Applied Linguistics, Volume 31, issue 5, 2010

APPLIN-31(5)Cover.qxd 10/7/10 2:51 PM Page 1 Articles REVIEWS Roy Harris and Christopher Hutton: De®nition in Theo...

Author: Oxford University

24 downloads 1425 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

APPLIN-31(5)Cover.qxd

10/7/10

2:51 PM

Page 1

Articles

REVIEWS

Roy Harris and Christopher Hutton: De®nition in Theory and Practice: Language, Lexicography and the Law NATALIA D. JACOBSEN Philip Seargeant: The Idea of English in Japan: Ideology and the Evolution of a Global Language DAVID COZY Ken Hyland and Giuliana Diani (eds): Academic Evaluation: Review Genres in University Settings JOHN M. SWALES Brian Richards, Michael H. Daller, David D. Malvern, Paul Meara, James Milton and Jeanine Treffers-Daller: Vocabulary Studies in First and Second Language Acquisition DALE BROWN

Corrigendum Notes on Contributors

Volume 31 Number 5 December 2010

Extended, Embodied Cognition and Second Language Acquisition DWIGHT ATKINSON Partial Word Form Learning in the Written Mode in L2 German and Spanish JOE BARCROFT and SUSANNE ROTT Language Policy and Latina Immigrants: An Analysis of Personal Experience and Identity in Interview Talk KENDALL A. KING and ANNA DE FINA NNES Doctoral Students in English-Speaking Academe: The Nexus between Language and Discipline YU-JUNG CHANG and YASUKO KANNO The Effects of Repetition and L1 Lexicalization on Incidental Vocabulary Acquisition CHUNTIEN CHEN and JOHN TRUSCOTT The Branding of English and The Culture of the New Capitalism: Representations of the World of Work in English Language Textbooks JOHN GRAY

APPLI ED LI NGUISTICS

CONTENTS

ISSN 0142-6001 (PRINT) ISSN 1477-450X (ONLINE)

Applied Linguistics Volume 31 Number 5 December 2010

Published in cooperation with AAAL American Association for Applied Linguistics AILA International Association of Applied Linguistics BAAL British Association for Applied Linguistics

OXFORD

www.applij.oxfordjournals.org


10/7/10

2:51 PM

Page 2

EDITORS

NOTES TO CONTRIBUTORS

Ken Hyland, Director, Centre for Applied English Studies, KK Leung Building, The University of Hong Kong, Pokfulam Road, Hong Kong Jane Zuengler, Nancy C. Hoefs Professor of English, University of Wisconsin-Madison, 6103 Helen C. White 600 North Park Street Madison, WI, 53706 USA Assistant to Jane Zuengler: Heather Carroll, University of Wisconsin-Madison

Articles submitted to Applied Linguistics should represent outstanding scholarship and make original contributions to the field. The Editors will assume that an article submitted for their consideration has not previously been published and is not being considered for publication elsewhere, either in the submitted form or in a modified version. Articles must be written in English and not include libelous or defamatory material. Manuscripts accepted for publication must not exceed 8,000 words including all material for publication in the print version of the article, except for the abstract, which should be no longer than 175 words. Additional material can be made available in the online version of the article. Such additions will be indexed in the print copy.

REVIEWS AND FORUM EDITOR Stef Slembrouck, Professor of English Linguistics and Discourse Analysis, Universiteit Gent, Vakgroep Engels, Rozier 44, B-9000 Gent, Belgium. <[email protected]> Assistant to Stef Slembrouck: Tine Defour, Universiteit Gent

ADVISORY BOARD Guy Cook, British Association for Applied Linguistics Aneta Pavlenko, American Association for Applied Linguistics Martin Bygate, International Association for Applied Linguistics Huw Price, Oxford University Press

EDITORIAL PANEL Karin Aronsson, Linko¨ping University David Block, London University Institute of Education Jan Blommaert, University of Jyva¨skyla¨ Deborah Cameron, University of Oxford Lynne Cameron, Open University (BAAL Representative) Tracey Derwing, University of Alberta Zoltań Do¨rnyei, University of Nottingham Patricia Duff, University of British Columbia Diana Eades, University of New England, Australia ZhaoHong Han, Columbia University (AAAL representative) Gabriele Kasper, University of Hawai’i at Manoa Claire Kramsch, University of California at Berkeley Angel Lin, University of Hong Kong Janet Maybin, Open University, UK Tim McNamara, University of Melbourne Junko Mori, University of Wisconsin-Madison Greg Myers, Lancaster University Susanne Niemeier, University Koblenz-Landau (AILA Representative) Lourdes Ortega, University of Hawai’i at Manoa Alastair Pennycook, University of Technology, Sydney Ben Rampton, King’s College, University of London Steven Ross, Kwansei Gakuin University Alison Sealey, University of Birmingham Antonella Sorace, University of Edinburgh Lionel Wee, National University of Singapore Applied Linguistics is published five times a year in February, May, July, September and December by Oxford University Press, Oxford, UK. Annual subscription price is £254/US$457/E381. Applied Linguistics is distributed by Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Periodicals postage paid at Rahway, NJ and at additional entry points. US Postmaster: send address changes to Applied Linguistics (ISSN 0142-6001), c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. # Oxford University Press 2010 All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior written permission of the Publishers, or a licence permitting restricted copying issued in the UK by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 9HE, or in the USA by the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. Typeset by Glyph International, Bangalore, India Printed by Bell and Bain Ltd, Glasgow

Applied Linguistics operates a double-blind peer review process. To facilitate this process, authors are requested to ensure that all submissions, whether first or revised versions, are anonymized. Authors’ names and institutional affiliations should appear only on a detachable cover sheet. Submitted manuscripts will not normally be returned. Forum pieces are usually reviewed by the journal Editors and are not sent for external review. Items for the Forum section are normally 2,000 words long. Contributions to the Forum section and offers to review book publications should be addressed to the Forum and Reviews Editor. For more detailed guidelines, see our website http://www.oxfordjournals.org/applij/for_authors/index.html

PROOFS Proofs will be sent to the author for correction, and should be returned to Oxford University Press by the deadline given.

OFFPRINTS On publication of the relevant issue, if a completed offprint form has been received stating gratis offprints are requested, 25 offprints of an article, forum piece or book review will be sent to the authors free of charge. Orders from the UK will be subject to a 17.5% VAT charge. For orders from elsewhere in the EU you or your institution should account for VAT by way of a reverse charge. Please provide us with your or your institution’s VAT number.

COPYRIGHT Acceptance of an author’s copyright material is on the understanding that it has been assigned to the Oxford University Press subject to the following conditions. Authors are free to use their articles in subsequent publications written or edited by themselves, provided that acknowledgement is made of Applied Linguistics as the place of original publication. Except for brief extracts the Oxford University Press will not give permission to a third party to reproduce material from an article unless two months have elapsed without response from the authors after the relevant application has been made to them. It is the responsibility of the author to obtain permission to reproduce extracts, figures, or tables from other works.

Applied Linguistics Journal online The full text of Applied Linguistics is available online to journal subscribers. Online access has a number of advantages: . quality PDFs ensure articles look the same as the print original and are easy to print out . access is easy—all you need is your subscription number or institutional IP address (see below) . online access is available ahead of print publication—so view while you await your print version! . access the text wherever you are (or from any part of your institution network if you have a library subscription) . perform searches by word or author across the full text of the articles of any part of the journal . download articles whenever you choose—you will be able to access past online issues as long as you have a current subscription . free sample copy available online . fully searchable abstracts/titles going back to volume 1 . Table of Contents email alerting service. The print version will continue to be available as previously. Institutions may choose to subscribe to the print edition only, online only, or both. Individual subscribers automatically receive both.

