scheme. Research Notes, 61, 3–5. Borg, S., & Burns, A. (2008). TESOL research: What? How? Why? Preconvention event, 2 April, 42nd Annual Convention, New York. Borg, S., & Lui, Y. (2013). Chinese English college teachers’ engagement with research. TESOL Quarterly, 47(2), 270–299. Breen, M., Candlin, C., Dam, L., & Gabrielsen, G. (1989). The evolution of a teacher training program. In K. Johnson (ed.), The second language curriculum (pp. 111–135). Cambridge: Cambridge University Press. Brindley, G. (1990). Towards a research agenda for TESOL. Prospect, 6(1), 7–26. Brumfit, C., & Mitchell, R. (1989). The language classroom as a focus for research. In C. Brumfit & R. Mitchell (eds.), Research in the language classroom (pp. 3–15). London: Modern English Publications and The British Council.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

180

Anne Burns

Bullock, D., & Smith, R. (eds.) (2015). Teachers research! Faversham: IATEFL. Burns, A. (1999). Collaborative action research for English language teachers. Cambridge: Cambridge University Press. Burns, A. (2000). Facilitating collaborative action research: Some insights from the AMEP. Prospect, 15(3), 23–34. Burns, A. (2010). Doing action research in English language teaching: A guide for practitioners. New York: Routledge. Burns, A. (2011). Action research in the field of second language teaching and learning. In E. Hinkel (ed.), Handbook of research in second language teaching and learning (Vol. 2, pp. 237–253). New York: Routledge. Burns, A. (2016). From reluctant teacher to teacher educator: A “brilliant” career. In R. Ellis (ed.), Being and becoming an applied linguist (pp. 301–330). Amsterdam: John Benjamins. Burns, A. (2017). “This life-changing experience”: Teachers becoming action researchers. In G. Barkhuizen (ed.), Reflections on language teacher identity research (pp. 133–138). New York: Routledge. Burns, A., & Brandon, K. (2017). Action research in ELICOS. Impact and insights from a national teacher development program. In A. Burns & H. Khalifa (Eds.), Second language assessment and action research (pp. 315–327). SiLT Series. Cambridge, England: Cambridge University Press. Burns, A., Dikilitaş, K., Smith, R., & Wyatt, M. (eds.) (2017). Developing insights into teacher-research. Faversham: IATEFL. Retrieved from http:// resig.weebly.com/uploads/2/6/3/6/26368747/burns_dikilitas_smith___ wyatt_eds._2017.pdf. Burns, A., & Kurtoglu-Hooton, M. (2016). Using action research to explore technology: International perspectives. London: The British Council. Burns, A., & Khalifa, H. (eds.) (2017). Second language assessment and action research. Cambridge: Cambridge University Press. Burns, A., & McPherson, P. (2017). Action research as iterative design: Implications for English language education research. In S.-A. Mirhosseini (ed.), Reflections on qualitative research in language and literacy education (pp. 105–120). New York: Springer. Burns, A., Westmacott, A., & Hidalgo Ferrer, A. (2016). Initiating an action research programme for university EFL teachers: Early experiences and responses. Iranian Journal of Language Teaching Research, 4(3), 55–73. Castellano, J., Mynard, J., & Rubesch, T. (2011). Action research: Student technology use in a self-access center. Language Learning and Technology, 15(3), 12–27. Çelik, S., & Dikilitaş, K. (2015). Action research as a professional development strategy. In S. Borg & H. Santiago Sanchez (eds.), International perspectives on teacher research (pp. 125–138). Basingstoke: Palgrave Macmillan. Corey, S. M. (1954). Action research in education. Journal of Educational Research, 47(5), 375–380,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

181

Cárdenas, M. L. (2003). Teacher researchers as writers: A way to sharing findings. Colombian Applied Linguistics Journal, 5(5), 49–64. Carr, W., & Kemmis, S. (1986). Becoming critical: Knowing through action research. Geelong: Deakin University Press. Castro Garcés, A. Y., & Martínez Granada, L. (2016). The role of collaborative action research in teachers’ professional development. PROFILE Issues in Teachers’ Professional Development, 18(1), 39–54. Chan, C. (2016). School–university partnerships in English language teacher education. Tensions, complexities and the politics of collaboration. New York: Springer. Clark, M. A. (1994). The dysfunctions of the theory/practice discourse. TESOL Quarterly, 28(1), 9–26. Crookes, G. (1993). Action research for English language teachers: Going beyond teacher research. Applied Linguistics, 14(2), 130–144. Crookes, G., & Chandler, P. M. (2001). Introducing action research into the education of postsecondary foreign language teachers. Foreign Language Annals, 34(2), 131–140. Dajani, M. (2015). Preparing Palestinian reflective English language teachers through classroom-based action research. Australian Journal of Teacher Education, 40(3), 116–139. Dikilitaş, K., & Griffiths, C. (2017). Developing language teacher autonomy through action research. New York: Springer. Dikilitaş, K., & Mumford, S. E. (2016). Supporting the writing up of teacher research: peer and mentor roles. ELT Journal, 70(4), 371–381. Dikilitaş, K., Smith, R., & Trotman, W. (eds.) (2015). Teacher-researchers in action. Faversham: IATEFL. Retrieved from http://resig.weebly.com/ uploads/2/6/3/6/26368747/teacher-researchers_in_action.pdf. Dikilitaş, K., Wyatt, M., Hanks, J., & Bullock, D. (eds.) (2016). Teachers engaging in research. Faversham: IATEFL. Downloadable from http:// resig.weebly.com/uploads/2/6/3/6/26368747/teachers_engaging_in_ research.pdf. Dörnyei, Z. (2007). Research methods in applied linguitics. Oxford: Oxford University Press. Edge, J., & Richards, K. (eds.) (1994). Teachers develop teachers research. London: Heinemann. Edwards, E., & Burns, A. (2016a). Language teacher action research: Achieving sustainablity: ELT Journal, 70(1), 6–15. Edwards, E., & Burns, A. (2016b). Language teacher-researcher identity negotiation: An ecological perspective. TESOL Quarterly, 50(3), 735–745. Esposito, J., & Smith, S. (2006). From reluctant teacher to empowered teacher-researcher. One educator’s journey towards action research. Teacher Education Quarterly, 33(3), 45–60. Fanselow, J. (1987). Breaking rules: Generating and exploring alternatives in language teaching. New York: Longman.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

182

Anne Burns

Field, J., Graham, A., Griffiths, E., & Head, K. (eds.) (1997), Teachers develop teachers research 2. Whitstable: IATEFL. Freeman, D. (1998). Doing teacher research. Boston, MA: Heinle & Heinle. Freeman, D., & Richards, J. C. (eds.) (1996). Teacher learning in language teaching. New York: Cambridge University Press. Hei, K. C., & David, M. K. (2017). Empowering language teachers through action research: Two case studies from Malaysia. English Review: Journal of English Education, 5(2), 163–174. Hine, G. S. C. (guest ed.) (2013). The importance of action research in teacher education programs [Special issue]. Issues in Educational Research, 23(2). Holliday, A. (2001). Doing and writing qualitative research. London: Sage. Kutlay, N. (2013). A survey of English language teachers’ views of research. Procedia—Social and Behavioral Sciences, 70, 188–206. Lewin, K. (1948). Resolving social conflicts: Selected papers on group dynamics, ed. G. W. Lewin. New York: Harper & Row. Mangion, L., & Stokes, P. (2017). Using cloud computing to increase collaboration in developing writing skills. Research Notes, 67, 7–12, McKay, S. (2006). Researching second language classrooms. Mahwah, NJ.: Lawrence Erlbaum. McFarland, K. P., & Stansell, J. C. (1993). Historical perspectives. In L. Patterson, C. M. Santa, C. G. Short, & K. Smith (eds.), Teachers are researchers: Reflection and action (pp. 12–18). Newark, DE: International Reading Association. Mathew, R., Eapen, R. L., & Tharu, J. (eds.) (2000). The language currriculum: Dynamics of change, Vol. 1. The outsider perspective. Report of the International Seminar 1995. Hyderabad: Orient Longman for the Central Institute for English and Foreign Language Studies. McDonough, J., & McDonough, S. (1997). Research methods for language teachers. London, England: Arnold. McDonough, K. (2006). Action research and the professional development of graduate teaching assistants. The Modern Language Journal, 90(1), 33–47. McKay, S. L. (2006). Researching second language classrooms. New York: Routledge. McKay, S. L. (2009). Second language classroom research. In A. Burns & J. C. Richards (eds.), The Cambridge guide to second language teacher education (pp. 281–288). New York: Cambridge University Press. Mehrani, M. B. (2016). A narrative study of Iranian EFL teachers’ experiences of doing action research. Iranian Journal of Language Teaching Research, 5(1), 93–112. MEST (2016). The National Curriculum Policy. Nairobi: Ministry of Education, Science and Technology. Mok, A. (1997). Student empowerment in an English language enrichment programme: An action research project in Hong Kong. Educational Action Research, 5(2), 305–320.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

183

NCTE (2009). National curriculum framework for teacher education. New Delhi: National Council for Teacher Education. Negi, J. S. (2016). Improving teaching through action research; Perceptions, practices and problems (3Ps): Voices from secondary level teachers in an EFL context. ELT Voices, 6(4), 18–30. NFL2020 (2014). Competency framework for English language teachers. Hanoi: Ministry of Education. Norton, B. (2000). Identity and language learning: Gender, ethnicity and educational change. Harlow: Longman and Pearson Education. Nunan, D. C. (1987). The teacher as curriculum developer. Adelaide: National Curriculum Resource Centre. Nunan, D. C. (1989). Understanding language classrooms. A guide for teacherinitiated action. Hemel Hempstead: Prentice Hall. O’Brien, T., Newnham, G., & Tinker, L. (2000). Collaborative practice-based research in supporting the language development of primary level bilingual children: A case study. Educational Action Research, 8(1), 43–64. Paltridge, B., & Phakiti, A. (eds.) (2015). Research methods in applied linguistics. London: Bloomsbury. Perrett, G. (2003). Teacher development through action research: A case study in focused action research. Australian Journal of Teacher Education, 27(2), 1–10. Pickering, G., & Gunashekar, P. (eds.) (2015). Innovation in language teacher education. London: The British Council. Rainey, I. (2000). Action research and the English as a foreign language practitioner: Time to take stock. Educational Action Research, 8(1), 65–91. Rajuan, M. (2015). Practice and principles of pre-service action research. In S. Borg & H. Santiago-Sanchez (eds.), International perspectives on teacher research (pp. 139–151). Basingstoke: Palgrave Macmillan. Ramani, W. (1987). Theorizing from the classroom. ELT Journal, 41(1), 3–11. Rebolledo, P., Smith, R., & Bullock, D. (eds.) (n.d.). Champion teachers: Stories of exploratory action research. London: The British Council. Richards, J. C., & Lockhart, C. (1994). Reflective teaching in second language classrooms. New York: Cambridge University Press. Richards, K. (2003). Qualitative inquiry in TESOL. Basingstoke: Palgrave Macmillan. Richards, K., Ross, S., & Seedhouse, P. (2012). Research methods for applied language studies. New York: Routledge. Roberts, J. (1986). Action research: An introduction. Teacher Development SIG Newsletter, No. 4. Faversham: IATEFL. Roberts, J. (1998). Language teacher education. London: Arnold. Somekh, B., & Zeichner, K. (2009). Action research for educational reform: Remodelling action research theories and practices in local contexts. Educational Action Research, 17(1), 5–21. Stenhouse, L. (1975). An introduction to curriculum research and development. London: Heinemann.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

184

Anne Burns

TESOL (2004). TESOL 2004 Annual Report. Alexandria, VA: TESOL International. Retrieved from http://www.tesol.org/docs/annual-reports/ arp2004.pdf?sfvrsn=4. Tinker Sachs, G. (ed.) (2002), Action research in English language teaching. Hong Kong: City University of Hong Kong. Tran, A., Burns, A., & Ollerhead, S. (2017). ELT lecturers’ experiences of a new research policy: Exploring emotion and academic identity, System, 67, 65–76. Trent, J. (2010). Teacher education as identity construction: Insights from action research. Journal of Education for Teaching: International Research and Pedagogy, 36(2), 153–168. van Lier, L. (1988). The classroom and the language learner. London: Longman. van Lier, L. (1994). Action research. Sintagma, 6, 31–37. Viera, F. (2003). Addressing constraints on autonomy in school contexts: Lessons from working with teachers. In D. Palfreyman & R. C. Smith (eds.), Learner autonomy across cultures: Language education perspectives (pp. 220–239). Basingstoke: Palgrave Macmillan. Vilches, J. U. (2007). Teacher autonomy: A critical review of the research and concept beyond applied linguistics. Í  kala, Revista de Lenguaje y Cultura, 12(18), 245–275. Wallace, M. (1991). Training foreign language teachers: A reflective approach. Cambridge: Cambridge University Press. Wallace, M. (1998). Action research. Cambridge: Cambridge University Press. Wang, Q., & Zhang, H. (2014). Promoting teacher autonomy through university–school collaborative research. Language Teaching Research, 18(2), 222–241. Wyatt, M. (2010). Teachers researching their own practice. ELT Journal, 65(4), 417–425. Wyatt, M., Burns, A., & Hanks, J. (2016). Teacher practitioner research: Reflections on an online discussion. TESL-EJ, 20(1), 1–22. Xie, J. (2015). Learning to do teacher research independently: Challenges and solutions. In S. Borg & H. Santiago Sanchez (eds.), International perspectives on teacher research (pp. 47–56). Basingstoke: Palgrave Macmillan. Xu, Y. (2014). Becoming researchers: A narrative study of Chinese university EFL teachers’ research practice and their professional identity construction. Language Teaching Research, 18(2), 242–259. Yoshida, T., & Kambara, K. (2014). EFL teachers’ conceptual development and the transformation of teaching through narratives in the E-portfolio. In B. S. Said & L. Zhang (eds.), Language teachers and teaching: Global perspectives, local initiatives (pp. 219–236). New York: Routledge. Yuan, R. E., & Lee, I. (2015). Action research facilitated by university–school collaboration. ELT Journal, 69(1),1–10.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

Action Research: Developments, Characteristics, and Future Directions

185

Yuan, R. E., & Burns, A. (2016). Teacher identity development through action research: A Chinese experience. Teachers and Teaching: Theory and Practice, 23(6), 729–749. Yucel, M., & Bos, V. (2015). Action research as a means of stepping out of the teaching comfort zone. In S. Borg & H. Santiago Sanchez (eds.), International perspectives on teacher research (pp. 29–46). Basingstoke: Palgrave Macmillan. Yucel, B., & Gűndoğdu, M. A. (2017). Mentoring teachers as researchers. In A. Burns, K. Dikilitaş, R. Smith, & M. Wyatt (eds.), Developing insights into teacher research (p. 29–38). Faversham: IATEFL.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:32:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.008

8 Classroom Observation Research Nina Spada 8.1 Introduction This chapter is about second and foreign language (L2) classroom observation research, with a focus on observation instruments developed within the interaction analysis tradition. It begins with a brief description of classroom observation research and a comparison of interaction analysis with other approaches. This is followed by a general description of how observation instruments are organized, how they differ in terms of number and type of categories, coding procedures, units of analysis, and overall strengths and weaknesses. Four observation schemes are then described in detail with specific examples and illustrations of their categories along with information about the reasons for their development, the contexts in which they have been used, the type of research undertaken with them, and brief summaries of the findings associated with each. Throughout, there is an examination of how interaction analysis observation schemes have evolved in relation to the different phases of L2 classroom research, including the global method comparison studies, process and process-product research, and the shift from macro-level to micro-level descriptions of classroom behaviours. The chapter concludes with a discussion of the benefits of multi-method approaches to L2 classroom observation research.

8.2  Approaches to Classroom Observation Research Observation research is concerned with describing what goes on in L2 classrooms between teachers and learners. It documents pedagogical practices and procedures including the content of instruction and how it is organized and delivered. It includes descriptions of the kind of language that teachers and students use and the nature of the linguistic interactions that take place between them. Classroom observation also describes how

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

187

teachers and learners are socially organized into different groups (e.g., whole class, pairs), the dynamics of the groupings, and the type of activities that take place within them. Another focus of observation research is affective behaviours such as the motivation practices of teachers and the levels of engagement on the part of learners. Depending on the approach taken, classroom observation research can be broad-based and open-ended or narrowly focused and closed, structured or unstructured, objective or subjective, and quantitative or qualitative. Chaudron (1988a) distinguished three main approaches to classroom observation: interaction analysis; discourse analysis; and ethnography. Interaction analysis has its roots in sociological investigations of group processes which led to the development of systems to observe and analyse classrooms in terms of the relationships between teachers and learners, and the activities in which are engaged. Early versions of observation schemes developed within the interaction analysis tradition emphasized the pedagogic functions of teacher talk (e.g., teacher gives directions, teacher corrects) and affective characteristics (e.g., teacher encourages, teacher praises). Later schemes included a larger number of categories to describe the language used by teachers (and learners) in linguistic and communicative terms. These systems typically consist of discrete categories of language use and do not describe how they combine to form classroom discourse (but see the FOCUS instrument below). Observation research that describes the nature and structure of the flow of verbal interactions between teachers and students falls within the discourse analysis tradition.1 This approach arose from a linguistics perspective and examines the language used by teachers and students as a sequence of interconnected moves and communicative acts described in terms of structure and pedagogical function. One of the earliest discourse analysis systems, developed by Sinclair and Coulthard (1975), identified the well-known Initiation–Response–Feedback (IRF) exchange structure, which has had an enormous impact on our understanding of the ways in which teachers and students communicate. The third tradition that Chaudron described is ethnographic—a holistic, qualitative approach to classroom observation that examines behaviours in social and cultural terms (Watson-Gegeo, 1988). Within this approach, detailed field notes supplemented with audio and video recordings are used to collect the observation data, which are analysed qualitatively and supplemented with other data sources (e.g., interviews). Ethnographic research is typically longitudinal in nature and can be conducted by participant or non-participant observers. A particular type of ethnography used in L2 classroom research is the ethnography of communication (Duff, 2002; Harklau, 1994). 1

 Related approaches include conversational analysis (Markee, 2000; Seedhouse, 2004) and critical discourse analysis (Fairclough, 1995).

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

188

N i n a S pa da

8.3  Brief History of Interaction Analysis Schemes Classroom observation schemes developed within the interaction analysis tradition consist of a set of pre-determined categories to document classroom events and behaviours. Because observation instruments were initially used to evaluate teaching effectiveness, the focus was on teacher behaviours. One well-known example is FIAC (Flander’s Interaction Analysis Categories), a scheme that was developed in the field of education which consisted of ten categories of “effective teaching” divided into those that reflected direct influence (e.g., lecturing, giving directions) and indirect influence (e.g., asking questions, accepts feelings; Flanders, 1970). A modified version of Flanders’ scheme for use in L2 classrooms— the Foreign Language Interaction System (FLINT)—was developed by Moskowitz (1976). The underlying assumption for both observation schemes was that indirect teaching practices were more effective for learning than direct ones. This was based entirely on expert opinion and consensus rather than research evidence. The categories were used to help teachers-in-training discover the extent to which their own teaching practices matched them or not, thus reflecting more or less effective instruction. The prescriptive nature of the early observation schemes was directly tied to the audio-lingual method which was the broadly accepted and widely implemented “scientific approach” to L2 teaching of the era. The development and use of observation schemes for research purposes came later and largely in response to the global method comparison studies in the 1960s. These were large-scale studies that compared traditional grammar-based pedagogies with audio-lingual instruction. Two frequently cited method comparison studies in the United States, known as the Colorado Project (Scherer and Wertheimer, 1964) and the Pennsylvania Project (Smith, 1970), investigated college level learners of German and high school students of French and German respectively. The combined findings of the studies indicated that while students in the audio-lingual group did better at speaking and students in the grammar-translation classes did better in reading and writing, there were no significant longterm differences in learners’ language abilities that could be attributed to the teaching method. Similar findings were obtained in a large-scale method comparison study in Sweden known as the GUME project (Levin, 1972) in which an implicit (i.e., inductive) and explicit (i.e., deductive) version of the audio-lingual method were compared. The inconclusive results of these studies pointed to a number of limitations, including the lack of control over teacher and learner variables as well as the use of tests that were biased in favour of more traditional grammar-based instruction. The most serious concern was the absence of any detailed systematic information about the instructional processes. That is, there was no reliable

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

189

evidence to demonstrate how the instructional approaches were delivered and, furthermore, whether they were implemented differently. The failure of the global method comparison studies led to a period of process research in which the focus shifted away from investigating learning outcomes (i.e., product) in relation to particular methods to investigating what actually goes on in L2 classrooms through detailed descriptions of teacher and learner behaviours (i.e., process). The shift to process-based research did not mean that method comparison studies ceased to exist. On the contrary, there continued to be studies to investigate a range of methodological differences including comparisons of communicative and more traditional approaches (e.g., Savignon, 1972), different versions of communicative and task-based approaches (e.g., Beretta & Davies, 1985; Prabhu, 1987), and comparisons between input- and output-based instructional programmes (Lightbown et al., 2002; Shintani & Ellis, 2010). These studies were much smaller in scale than the global method comparison studies in terms of number of parti­ cipants and focus. Like the earlier global method comparison studies, they did not include an observation component and the findings were inconclusive. In an analysis and review of sixteen of these studies, Ellis (2012) points to the considerable heterogeneity of the research in terms of parti­ cipants, contexts, types of instruction, methods of testing, etc. He also discusses the methodological problems associated with many of the studies, including comparisons of non-equivalent groups, no attempt to establish that the instructional treatments were implemented in ways that were consistent with their descriptions, no control groups, no account of individual learner differences, and in most cases no efforts to guard against test bias. Factors such as these make comparisons between studies difficult and necessarily lead to tentative conclusions.

8.4 Observation Schemes: Coding Procedures and Categories In a review of L2 classroom observation research, Long (1980) identified twenty-two observation schemes developed for L2 teaching. Almost ten years later in Chaudron’s (1988a) review of research in L2 classrooms, he added three observation schemes that had been developed in the interim: the Target Language Observation Scheme (TALOS) (Ullman & Geva, 1985), the Communicative Interaction (CI) system developed by Mitchell, Parkinson, and Johnstone (1981), and the Communicative Orientation of Language Teaching (COLT) observation scheme (Allen, Fröhlich, & Spada, 1984; Spada & Fröhlich, 1995). To my knowledge, these are the last comprehensive observation instruments developed within the interaction

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

190

N i n a S pa da

analysis tradition that cover a wide range of pre-determined categories for the documentation and measurement of pedagogic and linguistic behaviours. As reported in Long (1980) and Chaudron (1988a), observation instruments differ in terms of coding procedures, number and type of categories, units of analysis, focus, and purpose. In this section, I discuss the first three features leaving an examination of differences in focus and purpose to section 8.5 where four individual observation schemes are described and discussed. With respect to coding procedures, some observation instruments use a category system in which each behaviour or event is coded every time it occurs. Others use a sign system in which behaviours or events are coded once during a specified time period (e.g., every three minutes). A rating scale is also used with some observation schemes. This typically takes place after the observation session and involves making decisions about the estimated frequency of behaviours or events along a scale ranging from “very frequent” to “never” or “high” to “low”. Another aspect of the coding procedures is whether the coding takes place in “real time”—as events or behaviours unfold in the classroom—or not. If the coding takes place post-hoc, it is done using audio or video recordings. Decisions about realtime or post-hoc coding depend to a large extent on the type of categories. Observation schemes also differ widely in terms of the number of categories, ranging from fewer than ten categories on simple schemes to over 150 with more complex instruments. While it can be challenging to work with observation schemes that have numerous coding categories, it depends on the type of categories. For example, low-inference categories record objective, overt, easily observable behaviours such as skill focus (e.g., speaking, listening), participant organization (teacher-fronted, pair work), and use of first languages (L1). High-inference categories involve judgment and thus are more subjective in nature (e.g., deciding whether students are motivated, whether teachers ask questions to which they know the answers). If an observation scheme has numerous categories but the majority of them are low-inference, it is easier to use than an observation scheme with fewer categories that are high-inference in nature. This raises issues about the reliability of observation categories discussed in section 8.4.1 below.

8.4.1 Observation Schemes: Reliability, Validity, and Units of Analysis Despite the importance of doing so, few L2 classroom observation researchers working within the interaction analysis tradition have confirmed the reliability of their observations and analyses. This includes inter-rater reliability to verify whether multiple users of a scheme are making the same coding decisions when using the categories to document classroom

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

191

behaviour. It also includes intra-rater reliability to determine whether an individual observer makes the same coding decisions consistently. Procedures for establishing reliability appear to be limited to observation schemes that have been developed most recently for research purposes. For example, the developers of the CI scheme (Mitchell, Parkinson, & Johnstone, 1981) double coded both early and late observations of several classes and reported reliability estimates consistently. Fröhlich, Spada, and Allen (1985) indicate high levels of reliability using the COLT scheme but do not report reliability estimates. While it is essential that developers of an observation scheme carry out reliability testing (and report it), it is equally important for users of the instruments to do so given that variation occurs across users of research instruments in different contexts. Establishing the validity of observation categories is equally important. This concern was raised by Long (1980) almost forty years ago. At the time he wrote that “in addition to categories designed specifically for the description of classroom processes, it would seem of crucial importance that instruments contain categories which current theory would suggest are relevant to classroom language acquisition” (p. 20). Again, only a few research teams associated with the more recently developed instruments have reported on efforts to establish validity. For example, in the development of TALOS, Ullman and Geva (1985) established validity by comparing the coding results of low-inference categories with those of the high-inference categories. A match between the two was considered evidence of construct validity because both sets of categories were developed using the same theoretical constructs. Fröhlich, Spada, and Allen (1985) set out to establish the validity of the COLT scheme by establishing a relationship between programmatically defined degrees of communicative language teaching with categories on the scheme designed to measure the communicative orientation of L2 programmes. They report that the instrument had criterion-related validity because it succeeded in identifying differences in the programmes based on the relative frequency of pedagogic and linguistic behaviours captured in both parts of the scheme. They also calculated a score based on the frequencies of five selected categories to characterize the degree of communicative orientation across the programmes to establish further validation of the scheme. Criticisms have been raised about this procedure, however, including the arbitrariness of the five selected categories and the equal weighting of them (Chaudron, 1991). Another methodological issue related to classroom observation schemes is that they vary with respect to the unit of analysis. For example, researchers analyse classroom events on smaller or larger segments such as linguistic or functional units (e.g., utterances or moves) or pedagogical units (e.g., activities or episodes). While these differences reflect

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

192

N i n a S pa da

important distinctions and assumptions about relationships between language, pedagogy, and learning, they make the interpretation of L2 observational data and comparisons across studies difficult. Chaudron (1988b) illustrates this problem with reference to a specific feature of classroom behaviour—teacher language choice—measured using different units of analysis: … studies may measure teacher language choice (whether the first, or the target language) in units of time, of teacher utterances, or of functional moves. These differences in measurement all involve different assumptions about the psychological relationships between language choice and student learning. For instance, a measure of language use by fixed time units (every three minutes) may overestimate the actual amount of language spoken, if the teacher speaks more slowly in that language relative to the other, or if more silence occurs in that time period. (p.13)

He goes on to argue that temporal versus categorical measures also assume a different value in terms of learning and raises the question as to whether the knowledge a learner derives from fifteen minutes of exposure to the L2 is the same type of knowledge that is derived from fifteen question-and-answer exchanges. The methodological challenges associated with classroom observation research within the interaction tradition point to the need for more quantitative research to establish the validity of observation categories, their reliability, and the use of comparable units of analysis to permit systematic comparisons that can lead to the generalization of findings. There appears to be little interest in conducting such research, however, given that the last studies to investigate some of these issues took place in the late 1980s. This appears to be related to a shift that took place around that time from using broad-based comprehensive schemes with numerous observation categories to the use of more narrowly focused descriptions of one or two classroom behaviours that were examined in relation to learning outcomes. This is discussed in more detail in sections 8.5 and 8.6. In the next section, some of the strengths and limitations of observation schemes developed within the interaction analysis tradition are examined.

8.4.2  Observation Schemes: Advantages and Disadvantages The advantage of using observation schemes is that they allow researchers to see directly what teachers and learners are doing in the L2 classroom rather than having to rely on what they say they do or on the use of external labels (e.g., audio lingual, communicative) to describe what they do. Observation schemes also permit an objective and systematic account of classroom events and behaviours that enables cross-study comparisons in different classrooms and educational contexts. This in turn allows for an

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

193

increase in the generalizability of research. In addition, the use of observation instruments facilitates the description and measurement of pedagogical change over time which can be effectively used in programme revision and evaluation. Importantly, structured observation makes the demanding task of describing complex and rich classroom processes do­­ able and easier to use when compared with non-systematic descriptions. Furthermore, observation schemes can and should be revised and adapted to suit the specific goals of the research. As Spada and Fröhlich (1995) point out, “observation schemes are simply tools for research and they should serve rather than direct it” (p. 10). The disadvantages that come with the use of observation schemes developed within the interaction analysis tradition are the flip side of some of the advantages. For example, while observation schemes provide a record of what is observable, we know that a good deal of what goes on in classrooms is not observable (e.g., mental processes). In addition, because these observation schemes consist of a set of pre-determined categories and the observer is coding for the categories “on the page”, other pedagogical behaviours that could serve important roles are not coded and perhaps not even noticed. This renders observation schemes less sensitive to emergent information and is one of the arguments for conducting qualitative and ethnographic observation research. Furthermore, while it is true that observation schemes have a way of simplifying complex pedagogical and linguistic behaviours, they may reduce the complexity to a point where it could result in an incomplete or distorted representation of events and behaviours. Finally, while some efforts have been made to establish the reliability and validity of observation schemes, they have been minimal. In the next section four observation schemes that are well-known within the interaction analysis tradition are examined.

8.5  Four Observation Schemes: FOCUS, TALOS, CI, & COLT Observation schemes share several common categories, such as the grouping of participants (e.g., individual, pair, group), content or topic of lesson (linguistic or other), language focus (e.g., vocabulary, grammar, etc.), and language use (L1, L2). However, they vary in terms of how these categories are defined and operationalized, the addition of other categories to document different aspects of classroom behaviours, and their overall structure and organization. These differences reflect divergence in their purpose and focus but also influences from various disciplines (e.g., linguistics, education) that have shaped their conceptualization and development. One of the earliest interaction schemes that was developed for both research and teacher education purposes is Fanselow’s (1977) Foci for

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

194

N i n a S pa da

Observing Communications Used in Settings (FOCUS). Following the early work of Bellack et al. (1966) on first language classroom discourse, FOCUS uses the move2 as the unit of analysis and includes structuring, soliciting, responding, and reacting as the major criteria for segmenting the classroom interaction. These moves, produced by the teacher or the student, are then described in terms of the mediums used to communicate (linguistic, non-linguistic, para-linguistic), how the mediums are used to communicate areas of content (attend, characterize, present, re-present), and what areas of content are communicated (language systems, life, procedure, subject matter). Within these major categories are over seventy sub-categories that more precisely describe the nature of the communication. For example, under “characterize” in the category describing how the mediums are used to communicate content, there are five ways in which this can be done (differentiate, evaluate, examine, illustrate, label). Included under “language systems” in the category describing the different areas of content communicated are eight sub-categories (e.g., contextual, grammatical, literary, meaning, mechanics of writing). This observation scheme is complex, and even though it consists of discrete categories, its goal is to show how they combine to form larger stretches of discourse. This is an important feature, and it is what distinguishes FOCUS from other schemes within the interaction analysis tradition. Because of its detailed nature and focus on individual moves by teachers and learners, it is not an instrument that can be used in real-time coding. Unlike earlier interaction analysis schemes, FOCUS is not restricted to describing (or prescribing) the behaviours or events that are associated with a specific language teaching method. The categories are linguistic rather than pedagogic in nature, and this is likely why Fanselow claimed that FOCUS could be used to describe communication both inside and outside classroom environments. Unfortunately, it appears that FOCUS has not been used to a great extent in either context. This may be due to the time-consuming nature of using the scheme. Two observation schemes that have more of a pedagogical focus are the Target Language Observation Scheme (TALOS) developed for use in a French second language programme evaluation study (Ullman & Geva, 1985) and the Communicative Interaction (CI) scheme developed to investigate features of communicative interaction in French foreign language classrooms (Mitchell, Parkinson, & Johnstone, 1981). TALOS consists of two separate but related sections. One is a low-inference real-time coding scheme that describes different types of classroom activities (e.g., drills, free communication), content (e.g., sound, phrase, discourse), focus (eg., grammar, culture), skill focus (e.g., listening, reading), teacher language 2

 A move is a functional-linguistic unit that can consist of one or two words or stretches of talk.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

195

(e.g., explain, correct, question), student language (e.g., word, sentence, type of question), and use of L1 and L2. The other part of TALOS is a high-inference rating scale to be completed after the observation session. This requires the observer to rate on a five-point scale (extremely low to extremely high) judgments about the overall degree of occurrence of similar events or behaviours. The unit of analysis is time for the low-inference portion of the scheme, that is, behaviours are coded every minute throughout the observation period and coded once afterwards for the high inference categories. TALOS was used to evaluate French second language classes in elementary schools, and the results indicated that, for the most part, the classes were teacher-centred and focused on the formal features of language with little evidence of functional or communicative teaching. Mitchell, Parkinson, and Johnstone (1981) and Mitchell (1985a) set out to determine the extent to which French foreign language teachers who shared a commitment to the communicative approach used the target language to provide extensive exposure to message-based input. In their CI observation scheme, there are two units of analysis: pedagogical moves and lesson segments. Within pedagogical moves, the functional (e.g., metalinguistic talk, organizing and activity instructions, discipline) and structural (e.g., lexical range, syntactic complexity, discourse markers) aspects of the teachers’ language are coded along with strategies they employ for the negotiation of meaning and the repair of breakdowns in communication. For lesson segments, categories are coded in terms of class organization (e.g., individual, whole class), language activities (e.g., drill, translation, communicative foreign language use), topic of discourse (e.g., language, organization, real life), and mode of involvement (e.g., listening, doing, writing). The results indicated considerable quantities of foreign language input and interaction; however, most of it was language-related in structure-based activities rather than messageoriented in communicative practice. Like the FOCUS observation scheme, it appears that neither the TALOS nor CI schemes have been used to a great extent outside the contexts in which they were originally developed. While TALOS and CI are easier to use and are less time-consuming in terms of coding and analyses than FOCUS, it is not clear why so little L2 classroom research using these schemes has been conducted. This is probably due to the shift over the past twenty-five years from using comprehensive schemes that capture a wide range of pedagogic and linguistic behaviours to using smaller-scale systems targeted to specific domains of linguistic and pedagogical practice in order to explore relationships between L2 instruction and learning outcomes. The Communicative Orientation of Language Teaching (COLT) observation scheme is used to describe L2 classrooms at the level of activity type along with the verbal interactions that take place within them. COLT is

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

196

N i n a S pa da

described in greater detail here than the other schemes because it has been used extensively in L2 classroom research, has influenced the development of other schemes, and appears to be the last broad-based comprehensive L2 observation scheme to have been developed in the field of L2 classroom research. COLT was developed in the early 1980s to describe differences in the communicative orientation of language teaching (process research) and to determine whether and how this contributes to differences in L2 learning outcomes (process-product research). One of the ways in which COLT has been distinguished from other interaction analysis observation schemes is the effort that was made to validate the categories based on theories of communicative competence, communicative language teaching, and research in first and second language acquisition (Allen, Fröhlich, & Spada, 1984). The COLT scheme consists of two parts: Part A, which involves realtime coding of instructional behaviours at the level of activity, and Part B, which is a detailed post-hoc coding of the verbal interactions between teachers and students and students interacting with each other using audio-recordings and transcriptions (see Appendix 1 for Part A and B of the COLT scheme). Part A consists of five major features that are divided into thirty-three categories and subcategories: activity type (open-ended); participant organization (e.g., whole class, group work, individual); content (e.g., language, other topics); student modality (e.g., listening, speaking); and materials (e.g., visual, minimal, controlled). Part B consists of seven major features divided into forty categories/subcategories that are divided equally to describe teacher and student speech: use of target language (L1 or L2); information gap (giving or requesting information); sustained speech (ultraminimal, minimal, sustained); reaction to code or meaning (form or message); incorporation of student utterances (e.g., correction, repetition, clarification request); discourse initiation (selfinitiated turns by students); and restriction of linguistic form (e.g., restricted, unrestricted). COLT has been used in several process-oriented studies to determine differences in the communicative orientation of L2 instruction. In the first such study, Fröhlich, Spada, and Allen (1985) used COLT Part A and B to observe instruction in three different French language programmes (i.e., core, extended, immersion) and one ESL programme. Prior to observing the classes, the researchers predicted that some classes would be more communicatively oriented than others based on previous knowledge of and observations in the programmes. One of the main goals of that study was to determine whether the categories in the scheme were successful in differentiating between more or less communicatively oriented instruction. This was confirmed. Subsequently, other studies have used COLT to describe instructional processes along communicative dimensions for a

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

197

variety of purposes (see Spada & Fröhlich, 1995 for summaries of earlier studies and Lyster & Mori, 2006 and Simard & Jean, 2011 for more recent research). Other observation schemes with different foci have also been modelled on COLT. For example, Guilloteaux and Dörnyei (2008) developed the Motivational Orientation of Language Teaching (MOLT), an instrument to describe the motivational practices of teachers and learners in foreign language classrooms. COLT has also been used in combination with more qualitative approaches to L2 classroom research (e.g., Fazio & Lyster, 1998). There are a few studies that have used COLT to explore whether differences in the communicative orientation of L2 instruction contribute to differences in learning outcomes. In a large-scale process-product study with school-age French language learners, Allen and Carroll (1987) investigated correlations between the COLT Part A and Part B categories in relation to learners’ performance on oral and written language tests that measured a wide range of learners’ L2 ability (e.g., grammatical, sociolinguistic, discourse). While several of the COLT categories positively correlated with learner achievement (e.g., focus on form, information gap, sustained speech), few of the relationships were statistically significant. In a separate analysis of the classroom data, the researchers grouped the COLT features into binary oppositions to characterize the target classes from most experiential to most analytic and found that there were positive correlations between both types of instruction and learning outcomes. In a study with adult ESL learners, Spada (1987) used the COLT categories to distinguish between more or less communicative classes and found minimal differences between them in terms of learning outcomes. Similar findings have been reported in other studies using COLT with different groups of learners (e.g., McKay, 2006). This raises questions about whether the COLT categories are valid predictors of learning outcomes. It also raises a more general concern as to whether efforts to establish process-product relationships using comprehensive observation schemes like COLT are feasible or realistic. That is, COLT may be too blunt an instrument to investigate process-product relationships. This is likely compounded when, not only are the classroom process variables measured broadly, but also the product variables (i.e., learner outcomes) via general language proficiency measures. This may explain why the use of broad-based observation schemes has apparently ceased to exist in studies investigating relationships between L2 instruction and L2 learning.3 Instead, researchers have sought to carry out more fine-grained process-product research focused on a narrower  The publication dates of research reports using comprehensive observation schemes such as COLT have dropped

3

precipitously since the late 1980s.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

198

N i n a S pa da

range of instructional features that are hypothesized to be predictors of L2 learning and examined their effects on specific features of language use and development.4 This represents a move from macroscopic to microscopic description (Spada & Lyster, 1997) or what McKay (2006) refers to as “generic/limited”. Mitchell (1985b) characterizes it as a shift from “global to single-issue research” and explains it in the following way: The complex and multi-layered nature of L2 classroom processes … has led many researchers to shift away from the attempt to conduct comprehensive analyses of lesson data, with the aim of encompassing all possible features and variables which may have significance for improving the effectiveness of the classroom as a language learning environment. Recognising that we do not yet have the theoretical understanding of the overall processes of teaching and learning a second language which would make this a practicable goal, these researchers have instead turned to the more detailed study of particular features and mechanisms of the classroom instructional process, which may ultimately contribute to the development of a global understanding. (p. 335)

This does not mean that broad-based observation schemes like COLT do not serve a useful purpose in L2 classroom research. On the contrary, the strength of comprehensive instruments is precisely in their capacity to provide the big picture. An example of where the use of COLT provided critical information in a single-issue study is discussed later in this chapter (Lyster & Mori, 2006). We now turn our attention to a few small-scale process and process-product studies.