CONTRIBUTORS There is no need for contributors to format their articles any differently; online files are produced automatically from the final page proofs of the journal. However, if you know that an item in your list of references is available online, please supply the URL. If you have your own website, you are welcome to include the URL with your contact address in your biodata.

ADVANCE ACCESS Applied Linguistics now has Advance Access articles. These are papers that have been copyedited and typeset but not yet paginated for inclusion in an issue of the journal. More information, including how to cite Advance Access papers, can be found online at http://www.applij.oxfordjournals.org.

Applied Linguistics Subscription Information

A subscription to Applied Linguistics comprises 5 issues. Annual Subscription Rate (Volume 31, 5 issues, 2010) Institutional. Print edition and site-wide online access: £254.00/US$457.00/E381.00; Print edition only: £233.00/US$419.00/E350.00; Site-wide online access only: £212.00/US$382.00/E318.00.

Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, UK. Email: [email protected]. Tel (and answerphone outside normal working hours): +44 (0)1865 353907. Fax: + 44 (0)1865 353485. In Japan, please contact: Journals Customer Services, Oxford Journals, Oxford University Press, Tokyo, 4-5-10-8F Shiba, Minato-ku, Tokyo 108-8386, Japan. Tel: þ81 3 5444 5858. Fax: þ81 3 3454 2929.

Personal. Print edition and individual online access: £82.00/US$164.00/E123.00.

subscribe to applied linguistics

Please note: US$ rate applies to US & Canada, Euros applies to Europe, UK£ applies to UK and Rest of World.

For new subscriptions and recent single issues only. Current subscribers will automatically receive a renewal form.

Prices include postage by surface mail, or for subscribers in the USA and Canada by airfreight, or in India, Japan, Australia and New Zealand, by Air Speeded Post. Airmail rates are available on request. There are other subscription rates available for members of AAAL, BAAL, AILA, and LSA, for a complete listing please visit www.applij.oxfordjournals.org/subscriptions. Full prepayment, in the correct currency, is required for all orders. Orders are regarded as firm and payments are not refundable. Subscriptions are accepted and entered on a complete volume basis. Claims cannot be considered more than FOUR months after publication or date of order, whichever is later. All subscriptions in Canada are subject to GST. Subscriptions in the EU may be subject to European VAT. If registered, please supply details to avoid unnecessary charges. For subscriptions that include online versions, a proportion of the subscription price may be subject to UK VAT. Personal rate subscriptions are only available if payment is made by personal cheque or credit card and delivery is to a private address. The current year and two previous years’ issues are available from Oxford Journals. Previous volumes can be obtained from the Periodicals Service Company at http://www.periodicals. com/oxford.html or Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Email: [email protected]. Tel: +1 (518) 537 4700. Fax: +1 (518) 537 5899. For further information, please contact: Journals Customer Service Department,

Please complete the form below and return it to: Journal Customer Service Department (please see above). Please record my subscription to Applied Linguistics, starting with Volume__________ (Subscriptions start with the March issue and can be accepted for complete volumes only.) Please send me the following single issue(s) Volume_________ Issue_________ Name (BLOCK CAPITALS please) _________________________________________ Adresss __________________________________ _________________________________________ _________________________________________ City _____________________________________ Country _________________________________ Postcode _________________________________ I enclose the correct payment of (see rates above): £/US/E __________________________________ Please debit my credit card: American Express / Mastercard / Visa (delete as appropriate) Card number: __|__|__|__|__|__|__|__|__|__|__|__|__|__|__|__|__ Expiry date: |__|__|__| Signature ________________________________ œ Please tick this box if you do NOT wish to receive details of related products and services of OUP and other companies that we think may be of interest.

Aims Applied Linguistics publishes research into language with relevance to real-world problems. The journal is keen to help make connections between fields, theories, research methods, and scholarly discourses, and welcomes contributions which critically reflect on current practices in applied linguistic research. It promotes scholarly and scientific discussion of issues that unite or divide scholars in applied linguistics. It is less interested in the ad hoc solution of particular problems and more interested in the handling of problems in a principled way by reference to theoretical studies. Applied linguistics is viewed not only as the relation between theory and practice, but also as the study of language and language-related problems in specific situations in which people use and learn languages. Within this framework the journal welcomes contributions in such areas of current enquiry as: bilingualism and multilingualism; computer-mediated communication; conversation analysis; corpus linguistics; critical discourse analysis; deaf linguistics; discourse analysis and pragmatics; first and additional language learning, teaching, and use; forensic linguistics; language assessment; language planning and policies; language for special purposes; lexicography; literacies; multimodal communication; rhetoric and stylistics; and translation. The journal welcomes both reports of original research and conceptual articles. The Journal’s Forum section is intended to enhance debate between authors and the wider community of applied linguists (see Editorial in 22/1) and affords a quicker turnaround time for short pieces. Forum pieces are typically responses to a published article, a shorter research note or report, or a commentary on research issues or professional practices. The Journal also contains a Reviews section. Applied Linguistics is covered by the following abstracting/indexing services: Bibliographie Linguistique/Linguistic Bibliography, BLonline, British Education Index, Current Index to Journals in Education, ERIC (Education Resources Information Centre), International Bibliography of the Social Sciences, ISI: Social Sciences Citation Index, Research Alert, Current Contents/Social and Behavioral Sciences, Social Scisearch, Sociological Abstracts: Language and Linguistics Behaviour Abstracts, Language Teaching, MLA Directory of Periodicals, MLA International Bibliography, PsycINFO, Sociological Abstracts, Zeitschrift fu¨r Germanistische Linguistik.

ADVERTISING Inquiries about advertising should be sent to Linda Hann, Oxford Journals Advertising, 60 Upper Broadmoor Road, Crowthorne, RG45 7DE, UK. Email: [email protected]. Tel/Fax: +44 (0)1344 779945.

PERMISSIONS For information on how to request permissions to reproduce articles/information from this journal, please visit www.oxfordjournals.org/permissions.

DISCLAIMER Statements of fact and opinion in the articles in Applied Linguistics are those of the respective authors and contributors and not of Applied Linguistics or Oxford University Press. Neither Oxford University Press nor Applied Linguistics make any representation, express or implied, in respect of the accuracy of the material in this journal and cannot accept any legal responsibility or liability for any errors or omissions that may be made. The reader should make his/her own evaluation as to the appropriateness or otherwise of any experimental technique described.


10/7/10

2:51 PM

Page 2

EDITORS

NOTES TO CONTRIBUTORS

Ken Hyland, Director, Centre for Applied English Studies, KK Leung Building, The University of Hong Kong, Pokfulam Road, Hong Kong Jane Zuengler, Nancy C. Hoefs Professor of English, University of Wisconsin-Madison, 6103 Helen C. White 600 North Park Street Madison, WI, 53706 USA Assistant to Jane Zuengler: Heather Carroll, University of Wisconsin-Madison

Articles submitted to Applied Linguistics should represent outstanding scholarship and make original contributions to the field. The Editors will assume that an article submitted for their consideration has not previously been published and is not being considered for publication elsewhere, either in the submitted form or in a modified version. Articles must be written in English and not include libelous or defamatory material. Manuscripts accepted for publication must not exceed 8,000 words including all material for publication in the print version of the article, except for the abstract, which should be no longer than 175 words. Additional material can be made available in the online version of the article. Such additions will be indexed in the print copy.

REVIEWS AND FORUM EDITOR Stef Slembrouck, Professor of English Linguistics and Discourse Analysis, Universiteit Gent, Vakgroep Engels, Rozier 44, B-9000 Gent, Belgium. <[email protected]> Assistant to Stef Slembrouck: Tine Defour, Universiteit Gent

ADVISORY BOARD Guy Cook, British Association for Applied Linguistics Aneta Pavlenko, American Association for Applied Linguistics Martin Bygate, International Association for Applied Linguistics Huw Price, Oxford University Press

EDITORIAL PANEL Karin Aronsson, Linko¨ping University David Block, London University Institute of Education Jan Blommaert, University of Jyva¨skyla¨ Deborah Cameron, University of Oxford Lynne Cameron, Open University (BAAL Representative) Tracey Derwing, University of Alberta Zoltań Do¨rnyei, University of Nottingham Patricia Duff, University of British Columbia Diana Eades, University of New England, Australia ZhaoHong Han, Columbia University (AAAL representative) Gabriele Kasper, University of Hawai’i at Manoa Claire Kramsch, University of California at Berkeley Angel Lin, University of Hong Kong Janet Maybin, Open University, UK Tim McNamara, University of Melbourne Junko Mori, University of Wisconsin-Madison Greg Myers, Lancaster University Susanne Niemeier, University Koblenz-Landau (AILA Representative) Lourdes Ortega, University of Hawai’i at Manoa Alastair Pennycook, University of Technology, Sydney Ben Rampton, King’s College, University of London Steven Ross, Kwansei Gakuin University Alison Sealey, University of Birmingham Antonella Sorace, University of Edinburgh Lionel Wee, National University of Singapore Applied Linguistics is published five times a year in February, May, July, September and December by Oxford University Press, Oxford, UK. Annual subscription price is £254/US$457/E381. Applied Linguistics is distributed by Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. Periodicals postage paid at Rahway, NJ and at additional entry points. US Postmaster: send address changes to Applied Linguistics (ISSN 0142-6001), c/o Mercury International, 365 Blair Road, Avenel, NJ 07001, USA. # Oxford University Press 2010 All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior written permission of the Publishers, or a licence permitting restricted copying issued in the UK by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 9HE, or in the USA by the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. Typeset by Glyph International, Bangalore, India Printed by Bell and Bain Ltd, Glasgow

Applied Linguistics operates a double-blind peer review process. To facilitate this process, authors are requested to ensure that all submissions, whether first or revised versions, are anonymized. Authors’ names and institutional affiliations should appear only on a detachable cover sheet. Submitted manuscripts will not normally be returned. Forum pieces are usually reviewed by the journal Editors and are not sent for external review. Items for the Forum section are normally 2,000 words long. Contributions to the Forum section and offers to review book publications should be addressed to the Forum and Reviews Editor. For more detailed guidelines, see our website http://www.oxfordjournals.org/applij/for_authors/index.html

PROOFS Proofs will be sent to the author for correction, and should be returned to Oxford University Press by the deadline given.

OFFPRINTS On publication of the relevant issue, if a completed offprint form has been received stating gratis offprints are requested, 25 offprints of an article, forum piece or book review will be sent to the authors free of charge. Orders from the UK will be subject to a 17.5% VAT charge. For orders from elsewhere in the EU you or your institution should account for VAT by way of a reverse charge. Please provide us with your or your institution’s VAT number.

COPYRIGHT Acceptance of an author’s copyright material is on the understanding that it has been assigned to the Oxford University Press subject to the following conditions. Authors are free to use their articles in subsequent publications written or edited by themselves, provided that acknowledgement is made of Applied Linguistics as the place of original publication. Except for brief extracts the Oxford University Press will not give permission to a third party to reproduce material from an article unless two months have elapsed without response from the authors after the relevant application has been made to them. It is the responsibility of the author to obtain permission to reproduce extracts, figures, or tables from other works.


10/7/10

2:51 PM

Page 1

Articles

REVIEWS

Roy Harris and Christopher Hutton: De®nition in Theory and Practice: Language, Lexicography and the Law NATALIA D. JACOBSEN Philip Seargeant: The Idea of English in Japan: Ideology and the Evolution of a Global Language DAVID COZY Ken Hyland and Giuliana Diani (eds): Academic Evaluation: Review Genres in University Settings JOHN M. SWALES Brian Richards, Michael H. Daller, David D. Malvern, Paul Meara, James Milton and Jeanine Treffers-Daller: Vocabulary Studies in First and Second Language Acquisition DALE BROWN

Corrigendum Notes on Contributors

Volume 31 Number 5 December 2010

Extended, Embodied Cognition and Second Language Acquisition DWIGHT ATKINSON Partial Word Form Learning in the Written Mode in L2 German and Spanish JOE BARCROFT and SUSANNE ROTT Language Policy and Latina Immigrants: An Analysis of Personal Experience and Identity in Interview Talk KENDALL A. KING and ANNA DE FINA NNES Doctoral Students in English-Speaking Academe: The Nexus between Language and Discipline YU-JUNG CHANG and YASUKO KANNO The Effects of Repetition and L1 Lexicalization on Incidental Vocabulary Acquisition CHUNTIEN CHEN and JOHN TRUSCOTT The Branding of English and The Culture of the New Capitalism: Representations of the World of Work in English Language Textbooks JOHN GRAY

APPLI ED LI NGUISTICS

CONTENTS

ISSN 0142-6001 (PRINT) ISSN 1477-450X (ONLINE)

Applied Linguistics Volume 31 Number 5 December 2010

Published in cooperation with AAAL American Association for Applied Linguistics AILA International Association of Applied Linguistics BAAL British Association for Applied Linguistics

OXFORD

www.applij.oxfordjournals.org

APPLIED LINGUISTICS Volume 31 Number 5 December 2010 CONTENTS Articles Extended, Embodied Cognition and Second Language Acquisition DWIGHT ATKINSON Partial Word Form Learning in the Written Mode in L2 German and Spanish JOE BARCROFT and SUSANNE ROTT Language Policy and Latina Immigrants: An Analysis of Personal Experience and Identity in Interview Talk KENDALL A. KING and ANNA DE FINA NNES Doctoral Students in English-Speaking Academe: The Nexus between Language and Discipline YU-JUNG CHANG and YASUKO KANNO The Effects of Repetition and L1 Lexicalization on Incidental Vocabulary Acquisition CHUNTIEN CHEN and JOHN TRUSCOTT The Branding of English and The Culture of the New Capitalism: Representations of the World of Work in English Language Textbooks JOHN GRAY REVIEWS Roy Harris and Christopher Hutton: Definition in Theory and Practice: Language, Lexicography and the Law NATALIA D. JACOBSEN Philip Seargeant: The Idea of English in Japan: Ideology and the Evolution of a Global Language DAVID COZY Ken Hyland and Giuliana Diani (eds): Academic Evaluation: Review Genres in University Settings JOHN M. SWALES Brian Richards, Michael H. Daller, David D. Malvern, Paul Meara, James Milton and Jeanine Treffers-Daller: Vocabulary Studies in First and Second Language Acquisition DALE BROWN