8.6 Narrowly-Focused Observation Systems: Process and Process-Product Research Using COLT as an example of a macroscopic scheme, Spada and Lyster (1977) describe the development of a microscopic system (i.e., error treatment model) used to describe teachers’ reactions to learners’ errors and their immediate responses to this feedback (Lyster & Ranta, 1997). The error treatment model was inspired by two categories on COLT Part B (reaction to form or message and incorporation of student utterances) as well as other feedback-on-error models used in L2 classroom research. Given the researchers’ focus, they also created additional categories that could more fully describe the oral linguistic behaviours of teachers and learners in the error treatment sequence. Seven types of corrective feedback are 4

 Although these approaches are not addressed here, it is important to emphasize that another response to the use of comprehensive observation schemes has been to carry out more qualitative studies, including ethnographic research of classroom processes.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

199

included in the model (e.g., recasts, elicitation, repetition) along with several different types of learner repair (e.g., repetition, incorporation) and needs repair (e.g., same error, partial repair). Since its development, this system has been used in many descriptive studies of oral corrective feedback in a wide range of contexts, as well as in numerous experimental and quasi-experimental studies (see Lyster, 2007 and Lyster, Saito, & Sato, 2013 for summaries of this work). The results from the process-oriented research using the error treatment model have revealed that recasts are the preferred type of corrective feedback used by L2 instructors in second and foreign language classrooms, and this is particularly the case in communicative and content-based language teaching contexts. The overall results of studies investigating the efficacy of oral corrective feedback in promoting L2 learning (process-product) indicate that it plays a beneficial role (Russell & Spada, 2006). Questions remain, however, about whether certain types of corrective feedback are more effective than others (Li, 2010; Lyster & Saito, 2010; Mackey & Goo, 2007). Conflicting results suggest the effects of different types of corrective feedback are dependent on several factors, including the context in which the research was undertaken (e.g., classroom versus laboratory), the type of target feature, the age and proficiency level of learners, and the overall communicative orientation of the instructional context in which the corrective feedback is provided. Based on these findings Lyster, Sato, and Saito (2013) conclude that teachers should use a variety of different types of corrective feedback, echoing what Lyster and Ranta (1997) concluded fifteen years earlier and reinforcing the words of Ammar and Spada (2006) that “one size does not fit all” (p. 566). Other examples of observation research taking a limited perspective include studies focused on the quantity and quality of teachers’ use of the L1 in L2 classrooms. Considerable variation has been observed both in terms of how much teachers use the L1 and for what purposes (e.g., classroom administration, grammar instruction, empathy or solidarity; Kim & Elder, 2005; Polio & Duff, 1994; Macaro, 2001). Research targeting other teacher behaviours has examined the use of metalanguage. This has primarily included observations of different types of metalanguage (grammatical, sociolinguistic, pragmatic) (Borg, 1998) and the extent to which teachers use metalanguage in communicative tasks (Basturkmen, Loewen, & Ellis, 2002). Classroom research investigating the variables of teacher L1 use and metalanguage have been primarily descriptive (i.e., process-oriented) with few studies investigating their effects on L2 learning (i.e., process-product). An area of narrowly focused L2 classroom observation research, where process-product relationships have been explored, is in teacher

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

200

N i n a S pa da

questioning behaviour. This research has focused on “display” versus “referential” questions, and over the years many studies have reported disproportionately higher numbers of display questions. This is considered problematic because referential questions are characterized as more authentic in nature, more reflective of “real” communicative exchanges, and thus likely to lead to more successful L2 learning. Research that has investigated the effects of types of teacher questions on learner output and L2 learning, however, has been mixed. While some studies have provided evidence that referential questions lead to longer and more complex language use on the part of learners (Brock, 1986; Nunan, 1987), others have indicated the reverse (Long & Sato, 1984; McNeil, 2012). An argument in support of the use of display questions is that they provide scaffolding support for learners and serve important pedagogic and interaction functions (Lee, 2006; McCormick & Donato, 2000). In a large-scale process-product study of teacher questions and learner output in content and languageintegrated learning (CLIL) classes, Dalton-Puffer (2007) investigated the distribution of four different question types (display or referential; open or closed) and their goals (e.g., questions for fact, explanation, reasons, opinions). Results indicated that open-ended questions asking learners for reasons and outcomes were ones that led to the most complex language outcomes. Decisions about whether to use a broad-based or narrow focus to classroom observation is clearly dependent on one’s research goals. There may also be times when both are necessary. An example of this comes from a comparative study of corrective feedback in Japanese and French immersion programmes (Lyster & Mori, 2006). Using the error treatment model described above (Lyster & Ranta, 1997), the researchers documented the teachers’ corrective strategies and learners’ immediate responses to them. The results revealed that the teachers’ corrective feedback strategies in both programmes were the same, but the students’ responses were different. This was perplexing given that both programmes were based on the principles of communicative and content-based language teaching and were assumed to reflect similar pedagogical practices. The researchers decided to use COLT Part A to explore whether there were any differences in the overall communicative orientation of the classes across the two settings that might help to provide an explanation for the observed differences in the corrective feedback interactions. The results indicated a more analytic pedagogic orientation in the Japanese immersion classrooms with a greater focus on accuracy and repetition compared with the French immersion classes in which a more experientially oriented pedagogy was implemented. The researchers would not have been able to fully interpret the differences between these programmes without the use of a comprehensive observation scheme:

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

201

Using two coding schemes to support comparisons of instructional settings can be seen as a collaborative innovation … studies that compare classrooms along a single dimension such as interactional feedback often lack other relevant descriptive data to support the comparison.  (Lyster & Mori, 2006, p.278)

8.7 Conclusion In this chapter, I have summarized the history, development, and use of L2 classroom observation instruments within the interaction analysis tradition. This has been one of the main approaches to conducting observation research in L2 classrooms for many decades and, like all research methods, comes with its strengths and limitations. Other traditions to classroom observation such as discourse analysis and ethnography represent different conceptual frameworks and use divergent methodologies and procedures in the collection, analysis, and interpretation of classroom observation data. These three approaches to classroom observation produce different types of knowledge, insights, and perspectives on L2 classroom life. While some might view them as incompatible, they can also be viewed as complementary. To be sure, most L2 classroom researchers agree that using multiple approaches and methods is preferable to using one. Indeed, a multi-method approach combining objective and subjective elements, quantitative and qualitative data collection and analysis, structured and unstructured approaches, and macro- and micro-level perspectives is considered to be the best way to achieve a complete description of the highly complex and wide range of linguistic, pedagogic, and social behaviours that occur in L2 classrooms. Such a view is consistent with the mixed-methods movement that has been rapidly growing within the field of applied linguistics and other disciplines in recent years. According to Dörnyei (2007), “the combining of several research strategies [in mixed methods classroom research] can broaden the scope of the investigation and enrich the researcher’s ability to draw conclusions” (p. 186). This is consistent with what Allwright and Bailey (1991) argued almost thirty years ago before the term mixed methods was introduced. Similar recommendations have been made by several other L2 researchers, including Nunan (2005), who states, “classroom researchers appear increasingly reluctant to restrict themselves to a single data collection technique, or even a single research paradigm” (p. 237). In conclusion, the evolution of L2 classroom observation within the interaction analysis tradition has taken it from a focus on teacher education to research, from prescription to description, from process to processproduct research, and from comprehensive, broad-based documentations

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

202

N i n a S pa da

of L2 classroom behaviours to more narrowly focused descriptions. While the shift to small-scale process-product research has permitted more feasible investigations of relationships between L2 teaching and learning, it does not negate the need for comprehensive schemes to provide an overall view of what goes on in L2 classrooms. Nevertheless, more research is needed to address some basic methodological issues associated with the use of comprehensive classroom observation schemes. This includes the empirical validation of observation categories, which can be explored through quantitative multi-factorial research designed to investigate the extent to which specific instructional variables and/or clusters of variables are stronger (or weaker) predictors of learning outcomes.

References Allen, J. P. B., & Carroll, S. (1987). Evaluation of classroom processes in a Canadian core French programme. Evaluation & Research in Education, 1, 49–61. Allen, J. P. B., Fröhlich, M., & Spada, N. (1984). The communicative orientation of second language teaching: An observation scheme. In J. Handscombe, R. Orem, & B. Taylor (eds.), On TESOL ’83 (pp. 232–252). Washington, DC: TESOL. Allwright, D., & Bailey, K. (1991). Focus on the language classroom: An introduction to classroom research for language teachers. Cambridge: Cambridge University Press. Ammar, A., & Spada, N. (2006). One size fits all? Recasts, prompts, and L2 Learning. Studies in Second Language Acquisition, 28, 543–574. Basturkmen, H., Loewen, S., & Ellis, R. (2002). Metalanguage in focus on form in the communicative classroom, Language Awareness, 11, 1–13. Bellack, A., Herbert, A., Kliebard, M., & Smith Jr., F. L. (1966). The language of the classroom. New York: Teachers College Press. Beretta, A., & Davies, A. (1985). Evaluation of the Bangalore Project. ELT Journal, 39, 121–127. Borg, S. (1998). Teachers’ pedagogical systems and grammar teaching: A qualitative study. TESOL Quarterly, 32, 9–38. Brock, C. (1986). The effects of referential questions on ESL classroom discourse. TESOL Quarterly, 20, 47–59. Chaudron, C. (1988a). Second language classrooms: Research on teaching and learning. New York: Cambridge University Press. Chaudron, C. (1988b). Classroom research: Recent methods and research findings. In G. Kasper (ed.), AILA review: Classroom research (Vol. 5, pp. 10– 19). Amsterdam: John Benjamins. Chaudron, C. (1991). Validation in second language classroom research and the evaluation of teaching methods. In R. Phillipson, E. Kellerman,

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

203

L. Selinker, M. Sharwood Smith, & M. Swain (eds.), Foreign language pedagogy: A commemorative volume for Claus Faerch (pp. 187–196). Clevedon: Multilingual Matters. Dalton-Puffer, C. (2007). Discourse in content and language integrated learning (CLIL) classrooms. Amsterdam: John Benjamins. Dörnyei, Z. (2007). Research methods in applied linguistics. Oxford: Oxford University Press. Ellis, R. (2012). Language teaching research and language pedagogy. Oxford: Wiley-Blackwell. Duff, P. A. (2002). The discursive co-construction of knowledge, identity, and difference: An ethnography of communication in the high school mainstream. Applied Linguistics, 22, 289–322. Fairclough, N. (1995). Critical discourse analysis. London: Longman. Fanselow, J. (1977). Beyond “Rashomon”—Conceptualizing and describing the teaching act. TESOL Quarterly, 10, 17–39. Fazio, L., & Lyster, R. (1998). Immersion and submersion classrooms: A comparison of instructional practices in language arts. Journal of Multilingual and Multicultural Development, 19, 303–317. Flanders, N. (1970). Analyzing teaching behavior. Reading, MA: Addison Wesley. Fröhlich, M., Spada, N., & Allen, P. (1985). Differences in the communicative orientation of L2 classrooms. TESOL Quarterly, 19, 27–57. Guilloteaux, M., & Dörnyei, Z. (2008). Motivating language learners: A classroom-oriented investigation of the effects of motivational strategies on student motivation. TESOL Quarterly, 42, 55–77. Harklau, L. (1994). ESL versus mainstream classes: Contrasting L2 learning environments. TESOL Quarterly, 28, 241–272. Kim, S., & Elder, C. (2005). Language choices and pedagogic functions in the foreign language classroom: A cross-linguistic functional analysis of teacher talk. Language Teaching Research, 9, 335–380. Lee, Y. (2006). Respecifying display questions: Interactional resources for language teaching. TESOL Quarterly, 40, 691–713. Levin, L. 1972. Comparative studies in foreign-language teaching. Stockholm: Almquist och Wiksell. Li, S. (2010). The effectiveness of corrective feedback in SLA: A meta-analysis. Language Learning, 60(2), 309–365. Lightbown, P., Halter, R., White, J., & Horst, M. (2002). Comprehensionbased learning: The limits of “do it yourself”. Canadian Modern Language Journal, 58, 427–464. Long, M. (1980). Inside the “black box”: Methodological issues in classroom research on language learning. Language Learning, 30, 1–42. Long, M., & Sato, C. (1984). Methodological issues in interlanguage studies: An interactionist perspective. In A. Davies, C. Criper, & A. Howatt (eds.), Interlanguage (pp. 253–280). Edinburgh: Edinburgh University Press.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

204

N i n a S pa da

Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced approach. Amsterdam: John Benjamins. Lyster, R., & Mori, H. (2006). Interactional feedback and instructional counterbalance. Studies in Second Language Acquisition, 28, 269–300. Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms. Studies in Second Language Acquisition, 19, 37–66. Lyster, R., & Saito, K. (2010). Oral feedback in classroom SLA: A metaanalysis. Studies in Second Language Acquisition, 32(2), 265–302. Lyster, R., Saito, K., & Sato, M. (2013). Oral corrective feedback in second language classrooms. Language Teaching, 46, 1–40. Macaro, E. (2001). Analysing student teachers codeswitching in foreign language classrooms: Theories and decision making. The Modern Language Journal, 85, 531–548. Mackey, A., & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (ed.), Conversational interaction and second language acquisition. Oxford: Oxford University Press. Markee, N. (2005). Conversation analysis for second language acquisition. In E. Hinkel (ed.), Handbook of research in second language teaching and learning (pp. 355–374). Mahwah, NJ: Erlbaum. McCormick, D., & Donato, R. (2000). Teacher questions as scaffolding assistance in an ESL classroom. In J. Hall & L. Verplaetse (eds.), Second and foreign language learning through classroom interaction (pp. 183–201). Mahwah, NJ: Lawrence Erlbaum. McKay, S. (2006). Researching second language classrooms. Mahwah, NJ: Lawrence Erlbaum. McNeil, L. (2012). Using talk to scaffold referential questions for English language learners. Teaching and Teacher Education, 28, 396–404. Mitchell, R. (1985a) Communicative interaction research project: Final report. Stirling: Department of Education, University of Stirling. Mitchell, R. (1985b) Process research in second language classrooms. Language Teaching, 18, 330–352. Mitchell, R., Parkinson, B., & Johnstone, R. (1981). The foreign language classroom an observational study. Stirling: University of Stirling. Moskowitz, G. (1976). The FLINT system: An observational tool for the foreign language classroom. In A. Simon & E. Boyer (eds.), Mirrors for behavior: An anthology of classroom observation instruments (pp. 125–157). Philadelphia, PA: Center for the Study of Teaching at Temple University. Nunan, D. (2005). Research methods in language learning. Cambridge: Cambridge University Press. Nunan, D. (1987) Communicative language teaching: Making it work. ELT Journal, 41, 136–145. Polio, C., & Duff, P. A. (1994). Teachers’ language use in university foreign language classrooms: A qualitative analysis of English and target language alternation. The Modern Language Journal, 78, 313–326.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Classroom Observation Research

205

Prabhu, N. S. (1987). Second language pedagogy. Oxford: Oxford University Press. Russell, J., & Spada, N. (2006). Corrective feedback makes a difference: A meta-analysis of the research. In J. Norris & L. Ortega (eds.), Synthesizing research on language learning and teaching (pp. 133–164). Amsterdam: John Benjamins. Savignon, S. (1972). Communicative competence: An experiment in foreign language teaching. Philadelphia, PA: Center for Curriculum Development. Scherer, A., & Wertheimer, M. (1964). A psycholinguistic experiment in foreign language teaching. New York: McGrawHill. Seedhouse, P. (2004). The interactional architecture of the language classroom: A conversation analysis perspective. Malden, MA: Blackwell. Shintani, N., & Ellis, R. (2010). The incidental acquisition of plural -s by Japanese children in comprehension-based and production-based lessons: A process-product study. Studies in Second Language Acquisition, 32(4), 607–637. Simard, D., & Jean, G. (2011). Exploration of L2 teachers’ use of pedagogical interventions. Language Learning, 61, 759–785. Sinclair, J., & Coulthard, M. (1975). Towards an analysis of discourse. Oxford: Oxford University Press. Smith, P. (1970). A comparison of the audiolingual and cognitive approaches to foreign language instruction: The Pennsylvania foreign language project. Philadelphia, PA: Center for Curriculum Development. Spada, N. (1987). Relationships between instructional differences and learning outcomes: A process-product study of communicative language teaching. Applied Linguistics, 8, 131–161. Spada, N., & Fröhlich, M. (1995). COLT. Communicative orientation of language teaching observation scheme. Sydney: National Center of English Language Teaching and Research. Spada, N., & Lyster, R. (1997). Microscopic and macroscopic views of second language classrooms. TESOL Quarterly, 31, 787–795. Ullman, R., & Geva, E. (1985). Expanding our evaluation perspective: What can classroom observation tell us about Core French programs? The Canadian Modern Language Review, 42, 307–323. Watson-Gegeo, K. A. (1988). Ethnography in ELS: Defining the essentials. TESOL Quarterly, 22, 575–592.

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

ACTIVITIES & EPISODES

2

TIME

1

T↔S/C

3

S↔S/C

4

Choral

5

Same task

6

Different tasks

7

Same task

8

Different tasks

9

Procedure

10

Discipline

11

Form

12

Function

13

14

15

Socioling.

Language Discourse

Group Indiv. Manag.

Other topics

16

17

Broad

Class Teacher/Text

18

19

Teacher/Text/Stud.

Page CONTENT CONTROL

20

Student

CONTENT

21

Listening

Date STUDENT MODALITY

22

Speaking

PARTICIPANT ORGANISATION

23

Reading

Subject

Narrow

Visit No

24

Other

25

26

27

28

29

30

Source

MATERIALS Type Text Minimal

Observer

Extended

Lesson (min.)

Audio

Grade(s)

Visual

Teacher

L2-NNS

School

Writing

© Spada & Fröhlich 1995

31

L2-NS

COLT PART A

32

L2-NSA

Communicative Orientation of Language Teaching Observation Scheme

33

Student-made

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 21 Jul 2019 at 07:39:37, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.009

A

E

M

L1

2

Off task

1

L2

3

Predict.

4

6

7

Giving Info.

Unpred.

5

Request. Info. Genuine requ.

Pseudo requ.

Reaction Sustained to form/ speech message

Minimal

8

Sustained

9

Form

10

Message

11

Correction

12

Incorporation of student utterances

Repetition

13

Paraphase

14

Comment

15

Expansion

16

Clarif. request

17

Elab. request

18

Discourse initiation

19

Target language

L1

20

21

L2

22

23

24

25

Request. Info.

Giving Info.

Sustained speech

Ultra-minimal

26

Minimal

27

Sustained

28

Form restriction

Choral

29

30

31

Reaction to form/ message

STUDENT VERBAL INTERACTION

Restricted

Information gap

Unrestricted

TEACHER VERBAL INTERACTION

32

Form

Information gap

Unpred.

Predict.

Date of visit

33

Message

Target language

Genuine Requ.

Pseudo Requ.

Coder

34

Correction

Subject

Incorporation of student/teacher utterances

35

Repetition

Teacher

36

Paraphase

© Spada & Fröhlich 1995

37

Comment

School

38

Expansion

Communicative Orientation of Language Teaching Observation Scheme

39

Clarif. request

COLT PART B: COMMUNICATIVE FEATURES

40

Elab. request

9 Psycholinguistic and Neurolinguistic Methods Leah Roberts 9.1 Introduction Language learning involves more than acquiring knowledge of the target language; learners must also be able to put their knowledge to use during real-time processing, and in the past twenty years, sophisticated psycholinguistic and neurolinguistic methods have been employed to allow for a deep investigation into these processes. A number of techniques, such as self-paced reading and cross-modal priming rely on speed responses (for instance a button-push) to infer the underlying comprehension processes that are involved in the real-time processing of linguistic material. Specifically, in comparison to a control condition, a slower response indicates difficulty in, for instance, lexical access, or the processing of ambiguous, complex, or ungrammatical input, at certain points in a sentence. Eye-tracking during reading can also be used to examine lexical and sentence processing in a similar way, but this method can add detail to the study of the comprehension processes via the examination of specific eye-movement measures thought to tap early (first fixations) and later comprehension processes (go-past times and regressions). Currently, expectation/prediction is an important topic in second language acquisition (SLA) processing research (e.g., Hopp, 2013; Kaan, 2014), and visualworld eye-tracking can be usefully employed in this regard, as one can explore listeners’ expectations of, for instance, a continuation of a sentence, or a referent for a pronoun. As well as being interested in the kinds of words or sentences that cause processing difficulty, language and language learning researchers also investigate types of information that are being accessed or applied during online comprehension, and neurolinguistic methods can be used for this purpose, in particular electroencephalography (EEG). This is because certain types of violations and/or unexpected linguistic material (syntactic, semantic, pragmatic) can elicit different types of event-related potential (ERP) components in the EEG signal, both in terms of polarity (negative versus positive) and timing (i.e., the onset of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

209

the component). So, for instance, difficulty with semantic integration of a current linguistic element may elicit a negative-going waveform, with an onset of approximately 400 ms following the beginning of the critical word (the so-called N400 component). The EEG technique thus provides highly time-sensitive data; however, it does not provide reliable topological information. For this, functional magnetic resonance imaging (fMRI) is more useful, as it can tell us which parts of the brain and which networks of neurons are most activated under various experimental conditions. In sum, using these techniques researchers can examine sentence processing or parsing procedures (activation of words and their integration into the ongoing analysis of a sentence) and for second language (L2) researchers, often this entails a comparison with native speakers’ (NS’) processing, informing theories of (L2) parsing (e.g., Clahsen & Felser, 2006; Cunnings, 2017; Kaan, 2014). For L2 researchers, these methods can also be employed to investigate the types of language that are assumed to be grammatical (or not) for language learners, in particular, looking at implicit knowledge, in comparison to explicit or metalinguistic knowledge which is arguably tapped into via offline tasks such as grammaticality judgments (Ellis, 2005; Tokowicz & MacWhinney, 2005). Focusing on such questions can speak to issues in SLA theory, for instance, whether late (post-puberty) L2 learners’ persistent problems with morphosyntax reflect a representational deficit (e.g., Hawkins & Liszka, 2003; Tsimpli & Dimitrakopoulou, 2007) or a processing problem, caused by difficulty in accessing appropriate grammatical knowledge online (e.g., Goad & White, 2006; Prévost & White, 2000; for a review, see Slabakova, 2009). Psycholinguistic and neurolinguistic techniques can also be applied to questions surrounding L2 learners’ ability to acquire phenomena at the (external) interface levels, such as syntax-discourse (e.g., Sorace, 2011; Sorace & Serratrice, 2009; Tsimpli & Sorace, 2006). In this chapter, I will discuss the use of such techniques in SLA research and critically appraise their ability to answer important theoretical topics in SLA research.

9.2  Psycholinguistic Methods 9.2.1  Self-Paced Reading Research on monolingual sentence processing has shown that it is incremental, that is, immediately on encountering each word in the input, grammatical knowledge is applied (the input is parsed). Furthermore, semantic, pragmatic, discourse, and world knowledge are accessed, and the plausibility and appropriateness of the concurrent analysis is assessed (for detailed discussion on parsing models, see van Gompel, 2013). By far the

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

210

Leah Roberts

most common method in L2 studies for investigating real-time sentence comprehension in general and incremental processing in particular has been self-paced reading (SPR; also known as the moving window technique; Just, Carpenter, & Woolley, 1982). This has been used both to examine readers’ parsing procedures as they integrate each word in the incoming input to create an interpretation of the sentence and the extent to which they are sensitive online to (un)grammatical phenomena. I will discuss each of these applications in the following section.

9.2.1.1  Self-Paced Reading and Incremental Processing Evidence for incremental processing in both native speakers and L2 learners has most often come from self-paced reading studies of temporarily ambiguous (garden-path) sentences like 1a (Juffs & Harrington, 1995; 1996). The time it takes for readers to read through a sentence is recorded via button pushes which bring up each new word of the input. Specifically, readers of such sentences often initially analyse a temporarily ambiguous determiner phrase (DP) (the water) as a direct object of a preceding verb (drank). This evidence for incremental processing is shown by processing slowdowns (slower button-pushes) on and following the disambiguating word, in this case when encountering the main verb (proved). Readers must then reanalyse the DP as subject of this main verb. Critically, reading times (RTs) are measured in comparison to conditions, in which, for instance, the previously encountered verb does not subcategorize for the DP, “arrived” (1b). (1)

a. After Bill drank the water proved to be poisoned. b. After Sam arrived the guests began to eat.

Juffs and Harrington (1995; 1996) found that their Chinese L2 learners performed similarly to native speakers, with garden-path sentences leading to slower processing than those containing intransitive verbs (1b), demonstrating both incremental processing and a sensitivity to subcategorization information in online processing. Incremental processing has also been found in a number of other L2 processing studies employing the SPR method (Dussias and Cramer Scaltz, 2008; Juffs & Harrington, 1995; 1996; Roberts & Felser, 2011; Juffs & Rodriguez, 2014). In fact, L2 learners are often highly sensitive to lexical-semantic information during parsing. For instance, Dussias and Cramer Scaltz (2008) found a similar sensitivity to the semantic biases of verbs. The reading times of their Spanish learners of English in an SPR task demonstrated that the learners had comparative processing difficulty when sentence continuations went against the verb’s bias for taking either a direct object (2a; confirmed the rumour could mean …) or a sentence complement (2b; admitted the mistake …). (2)

a. The CIA director confirmed the rumour could mean a security leak. b. The ticket agent admitted the mistake when he got caught.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

211

Although the learners appeared native-like in their processing of these sentences, when they encountered verbs whose biases for taking sentence complements or direct objects were different between Spanish and English, this influenced their parsing, indicating an L1 transfer effect (see Frenck-Mestre & Pynte, 1997, for similar findings using eye-tracking during reading). These data and others from SPR studies of the processing of garden-path sentences show that, like native speakers, learners can incrementally process the input and can access and employ rapidly subcategorization information of L2 verbs during the real-time interpretation of sentences. However, there is evidence that learners may differ from native speakers in that they appear to have more difficulty in recovering from misanalyses of the input. For example, the learners in Juffs and Harrington (1996) were found to have problems correctly judging items that were temporarily ambiguous but ultimately grammatical (Sam warned the student cheated on the exam). Researchers can therefore employ end-of-sentence tasks like comprehension questions or acceptability judgments to assess the extent to which readers recover from processing problems. Roberts and Felser (2011) also found that learners had difficulty recovering from misanalyses in an SPR study with advanced Greek learners of English. Rather than employing a metalinguistic task together with the SPR experiment, the authors manipulated the pragmatic plausibility of their experimental items as a diagnostic to investigate recovery from temporary ambiguities (see also Pickering & Traxler, 1998), as shown in 3. (3)

The journalist wrote the book (the girl) had amazed all the judges.

The learners were garden-pathed as shown by the fact that their reading times indicated that they attempted to integrate the ambiguous noun phrase (book/girl) with the main verb (wrote). Specifically, they were slower when this misanalysis led to an implausible sentence fragment (wrote the girl) in comparison to plausible conditions (wrote the book). Following this, the plausibility of their initial analysis affected how fast they were to read the remainder of the sentence, an indication of how comparatively easy recovery was. That is, on and beyond the disambiguating verb phrase (had amazed), RT patterns were reversed such that processing difficulty was elicited in the plausible condition (wrote the book had amazed all the judges) compared to the implausible condition (wrote the girl had amazed all the judges). The processing assumption here is that readers are more strongly committed to a semantically (or pragmatically) initial plausible interpretation (wrote the book had amazed), which is harder to abandon in the face of new evidence in comparison to an initial analysis that was implausible. In sum, the learners in this study showed online recovery from their initial misanalyses. This online recovery effect for ambiguous complement

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

212

Leah Roberts

clause constructions (3) was not, however, in evidence with more structurally complex sentences, such as preposed adjuncts (4), where the learners were found to be garden-pathed, but no online recovery (as in a reversal of reading times) was observed. (4)

While the band played the song (the beer) pleased all the customers.

In sum, the results from SPR garden-path studies show that, like native speakers, learners incrementally process the input, but information from their L1 (such as differences in subcategorization properties of verbs) and structural complexity can affect their online processing decisions and the ability to incrementally process the sentence following any initial misanalysis.

9.2.1.2  Self-Paced Reading and Assessing Grammatical Knowledge SLA researchers have also used SPR to good effect in the study of L2 learners’ grammatical knowledge. Specifically, studies focus on the extent to which learners are sensitive to grammatical violations during L2 processing (e.g., Hopp, 2010; Roberts & Liszka, 2013; Sagarra and Herschensohn, 2011), and the task can therefore be employed to assess a learner’s current interlanguage (Selinker, 1972). It is important to note that L2 learners (of sufficient proficiency) most often perform like native speakers when they are required to undertake a metalinguistic task at the same time (e.g., Hopp, 2010). A number of word-by-word SPR studies have found that learners may not slow down when processing grammatical violations, yet they are able to assess the items as unacceptable or ungrammatical in a separate offline task, particularly if the feature being examined is not instantiated in their L1 (e.g., for subject–verb agreement violations with Chinese L2 learners of English, see Jiang, 2004, and for Spanish and Chinese L2 learners, see Lee, 2002). For example, Roberts and Liszka (2013) examined advanced French and German L2 learners’ ability to employ their knowledge of English tense/aspect in a word-by-word SPR task, focusing on mismatches between a fronted adverbial and the following tensed verb. (5) Yesterday/*Since yesterday, John went swimming three times. (Past Simple). (6) Since yesterday/*Yesterday, John has been swimming three times. (Present Perfect).

Both groups of learners demonstrated explicit knowledge of the constructions under investigation in a cloze production task and an offline acceptability judgment task. Despite this, online, an effect of the learners’ L1 was in evidence. That is, for the past simple (5), both the Germans and the French were slower to read the ungrammatical items, whereas only the French learners showed the same sensitivity to the violation in the present perfect condition (6). The authors argue that these findings can be

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

213

explained in terms of transfer, that is, tense is instantiated in all the languages of concern (English, French, German), but aspect is only grammaticalized in French and English. Such findings support the theoretical view that only features available from the L1 are ultimately fully acquirable to native-like levels (see Hawkins & Liszka, 2003) or that explicit knowledge is obtainable, but it may not be possible to employ such grammatical knowledge in online tasks, which is assumed to tap more implicit knowledge (Ellis, 2005; for eye-tracking evidence, see Keating, 2009). However, a number of researchers argue against this position. For instance, Hopp (2010) reports the results of a series of experiments using traditional grammaticality and speeded grammaticality judgment tasks (GJTs), as well as SPR for comprehension with advanced and near-native Russian, Dutch, and English learners of German. The author examined the learner’s real-time grammaticality judgments on sentences involving number (7) and case agreement (8) violations. Sensitivity to the ungrammaticalities was assumed to be in evidence in reading time slow-downs at the point at which the ungrammaticality occurs (haben in 7 and Hund in 8). (7) *Er glaubt, dass der Förster im vorigen Jahr den Angler umgebracht haben. (He believes that the forester (SG) in the previous year the fisherman (SG) killed have (PL)). (8) *Er glaubt, dass die seit langem vermisste Hund im Garten gefunden wurde. (He believes that the (FEM) since long missed (MASC) dog in the garden found was.)

The author examined word-by-word reading times as well as offline grammaticality judgments. The highly advanced learners performed like native speakers, being slower to process the ungrammatical in comparison to the grammatical items and judging the former as less acceptable. One could argue, however, that asking learners to make metalinguistic judgments during online processing can lead them to pay more attention to grammatical information, and thus they are more likely to display “online” knowledge akin to native speakers under these circumstances. The extent to which such tasks actually tap truly implicit knowledge requires much further investigation (see also Roberts, 2013, for more discussion).

9.2.2 Eye-Tracking Eye-tracking is another common technique that has been successfully employed to investigate real-time processing in the L2, for the study of reading processes and comprehension during auditory processing (the visual-world paradigm). Research using the former method has often focused on temporary ambiguities and ungrammaticalities, as with SPR, and also on various types of referential or anaphor resolution. Anaphoric processes such as pronoun resolution can also be investigated by use of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

214

Leah Roberts

the visual-world technique, and, more recently, it has been employed to look at L2 learners’ predictive processing. Some of these studies will be discussed in the following sections.

9.2.2.1 Eye-Tracking During Reading—Examining Ambiguity, Ungrammaticality, and Referential Processing During reading, rapid eye movements are made from one fixation point to another. These movements or saccades are too fast for new information to be processed (Liversedge, Paterson, & Pickering, 1998; Rayner, 1998, 2009), but in between them, the eyes remain stationary (fixations) long enough for word recognition to take place (Rayner, 1998, 2009). Comprehension processes are inferred according to the number and duration of such fixations. Thus, as is the case with slower button-pushes during SPR, longer fixation times and a greater number of fixations can indicate processing difficulty. As well as moving progressively from left-to-right through a text while reading, often the eyes move from right-to-left. These latter movements are regressions and may be elicited by comprehension difficulties and failures. For instance, on encountering an unexpected or ambiguous word, a reader may regress to earlier parts of the sentence or discourse in an attempt at comprehension. Alternatively, highly predictable and frequent words are often skipped over. Therefore, taken together, via eye-movement recordings, we can see what has been fixated on, re-fixated on (and for how long), or indeed skipped altogether, during the uninterrupted processing of the input, which provides a highly rich account of real-time and comparatively natural comprehension processes (FrenckMestre, 2005), and this is therefore a great strength of this method, particularly in comparison to SPR (e.g., Duyck et al., 2007). Another strength of this technique in comparison to SPR is that early versus late processes in reading comprehension can be teased apart. Specifically, it is thought that first fixations and first pass durations reflect the earliest stages of processing, such as lexical access and immediate integration of semantic and grammatical information. Total reading times are assumed to reflect later stages of processing, such as re-integration following difficulties and/or recovery from misanalyses (Paterson, Liversedge, & Underwood, 1999; Rayner et al., 1989). Eye-tracking during reading has been used to good effect in the study of L2 learners’ incremental processing, with similar findings to those outlined above for SPR. However, given the greater sensitivity of this method versus SPR, a more fine-grained picture of processing can be obtained. For instance, in an early study by Frenck-Mestre and Pynte (1997), advanced L2 learners of French read sentences in French which were similarly ambiguous in both languages (9), where the critical prepositional phrase (PP) is ambiguous in that it can attach to either the preceding verb, “VP-attachment” (9a), or noun, “NP-attachment” (9b).

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

215

(9) a. VP-attachment Il accuse son chef de meurtre mais il ne peut pas fournir de preuve. “He accuses his boss of murder but he cannot provide proof.” b. NP-attachment Il accuse son chef de service mais il ne peut pas fournir de preuve. “He accuses his head of department but he cannot provide proof.”

The authors found that both learners and native speakers processed the VP-attachment items more quickly than the NP-attachment constructions, as per earlier monolingual research findings (Rayner, Carlson, & Frazier, 1983). A second set of items used monotransitive verbs (10), to see whether the L2 learners, like native speakers, would be guided in their early parsing by the lexical information in the initial verb, and thus prefer NP-attachment (10b) and VP-attachment (10a) because only one argument is allowed. The predictions were indeed borne out, as in the earliest measures of processing (first fixations and first pass times) an NP-attachment preference was observed in the shorter fixation durations in comparison to the VP-attachment condition. Thus, since the L2 learners’ parsing decisions were affected by lexical information (subcategorization properties of the verbs) in the very early parsing stages, the authors argue that their processing (at least in their use of lexical-semantic information) during real-time comprehension appears to be qualitatively highly similar to native speakers’. (10) a. VP-attachment Il connaît la femme de vue mais ne se rappelle plus son nom. “He knows the woman by sight but no longer remembers her name.” b. NP-attachment Il connaît la femme de chambre mais ne se rappelle plus son nom. “He knows the chambermaid but no longer remembers her name.”

Early processing effects were also observed in a second experiment. The authors examined the incremental processing of garden-path experiments with sentences which were unambiguous in the target language (French) but temporarily ambiguous in English, using verbs which were intransitive in French (obéir) but optionally transitive in the learners’ L1, English (obey), which would thus lead to a “garden-path effect” if translated into the learners’ L1 (11). (11) Chaque fois que le chien obéissait la jolie petite fille montrait sa joie. “Every time the dog obeyed the pretty little girl showed her approval.”

Later fixation time measures (total reading times) showed no differences between the two groups, but for the learners, the very earliest fixations reflected a brief slowdown on those items which were ambiguous in their L1. In sum, initial incremental processing in the L2 was affected by cross-linguistic differences, which were not picked up in later processing measures.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

216

Leah Roberts

Eye-tracking during reading is therefore a useful tool to investigate early versus late processes during reading and can therefore offer a more detailed picture of L2 processing than SPR. The eye-tracking technique has also been used to test sensitivity to violations. For example, in Keating (2009), the knowledge and processing of grammatical gender in adult English L2 learners of Spanish was examined, since earlier research had found conflicted findings as to whether English learners were able to demonstrate knowledge of this feature (Franceschina, 2001; Sabourin, Stowe, & de Haan, 2006; White et al., 2004). The author used constructions in which the distance between the NP and the postnominal adjective differed in distance (12). This is an interesting construction to test, because readers need to incrementally process the input while keeping the noun in working memory until they integrate it with its modifier, which is thought to put the processing system under pressure (Gibson et al., 1996). (12)





a. In the DP [IP Una casa pequeña [VP cuesta mucho en San Francisco.]] “A small house costs a lot in San Francisco.” b. In the matrix clause VP [IP La casa [VP es bastante pequeña y necesita muchas reparaciones.]] “The house is quite small and needs a lot of repairs.” c. In the subordinate clause VP [IP Una casa [VP cuesta menos [CP si [VP es pequeña y necesita reparaciones.]]]] “A house costs less if it is small and needs repairs.”

Processing difficulty (and by extention an online sensitivity to gender agreement errors) was reflected in the eye-tracking data when there was a mismatch in gender between the two (e.g., un libr-o [masc] … pequeñ-a [fem]) by both longer fixation times on the critical adjective (e.g., pequeña) and the number of regressions from the adjective back to the modifying noun. Keating’s learners differed in their proficiency, and only those advanced learners of Spanish showed reading profiles akin to native speakers on the longer distance items (12b and c). The less proficient learners showed the same fixation patterns as the natives and higher proficient learners (longer total reading times and more regressions) only in the “local” dependency condition, where the noun and modifier were adjacent (12a). Such data is argued to be supportive of SLA theories that predict that learners are indeed able to ultimately attain grammatical features not available from transfer from the L1 (c.f. the Full Transfer Full Access Hypothesis; Schwartz & Sprouse, 1996), contra the findings of many offline studies (Franceschina, 2001; White et al., 2004) and against competing theories such as the Failed Functional Features Hypothesis (Hawkins & Liszka, 2003). Furthermore, such data suggest that L2 learners, if they are of high enough proficiency, do not have more difficulty than native speakers in computing structural dependencies across clause boundaries (against the Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

217

predictions of the Shallow Structure Hypothesis; Clahsen & Felser, 2006). Keating (2009) did not find effects of ungrammaticality in first fixation for any of his groups, but rather the violation was reflected only in later processing measures (total reading times or regressions after the critical regions). This may be because in his task the participants were instructed to read for meaning and to undertake a meaning-related task (translation matching) after each sentence. In contrast to this, Foucart and FrenckMestre’s (2012a) gender violation sentences (13) elicited longer first pass fixations, as well as total fixation times and regressions, and this may be because the participants were asked to perform an acceptability judgment task at the same time, which arguably directed them to the violations. (13) Au printemps les pommes (fem) son vertes (fem) (*verts [masc]) sur cet arbre. “In spring apples are green on this tree.”

In sum, as with SPR studies, it is important to consider the task(s) that learners (and NSs) are asked to perform together with the online experiment. Overall, it seems that the more metalinguistic the accompanying task, the earlier processing effects are observed, and the more native-like learners appear to be, but this may be caused by the fact that making judgments can push learners to employ more explicit knowledge during online processing (see Roberts, 2013 for an overview). The eye-tracking technique has also been used to explore referential processing in L2 learners and questions concerning the online application of syntactic constraints (e.g., binding). An example comes from reading studies by Felser, Sato, and Bertenshaw (2009) and Felser and Cunnings (2011) who looked at the processing of reflexive pronouns by advanced Japanese and German learners of English. The authors investigated whether, like native speakers, their learners would attempt to link immediately a reflexive anaphor with a grammatically appropriate antecedent NP in the earlier discourse (according to Principle A of the Binding Theory; Chomsky, 1981). Short texts were read (14) which included an introductory sentence containing two referents (e.g., John and Richard), and which were then reintroduced. Only the local referent was “binding-accessible” (Richard) and the other was “binding-inaccessible” but prominent in the discourse (John/Jane). Of interest were reading times on the reflexive pronoun himself/herself and the following few words (to catch spill-over effects). (14) John (Jane) and Richard were very worried in the kitchen of the expensive restaurant. John noticed that Richard had cut himself (herself) with a very sharp knife.