599

623

651

671

693

714

734

737

740

744

Corrigendum

748

NOTES ON CONTRIBUTORS

749

Applied Linguistics: 31/5: 599–622 ß Oxford University Press 2010 doi:10.1093/applin/amq009 Advance Access published on 19 March 2010

Extended, Embodied Cognition and Second Language Acquisition DWIGHT ATKINSON Department of English, Purdue University, Indiana, USA E-mail: [email protected]

A highly specific view of cognition—cognitivism—has dominated second language acquisition (SLA) studies. As one SLA pioneer describes it: Even in those early days, we believed that we were witnessing the birth of a new field—one that did not see language as behavior, one that no longer ignored the mind, one that put cognitivism squarely at the forefront of its explanations. As it turns out, it was a powerful birthright. It is fair to say that a cognitivist view has dominated the field ever since. (Larsen-Freeman 2007: 775) Yet conceptions of cognition have changed radically over the past century. Most recently, ‘a sea change in research and theory’ (Semin and Cacioppo 2008: 140) has occurred—toward extended and embodied views of cognition. Extended cognition conceptualizes mind/brain as inextricably tied to the external environment, while embodied cognition views cognitive activity as grounded in bodily states and action. These two approaches are related because bodies link minds to the world—we experience, understand, and act on the world through our bodies. As a result, extended and embodied cognition are sometimes grouped together. This article begins by briefly locating cognitivism in SLA studies. Next, extended and embodied views of cognition are reviewed and synthesized. An approach to SLA based on this synthesis—a sociocognitive approach—is then described in three principles. Finally, extended, embodied cognition-for-SLA is exemplified in a videotaped English as a Foreign Language (EFL) tutoring interaction.

Downloaded from applij.oxfordjournals.org by guest on December 31, 2010

A cognitivist approach to cognition has traditionally dominated second language acquisition (SLA) studies. In this article, I examine two alternative approaches— extended cognition and embodied cognition—for how they might help us conceptualize SLA. More specifically, I present: (i) summaries of extended and embodied cognition, followed by reasons why the two can be treated as a single, synthetic perspective; (ii) an approach to SLA grounded in an extended, embodied view of cognition—i.e. a sociocognitive approach—in three principles; and (iii) a naturally occurring example of extended, embodied cognition-for-SLA.

600 EXTENDED, EMBODIED COGNITION AND SECOND LANGUAGE ACQUISITION

COGNITION IN MAINSTREAM SLA STUDIES Following Descartes (1637/1960), cognitivism views the mind/brain as the self-sufficient source of cognition. In SLA studies, its locus classicus is Sharwood Smith’s (1991) declaration that while social context is the icing in L2 learning, cognition is the cake. This perspective, which has dominated the field, conceptualizes SLA in terms of what learners know—their internalized linguistic competence. In Doughty and Long’s influential view:

This is a paradigmatic statement of the mainstream, cognitivist view of SLA—one which recapitulates Chomsky’s radical severing of competence from performance and assumes the centrality of ‘representation and computation . . . [in] a pre-eminently cognitive, information-processing approach’ to SLA (Long and Doughty 2003: 867). It also suggests that SLA studies is a cognitive science.

EXTENDED COGNITION Yet cognitive science is many things, and to many people. Philosophers Clark and Chalmers (1998) coined the term ‘extended cognition’ to describe their distinctive take on what is variously called situated cognition, distributed cognition, grounded cognition, and active externalism. Based substantially on Clark’s work, I see four propositions as underlying the extended cognition thesis.

Cognition is environmental Instead of being the self-contained logical system posited by cognitivism (Boden 2006), cognition depends heavily on the external environment. Mundane examples abound: We wake, sleep, eat, and shift activities according to clocks and calendars; do math with calculators or on paper; leave our keys by the door on returning home so we can find them later on; refrigerate food in see-through containers; store cleaning supplies under the sink; arrange


Much current SLA research and theorizing shares a strongly cognitive orientation . . . The focus is firmly on identifying the nature and sources of the underlying L2 knowledge system, and on explaining developmental success and failure. Performance data are inevitably the researchers’ mainstay, but understanding underlying competence, not the external verbal behavior that depends on that competence, is the ultimate goal. Researchers recognize that SLA takes place in a social context, of course, and accept that it can be influenced by that context . . . However, they also recognize that language learning, like any other learning, is ultimately a matter of change in an individual’s internal mental state. As such, research on SLA is increasingly viewed as a branch of cognitive science. (2003a: 4)

D. ATKINSON

601

books and papers alphabetically or by subject; talk problems out in conversation; make group decisions; consult shopping lists or scan store shelves to help us shop; and literacy pervades our lives. Even ‘higher-order’ cognitive activities depend on the environment, as Clark’s (2001: 142) account of his academic writing process suggests:

Without question, cognition does sometimes operate in less environmentally dependent ways.1 But that we so frequently rely on the world when cognizing is not incidental. I present reasons for cognition’s deep environmentalism below. Clark and Chalmers (1998) propose a ‘parity principle’ (Wilson and Clark 2009) for determining the scope of extended cognition: If, as we confront some task, a part of the world functions as a process which, were it done in the head, we would have no hesitation in recognizing as part of the cognitive process, then that part of the world is . . . part of the cognitive process. (71) This claim is the heart of the extended cognition thesis: The environment is part of cognition itself, at least when fundamentally abetting onboard cognitive resources.

Cognition is adaptive This proposition holds that the primary purpose of cognition is to facilitate adaptivity to an uncertain environment (Barsalou 2008). It has two parts. First, rather than having evolved to address the abstract, logic-like problems of cognitivism, cognition is first and foremost flexible, on-line adaptive intelligence (Wheeler 2005). Cognition thus subserves adaptive action: crossing the street safely; staying warm and dry in cold rain; or knowing the right time, place, and partners for business, political action, or romance.


The brain supported some rereading of old texts, materials, and notes. While rereading these, it responded by generating a few fragmentary ideas and criticisms. These ideas and criticisms were then stored as more marks on paper, in margins, on computer discs, etc. The brain then played a role in reorganizing these data on clean sheets, adding new on-line reactions and ideas. The cycle of reading, responding, and external reorganization is repeated, again and again. Finally, there is a product . . . But this intellectual product owes a lot to those repeated loops out into the environment. Credit belongs to the embodied, embedded agent in the world. The naked biological brain is just a part (albeit a crucial and special part) of a spatially and temporally extended process, involving lots of extraneural operations, whose joint action creates the intellectual product. There is thus a real sense . . . in which . . . the ‘problem-solving engine’ is . . . the whole caboodle. . .: the brain and body operating within an environmental setting.


The second part of this proposition concerns the understanding that real-time cognitive resources are extremely limited (Schmidt 2001). If so, but fluent cognition is crucial for survival, then these resources must be guarded jealously. We have therefore evolved breathtaking abilities to offload cognition onto the world, as formalized in Clark’s cleverly named 007 Principle:

Thus, even where cognizers can solve problems using just the naked brain, it is usually cognitively cheaper and therefore more adaptive to bring in the environment. For instance, I could have memorized (with great effort) the talk this article is based on; instead, I used notes and Power Point to enhance my cognitive resources.

The environment is highly structured for cognitive activity This proposition is the converse of the first two. It also has two parts. First, the natural environment is rich in cognitive support structures, sometimes due to human–environment coevolution. Consequently, ‘the world is [often] its own best model’ for cognition (Brooks 1990: 5). That is, contra cognitivism’s decontextualized internal world models or representations, the natural environment often guides cognition-for-action. Thus, before leaving home I usually peek outside: If it’s sunny I know to apply sunscreen, if it’s rainy to take an umbrella. Another example: I once received these cognition-for-worldlydirection-finding instructions: ‘Turn right at the pond,’ and ‘Ours is the house with the large pine trees.’ Second, human environments are often designed for fluid, low-dimensional cognitive action. Modern homes, for instance, have their light switches by doors, door knobs uniformly positioned, closets located just where stored items can be accessed when needed, and kitchens organized in highly memory-enhancing ways, thereby enabling fluent cognition-for-cookingand-serving. Organizations of space and material thus profoundly afford the cognition-action complexes comprising everyday human behavior. In ground-breaking studies, anthropologists have indicated how human-designed environments support cognition-for-action. Hutchins (2005), for example, suggests that a line of people waiting for a concert is cognized as a ‘blend’ of the perceptual universal of linear salience, internalized cultural notions like ‘first-come, first-served’ (1559), and real-world structure, i.e. the line itself. To use this line for worldly action—e.g. to know how many precede you in it—requires no elaborate internal representation: Just look.


In general, evolved creatures will neither store nor process information in costly ways when they can use the structure of their environment . . . as a convenient stand-in for the informationprocessing operations concerned. That is, know only as much as you need to know to get the job done. (1989: 64)

D. ATKINSON

603

Cognition is shared and distributed

EMBODIED COGNITION In conceptualizing thought as the manipulation of logical symbols, the leaders of the cognitive revolution forgot that cognitive processes are also biological. The result was functionalism—the doctrine that cognition is implementation-neutral (Boden 2006). But cognitive science took a biological turn in the 1990s, leading to extensive research on embodied cognition. I briefly summarize this research here.2

Cognition, perception, and motor action are integrated activities Cognitivists modularize cognition, strictly dividing it from perception on the input side and motor action on the output side (Barsalou 2008). Recent


The extended cognition thesis emphasizes non-human affordances over human co-cognition. Yet shared cognition contributes crucially to human activity. Everyday conversation, for instance, is so highly coordinated that ‘the notion of an utterance as the sole product of a speaker, or . . . a mind, could hardly have been entertained had real talk-in-interaction’ been a central focus of linguistics (Schegloff et al. 1996: 20). Consider latching, the very common practice of positioning one turn-at-talk after another without a gap. Latching requires not just the ability to predict precisely when the preceding speaker will stop, but also what she will say, since the following turn must be relevant (i.e. appropriate, contentful, and formally congruent— Sperber and Wilson 1986; Schegloff 2007) vis-a`-vis the preceding one. Likewise, the phenomenon of one speaker completing another’s turn is predicated on high-level co-cognition; both this and the foregoing example can be explained only via profound sociocognitive anticipation (Kinsbourne and Jordan 2009)—the ability to know what others are about to do—and synchrony (Semin 2007)—the ability to adjust one’s own real-time behavior accordingly. There is converging evidence that anticipation and synchrony are built into human sociocognition. Research on infants, for example, shows apparently innate anticipatory and synchronization abilities such as proto-turn-taking (Foster 1990), an activity which has serious consequences for caregiver– child bonding soon after birth. Infants likewise actively imitate their caregivers, and begin to track their gaze late in the first year. Levinson’s (2006) and Tomasello’s (2008) research on what the former terms the ‘interaction engine’—the suite of innate abilities enabling the close ‘intention-reading’ that underlies all human communication, as described in more detail below—further supports the idea that humans are highly attuned co-cognizers. The same is true for the human ’mirror neuron system,’ also described below.


research, however, suggests the integrated nature of these domains. Thus, mirror neurons, a class of neurons in the cerebral cortex, activate not just when we perform motor actions, but also when perceiving others performing those same actions (Rizzolatti and Craighero 2004). By providing evidence for the shared neural coding of self- and other-action, mirror neurons appear to account for action-oriented understanding, imitative learning, and synchronized behavior.

Cognitive representations are embodied and action-oriented

recurrent patterns of bodily experience, ‘image’-like in that they preserve the topological structure of the perceptual whole, . . . [which are] operating dynamically in and across time, are realized as activation patterns . . . in and between topological neural maps, . . . link sensorimotor experience to conceptualization and language, and . . . afford ‘normal’ pattern completions that can serve as a basis for inference. (36) Image schemas are therefore analog representations based on cumulative sensory experience ‘that help solve . . . adaptive problems . . . in complex physical environments’ (Gibbs 2006: 69). Even abstract thought and language are


Cognitivism holds that cognitive representations are abstract, symbolic, and amodal (Harnish 2002); embodied cognition provides three alternative (if overlapping) understandings: simulation; analogical representations; and image schemas. Simulation is the cognitive ‘reenactment of perceptual, motor, and introspective states acquired during experience’ (Barselou 2008: 618). Extending one of Barsalou’s examples, the act of sitting in a chair is accompanied by the integrated, multimodal recording of sensory experience (e.g. of the chair’s texture), action (the neuromuscular activity of sitting), and introspection (e.g. expecting the chair to be comfortable). These representations are simultaneously combined with multimodal representations of past acts of sitting. Then, on next encountering a chair, this perception-actionintrospection complex is activated, enabling comprehension that the object is a chair. Analogical representations store patterns of how our bodies ‘mesh’ with the environment—these patterns incorporate environmental information (Glenberg 1997). Adapting from Glenberg, a path through the woods is remembered via the means it affords for passing from point A to B: space for our body between two trees here, an uneven walking surface, a sequence of stones across a stream there. Analogical representations are thus directly action-oriented, and summarize embodied experience. Johnson and Rohrer (2007) described image schemas as:

D. ATKINSON

605

cognized as image schemas: In ‘She launched into her spiel,’ ‘launched into’ is comprehended via conceptual metaphors based on the literal meaning of ‘launch’ and ‘into.’ Thus, all thought is embodied, no matter how abstract or figurative.

Empirical evidence for embodied cognition There is substantial empirical evidence for embodied cognition, much of it experimental. Barsalou (2008), Gibbs (2006), Glenberg (1997), and Semin and Smith (2008) provide partial reviews. Major findings include:

Embodied states affect and are affected by cognition (Barsalou 2008) Brain regions previously thought to be reserved for cognition are active during perception and motor action (Iverson and Thelan 1999) Neural mechanisms underlying cognition show embodiment effects (Rizzolatti and Craighero 2004) Memory shows embodiment effects (Glenberg 1997) Understanding objects depends on embodied experience with them (Carlson and Kenny 2005) Emotions enable cognition/understanding, e.g. brain damage causing loss of evaluative ability renders decision-making impossible (Damasio 1994) Bodily orientation affects cognition (Lempert and Kinsbourne 1982) Cognitive development depends on opportunities for embodied action (Smith 2005) Synchronized bodily states and action promote shared cognition (Semin and Cacioppo 2008)

Embodied cognition and language Embodied cognition researchers study language processing as a core function of cognition. Barsalou (2008), Gibbs (2006), Glenberg (1997), Iverson and Thelen (1999), and Johnson and Rohrer (2007) review findings, which include:

The brain’s ‘language areas’ activate during sensorimotor action (Bonda et al. 1994) Brain ‘motor areas’ activate during speech (Hauk et al. 2004) Verbalization of memory is facilitated when assuming original body posture during recall (Dijkstra et al. 2007) Linguistic tasks are facilitated when accompanied by action (Rieser et al. 1994) Descriptions of spatial associations are comprehended faster than those of spatial dissociations (Glenberg et al. 1987) Words with high ‘body-object interaction’ ratings are recognized faster than those without (Saikaluk et al. 2008) Speech and gesture emerge together in infancy (Iverson and Thelen 1999)



SYNTHESIZING EXTENDED AND EMBODIED APPROACHES TO COGNITION Although treated separately so far, extended and embodied views of cognition are broadly compatible and sometimes grouped together. The unifying idea is that cognition is naked and amodal only when conceptualized abstractly. Studied biologically, cognition is embedded—both in body and world. In a review entitled ‘Grounded Cognition,’ Barsalou (2008: 618–19) synthesizes embodied and extended views of cognition:

Similarly, in a chapter entitled ‘Embodied, Situated, and Distributed Cognition’, Clark (1998) describes these forms of cognition in a loosely united framework. That he retains the individual terms suggests disciplinary differences, while his synoptic treatment points toward synthesis. In fact, extended cognition has been studied largely by philosophers and anthropologists, and embodied cognition by psychologists, linguists, and neuroscientists. In this article, I adopt the view that extended and embodied cognition coalesce in a larger project—the study of cognition-for-worldly-action. At the same time, I retain separate terms in my formulation—extended, embodied cognition—to indicate that researchers often distinguish them.

EXTENDED, EMBODIED COGNITION AND SLA Extended, embodied perspectives on cognition have entered the cognitive science mainstream, but are just beginning to influence SLA studies. I try to promote their development here, working within a sociocognitive approach to SLA (Atkinson 2002; Atkinson et al. 2007; Churchill et al. 2010). Specifically, I introduce three SLA principles based on extended, embodied cognition: (1) The Inseparability Principle: Mind, body, and world work together in learning/ SLA; (2) The Learning-is-adaptive Principle: Learning/SLA facilitates survival and prosperity in complex environments; and (3) The Alignment Principle: A major engine of learning/SLA is alignment—the means by which we effect interaction.

The Inseparability Principle and SLA To appreciate the inseparability principle, first consider four questions vis-a`-vis Pictures 1–4, taken from naturally occurring interaction: (1) Who is the person in the picture?, (2) What is she doing?, (3) Where and when?, (4) Why? The aim of this


Conceptions of grounded cognition take many different forms . . . Some accounts . . . focus on the roles of the body in cognition, based on widespread findings that bodily states can cause cognitive states and be effects of them . . . Most accounts of grounded cognition, however, focus on the roles of simulation in cognition . . . Still other accounts . . . focus on situated action, social interaction, and the environment.

D. ATKINSON

607

exercise is to suggest that SLA is more than just a cognitive input/restructuring/output process, and what some of that ‘more’ may be.3

Possible answers to the ‘who’ question here are: ‘A young woman’, ‘An adolescent’. The ‘what’ question might be answered: ‘She’s thinking’ or ‘She’s gazing at something.’ The remaining questions cannot be answered because the context has been removed. Like the cognitivist cognizer, this thinker is an isolated entity, cognizing for no apparent purpose in an ecosocial vacuum. Yet she differs in being embodied—she is not just the abstract mechanism described, for instance, in Corder’s SLA field-establishing paper: ‘The internal structure of the (language acquisition) device, i.e., the learner, has gone relatively unexplored’ (1967/1981: 12).

Picture 2


Picture 1


Who: Based on her possible actions (see below) and age, this may be a student, perhaps in secondary school. What: We now see what she may be cognizing about: Papers lie before her, and a pen is barely visible in her right hand.4 She may thus be thinking in order to complete a school-sponsored assignment. Where and when: She is sitting at a table, perhaps in a classroom given the above. Why: Given the above, her cognition may be part of the larger sociocognitive action of completing a school-sponsored assignment, partly or conventionally in order to learn. Remove the assignment, and the nature of her cognition will likely change as well.

Who: We now see that the young woman has an older companion. Both perform similar activities—at least each has a literacy artifact before her and pen in hand. They even share their physical orientation: left hand/arm on artifact, right pen-holding hand to head, and head tilted right. Are they student and teacher? What: They appear to be studying together, the younger person completing an assignment, the older reading a book. Where and when: In a classroom? During class time? Why: No new information, but given the above, the older woman could be playing a guiding role.


Picture 3

D. ATKINSON

609

Here, possible answers start to blend together. Where/who/what: They are in a home, sitting (perhaps) at a dining table. This suggests a family relationship. This in turn suggests that homework is being completed, the older woman assisting. When: If the younger woman is a student, perhaps it is a weekend or after school. In fact, the younger woman (‘Ako’) is a Japanese junior high school student completing English grammar exercises to prepare for high school entrance exams. Her aunt (‘Tomo’) assists her as they sit at Tomo’s dining table. Ako has just read out a Japanese sentence, and repeated an adverbial as a place to start translating. Tomo has confirmed this choice by latching Ako’s utterance with an agreement marker and repeating the adverbial, closely shadowing Ako’s intonation. Now, as Ako ponders the correct translation, Tomo behaves in two seemingly contrasting ways: She looks at the book before her, having earlier told Ako she would help her when absolutely necessary but that otherwise she was on her own; and she aligns her body with Ako’s, seeming to signal ‘We’re in this together.’ One might even venture that the shared physical orientation suggests co-cognition—that the exercise answer will be highly co-constructed, as indeed it turned out to be (Atkinson et al. 2007). This pictorial illustration has three implications for SLA. First, it suggests that people cognize/learn not just mentally, but in environments comprised of bodies, cognitive tools, social practices, and environmental features. If, as the inseparability principle argues, such contexts crucially affect cognition/learning,5 then they cannot be treated as optional extras. Goodwin (2003: 20) makes this point specifically regarding embodiment: ‘The positioning, actions, and orientation of the body in the environment are crucial to how participants understand what is happening and build action together.’ Regarding learning,


Picture 4


this suggests that: (i) Learning is more discovering how to align with the world than extracting knowledge from it (Ingold 2000); and (ii) By being environmentally embedded, knowledge/cognition is made public and thereby learnable (Goodwin 2003). Indeed, Gee (1995) makes the latter a general principle of learning:

The second implication of this illustration concerns the quality of cognition/ learning (van Lier 2002). The personal relationships learning involves, its role in identity construction (Norton 2000), where and under whose sponsorship it occurs, and how it is embodied and enacted fundamentally influence its outcome. Unlike computers, humans don’t just process—they find value and meaning. This affects learners’ engagement with learning opportunities, including whether they engage at all. The qualitative nature of learning is well known in L2 teaching, but has not played a major role in mainstream SLA studies. This is because cognitivism views cognition as computational and amodal, and because it employs research methods which neutralize the quality of learning experiences in order to put them on common quantitative scales. Researchers thus negate what good teachers know—that learning depends on its meaningfulness to learners themselves. Third, if cognition/learning is complex and multimodal, as the illustration suggests, then it must be studied complexly and multimodally. Quoting Goodwin (2003) again, ‘Rather than being lodged in a single modality . . ., many forms of human action are built through the juxtaposition of quite diverse materials, including the actor’s body, the bodies of others, language, structure in the environment, and so on’ (22–3). An extended example of multimodal SLA is provided below, so I will not elaborate here.