For items in which there was a mismatch between the gender of the pronoun (herself) and the local, binding-accessible antecedent (Richard), longer first pass fixations as well as total reading times, reflecting later processing,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

218

Leah Roberts

were in evidence for the native speakers. The learners’ fixation patterns were similar to those of the natives in this regard. However, for both the German and the Japanese learners, a fleeting effect of “interference” was observed, in that first fixation and first pass reading times were longer on the reflexive anaphor when there was a gender mismatch. That is, despite only German being structurally similar to English in the binding of reflexives, for both groups, when the discourse prominent, binding-inaccessible referent differed in gender from the reflexive pronoun (Jane … Richard … himself; see also Roberts, Gullberg, & Indefrey, 2008, for similar “disruption” effects in L2 processing of pronouns). The results could be argued to support theories of L2 acquisition that argue that L2 learners may have trouble with phenomena that lie at the (external) interfaces (e.g., between discourse-pragmatics and syntax; c.f. the Interface Hypothesis; Sorace, 2011; Tsimpli & Sorace, 2006) and that, therefore, differences may reflect processing rather than underlying competence issues.

9.2.2.2  The Visual-World Paradigm: Predictive Processing In the visual-world paradigm, participants listen to sentences while their eyes are tracked as they look at a visual array. For example, research has shown that within 200  ms of hearing a word in the input, the eyes will move to an image of the word that appears on the screen (Altmann, 2011). Thus, the visual-world eye-tracking can be successfully used to investigate word recognition (e.g., Marian & Spivey, 2003) and anaphor resolution, for instance, the real-time interpretation of subject pronouns in auditory discourse (Ellert, 2011; Wilson, 2009). Of recent theoretical interest has been the study of predictive processing in L2 comprehension, and this technique has been proved to be useful in this regard. For instance, Dussias et al. (2013) recorded the eye-movements of English and Italian learners of Spanish as they heard sentences, while looking at a display on which appeared a picture of a masculine (15) and feminine noun. Native speakers have been found to look at the picture of the corresponding noun as soon as they hear the determiner (in 15 el); thus they predict immediately the following noun on the basis of the gender of the determiner (Lew-Williams & Fernald, 2010). (15) El estudiante estaba dibujando el reloj que vio ayer. “ The student was drawing the clock that he saw yesterday.”

The learners in the Dussias et al. study did not pattern with the native speakers, as their eyes moved towards the correct target only on hearing the noun itself (reloj), including even the Italian learners, in whose L1 gender is instantiated. Similar findings have been observed in other visual-world eye-tracking studies with other languages (e.g., Grüter, Lew-Williams, & Fernald, 2012; Grüter, Rohde, & Schafer, 2014; Hopp, 2013; Martin et al., 2013), although the ability appears to increase with greater proficiency (see Hopp, 2013). Such findings have led to the theory that L2 learners may

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

219

be less able to predict upcoming information during real-time processing because their (lexical) representations are less established than those of native speakers (c.f., Reduced Ability to Generate Expectations during language processing; Grüter, Rohde, & Schafer, 2014; Kaan, 2014).

9.3  Neurolinguistic Methods 9.3.1 EEG/ERPs Most L2 processing studies to date have employed psycholinguistic methods such as SPR and eye-tracking, and such techniques can arguably tap implicit processes to some extent. However, a more sensitive method for this purpose is to take EEG (electroencephalogram) recordings while comprehenders are processing the input, because it can provide highly finegrained temporal resolution information. The use of this technique most often involves the examination of which event-related potential (ERP) components (which reflect the simultaneous activation of populations of neurons which are time-locked to a specific stimulus such as a word) are elicited by different types of violation, and thus can be seen to be of great interest to SLA researchers. ERP responses are time-locked in the EEG recording to the word or segment of the anomalous item in comparison to a control condition (e.g., The winner of the big trophy has/* have proud parents), and researchers compare the brain responses to the grammatical versus the ungrammatical items during reading or listening. Language researchers have found this technique useful because different types of violations elicit qualitatively and quantitatively different ERP components. For instance, lexical integration problems consistently elicit a negative-going waveform with a peak in amplitude at approximately 400  ms following the onset of the critical word (the N400 effect). The N400 component has been reliably elicited in L2 processing studies, particularly with learners of high proficiency (Kotz, Holcomb, & Osterhout, 2008; Proverbio, Cok, & Zani, 2002; Weber & Lavric, 2008; Zawiszewski et al., 2011). Much sentence processing research has, however, focused on grammatical processing. Syntactic agreement and integration violations tend to provoke the so-called P600 component, a positive-going wave, peaking at about 600  ms, in both monolingual and L2 studies (e.g., Alemán Bañón, Fiorentino, & Gabriele, 2012; Hagoort & Brown, 1999; Kolk et al., 2003; Nevins et al., 2007; Wicha, Moreno, & Kutas, 2004), and in some studies the P600 is preceded by a left-anterior negativity (LAN) at approximately 200–300 ms (e.g., Barber & Carreiras, 2005; O’Rourke & Van Petten, 2011). There is debate in the literature as to what exactly the P600 may be indexing. Some argue that it reflects later, repair operations (semantic and syntactic reanalysis, for instance), and it is therefore assumed at least to be an

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

220

Leah Roberts

indirect correlate to syntactic processing in contrast to the lesser observed LAN effect, which is elicited by earlier, more automatic processes (e.g., Friederici, Hahne, & Mecklinger, 1996; Molinaro, Barber, & Carreiras, 2011; Molinaro et al., 2015). Some studies have found a comparable P600 effect in L2 learners to that of native speakers when processing agreement violations, for instance in a study by Tokowicz and MacWhinney (2005), who examined violation processing in English L2 learners of Spanish. However, such effects appear to be strongest when the constructions under investigation are similar in the learners’ L1 and L2 (tense violations; 16a). The P600 effect may be reduced or even not appear for grammatical phenomena that differ between the two languages (determiner–noun number agreement; 16b) or are unique to the target language (gender agreement; 16c). (16)

a. Su abuela *cocinando/cocina muy bien. His grandmother *cooking/cooks very well. b. *El/Los niños están jugando. *The (s.)/the (pl.) boys are playing. c. Ellos fueron a *un/una fiesta. They went to *a (m.)/a (f.) party.

Such evidence of an influence of a learner’s L1 have been observed in a number of other ERP studies (Chen et al., 2007; Foucart & FrenckMestre, 2012b; Ojima, Nakata, & Kakigi, 2005; Sabourin and Stowe, 2008; Zawiszewski et al., 2011) and have led to the view that constructions that are different from or not available in a learner’s L1 are not salient enough to be picked up by such implicit techniques. However, in comparison to psycholinguistic studies of L2 processing, ERP work is rare, and it may be that a number of factors influence the extent to which native-like components can be provoked, such as age of onset of L2 acquisition (Weber-Fox and Neville, 1996), proficiency (Ojima, Nakata, & Kakigi, 2005; Rossi et al., 2006), and type of exposure (Caffarra et al., 2015, for a detailed overview). One should take care in the interpretation of ERP data, however, since there is recent evidence that individuals may differ in their responses to the same input stimuli in different ways, specifically, that there may be some individuals who are more N400- and some who are more P600dominant (Tanner et al., 2013; Tanner, Inoue, & Osterhout, 2014). It has been argued that the biphasic responses that have been observed in the processing of agreement violations may be artefactual, caused by researchers averaging over these different sub-groups of participants (see Tanner & van Hell, 2014, for a review and discussion). Of great interest to SLA researchers is language learners’ development over time, and there have been some studies in this area using the ERP technique in longitudinal (Davidson & Indefrey, 2011; McLaughlin et al., 2010; Morgan-Short et al., 2010) rather than cross-sectional studies (e.g.,

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

221

Hahne & Friederici, 2001; Friederici, Steinhauer, & Pfeifer, 2002; Rossi et al., 2002). For instance, Osterhout et al. (2006) tested beginning learners of French over time, on subject–verb agreement violations. Following one month’s training, the ERP study found that the violations elicited an N400. After four months of classroom training, a P600 effect was observed for the same grammatical violations. This suggests that in the very early stages of acquisition, grammatical violations are treated in a comparable way to infrequent lexical items, and only later do grammatical processes come into play (Steinhauer, White, & Drury, 2009; Osterhout et al., 2006; see Roberts et al., 2018 for discussion).

9.3.2 fMRI Another neurolinguistic technique that can be employed to examine L2 processing is functional magnetic resonance imaging (fMRI), although there are comparably fewer studies employing this methodology. While the ERPs can offer highly time-sensitive data, fMRI can address questions regarding which areas of the brain may be active during language processing, as the method records activation via minute changes in blood flow in the brain. Complete sentences are presented to participants (in contrast to most ERP studies), while activation is charted across the entire sentence. Hagoort and Indefrey (2014) present a meta-analysis of fMRI language processing studies and conclude that different populations of neurons may be recruited for the processing of semantic versus syntactic violations, specifically, violations that involve syntactic phenomena elicit stronger activation in left-posterior inferior frontal gyrus (IFG; Brodmann Area 44) compared to semantic violations. Another area relevant for complex syntactic processing is the left inferior frontal gyrus (LIFG), an area that also appears to subserve working memory (see Friederici, 2011, for a comprehensive review). Most fMRI research with L2 populations shows that there is little qualitative difference between learners and natives, particularly when the learners are of a high proficiency. Specifically, it seems that the same brain regions are involved in language processing, but in L2 learners, activation is larger and more extended, which arguably reflects the greater processing cost involved in L2 versus L1 processing in general (Rüschemeyer et al., 2005; Rüschemeyer, Zysset, & Friederici, 2006), and similar effects have been reported for late versus early L2 learners (Hernandez, Hofmann, & Kotz, 2007; see Roncaglia-Denissen & Kotz, 2016, for an overview). One study on L2 grammatical processing using fMRI which found that L2 learners differed from native speakers is that of Suh et al. (2007). The authors investigated brain activation in KoreanEnglish bilinguals’ processing of complex (embedded; 17) and less complex (conjoined; 18) constructions in both languages.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

222

Leah Roberts

(17) (18)

The director that the maid introduced ignored the farmer. The maid introduced the director and ignored the farmer.

The authors found involvement of the areas predicted for language processing for both L1 and L2 processing; however, only a difference between the embedded and the conjoined condition was found during L1 (Korean) processing, with greater activation for the former, more complex construction. In contrast, no difference was observed between the two sentence types during L2 processing. One could argue that these results support the view that L2 grammatical processing may be fundamentally different from L1 processing (c.f. Clahsen & Felser, 2006); however, given that there are so few fMRI studies, and that most have found parallel types of activation, much more research with this method is needed before solid conclusions can be made.

9.4 Conclusion Psycholinguistic and neurolinguistic methods have been applied in the area of SLA research relatively recently, and there is much more research employing psycholinguistic techniques, particularly self-paced reading. Nevertheless, such online methods can be employed to great effect in the study of what kinds of information L2 learners may be able to employ during their real time comprehension of the target input. SPR and eyetracking research has provided evidence that L2 learners can incrementally process the input like native speakers and can make use very rapidly of lexical-semantic information to inform parsing decisions. They may be at a disadvantage, however, with more highly complex constructions and those involving structural reanalysis processes. Learners of high proficiency may be sensitive online to grammatical violations, particularly those which involve features instantiated in their L1. However, researchers need to take care in choosing which methods to combine with online tasks, since there is consistent evidence that L2 learners process more akin to native speakers when required to make metalinguistic judgments during comprehension. Therefore, if one wishes to investigate implicit processes, such supplementary tasks should be avoided. Neurolinguistic methods such as ERP and fMRI recording can offer insights into implicit processes, but we are only beginning to understand if, and, if so, to what extent, L2 learners’ processing involves the same brain regions and underlying processes as those of native speakers. Nevertheless, as this overview makes clear, data from psycholinguistic and neurolinguistic methods can address questions of central concern to SLA researchers, such as how a learner’s L1 may influence their knowledge and processing of the L2, what

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

223

effects there may be of individual differences such as proficiency, and, by extension, how such data can help push forward theories of SLA in general.

References Alemán Bañón, J., Fiorentino, R., & Gabriele, A. (2012). The processing of number and gender agreement in Spanish: An event-related potential investigation of the effects of structural distance. Brain Research, 1456, 49–63. Altmann, G. (2011). Language can mediate eye movement control within 100 milliseconds, regardless of whether there is anything to move the eyes to. Acta Psychologica, 137, 190–200. Barber, H., & Carreiras, M. (2005). Grammatical gender and number agreement in Spanish: An ERP comparison. Journal of Cognitive Neuroscience, 17, 137–153. Caffarra S., Molinaro, N., Davidson, D., & Carreiras, M. (2015). Second language syntactic processing revealed through event-related potentials: An empirical review. Neuroscience and Biobehavioural Reviews, 51, 31–47. Chen, L., Shu, H., Liu, Y., Zhao, J., & Li, P. (2007). ERP signatures of subject– verb agreement in L2 learning. Bilingualism, Language and Cognition, 10, 161–174. Chomsky, N. (1981). Lectures in government and binding. Dordrecht: Foris. Clahsen, H., & Felser, C. (2006). How native-like is non-native language processing? Trends in Cognitive Sciences, 10, 564–570. Cunnings, I. (2017). Parsing and working memory in bilingual sentence processing. Bilingualism: Language and Cognition, 20, 659–678. Davidson, D., & Indefrey, P. (2011). Error-related activity and correlates of grammatical plasticity. Frontiers in Psychology, 2, 219. Dussias, P. E., & Cramer Scaltz, T. R. (2008). Spanish-English L2 speakers’ use of subcategorization bias information in the resolution of temporary ambiguity during second language reading. Acta Psychologia, 128, 501–513. Dussias, P. E., Valdés Kroff, J. R., Guzzardo Tamargo, R. E., & Gerfen, C. (2013). When gender and looking go hand in hand: Grammatical gender processing in L2 Spanish. Studies in Second Language Acquisition, 35, 353–387. Duyck, W., Van Assche, E., Drighe, D., & Hartsuiker, R. (2007). Visual word recognition by bilinguals in a sentence context: Evidence for nonselective lexical access. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33, 663–679. Ellert, M. (2011). Ambiguous Pronoun Resolution in L1 and L2 German and Dutch. Unpublished PhD thesis, Radboud University, Nijmegen.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

224

Leah Roberts

Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language. A psychometric study. Studies in Second Language Acquisition, 27, 141–172. Felser, C., & Cunnings, I. (2011). Processing reflexives in English as a second language: The role of structural and discourse-level constraints. Applied Psycholinguistics, 33, 571–603. Felser, C., Sato, M., & Bertenshaw, N. (2009). The online application of binding Principle A in English as a second language. Bilingualism: Language and Cognition, 12, 485–502. Foucart, A., & Frenck-Mestre, C. (2012a). Can late L2 learners acquire grammatical features? Evidence from ERPs and eye-tracking. Journal of Memory and Language, 66, 226–248. Foucart, A., & Frenck-Mestre, C. (2012b). Grammatical gender processing in L2: Electrophysiological evidence of the effect of L1-L2 syntactic similarity: Erratum. Bilingualism: Language and Cognition, 15(1), 202. Franceschina, F. (2001). Morphological or syntactic deficits in near-native speakers? An assessment of some current proposals. Second Language Research, 17, 213–247. Frenck-Mestre, C. (2005). Eye-movement recording as a tool for study­ing syntactic processing in a second language: A review of methodologies and experimental findings. Second Language Research, 21, 175–198. Frenck-Mestre, C., & Pynte, J. (1997). Syntactic ambiguity resolution while reading in second and native languages. Quarterly Journal of Experimental Psychology, section A: Human Experimental Psychology, 50A, 119–148. Friederici, A. D. (2011). The brain basis of language processing: From structure to function. Physiological Reviews, 91(4), 1357–1392. Friederici, A. D., Hahne, A., & Mecklinger, A. (1996). Temporal structure of syntactic parsing: Early and late event-related brain potential effects. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22, 1219–1248. Friederici, A., Steinhauer, K., & Pfeifer, E. (2002). Brain signatures of artificial language processing: Evidence challenging the critical period hypothesis. Proceedings of the National Academy of Science, 99, 529–534. Gibson, E., Pearlmutter, N., Canseco-Gonzalez, E., & Hickok, G. (1996). Recency preference in the human sentence processing mechanism. Cognition, 59, 21–39. Goad, H., & White, L. (2006). Ultimate attainment in interlanguage grammars: A prosodic approach. Second Language Research, 22(3), 243–268. Grüter, T., Lew-Williams, C., & Fernald, A. (2012). Grammatical gender in L2: A production or a real-time processing problem? Second Language Research, 28(2), 191–215. Grüter, T., Rohde, H., & Schafer, A. J. (2014). The role of discourse-level expectations in nonnative speakers’ referential choices. In W. Orman and M. J. Valleau (eds.), Proceedings of the 38th annual Boston University Conference on Language Development, BUCLD 38 (pp. 179–191). Somerville, MA: Cascadilla Proceedings Project.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

225

Hagoort, P., & Brown, C. M. (1999). Gender electrified: ERP evidence on the syntactic nature of gender processing. Journal of Psycholinguistic Resesarch, 28, 715–728. Hagoort, P., & Indefrey, P. (2014). The neurobiology of language beyond single words. Annual Review of Neuroscience, 37, 347–362. Hahne, A., & Friederici, A. D. (2001). Processing a second language: Late learners’ comprehension mechanisms as revealed by event-related brain potentials. Bilingualism, Language and Cognition, 4(2), 123–141. Hawkins, R., & Liszka, S. (2003). Locating the source of defective past tense marking in advanced L2 English speakers. In R. Hout, A. Hulk, F. Kuiken, & R. Towell (eds.), The lexicon–syntax interface in second language acquisition (pp. 21–44). Amsterdam: John Benjamins. Hernandez, A. E., Hofmann, J., & Kotz, S. A. (2007). Age of acquisition modulates neural activity for both regular and irregular syntactic functions. NeuroImage, 36(3), 912–923. Hopp, H. (2010). Ultimate attainment in L2 inflectional morphology: Performance similarities between non-native and native speakers. Lingua, 120, 901–931. Hopp, H. (2013). Semantics and morphosyntax in predictive L2 sentence processing. International Review of Applied Linguistics in Language Teaching, 53, 277–306. Jiang, N. (2004). Morphological insensitivity in second language processing. Applied Psycholinguistics, 25, 603–634. Juffs, A., & Harrington, M. (1995). Parsing effects in second language sentence processing: Subject and object asymmetries in wh-extraction. Studies in Second Language Acquisition, 17, 483–516. Juffs, A., & Harrington, M. (1996). Garden path sentences and error data in second language processing research. Language Learning, 46, 286–324. Juffs, A., & Rodriguez, G. A. (2014). Second language sentence processing. New York: Routledge. Just, M. A., Carpenter, P. A., & Woolley, J. D. (1982). Paradigms and processes in reading comprehension. Journal of Experimental Psychology: General, 111, 228–238. Kaan, E. (2014). Predictive processing in L1 and L2. What’s different? Linguistic Approaches to Bilingualism, 4, 257–282. Keating, G. D. (2009). Sensitivity to violations in gender agreement in native and nonnative Spanish: An eye-movement investigation. Language Learning, 59(3), 503–535. Kolk, H. H. J., Chwilla, D. J., Van Herten, M., & Oor, P. J. W. (2003). Structure and limited capacity in verbal working memory: A study with eventrelated potentials. Brain and Language, 85, 1–36. Kotz, S. A., Holcomb, P., & Osterhout, L. (2008). ERPs reveal comparable syntactic sentence processing in early bilinguals and monolinguals. Acta Psychologica, 128, 514–527.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

226

Leah Roberts

Lee, M. W. E. (2002). Implementing subject–verb number agreement in a non-native language. Poster presented at the 15th CUNY Conference on Human Sentence Processing, New York. Lew-Williams, C., & Fernald, A. (2010). Real-time processing of gendermarked articles by native and non-native Spanish speakers. Journal of Memory and Language, 63(4), 447–464. Liversedge, S., Paterson, K., & Pickering, M. (1998). Eye movements and measures of reading time. In G. Underwood (ed.), Eye guidance in reading and scene perception (pp. 55–75). Oxford: Elsevier. Marian, V., & Spivey, M. (2003). Competing activation in bilingual language processing: Within- and between-language competition. Bilingualism: Language and Cognition, 6, 97–115. Martin, C., Thierry, G., Kuipers, J.-R., Boutonnet, B., Foucart, A., & Costa, A. (2013). Bilinguals reading in their second language do not predict upcoming words as native readers do. Journal of Memory and Language, 69(4), 574–588. McLaughlin, J., Tanner, D., Pitkanen, I., Frenck-Mestre, C., Inoue, K., Valentine, G., & Osterhout, L. (2010). Brain potentials reveal discrete stages of L2 grammatical learning. Language Learning, 60(S2), 123–150. Molinaro, N., Barber, H. A., Caffarra, S., & Carreiras, M. (2015). On the left anterior negativity (LAN): The case of morphosyntactic agreement: A reply to Tanner et al. Cortex, 66, 156–159. Molinaro, N., Barber, H. A., & Carreiras, M. (2011). Grammatical agreement processing in reading: ERP findings and future directions. Cortex, 47, 908–930. Morgan-Short, K., Sanz, C., Steinhauer, K., & Ullman, M. T. (2010). Second language acquisition of gender agreement in explicit and implicit training conditions: An event-related potential study. Language Learning, 60, 154–193. Nevins, A., Dillon, B., Malhotra, S., & Phillips, C. (2007). The role of feature number and feature type in processing Hindi verb agreement violations. Brain Research, 1164, 81–94. Ojima, S., Nakata, H., & Kakigi, R. (2005). An ERP study of second language learning after childhood: Effects of proficiency. Journal of Cognitive Neuroscience, 17(8), 1212–1228. O’Rourke, P. L., & Van Petten, C. (2011). Morphological agreement at a distance: Dissociation between early and late components of the event related brain potential. Brain Research, 1392, 62–79. Osterhout, L., McLaughlin, J., Pitkänen, I., Frenck-Mestre, C., & Molinaro, N. (2006). Novice learners, longitudinal designs, and event-related potentials: A means for exploring the neurocognition of second-language processing. Language Learning, 56(1), 199–230. Paterson, K., Liversedge, S., & Underwood, G. (1999). The influence of focus operators on syntactic processing of short relative clause sentences. The Quarterly Journal of Experimental Psychology, 52, 717–737.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

227

Pickering, M. J., & Traxler, M. J. (1998). Plausibility and recovery from garden-paths: An eye-tracking study. Journal of Experimental Psychology: Learning, Memory and Cognition, 24, 940–961. Prévost, P., & White, L. (2000). Missing surface inflection or impairment in second language acquisition? Evidence from tense and agreement. Second Language Research, 16, 103–133. Proverbio, A. M., Cok, B., & Zani, A. (2002). Electrophysiological measures of language processing in bilinguals. Journal of Cognitive Neuroscience, 14(7), 994–1017. Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124, 372–422. Rayner, K. (2009). The 35th Sir Frederick Bartlett Lecture: Eye movements and attention during reading, scene perception, and visual search. Quarterly Journal of Experimental Psychology, 62, 1457–1506. Rayner, K., Carlson, M., & Frazier, L. (1983). The interaction of syntax and semantics during sentence processing. Journal of Verbal Learning and Verbal Behavior, 22, 358–374. Rayner, K., Sereno, S., Morris, R., Schmauder, A., & Clifton, C. (1989). Eye movements and online language comprehension processes. Language and Cognition Processes, 4, 21–49. Roberts, L. (2013). Sentence processing in bilinguals. In R. van Gompel (ed.), Sentence processing. Current issues in the psychology of language (pp. 221– 246). Hove: Psychology Press. Roberts, L., & Felser, C. (2011). Plausibility and recovery from garden-paths in second language sentence processing. Applied Psycholinguistics, 32, 299–331. Roberts, L., González Alonso, J., Pliatsikas, C., & Rothman, J. (2018). Evidence from neurolinguistic methodologies: Can it actually inform linguistic/language acquisition theories and translate to evidence-based applications? Second Language Research, 34, 125–143. Roberts, L., Gullberg, M., & Indefrey, P. (2008). Online pronoun resolution in L2 discourse: L1 influence and general learner effects. Studies in Second Language Acquisition, 30, 333–357. Roberts, L., & Liszka, S. (2013). Processing tense/aspect-agreement violations online in the second language: A self-paced reading study with French and German L2 learners of English. Second Language Research, 29, 413–439. Roncaglia-Denissen, M. P., & Kotz, S. A. (2016). What does neuroimaging tell us about morphosyntactic processing in the brain of second language learners? Bilingualism: Language and Cognition, 19, 665–973. Rossi, S., Gugler, M., Friederici, A., & Hahne, A. (2002). The impact of proficiency on second language processing of German and Italian: Evidence from event-related potentials. Journal of Cognitive Neuroscience, 18, 2030–2048. Rossi, S., Gugler, M. F., Friederici, A. D., & Hahne, A. (2006). The impact of proficiency on syntactic second-language processing of German

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

228

Leah Roberts

and Italian: Evidence from event-related potentials. Journal of Cognitive Neuroscience, 18(12), 2030–2048. Rüschemeyer, S.-A., Fiebach, C. J., Kempe, V., & Friederici, A. D. (2005). Processing lexical semantic and syntactic information in first and second language: fMRI evidence from German and Russian. Human Brain Mapping, 25(2), 266–286. Rüschemeyer, S.-A., Zysset, S., & Friederici, A. D. (2006). Native and nonnative reading of sentences: An fMRI experiment. NeuroImage, 31(1), 354–365. Sabourin, L., & Stowe, L. A. (2008). Neurobiology of Language Learning. In F. M. Hult & B. Spolsky (eds.), The handbook of educational linguistics (pp. 27–37). Malden, MA/Oxford: Blackwell. Sabourin, L., Stowe, L. A., & de Haan, G. J. (2006). Transfer effects in learning a second language grammatical gender system. Second Language Research, 22, 1–29. Sagarra, N., & Herschensohn, J. (2011). Proficiency and animacy effects on L2 gender agreement processes during comprehension. Language Learning, 61(1), 80–116. Schwartz, B., & Sprouse, X. (1996). L2 cognitive states and the Full Transfer/ Full Access model. Second Language Research, 12, 40–72. Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics in Language Teaching (IRAL), 10(3), 209–231. Slabakova, R. (guest ed.) (2009). The Fundamental Difference Hypothesis twenty years later [Special issue]. Studies in Second Language Acquisition, 2. Sorace, A. (2011). Pinning down the concept of “interface” in bilingualism. Linguistic Approaches to Bilingualism, 1(1), 1–33. Sorace, A., & Serratrice, L. (2009). Internal and external interfaces in bilingual language development: Beyond structural overlap. International Journal of Bilingualism, 13, 195–210. Steinhauer, K., White, E. J., & Drury, J. E. (2009). Temporal dynamics of late second language acquisition: Evidence from event-related brain potentials. Second Language Research, 25(1), 13–41. Suh, S., Yoon, H. W., Lee, S., Chung, J.-Y., Cho, Z.-H., & Park, H. (2007). Effects of syntactic complexity in L1 and L2; An fMRI study of KoreanEnglish bilinguals. Brain Research, 1136(1), 178–189. Tanner, D., Inoue, K., Osterhout, L. (2014). Brain-based individual differences in online L2 grammatical comprehension. Bilingualism, Language and Cognition, 17(2), 277–293. Tanner, D., McLaughlin, J., Hershensohn, J., & Osterhout, L. (2013). Individual differences reveal stages of L2 grammatical acquisition: ERP evidence. Bilingualism: Language and Cognition, 16(2), 367–382.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Psycholinguistic and Neurolinguistic Methods

229

Tanner, D., & van Hell, J. G. (2014). ERPs reveal individual differences in morphosyntactic processing. Neuropsychologia, 56, 289–301. Tokowicz, N., & Macwhinney, B. (2005). Implicit and explicit measures of sensitivity to violations in second language grammar: An event-related potential investigation. Studies in Second Language Acquisition, 28, 173–204. Tsimpli, I.-M., & Dimitrakopoulou, M. (2007). The Interpretability Hypothesis: Evidence from wh- interrogatives in second language acquisition. Second Language Research, 23, 215–242. Tsimpli, I.-M., & Sorace, A. (2006). Differentiating interfaces: L2 ­performance in syntax-semantics and syntax-discourse phenomena. In D. Bamman, T. Magnitskaia, & C. Zaller (eds.), Proceedings of the 30th annual Boston University Conference on Language Development, BUCLD 30 (pp. 653–664). Somerville, MA: Cascadilla Press. van Gompel, R. (2013). Sentence processing. Current issues in the psychology of language. Hove: Psychology Press. Weber, K., & Lavric, A. (2008). Syntactic anomaly elicits a lexico-semantic (N400) ERP effect in the second language but not the first. Psychophysiology, 45, 920–925. Weber-Fox, C., & Neville, H. J. (1996). Maturational constraints on functional specializations for language processing: ERP and behavioral evidence in bilingual speakers. Journal of Cognitive Neuroscience, 8(3), 231–256. Wicha, N. Y., Moreno, E. M., & Kutas, M. (2004). Anticipating words and their gender: An event-related brain potential study of semantic integration, gender expectancy and gender agreement in Spanish sentence reading. Journal of Cognitive Neuroscience, 16, 1272–1288. Wilson, F. (2009). Processing at the syntax–discourse interface in second language acquisition. Unpublished PhD thesis, University of Edinburgh. White, L., Valenzuela, E., Kozlowska-Macgregor, M., & Leung, Y.-K. I. (2004). Gender and number agreement in nonnative Spanish. Applied Psycholinguistics, 25, 105–133. Zawiszewski, A., Gutiérrez, E., Fernández, B., & Laka, I. (2011). Language distance and non-native syntactic processing: Evidence from event-­ related potentials. Bilingualism, Language and Cognition, 14(3), 400–411.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:36:09, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.010

10 Interaction in L2 Learning Jaemyung Goo

10.1 Introduction Cognitive interactionists claim that interaction provides valuable opportunities for learners to refine and restructure their interlanguage by drawing their focal attention to linguistic code features during negotiation for meaning (see Gass, 1997, 2003; Gass & Mackey, 2006, 2015; Long, 1996, 2007; Mackey, 2012; Mackey, Abbuhl, & Gass, 2012; Pica, 1994, 1996). Negotiated interaction activates cognitive learning processes that involve processing (modified) input, receiving corrective feedback (CF), and producing (modified) output, during which learner attention is directed to L2 linguistic features, leading to noticing and in turn L2 development. Over two decades of empirical research designed to explore the link between interaction and actual learning has yielded abundant evidence that clearly indicates that interaction precipitates L2 learning (see Mackey, 2012 for a review of interaction research; Keck et al., 2006 and Mackey & Goo, 2007 for meta-analytic reviews of early interactionacquisition studies). A number of interactional features or processes and learner-internal and -external factors mediating (assumed to mediate) the extent of interaction effects (e.g., modified output opportunities, noticing, CF type, task type and complexity, working memory, language aptitude, etc.) have been investigated in various learning contexts and from diverse research perspectives. Although the overall results contribute to a great extent to deepening our understanding of how interaction aids L2 learning, the results also bespeak a complex and multifaceted nature of interaction-based learning that involves a multitude of potential mediating factors interacting with each other. In this chapter, I first illustrate the Interaction Hypothesis along with relevant research findings, then discuss several mediating factors in terms of whether they influence interactional

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

234

Jaemyung Goo

benefits and, if so, how, and end the chapter with a brief summary of findings and implications for teaching.

10.2 Interactional Modifications During Negotiation for Meaning Long’s (1996) updated Interaction Hypothesis, vis-à-vis its earlier version focusing on the relationship among interactional and linguistic adjustments and L2 comprehension assumed to lead to L2 acquisition, presents a more specific framework involved in interaction-driven L2 learning with great emphasis placed on the nontrivial roles of such additional components as noticing, L2 processing capacity, and corrective feedback (CF). He proposed that environmental contributions to acquisition are mediated by selective attention and the learner’s developing L2 processing capacity, and that these resources are brought together most usefully, although not exclusively, during negotiation for meaning. Negative feedback obtained during negotiation work or elsewhere may be facilitative of L2 development, at least for vocabulary, morphology, and language-specific syntax, and essential for learning certain specifiable L1–L2 contrasts. (p. 414)

That is, negotiation for meaning contributes to L2 development by providing propitious conditions for, and facilitating, attentional and psycholinguistic processes involved in L2 learning. One critical concept to note in the updated Interaction Hypothesis is negotiation for meaning, which he defines as the process in which, in an effort to communicate, learners and competent speakers provide and interpret signals of their own and their interlocutor’s perceived comprehension, thus provoking adjustments to linguistic form, conversational structure, message content, or all three, until an acceptable level of understanding is achieved. (p. 418)

In other words, negotiation for meaning enhances the comprehensibility of the message content, and also renders certain linguistic features more salient so that they become more available for acquisition. Similarly, Pica (1994), underlining the significance of negotiation, contends that “the twofold potential of negotiation—to assist L2 comprehension and draw attention to L2 form—affords it a more powerful role in L2 learning than has been claimed so far” (p. 508). On the specific role of negotiation for meaning in L2 learning, Long (1996) states that negotiation for meaning, and especially negotiation work that triggers interactional adjustments by the NS or more competent interlocutor, facilitates acquisition because it connects input, internal learner capacities, particularly selective attention, and output in productive ways.  (pp. 451–452)

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

235

Negotiation for meaning functioning as one fundamental element of interaction provides breeding grounds for, and effectuates, interaction-driven L2 acquisition by combining and coordinating interactional features and cognitive learning processes in an optimal way. Negotiated interaction can be initiated by either L2 learners or their interlocutors. Learner-initiated negotiation induces interactionally modified input, and interlocutor-initiated negotiation via three typical negotiation moves (i.e., clarification requests, confirmation checks, comprehension checks) or negative feedback may elicit output modifications when needed. Benefits of negotiation for meaning transpire in either direction. For example, early interaction research focused on comprehension through input modifications in interaction contexts (e.g., Ellis, Tanaka, & Yamazaki, 1994; Gass & Varonis, 1994; Long, 1985; Loschky, 1994; Pica, 1992; Pica, Young, & Doughty, 1987). Overall results showed that interactionally modified input facilitated L2 comprehension and often resulted in better comprehension than premodified input (e.g., Ellis, Tanaka, & Yamazaki, 1994; Pica, Young, & Doughty, 1987; Loschky, 1994). Also, input modifications made for L2 learners during negotiated interaction contributed to boosting subsequent L2 production (Gass & Varonis, 1994). Based on his findings of nonsignificant between-group differences in terms of L2 learning, despite better comprehension observed in the performance of the interactionally modified input condition, Loschky (1994) noted that “posing a linear relationship between comprehension of input and intake of the structures contained therein may be untenable” (p. 320). Therefore, interaction researchers began to investigate the link between interaction and actual L2 learning. Mackey’s (1999) study is one of the earliest empirical attempts in that respect. She found that active participation in negotiated interaction led to the production of developmentally more advanced English question forms than simple observations or passive participation in interaction without negotiation. Given these early empirical findings, it is evident that negotiated interaction functions as a critical bridge that connects input with learning, increasing and maximizing the value of input as a fundamental element of language learning. Echoing the significant role of negotiated interaction and its relation to input, Swain (1985) states that “it is not input per se that is important to second language acquisition but input that occurs in interaction where meaning is negotiated” (p. 246). Mackey (2012) also suggests that “through negotiation of meaning, input can be uniquely tailored to individuals’ strengths, weaknesses, and communicative needs, providing language that suits their distinct developmental levels” (p. 12).

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

236

Jaemyung Goo

Interlocutor-initiated negotiation involving negotiation moves and/ or negative feedback motivates output modifications by the L2 learner, another crucial aspect of interaction. Building on her seminal work (1985) that clearly indicated the need for L2 production opportunities, Swain proposed the Output Hypothesis (see also Swain, 1995, 2005) and argued that L2 “learners need to be pushed to make use of their resources; they need to have their linguistic abilities stretched to their fullest; they need to reflect on their output and consider ways of modifying it to enhance comprehensibility, appropriateness, and accuracy” (1993, pp. 160–161). Swain (1995, 2005) propounded three functions of pushed output in SLA, in addition to its obvious role in enhancing fluency and automatization: noticing the interlanguage (IL)–L2 gap, hypothesis testing, and metalinguistic reflection on L2 features. Due to these functions, according to Swain (1995), output may stimulate learners to move from the semantic, open-ended, non-deterministic, strategic processing prevalent in comprehension to the complete grammatical processing needed for accurate production. Output, thus, would seem to have a potentially significant role in the development of syntax and morphology. (p. 128)

Swain’s (1985, 1995, 2005) Output Hypothesis changed the conceptual paradigm with respect to output, extending its role as an integral part of L2 learning, not just as a simple means of practising what has already been learned or eliciting information. Empirical studies have been conducted to identify patterns of output modifications, and they have shown that different types of negotiation moves or negative feedback led to different types of output modifications (e.g., Pica et al., 1989; Shehadeh, 1999, 2001; Van den Branden, 1997). Nevertheless, relevant research findings, in general, accord more or less with expected benefits of output modifications (e.g., Egi, 2010; Ellis & He, 1999; Izumi et al., 1999; Loewen, 2005; McDonough, 2005; Nobuyoshi & Ellis, 1993). For example, McDonough (2005) showed that modified output production was the only significant predictor of English question development. Also, Loewen (2005), having examined 491 focus-on-form episodes, found that successful immediate modified output was a significant predictor of learner accuracy observed in individualized tailor-made post-tests. As for noticing, Egi’s (2010) study revealed that repairs (i.e., successful modified output), but not needsrepairs, were associated with learner noticing of both the corrective intent of recasts and their linguistic gap (see also Gurzynski-Weiss & Baralt, 2015, for similar results). Notwithstanding the potentially crucial role of negotiation for meaning, Foster (1998) questioned the ecological validity of research findings regarding negotiated interaction and analysed classroom-based interaction data.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

237

Based on her observation that there occurred only a few instances of negotiation for meaning (see also Foster & Ohta, 2005; Pica, 2002 for similar results), she suggested that it is the type of setting that determines the amount of negotiation for meaning, with little or no negotiation in actual classroom contexts and much more negotiation in tightly controlled research contexts. In their study designed to compare classroom and laboratory settings in terms of negotiation for meaning, Gass, Mackey, and Ross-Feldman (2005) found few differences in the amount of negotiation and interaction patterns, and, instead, observed the impacts of task type on learner interaction. These studies function as a reminder that task features including task type (see Pica, Kanagy, & Falodun, 1993) and complexity-based task sequencing (e.g., Robinson, 2011b) should be taken into consideration to ensure that ample negotiated interaction transpires in classroom contexts. Of course, more research on this issue of laboratory versus classroom contexts in regard to instances of negotiation appears necessary. Digging deeper into the positive potential of negotiation for meaning, interaction researchers have shown a great interest in CF with respect to how different CF moves create L2 learning opportunities, why certain CF moves are relatively more effective at promoting L2 development than others, and whether their positive effects are mediated by certain factors, to which I now turn.

10.3  Corrective Feedback, Noticing, and L2 Learning CF, claimed to serve as an effective tool for drawing learners’ attention to L2 forms, as another crucial feature of interaction, has received a great amount of attention from interaction researchers and, as such, has generated much discussion and numerous empirical studies on its potential relation to L2 development (see Brown, 2016; Goo & Mackey, 2013; Li, 2010; Lyster & Ranta, 2013; Lyster & Saito, 2010; Lyster, Saito, & Sato, 2013; Mackey & Goo, 2007; Nassaji, 2016, for overall reviews and meta-analyses). It has been observed that various types of CF, albeit not in an equal proportion, have been employed in L2 interactional learning contexts. Lyster and Ranta (1997) identified six types of CF moves in their French immersion classroom data: explicit correction, recasts, repetitions, elicitations, clarification requests, and metalinguistic feedback. These CF moves are often categorized as either input-providing (e.g., explicit correction, recasts) or output-prompting (e.g., clarification requests, elicitations), and also as either explicit (e.g., metalinguistic feedback) or implicit feedback (e.g., recasts), and compared in terms of learner responses or reactions to CF and their effects on L2 learning. Brown (2016), in his recent metaanalysis of classroom-based CF studies, observed that recasts were the

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

238

Jaemyung Goo

most frequently used CF moves that teachers provided in L2 classrooms (57%), followed by prompts (30%), i.e., repetition, elicitation, clarification, and metalinguistic feedback. Further, L2 learners received significantly more CF moves on their grammatical errors (42.7%) than on lexical and phonological errors (27.6% and 22.4%, respectively). Lyster and Ranta (1997), as regards L2 learners’ immediate responses or reactions to CF, which they called uptake, the pedagogic significance or value of which interaction researchers have doubted (e.g., Gass, 2003; Mackey & Philp, 1998; McDonough & Mackey, 2006), found that although recasts were the most frequently used feedback type, they induced the least amount of uptake. Later studies, however, showed somewhat differing results, indicating the amount of learner uptake following recasts and other CF moves may vary depending upon L2 learning contexts (e.g., Ellis, Basturkmen, & Loewen, 2001; Lyster & Mori, 2006; Sheen, 2004). Findings suggest that recasts lead to substantial uptake in some instructional settings where teachers and learners focus on language as form (Sheen, 2004) and interactional contexts where explicit language-focused exchanges are involved (Oliver & Mackey, 2003): see Lyster and Mori’s (2006) idea on the role of instructional focus in the choice of CF moves. Brown (2016) found that educational contexts (level of education) influenced teachers’ CF choices with more recasts supplied to university or adult learners and elementary-level learners than prompts and with more prompts to high school learners than recasts. In addition to these and other relevant observations regarding learner responses to various CF moves during negotiated interaction, and given their pedagogical and acquisitional values, CF moves have been extensively researched in terms of their relative efficacy (e.g., recasts versus prompts or implicit versus explicit) in the development of specific L2 target features, with recasts at the epicentre of much scholarly attention (see Brown, 2016; Goo & Mackey, 2013; Lyster & Ranta, 2013; Lyster, Saito, & Sato, 2013; Nassaji, 2016, for recent reviews). This overwhelming interest in recasts has stemmed not just from their observed high frequency in L2 learning contexts but also from cognitive benefits that likely effectuate noticing, which in turn promotes L2 learning. Recasts are provided immediately following the learner’s erroneous utterance, the intended meaning of which has already been comprehended, that is, more attentional resources can be directed to the linguistic code features targeted in recasts (e.g., Doughty, 2001; Long, 1996, 2007). These freed-up attentional resources, along with an immediate juxtaposition of the learner’s erroneous utterance and the recast provided, render it highly plausible to go through a cognitive comparison of the two versions that are in contrast and perceive the form contained in the recast as a correct alternative to their erroneous form. Long (2007) states that “recasts convey

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

239

needed information about the target language in context, when interlocutors share a joint attentional focus, and when the learner already has prior comprehension of at least part of the message, thereby facilitating form– function mapping” (p. 77). He further notes that “learners are vested in the exchange, as it is their message that is at stake, and so will probably be motivated, and attending, conditions likely to facilitate noticing of any new linguistic information in the input” (pp. 77–78). Regarding the effectiveness of recasts, one recent study (Nassaji, 2017) deserves attention due to its interesting results. Nassaji compared intensive recasts with extensive recasts in their effects on the development of English articles. The intensive group received recasts on article errors only, whereas the extensive group received recasts on any L2 errors. Extensive recasts turned out to be more beneficial than intensive recasts, contrary to the assumption that intensive recasts direct L2 learners’ focused attention to the targeted L2 feature more effectively than extensive recasts, and, thus, likely lead to more noticing and learning as a result. He attributed the findings to learners’ increased sensitivity to recasts due to an overall greater number of recasts provided for the extensive group (seventy per cent more compared to the intensive group), leading to the enhanced saliency and noticeability of recasts as negative feedback. It is suspected, then, that this heightened noticing of overall recasts as CF, regardless of the target, may have effectuated the increased awareness of recasts provided in response to errors on other linguistic features and aided the learning of those features not in the main focus of instruction. Learning this way, albeit not measured in his study, would bespeak further evidence of recasts promoting incidental learning of L2 features outside of instructional focus. This, of course, needs to be tested in future research. Nevertheless, the effectiveness of recasts, especially in comparison with other CF moves, has been questioned and, as such, tested empirically. Although some studies showed recasts were slightly more beneficial than, or as effective as, other CF moves such as prompts and explicit correction (e.g., Adams, Nuevo, & Egi, 2011; Goo, 2012; Lyster & Izquierdo, 2009; McDonough, 2007; Nassaji, 2009; see Mackey & Goo, 2007; Goo & Mackey, 2013, for reviews on general benefits of recasts), far more evidence has been obtained in favour of other CF moves (e.g., Ammar, 2008; Ammar & Spada, 2006; van de Guchte et al., 2015; Ellis, 2007; Ellis, Loewen, & Erlam, 2006; Goo, 2016; Gooch, Saito, & Lyster, 2016; Lyster, 2004; Sheen, 2007; Yang & Lyster, 2010; Yilmaz, 2012, 2013a; see also Lyster & Ranta, 2013; Lyster & Saito, 2010; Lyster, Saito, & Sato, 2013, for reviews). Lyster and colleagues (e.g., Lyster, Saito, & Sato, 2013) suggested that prompts are more effective at facilitating L2 development than recasts due to the nature of prompts as placing cognitive demands for self-repair on L2 learners, as supported by Swain’s Output Hypothesis (Swain, 1985, 1995).

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

240

Jaemyung Goo

Goo and Mackey (2013), however, raised several methodological issues, questioning the validity and reliability of previous findings on recasts versus prompts (see Goo & Mackey, 2013; Lyster & Ranta, 2013 for more discussion and debate on recasts versus prompts). Ellis (2015) notes that the recasts-versus-prompts debate may be unnecessary because recasts may function better than prompts, and vice versa depending on particular learning contexts, for instance, with recasts providing positive evidence of new linguistic items for L2 learners to acquire (e.g., Goo & Mackey, 2013; Long, 2007) and prompts providing opportunities for learners to refine their prior knowledge of target items via self-repair (e.g., Lyster, Saito, & Sato, 2013). Thus, he suggests that L2 learners may benefit most from a mixture of these two CF types. Also, given varied factors (e.g., social and contextual) may affect the effectiveness of different CF moves in a different way; as Ortega (2009) noted earlier, it may be “necessary to reconceptualize the all-or-nothing comparative approach that has characterized L2 research on negative feedback to date” (p. 76). However, considering that noticing precipitates L2 development (Schmidt, 1990, 1995, 2001), it should be noted that the effectiveness of CF moves during negotiation for meaning depends, to a great extent, on whether L2 learners notice CF moves, and, if so, how they perceive different CF moves, which has also invited L2 researchers’ attention (e.g., Bao, Egi, & Han, 2011; Egi, 2007, 2010; Gurzynski-Weiss & Baralt, 2014, 2015; Mackey, Gass, & McDonough, 2000; Philp, 2003). A study by Mackey, Gass, and McDonough (2000), widely cited in this respect, examined two L2 groups about their perceptions of CF—English as a second language (ESL) and Italian as a foreign language (IFL) learners—and found that, whereas the learners were relatively more accurate in their perceptions of lexical and phonological feedback, their perceptions of morphosyntactic feedback were generally inaccurate. They further observed that morphosyntactic feedback was provided predominantly in the form of recasts (their ESL data), which indicates recasts on morphosyntactic errors may sometimes go unnoticed. Similarly, Gurzynski-Weiss and Baralt (2014) reported that L2 Spanish learners’ noticing of morphosyntactic feedback was relatively inaccurate compared to lexical and semantic feedback in both face-to-face (FTF) and computer-mediated communication (CMC) modes. More specifically, Egi (2007) observed that L2 learners tended to interpret long recasts with three or more changes as responses to content (see Lyster, 1998a, 1998b, for a similar position about recasts) and short recasts with one or two corrective changes as linguistic evidence (positive, negative, or both). She suggested the length and number of changes in recasts may influence L2 learners’ interpretations of recasts (see Philp, 2003, for similar results). These characteristics of recasts also appear to affect the amount of learner uptake and L2 learning as well (e.g., Loewen & Philp, 2006; Sheen, 2006). It

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

241

should be noted, however, that uptake, in and of itself, may not provide an accurate representation of learner noticing (e.g., Bao, Egi, & Han, 2011). In a similar vein, repairs, but not needs-repairs, have been found to be associated with learner noticing of CF moves (e.g., Egi, 2010; Gurzynski-Weiss & Baralt, 2015). These studies indicate that, although noticing and modified output production are not isomorphic (e.g., Goo & Mackey, 2013; Long, 2007; Mackey & Philp, 1998), learner noticing may still be reflected in the production of modified output, as Chaudron (1977) noted. Nonetheless, as Mackey (2006) observed, it should be remembered that, although noticing may activate cognitive processing facilitative of L2 learning, noticing does not always lead to L2 learning. Furthermore, noticing may or may not transpire depending upon several mediating factors (e.g., target type, CF type, task type, L2 proficiency, cognitive capacities, etc.), resulting in varying degrees of L2 learning through interaction (see also Nakatsukasa, 2016, for a recent empirical attempt to examine gestures in noticing and learning). Due to their nontrivial roles in interaction-driven L2 learning, some of these mediating variables that have been found to influence noticing, interaction, and/or overall L2 learning are discussed in the following sections.

10.4  Factors Mediating Interaction Effects Benefits of CF moves in L2 learning have also been examined in regard to L2 proficiency (and developmental readiness), with results indicating that more proficient learners are better able to profit from recasts (e.g., Ammar, 2008; Ammar & Spada, 2006; Mackey & Philp, 1998; Trofimovich, Ammar, & Gatbonton, 2007). Some studies (e.g., Ammar, 2008; Ammar & Spada, 2006) suggested that low proficient learners may benefit more from prompts than recasts. Target type has also been found to mediate the effectiveness of CF moves (e.g., Adams et al., 2011; van de Guchte et al., 2015; Ellis, 2007; Jeon, 2007; Long, Inagaki, & Ortega, 1998; Mackey, 2006; Yang & Lyster, 2010), more often than not, with different CF moves differentially affecting the learning of different target features (but see Yilmaz, 2012, 2013a, for relatively consistent findings in favour of explicit feedback). Long (2007) suggests that unobtrusive implicit negative feedback (e.g., recasts) works for salient linguistic features, and more explicit types of feedback (e.g., metalinguistic feedback) for communicatively redundant and phonologically nonsalient linguistic features (e.g., past tense –ed). He also notes that recasts may be more effective for linguistic structures and forms that are hard to learn and thus require long-term treatments, compared to easy structures and forms that short-term treatments of explicit feedback may be sufficient to learn. This interaction between target type

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

242

Jaemyung Goo

and CF type is still an ongoing empirical question that merits further research. Interlocutor type and characteristics have also been examined as a mediating variable that affects L2 performance during interaction (e.g., Gurzynski-Weiss, 2016; Mackey, Oliver, & Leeman, 2003; Mackey, Polio, & McDonough, 2004; Polio, Gass, & Chapin, 2006; Sato & Lyster, 2007; see Gurzynski-Weiss, 2017, for a review). Learners likely produce more modified output when interacting with their non-native speaker (NNS) peers than with native speaker (NS) interlocutors, and CF type, amount of CF, and opportunities for modified output have also been found to differ the type of dyad, that is, NS–NNS versus NNS–NNS dyads (e.g., Mackey, Oliver, & Leeman, 2003; Sato & Lyster, 2007). As shown in Mackey, Oliver, & Leeman’s (2003) study, age may also affect language learning opportunities. As for teachers as interlocutors, their teaching experience has been found to be an important factor that influences interactional patterns and decisions on CF choices and use (e.g., Mackey, Polio, & McDonough, 2004; Gurzynski-Weiss, 2016), and enables more accurate perceptions of learner performance (e.g., Polio, Gass, & Chapin, 2006), with SLA training and teachers’ L1 also playing a role (Gurzynski-Weiss, 2016). Lastly, task complexity, given “task demands are a powerful determinant of what is noticed” (Schmidt, 1990, p. 143), has attracted a great amount of attention and yielded a recent surge of empirical research on whether it mediates L2 performance and development, and, if so, how (e.g., Baralt, 2013; Gilabert, Barón, & Llanes, 2009; Kim, 2009, 2012; Nuevo, 2006; Nuevo, Adams, & Ross-Feldman, 2011; Révész, 2009, 2011; Révész, Sachs, & Hama, 2014; Révész, Sachs, & Mackey, 2011). Much research has been motivated by, and designed specifically to test, Robinson’s (2001a, 2003, 2007, 2011a, 2011b) Cognition Hypothesis. Increasing task complexity along resource-directing dimensions (e.g., ± reasoning, ± few elements, ± here-and-now), according to the Cognition Hypothesis, “has the potential to connect cognitive resources, such as attention and memory, with effort at conceptualization and the L2 means to express it” (Robinson, 2011b, p. 14), engendering greater accuracy and complexity of production, and thus aiding L2 learning. Robinson (2007) notes that increased cognitive or conceptual demands of tasks generate “more noticing of task relevant input, and heightened memory for it, and so lead to more uptake of forms made salient in the input through various focus on form interventions” (p. 23). This line of research has observed that more complex tasks led to more negotiation and language learning opportunities with task type and proficiency mediating the overall impacts (e.g., Gilabert, Barón, & Llanes, 2009; Kim, 2009; Révész, 2011; Robinson, 2001b). In terms of uptake/modified output and L2 learning, findings are far from being conclusive, with some showing (partial) supportive evidence for the Cognition Hypothesis (e.g., Baralt, 2013; Kim, 2012) and others providing counterevidence (e.g.,

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

243

Nuevo, 2006; Nuevo, Adams, & Ross-Feldman, 2011; Révész, Sachs, & Hama, 2014; Révész, Sachs, & Mackey, 2011). Furthermore, Révész’ (2009) findings suggest that task complexity interacts with the effectiveness of recasts, depending upon the type of resource-directing and resourcedispersing dimension, and related task types. As indicated, overall results are not straightforward, which necessitates much more empirical research to uncover a potential link between task complexity and actual L2 learning through interaction.

10.5 WM, Language Aptitude, and Other Individual Differences Cognitive interactionists have also explored whether individual differences (ID) in cognitive capacities are associated with selective attention and noticing that likely determine the extent of beneficial impacts of negotiation for meaning on L2 learning. Working memory (WM) is one such area of interest (see Wen, 2016, for an extensive review on WM and SLA research), and much attention has been drawn to WM within the interactionist framework to investigate its role in the extent to which L2 learners benefit from interaction (e.g., Ahmadian, 2012; Goo, 2012, 2016; Kim, Payant, & Pearson, 2015; Kormos & Trebits, 2011; Li, 2013; Mackey et al., 2010; Mackey et al., 2002; Mackey & Sachs, 2012; Révész, 2012; Sagarra, 2007; Sagarra & Abbuhl, 2013; Trofimovich, Ammar, & Gatbonton, 2007; Yilmaz, 2013b). Overall research findings suggest that high-working memory capacity (WMC) learners are likely to benefit more from interaction compared to low-WMC learners in terms of noticing of CF (e.g., Kim, Payant, & Pearson, 2015; Mackey et al., 2002) and modified output production (e.g., Mackey et al., 2010; Sagarra, 2007), and actual L2 learning through CF (e.g., Goo, 2012; Li, 2013; Mackey & Sachs, 2012; Révész, 2012; Sagarra, 2007; Sagarra & Abbuhl, 2013; Yilmaz, 2013b). As for the relationship between WM and different CF moves, nevertheless, somewhat mixed results have been reported. For instance, Goo (2012) found WM was predictive of the effectiveness of implicit feedback (i.e., recasts), as expected given their role in cognitive comparisons (see Long, 2015), but not that of explicit feedback (i.e., metalinguistic feedback), in the development of an L2 grammatical feature (see also Sagarra, 2007; Sagarra & Abbuhl, 2013, for links between recasts and WM), whereas Li (2013) and Yilmaz (2013b) showed the exact opposite. More recently, task design and implementation variables such as task planning and task complexity were investigated in terms of whether they interact with L2 learners’ WMC (e.g., Ahmadian, 2012; Kim, Payant, & Pearson, 2015; Kormos & Trebits, 2011). Ahmadian (2012) found that WM was significantly correlated

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

244

Jaemyung Goo

with accuracy and fluency, but not with complexity, of oral production under an unpressured careful online task-planning condition. On the contrary, Kormos and Trebits (2011) observed WM affected the complexity of learner output only in the cognitively less complex task. Unlike what the Cognition Hypothesis (Robinson, 2001a, 2001b, 2011) predicts, WMC did not influence learner performance on the more complex task. More recently, Kim, Payant, & Pearson (2015) explored potential relationships among task complexity, WM, and L2 development during interaction. WM was found to be significantly associated with the noticing of recasts provided during interaction, regardless of task complexity, and significantly predictive of English question development. The cognitive advantage of having high WMC was observable only in the complex task condition, which confirms Robinson’s (2011b) suggestion that “there should be less variation between learners in performing simple tasks than there is when performing more complex versions” (p. 19). However, important caveats are ahead. WM performance and its relation to L2 learning may vary depending upon the type of WM task (e.g., Révész, 2012) and the type of WM scoring procedure (e.g., Leeser & Sunderman, 2016). Thus, care must be taken when interpreting data as to the nature of WM in L2 research and its effects on interaction-driven L2 learning. Language aptitude has also been researched as another potentially critical cognitive factor that may determine the extent of beneficial effects of interactional features, especially corrective feedback (e.g., Li, 2013; Sheen, 2007; Trofimovich, Ammar, & Gatbonton, 2007; Yilmaz, 2013b; Yilmaz & Granena, 2016; see also Li, 2017, for a brief review of research on language aptitude in interactional contexts). Overall findings are rather mixed, and, as shown in relation to the role of WM in interaction, the current state of affairs is still far from providing a clear picture of whether and how it relates to interaction-driven L2 learning, with some studies showing evidence of a significant role of language aptitude in the learning of L2 features through implicit feedback (e.g., Li, 2013; Trofimovich, Ammar, & Gatbonton, 2007) and others a significant relationship between language aptitude and beneficial effects of explicit, but not implicit, feedback (e.g., Sheen, 2007; Yilmaz, 2013b; Yilmaz & Gisela, 2016). Obviously, any conclusions are still premature with respect to the role of language aptitude in interaction-driven L2 learning, given various factors such as the type of language and target, type of aptitude measure, treatment measures (e.g., operationalization of feedback conditions), and so forth. More research needs to be conducted to obtain a better understanding of the role of language aptitude and its relation to other cognitive variables (e.g., WM, attention control, etc.) as well. Other moderating variables such as language anxiety, self-confidence, cognitive style, creativity, etc. have also been investigated in terms

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

245

of the extent to which they interact with interactional features (e.g., McDonough, Crawford, & Mackey, 2015; Nakatsukasa, 2016; Rassaei, 2015a, 2015b; Révész, 2011; Sheen, 2008). As regards language anxiety in relation to L2 learning through recasts, whereas Sheen (2008) showed relative advantages for low-anxiety learners vis-à-vis their high-anxiety counterparts, Rassaei (2015a) found the opposite results. Rassaei also observed that low-anxiety learners benefited more from metalinguistic feedback than from recasts, and the reverse was true for high-anxiety learners. Adding further confusion, Révész (2011) reported no impact of language anxiety (also linguistic self-confidence and self-perceived communicative competence) on L2 learners’ speech production and the quality and quantity of language-related episodes (LREs), regardless of the level of task complexity. More research, obviously, seems necessary to untangle this rather conflicting puzzle. Research on other individual difference (ID) variables (e.g., cognitive style, creativity) rarely explored in interaction contexts has also emerged in recent years and contributed to providing further insights into how much and in what way interaction benefits L2 development. Rassaei (2015b), for instance, observed that field-independent (FI) learners benefited significantly more from recasts when compared to field-dependent (FD) learners, which indicates FI learners’ superior cognitive ability to manipulate selective attention for noticing recasts. In another interesting attempt to broaden the scope of interaction-based research, McDonough, Crawford, & Mackey (2015) explored whether there was any noticeable association between creativity and L2 task performance. Creativity was significantly correlated with the production of some linguistic features, but not with that of other features. The findings suggest that creativity may, to some extent, function as another moderator variable affecting L2 learners’ task performance (see also Albert, 2011, for creativity and task complexity). The above-mentioned studies, designed to examine whether and how varied ID factors mediate interaction-driven L2 learning, have contributed to diversifying interaction research. However, relevant findings are a long way from offering a complete picture of the specific functions of these ID variables. To make matters more complicated, some ID variables (e.g., language aptitude, WM, cognitive style, etc.) interact with other mediating factors (e.g., target type, CF type, etc.). Long (2015) claimed that it is the interaction of input sensitivity (a constant within the individual, but varying across individuals) and perceptual saliency (which varies across structures) that has the potential to account for success and failure at the level both of individual learners and individual structures. (Emphases in original; p. 60)

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

246

Jaemyung Goo

More empirical research including replication studies needs to be conducted to allow us to better understand these already complicated cognitive phenomena that various ID variables have brought to our attention.

10.6  Conclusion and Implications for Teaching Interaction researchers have obtained clear evidence that L2 learners benefit from negotiated interaction because it affords L2 learners potential learning opportunities to receive modified input and CF on their erroneous output and produce modified output in response to their interlocutors’ CF moves (see Gass & Mackey, 2015; Mackey, 2012, for reviews). Early days of interaction research observed that negotiation for meaning facilitated L2 comprehension and subsequent oral production. Beginning in the mid 1990s, researchers began to investigate a link between interaction and actual learning, and numerous studies have evidenced benefits of interaction ever since. Meanwhile, a good number of descriptive, observational, and empirical studies have also examined specific features of interaction (e.g., CF, uptake, modified output, noticing, etc.) in terms of how they relate to L2 learning and/or learning opportunities. Overall findings in regard to the effectiveness of CF suggest that CF is facilitative of L2 development with recasts more effective than, or as effective as, other CF moves (e.g., prompts, explicit corrections) in some studies, but less effective in others (see Lyster, Saito, & Sato, 2013; Nassaji, 2016, for recent reviews). It should be noted, however, that accumulated evidence bespeaks relatively more beneficial roles of prompts compared to recasts. As for uptake, the amount of uptake following recasts may vary depending upon learning contexts, with relatively more uptake in form-oriented, and less in meaning-oriented, contexts. Successful modified output (repairs), but not needs-repairs, has been found to be associated with actual learning as well as learner noticing. In terms of language areas, relevant findings suggest that L2 learners are not quite accurate in their perceptions of morphosyntactic feedback, compared to phonological and lexical or semantic CF. Previous research findings also indicate that characteristics of recasts (e.g., length, number of changes, etc.) tend to mediate learner noticing and overall effects of recasts, with recasts involving short and one or two changes likely to be noticed more often than those involving long and three or more changes. Noticing, L2 learning opportunities, and the effectiveness of CF can be mediated by various factors, such as type of target, developmental readiness, interlocutor characteristics, task complexity, and so forth. We have also obtained some evidence of mediating roles of learner-internal ID variables (e.g., WM, language aptitude, cognitive style, etc.), however,

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

247

with some studies showing conflicting results and depicting a rather complex picture of the issue, because these ID variables often interact with other mediating variables with possibilities of certain ID variables interplaying with other ID variables. As far as these mediating factors are concerned, interaction research to date has not yet arrived at any level of conclusive end. Given that the field of SLA has expanded its theoretical and experimental boundaries by adopting, adapting, and borrowing varied research methods and theoretical frameworks, our burden of proof has been diversified along with quite a few factors that may mediate the extent to which L2 learners benefit from negotiated interaction. The topics (and relevant studies) that I illustrated in this chapter are just a few of many potential variables of that kind. Accordingly, these features have been discussed and investigated extensively at the centre of current interaction research. Although more than thirty years of interaction research have no doubt revealed that interaction precipitates L2 learning, we now have much more complicated issues to deal with than when this idea of interaction first emerged, and far more questions than answers when it comes to how it does so. From a pedagogical standpoint, task-based language teaching (TBLT) creates an optimal condition for interaction-driven L2 learning (see Long, 2015, for a recent volume on TBLT) because negotiation for meaning, during which “important brief opportunities for attention to linguistic code features, and for explicit learning (cf. explicit teaching) to improve implicit input processing, occur” (Long, 2015, p. 53), likely abounds in most interactional tasks that require L2 learners’ active involvement. However, given that various mediating factors (e.g., task type, task complexity, target type, proficiency, etc.) can affect the amount of negotiation for meaning and overall task performance, teachers should be cautious when designing tasks to ensure learners are provided with enough opportunities for negotiated interaction befitting their proficiency, but in a rather challenging way. Well-designed two-way, rather than one-way, tasks with complexity manipulated from simple to complex and with enough cognitive challenges are most likely to elicit much negotiated interaction and afford L2 learners ample opportunities to engage in cognitive processing of the target language with input and output often modified by means of negotiation moves and negative feedback (see Long, 2015, for more suggestions). Of course, focus on form (e.g., Long, 1991, 2000; Long & Robinson, 1998; see also Loewen, 2015, for a recent review), which refers to overtly drawing “students’ attention to linguistic elements as they arise incidentally in lessons whose overriding focus is on meaning, or communication” (Long, 1991, p. 46), should be embedded in TBLT contexts. Long (2015) further elaborates that focus on form involves

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

248

Jaemyung Goo

reactive use of a wide variety of pedagogic procedures (PPs) to draw learners’ attention to linguistic problems in context, as they arise during communication (in TBLT, typically as students work on problem-solving tasks), thereby increasing the likelihood that attention to code features will be synchronized with the learner’s internal syllabus, developmental stage, and processing ability. (p. 27)

Focus on form can be achieved through various attention-drawing techniques, including CF moves (see Doughty & Williams, 1998, for a wider taxonomy as compared to Long’s definition of focus on form). Given interaction researchers have obtained enough empirical findings that clearly suggest CF moves during interaction foster L2 development (see Mackey, 2012; Nassaji, 2016, for reviews), focus on form through various CF moves, not just recasts or just prompts, must be considered as an important teaching practice, and implemented as such. Lastly, Long’s (2009, 2015) list of ten methodological principles for TBLT deserves much attention from language teachers, as they represent previous research findings (e.g., use task as the unit of analysis, focus on form, promote learning by doing, encourage inductive learning, etc.). The proposed principles (see Long, 2015, for detailed discussion) are an important source of reference that language teachers should consult when they plan, design, and reflect on classroom activities and various types of in-class phenomena.

References Adams, R., Nuevo, A.-M., & Egi, T. (2011). Explicit and implicit feedback, modified output, and SLA: Does explicit and implicit feedback promote learning and learner–learner interactions? The Modern Language Journal, 95(S), 42–63. Ahmadian, M. J. (2012). The relationship between working memory capacity and L2 oral performance under task-based careful online planning condition. TESOL Quarterly, 46, 165–175. 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: Researching the Cognition Hypothesis of language learning and performance (pp. 239–265). Amsterdam: John Benjamins. Ammar, A. (2008). Prompts and recasts: Differential effects on second language morphosyntax. Language Teaching Research, 12, 183–210. Ammar, A., & Spada, N. (2006). One size fits all? Recasts, prompts, and L2 learning. Studies in Second Language Acquisition, 28, 543–574. Bao, M., Egi, T., & Han, Y. (2011). Classroom study on noticing and recast features: Capturing learner noticing with uptake and stimulated recall. System, 39, 215–228.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

249

Baralt, M. (2013). The impact of cognitive complexity on feedback efficacy during online versus face-to-face interactive tasks. Studies in Second Language Acquisition, 35, 689–725. Brown, D. (2016). The type and linguistic foci of oral corrective feedback in the L2 classroom: A meta-analysis. Language Teaching Research, 20, 436–458. Chaudron, C. (1977). A descriptive model of discourse in the corrective treatment of learners’ errors. Language Learning, 27, 29–46. Doughty, C. (2001). Cognitive underpinnings of focus on form. In P. Robinson (ed.), Cognition and second language instruction (pp. 206–257). Cambridge: Cambridge University Press. Doughty, C., & Williams, J. (eds.) (1998). Focus on form in classroom second language acquisition. Cambridge: Cambridge University Press. Egi, T. (2007). Interpreting recasts as linguistic evidence: The roles of linguistic target, length, and degree of change. Studies in Second Language Acquisition, 29, 511–537. Egi, T. (2010). Uptake, modified output, and learner perceptions of recasts: Learner responses as language awareness. The Modern Language Journal, 94, 1–21. Ellis, R. (2007). The differential effects of corrective feedback on two grammatical structures. In A. Mackey (ed.), Conversational interaction in second language acquisition: A collection of empirical studies (pp. 339–360). Oxford: Oxford University Press. Ellis, R. (2015). Understanding second language acquisition (2nd edn.). Oxford: Oxford University Press. Ellis, R., Basturkmen, H., & Loewen, S. (2001). Learner uptake in communicative ESL lessons. Language Learning, 51, 281–318. Ellis, R., & He, X. (1999). The roles of modified input and output in the incidental acquisition of word meanings. Studies in Second Language Acquisition, 21, 285–301. Ellis, R., Loewen, S., & Erlam, R. (2006). Implicit and explicit corrective feedback and the acquisition of L2 grammar. Studies in Second Language Acquisition, 28, 339–368. Ellis, R., Tanaka, Y., & Yamazaki, A. (1994). Classroom interaction, comprehension and the acquisition of L2 word meanings. Language Learning, 44(3), 449–491. Foster, P. (1998). A classroom perspective on the negotiation of meaning. Applied Linguistics, 19, 1–23. Foster, P., & Ohta, A. S. (2005). Negotiation for meaning and peer assistance in second language classrooms. Applied Linguistics, 26, 402–430. Gass, S. M. (1997). Input, Interaction, and the Second Language Learner. Mahwah, NJ: Lawrence Erlbaum Associates. Gass, S. M. (2003). Input and interaction. In C. J. Doughty & M. H. Long (eds.), The handbook of second language acquisition (pp. 224–255). Oxford: Blackwell.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

250

Jaemyung Goo

Gass, S. M., & Mackey, A. (2006). Input, interaction and output: An overview. AILA Review, 19, 3–17. Gass, S. M., & Mackey, A. (2015). Input, interaction, and output in second language acquisition. In B. VanPatten & J. Williams (eds.), Theories in second language acquisition (pp. 180–206). New York: Routledge. Gass, S. M., Mackey, A., & Ross-Feldman, L. (2005). Task-based interactions in classroom and laboratory settings. Language Learning, 55, 575–611. Gass, S. M., & Varonis, E. (1994). Input, interaction, and second language production. Studies in Second Language Acquisition, 16, 282–302. Gilabert, R., Baron, J., & Llanes, M. (2009). Manipulating task complexity across task types and its influence on learners’ interaction during oral performance. International Review of Applied Linguistics, 47, 367–395. Goo, J. (2012). Corrective feedback and working memory capacity in interaction-driven L2 learning. Studies in Second Language Acquisition, 34, 445–474. Goo, J. (2016). Corrective feedback and working memory capacity: A replication. In G. Granena, D. O. Jackson, & Y. Yilmaz (eds.), Cognitive individual differences in second language processing and acquisition (pp. 279–302). Amsterdam: John Benjamins. Goo, J., & Mackey, A. (2013). The case against the case against recasts. Studies in Second Language Acquisition, 35, 127–165. Gooch, R., Saito, K., & Lyster, R. (2016). Effects of recasts and prompts on L2 pronunciation development: Teaching English /ɹ/ to Korean adult EFL learners. System, 60, 117–127. Gurzynski-Weiss, L. (2016). Factors influencing Spanish instructors’ in-class feedback decisions. The Modern Language Journal, 100, 255–275. Gurzynski-Weiss, L. (2017). L2 instructor individual characteristics. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 451–467). New York: Routledge. Gurzynski-Weiss, L., & Baralt, M. (2014). Exploring learner perception and use of task-based interactional feedback in FTF and CMC modes. Studies in Second Language Acquisition, 36, 1–37. Gurzynski-Weiss, L., & Baralt, M. (2015). Does type of modified output correspond to learner noticing of feedback? A closer look in face-to-face and computer-mediated task-based interaction. Applied Psycholinguistics, 36, 1393–1420. Izumi, S., Bigelow, M., Fujiwara, M., & Fearnow, S. (1999). Testing the Output Hypothesis: Effects of output on noticing and second language acquisition. Studies in Second Language Acquisition, 21, 421–452. Jeon, S. (2007). Interaction-driven L2 learning: Characterizing linguistic development. In A. Mackey (ed.), Conversational interaction in second language acquisition (pp. 379–403). Oxford: Oxford University Press. Keck, C. M., Iberri-Shea, G., Tracy-Ventura, N., & Wa-Mbaleka, S. (2006). Investigating the empirical link between task-based interaction and acquisition: A meta-analysis. In J. M. Norris & L. Ortega (eds.), Synthesizing

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

251

research on language learning and teaching (pp. 91–131). Amsterdam: John Benjamins. Kim, Y. (2009). The effects of task complexity on learner–learner interaction. System, 37, 254–268. Kim, Y. (2012). Task complexity, learning opportunities and Korean EFL learners’ question development. Studies in Second Language Acquisition, 34, 627–658. Kim, Y., Payant, C., & Pearson, P. (2015). The intersection of task-based interaction, task complexity, and working memory: L2 question development through recasts in a laboratory setting. Studies in Second Language Acquisition, 37, 549–581. Kormos, J., & 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. 267–285). Amsterdam: John Benjamins. Leeser, M. J., & Sunderman, G. L. (2016). Methodological implications of working memory tasks for L2 processing research. In G. Granena, D. O. Jackson, & Y. Yilmaz (eds.), Cognitive individual differences in second language processing and acquisition (pp. 279–302). Amsterdam: John Benjamins. Li, S. (2010). The effectiveness of corrective feedback in SLA: A metaanalysis. Language Learning, 60, 309–365. Li, S. (2013). The interactions between the effects of implicit and explicit feedback and individual differences in language analytic ability and working memory. The Modern Language Journal, 97, 634–654. Li, S. (2017). Cognitive differences and ISLA. In S. Loewen & M. Sato (eds.), The Routledge handbook of instructed second language acquisition (pp. 396– 417). New York: Routledge. Loewen, S. (2005). Incidental focus on form and second language learning. Studies in Second Language Acquisition, 27, 361–386. Loewen, S. (2015). Introduction to instructed second language acquisition. New York: Routledge. Loewen, S., & Philp, J. (2006). Recasts in the adult English L2 classroom: Characteristics, explicitness, and effectiveness. The Modern Language Journal, 90, 536–556. Long, M. H. (1985). Input and second language acquisition theory. In S. M. Gass & C. G. Madden (eds.), Input in second language acquisition (pp. 377– 393). Rowley, MA: Newbury House. Long, M. H. (1991). Focus on form: A design feature in language teaching methodology. In K. De Bot, R. B. Ginsberg, & C. Kramsch (eds.), Foreign language research in cross-cultural perspective (pp. 39–52). Amsterdam: John Benjamins. Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. Richie & T. K. Bhatia (eds.), Handbook of second language acquisition (pp. 413–468). New York: Academic Press.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

252

Jaemyung Goo

Long, M. H. (2000). Focus on form in task-based language teaching. In R. L. Lambert & E. Shohamy (eds.), Language policy and pedagogy (pp. 179–192). Amsterdam: John Benjamins. Long, M. H. (2007). Problems in SLA. Mahwah, NJ: Lawrence Erlbaum Associates. Long, M. H. (2009). Methodological principles for language teaching. In M. H. Long & C. J. Doughty (eds.), The handbook of language teaching (pp. 373–394). Malden, MA: Wiley-Blackwell. Long, M. H. (2015). Second language acquisition and task-based language teaching. Malden, MA: Wiley-Blackwell. Long, M. H., Inagaki, S., & Ortega, L. (1998). The role of implicit negative feedback in SLA: Models and recasts in Japanese and Spanish. The Modern Language Journal, 82, 357–371. Long, M. H., & Robinson, P. (1998). Focus on form: Theory, research, and practice. In C. Doughty & J. Williams (eds.), Focus on form in classroom second language acquisition (pp. 15–41). Cambridge: Cambridge University Press. Loschky, L. (1994). Comprehensible input and second language acquisition: What is the relationship? Studies in Second Language Acquisition, 16(3), 305–325. Lyster, R. (1998a). Recasts, repetition, and ambiguity in L2 classroom discourse. Studies in Second Language Acquisition, 20, 51–81. Lyster, R. (1998b). 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). Differential effects of prompts and recasts in formfocused instruction. Studies in Second Language Acquisition, 26, 399–432. Lyster, R., & Izquierdo, J. (2009). Prompts versus recasts in dyadic interaction. Language Learning, 59, 453–498. Lyster, R., & Mori, H. (2006). Interactional feedback and instructional counterbalance. Studies in Second Language Acquisition, 28, 269–300. Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms. Studies in Second Language Acquisition, 19, 37–66. Lyster, R., & Ranta, L. (2013). Counterpoint piece: The case for variety in corrective feedback research. Studies in Second Language Acquisition, 35, 167–184. Lyster, R., & Saito, K. (2010). Oral feedback in classroom SLA: A meta-analysis. Studies in Second Language Acquisition, 32, 265–302. Lyster, R., Saito, K., & Sato, M. (2013). Oral corrective feedback in second language classrooms. Language Teaching, 46, 1–40. Mackey, A. (1999). Input, interaction, and second language development: An empirical study of question formation in ESL. Studies in Second Language Acquisition, 21, 557–587. Mackey, A. (2006). Feedback, noticing and instructed second language learning. Applied Linguistics, 27, 405–430.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

253

Mackey, A. (2012). Input, interaction, and corrective feedback in L2 classrooms. Oxford: Oxford University Press. Mackey, A., Abbuhl, R., & Gass, S. (2012). Interactionist approach. In S. Gass & A. Mackey (eds.), The Routledge handbook of second language acquisition (pp. 7–24). New York: Routledge. Mackey, A., Adams, R., Stafford, C., & Winke, P. (2010). Exploring the relationship between modified output and working memory capacity. Language Learning, 60, 501–533. Mackey, A., Gass, S. M., & McDonough, K. (2000). How do learners perceive interactional feedback? Studies in Second Language Acquisition, 22, 471–497. Mackey, A., & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (ed.), Conversational interaction in second language acquisition: A collection of empirical studies (pp. 407–452). Oxford: Oxford University Press. Mackey, A., Oliver, R., & Leeman, J. (2003). Interactional input and the incorporation of feedback: An exploration of NS–NNS and NNS–NNS adult and child dyads. Language Learning, 53, 35–66. Mackey, A., & Philp, J. (1998). Conversational interaction and second language development: Recasts, responses, and red herrings? The Modern Language Journal, 82, 338–356. Mackey, A., Philp, J., Egi, T., Fujii, A., & 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). Amsterdam: Benjamins. Mackey, A., Polio, C., & McDonough, K. (2004). The relationship between experience, education and teachers’ use of incidental focus-on-form techniques. Language Teaching Research, 8, 301–327. Mackey, A., & Sachs, R. (2012). Older learners in SLA research: A first look at working memory, feedback, and L2 development. Language Learning, 62, 704–740. McDonough, K. (2005). Identifying the impact of negative feedback and learners’ responses on ESL question development. Studies in Second Language Acquisition, 27, 79–103. McDonough, K. (2007). Interactional feedback and the emergence of simple past activity verbs in L2 English. In A. Mackey (ed.), Conversational interaction in second language acquisition: A collection of empirical studies (pp. 323–338). Oxford: Oxford University Press. McDonough, K., Crawford, W. J., & Mackey, A. (2015). Creativity and EFL students’ language use during a group problem-solving task. TESOL Quarterly, 49, 188–199. McDonough, K., & Mackey, A. (2006). Responses to recasts: Repetitions, primed production, and linguistic development. Language Learning, 56, 693–720. Nakatsukasa, K. (2016). Efficacy of recasts and gestures on the acquisition of locative prepositions. Studies in Second Language Acquisition, 38, 771–799.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