The learning-is-adaptive principle and SLA Given that cognition exists to support adaptive action, sociocognitive approaches to SLA posit that learning’s main purpose is to increase adaptive functionality. This seems commonsensical—what else could learning be? Yet mainstream learning theory assumes the opposite: that learning means detaching knowledge from the world and internalizing it. As evidence that learning is adaptive, consider our finest learners: Children learn so quickly and well that some deny they are learning at all. Thus, for Chomsky, ‘It’s probably a mistake to even use the word ‘learning’ in connection with language acquisition’ (Equinox Films 1995). Yet children clearly do


The meanings of the parts of new systems, whether words, visual symbols, actions, or objects, must initially be rendered public and overt, so that the learner can see the connections between the signs and their interpretations. And this is done . . . by the ways in which words, actions, and social interaction are integrally intertwined. (p. 337)

D. ATKINSON

611

The alignment principle and SLA Atkinson et al. (2007) define alignment as ‘the complex means by which human beings effect coordinated interaction, and maintain that interaction in dynamically adaptive ways’ (169). Here, ‘interaction’ includes action both between human beings, and humans and their non-human environments. Regarding human interaction, sociality has alignment at its core: ‘A social relationship may be said to exist when several people reciprocally adjust their behavior to each other with respect to the meaning . . . they give . . . it, and


have a serious need to learn: As infantile dependence decreases, they must survive in a complex world. And this world is largely social. Children therefore learn best what they need most to be treated as humans—their social groups’ practices, including language. The learning-is-adaptive principle has four linked implications for learning/ SLA: (1) Learning/SLA is relational: Learning-as-adaptive-behavior concerns how to relate—how to articulate with one’s environment. For SLA, this means that L2s are learned as social practices—practices employed by the language’s users to accomplish social action; (2) Learning/SLA is experiential, participatory, and guided: One learns to relate by relating—learning is experiential. Of course, SLA involves a wide range of activities and motivations, but its paradigm route and goal centrally involve communicating with L2 users to effect social action, i.e., participation in social situations for conventional social purposes. This is a far cry from seeing acquisition-for-internalization as route and goal of SLA.6 Yet learning only from raw experience is risky business: Learning-as-participation is therefore frequently guided—mediated or scaffolded by various actors and structures; (3) Learning/SLA is public:7 One crucial way learning is guided is by externalizing its object while focusing the learner’s attention (Schmidt 2001) on it. That is, guides focus learners on particular learner-world relationships as relationships to be learned. These need to be learned because, as learner-world relationships, their learning enhances learner-world alignment. Ingold (2000: 37) calls such learning ‘education of attention,’ describing how, as they move through the woods with their elders, novice indigenous hunters are ’educated’ to attend to specific environmental cues as markers of the disposition of their prey. In Churchill et al. 2010, we show a similar process at work as Ako interactively learns relationships among (English-specific conceptions of) past, finished worldly action, English simple past-tense forms, and simple past-tense-cueing adverbials through Tomo’s repeated use of ‘symbiotic gestures’ (Goodwin 2003)—gestures which serve to align attention, embodied action, and the sociomaterial world; (4) Learning/ SLA is aligning, and learning to align: This final point summarizes the preceding three: Learning is a process of alignment—of continuously and progressively fitting oneself to one’s environment, often with the help of guides. I expand on this point immediately below in describing the final principle of a sociocognitive approach to SLA.


EXTENDED, EMBODIED COGNITION-FOR-SLA: AN EXAMPLE In this section I describe a case of extended, embodied—i.e. sociocognitive— cognition-for-SLA. The example comes from the videotaped EFL tutoring


when this reciprocal adjustment determines the form . . . it takes’ (Weber 1922/ 1978: 30). Levinson (2006) proposes that humans possess a formidable ‘interaction engine’—an evolutionarily evolved set of capabilities which make us fundamentally interactive and social. Language is central here, but is built on more basic interactive abilities: (i) theory of mind—the ability to determine others’ intentions and what they know of our own; (ii) inferential skills—the ability to infer meaning beyond what is expressed; (iii) cooperative skills—the ability to coordinate behavior in achieving common goals; and (iv) interactive skills—means of responding in real time to attempts to interact, e.g. turn-taking, gaze, and bodily orientation. All four skill sets are fully displayed, according to Levinson, in mundane social interaction. Levinson’s theory powerfully informs the concept of alignment: All animals interact socially, but nowhere near the level of humans. Yet Levinson gives insufficient attention to the public nature of interaction—the great contributions others’ actions-within-contextual-frames (i.e. actions vis-a`-vis the sociomaterial environment) make to understanding them. Thus, requesting salt while gazing at your dining companion and gesturing toward the salt shaker shows your meaning substantially in your environmentally framed actions. Regarding human–environment interaction, our online ability to align our behavior with our surroundings is largely what keeps us alive. This includes learning from those surroundings—finding exploitable features that enhance adaptivity. We also engineer our environments for adaptive advantage, as previously described. But what does alignment mean for SLA? First, it means having formidable pre-existing capacities for interacting without necessarily sharing a language. That is, all language learners have powerful interaction engines supporting their learning at every turn. Second, SLA itself is a process of alignment—of learning the ‘differences that make a difference’ (Bateson 1972: 459) in the L2 environment. This is what our guides teach us as we engage with the environment: A requisite idiom or formula here, a form-function relationship there—to do that, you need to say this. By trying to align with our environment—by learning to behave in ecosocially adaptive ways—we become ‘enskilled’ (Ingold 2000: 5). Third, as noted above, alignment has a public face: Our aligning/learning behaviors extend into the world, and as our guides facilitate and respond to them, alignment is further externalized. Ultimately, sociocognitive approaches to SLA are based on this tripartite premise: (i) Mind, body, and world are in continuous processes of interactive alignment; (ii) These processes are partly public; and (iii) In being public, they are learnable. Thus, if cognition is the site of learning, it is extended, embodied cognition that makes learning possible, at least in part.

D. ATKINSON

613

Example In this example, Ako and Tomo are completing the final item in an exercise section devoted to the present perfect tense, wherein the last two items involve converting present-perfect statements containing time adverbials into ‘how’ questions cued by the adverbial. Figure 1 shows the (completed) item; the parenthesized directions read, ‘Make a question about the underlined part.’ The transcript (Figure 2) begins with Ako quietly reading the exercise prompt aloud (line 1). By actively externalizing her language processing into the sociocognitive ‘problem space’ between herself and Tomo, Ako focuses their distributed cognition—their shared attention and problem-solving—on the task at hand. Then, after a pause, Tomo repeats the underlined adverbial, thus further focusing their shared cognition on this key part of the prompt (line 2). Note, too, the grammar exercise’s role in: (i) initiating cognition by providing its content; and (ii) organizing cognition, e.g. by highlighting the adverbial. As the episode begins, cognition is therefore thoroughly extended and ecological, circulating across and through Ako, Tomo, and the grammar worksheet-as-cognitive technology.