254

Jaemyung Goo

Nassaji, H. (2009). Effects of recasts and elicitation in dyadic interaction and the role of feedback explicitness. Language Learning, 59, 411–452. Nassaji, H. (2016). Interactional feedback in second language teaching and learning: A synthesis and analysis of current research. Language Teaching Research, 20, 535–562. Nassaji, H. (2017). The effectiveness of extensive versus intensive recasts for learning L2 grammar. The Modern Language Journal, 101, 353–368. Nobuyoshi, J., & Ellis, R. (1993). Focused communication tasks and second language acquisition. English Language Teaching Journal, 47, 203–210. Nuevo, A.-M. (2006). Task complexity and interaction: L2 learning opportunities and development. Unpublished PhD dissertation, Georgetown University, Washington, DC. Nuevo, A.-M., Adams, R., & Ross-Feldman, L. (2011). Task complexity, modified output, and L2 development in leaner–learner interaction. In P. Robinson (ed.), Second language task complexity: Researching the Cognition Hypothesis of language learning and performance (pp. 175–201). Amsterdam: John Benjamins. Oliver, R., & Mackey, A. (2003). Interactional context and feedback in child ESL classrooms. The Modern Language Journal, 87, 519–533. Ortega, L. (2009). Understanding second language acquisition. London: Hodder Education. 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. Pica, T. (1992). The textual outcomes of native speaker–non-native speaker negotiation: What do they reveal about second language learning? In C. Kramsch & S. McConnell-Ginet (eds.), Text and context: Cross-disciplinary perspectives on language study (pp. 198–237). Lexington, MA: D.C. Heath and Company. Pica, T. (1994). Research on negotiation: What does it reveal about second language learning conditions, processes, and outcomes? Language Learning, 44, 493–527. Pica, T. (1996). Do second language learners need negotiation? International Review of Applied Linguistics in Language Teaching (IRAL), 34, 1–21. Pica, T. (2002). Subject-matter content: How does it assist the interactional and linguistic needs of classroom language learners? The Modern Language Journal, 86, 1–19. Pica, T., Holliday, L., Lewis, N. E., & Morgenthaler, L. (1989). Comprehensible output as an outcome of linguistic demands on the learner. Studies in Second Language Acquisition, 11, 63–90. Pica, T., Kanagy, R., & Falodun, J. (1993). Choosing and using communication tasks for second language instruction. In G. Crookes & S. M. Gass (eds.), Tasks and language learning (pp. 9–34). Clevedon: Multilingual Matters.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

255

Pica, T., Young, R., & Doughty, C. (1987). The impact of interaction on comprehension. TESOL Quarterly, 21, 737–58. Polio, C., Gass, S., & Chapin, L. (2006). Using stimulated recall to investigate native speaker perceptions in native–nonnative speaker interaction. Studies in Second Language Acquisition, 28, 237–267. Rassaei, E. (2015a). Recasts, field dependence/independence cognitive style, and L2 development. Language Teaching Research, 19, 499–518. Rassaei, E. (2015b). Oral corrective feedback, foreign language anxiety and L2 development. System, 49, 98–109. Révész, A. (2009). Task complexity, focus on form, and second language development. Studies in Second Language Acquisition, 31, 437–470. Révész, A. (2011). Task complexity, focus on L2 constructions, and individual differences: A classroom-based study. The Modern Language Journal, 95(S), 162–181. Révész, A. (2012). Working memory and the observed effectiveness of recasts on different L2 outcome measures. Language Learning, 62, 93–132. Révész, A., Sachs, R., & Hama, M. (2014). The effects of task complexity and input frequency on the acquisition of the past counterfactual construction through recasts. Language Learning, 64, 615–650. Révész, A., Sachs, R., & Mackey, A. (2011). Task complexity, uptake of recasts, and L2 development. In P. Robinson (ed.), Second language task complexity: Researching the Cognition Hypothesis of language learning and performance (pp. 203–235). Amsterdam: John Benjamins. Robinson, P. (2001a). Task complexity, cognitive resources, and syllabus design: A triadic framework for examining task influences on SLA. In P. Robinson (ed.), Cognition and second language instruction (pp. 287–318). Cambridge: Cambridge University Press. Robinson, P. (2001b). Task complexity, task difficulty, and task production: Exploring interactions in a componential framework. Applied Linguistics, 22, 27–57. Robinson, P. (2003). The Cognition Hypothesis, task design, and adult taskbased language learning. Second Language Studies, 21, 45–105. Robinson, P. (2007). Criteria for classifying and sequencing pedagogic tasks. In M. P. García Mayo (ed.), Investigating tasks in formal language learning (pp. 7–26). Clevedon: Multilingual Matters. Robinson, P. (2011a). Task-based language learning: A review of issues. Language Learning, 61(S1), 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. 3–37). Amsterdam: Benjamins. Sagarra, N. (2007). From CALL to face-to-face interaction: The effect of computer-delivered recasts and working memory on L2 development. In A. Mackey (ed.), Conversational interaction in second language acquisition:

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

256

Jaemyung Goo

A collection of empirical studies (pp. 229–248). Oxford: Oxford University Press. Sagarra, N., & Abbuhl, R. (2013). Optimizing the noticing of recasts via computer-delivered feedback: Evidence that oral input enhancement and working memory help second language learning. The Modern Language Journal, 97, 196–216. Sato, M., & Lyster, R. (2007). Modified output of Japanese EFL learners: Variable effects of interlocutor versus feedback types. In A. Mackey (ed.), Conversational interaction in second language acquisition: A collection of empirical studies (pp. 123–142). Oxford: Oxford University Press. Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129–158. Schmidt, R. (1995). Consciousness and foreign language learning: A tutorial on the role of attention and awareness in learning. In R. Schmidt (ed.), Attention and awareness in foreign language learning (pp. 1–63). Honolulu, HI: University of Hawai’i Press. Schmidt, R. (2001). Attention. In P. Robinson (ed.), Cognition and second language instruction (pp. 3–32). 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: A collection of empirical studies (pp. 301–322). Oxford: Oxford University Press. Sheen, Y. (2008). Recast, language anxiety, modified output and L2 learning. Language Learning, 58, 835–874. Shehadeh, A. (1999). Non-native speakers’ production of modified comprehensible output and second language learning. Language Learning, 49, 627–675. Shehadeh, A. (2001). Self- and other-initiated modified output during taskbased interaction. TESOL Quarterly, 35, 433–457. Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. G. Madden (eds.), Input in second language acquisition (pp. 235–253). Rowley, MA: Newbury House. Swain, M. (1993). The Output Hypothesis: Just speaking and writing aren’t enough. The Canadian Modern Language Review, 50, 158–164. Swain, M. (1995) Three functions of output in second language learning. In G. Cook and B. Seidelhofer (eds.) Principle and practice in applied linguistics: Studies in honour of H. Widdowson (pp. 125–144). Oxford: Oxford University Press.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

Interaction in L2 Learning

257

Swain, M. (2005). The Output Hypothesis: Theory and research. In E. Hinkel (ed.), Handbook on research in second language teaching and learning (pp. 471–484). Mahwah, NJ: Lawrence Erlbaum. Trofimovich, P., Ammar, A., & Gatbonton, E. (2007). How effective are recasts? The role of attention, memory, and analytic ability. In A. Mackey (ed.), Conversational interaction in second language acquisition (pp. 171–195). Oxford: Oxford University Press. van de Guchte, M., Braaksma, M., Rijlaarsdam, G., & Bimmel, P. (2015). Learning new grammatical structures in task-based language learning: The effects of recasts and prompts. The Modern Language Journal, 99, 246–262. Van den Branden, K. (1997). Effects of negotiation on language learners’ output. Language Learning, 47, 589–636. Wen, Z. (2016). Working memory and second language learning: Towards an integrated approach. Bristol: Multilingual Matters. Yang, Y., & 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. Yilmaz, Y. (2012). The relative effects of explicit correction and recasts on two target structures via two communication modes. Language Learning, 62, 1134–1169. Yilmaz, Y. (2013a). The relative effectiveness of mixed, explicit and implicit feedback in the acquisition of English articles. System, 41, 691–705. Yilmaz, Y. (2013b). Relative effects of explicit and implicit feedback: The role of working memory capacity and language analytic ability. Applied Linguistics, 34, 344–368. Yilmaz, Y., & Gisela, G. (2016). The role of cognitive aptitudes for explicit language learning in the relative effects of explicit and implicit feedback. Bilingualism: Language and Cognition, 19, 147–161.

Downloaded from https://www.cambridge.org/core. UB der LMU München, on 19 Jun 2020 at 14:34:11, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.011

11 Speaking Dustin Crowther and Susan M. Gass 11.1 Introduction In describing the difficulties of establishing a definition for the construct second language (L2) “speaking”, Fulcher (2014) notes that “speaking is the verbal use of language to communicate with others” (p. 23). Such a broad definition is of course limited, as what enables successful spoken communication is variable. For example, Fulcher lists the following considerations: pronunciation and intonation; accuracy and fluency; strategies for speaking; structuring speech; speaking in context; and interactional competence. Such variability is evident when considering how high-stakes language assessments (e.g., the American Council on the Teaching of Foreign Languages (ACTFL), instruments based on the Common European Framework of Reference, International English Language Testing System (IELTS), Test of English as a Foreign Language (TOEFL)) define speaking within their frameworks. For example, whereas the IELTS speaking rubric includes categories for Fluency and Coherence, Lexical Resource, Grammatical Range and Accuracy, and Pronunciation, the TOEFL speaking rubrics emphasize General Description, Delivery, Language Use, and Topic Development.1 A quick glance across the two rubrics would seemingly indicate that IELTS emphasizes the linguistic and TOEFL the contextual considerations of L2 speaking. While IELTS and TOEFL are English specific, the ACTFL guidelines were developed to be language neutral, being relevant to a wide range of languages. While making reference to linguistic dimensions (phonology, lexicon, syntax), ACTFL emphasizes the effectiveness in which speakers are expected to carry out a range of tasks. For example, a superior (a term used to designate one of five broad proficiency levels) speaker will be able to communicate with accuracy and fluency in order to participate fully and effectively in conversations on a variety of topics in formal and informal settings … present their opinions on a number of issues of interest to them, such as social and political issues, and provide structured arguments to support these opinions. (ACTFL, 2012, p. 5)  Refer to https://www.ielts.org/en-us/ielts-for-organisations/ielts-scoring-in-detail for the IELTS speaking rubric and to

1

https://www.ets.org/toefl/ibt/scores/understand for the TOEFL speaking rubrics.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

259

Recognizing this variability in defining L2 speaking, in this chapter, we address the construct from macro- and micro-perspectives. We first draw upon a large database of oral proficiency interviews across foreign languages (e.g., French, Spanish, Russian, and Chinese) to see what gains universitylevel learners in the US make across multiple semesters of study. We then review phonology-, fluency-, and sociolinguistic-based variables associated with listener perception of L2 speech performance. We conclude with a discussion on developmental variables (e.g., age, immersion, study abroad) that may influence the extent to which L2 learners may develop their L2 speaking ability.

11.2  Attainment: What are Reasonable Expectations? We explore attainment from the perspective of a large dataset that considered proficiency attainment across different skills: speaking, listening, and reading. The data come from a three-year study (2014–2017) that considered proficiency results across multiple languages and multiple years of university study.2 In that study, data were collected from students enrolled in Chinese, French, Russian, or Spanish classes at a large Midwestern US university. Table 11.1 reflects the numbers of tests administered. Standardized tests, based on ACTFL standards, were used to assess speaking, listening, and reading. For speaking, a computerized test (OPIc) was administered online through Language Testing International, who was responsible for grading the oral sample provided by the student.

Table 11.1.  Number of students who took proficiency tests administered at Michigan State University from 2014–2017 in four languages. 2014–2015

2015–2016

2016–2017*

Chinese

250

272

162

French

526

510

301

Russian

116

115

57

Spanish

1155

962

936

Total

2047

1859

1456

Note. Tests were not administered in Fall 2016, so the 2016–2017 numbers only include Spring 2017.

 Partial funding for this research comes from the National Security Education Program’s Language Flagship

2

Proficiency Initiative, Award Number 0054-MSU-22-PI-280-PO2, granted to Paula Winke and Susan Gass.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

260

D ustin C rowt h er and S usan M . G ass

8 7 6 5 4 3 2 1

AL IL NM Year 1

Year 2 Speaking

Year 3 Listening

Year 4 Reading

Figure 11.1.  ACTFL spring 2017 results for speaking, listening, and reading. Note: Years 1–4 on the x-axis indicate years of study (e.g., first-year courses through fourth-year courses).

The listening and reading tests were computer-graded, and scores were determined immediately. The OPIc used an avatar to deliver questions. Following the ACTFL scale, students were given ratings in one of five categories: (1) Novice; (2) Intermediate; (3) Advanced; (4) Superior; and (5) Distinguished. The three lowest levels were subdivided into sublevels low (L), mid (M), and high (H). Thus, there are essentially eleven possible ratings, although none of our students reached distinguished, and very few reached superior in any skill. The majority of speakers fell into the Novice (N), Intermediate (I), and Advanced (A) bands. More information on the descriptors can be found in ACTFL (2012). For purposes of displaying the results, each level was converted into a numerical value, ranging from 1 (Novice low (NL)) to 10 (Superior). Figure 11.1 shows the results from the final year of our study (Spring 2017). As can be seen, at the end of four years of study, speaking was the lowest skill attained, with students not even reaching the Intermediate High level. The increase in speaking skills over a four-year period is somewhat flat when compared to the rapid jump in the other skills, in particular, from year two to year three. Although both of these observations may be a reflection of curricular emphases, it is noteworthy that these data do include students with study abroad experience where speaking was undoubtedly a part of a student’s daily life. What we have seen thus far is that for skill development in a US foreign language setting, speaking is the weakest skill. This is surprising given the emphasis on the importance of speaking from an L2 acquisition perspective, where the Interaction Approach has been on the forefront of research and emphasizes the important role of output, a topic to which we turn next.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

261

11.3  The Importance of Speaking: Output Output, in its simplest form, refers to the production of language, and incorporates oral, written, or signed production. The construct of output became theoretically important in the 1980s when Swain (1985) introduced the notion of comprehensible output, further elaborated on in Swain (1995, 2005). Output “may force the learner to move from semantic processing to syntactic processing” (1985, p. 249). In other words, using language productively (as opposed to receptively) forces learners to join sounds and put words in a particular order. Comprehensible output refers to the benefits derived when learners are “pushed toward the delivery of a message that is not only conveyed, but that is conveyed precisely, coherently, and appropriately’ (1985, p. 249). Swain (1995) further claimed that output may stimulate learners to move from the semantic, open-ended, nondeterministic, strategic processing prevalent in comprehension to the complete grammatical processing needed for accurate production. Output, thus, would seem to have a potentially significant role in the development of syntax and morphology. (p. 128)

Thus, speaking, as an activity, coupled with feedback from an interlocutor, is a necessary (although not sufficient) part of learning. This is so because in an interactive context, feedback is frequent and draws a learner’s attention to some (usually erroneous) part of his/her spoken utterance, enabling the learner to focus on that part of language and make corrections, as may be appropriate. The Interaction Approach (Long, 1996; Gass & Mackey, 2006, 2007; Mackey, Abbuhl, & Gass, 2012) includes a number of important constructs, of which output is one and feedback another. From a theoretical standpoint, feedback is part of what is known as negative evidence, that is, information to a speaker that there is something that differs from acceptable norms of language. Negative evidence is not a necessary condition for child language acquisition but may play a different role in the case of L2 learning (see for example White, 1991). Within a classroom context, feedback is clearly prevalent and is in fact common although not always pervasive as Zyzik and Polio (2008) showed in the context of Spanish literature classrooms. Loewen (2003) also found variation in feedback even within the same language school and from student to student. In other words, speaking, coupled with feedback, may be important for L2 learning, but caution is warranted because of the lack of consistency in which feedback is given to oral production and the selective nature of L2 learners in incorporating feedback (see Mackey, Gass, & McDonough, 2000). While acknowledging the importance of producing comprehensible output and receiving feedback on this output, the previously referenced ACTFL data make clear that there is potential concern in regard to

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

262

D ustin C rowt h er and S usan M . G ass

speaking development (here limited to a US context) (cf. Winke, Gass, & Heinrich, 2019; Rubio et al., 2018). However, such data do not allow us to pinpoint what it is specifically about L2 learners’ speaking that is not developing. To discuss further, we begin by considering possibly the most intuitive link to L2 speaking, pronunciation.

11.4  Global Perception of L2 Speech The above analysis emphasizes holistic speaking gains, as measured by Language Testing International raters, who have been specifically trained to ensure reliable and unbiased assessments across speakers. However, empirical research into L2 speech production has also frequently employed listeners, a population more variable in membership, and usually employed to provide a more global assessment of speech production (Yan & Ginther, 2018). Within such studies, researchers’ interest is focused primarily on how naïve listeners (i.e., without formal training) perceive the speech of L2 users (e.g., Derwing & Munro, 1997; Munro & Derwing, 1999), and which linguistic measures of speech influence this perception (e.g., Crowther et al., 2015; Trofimovich & Isaacs, 2012; Varonis & Gass, 1982). Such research has emphasized three key constructs, here defined following Derwing and Munro (2015, p. 5): a. accentedness—how distinguishable an L2 learners’ speech pattern is from that of a member of the target speech community, b. comprehensibility—how easy or difficult to understand a listener finds an L2 speaker’s utterance to be, c. intelligibility—how accurately a listener understands an L2 speaker’s intended message. Importantly, it has been argued that perceived accentedness is partially independent from the two measures of understanding, comprehensibility and intelligibility (Derwing & Munro, 2015). In essence, while low comprehensible or intelligible speakers will nearly always possess a heavy accent, simply possessing a heavy accent is not an automatic predictor of low comprehensibility or intelligibility. As such, there has been a strong belief in L2 pronunciation scholarship that pedagogical approaches towards developing L2 speech should place a greater emphasis on the ability to produce understandable before native-like speech, referred to by Levis (2005) as the Intelligibility and Nativeness principles, respectively. Although studies in this vein have placed a primary emphasis on L2 English, recent years have seen an increased focus on additional languages, including Dutch (Caspers, 2010), French (Bergeron & Trofimovich, 2017), German (O’Brien, 2014), Japanese (Saito & Akiyama, 2017), Korean (Isbell, Park, & Lee, 2019), and Spanish (Nagle, 2018).

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

263

11.4.1  Linguistic Dimensions of L2 Speaking L2 pronunciation scholarship has identified a wide range of linguistic dimensions associated with listeners’ perception of L2 speech. While accentedness is intuitively linked to phonological measures, comprehensibility is generally more diverse, encompassing phonology, fluency, and lexicogrammar measures (e.g., Crowther et al., 2015; Trofimovich & Isaacs, 2012). An emphasis here is placed on comprehensibility (ease of understanding) over intelligibility (accuracy), as the scalar approach of the former is more in line with how “speaking” is operationalized in high stakes assessment rubrics (Isaacs & Trofimovich, 2012). It should be noted, however, that it does not inform us on how accurate understanding was, which is often measured at the segment, word, and phrasal levels through techniques such as orthographic transcription, truth-value judgments, and forced-choice tasks (Thomson, 2018). Many of the studies described above are of a one-shot nature and fail to tell us much with regard to development. However, some insight can be gleaned from the works of Derwing, Munro, Thomson, and colleagues, who have published some of the few pronunciation studies using longitudinal data. Focusing on the pronunciation gains of first language (L1) Mandarin and L1 Slavic learners of English in Canada, the authors have looked at accentedness, comprehensibility, and fluency development ranging from ten months to seven years (e.g., Derwing & Munro, 2013; Derwing, Munro, & Thomson, 2007; Derwing et al., 2009; Derwing, Thomson, & Munro, 2006; Munro, Derwing, & Thomson, 2015). Derwing and Munro (2013) looked at gains in perceived accentedness, comprehensibility, and fluency over the entire seven-year period, which covered time spent in ESL classes and beyond. For their Slavic learners, Derwing and Munro observed continued development in comprehensibility and fluency over the seven years, but not for accentedness, which appeared to level off after two years. However, for the Mandarin learners, who began as markedly worse than the Slavic learners, no perceptual gains were found across the three global measures during the seven-year period. In discussing their findings, Derwing and Munro highlight “the complex interplay of first language, the extent of linguistic interactions in the L2, and overall WTC [willingness to communicate]” (p. 177), all topics of focus later in this chapter.

11.5  Measures of Fluency When considering Derwing and Munro (2013), it should be noted that “fluency” as a construct has been variable in its definition (Chambers, 1997). For example, Derwing and Munro (2015) define it as the ease of flow of L2 speech, typically in reference to the presence or absence of pauses and

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

264

D ustin C rowt h er and S usan M . G ass

other dysfluency markers. Lennon (1990) discusses “broad” (global speaking ability) versus “narrow” (ease of delivery) fluency. The variability in how L2 scholars have defined and operationalized fluency makes it difficult to draw clear generalizations across studies and develop a theory of L2 fluency acquisition (Segalowitz, 2016). With an aim to explain rather than describe fluency, Segalowitz (2010) states that L2 fluency should focus on features of L2 performance that are reliable indicators of how efficiently a speaker is able to mobilize and temporally integrate, in a nearly simultaneously way, the underlying processes of planning and assembling an utterance in order to perform a communicatively acceptable speech act. (p. 165)

In addition to the importance of motivation and social context in developing L2 fluency, Segalowitz (2010) highlights the potential link between cognitive and utterance fluency. Cognitive fluency refers to learners’ processing efficiency, which includes “the speed and efficiency of semantic retrieval, the handling of the attention-focusing demands inherent in utterance construction, operations in working memory, among others” (Segalowitz, 2016, p. 82). Utterance fluency, in turn, refers to the temporal measures associated with spoken production, including speech rate and hesitation and pausing phenomena. Viewing Segalowitz’s framework for L2 fluency at its most basic, increased L2 speech production “sharpens the learner’s cognitive-perceptual systems so that these cognitive operations become rapid, efficient, and fluid, resulting in speech output that is fluent” (Segalowitz, 2016, p. 89). In other words, increased language use experience leads to more efficient processing, which in turn leads to more fluent output. Pedagogically, if we accept fluency development as a result of increased usage experience, then we should ideally provide learners with extensive speaking practice in the classroom (e.g., Derwing et al., 2009; Nation, 1989), which in turn will enable the generation and automatization of useful, reusable utterances (see Gatbonton & Segalowitz, 2005). While not a “theory” of fluency development, such a pedagogical approach aligns strongly with an Interaction Approach to L2 development, as described earlier in this chapter.

11.6  Communicative Competence An emphasis on communicative competence can be traced back nearly fifty years, where Hymes (1971, 1972) argued that speakers need not only be able to speak a language but be able to use this language appropriately within and across various social situations, taking into account different settings and interlocutors. Extending to L2 speaking development, Canale and Swain (1980) identified four types of competences:

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

265

• Grammatical—knowledge of the rules of morphology, syntax, sentencegrammar semantics, and phonology, and lexical knowledge. • Sociolinguistic—appropriateness of utterances within specified communicative events. • Strategic—verbal and non-verbal strategies to compensate for breakdowns in communication. • Discourse—language across specific domains (added later in Canale, 1983, p. 9). Here we provide an emphasis on communicative competence as the writings of Hymes, Canale, and Swain were highly influential in the development of Communicative Language Teaching in North America and Europe during the 1970s and 1980s (see Duff, 2014, for an overview of the history of communicative language teaching). Communicative language teaching not only has strong ties to the Interaction Approach, it has also informed more contemporary L2 pedagogical methods, such as content- and taskbased instruction (see Nassaji & Kartchava and Long, Lee, & Hillman, in this volume, respectively). Since the works of Hymes, the term communicative competence has continued to evolve. Byram (1997) proposed intercultural communicative competence, with a pedagogical emphasis on raising L2 learners’ recognition of how different cultural contexts may impact how the language they produce is perceived and interpreted. Risager (2007) argued for a need to develop L2 speakers into multilingually and multiculturally aware global citizens, and Kumaravadivelu (2008) proposed L2 speakers learn from other cultures as opposed to about them. Drawing from these different views on communicative competence, the term interculturality has increased in usage (Baker, 2015; Dervin & Risager, 2015). Though not all of the above directly addresses “speaking”, each can potentially shape an interactive event, which in turn shapes the input, output, and feedback one receives or produces. Considering the importance that the Interaction Approach prescribes to each, we cannot look past the role of linguistic and cultural diversity in language use (e.g., Baker, 2015; De Costa, 2012, 2014).

11.7  Individual Differences There are a range of individual-specific variables that may influence the extent and rate in which learners develop their ability to speak in their L2. While there is not space in one chapter to address all potential considerations, below we present five of the variables that have received strong empirical focus. We present them in alphabetical order to avoid placing priority on any one measure, as all have been argued to impact L2 speech development to different extents.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

266

D ustin C rowt h er and S usan M . G ass

11.7.1 Age In this section, we consider age as a factor in predicting language proficiency. Most of the literature related to age has been situated within what is known as the Critical Period or Sensitive Period. The main theoretical notion is that after a certain age, native-like L2 proficiency is not possible. Arguments have appeared on both sides of this debate. As a case in point, Gregg (1996) made an unequivocal statement that “truly nativelike competence in an L2 is never attained” (p. 52). Brain maturation is often used as an explanation for this phenomenon. However, others have focused on exceptions to the idea that late onset cannot result in native-like proficiency in an L2. Abrahamsson and Hyltenstam (2009), in a detailed study of L2 learners of Swedish, investigated this issue by first identifying a group of individuals who appeared to be native-like Swedish speakers by means of self-identification as well as perception by others. This was followed by an “in-depth scrutiny of actual, linguistic nativelikeness” (p. 260). This scrutiny encompasses a wide range of areas including phonetics, perception embedded in white noise and babble noise, written and auditory judgments, inferencing, and formulaic language. 195 participants were divided into two groups: (1) those who began their Swedish learning before age 12 (n = 107) and those who began at age 12 or later (n = 88). The average age of the former group, at the time of testing was 28–29 years and the average age for the older onset age group was 41–42 years. Other important variables, such as length of residence, were controlled. Based on self and listener ratings, most (156 of 195) did pass for native speakers. However, following a detailed linguistic scrutiny of native-likeness of forty-one participants (thirty-one in the early age group; ten in the late age group), only a small number of the early age learners and none of the late learners could be classified as native-like. One important takeaway from this study, given our focus on oral skills, is that the late learner who most resembled a native speaker (seven of ten measures of proficiency) deviated from native-speaker norms on the phonetic aspects of production and perception, suggesting that native-like pronunciation is, indeed, virtually impossible. Piske, MacKay, and Flege (2001) investigated Italian (L1)-English (L2) speakers who differed on their self-reported use of Italian. Some were early learners (learned English as children) and others were late learners (learned English as adolescents). Within these two broad categories were low-use participants and high-use participants. Gender was also a variable although it turned out not to be a predictor of accent. Three sentences were read by each of the participants and each sentence was then rated on a nine-point scale ranging from no foreign accent to very strong foreign accent. The results supported previous studies in that age of learning had an effect on pronunciation, but what was particularly interesting was the

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

267

finding that amount of L1 use (in this case Italian) impacted both early and late learners equally, suggesting this important factor independent of age of learning. Those who used Italian often tended to have stronger accents in English. Muñoz (2014) considered long-term effects of onset of L2 learning on oral performance (Spanish speakers learning English). The study also investigated the not unrelated issue of input, which is carefully described through instructional input (number of curricular and extracurricular hours and number of years and study abroad [amount of time spent in an immersion environment]). Her oral production measures included speech rate, lexical diversity, syntactic complexity, and accuracy. Performance was more closely associated with input characteristics than with starting age. She supports DeKeyser (2000), who argues that children use implicit learning mechanisms which require more input than may be available in an instructional setting. Older learners are advantaged in a school setting where more explicit learning takes place. In other words, starting age is not particularly relevant given the different learning mechanisms used by children versus adolescents. What has greater predictability for oral production is cumulative exposure and quality of contact with native speakers (for a more detailed discussion of age-related factors, see Muñoz, this volume). The role of age in L2 speaking development is likely to remain an important topic of inquiry, as demonstrated in recent publications such as Saito et al. (2018), who consider the effect of age of arrival and length of residence on low-, mid-, and high-level L2 fluency performance.

11.7.2  Ethnic Group Affiliation The strength of pride and loyalty towards one’s ethnic group has been shown to both positively and negatively affect L2 development. Drawing from the at times tense multicultural and multilingual context in Québec, Canada, Gatbonton and colleagues have considered the relationship between ethnic group affiliation and L2 proficiency (Gatbonton & Trofimovich, 2008) and ethnic group affiliation and pronunciation accuracy (Gatbonton, Trofimovich, & Magid, 2005; Gatbonton, Trofimovich, & Segalowitz, 2011). Specifically, they considered the English proficiency of French-dominant bilinguals. Highlighting the multidimensional nature of ethnic group affiliation, Gatbonton and Trofimovich (2008) indicated that English learners with strong ethnic group identification and a more positive orientation towards the L2 group demonstrated higher L2 proficiency, while those with strong support for their ethnic group’s sociopolitical aspirations tended to demonstrate lower L2 proficiency. This negative association between political ethnic group affiliation and L2 proficiency carried

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

268

D ustin C rowt h er and S usan M . G ass

directly into pronunciation accuracy (Gatbonton, Trofimovich, & Magid, 2005; Gatbonton, Trofimovich, & Segalowitz, 2011). A key mediating factor of these relationships was self-reported L2 use. Though they were unable to empirically pursue this finding further, Gatbonton and colleagues hypothesize that greater ethnic group affiliation (specifically in regard to political beliefs) likely led to reduced L2 contact and usage, which in turn led to “poorer development of psycholinguistic perception and production processes underlying L2 oral proficiency” (Gatbonton, Trofimovich, & Segalowitz, 2011, p. 199). The above studies provide just one example of how learner identity may impact L2 development. And, in essence, it refers us back to quantity and quality of language contact. Those looking for more information on this topic are referred to Darvin and Norton (this volume), who discuss how identity interacts with L2 acquisition.

11.7.3  First Language The strength of impact that a learner’s L1 has on their L2 development has been a highly discussed topic in L2 research (see Foley & Flynn, 2013, for a more general review of L1 effects in L2 development). Though also highlighting considerations beyond L1 influence, Eckman (2004) states “L1 influence has always been, and still is, a factor in explaining L2 pronunciation” (p. 543). For example, a key area of empirical interest has been the acquisition of English /r/ and /l/ by L1 Japanese learners (e.g., Aoyama et al., 2004; Saito & Lyster, 2012). For such learners, for whom these two phonemes are non-contrastive in their native language, the ability to discriminate both perceptually and productively has proven problematic. Such findings have provided support for the belief that before L2 learners can produce two distinct sounds, they must be able to perceive them as distinct (although see Gass, 1984, for a study in which production preceded perception). While we do not have space to provide an in-depth review of the sources of perceptual difficulties, one key consideration, which ties back to our previous discussion on age, is that our phonemic inventory is often set by the time we are one year old, and, as such, our ability to differentiate between non-native phonemic contrasts decreases (e.g., Kuhl, 2004; Werker & Tees, 2005; also, see Patricia Kuhl’s 2011 Ted Talk “The Linguistic Genius of Babies” for a more visual demonstration of this effect). Two key models of pronunciation, Best’s Perceptual Assimilation Model (PAM; e.g., Best, 1993, 1994) and Flege’s Skill Learning Model (SLM; e.g., Flege, 1995, 1999), attempt to account for the difficulties L2 learners face in phonemic acquisition. Both models are based in listener perception, an area of research surprisingly more frequent in psycholinguistics than it is in L2 acquisition (Derwing & Munro, 2015). In PAM, adult L2 learners perceive foreign sounds in relationship to their own phonemic inventory.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

269

In essence, they attempt to assimilate these new sounds into an already existing category. For example, Best and Tyler (2007) note how L1 English learners of French are likely to recognize French /r/ and English /r/ as phonological equivalents, despite different phonetic realizations. In SLM, it is hypothesized that new sounds that are more perceptually similar to existing sounds in the L1 will be more difficult to acquire than new sounds that are perceptually different. Guided by the SLM assumption that speech production eventually falls in line with speech perception (e.g., Thomson, Neary, & Derwing, 2009), there has been an increase in approaches towards addressing L2 speakers’ perception of foreign sounds (e.g., Hardison, 2005; Okuno & Hardison, 2016; Thomson, 2011, 2012). Also, please see Sakai and Moorman (2018) for a twenty-five-year meta-analytic review of the effectiveness of perception training on L2 phoneme production. While the above models focus on phonemic acquisition, evidence also exists for L1 influence in suprasegmental acquisition. For example, Trofimovich and Isaacs (2012) ran correlations between listener ratings of the accentedness and comprehensibility of L1 French speakers of English and nineteen different linguistic measures of speech. For both accentedness and comprehensibility, strong correlations (r > .70) were found for word stress and rhythm (even stronger than for individual sounds). The authors linked this strong association to the fact that word stress is non-contrastive in French and rhythm generally lacks the alterations found in English. The authors state that ‘unpredictable’ stress placement (word stress) and alternations in stress (rhythm) in English would pose a major learning challenge for L1 French speakers, regardless of proficiency level, and would be a salient factor influencing listeners’ perception of both accent and comprehensibility.  (p. 914)

Alluding to the number of L2 English speakers with word stress and rhythm difficulties (from a variety of L1 backgrounds), they emphasize that a high global importance be placed on these two measures. Interested readers might consider recent publications such as Tao and Taft (2017), who compared the effect of home language exposure on L2 perception and production, or Munro (2018), who reviews the potential flaws of utilizing a contrastive approach towards predicting L2 pronunciation difficulties.

11.7.4  Study Abroad Many studies have emphasized the benefits of study abroad and, in particular, the impact that studying abroad can have on learner proficiency gains, often focusing on oral proficiency. In fact, the 2017 American Academy of Arts and Sciences report on language learning in the United

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

270

D ustin C rowt h er and S usan M . G ass

States included policy statements among which was the goal to increase access to study abroad. They explicitly highlighted the benefits for “students to travel, experience other cultures, and immerse themselves in languages as they are used in everyday interactions and across all segments of society” (American Academy of Arts and Sciences, 2017, p. 27.) When attempting to understand the value of study abroad, one must remember that an understanding of language gains (speaking and other skills) must take into account learner internal variables as well as contextual variables, including length of stay and living contexts (see Sanz & MoralesFront, 2018, for  extensive coverage of study abroad-related issues). One also has to consider speaking in relation to other skills. For example, Davidson (2010) found that speaking gains in a study abroad context could be predicted on the basis of pre-departure listening scores. Davidson and Shaw (2019) further demonstrate through a large database of learners in study abroad contexts that gains in speaking proficiency vary depending on the starting point, with gains being greater at lower levels than at higher levels. Despite the obvious benefits that accrue from studying abroad (see Jackson and Schwieter, this volume, for a detailed discussion of terminology and differences in study abroad programmes), location of study is not the only answer to increased proficiency. For example, Freed, Segalowitz, and Dewey (2004) found greater benefits for the amount of production (see earlier section on output) in which learners engaged, rather than the venue where learning takes place. Importantly, a study abroad venue provides opportunities for production that are not available in a foreign language context. In general, however, results are mixed. Magnan & Back (2007) focused on semester-long study abroad programmes in France. They found that language gains did occur; students either maintained the same level of proficiency or exhibited gains as measured through oral proficiency interview (OPI) scores. However, commonly-held beliefs about living contexts and the use of authentic media as factors that differentiated learners was not maintained (see Winke & Gass, 2018a for an example of the relevance of authentic materials in reaching advanced levels of proficiency). In a more fine-grained analysis that included detailed language contact information, Hardison (2014b) examined changes in a number of areas following a study abroad experience. In addition to areas not directly related to speaking, she considered changes in oral production, namely, pronunciation, fluency, and accentedness, as determined by raters of speech samples. She further incorporated into her study results from surveys measuring motivation, affect, and self-assessment of oral competence. Finally, she gathered information on the amount and type of language use. A general finding was that study abroad did result in changes

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

271

in all oral skills as well as positive changes in socio-affective measures. Interesting, although not surprising was a pre-study abroad predictor of affect. Those who showed greater positive affect spent more time interacting with native speakers while abroad and had increased positive affect following their study abroad. However, despite this, the study did not find a significant relationship between contact and oral skill improvement. In a study that only focused on pronunciation, Muñoz and Llanes (2014) compared adults and children in two contexts: study abroad (three months) and at home. There were four groups of participants, all Catalan-Spanish learners of English, twenty-eight children (ages 10–11), of whom thirteen had studied abroad, and nineteen adults, fifteen of whom had studied abroad. Recordings were made (pre-study and post-study abroad for those in the study abroad group and at the same interval for the at home group). Listeners rated the samples on a 1–7 Likert scale indicating the perceived degree of foreign accent. Their findings suggest that the learning environment is a greater predictor than age for pronunciation improvement. The child participants showed greater improvement than adults although the difference was not significant. The greater improvement is likely to be not unrelated to the fact that the children interacted more with native speakers than the adults did. They made the important observation that the short amount of time may be insufficient for a stronger pronunciation impact to be noticeable. Segalowitz and Freed (2004), in their comparison of students (English learners of Spanish in Spain) in a study abroad context and at home, found benefits for students in a study abroad context based on speech-specific measures (e.g., temporal and hesitation) as well as holistic measures (ACTFL OPI). But, as they pointed out, learning is complex and there is no single indicator of success. One cannot begin to understand speaking gains without understanding cognitive abilities and the amount of language contact. Perhaps most important is the proficiency level at which one begins study abroad, as the Davidson (2010) and Davidson and Shaw (2019) studies mentioned earlier noted. Individual variables cannot be overemphasized, and, like all facets of language learning, the path to success is not linear. These findings are further supported by a study by Leonard and Shea (2017) dealing with English speakers learning Spanish (Argentina). Unlike the Segalowitz and Freed study, theirs was not a comparison study. Their measures of success were based on complexity, accuracy, and fluency. Again, individual variation was noted, with linguistic knowledge prior to study abroad a predictor of gains, as was their pre-departure processing speed, measured by a picture-naming task and a sentence-picture verification task. With ongoing increases in global mobility, interest in study abroad research is unlikely to wane. Aside from Leonard and Shea (2017) and Davidson and Shaw (2019), discussed above, other recent publications of

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

272

D ustin C rowt h er and S usan M . G ass

note include those by Segalowitz, French, and Guay (2018), who investigate fluency gains during short-term immersion, and Zaykovskaya, Rawal, and De Costa (2017), who present a case study highlighting how learner beliefs and attitudes can impact opportunities for communication during study abroad.

11.7.5  Willingness to Communicate In previous sections we have discussed contact with the L2 community and the advantage that some learners take of this opportunity (see for example the discussion on language use by Piske, MacKay, & Flege (2001), as well as the discussion on the value of output earlier in this chapter). As LeVelle and Levis (2014) note, “social involvement may be an important strategy for those who want to improve their pronunciation” (p. 103). This builds on the argument by MacIntyre et al. (1998), who define this construct as “a readiness to enter into discourse at a particular time with a specific person or persons, using a L2” (p. 547). As Moyer (2014) points out, younger learners are less constrained and are more willing to engage with speakers of the target language than are older learners, who, she speculates, may be more self-conscious and less willing to take risks. In work by Winke and Gass (2018b) on returning study abroad students, students expressed this concept directly. Students reported that they “had a ton more confidence”, and, as a result, they were “more willing to speak in class” (p. 538). More willingness to speak in class is likely to lead to more output, so it is not unexpected that recent publications, such as Khajavy, MacIntyre, and Barabadi (2017) and Peng, Zhang, and Chen (2017), have emphasized how willingness to communicate may be or can be pedagogically addressed. We have described many individual characteristics of language learners. Over the years, many have been seen individually as having a direct relationship to language learning, but as we have shown in this section, there is a complex relationship between many (if not all) as predictors of global aspects of oral production and/or specific areas of production related to pronunciation and/or fluency.

11.8  Pedagogical Intervention A final, important consideration is the role of the classroom in developing L2 speaking. To begin, explicit pronunciation instruction is generally seen as effective (e.g., Lee, Jang, & Plonsky, 2015; Saito, 2012). Previous pronunciation instruction studies have targeted both global measures (accentedness, comprehensibility, intelligibility) as well as specific linguistic measures (e.g., segmental accuracy, word stress accuracy, speech rhythm). While both empirical- (Lee, Jang, & Plonsky, 2015) and classroom- (Foote,

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

273

Holtby, & Derwing, 2011; Foote et al., 2016; Hardison, 2014a) orientated studies have indicated a pronunciation instruction emphasis on segments (i.e., individual sounds), studies emphasizing understandable speech have argued for a shared-emphasis on segmental and suprasegmental (word stress, intonation, rhythm) features (e.g., Crowther et al., 2015; Trofimovich & Isaacs, 2012). Several example intervention studies of note include Derwing, Munro, and Wiebe (1998) and Nagle (2018). Focusing on L2 English, Derwing, Munro, and Wiebe compared the effects of two treatment approaches (segmental, global) to that of a control group on a sentence read-aloud task and a narrative task (target language was English). The segmental group spent twenty minutes a day on the production of individual sounds, while the global group received twenty minutes on “features such as speaking rate, intonation, rhythm, projection, word stress, and sentence stress” (p. 399). Each group’s performance on the three tasks (before and after treatment) was then rated by forty-eight native-English speakers for accentedness and comprehensibility on nine-point Likert scales. While all three groups showed improvement in accentedness (i.e., perceived nativelikeness) during the sentence read-aloud task, only the two treatment groups showed improvement in comprehensibility (i.e., ease of understanding). More importantly, when it came to the narrative task, which required the participants to describe a picture narrative, only the global group showed improvement from pre- to post-treatment, and only for comprehensibility. Focused on L2 Spanish learners in a communicative-based university-level classroom in the US, Nagle (2018) tracked accentedness and comprehensibility over a year-long period, with an additional predictor variable being motivation. Though classroom instructors indicated limited attention to pronunciation in their lesson, learners demonstrated general improvement in their accentedness and comprehensibility. Interestingly, Nagle found no relationship between motivation measures (specifically Dörnyei’s [2009] L2 Motivational Self System) and increases in comprehensibility, although learners who tended to demonstrate increased effort throughout the year did increase in perceived accentedness. While such findings would provide support for communicative language teaching in regard to increasing the comprehensibility of learners’ speech, Nagle cautions that further research on the relationship between L2 speaking skills and motivation is needed.

11.9 Conclusion At the outset we quoted Fulcher (2014), who defined L2 speaking as “the verbal use of language to communicate with others” (p. 23). Throughout this chapter we have highlighted a range of considerations when attending to

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

274

D ustin C rowt h er and S usan M . G ass

L2 speaking development, including a review of phonology-, fluency-, and sociolinguistic-based variables. What is clear is that the difficulty in defining L2 speaking as a construct is in line with the complex nature of the variables that inform L2 speaking development. A consistent theme, however, is that actual L2 production (i.e., output) appears vital to speaking development. Beyond the importance of the Interaction Approach, evidence of this exists when we consider our discussion of individual differences such as age, ethnic group affiliation, study abroad, and willingness to communicate (and even pedagogy if we take Nagle, 2018, into account). As complex a relationship as the variables associated with L2 speaking development may have, ultimately it would appear that the basis for such development is in the opportunity for continued language production. As such, this may also help to explain the lower ACTFL performance highlighted at the outset of our chapter, where for all learners, studying in a foreign language context was the norm, even though some may have had some limited experience abroad. Moving forward, we suggest several necessary empirical considerations in addressing L2 speaking development. Much research highlighted above places English as the target of acquisition. While this is understandable due to the significant international position that English now holds (e.g., Low & Pakir, 2018), we must also recognize the increased presence of languages such as Chinese (e.g., Chan, 2018; Sharma, 2018) in our global landscape. Recognizing the complex nature of L2 speaking development, understanding how this development may differ across target languages is key. Directly related to non-English target languages is the need to consider non-L1 English speaking learners. For example, while highly informative, the ACTFL data presented at the outset of this chapter is primarily focused on L1 English learners of foreign languages. Such findings do not allow us to distinguish between challenges specific to this population versus those from different L1 populations (Long & Geeslin, 2017). Finally, we echo Derwing and Munro’s (2013) call for more longitudinal research, focused on not only common measures of L2 pronunciation, but also fluency, communicative competence, and pedagogical considerations as well.

References Abrahamsson, N., & Hyltenstam, K. (2009). Age of onset and nativelikeness in a second language: Listener perception versus linguistic scrutiny. Language Learning, 59, 249–306. ACTFL. (2012). ACTFL proficiency guidelines. Alexandria, VA: ACTFL. Retrieved from https://www.actfl.org/publications/guidelines-and-manuals/actflproficiency-guidelines-2012.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

275

American Academy of Arts and Sciences (2017). America’s languages: Investing in language education for the 21st century. Cambridge, MA: American Academy of Arts and Sciences. Retrieved from https://www.amacad.org/content/ Research/researchproject.aspx?i=21896. Aoyama, K., Flege, J. E., Guion, S. G., Akahane-Yamada, R., & Yamada, T. (2004). Perceived phonetic dissimilarity and L2 speech learning: The case of Japanese /r/ and English /l/ and /r/. Journal of Phonetics, 32(2), 233–250. Baker, W. (2015). Culture and identity through English as a lingua franca. Berlin: De Gruyter Mouton. Bergeron, A., & Trofimovich, P. (2017). Linguistic dimensions of accentedness and comprehensibility: Exploring task and listener effects in second language French. Foreign Language Annals, 50(3), 547–566. Best, C. T. (1993). Emergence of language-specific constraints in perception of non-native speech: A window on early phonological development. In B. de Boysson-Bardies, S. de Schonen, P. W. Jusczyk, P. McNeilage, & J. Morton (eds.), Developmental neurocognition: Speech and face processing in the first year of life (pp. 289–304). Dordrecht: Kluwer Academic. Best, C. T. (1994). The emergence of native-language phonological influences in infants: A Perceptual Assimilation Model. In J. C. Goodman & H. C. Nusbaum (eds.), The development of speech perception: The transition from speech sounds to spoken words (pp. 167–224). Cambridge, MA: MIT Press. Best, C. T., & Tyler, M. D. (2007). Nonnative and second-language speech perception: Commonalities and complementarities. In O.-S. Bohn & M. J. Munro (eds.), Language experience in second language speech learning (pp. 13–34). Amsterdam: John Benjamins. Byram, M. (1997). Teaching and assessing intercultural communicative competence. Clevedon: Multilingual Matters. Canale, M. (1983). From communicative competence to communicative language pedagogy. In J. Richards & R. Schmidt (eds.), Language and communication (pp. 2–27). London: Longman. Canale, M., & Swain, M. (1980). Theoretical basis of communicative approaches to second language teaching and testing. Applied Linguistics, 1, 1–47. Caspers, J. (2010). The influence of erroneous stress position and segmental errors on intelligibility, comprehensibility and foreign accent in Dutch as a second language. Linguistics in the Netherlands, 27(1), 17–29. Chambers, F. (1997). What do we mean by fluency? System, 25, 535–544. Chan, J. Y. H. (2018). Attitudes and identities in learning English and Chinese as a lingua franca: A bilingual learners’ perspective. Journal of Multilingual and Multicultural Development, 39(9), 759–775. Crowther, D., Trofimovich, P., Isaacs, T., & Saito, K. (2015). Does a speaking task affect second language comprehensibility? The Modern Language Journal, 99, 80–95.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

276

D ustin C rowt h er and S usan M . G ass

Davidson, D. E. (2010). Study abroad: When, how long, and with what results? New data from the Russian front. Foreign Language Annals, 43, 6–26. Davidson, D. E., & Shaw, J. (2019). A cross-linguistic and cross-skill perspective on L2 development in study abroad. In P. Winke and S. Gass (eds.), Foreign language proficiency in higher education (pp. 217–242). Cham: Springer. De Costa, P. I. (2012). Constructing SLA differently: The value of ELF and language ideology in an ASEAN case study. International Journal of Applied Linguistics, 22, 205–224. De Costa, P. I. (2014). Cosmopolitanism and English as a lingua franca: Learning English in a Singapore school. Research in the Teaching of English, 49, 9–30. DeKeyser, R. (2000). The robustness of critical period effects in second language acquisition. Studies in Second Language Acquisition, 22, 499–533. Dervin, F., & Risager, K. (2015). Researching identity and interculturality. New York: Routledge. Derwing, T. M., & Munro, M. J. (1997). Accent, intelligibility, and comprehensibility. Studies in Second Language Acquisition, 20(1), 1–16. Derwing, T. M., & Munro, M. J. (2013). The development of L2 oral language skills in two L1 groups: A 7-year study. Language Learning, 63, 163–185. Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidencebased perspectives for L2 teaching and research. Philadelphia, PA: John Benjamins. Derwing, T. M., Munro, M. J., & Thomson, R. I. (2007). A longitudinal study of ESL learners’ fluency and comprehensibility development. Applied Linguistics, 29, 359–380. Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favor of a broad framework for pronunciation instruction. Language Learning, 48, 393–410. Derwing, T. M., Munro, M. J., Thomson, R. I., & Rossiter, M. J. (2009). The relationship between L1 fluency and L2 fluency development. Studies in Second Language Acquisition, 31(4), 533–557. Derwing, T. M., Thomson, R. I., & Munro, M. J. (2006). English pronunciation and fluency development in Mandarin and Slavic speakers. System, 34, 183–193. Dörnyei, Z. (2009). The L2 motivational self system. In Z. Dörnyei & E. Ushioda (eds.), Motivation, language identity and the L2 self (pp. 9–42). Toronto: Multilingual Matters. Duff, P. A. (2014). Communicative language teaching. In M. Celce-Murcia, D. M. Brinton, & M. A. Snow (eds.), Teaching English as a second or foreign language (4th edn., pp. 15–30). Boston, MA: National Geographic Learning/ Heinle. Eckman, F. (2004). From phonemic differences to constraint rankings: Research on second language phonology. Studies in Second Language Acquisition, 26(4), 513–549.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

277

Flege, J. E. (1995). Second language speech learning: Theory, findings, and problems. In W. Strange (ed.), Speech perception and linguistic experience: Issues in cross-language research (pp. 233–276). Timonium, MD: York Press. Flege, J. E. (1999). Age of learning and second language speech. In D. Birdsong (ed.), Second language acquisition and the Critical Period Hypothesis (pp. 101–131). Mahwah, NJ: Lawrence Erlbaum Associates. Foley, C., & Flynn, S. (2013). The role of native language. In J. Herschenson & M. Young-Scholten (eds.), The Cambridge handbook of second language acquisition (pp. 97–113). Cambridge: Cambridge University Press. Foote, J. A., Holtby, A. K., & Derwing, T. M. (2011). Survey of the teaching of pronunciation in adult ESL programs in Canada, 2010. TESL Canada Journal, 29(1), 1–22. Foote, J. A., Trofimovich, P., Collins, L., & Urzúa, F. S. (2016). Pronunciation teaching practices in communicative second language classes. The Language Learning Journal, 44(2), 181–196. Freed, B., Segalowitz, N., & Dewey, D. (2004). Context of learning and second language fluency in French: Comparing regular classroom, study abroad, and intensive domestic immersion programs. Studies in Second Language Acquisition, 26, 275–301. Fulcher, G. (2014). Testing second language speaking. New York: Routledge. Gass, S. (1984). Development of speech perception and speech production abilities in adult second language learners. Applied Psycholinguistics, 5(1), 51–74. Gass, S., & Mackey, A. (2006). Input, interaction and output: An overview. AILA Review, 19, 3–17. Gass, S., & Mackey, A. (2007). Input, interaction and output in second language acquisition. In J. Williams and B. VanPatten (eds.), Theories in second language acquisition (pp. 175–199). Mahwah, NJ: Lawrence Erlbaum Associates. Gatbonton, E., & Segalowitz, N. (2005). Rethinking communicative language teaching: A focus on access to fluency. Canadian Modern Language Review, 61(3), 325–353. Gatbonton, E., & Trofimovich, P. (2008). The ethnic group affiliation and L2 proficiency link: Empirical evidence. Language Awareness, 17(3), 229–248. Gatbonton, E., Trofimovich, P., & Magid, M. (2005). Learners’ ethnic group affiliation and L2 pronunciation accuracy: A sociolinguistic investigation. TESOL Quarterly, 39(3), 489–511. Gatbonton, E., Trofimovich, P., & Segalowitz, N. (2011). Ethnic group affiliation and patterns of development of a phonological variable. The Modern Language Journal, 95(2), 188–204. Gregg, K. (1996). The logical and developmental problems of second language acquisition. In W. C. Ritchie & T. K. Bhatia (eds). Handbook of second language acquisition (pp. 49–81). San Diego, CA: Academic Press.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

278

D ustin C rowt h er and S usan M . G ass

Hardison, D. M. (2005). Second-language spoken word identification: Effects of perceptual training, visual cues, and phonetic environment. Applied Psycholinguistics, 26(4), 579–596. Hardison, D. M. (2014a). Phonological literacy in L2 learning and teaching. In J. M. Levis & A. Moyer (eds.), Social dynamics in second language accent (pp. 195–218). Boston, MA: Walter de Gruyter. Hardison, D. M. (2014b). Changes in second-language learners’ oral skills and socio-affective profiles following study abroad: A mixed-methods approach. Canadian Modern Language Review, 70(4), 415–444. Hymes, D. (1971). On communicative competence. Philadelphia, PA: University of Pennsylvania Press. Hymes, D. (1972). On communicative competence. In J. B. Pride & J. Holmes (eds.), Sociolinguistics (pp. 269–293). Harmondsworth: Penguin. Isaacs, T., & Trofimovich, P. (2012). Deconstructing comprehensibility: Identifying the linguistic influences on listeners’ L2 comprehensibility ratings. Studies in Second Language Acquisition, 34, 475–505. Isbell, D., Park, O.-S., & Lee, K. (2019). Learning Korean pronunciation: Effects of instruction, proficiency, and L1. Journal of Second Language Pronunciation, 5(1), 13–48. Khajavy, G. H., MacIntyre, P. D., & Barabadi, E. (2017). Role of the emotions and classroom environment in willingness to communicate: Applying doubly latent multilevel analysis in second language acquisition research. Studies in Second Language Acquisition, 40(3), 605–624. Kuhl, P. K. (2004). Early language acquisition: Cracking the speech code. Nature Reviews Neuroscience, 5, 831–843. Kumaravadivelu, B. (2008). Cultural globalization and language education. New Haven, CT: Yale University Press. Lee, J., Jang, J., & Plonsky, L. (2015). The effectiveness of second language pronunciation instruction: A meta-analysis. Applied Linguistics, 36, 345–366. Lennon, P. (1990). Investigating fluency in EFL: A quantitative approach. Language learning, 40(3), 387–417. Leonard, K. R., & Shea, C. E. (2017). L2 speaking development during study abroad: Fluency, accuracy, complexity, and underlying cognitive factors. The Modern Language Journal, 101(1), 179–193. LeVelle, J., & Levis, J. (2014). Understanding the impact of social factors in L2 pronunciation: Insights from learners. In J. Levis & A. Moyer (eds.), Social dynamics in second language accent (pp. 97–118). Boston, MA: DeGruyter Mouton. Levis, J. M. (2005). Changing contexts and shifting paradigms in pronunciation teaching. TESOL Quarterly, 39, 369–377. Loewen, S. (2003). Variation in the frequency and characteristics of incidental focus on form. Language Teaching Research, 7(3), 315–345.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

279

Long, A. Y., & Geeslin, K. (2018). Spanish second language acquisition across the globe: What future research on non-English-speaking learners will tell us, Hispania, 100(5), 205–210. Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. Ritchie & T. Bhatia (eds.), Handbook of second language acquisition (pp. 413–468). San Diego, CA: Academic Press. Low, E. L., & Pakir, A. (2018) World Englishes: Rethinking paradigms. New York: Routledge. MacIntyre, J., Dörnyei, Z., Clément, R., & Noels, K. (1998). Conceptualizing willingness to communicate in a L2: A situational model of L2 confidence and affiliation. The Modern Language Journal, 82, 545–562. Mackey, A., Abbuhl, R., & Gass, S. (2012). Interactionist approach. In S. Gass & A. Mackey (eds.), Handbook of second language acquisition (pp. 7–23). New York: Routledge. Mackey, A., Gass, S., & McDonough, K. (2000). How do learners perceive interactional feedback? Studies in Second Language Acquisition, 22(4), 417–497. Magnan, S. S., & Back, M. (2007). Social interaction and linguistic gain during study abroad. Foreign Language Annals, 40(1), 43–61. Moyer, A. (2014). What’s age got to do with it? Accounting for individual factors in second language accent. Studies in Second Language Learning and Teaching, 4, 443–464. Muñoz, C. (2014). Contrasting effects of starting age and input on the oral performance of foreign language learners. Applied Linguistics, 35(4), 463–482. Muñoz, C., & Llanes, Á. (2014). Study abroad and changes in degree of foreign accent in children and adults. The Modern Language Journal, 98(1), 432–449. Munro, M. J. (2018). How well can we predict second language learners’ pronunciation difficulties? CATESOL Journal, 30(1), 267–281. Munro, M. J., & Derwing, T. M. (1999). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 49(S1), 285–310. Munro, M. J., Derwing, T. M., & Thomson, R. (2015). Setting segmental priorities for English learners: Evidence from a longitudinal study. International Review of Applied Linguistics in Language Teaching (IRAL), 53(1), 39–60. Nagle, C. (2018). Motivation, comprehensibility, and accentedness in L2 Spanish: Investigating motivation as a time-varying predictor of pronunciation development. The Modern Language Journal, 102(1), 199–217. Nation, P. (1989). Improving speaking fluency. System, 17(3), 377–384. O’Brien, M. G. (2014). L2 learners’ assessment of accentedness, fluency, and comprehensibility of native and nonnative German speech. Language Learning, 64(4), 715–748.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

280

D ustin C rowt h er and S usan M . G ass

Okuno, T., & Hardison, D. M. (2016). Perception–production link in L2 Japanese vowel duration: Training with technology. Language Learning & Technology, 20(2), 61–80. Peng, J. E., Zhang, L., & Chen, Y. (2017). The mediation of multimodal affordances on willingness to communicate in the English as a foreign language classroom. TESOL Quarterly, 51(2), 302–331. Piske, T., MacKay, I. R., & Flege, J. E. (2001). Factors affecting degree of foreign accent in an L2: A review. Journal of phonetics, 29(2), 191–215. Risager, K. (2007). Language and culture pedagogy: From a national to transnational paradigm. Clevedon: Multilingual Matters. Rubio, F., Hacking, J., Soneson, D., Winke, P., & Gass, S. (2018). Outcomes and observed trends from the Flagship Proficiency Initiative Paper presented at the 2018 Language Flagship Meeting, May, Philadelphia, PA. Saito, K. (2012). Effects of instruction on L2 pronunciation development: A synthesis of 15 quasi-experimental intervention studies. TESOL Quarterly, 46, 842–854. Saito, K., & Akiyama, Y. (2017). Linguistic correlates of comprehensibility in second language Japanese speech. Journal of Second Language Pronunciation, 3(2), 199–217. Saito, K., & Lyster, R. (2012). Effects of form-focused instruction and corrective feedback on L2 pronunciation development of /ɹ/ by Japanese learners of English. Language Learning, 62(2), 595–633. Saito, K., Ilkan, M., Magne, V., Tran, M., & Suzuki, S. (2018) Acoustic characteristics and learner profiles of low, mid and high-level second language fluency. Applied Psycholinguistics, 39(3), 593–617. Sakai, M., & Moorman, C. (2018). Can perception training improve the production of second language phonemes? A meta-analytic review of 25 years of perception training research. Applied Psycholinguistics, 39(1), 187–224. Sanz, C., & Morales-Front, A. (eds). (2018). Handbook of study abroad research and practice. New York: Routledge. Segalowitz, N. (2010). Cognitive bases of second language fluency. New York: Routledge. Segalowitz, N. (2016). Second language fluency and its underlying cognitive and social determinants. International Review of Applied Linguistics in Language Teaching (IRAL), 54(2), 79–95. Segalowitz, N., & Freed, B. F. (2004). Context, contact, and cognition in oral fluency acquisition: Learning Spanish in at home and study abroad contexts. Studies in Second Language Acquisition, 26(2), 173–199. Segalowitz, N., French, L., & Guay, J. D. (2018). What features best characterize adult second language utterance fluency and what do they reveal about fluency gains in short-term immersion? Canadian Journal of Applied Linguistics/Revue canadienne de linguistique appliquée, 20(2), 90–116. Sharma, B. K. (2018). Chinese as a global language: Negotiating ideologies and identities. Global Chinese, 4(1), 1–10.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

Speaking

281

Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (eds.), Input in second language acquisition (pp. 235–256). New York: Newbury House. Swain, M. (1995). Three functions of output in second language learning. In G. Cook & B. Seidlhofer (eds.), Principle and practice in applied linguistics: Studies in honor H. G. Widdowson (pp. 125–144). Oxford: Oxford University Press. Swain, M. (2005). The Output Hypothesis: Theory and research. In E. Hinkle (ed.), Handbook of research in second language learning and teaching (pp. 471–484). New York: Routledge. Tao, L., & Taft, M. (2017). Effects of early home language environment on perception and production of speech. Bilingualism: Language and Cognition, 20(5), 1030–1044. Thomson, R. I. (2011). Computer assisted pronunciation training: Targeting second language vowel perception improves pronunciation. CALICO Journal, 28, 744–765. Thomson, R. I. (2012). Improving L2 listeners’ perception of English vowels: A computer-mediated approach. Language Learning, 62, 1231–1258. Thomson, R. I. (2018). Measurement of accentedness, intelligibility, and comprehensibility. In O. Kang & A. Ginther (eds.), Assessment in second language pronunciation (pp. 11–29). New York: Routledge. Thomson, R. I., Nearey, T. M., & Derwing, T. M. (2009). A modified statistical pattern recognition approach to measuring the crosslinguistic similarity of Mandarin and English vowels. The Journal of the Acoustical Society of America, 126(3), 1447–1460. Trofimovich, P., & Isaacs, T. (2012). Disentangling accent from comprehensibility. Bilingualism: Language and Cognition, 15, 905–916. Varonis, E. M., & Gass, S. M. (1982). The comprehensibility of nonnative speech. Studies in Second Language Acquisition, 4, 114–136. Werker, J. F., & Tees, R. C. (2005). Speech perception as a window for understanding plasticity and commitment in language systems of the brain. Developmental Psychobiology, 46(3), 233–251. White, L. (1991). Adverb placement in second language acquisition: Some effects of positive and negative evidence in the classroom. Interlanguage Studies Bulletin, 7(2), 133–161. Winke, P., & Gass, S. (2018a). Individual differences in advanced proficiency. In P. Malovrh and A. Benati (eds). Handbook of Advanced Language Proficiency (pp. 157–178). New York: Routledge. Winke, P., & Gass, S. (2018b). When some study abroad: How returning students realign with the curriculum and impact learning. In C. Sanz & A. Morales-Front (eds.), Handbook of study abroad research and practice (pp. 527–543). New York: Routledge. Winke, P., Gass, S., & Heidrich, E. (2019). Modern-day foreign language majors: Their goals, attainment, and fit with a 21st century curriculum.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

282

D ustin C rowt h er and S usan M . G ass

In P. Winke & S. Gass (eds.), Foreign language proficiency in higher education (pp. 93–113). Cham: Springer. Yan, X., & Ginther, A. (2018). Listeners and raters: Similarities and differences in evaluation of accented speech. In O. Kang & A. Ginther (eds.), Assessment in second language pronunciation (pp. 67–88). New York: Routledge. Zaykovskaya, I., Rawal, H., & De Costa, P. I. (2017). Learner beliefs for successful study abroad experience: A case study. System, 71, 113–121. Zyzik, E., & Polio, C. (2008). Incidental focus on form in university Spanish literature courses. The Modern Language Journal, 92(1), 53–70.

Downloaded from https://www.cambridge.org/core. University of Groningen, on 08 Aug 2019 at 02:48:17, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.012

12 Second Language Listening: Current Ideas, Current Issues John Field 12.1 Introduction This chapter starts (12.2) by mentioning the drawbacks of the approach conventionally adopted in second language (L2) listening instruction— in particular, its focus on the products of listening rather than the processes that contribute to it. It then offers an overview of our present understanding of what those processes are, drawing upon research findings in psycholinguistics, phonetics, and applied linguistics. section 12.3 examines what constitutes proficient listening and how the performance of an L2 listener diverges from it; and section 12.4 considers the perceptual problems caused by the nature of spoken input. Subsequent sections then cover various areas of research in L2 listening. Section 12.5 provides a brief summary of topics that have been of interest to researchers over the years; and section 12.6 reviews the large body of research into listening strategies. Section 12.7 then covers a number of interesting issues that have come to the fore in recent studies: multimodality; levels of listening vocabulary; cross-language phoneme perception; the use of a variety of accents; the validity of playing a recording twice; text authenticity; and listening anxiety. A final section (12.8) identifies one or two recurring themes that have arisen and considers how listening instruction is likely to develop in the future.

12.2  The Comprehension Approach The teaching of L2 listening for its own sake, rather than as an adjunct to the teaching of speaking, grammar, or pronunciation, began as recently as the late 1960s (Rivers, 1966; Morley, 2001). The approach adopted, based

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

284

John Field

on answering comprehension questions, had its roots in the teaching and testing of reading. Later, with the advent of communicative approaches to language teaching in the late 1970s, preferences shifted away from formal questions and towards setting tasks (Ur, 1984, Chapter 5): completing simple grids, filling in simulated forms, or marking maps and plans. This had the advantage of reducing the reading and writing load in what is supposedly a listening exercise and aligning the exercise more closely to what takes place in real-world encounters. Nevertheless, there have long been methodological concerns about the Comprehension Approach. It can be viewed as teacher-centric (the instructor controlling both recording and questions) and as associating listening too closely with reading, despite obvious differences of input. It has been criticized (Brown, 1986) for testing the skill rather than teaching it, though some kind of probing would seem to be unavoidable, given that listening is an internalized process taking place in the mind of the listener. Perhaps the most important concern is that the approach represents “comprehension” narrowly in terms of correct answers to questions but does not provide a means for analysing and repairing what causes listening to go wrong. It focuses on the products of listening but does not shed light on the processes that give rise to those products. It provides practice but does not necessarily advance the listening competence of the learners. Brown (1986) argued that what practitioners need is a set of techniques that enable them to approach listening performance diagnostically. In a similar vein, Field (1998, 2008a) suggested that it is possible to retain the framework of the existing Comprehension Approach but to shift its focus. Instead of simply classifying responses as right or wrong, the teacher might ask learners to justify their answers, thus providing insights into what they think they have heard and where they have gone astray. The information can then lead to small-scale remedial exercises that practise the problems that have arisen. The approach was extended by Harding, Alderson, and Brunfaut (2015), who propose diagnostic tests that enable an informed instructor to identify individual listening difficulties so as to provide feedback on them. The function of teacher-as-advisor represents an attractive alternative to the conventional teacher-centred model. But a diagnostic approach cannot succeed unless practitioners have better information about (a) the precise nature of the skill that they are aiming to inculcate in learners; (b) the nature of the signal that listeners have to deal with and where problems are likely to arise in a second language context; and (c) how L2 listeners handle breakdowns of comprehension. Unsurprisingly, it is on these areas that much recent discussion of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

285

second language listening has focused. They form the substance of the three sections that follow.

12.3  Theoretical Background 12.3.1  Modelling Listening With a view to providing clearer goals for instructors, attention has turned to the demands that listening imposes on the listener. This originated in the late 1970s, with the notion of skills-based instruction. Established approaches to the teaching of all four language skills were challenged on the grounds that activities needed to reflect the cognitive realities of putting the skills to use. The view that emerged was componential: based on the assumption that the skills could be divided into constituent parts and that those processes could then be practised individually. This gave rise to an eclectic list of examples of what became known as “sub-skills” (Munby, 1978). The concept was applied creatively to the teaching of reading, where it led to a rethinking of priorities and task types (Grellet, 1981; Nuttall, 1996). However, it did not have the same impact on listening instruction, despite a promising initiative by Richards (1983), who provided detailed taxonomies of likely sub-skills in both general and academic contexts. Material writers were slow to devise new exercise types to practise these components of listening; it also proved difficult to grade them in relation to different proficiency levels. But perhaps the most important concern was that sub-skills were identified intuitively rather than on hard research evidence of how individuals actually listen. The work of phoneticians, speech scientists, and psycholinguists had already begun to shed light on the complex demands associated with turning a speech signal into words, clauses, and ideas; but these findings were slow to find their way into second language theory and practice. Responding to the need, Rost (1992, 2002) offered a detailed account of the processes involved in listening, citing current theory in psychology and discourse analysis. Also needed, however, was a model of how one component supports another. One such was found in a proposal by Anderson (2000), much quoted in L2 listening contexts. It represents listening as falling into three phases: perceptual processing (converting sounds into words); parsing (converting words into syntactic patterns); and, rather misleadingly, “utilization” (converting syntactic patterns into units of meaning). A more detailed model has since been proposed by two leading first language (L1) listening researchers, Cutler and Clifton (1999). The simplified

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

286

John Field

Input decoding

Lexical search

Parsing

Meaning construction

Discourse construction Figure 12.1.  Simplified cognitive model of the listening process (Cutler & Clifton, 1999; Field, 2008a).1

version in Figure 12.1 offers a framework for understanding the various types of operation2 which L2 learners have to master. Note that a single meaning-level phase in Cutler and Clifton (1999) is here divided into two, to accommodate the important distinction made by discourse analysts (e.g., Brown & Yule, 1983) between a meaning-based level of operation and a discourse-based one. The downward arrows in the figure indicate a “bottom-up” progression from smaller to larger units of language and from linguistic information based upon strings of words to more abstract representations based upon meaning. However, the upward arrows remind us that this is not a purely linear progression. Listening is an interactive operation which draws upon multiple cues as to speech content; larger units sometimes operate in a top-down manner to assist the recognition of smaller ones. For example, knowledge of the spoken form of a whole word might influence what we think we hear at phoneme level. The relevance of a model of this kind to an L2 instructor or test designer is that it provides an empirically supported framework for the various levels of performance that a listener has to command. Each phase consists of a set of coordinated processes whose purposes can be specified. For example, lexical search necessarily entails: identifying possible word matches for a string of sounds; trading word matches against each other until a correct one has been chosen; accessing the sense or range of senses represented by that word; accessing information about the word’s syntactic properties and likely collocations; and checking the word in relation to surrounding co-text. These processes provide the basis for descriptors  The model proposed by Lynch (2009) draws upon Levelt (1993) but essentially features the same stages.

1

 The term “level” is generally used in the psycholinguistics literature. Here “operation” or “phase” are widely used to

2

avoid confusion with levels of proficiency. An operation is subserved by multiple processes.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

287

outlining what a listener is capable of at different levels of proficiency and the points at which understanding may break down. They also serve to identify a range of possible targets that can be practised individually by means of small-scale exercises (Field, 2008a). The first three operations are perceptual. Listeners have to decode the input—i.e., they have to match the acoustic sensations reaching their ears to the sounds of the target language (Fowler & Magnuson, 2012). Next, they have to identify oral word forms from clusters of these sounds. The currently most favoured account of lexical recognition (McQueen, 2007) is that possible word matches compete in the mind of the listener until one wins out as the closest fit to what is in the signal. Once identified, the form unlocks an entry in the listener’s mental vocabulary store and enables him/ her to access word meanings (Aitchison, 2008). Of course, it is not always as simple as that. A word form may unlock multiple senses (compare the word RUN in run a mile, run a business, run a bath, run an idea past somebody), and allowance has to be made for homophones like /raɪt/. Several possible interpretations of a word form may have to be held in reserve until the evolving co-text makes clear which one the speaker intended. At a third stage, known as parsing (Gompel & Pickering, 2007), a listener has to trace a syntactic pattern in a group of identified words. This entails building up a string of words one by one as they occur and then holding them in the mind until such time as the clause is complete. There are likely to be limitations on how many words an early-stage L2 listener can retain in this way because of the attention that needs to be allocated to recognizing newly incoming ones. A useful boundary signal may occur when the speaker pauses briefly to plan the subsequent clause. However, the identification of a syntactic pattern mainly takes place while the clause is still being uttered, using knowledge of frequent target-language structures. Some languages (e.g., English) adhere quite rigidly to a Subject–Verb–Object sequence, while with others (e.g., Italian), flexibility of word order has to be allowed (Bates & MacWhinney, 1989). Wrong assumptions can lead to a garden-path effect (Tanenhaus & Trueswell, 1995) where initial expectations have to be changed. Evidence suggests that, after forming an incorrect hypothesis of this kind, L2 listeners are more reluctant than L1 listeners to abandon it (Field, 2008b). All these perceptual operations are complicated by the time-constrained nature of listening, which differentiates it markedly from reading. The signal in listening is transitory, and has to be processed online, as it is received. Short of getting a speaker to repeat, a listener cannot check back that a word or a syntactic structure has been correctly identified. What is more, the pace at which listening takes place is determined by the speaker, whereas a reader can modify reading speed according to the

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

288

John Field

difficulty of the content or the goals of the task. The result is that listening emerges as a much more tentative activity than reading, with even L1 listeners sometimes failing to identify a word until up to three words later (Grosjean, 1985). This is the kind of uncertainty that undermines the confidence of an L2 listener and creates listening anxiety. The output from the perceptual operations described is a piece of information that is no longer in the form of language but remains neutral in terms of the interpretations that the listener might come to place on it. The next two phases are therefore best described as conceptual. They relate the information to its immediate context and to the wider speech event in which it occurs. Firstly, the listener has to draw on external information (world knowledge, knowledge of the context, knowledge of the speaker, etc.) in order to gain a clearer idea of the relevance of the utterance that has just been heard. This may include (Brown, 1995): • recognizing the speaker’s attitude or intentions; • drawing inferences about a point that the speaker has implied but not explicitly expressed; • linking anaphors such as pronouns to the items or ideas to which they refer; • tracing a logical link between a new utterance and the one that preceded it. Finally, information derived from the current utterance has to be connected to what has been said so far, in order to construct a discourse representation of the overall direction of the conversation or talk. This might include (Gernsbacher, 1990): • • • •

distinguishing major points from minor ones; identifying the speaker’s main point or overall purpose; tracing logical links between different points that have been made; monitoring understanding to ensure that new points are consistent with what has been said before.

12.3.2  The L2 Listener Clearly, the procedure that has been outlined is applied less than perfectly by many L2 listeners. Instructors tend to assume that the main cause is gaps in knowledge, i.e., uncertain phonological values, inadequate vocabulary, or incomplete syntactic information. Problems at the conceptual level might reflect an incomplete understanding of pragmatics or cultural norms, which prevents the listener from interpreting the speaker’s intentions. While linguistic and cultural knowledge are necessary components of skilled L2 listening, they are by no means sufficient ones. Equally if

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

289

not more important in current listener-based accounts is the part played by expertise. Expertise, as classically conceived (Ericsson & Smith, 1991; Feltovich, Prietula, & Ericsson, 2006; Johnson, 2005), entails the achievement of a high degree of automaticity when performing an operation, whether driving a car, opening a door, or making sense of a piece of speech. The way a competent listener maps from incoming speech to a word in their vocabulary store depends critically on how accessible the vocabulary information is and how automatically the connection is made between a group of sounds and a lexical entry. (For discussions of the contribution of automaticity to fluent language performance, see Segalowitz, 2010, pp. 78–104; Segalowitz, 2016.) Processing information automatically rather than in a controlled way results in speed and accuracy of performance, but its most critical contribution is to reduce the attention that an individual needs to give to basic operations. Attentional capacity in human beings is limited (Pashler & Johnson, 1998); so, if fundamental processes are handled automatically, it enables one to allocate more attention to larger-scale issues. The relevance to L2 listening will be obvious. Early stage L2 listeners find that they need to focus heavily on decoding, lexical search, and parsing, with the result that they have little or no capacity left for the types of conceptual operation associated with meaning and discourse construction. For example, they may report words that approximate phonologically to the target but ignore wider co-textual and contextual cues (Hansen & Jensen, 1994; Pemberton, 2003; Field, 2004). In a test situation with time to reflect on answers, they may fall back on general guesses based upon the topic, but such guesses may have little connection with board perceptual evidence. An imaginative, large-scale research project by Tsui and Fullilove (1998) illustrates this well. The researchers classified items in a listening test into two types: those that related to prior world or topic knowledge and those that did not. They discovered that high-scoring candidates performed much better than low-scoring ones on the second type and on questions that were global rather than local. In other words, proficient listeners were able to integrate perceptual and conceptual information in a way that less proficient ones were not. By contrast, the topic-based guesses made by weaker listeners were sometimes erroneous because they were not backed up by an understanding of the input. Through practice and experience (Anderson, 1983), learners achieve a degree of automaticity in handling perceptual processes; only then are they able to commit adequate attention to information at conceptual levels. Meanwhile, they may be heavily reliant upon the use of compensatory strategies (Cohen, 1998) to supply gaps in what they have been capable of decoding. There are implications here for L2 listening programmes.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

290

John Field

a. In the early stages of L2 acquisition, programmes need to focus quite intensively on perceptual processes, with a view to enhancing automaticity through practice (Hulstijn, 2003). They can reasonably require learners to report factual information and the main point of a recording, but should not venture into questions that require interpretation or the reporting of a discourse-level line of argument. b. Programmes should recognize that the early stages of acquiring L2 listening may be frustrating for learners; motivation can be sustained by equipping them with strategies that enable them to compensate for gaps in understanding. See section 12.6. Citing evidence from learner transcriptions, Field (2018) suggests that a breakthrough occurs about Level B1+ in the Common European Framework (CEFR; Council of Europe, 2018), when the majority of words in a recording are correctly identified and there is emerging evidence of automatic processing.

12.4  The Nature of Connected Speech 12.4.1  Variability of Form There was a tendency until recently to downplay the role of perceptual processes in L2 listening—possibly a reaction to early approaches to listening instruction based upon ear-training, minimal pairs, and dictation (Brown, 1990, pp. 144–146). A second factor was the long-standing fallacy that L2 listeners experiencing problems of word recognition could always fall back strategically on “context” to assist them. The argument does not bear scrutiny: without a bare minimum of accurately-decoded text, a listener has no context or co-text available upon which to draw (see Macaro, Graham, & Vanderplank, 2007, p. 171, for a similar comment). It also became recognized that many incorrect answers to comprehension questions originated, not in failures of general understanding, but in failures of recognition at word or clause level. The result was a revived interest in the nature of the speech signal and the challenges it poses (Lynch, 2006). The experience of an L2 listener was distinguished sharply from that of an L2 reader, not least because of the major differences between written and spoken input. There are no consistent gaps between words in listening as there are in reading. Reading possesses a standardized spelling system which generally provides a single form for each word; by contrast, word forms in listening are highly variable. And while words in reading vary only a little in prominence (e.g., through the use of italics), those in listening vary enormously in both emphasis and duration. The comments that follow explore in a summative way the implications of these differences

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

291

of input. They draw upon examples from English; but L2 listening in any language presents similar obstacles to understanding (Sebastián-Gallés, 2008). For extended accounts of the phenomena described, see: Brown (1990); Brown & Kondo-Brown (2006); Field (2008a, Chapter 9); Shockey (2003).

12.4.1.1  The Elusive Phoneme Listening instructors sometimes forget that a phoneme is an abstract unit and one that does not have a neat one-to-one relationship with a particular combination of accoustic cues (Field, 2014). This is partly due to co-articulation: the effect upon a phoneme of those that precede or follow it within the same syllable. In the 1960s, researchers at the Haskins laboratories in the US were unable to find sets of acoustic features that served to uniquely identify certain English consonants (Nygaard & Pisoni, 1995, pp. 64–68). As for vowels, what distinguishes them perceptually is a combination of two frequency formants which differ according to the pitch of the speaker’s voice (Strange, 1999, pp. 153–160). It has been suggested (e.g., Pisoni & Luce, 1987, p. 26) that the phoneme might not itself constitute a unit of analysis within listening.3 Instead, listeners may well operate at the level of the syllable (Mattys & Melhorn, 2005), which is not only a more reliable unit but one whose likelihood of occurrence can be recognized. Alternatively, it may be that listeners map directly from a group of sounds to a whole word (Klatt, 1979).

12.4.1.2  Word Boundaries Speakers hesitate occasionally and pause briefly to plan their next utterance; but there are no systematic gaps between words in listening as there are in writing. It is the listener who has to decide where one word ends and the next begins, a process known as lexical segmentation (Cutler, 2012, pp. 117–189). It seems that, across languages, different cues are used in order to locate where word boundaries are most likely to fall (Cutler, 1997). They include lexical stress. This can provide a failsafe indicator in the case of languages like Finnish, Czech, or Polish, which have fixed stress on initial or final syllables. Or it can provide broad indications in the case of (say) English, where around ninety per cent of content words are either monosyllabic or characterized by an initial stressed syllable (Cutler, 1990). Other cues might be phonotactic constraints (e.g., an illegal cluster of consonants indicating the end of one word and the onset of the next; see McQueen, 1998), units such as the mora in Japanese (Cutler & Otake, 1994) or the presence of familiar prefixes and suffixes. Because segmentation cues are so language-specific, there is evidence that second language learners transfer their native techniques to L2 contexts (Cutler, 2012,  Our ability to identify phonemes in speech may be a product of literacy rather than a precursor (Morais et al., 1979).

3

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

292

John Field

pp. 129–132). However that does not mean that they cannot be acquired, even through simple exposure to the target language (Field, 2001).

12.4.1.3  Word Variation Perhaps the greatest challenge faced by the L2 listener is the variation in the ways in which words are articulated in connected speech. This is partly a question of the personal style of the speaker and the type of speech event: Laver (1994), for example, identifies seven different versions of the word actually, ranging from most casual to most formal. But it is also systemic, reflecting the way in which speakers try to move with minimal effort from one articulatory position to another. The major types of phonetic change appear in the panel below. These are not markers of casual delivery, but are present in all natural connected speech: Resyllabification before a word initial vowel: went in → wen tin Liaison, introducing a consonant between vowels: to (w)eat, the (y)end Assimilation, anticipating the next articulatory position: tem people, hop bath Elision, especially within a complex consonant cluster: next spring → neck spring

A little-noticed elision effect is that word-final inflections in English are often absent. Brown (1990) reports that, even in samples of careful speech (speakers on BBC radio), more than fifty per cent of instances of /t/ and /d/ between consonants were omitted, even when they represented past simple and past participle markers. These variations often result in word-level forms that differ enormously from the citation forms an L2 listener will have encountered in a vocabulary class. Given this, how does a listener, whether in L1 or L2, succeed even in recognizing known words? A widely supported “exemplar” account would have it that a language user stores multiple variants of each word as heard in different voices and different contexts (Goldinger, 1998). Another explanation is that lexical recognition does not reach its final stage until there is sufficient co-text to enable a word to be confidently identified. As Shockey (2003) puts it: “perceivers of casual speech seem quite comfortable with building up an acoustic sketch as the utterance is produced, the details of which are filled in when enough information becomes available” (p. 103).

12.4.1.4  Perceptual Prominence The most reliable words for an L2 listener are those given prominence by means of tonic stress and more precise articulation. However, the listener also pays a price in that the surrounding words become correspondingly compressed (Cauldwell, 2013). Even function words already

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

293

in weak form are reduced in duration, with the result that some of the most frequent words of the language can be among the most difficult to recognize. Field (2008c) reports much greater accuracy in L2 recognition of content words than of functors, partly because listener attention is focused upon the meaning-bearing items in an utterance, but also because of their relative salience. The identification of words within utterances becomes more accurate as L2 learners come to recognize many highly recurrent formulaic chunks of language in connected speech (Wray, 2002)—not simply multi-word items like in front of but also markers (in other words), functional formulae (never mind, if I were you), and syntactic patterns (should have done). This advance appears to occur around CEFR level B1+ (Field, 2018).

12.4.2  The Variable Speaker Another major variable that has to be dealt with is the voice of the speaker. When hearing a speaker for the first time, a listener has to go through a process of adjustment, known as normalization (Johnson, 2008). L1 listeners are capable of making rapid and highly automatic assessments of a voice’s pitch, the speed at which speech is being delivered, and the range of the pitch movements (Pisoni, 1997). This is necessary, as noted earlier, because the first of these enables us to distinguish vowels. The second may affect the recognition of consonants such as /b/ and /w/ (Miller, 1990). Some idea of the important role that normalization plays in speech events can be gathered from our extraordinary capacity to recognize a speaker’s voice when we hear it again, even by telephone. Clearly the process does not come as readily to listeners when the voices in question are using a second language, with its own distinctive pitch ranges and rhythms. In instruction, this validates the convention of a preliminary play of a recording or, at the very least, a short delay of ten to fifteen seconds before any information in the recording is targeted by a question. The need to normalize to voices is also, of course, greater when the input is in audio form, with no visual information to help differentiate speakers. Two voices in a dialogue do not pose a major challenge to an early-level learner, especially if the convention is observed of making one male and one female; but where there are three or more voices, the demands upon the listener are inevitably stepped up.

12.4.3  The Variable Input: Implications for Practice Given the perceptual difficulties experienced by the L2 listener, what steps can be taken to improve learner performance? One solution is to make use of some form of transcription task (Field, 2008a (Chapters

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

294

John Field

9–10))—not the stilted dictation of yore, but an exercise supplementing a comprehension task in which examples of one or more of the features mentioned above are excised from the recording and learners are asked to transcribe them. In more progressive language schools, instructors have begun to develop archives of materials that consist either of excerpts with which previous learners have experienced difficulty or of samples of features such as lexical segmentation, assimilation, elision, or the chunking of words. A valuable precedent has been set by Cauldwell’s (2013) analyses of natural speech; and some materials writers now include post-comprehension exercises of this kind in their coursebooks. A problem with this approach (and indeed with class-based instruction generally) is that different listeners experience different problems, especially so far as perception is concerned. Listening thus lends itself particularly well to self-access work, where learners can listen again and again to any short sections of a text which they find hard to process. This is the direction that much listening instruction seems likely to take in the future. It might be in the form of downloadable listening homework or of practice sessions in a dedicated listening centre. Hulstijn (2003) reports on 123LISTEN, a programme which enables instructors to prepare self-teach listening materials based upon clips. In a groundbreaking computer-based initiative for French universities (http://innovalangues.fr), an autonomous learner will be able to time-mark sections of speech that are problematic, and then access samples of similar phonetic patterns for ear-training practice.

12.5  Issues in Second Language Listening Research There has been more research into second language listening over the past twenty years than is sometimes suggested, and it is hard to do justice to it in the space of a chapter of this length. The sections that follow must, of necessity, be selective. They provide an overview of what has been the largest area of study (the use of listening strategies), and then focus on seven topics which have previously been little represented in reviews such as this—either because they have come to the forefront recently or because they have tended to be neglected by commentators. They are chosen because they seem likely to stimulate discussion within the field in coming years. Before moving on to specifics, it may be useful to provide (Figure 12.2) a general outline of the aspects of L2 listening that have received the attention of researchers over the years. Historically, research questions have been very much concerned with difficulty—especially with the impact of

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

295

Figure 12.2.  Topics favoured in L2 listening research.

materials and tasks on listener performance. What appears in Figure 12.2 is inevitably selective, but it draws upon previous reviews such as Rubin (1994) and Bloomfield et al. (2010), as well as the author’s own validation work on the testing of listening (Field, 2013, 2018). Italics mark topics that will be covered in the sections that follow.

12.6  Listening Strategies4 Of the four language skills, listening appears to be the most dependent upon strategy use because of the input difficulties discussed in section 12.4 and the consequent need for listeners to focus attention upon perceptual processes to the exclusion of wider meaning. Strategy training equips  The strategies considered here are solely compensatory strategies (Cohen’s (1998) strategies of use).

4

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

296

John Field

listeners to extract more information from a listening passage than they would be able on the sole basis of recognized language. The result is to boost self-efficacy (Graham, 2011) at times when the acquisition of listening proficiency may seem a slow and uncertain process. At this point, attention should be drawn to a terminological problem in the way listening strategies are often represented. It is important to make a distinction between: • processes intrinsic to normal listening such as those profiled in section 12.3, which constitute what a language learner needs to acquire for long-term performance; • strategies, expedients used periodically when an individual’s listening competence cannot match the demands of a listening task.5 This might seem pedantic, and, to be sure, there are grey areas between the two. Expert listeners still need to use strategies occasionally (e.g., to identify words in noise), and some operations, like checking understanding, are part of normal listening as well as something that less-experienced listeners are advised to do. Nevertheless, there is an important pedagogical point at stake. Instructors need to be able to distinguish between activities which hopefully, long-term, will become fully integrated into the behaviour of a listener and highly automatic and others which are short-term expedients to overcome an inadequate grasp of the second language. Afflerbach, Pearson, and Paris (2008) make a similar point in relation to reading. There has been much research on listening strategies in recent years. However, as Macaro, Graham, and Vanderplank (2007) acknowledge in their perceptive review of the literature, it is sometimes difficult to draw generalizable conclusions. This they partly attribute to the rather fragmented nature of the field in terms of research goals, criteria applied, and populations studied. One might also add that researchers in this area are unfortunate in lacking a body of empirically based theory upon which to construct a general model; they often have to fall back on intuitive categories not unlike the original “sub-skills”. The ways in which these categories are connected may well be unclear. Early attempts to classify strategies more systematically in terms of their purpose (e.g., Faerch & Kasper, 1983; Rost, 1992; Dörnyei & Scott, 1997) do not seem to have led to any consensus. However, commentators usually observe Cohen’s (1998) distinction between “strategies of learning” and “strategies of use”, and there is widespread use of O’Malley and Chamot’s (1990) classification of three strategy types: metacognitive; cognitive; and social-affective. The difficulty in researching listening strategies also lies in the internalized nature of the skill, which severely restricts the types of research  See Field (2000) for an earlier discussion of this issue.

5

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

297

methodology that can be used. Questionnaires are notoriously open to over-reporting, especially if a respondent views strategy use as the mark of a good learner.6 In addition, the questionnaires that have been designed (including those which provided evidence for Vandergrift et al.’s (2006) MALQ) reflect the problems of reliably and validly quantifying strategy use. What evidence do we have for the psychological reality of a given strategy and how context-dependent is it? How interdependent might some strategies be? Above all, is reported frequency or familiarity of strategy use necessarily an indicator of appropriate and effective use? An alternative research methodology often employed is retrospective verbal report. Here, there are also constraints: the reporting clearly cannot interrupt the listening task, and the time-lapse before it occurs has to be short to avoid memory effects. Despite these obstacles, a number of researchers (e.g., Goh, 1998; Graham, 2006; Field, 2012) have used recall to effect, and elicited detailed evidence of the strategies used by learners. There has been particular research interest in metacognition (extensively reviewed in Vandergrift & Goh, 2012). By contrast, cognitive and socio-affective strategies are under-represented in the literature, which seems unfortunate, given the needs of lower-proficiency learners.7 One reason for the preference may be that metacognitive activities involve intentionality and are therefore more accessible to report. However, a precautionary note needs to be sounded here: because they are more reportable, they are likely to feature disproportionately in any verbal reports or questionnaires aiming to quantify frequency of use. There is a tendency in the metacognitive literature (Teichert, 1996; Herron et al., 1998; Chung, 1999) to focus on one type of metacognitive strategy, namely, “advanced organizer” techniques that improve classroom performance (an example might be reflecting on a topic before listening to the recording). While strategies of this kind may be useful in teaching and testing contexts, they cannot be said to contribute greatly to real-world listening. Two general research questions have been widely investigated. The first seeks evidence of a relationship between level of strategy use and listening success. The cause-and-effect assumptions underlying this research are sometimes unclear. Does evidence of an association suggest that employing listening strategies enables better listening performance across listeners? Or does it indicate that higher-level listeners are more inclined to strategy use? However, this area of study has led to some interesting questions about additional variables that have to be considered when investigating the contribution of strategic competence to listening  See Vogely’s (1995) report of discrepancies between students’ perceptions of what makes a good listener and

6

their actual behaviour.  Notable exceptions are Goh (2000) and Graham (2006).

7

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

298

John Field

outcomes (Macaro, Graham, & Vanderplank, 2007). It has been suggested that researchers may also need to control for: (a) linguistic knowledge; (b) a facility for acquiring languages; and/or (c) L1 listening competence. For a study that focuses on the third of these, see Vandergrift (2006). An aspect of listener background surprisingly little investigated has been the extent to which strategy use varies according to individual temperament or whether it is culturally determined. It may well be that the educational formation of learners from certain parts of the world encourages speculative thinking, while the formation of others restrains them from making guesses until sufficient evidence is available (Braxton, 1999). A second major topic has been the effect of training upon strategy use or upon listening proficiency. Macaro, Graham, and Vanderplank (2007) report largely unclear results. This may partly be because many studies have employed a longitudinal methodology with control groups and preand post-testing—an approach that, in listening research, is vulnerable to variables such as participants’ degree of exposure to L2 sources during the trial period.8 That said, Rost and Ross (1991) demonstrated clear improvements in performance after first identifying the strategies used by higher-proficiency learners and then teaching them to lower-proficiency ones. Vandergrift and Tafaghodtari (2010) report improved post-test scores among lower-proficiency participants following training. Graham (2006) and Goh and Taib (2006) cite unambiguous quotes by individual young learners on the benefits of training. Arguably, much may depend upon the precise form that the instruction takes. O’Malley and Chamot (1990) identify two main types. A “direct” approach begins by raising strategy awareness. Strategies are presented individually and may even be named. They are then modelled and practised in controlled tasks. This has become a format for listening instruction in various parts of the world, particularly the US (Mendelsohn, 1994). A potential problem lies in separating out individual strategies: Vandergrift (2004) comments that “The apparent interconnectedness of strategy use may explain why individual strategies instruction has not been as successful as hoped” (p. 12). In an “indirect” approach, learners engage in conventional listening comprehension tasks before going on to discuss both the strategies they used and how effective they proved to be. This option is favoured by those who see compensatory strategy use as a problem– solution situation (Field, 2008a), in which the strategic response of the listener varies according to the nature of the problem and to some extent the temperament of the individual. Vandergrift (2004) recommends a mixed approach.  This is especially the case with learners living in an L2 environment.

8

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

299

12.7  Some Other Current Issues in L2 Listening 12.7.1  Visual Input After an upsurge in the 1980s in the production of video-based language courses, publishers seem more reluctant these days to support their materials with visual content. However, improvements in the quality of digitized sound mean that instructors now have access to a large amount of promising authentic material via the internet. Hence the continuing research interest in the impact of video versus audio material upon the L2 listening experience. Intuitively, it would seem that video is more representative of realworld listening. It provides a visible context for the speech event; it enables speakers to be identified physically; and it gives access to the gestures and facial expressions of the speakers. The much-quoted McGurk effect (McGurk & McDonald, 1976) demonstrates that phonetic judgments are influenced by lip movements; there is even neurological evidence (Green, 1998) that visual information becomes integrated into the listening process at a very early stage. By contrast, audio material reflects only a small range of circumstances (phone conversations, radio broadcasts). Curiously, the evidence that visual information enhances L2 listening comprehension is more mixed than might be assumed. Sueyoshi and Hardison (2005) reported improved results from participants when they had sight of the speaker’s face and gestures. By contrast, Coniam (2001) found no difference between scores on a video version of a test as against an audio one; strikingly, eighty per cent of the video group reported that the visual support had not assisted their comprehension. Gruba (2004) also interviewed learners, this time after they had watched news clips, and concluded that visual information can bring benefits to higher proficiency listeners but potentially confuse weaker ones. Yet another result was reported by Suvorov (2009), who directly compared performance on audio recordings with performance on recordings supported by still pictures and by video. The scores for the video-assisted material were significantly lower than those for the other two conditions. Perhaps the best way of interpreting these conflicting findings is to envisage visual material as imposing increased cognitive demands upon the listener, who has to integrate an extra source of information into the mental model being constructed. This may be particularly difficult for earlystage listeners, given their need to focus much attention on perceptual processing (section 12.3.2). However, a recent study by Batty (2015), using Rasch modelling, failed to find an interaction between medium of presentation and proficiency level, or indeed between medium and text type (monologue versus conversation versus lecture).

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

300

John Field

Two relatively unexplored areas of multimodality concern written stimuli. Firstly, if we are to fully represent the circumstances of academic listening, we need a better understanding of the way Powerpoint slides are processed by the listener and of how they shape the information that is derived. There is some tentative evidence of the interaction between slides and voice in an eye-tracking study by Suvorov (2015). A second topic is the part that subtitles might play in assisting self-access listening to video material. A preliminary study by Guichon and McLornan (2008) suggests that L2 subtitling is more beneficial than L1 because it causes less interference at a lexical level.

12.7.2 Vocabulary There has been recent interest amongst vocabulary scholars in whether successful performance in a listening task requires the same level of vocabulary knowledge as success in a similar reading task. This follows a pilot study by Milton and Hopkins (2006) which used a test of phonological word forms to track how the aural repertoire of a group of school learners of English developed across eight grade levels. At every level, knowledge of the phonological forms of the target words lagged behind the knowledge of orthographic forms—and indeed the gap between the two became greater as the learners grew older and more proficient. What was striking about this finding was that the learners were from a Greek school, where one might have anticipated problems with an unfamiliar alphabet to delay orthographic development. In the main study, Milton and Hopkins (2006) went on to test two large groups of Greek and Arabic speakers from a range of proficiency levels. At the very lowest levels, there was evidence that words were first acquired phonologically by the Arabic speakers; but from a lower-intermediate stage, orthographic forms began to predominate, and, at advanced level, they sometimes exceeded oral forms by a large margin. On the basis of this and other research findings, Milton (2009) suggests a target oral vocabulary of around 3000 to 3500 word families at CEFR B2 level and about 4500 at C2, as compared with suggestions elsewhere (Nation, 2001) that readers may require 6000 or more. Milton partly attributes these lower vocabulary needs to the different content of spoken and written texts. He suggests (p. 179) that, as proficiency improves, reading and writing depend increasingly upon a learner’s ability to handle large quantities of infrequent vocabulary, whereas listeners may reach a ceiling in their vocabulary knowledge, after which increases in it are not reflected correspondingly in their test performance. An additional interpretation may perhaps be found in the way in which learners store vocabulary. Literate individuals rely heavily on the

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

301

written word to conserve information; they are thus much more adept at retaining written forms than spoken ones. What is more, those written forms are consistently reinforced during reading, whereas what learners encounter in listening is (see section 12.4) a highly variable version of the citation form they encounter in a vocabulary class. As already noted, the difficulty of recognizing words in connected speech also makes listening a more approximate operation than reading, one where strategy use may need to be relied upon to a greater extent. Indeed, one of the most widely used strategies for L2 listeners at all proficiency levels entails inferring the meanings of unknown or unrecognized words (Wesche & Paribakht, 2010), using a variety of cues. This may well make listeners more willing than readers to tolerate gaps in their vocabulary. Inevitably linked to this discussion is the notion of coverage, i.e., the percentage of known vocabulary that an L2 reader needs in a text to ensure “comfortable comprehension”. Applying findings from L2 reading, Nation (2001) suggests that listeners “need at least 95% coverage of the running words in the input to gain reasonable comprehension and to have reasonable success at guessing from context” (p. 114). However, the concept of “coverage” surely has to be a relative one: much depends upon the goals of the listener or reader, the discourse type (e.g., narrative versus discursive), and the information density of the text.9 Even without these provisos, van Zeeland and Schmitt (2013) reported accurate responses from both native and non-native listeners when a narrative text had ninety per cent coverage. They concluded that a vocabulary level of around 2000 word families would suffice for this kind of discourse, supporting Milton’s (2009) relatively low figure. Again, this may be evidence that the lexical content of listening and the more strategic approach of listeners mean that vocabulary needs are lower than in reading. As a final note, in their research into vocabulary knowledge and recognition in L2 listening, Milton and colleagues made use of a corpus of phonological word frequency (Milton, 2009). Surprisingly often, researchers have reported conclusions on listening vocabulary that are based upon orthographic corpora; this includes Staehr’s (2009) widely quoted account.

12.7.3  Cross-Language Phoneme Perception The extensive body of research into L1 and L2 phoneme perception has received only limited attention in the general L2 listening literature. As  Nearly fifty per cent of words in a corpus of speech samples analysed by McCarthy & Carter (1997) were function

9

words. So the notion of ninety-five per cent coverage would appear to represent one unfamiliar content word in every ten.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

302

John Field

already noted (Section 12.4.1), there is a widely accepted view that the phoneme does not provide a reliable unit of analysis because it is so variable.10 Nevertheless, we still need to address the question of precisely how an L2 learner (or for that matter, an infant acquiring an L1) manages to convert recurring clusters of acoustic cues into the sound system of a language—even if that system is at first mainly perceived in terms of syllables and words and only later forms the basis for phoneme constructs of the ship/sheep kind. A solution that has been widely supported by L1 speech scientists draws on exemplar models of how language is acquired and used (Pisoni, 1997; Goldinger, 1997; Pierrehumbert, 2001; Bybee, 2001; Nygaard, 2008). The assumption is that speech encounters leave long-term memory traces in a listener’s mind. The ability to trace phoneme values and word forms in connected speech is linked to the ability to recognize familiar voices, and results from the listener constantly finding analogies between new input and what was heard previously. This might seem highly improbable, but it accords with a recent neurological finding that the human brain is far slower at computational operations than was once believed but is capable of storing massive amounts of information in an easily retrievable way (Dąbrowska, 2004). The exemplar principle goes a long way towards explaining how our awareness of word frequencies and of formulaic chunks is acquired (Bybee, 2010). If one accepts this version of events, there are important conclusions so far as the second language listener is concerned. The auditory acquisition of phoneme values seems to depend not upon repetition exercises but upon exposure over time to multiple variants of phonemes and words spoken in a range of different voices. Kuhl’s “Perceptual Magnet” account of how infants acquire L1 phonemes (Hawkins, 1999a, pp. 188–191) provides a useful way of envisaging what happens. It represents less typical exemplars of a phoneme (or indeed of a syllable or word) as becoming marginalized over time while those that share certain common features are increasingly grouped around a central value. Within the second language acquisition process, we also need to understand how the influence of L1 affects the perception of L2 phoneme values. Best’s (1995) Perceptual Assimilation Model aims to account for the way in which naïve listeners respond to L2 speech, while Flege’s (1995) Speech Learning Model is more concerned with L2 developmental patterns. However, they agree on general principles (Best & Tyler, 2007). Their research suggests that the way in which a listener becomes familiar 10

 Research in this area is generally neutral as to whether the phoneme is a primary unit of analysis, though Hawkins (1999b, p. 259) recognizes that “it is reasonable to suggest that words can be identified before their constituent phonemes.”

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

303

with the phonemes of a second language is determined by whether a phoneme is perceived (a) as a good example of one in L1 (but is not necessarily so); (b) as a less satisfactory example; or (c) as markedly different to any in L1. It is not the third category that causes perceptual difficulties but the second and, above all, the first. These difficulties may impact word recognition and ultimately comprehension. Other researchers (e.g., Bohn, 1995) have adopted a finer-grained approach, investigating whether there are certain acoustic features (e.g., place of articulation in consonants) which assist L2 phoneme recognition more than others such as voicing. In another extended research programme (Piske, 2007), Flege and colleagues have investigated the impact upon L2 phoneme acquisition of external factors such as AOL (age at which learning began), length of residence, and language use. The findings are mixed. There is evidence that an early AOL does not necessarily lead to a native-like accent. Length of residence and language use appear to be interconnected: much depends upon the extent to which individuals continue to be exposed to L2 accented speech, thus reinforcing their own patterns of speech. This overview has necessarily been a brief one. For more detailed accounts, see Hawkins (1999b), Strange and Shafer (2008), and Nygaard (2008). Strange and Shafer (p. 163) mark out a direction for future research in proposing the use of sections of natural connected speech rather than the citation forms of words that have often been employed in these studies.

12.7.4 Accent There has been considerable research and discussion on the impact of accents, both those within the target language and L2 variants. Major et al. (2005) investigated how accents of different degrees of familiarity (standard taught forms, regional and international varieties of English) affected comprehension by both native and non-native listeners. The results indicated that the more familiar the accent, the better the level of understanding—but the effect was considerably greater in the case of the non-native group. A large-scale project by Ockey and French (2016) found that unfamiliar international accents (British and Australian) impacted adversely on the comprehension of candidates taking material based on the TOEFL listening test. It also demonstrated a correlation between accent familiarity and the score achieved. Floccia and colleagues (2006, 2009) investigated the effects of training. They monitored the recognition by native English listeners of regional and L2 accents, and concluded that extensive exposure was necessary to ensure full adaptation. An interesting study by Bradlow and Bent (2008),

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

304

John Field

using L1 listeners and Chinese-accented varieties of English, suggested that listeners succeeded in generalizing their training to new voices when they were exposed to multiple speakers, but not when they were exposed to a single one. A similar result was reported by Clopper and Pisoni (2004) in relation to regional varieties of US English. These accounts would seem to accord with the exemplar model of phonological and word acquisition mentioned earlier. An issue that has aroused considerable interest is whether L2 listeners find others with their own language background easier to understand than native speakers: what Bent and Bradlow (2003) term “the inter-language speech intelligibility benefit”. Imai, Walley, and Flege (2005) concluded that, for lower proficiency learners, there was little or no difference. Similarly, Major et al. (2002) and Munro, Derwing, and Morton (2006) produced only limited evidence of benefit. One conclusion by the latter was that a great deal depended upon whether the groups they studied came from a second-language background (and were trained by L1-accented teachers) or from a foreign-language one. In short, familiarity with a particular variety counts for more than the speaker’s origin. If a learner is studying English, Spanish, Portuguese, or a number of other languages with distinctive international varieties, then (given our globalized media and systems of communication) it makes obvious sense to expose them to any of those varieties that represent relatively large populations. The question (little addressed in the literature) is when. In light of the exemplar model of speech perception presented in section 12.7.3, it would seem that exposure plays a critical role. A first priority is to lay down sufficient traces of a single variety before moving on to others, so as to provide the learner with a set of consistent phonemic and lexical points of reference. A useful point might be CEFR B1+ level, once a degree of automaticity in basic perceptual processes has been achieved (Field, 2018). But what about local and regional varieties of the L2? Here, discussion sometimes gets hijacked by the argument that focusing teaching on a single generalized variety entails assumptions of “prestige”. It is important to separate out local loyalties from the simple practical need to equip earlystage learners with a single variety which provides them with a set of perceptual points of reference. To be sure, regional varieties of the target language might be introduced if they represent substantial populations—but only at a later stage and, of course, the local variety should be taught from the outset in those parts of the world where English serves as the language of communication (Harding, 2011). There remains the vexed issue of the desirability of including L2 accents in listening materials and tests. The argument generally put forward in the case of English (Jenkins, 2002) is that, these days, more spoken exchanges take place between two L2 speakers than between an L1

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

305

and an L2 speaker, a point well made by commentators such as Crystal (2003). The principle of representing international communication in this way appears uncontroversial—until one considers the practicalities. Communication is a two-way operation; one has to consider not simply the productive characteristics of the speaker but also how the receptive characteristics of the listener influence understanding (Flege, 1995). This entails a vast number of possible speaker L1–listener L1 combinations. Another obstacle is that targets such as “Chinese accented speech” or “Portuguese accented speech” represent large generalizations, failing to distinguish between the different effects upon L2 of Mandarin versus Cantonese and Brazilian Portuguese versus European. A further factor is that the form of speech produced by an L2 speaker is also determined by the variety that he/she was originally taught; in the case of English, this could be American, British, or Australian/New Zealand. This may sound theoretical, but it becomes a major concern when various L2 varieties (or indeed relatively minor L1 varieties) to which candidates may never have been exposed are employed in international tests of L2 listening. As has been pointed out (Taylor, 2006; Geranpayeh & Taylor, 2008), the risks of test bias are considerable, favouring those who may, by pure chance, have encountered some of these varieties, and disfavouring others who do not have the financial means to study or to travel abroad. The issue of accent in listening tests was addressed in a sensitive study by Harding (2011). While his findings were not conclusive, he raises the possibility of a weak ELF (English as a lingua franca) approach using only L2 speakers who are identified as “intelligible” (but how to ensure that they are equally intelligible to a wide range of listeners with different L1s?11). He also recognizes the need for a “Local Englishes” approach in domains where English is widely used as a medium of communication.

12.7.5  Double Play It would seem intuitively obvious that comprehension improves if a learner is permitted two or more plays of the recording. Findings by Cervantes and Gainer (1992) suggest that learners benefit from a second hearing, regardless of their level of proficiency. However, Chang and Read (2006) report a different result, with higher-level learners benefiting much more than lower-level ones, perhaps because the latter lacked adequate perceptual skills.

11

 Jenkins’s (2002) “Lingua Franca Core” does not provide an answer. Contrary to what is sometimes suggested, it is not a fail-safe guide to features of L2 English speech that ensure maximum intelligibility worldwide. It is improbable that such a generalization could ever be reliably achieved, given the multitude of possible speaker–listener L1 pairings. In fact, the original proposal for the Core was based upon a very small sample of speaker–listener dyads with no controls for the L1s of those represented.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

306

John Field

The question of whether to play a recording twice is especially an issue for testers, as it is a well-established convention in many listening tests. The main supporting argument is that, when using audio material, it compensates for not having access to paralinguistic signals or sight of the physical context of the speech event. It also gives listeners the opportunity to normalize to an unfamiliar speaker’s voice (section 12.4.2). In addition, with small-scale or computer-delivered tests, it enables the test questions to be delivered between the two plays (Sherman, 1997), thus considerably reducing the opportunity for test takers to use test-wise strategies based upon the wording of written items. However, the practice has been questioned on the grounds that, in a real-world situation, a listener is unable to hear a piece of spoken input more than once. Admittedly, that assertion is not entirely true these days: there are usually playback facilities for radio and TV programmes, university lectures, etc.; and in an interactive context, one can always ask the speaker for clarification. However, opponents also claim that the double-play convention unnaturally reduces the difficulty of a test; some major tests (particularly tests of academic listening such as TOEFL or IELTS) therefore restrict candidates to a single play. The lack of research on the effects of double-play in test conditions is surprising, especially after Henning (1990) suggested that scores from a single play discriminated better. However, a major study was recently undertaken by Ruhm et al. (2016) involving over 1250 low-level Austrian school learners. The recordings used were relatively short clips. Results on the shorter ones (twenty-five seconds or less) generally showed an improvement on the second play; scores on the longer ones (over sixty seconds) sometimes increased and sometimes even went down. The other recent study (Field, 2015) used two former IELTS recordings to investigate the impact of double play upon score bands and upon the cognitive behaviour of adult test takers. It found that double-play raised scores, but by a smaller amount than expected. Replaying the recordings did not appear to disproportionately favour test takers at any of the three proficiency levels tested; if anything, it marked out the differentials between score bands more sharply. In protocols following the task, a majority of test takers reported that they used the first play to orient themselves to the listening passage and the second to confirm their answers to questions. In other words, during at least one play, they were engaging in a type of listening that approximated more closely to real world behaviour than item-driven tasks usually permit. Several participants reported that hearing the recording twice had the effect of reducing listening anxiety (section 12.7.7). A more adventurous solution to the issue of whether to permit a double play or not is to allow the individual test taker to choose. This is obviously

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

307

only possible in computer-delivered tests, where the timing of the listening session can be flexible. A concern here is that the effort of decision-making about whether to rewind or not may impose additional stress upon a candidate.

12.7.6 Authenticity Materials writers and test producers have four possible sources for the recordings they choose: they can script them; they can ask actors to improvise a situation in the studio; they can use recordings of real-life speech events; or they can make use of semi-scripted materials, where an authentic event is transcribed and rerecorded under studio conditions. Test producers in particular find the fourth option attractive as it avoids complications associated with reproduction rights. It also enables them to manipulate the original text by editing its language and even introducing multiple-choice distractors. In this context, the notion of what constitutes an “authentic” text is sometimes stretched to its limits. Salisbury (2005) reports surprisingly that of ten experienced item writers studied, only one regularly used oral material as his point of departure, with others generating text from written sources such as magazine interviews. There are implications here for the extent to which tests of L2 listening predict real-world behaviour. In a study by Wagner and Toth (2014), learners of Spanish scored significantly higher when exposed to a rescripted version of a natural text than when hearing the original version. The authentic option has increasingly become a possibility open to instructors, given the wide availability of both oral and visual materials on the internet. However, there has been relatively little research into the effects upon learner comprehension and motivation of using these sources. Herron and Seay (1991) reported improvements in learners’ listening skills as a result of supplementing conventional materials with authentic ones. Wu and Stansfield (2001) constructed tests based on improvised material, but compromised spontaneity by recording a target conversation multiple times “until it was determined to be wholly authentic” (p. 196). Gilmore (2011) reports on the positive effects of exposure to authentic listening materials upon Japanese learners’ communicative competence. Much of the work in this area has focused on formal differences between authentic and scripted speech. In an interesting study that compared material within a single genre, Gilmore (2004) identified a predictably higher prevalence of false starts, repetitions, overlaps, and back-channelling in authentic sources, and very large differences in relation to fillers and pauses (the latter were not subdivided by type of pause). Field (2013) reports that test writers at lower levels sometimes succeed better

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

308

John Field

at emulating natural speech than those at higher levels, where there are pressures to increase the density of the information and the complexity of the links between it.

12.7.7  Listening Anxiety Perhaps partly because of the extensive use of comprehension questions, L2 listening places heavy processing demands upon learners. This has stimulated interest in the types of language learning anxiety (Horwitz, 2001) associated with them. A listening anxiety questionnaire designed by Kim (2000) has been widely used to establish the degree and causes of the phenomenon. A major factor appears to be the variable nature of the spoken input. Vogely (1999) found that around half the concerns reported by learners of Spanish at a US university related to this issue. A more recent study of trainee teachers by Bekleyen (2009) also records failure to recognize spoken words as a major challenge to confidence. Chang (2008) reports unsurprisingly that test situations add to the stress experienced in classroom settings, while an interesting study by Chang and Read (2008) identifies concrete ways of reducing anxiety in listening tests, including providing topical knowledge and replaying the recording. A second topic of interest is the possible relationship between anxiety and poor performance. The current consensus appears to be that the former causes the latter rather than vice versa (Kim, 2000; Zhang, 2013). A related question is whether anxiety results from a lack of strategic competence with which to resolve problems of comprehension.

12.8 Conclusion This review has covered a range of recent topics in L2 listening theory and research, chosen in part because they differ from those that have historically attracted attention. The purpose, particularly in section 12.7, was to show the diversity of current work in the field and to identify issues of likely interest to future investigators. A theme common to many of the studies mentioned is the need to achieve a deeper understanding of the mind of the L2 listener and the challenges he/she faces. This marks a shift in emphasis from past research, which quite often focused on how aspects of listening texts and tasks contributed to difficulty (Bloomfield et al., 2010). The theoretical background and much of the research quoted highlights the significant part played by the characteristics of connected speech. Instructors, researchers, and test designers cannot afford to lose sight of the transitory nature and the extreme variability of the medium by which

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

309

information reaches the ears of learners. Many failures of “comprehension” by lower-proficiency listeners are actually failures of perceptual processing (Field, 2012). Even if those listeners were capable of conceptual processes such as inferring connections between pieces of information (and they may well be in L1), they cannot employ them if they have to focus heavy attention on decoding, nor can they if there is insufficient decoded material to provide a basis for constructing meaning. A theme that has occurred several times in this review is the role of exposure. While there is no full agreement on precisely how L2 listeners acquire phonological knowledge, the most persuasive account seems to be the “episodic” or “exemplar” one, based on the assumption that we store traces of speakers’ voices in our minds, from which, over time, we manage to extract generalized phoneme and word values. This accounts for phenomena such as phoneme recognition, accent familiarity, awareness of word frequency, voice normalization, and the acquisition of chunks. But it also conduces to a view of the L2 listener as an individual whose behaviour is shaped by the number and quality of his/her encounters with the target language. Amongst other concerns, it sounds a cautionary note about longitudinal studies of listening development that focus on learners who are exposed on a daily basis to L2 influences outside the classroom. How might this awareness of the listener as an individual affect approaches to instruction? A flaw of the Comprehension Approach mentioned at the outset is that it tends to treat learners as uniform in their behaviour, with the instructor or materials writer anticipating which parts of a passage will prove problematic. Three solutions have been touched upon briefly. One is to adopt a diagnostic approach that follows up wrong answers by replaying the relevant section and identifying across class members why the problem occurred. This can be followed by remedial tasks focusing on the feature that caused the difficulty. A second is to discuss and develop compensatory listening strategies, particularly at lower proficiency levels, in order to sustain learners’ motivation and self-efficacy by enabling them to handle gaps in understanding in realworld contexts. A third approach is to transfer part of the instructional programme to self-access activities—an option that has become possible thanks to the wide availability of downloadable material. Much future listening practice seems likely to relocate from the classroom to a listening centre or to the home, where learners can listen to recordings as many times as they need, picking up on features that they as individuals find difficult and benefiting post-practice from a downloadable script. In time, materials designers may be able to add online remedial tasks linked to specific difficulties of audition that are likely to occur.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

310

John Field

References Afflerbach, P., Pearson, P. D. & Paris, S. G. (2008). Clarifying differences between reading skills and reading strategies. The Reading Teacher, 61, 364–373. Aitchison, J. (2008). Words in the mind (4th edn). London: Wiley. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press Anderson, J. R. (2000). Cognitive psychology and its implications (5th edn). New York: W. H. Freeman. Bates, E., & MacWhinney, B. (1989). Functionalism and the Competition Model. In B. MacWhinney & E. Bates (eds.), The cross-linguistic study of sentence processing (pp. 777–780). New York: Cambridge University Press. Batty, A. O. (2015). A comparison of video- and audio-mediated listening tests with many-facet Rasch modelling and differential distractor functioning. Language Testing, 32(1), 3–20. Bekleyen, N. (2009). Helping teachers become better English students: Causes, effects, and coping strategies for foreign language listening anxiety. System, 37(4), 664–675. Bent, T., & Bradlow, A. R. (2003). The interlanguage speech intelligibility benefit. Journal of the Acoustical Society of America, 114(3), 1600–1610. Best, C. T. (1995). A direct realist perspective on cross-language speech perception. In W. Strange (ed.). Speech perception and linguistic experience: Issues in cross-language research (pp. 171–204). Timonium, MD: York Press. Best, C. T., & Tyler, M. D (2007). Nonnative and second-language speech perception: Commonalities and complementarities. In M. J. Munro & O.-S. Bohn (eds.) Second language speech learning: The role of language experience in speech perception and production (pp. 13–34). Amsterdam: John Benjamins. Bloomfield, A., Wayland, S. C., Rhoades, E., Blodgett, A., Linck, J., & Ross, S. (2010). What makes listening difficult: Factors affecting second language comprehension. Report: University of Maryland. Bohn, O.-S. (1995). Cross-language speech perception in adults: First language transfer doesn’t tell it all. In W. Strange (ed.), Speech perception and linguistic experience: Issues in cross-language research (pp. 279–304). Timonium, MD: York Press. Bradlow, A. R., & Bent, T. (2008). Perceptual adaptation to non-native speech. Cognition, 106(2), 707–729. Braxton, M. A. (1999). Adult ESL language learning strategies: Case studies of preferred learning styles and perceived cultural influences in academic listening tasks. Unpublished PhD dissertation, Ohio State University, Columbus, OH. Brown, G. (1986). Investigating listening comprehension in context. Applied Linguistics, 7(3), 284–302.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

311

Brown, G. (1990). Listening to spoken English (2nd edn). Harlow: Longman. Brown, G. (1995). Speakers, listeners and communication. Cambridge: Cambridge University Press. Brown, G., & Yule, G. (1983). Discourse analysis. Cambridge: Cambridge University Press. Brown, J. D., & Kondo-Brown, K. (Eds.) (2006). Perspectives on teaching connected speech to second language speakers. Mãnoa, HI: University of Hawaii. Bybee, J. (2001). Phonology and language use. Cambridge: Cambridge University Press. Bybee, J. (2010). Language, usage and cognition. Cambridge: Cambridge University Press. Cauldwell, R. (2013). Phonology for listening. Birmingham: Speechinaction. Cervantes, R., & Gainer, G. (1992). The effects of syntactic simplification and repetition on listening comprehension. TESOL Quarterly, 26(4), 767–770. Chang, A. C.-S. (2008). Sources of listening anxiety in learning English as a foreign language. Perceptual and Motor Skills, 106(1), 21–34. Chang, A. C.-S., & Read, J. (2006). The effect of listening support on the listening performance of EFL learners. TESOL Quarterly, 40, 375–397. Chang, A. C.-S., & Read, J. (2008). Reducing listening test anxiety through various forms of listening support. TESL-EJ, 12(1), 1–25. Chung, J. M. (1999). The effects of using video texts supported with advance organizers and captions. Foreign Language Annals, 32(3), 295–307. Clopper, C. G., & Pisoni, D. B. (2004). Effects of talker variability on perceptual learning of dialects. Language and Speech, 47, 207–239. Cohen, A. (1998). Strategies of language learning and language use. Harlow: Longman. Coniam, D. (2001). The use of audio or video comprehension as an assessment instrument in the certification of English language teachers: A case study. System, 29, 1–14. Council of Europe (2018) Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Available online at https://rm.coe.int/cefr-companion-volume-with-new-descriptors-2018/1680787989. Crystal, D. (2003). English as a global language. Cambridge: Cambridge University Press. Cutler, A. (1990). Exploiting prosodic possibilities. In G. Altmann (ed.), Cognitive models of speech processing: Psycholinguistic and computational perspectives (pp. 105–121). Cambridge, MA: MIT Press. Cutler, A. (1997). The comparative perspective on spoken-language processing. Speech Communication, 21, 3–15. Cutler, A. (2012). Native listening. Cambridge, MA: MIT Press. Cutler, A., & Clifton, C. (1999). Comprehending spoken language: A blueprint of the listener. In C. M. Brown & P. Hagoort (eds.), The neurocognition of language (pp. 123–166). Oxford: Oxford University Press.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

312

John Field

Cutler, A., & Otake, T. (1994). Mora or phoneme? Further evidence for language-specific listening. Journal of Memory and Language, 33, 824–844. Dąbrowska, E. (2004). Language, Mind and Brain. Edinburgh: Edinburgh University Press. Dörnyei, Z., & Scott, M. L. (1997). Communication strategies in a second language: Definitions and taxonomies. Language Learning, 47(1), 173–210. Ericsson, K. A., & Smith, J. (eds.) (1991). Toward a general theory of expertise: Prospects and limits. Cambridge: Cambridge University Press. Faerch, G., & Kasper, G. (1983). Plans and strategies in foreign language communication. In G. Faerch & G. Kasper (eds.), Strategies in interlanguage communication. London: Longman. Feltovich, P. J., Prietula, M. J., & Ericsson, K. A. (2006). Studies of expertise from psychological perspectives. In K. A. Ericsson, N. Charness, P. J. Feltovich, & R. R. Hoffman (eds.), The Cambridge handbook of expertise and expert performance (pp. 41–67). Cambridge: Cambridge University Press. Field, J. (1998). Skills and strategies: Towards a new methodology for listening. ELT Journal, 54(2), 110–118. Field, J. (2000). “Not waving but drowning”: A reply to Tony Ridgway. ELT Journal, 54(2), 186–195. Field, J. (2001). Lexical segmentation in first and foreign language listening. Unpublished PhD dissertation, University of Cambridge. Field, J. (2004). An insight into listeners problems: Too much bottom-up or too much top-down? System, 32, 363–377. Field, J. (2008a). Listening in the language classroom. Cambridge: Cambridge University Press. Field, J. (2008b). Revising segmentation hypotheses in first and second language listening. System, 36, 35–51. Field, J. (2008c). Bricks or mortar: Which parts of the input does a second language listener rely on? TESOL Quarterly, 42(3), 1–22. Field, J. (2012). The cognitive validity of the lecture based question in the IELTS listening paper. In L. Taylor & C. Weir (eds.), IELTS collected papers 2: Research in reading and listening assessment (pp. 391–453). Cambridge: Cambridge University Press. Field, J. (2013). Cognitive validity. In A. Geranpayeh & L. Taylor (eds.), Examining listening (pp. 77–151). Cambridge: Cambridge University Press. Field, J. (2014). Myth: Pronunciation teaching needs to fix in the minds of learners a set of distinct consonant and vowel sounds. In L. Grant (ed.), Pronunciation myths (pp. 80–106). Ann Arbor, MI: University of Michigan Press. Field, J. (2015). The effects of single and double play upon listening test outcomes and cognitive processing. Report: British Council ARAGS Reports. Retrieved from http://www.britishcouncil.org/exam/aptis/ research/publications. Field, J. (2018). Rethinking the second language listening test: From theory to practice. Sheffield: Equinox.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

313

Flege, J. E. (1995). Second language speech learning: Theory, findings, and problems. In W. Strange (ed.), Speech perception and linguistic experience: Issues in cross-language research (pp. 233–276). Timonium, MD: York Press. Floccia, C., Butler, J., Goslin, J., & Ellis, L. (2009). Regional and foreign accent processing in English: Can listeners adapt? Journal of Psycholinguistic Research, 38(4), 379–412. Floccia, C., Goslin, J., Girard, F., & Konopczynski, G. (2006). Does a regional accent perturb speech processing? Journal of Experimental Psychology: Human Perception and Performance, 32(5), 1276–1293. Fowler, C., & Magnuson, J. S. (2012). Speech perception. In M. Spivey, K. McRae, & M. Joanisse (eds.) The Cambridge handbook of psycholinguistics (pp. 3–25). Cambridge: Cambridge University Press. Geranpayeh, A., & Taylor, L. (2008). Examining listening: Developments and issues in assessing second language listening. Cambridge ESOL Research Notes, 32, 2–5. Gernsbacher, M.-A. (1990). Language comprehension as structure building. Hillsdale, NJ: Erlbaum. Gilmore, A. (2004). A comparison of textbook and authentic interactions. ELT Journal, 58(4), 363–371. Gilmore, A. (2011). “I Prefer Not Text”: Developing Japanese learners’ communicative competence with authentic materials. Language Learning, 61(3), 786–819. Goh, C. (1998). How ESL learners with different listening abilities use comprehension strategies and tactics. Language Teaching Research, 2, 124–147. Goh, C. (2000). A cognitive perspective on language learners’ listening comprehension problems. System, 28, 55–75. Goh, C., & Taib, Y. (2006). Metacognitive instruction in listening for young learners. ELT Journal, 60(3), 222–232. Goldinger, S. D. (1997). Words and voices: Perception and production in an episodic lexicon. In K. Johnson & J. W. Mullennix (eds.), Talker variability in speech processing (pp. 33–66). San Diego, CA: Academic Press. Goldinger, S. D. (1998). Echoes of echoes? An episodic theory of lexical access. Psychological Review, 105, 251–279. Gompel, R. P. G., & Pickering, M. (2007). Syntactic parsing. In G. Gaskell (ed.), The Oxford handbook of psycholinguistics (pp. 289–398). Oxford: Oxford University Press. Graham, S. (2006). Listening comprehension: The learner’s perspective. System, 34(2), 165–182. Graham, S. (2011). Self efficacy and academic listening. Journal of English for Academic Purposes, 10(2), 113–117. Green, K. P. (1998). The use of auditory and visual information during phonetic processing: Implications for theories of speech perception. In R. Campbell & B. Dodd (eds.) Hearing by eye II: Advances in the psychology of speechreading and audiovisual speech (pp. 3–25). Hove: Psychology Press.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

314

John Field

Grellet, F. (1981). Developing reading skills. Cambridge: Cambridge University Press. Grosjean, F. (1985). The recognition of words after their acoustic offsets: Evidence and implications. Perception and Psychophysics, 38, 299–310. Gruba, P. (2004). Understanding digitized second language videotext. Computer Assisted Language Learning, 17(1), 51–82. Guichon, N., & McLornan, S. (2008). The effects of multimodality on L2 learners: Implications for CALL resource design. System, 36, 85–93. Hansen, C., & Jensen, C. (1994). Evaluating lecture comprehension. In J. Flowerdew (ed.) Academic listening: Research perspectives (pp. 241–268). Cambridge: Cambridge University Press. Harding, L. (2011). The use of speakers with L2 accents in academic English learning listening assessment: A validation study. Frankfurt am Main: Peter Lang. Harding, L., Alderson, J. C., & Brunfaut, T. (2015). Diagnostic assessment of reading and listening in a second or foreign language: Elaborating on diagnostic principles. Language Testing, 32(3), 317–336. Hawkins, S. (1999a). Auditory capacities and phonological development: Animal, baby and foreign listeners. In J. M. Pickett (ed.) The acoustics of speech communication: Fundamentals, speech perception theory, and technology (pp. 183–197). Needham Heights, MA: Allyn and Bacon. Hawkins, S. (1999b). Re-evaluating assumptions about speech perception: Interactive and integrative theories. In J. M. Pickett (ed.) The acoustics of speech communication: Fundamentals, speech perception theory, and technology (pp. 232–288). Needham Heights, MA: Allyn and Bacon. Henning, G. (1990). A study of the effects of variation of short term memory load, reading response length and processing hierarchy on TOEFL listening comprehension item performance. Princeton, NJ: ETS Report, 33. Herron, C., & Seay, I. (1991). The effect of authentic oral texts on student listening comprehension in the foreign language classroom. Foreign Language Annals, 24(6), 487–495. Herron, C., Cole, P., York, H., & Linden, P. (1998). A comparison study of student retention of foreign language video: Declarative versus interrogative advance organizer. The Modern Language Journal, 82(2), 237–247. Horwitz, E. K. (2001). Language anxiety and achievement. Annual Review of Applied Linguistics, 21, 112–126. Hulstijn, J. (2003). Connectionist models of language processing and the training of listening skills with the aid of multimedia software. Computer Assisted Language Learning, 16(5), 413–425. Imai, S., Walley, A. C., & Flege, J. E. (2005) Lexical frequency and neighborhood density effects on the recognition of native and Spanish-accented words by native English and Spanish listeners. Journal of the Acoustical Society of America, 117, 896–906. Jenkins, J. (2002). A sociolinguistically based, empirically researched pronunciation syllabus for English as an international language. Applied Linguistics, 23(1), 83–103.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

315

Johnson, K. (2005). The “general” study of expertise. In K. Johnson (ed.), Expertise in second language learning and teaching (pp. 11–33). Basingstoke: Palgrave Macmillan. Johnson, K. (2008). Speaker normalisation in speech perception. In D. B. Pisoni & R. E. Remez (eds.), The handbook of speech perception (pp. 363–389), Oxford: Blackwell. Kim, S. Y. (2000). Foreign language listening anxiety: A study of Korean students learning English. Unpublished PhD dissertation, University of Texas, Austin, TX. Klatt, D. H. (1979). Speech perception: A model of acoustic-phonetic analysis and lexical access. Journal of Phonetics, 7, 279–312. Laver, J. (1994). Principles of phonetics. Cambridge: Cambridge University Press. Levelt, W. J. M. (1993). The architecture of normal spoken language use. In G. Blanken, J. Dittman, H. Grimm, J. Marshall, & C.-W. Wallesch (eds.), Linguistic disorders and pathologies (pp. 1–15). Berlin: De Gruyter. Lynch, T. (2006). Academic listening: Marrying top and bottom. In A. Martinez-Flor & E. Uso-Juan (eds.), Current trends in learning and teaching the four skills within a communicative framework (pp. 91–110). Amsterdam: Mouton. Lynch, T. (2009). Teaching second language listening. Oxford: Oxford University Press. Macaro, E., Graham, S., & Vanderplank, R. (2007). A review of listening strategies: Focus on sources of knowledge and on success. In A. D. Cohen & E. Macaro (eds.), Language learner strategies (pp. 165–185). Oxford: Oxford University Press. Major, R. C., Fitzmaurice, S. F., Bunta, F., & Balasubramian, C. (2002). The effects of nonnative accents on listening comprehension: Implications for ESL assessment. TESOL Quarterly, 36(2), 173–190. Major, R. C., Fitzmaurice, S. F., Bunta, F., & Balasubramian, C. (2005). Testing the effects of regional, ethnic and international dialects of English on listening comprehension. Language Learning, 55(1), 37–69. Mattys, S. L., & Melhorn, J.F. (2005). How do syllables contribute to the perception of spoken English? Insight from the migration paradigm. Language and Speech, 48, 223–253. McCarthy, M., & Carter, R. (1997). Written and spoken vocabulary. In N. Schmitt & M. McCarthy (eds.), Vocabulary: Description, acquisition and pedagogy (pp. 20–39). Cambridge: Cambridge University Press. McGurk, H., & McDonald, J. (1976). Hearing lips and seeing voices. Nature, 264, 746–748. McQueen, J. M. (1998). Segmentation of continuous speech using phonotactics. Journal of Memory and Language, 39, 21–46. McQueen, J. M. (2007). Eight questions about spoken word recognition. In G. Gaskell (ed.), The Oxford handbook of psycholinguistics (pp. 37–54). Oxford: Oxford University Press.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

316

John Field

Mendelsohn, D. J. (1994). Learning to listen: A strategy-based approach for the second-language learner. San Diego, CA: Dominie Press. Miller, J. L. (1990). Speech perception. In D. N. Oscherson & H. Lasnik (eds.), An invitation to cognitive science (Vol. 1, pp. 119–157). Hillsdale NJ: Erlbaum. Milton, J. (2009). Measuring second language vocabulary acquisition. Bristol: Multilingual Matters. Milton, J., & Hopkins, N. (2006). Comparing phonological and orthographic vocabulary size. Canadian Modern Language Review, 63(1), 127–147. Morais, J., Cary, L., Alegria, J., & Bertelson, P. (1979). Does awareness of speech as a sequence of phones arise spontaneously? Cognition, 50, 323–331. Morley, J. (2001). Aural comprehension instruction: Principles and practice. In M. Celce-Murcia (ed.). Teaching English as a second or foreign language (3rd edn., pp. 81–106). Boston, MA: Heinle & Heinle. Munby, J. (1978). Communicative syllabus design. Cambridge: Cambridge University Press. Munro, M. J., Derwing, T. M., & Morton, S. (2006). The mutual intelligibility of foreign accents. Studies in Second Language Acquisition, 28, 111–131. Nation, I. S. P. (2001). Learning vocabulary in another language. Cambridge: Cambridge University Press. Nuttall, C. (1996). Teaching reading skills in a foreign language. Oxford: Heinemann. Nygaard, L. C. (2008). Linguistic and nonlinguistic properties of speech. In D. B. Pisoni & R. E. Remez (eds.), The handbook of speech perception (pp. 390–413), Oxford: Blackwell. Nygaard, L. C., & Pisoni, D. B. (1995). Speech perception: New directions in research and theory. In J. L. Miller & P. D. Eimas (eds.), Speech, language and communication (pp. 63–96). San Diego, CA: Academic Press. Ockey, G., & French, R. (2016). From one to multiple accents on a test of L2 listening comprehension. Applied Linguistics, 37(5), 693–715. O’Malley, J. M., & Chamot, U. (1990). Learning strategies in second language acquisition. Cambridge: Cambridge University Press. Palmeri, T. J., Goldinger, S. D., & Pisoni, D. B. (1993). Episodic encoding of voice attributes and recognition memory for spoken words. Journal of Experimental Psychology: Learning, Memory and Cognition, 19, 309–328. Pashler, H., & Johnston, J.C. (1998). Attentional limitations in dual task performance. In H. Pashler (ed.), Attention (pp. 155–189). Hove: Psychology Press. Pemberton, R. (2003). Spoken Word Recognition in a Second Language. Hong Kong: HK University of Science and Technology. Pierrehumbert, J. (2001). Exemplar dynamics: Word frequency, lenition and contrast. In J. Bybee & P. Hopper (eds.), Frequency effects and emergent grammar (pp. 137–158). Amsterdam: John Benjamins. Piske, T. (2007). Implications of James E. Flege’s research for the foreign language classroom. In M. J. Munro & O.-S. Bohn (eds.), Second language

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

317

speech learning: The role of language experience in speech perception and production (pp. 301–314). Amsterdam: John Benjamins. Pisoni, D. B. (1997). Some thoughts on “normalization” in speech perception. In K. Johnson & J. W. Mullennix (eds.), Talker variability in speech processing (pp. 33–66). San Diego, CA: Academic Press, Pisoni, D. B., & Luce, P. A. (1987). Acoustic-phonetic representation in word recognition. In U. H. Frauenfelder & L. K. Tyler (eds.), Spoken word recognition (pp. 21–52). Cambridge, MA: MIT Press. Richards, J. (1983). Listening comprehension: Approach, design, procedure. TESOL Quarterly, 17, 219–239. Rivers, W. (1966). Listening comprehension. The Modern Language Journal, 50(4), 196–204. Rost, M. (1992). Listening in language learning. Harlow: Longman. Rost, M. (2002). Teaching and researching listening. Harlow: Pearson. Rost, M., & Ross, S. (1991). Learner use of strategies in interaction: Typology and teachabilty. Language Learning, 41, 235–273. Rubin, J. (1994). A review of second language listening comprehension research. The Modern Language Journal, 78(2), 199–211. Ruhm, R., Leitner-Jones, C., Kulmhofer, A., Kiefer, T., Mlakar, H., & Itzlinger-Bruneforth, U. (2016). Playing the recording once or twice: Effects on listening test performances, International Journal of Listening, 30(1–2), 67–83. Salisbury, K. (2005). The edge of expertise: Towards an understanding of listening test item writing as professional practice. Unpublished PhD dissertation, Kings College London. Sebastián-Gallés, N. (2008). Cross language speech perception. In D. B. Pisoni & R. E. Remez (eds.), The handbook of speech perception (pp. 546–566). Oxford: Blackwell. Segalowitz, N. (2010). Cognitive bases of second language fluency. New York: Routledge. Segalowitz, N. (2016). Automaticity. In P. Robinson (ed.), The Routledge encyclopaedia of second language acquisition (pp. 53–57). Abingdon: Routledge. Sherman, J. (1997). The effect of question preview in listening comprehension tests. Language Testing, 14, 185–213. Shockey, L. (2003). Sound patterns of spoken English. Oxford: Blackwell. Staehr, L. S. (2009). Vocabulary knowledge and advanced listening comprehension in English as foreign language. Studies in Second Language Acquisition, 31(4), 577–607. Strange, W. (1999). Perception of vowels. In J. M. Pickett (ed.), The acoustics of speech communication (pp. 153–165). Needham Heights, MA: Alleyn & Bacon. Strange, W., & Shafer, V. L. (2008). Speech perception in second language learners: The re-education of selective perception. In J. G. Hansen Edwards & M. L. Zampini (eds.), Phonology and second language acquisition (pp. 153–191). Amsterdam: John Benjamins.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

318

John Field

Sueyoshi, A., & Hardison, D. M. (2005). The role of gesture and facial cues in second language listening comprehension. Language Learning, 55, 661–699. Suvorov, R. (2009). Context visuals in L2 listening tests: The effects of photographs and video vs. audio-only format. In C. A. Chapelle, H. G. Jun, & I. Katz (eds.), Developing and evaluating language learning materials. Ames, IA: Iowa State University. Suvorov, R. (2015). Interacting with visuals in L2 listening tests: An eye-tracking study. Report: British Council ARAGS Reports, London. Retrieved from https://www.britishcouncil.org/sites/default/files/interacting_with_­ visuals_in_l2_listening_tests_suvorov.pdf. Tanenhaus, M. K., & Trueswell, J. C. (1995). Sentence comprehension. In J. L. Miller & P. D. Eimas (eds.), Speech language and communication (pp. 217–262). San Diego, CA: Academic Press. Taylor, L. (2006). The changing landscape of English: Implications for language assessment. ELT Journal, 60(1), 51–60. Teichert, H. U. (1996). A comparative study using illustrations, brainstorming and questions as advance organizers. Modern Language Review, 80(4), 509–517. Tsui, A., & Fullilove, J. (1998). Bottom-up or top-down processing as a discriminator of L2 listening performance. Applied Linguistics, 19, 432–451. Ur, P. (1984). Teaching listening comprehension. Cambridge: Cambridge University Press. Van Zeeland, H., & Schmitt, N. (2013). Lexical coverage in L1 and L2 listening comprehension: The same or different from reading comprehension? Applied Linguistics, 34(4), 457–479. Vandergrift, L. (2004). Listening to learn or learning to listen? Annual Review of Applied Linguistics, 24, 3–25. Vandergrift, L. (2006). Second language listening: Listening ability or language proficiency? The Modern Language Journal, 90, 6–18. Vandergrift, L., & Goh, C. C. M. (2012). Teaching and learning second language listening. New York: Routledge. Vandergrift, L., Goh, C., Mareschal, C., & Tafaghodtari, M. H. (2006). The metacognitive awareness listening questionnaire: Development and validation. The Modern Language Journal, 90, 6–18. Vandergrift, L., & Tafaghodtari, M. H. (2010). Teaching learners how to listen does make a difference. Language Learning, 65, 470–497. Vogely, A. (1995). Perceived strategy use during performance on three authentic listening comprehension tasks. The Modern Language Journal, 79(1), 41–56. Vogely, A. (1999). Addressing listening comprehension anxiety. In D. J. Young (ed.), Affect in foreign language and second language learning (pp. 106– 123). New York: McGraw-Hill.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

Second Language Listening: Current Ideas, Current Issues

319

Wagner, E., & Toth, P. A. (2014). Teaching and testing L2 Spanish listening using scripted vs unscripted texts. Foreign Language Annals, 47(3), 404–422. Wesche, M. B., & Paribakht, T. S. (2010). Lexical inferencing in a first and second language. Bristol: Multilingual Matters. Wray, A. (2002). Formulaic language and the lexicon. Cambridge: Cambridge University Press. Wu, W. M., & Stansfield, C. W. (2001). Towards authenticity of task in test development. Language Testing, 18(2), 187–206. Zhang, X. (2013). Foreign language listening anxiety and listening performance: Conceptualizations and causal relationships. System, 41(1), 164–177.

Downloaded from https://www.cambridge.org/core. Columbia University Libraries, on 08 Aug 2019 at 02:49:50, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.013

13 Contemporary Perspectives on L2 Upper-Register Text Processing Elizabeth B. Bernhardt and Cici Malik Leffell 13.1 Introduction Second-language (L2) reading research is replete with intellectual tensions: reading is the second-language skill with the greatest durability and usability and yet it is the most under-researched of the second language areas. While rooms could be filled with volumes and articles on how oral language develops and how that knowledge can be brought into instruction, on input and output hypotheses about the active use of the oral portion of second-language acquisition, or on the importance of pragmatic functions of language and how learners acquire them, the same cannot be said of knowledge about how learners interact with written text. A further paradoxical element is introduced by the notion of linguistic environment. While the environment of oral language appears to be most salient, the environment of written language is both more substantial and more linguistically complex. Oral language forms are much more simple than written forms, and they are far more lexically constrained than written language. An additional tension is introduced by the actual function of written material in the modern world. Adults use literacy in the workplace, be it a classroom, laboratory, or retail store. Reading complicated written material, seeking, as Guthrie and Greaney (1991) note, “knowledge gain … personal empowerment … participation in society … or occupational effectiveness” (pp. 73–75) is of paramount importance to functioning in the modern world, whether from a piece of paper or from a screen. Yet, ironically, the field has made little progress in understanding this process from a second-language perspective. Finally, the field of second-language reading research has failed to confront difficulty levels of text. Difficulty resides not exclusively in linguistic forms, but in the conceptual level of the material often linked to the professional background of the reader.

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

Contemporary Perspectives on L2 Upper-Register Text Processing

321

Difficult text is rarely generic, but rather nuanced and referential, connected to networks of information with which a second-language reader may or may not be familiar. The purpose of this paper is to bring awareness to very serious issues that need to be investigated and resolved in order to ensure more successful comprehension processes for advanced second-language readers. After defining advanced-register reading and arguing for a “state of the field” that synthesizes perspectives on various dimensions of second-language reading research and theory, the paper will concentrate on the research base focused on the comprehension of complicated, extended L2 discourse. It will mention the interactive compensatory framework as well as research methodologies that can provide insights into L2 upper-register text processing, and it will end with some guidance regarding research directions.

13.2  Defining Advanced Reading The issue of advanced register has indeed been of concern in first-language (L1) research with recent foci on literacy in the workplace and independent reading. A report centred on common goals for elementary and secondary school learners, Measures of Text Difficulty (Nelson et al., 2012), is helpful in probing advanced-register text. In its Appendix, the report synthesizes, first, quantitative measures of text complexity that include formulas examining word and sentence length as well as cohesion and word difficulty. While increasingly sophisticated as compared with traditional readability formulas, these measures nevertheless focus on concrete visible measures of text complexity. The report, second, synthesizes qualitative elements such as structure (graphics and other print features); whether the text uses “literal” language as opposed to “figurative, ironic” language; what background information the text presupposes; and whether the text is “literary” or “informational” (National Governors Association for Best Practices & Council of Chief State School Officers, 2010, p. 5). In spite of the helpful nature of the synthesis, it is not totally germane to the topic at hand. That synthesis is focused on “texts representative of those required in typical first-year credit-bearing college courses and in workforce training programs” (National Governors Association for Best Practices & Council of Chief State School Officers, 2010, p. 4). These types of texts are essentially instructional in nature, meant for learners. But what of learners who must learn to be independent and to handle large quantities of text for their professional lives in a second language? Two relevant sources are available. The first is offered as part of the North American codification of proficiency published by the Interagency

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

322

E l i z a bet h B . B er n h a r d t a n d C i c i M a l i k L e f f ell

Roundtable. That system focuses often on text types that define Advanced Professional Proficiency, but includes a number of reader-oriented descriptors. The system notes in particular that [T]he individual’s experience with the written language is extensive enough that he/she is able to relate inferences in the text to real-world knowledge and understand almost all sociolinguistic and cultural references [and] the intent of writers’ use of nuance and subtlety. (Interagency Language Roundtable, n.d., para. 13)

The descriptors also indicate that the advanced reader is able to handle the unpredictability often found in literary and journalistic essays and is able to “understand the full ramifications of texts as they are situated in the wider cultural, political, or social environment” (Interagency Language Roundtable, n.d., para. 13). A second relevant perspective is offered by Bernhardt (2011). She notes: complexity lies in the processing of intricate, complicated and often times obscure linguistic and cultural features accurately while trying to comprehend content and while remaining distant from it in order to assess the content’s value and accuracy. To expand on the point, learning to read in the upper registers of a second language entails being able to process the minutiae of word and grammatical nuance while constructing a message and simultaneously remaining aloof from that construction in order to assess its content and intention. (pp. 19–20)

To concretize the challenge, the following paragraph taken from a London Review of Books review might be helpful: Speer was lucky not just in his looks. His life was a succession of felicitous opportunities which came his way without obvious effort. Of course he strove to make his luck. But he had too great a sense of entitlement to allow himself to be seen to scramble for preferment. He gave the impression that he was the insouciant recipient of chance’s beneficence. He was, however, genuinely lucky to be rejected as a student by the great Expressionist Hans Poelzig. He was obliged to study, initially reluctantly, under the super-twee Heinrich Tessenow, whose precursively völkisch, Arts and Crafts saccharine he was indifferent to, even though it would become the almost invariable idiom of bucolic settlements, agrarian expansion and, on a larger scale, of the Ordensburgen and Napola—schools for the Nazi elite. Unastonishingly, Tessenow was keenly anti-modernist, going on Luddite. Speer was impressed. He acquired a cast of mind and a gamut of tastes which would soon allow him to ingratiate himself with Hitler, whose ranting in a Berlin beerhall prompted him to join the National Socialists in January 1931, two years before they seized power. During that period Speer became, faute de mieux, the impecunious party’s interior designer of choice. His initial willingness to work without a fee eased his way. (Meades, 2016, para. 3)

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

Contemporary Perspectives on L2 Upper-Register Text Processing

323

The vocabulary level itself is daunting. Preferment, insouciant, super-twee, Luddite, faute de mieux, and impecunious are all words, according to MerriamWebster’s online dictionary, at the lower end of the 30,000 most common words in English. Syntax, too, is formidable, with a combination of extremely short sentences, He was impressed, to the extraordinarily long middle sentence regarding Speer’s teachers. Further, the level of German cultural knowledge, whether one immediately identifies Albert Speer as Speer, combined with a knowledge of art and architectural movements, lends a level of befuddling wonderment in trying to understand the brief passage if that information is missing. Finally, the challenge for the reader is to identify the paragraph as a character sketch within a 5,000word essay. How a second-language reader learns to perform tasks such as the above is unclear, and operationalizing all of the variables entailed in text complexity for research purposes is daunting. Quantitative measures such as word and sentence length remain important. Yet, unseen variables such as the academic, professional level of background knowledge of the reader, as well as subtlety and nuance within the text, are critical toward understanding what second language readers must cope with. Admittedly, the above text might provide an exaggeration for effect, but nevertheless it is critical to recognize that the task faced by second language readers when they are to use text in professional-level settings is something that deserves careful attention. Indeed, these are the challenges that all readers—native and non-native alike—face. The key difference is that non-native readers have potentially shallower linguistic and cultural resources. Programming texts so that learners feel simply confident and comfortable will not help the field approach a set of very serious issues.

13.3  The Research Landscape: 1991–2008 The last twenty-five years have seen the publication of a number of significant research reviews dedicated exclusively to the research in the study of second-language reading (Bernhardt, 1991; 2000; 2005; 2011). Consolidated in these essays and volumes are research findings that have remained relatively stable over the half-century as well as those that have highlighted newer emerging research areas. In 1991, more than 140 studies were considered that focused specifically on adult second-language reading. These studies were culled from Applied Linguistics; Canadian Modern Language Review; English Language Teaching Journal; Foreign Language Annals; French Review; Hispania; International Review of Applied Linguistics in Language Teaching (IRAL); Language Learning;

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

324

E l i z a bet h B . B er n h a r d t a n d C i c i M a l i k L e f f ell

Language Testing; The Modern Language Journal; Reading in a Foreign Language; RELC Journal; System; Studies in Second Language Acquisition; TESOL Quarterly; Die Unterrichtspraxis; Working Papers in Linguistics and Language Teaching; and On TESOL, and from a number of L1-oriented journals such as Reading Research Quarterly and Journal of Literacy Behavior. Those studies focused on the interest areas of word recognition; background knowledge; text structure, including authentic text modification; oral and aural dimensions of reading, including reading aloud; syntax and morphology; crosslingual or L1–L2 relationships; metacognition and affect; testing reading comprehension; and instruction. The largest portion of the studies emphasized cultural background knowledge, a key element in 1980s L1 research (Pearson et al., 1984). Another significant portion explored the vagaries of different assessment mechanisms—key elements in a new and burgeoning research area. A third examined crosslingual dimensions of the second language reading process. At the time, the specification of research variables, as well as descriptions of subject groups, literacy levels, materials used, and proficiency level in the second language were found wanting. The database was heavily influenced by first language models and theories which perforce did not consider two languages in simultaneous operation. In a retrospective and then prospective analysis, Bernhardt (2000) reviewed the decade of 1990–2000 and documented that “a number of areas remained consistent instances of research interest: affective factors, text structure, syntactic features, and word knowledge and instruction” (p. 798). A new research area, the relationship of reading to other modalities such as writing and listening, appeared. While affective factors became focused on the use of literary text and attitudes of learners toward such texts, text structure and its modification often led to research findings that shortened, abridged texts caused readers greater comprehension difficulties rather than fewer. Syntactic features examined phenomena such as reference and anaphoric resolution. Word knowledge research examined how and when learners choose to use dictionaries and whether particular kinds of dictionary displays were efficacious. Instruction focused often on the use of sustained free reading of authentic texts. The issue of language of assessment remained. Important studies clarified that reader performance was different based on the language in which comprehension questions were posed. A significant addition to the research area was a substantial emphasis on the L1–L2 relationship, which tried to resolve the question of whether a specific level of grammatical knowledge would be necessary before comprehension could begin. Articulated by Alderson and Urquhart (1984) as the language-versus-reading problem, a set of studies enabled two conclusions: that indeed language knowledge was a larger element than literacy knowledge and that these two features were intertwined

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

Contemporary Perspectives on L2 Upper-Register Text Processing

325

and interdependent (Bernhardt & Kamil, 1995). While one was larger (language proficiency), neither (L2 proficiency nor L1 literacy) was more important than the other; they co-exist. This review led to an expanded view of second-language reading and dimensions of second-language reading research that critically included the contribution of L1 literacy to any second-language literacy performance. In spite of moving the field forward with this definitive insight, the underspecification of variables in second-language reading research studies remained, continuing into the early 1990s. At that time, Bernhardt (2005) posited a theoretical model that emphasized the developmental impact of L1 literacy and L2 language knowledge on the reading comprehension process. A further review followed that employed the same methodology for culling articles as the 1991 and 2000 reviews. Bernhardt (2011) expanded the number of journals reviewed to include Research and Reading Instruction; ADFL Bulletin; CALICO Journal; Journal of Research in Reading; Reading and Writing; Second Language Research; Journal of Adolescent and Adult Literacy; Reading in a Foreign Language; and The Reading Matrix. This process yielded a pool of 250+ research studies. Critically, seated in the 2005 model, the review included an analysis of the potential acknowledgement of the influence of L1 literacy and L2 language knowledge in the database, conceptualized as the interactive or compensatory model of L2 Reading. Superseding psycholinguistic models of the reading process, this model entailed processing in multiple dimensions, not hierarchical in nature. Most of the topics of concern in the prior twenty years remained of interest. Background knowledge; technology use; how good and poor readers employ reading strategies; and assessment were also explored. Intrapersonal variables as well as transfer or L1–L2 relationships were included. Word recognition and phonological processing also remained of research interest, as did instruction, vocabulary gain from text, and dictionary use. Studies of morphosyntax as well as of genre and text features were also found within the database, but indicated minor if any impacts on comprehension.

13.4  The Research Landscape: 2008 and Beyond Using a methodology similar to those in previous reviews, that is, searching through the 2008 to 2016 volumes of the above-mentioned journals (including also The International Journal of Applied Linguistics & English Literature) for studies that examine readers of high school age and beyond, articles were first collated and then categorized by main theme or thrust, by the number of subjects and educational level, and by the first and second languages of interest. This process yielded a total of 332 research studies forming the current L2 reading landscape. In brief, the present

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

326

E l i z a bet h B . B er n h a r d t a n d C i c i M a l i k L e f f ell

landscape illustrates a similar research interest in topics that have been widely covered in the research literature over the years: a consistent interest in L1 and L2 relationships; in metacognition; in strategies and affective features of second language reading; and in vocabulary knowledge. Structural features of text, instruction, and technology also grew in interest. Some of the research areas explored across the decades still nominally exist, such as word recognition, background knowledge, and assessment. A key difference is the explosion of interest in extensive reading. To try to answer the question of the present review regarding complicated, extended L2 discourse, the 332 studies were queried for any findings that reported on reading comprehension of complicated, upper-register text. Sixty-one (18%) of the articles involved what could be considered advanced-level text but only eighteen (5% of the total) investigated reading comprehension of such texts. The largest group of studies encountered investigates metacognitive aspects of second language reading, and the vast majority of these focus on reading strategies. Only ten of the fifty-one total studies on metacognition feature participants who were presented with authentic texts aimed at adult readers or with what may be considered upper-register texts. Cheng (2008) had one doctoral candidate in electrical engineering (L1 Chinese) read authentic research articles and texts of other genres in English and observed that the student employed rhetorical and evaluative analysis to decipher genre. Wurr, Theurer, and Kim (2008/2009) presented three adult English language learners (ELLs) with readings from a student and popular anthology; they found retrospective miscue analysis to increase strategy awareness. Bell (2008a, 2008b, 2011) looked at framing by conducting three experiments on small groups of graduate student ELLs in Australia. She found that learners employ extratextual framing to adapt prior knowledge and reading behaviours to present context (2008a), that they undergo significant changes in the self-awareness and intertextual framing practices as they acclimate to a new culture (2008b), and that they employ intertextual framing skills to an increasing extent over the course of an academic year, which suggests greater metacognition (2011). The author used 750-word general interest pieces from a scientific magazine for all three of these studies. Urlaub (2013) found US university students of German to perceive strategy training as helpful for reading authentic literary texts, and Kasemsap and Lee (2015), in addition to using books intended for children, presented participants with excerpts from English language newspapers to help them conclude that language proficiency strongly predicts retrieval strategy use. Based on two English as a second language (ESL) undergraduates’ reading of “a model synthesis paper”, Zhao and Hirvela (2015) vouch that learners’ rhetorical reading strategy skills and their knowledge of synthesis and sources are important for synthesizing in L2. Of the ten articles on metacognition that use what

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

Contemporary Perspectives on L2 Upper-Register Text Processing

327

could be considered more complicated, advanced texts, only two investigate reading comprehension of them. Horiba (2013) tested participants with upper-register texts; she performed two experiments involving persuasive essays (356–433 words) in the ki-sho-ten-ketsu style (p. 101) from a Japanese newspaper. Findings indicate that instructions may encourage strategic L2 processing and that goal type affects the extent to which L2 proficiency and comprehension abilities participate in processing. Karimi (2015) presented seventy-six ELLs (L1 Persian) with expository texts “ranging from semi-technical to general interest” (p. 43) in genre (between 788 and 1562 words) and concluded that strategy-based instruction positively impacts strategy use and multiple documents comprehension. Vocabulary is the principal interest of the second-largest group of studies. Of the forty-six studies on vocabulary, only nine feature what might be considered complicated, advanced-level text, and only two of these investigate comprehension of such texts. Min (2008) presented undergraduates with four articles on two themes and stresses the importance of focused vocabulary exercises. Prichard (2008) looks into various aspects of dictionary use with different sorts of L2 readers in mind. He gave participants two news pieces (of 382 and 1120 words each) and a book section of 420 words and found that dictionary use may speed up students’ improvements in comprehension and vocabulary acquisition, that certain L2 learners may profit from learning how to look up words in a targeted manner, and that others need instruction in basic reading strategies before benefiting from dictionary use or instruction. Peters et al. (2009) used an authentic newspaper article in their experiment with Dutch-speaking learners of German; they argue that giving students tasks that require them to look up and process new words is a way to enhance vocabulary acquisition through reading. After reading the first eight chapters of a nineteenth-century novel, the English majors in Daskalovska’s (2014) Macedonian study showed that vocabulary size does not affect rate of acquisition. Elgort and Warren (2014) presented adult ELLs with embedded pseudowords in a long informational text and learned that particular learner characteristics and aspects of the text in question combine to affect L2 vocabulary learning through reading. The thirty-three native English speakers and fifty-nine ELLs in Reynolds’ (2014) study read a novel to reveal that native speakers incidentally obtain vocabulary through reading fiction more effectively than L2 learners, and Ender’s (2016) subjects’ reading of a newspaper article and an excerpt from a 1947 French novel suggests that implicit cognitive processes may more greatly benefit vocabulary development than deliberate, explicit processing strategies. Of the nine total papers on vocabulary in second language reading that incorporate upper-register texts, only two investigate reading comprehension. Alessi and Dwyer (2008) presented seventy-six US university students

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

328

E l i z a bet h B . B er n h a r d t a n d C i c i M a l i k L e f f ell

of Spanish with an article from a Spanish-language news website, its 431 words divided among three pages, and drew conclusions on vocabulary support: namely, that vocabulary support is more beneficial when simultaneous to, not before, reading, and that focusing on vocabulary before reading increases reading speed while help with vocabulary during reading enhances comprehension. Schmitt, Jiang, and Grabe (2011), for their part, base their findings of how 661 ELLs (aged 16–33) read a 757-word passage from a textbook as well as a 582-word reading from an Englishlanguage magazine; they assert that the amount of known words in a given text correlates with reading comprehension. The third most researched second-language area in the most recent database is extensive reading. These forty articles, roughly twelve per cent of all studies queried, generally report on the ways in which the practice benefits language learners’ motivation and perceptions. The vast majority of studies on extensive reading involve subjects who read graded readers, textbook passages, or other simplified texts. Interestingly, when authentic texts are used, they are commonly at a level intended for children, teenagers, or young adults, even when the study participants are postsecondary learners. Only a handful of researchers present their participants with authentic texts aimed at adult readers: Bordonaro (2011) encouraged participants to access authentic periodical and literary texts online; Beglar and Hunt (2014) had participants read authentic novels in addition to graded readers and other simplified texts; Ghiabi (2014) asked participants to read a short novel; and Pereyra (2015) instructed students to peruse journals, magazines, and a novel as well as graded readers. Only two authors from this group looked at reading comprehension of authentic texts intended for adult readers: Beglar, Hunt, and Kite (2012) chose three novels in addition to works from a collection of simplified texts, and Fernandez de Morgado (2009) had students find their own scientific articles to create a reading library from which the class could select readings. Beglar, Hunt, and Kite (2012) found extensive reading to improve reading comprehension abilities, especially when learners read simplified texts, and Fernandez de Morgado’s study results indicate that extensive reading does not significantly impact reading comprehension but that students view the practice as beneficial. Studies on the relationship between L1 and L2 often focus on differences between native speakers and L2 learners. The majority of studies in the group either get their data from participants working with readings from standardized tests or set out research questions that require that their subjects process smaller units of text, such as characters, phonemes, words, or individual sentences. Only two researchers in the cross-lingual group pull data from learners working with advanced text, and both of them touch upon comprehension in their findings. Alptekin and Erçetin

Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 01 Aug 2019 at 17:22:54, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108333603.014

Contemporary Perspectives on L2 Upper-Register Text Processing

329

(2010) presented native Turkish speakers training to become English teachers with a 2270-word autobiographical short story and observed that task processing is more precise in L1 than in L2 but that stora