Figure 1: Exercise item


session featured in Pictures 1–4, where the participants are Ako, a junior high school student, and her aunt Tomo, an experienced EFL teacher. They are completing a grammar-exercise worksheet in preparation for high school entrance exams. Two main forms of interaction occur during this session: grammar-focused interaction, conducted mostly in Japanese and in a fairly serious manner; and interaction around but not strictly focusing on the worksheet—this tends toward playfulness and sometimes features English. At a finer grain of analysis, various participation frameworks (Goffman 1981)—fluid configurations of actors, actions, roles/identities, contexts, and artifacts—are constituted, as described below. Despite its mundane appearance, the grammar worksheet itself is a dynamic actor in this interaction—a highly designed cognitive technology (Clark 2001) for producing, comprehending, and learning L2 forms (Churchill et al. 2010). Its effective deployment assumes considerable background, e.g. in its four writing and two numeric systems, outline-like format, complex sequential organization, and the highly socialized behaviors by which such technologies are conventionally used. The grammar worksheet and skills needed to deploy it are therefore part of an academic ‘form of life’ (Wittgenstein 1958).


01 02 03 04

05 06

11 12 13 14 15 16

17 18

19 20 21 22 23 24

Figure 2: Example transcript8


07 08 09 10

A: ((Reads exercise prompt with flat intonation)) oI have been busy for two weekso (3.5) T: For two [weeks A: [How [[longu? T: [[>Ok ok.< (1.5) oSaeteru.o ((Withdraws gaze and opens book before her)) Ok, ok. You’re sharp! A: ((Chuckles while writing answer)) (2.0) oHow longuo (1.8) T: ((Holds up pen, glances at it while clicking it shut, then shifts gaze back to worksheet)) Sakki to onaji Same as last one (.8) A: Havu T: Havu A: I ((Tilts head to right)) T: Un >demo ichiyoo anata ni kiiteru hooga< (.5) shizen dakara have you how long have you:? Right, but generally asking ‘you’ is more natural so have you how long have you? A: Have you:, busy? ((Writes on worksheet)) (2.0) T: >Nanka nukete nai?< Isn’t something missing? A: Been= T: =Been un Right A: [[Been busy T: [[Been busy un. (3.5) ((Speaks with great feeling)) oFor a long time ne? (2.8) Kotaeta no kizuita. ((L eans toward A, laughing slightly)) I have been busy for a long time.o Been busy, right. For a long time, right? Did you realize that I answered? I have been busy for a long time. A: ((Nods and chuckles while writing)) For a long time okaku no ka to omottao I thought I was supposed to write ‘for a long time.’ T: ((Laughingly)) >Chigau chigau chigau< How long have you been busy. ((Points to A with pen)) (.8) Chittomo I have never been busy desho. (1.5) How long have you been busy. (.5) ((Pokes A with pen)) oAko-san.o No, no, no! How long have you been busy? ‘Never ever, I have never been busy,’ right? How long have you been busy, Ako-san? (1.5) A: ((Looking at exercise sheet)) I havu been busy for two/i/ weeks ((Shifts gaze to T and laughs energetically)) T: Nande nani de, for whatto. (.5) How come [why Why? From what? For what? How come? Why? A: [I am sleep T: Sleep de, hai. Ok, from sleeping A: Yes [yes ((laughs)) T: [oKekkoo de gozaimasu doozoo ((taps pen on next item on worksheet) That is fine, please proceed.

D. ATKINSON

615


In line 3, Ako overlaps Tomo’s repetition of the adverbial as she tentatively converts it into How longu? Here, the overlap suggests highly aligned intercognition, while the final rising intonation seeks cross-cognitive regulation—Tomo’s confirmation. That Ako receives this even as she speaks—Tomo overlaps with Ok ok—again suggests high-level alignment. Tomo then adds an under-the-breath compliment, Saeteru (‘You’re sharp!’ line 4), and Ako, chuckling, writes How long on the worksheet and orally repeats it. Simultaneously, Tomo shifts her gaze to the book she has just opened. In line 6, Tomo returns her gaze to the worksheet and says Sakki to onaji (‘Same as last one’), referring to the immediately preceding, just-completed exercise item, whose answer reads How often has he read this book? Tomo thus invokes Clark’s 007 Principle (2001 and above): If the answer already exists in the environment, why waste cognitive resources puzzling it out? Ako assents by offering havu, thus importing the word order from the just-completed item while retaining the auxiliary verb inflection from the current one. Receiving Tomo’s confirmation (line 8), Ako tries the same procedure in her next guess, venturing I for the inverted subject pronoun. This time, however, Tomo responds more complexly: Starting with a confirmation (Un), she continues by suggesting you as more ‘natural’ than I—i.e. that How long have I been busy? is a low-frequency usage. She then concludes with have you how long have you:? Through this response Tomo performs multiple sociocognitive tasks: (i) confirming the partial correctness of Ako’s answer; (ii) suggesting a better answer; (iii) justifying the better answer; (iv) combining this answer with Ako’s immediately preceding answer, yielding have you; (v) summarizing their progress to this point with how long have you, thus stabilizing their shared cognition and providing a sociocognitive base from which to proceed; and (vi) via the elongated vowel and final rising intonation of you:?, prompting Ako to proceed. In line 11, Ako repeats the answer achieved to this point, Have you:, pauses briefly, and then ventures busy? Here, vowel lengthening, internal pausing, and final rising intonation mark a cognitive state of uncertainty. A fast-paced turn-sequence ensues: Tomo answers with Nanka nukete nai? (‘Isn’t something missing?’), Ako offers an assertive-sounding been, Tomo responds with a latched repetition and agreement marker (line 14), and the sequence concludes with in-unison production of the mutually constructed answer, been busy, to which Tomo adds un (‘right’). The highly synchronized and rhythmic nature of this exchange is hard to capture in words, but it is an important additional way in which language, thought, and action are deeply aligned here (see also lines 2–4 and 20–24). At this point, the full answer to the exercise item has been negotiated, and Ako is busy writing it down. The interaction now undergoes a marked shift in tone and participation framework. This begins as Tomo, after a long turn-internal pause, answers the question they have constructed, How long have you been busy?, with a quiet but emotion-filled For a long time ne? (line 16).9 In this period, Tomo was desperately trying to balance work, graduate



school, mothering, household duties, and caring for a declining parent-in-law, so this statement sounds like a heartfelt comment on her life. Her subsequent actions support this interpretation: She leans toward Ako, and with a hint of laughter quietly asks, Do you realize that I answered? She then states the full form of her answer: I have been busy for a long time (line 16). In this turn-at-talk, Tomo performs several actions relevant to sociocognitive SLA. First, she imbues the lifeless grammar item with real-life meaning. Second, she apparently tries to engage Ako in real-life communication, as suggested by the highly personal nature of her talk, her new and more intimate physical orientation, her use of the alignment-seeking particle ne (Maynard 1993), her subsequent behavior (see below), and by the fact that she does so elsewhere in the tutoring session (Atkinson et al. 2007: 180–2). Third, Tomo expresses her feelings in English—English is thus now being used to communicate rather than to produce grammatical forms. Fourth, she seems to abdicate the teacher/tutor role, becoming an interlocutor with a real need to talk. Larger implications of these points are discussed below. Ako responds by nodding and chuckling as she finishes writing the answer, and then, eyes still on worksheet, leans back and replies, For a long time kaku no ka to omotta (‘I thought I was supposed to write ’for a long time,’’ line 17). This elicits an animated Chigau chigau chigau (‘No, no, no!’) from Tomo, said quickly with laughter and tight shakes of her head. Ako responds with laughter as Tomo leans close and asks, How long have you been busy, bringing her pen into Ako’s field of vision above the worksheet and pointing at her with it. Tomo then pauses, apparently expecting an answer, but none comes. Next, Tomo extends her utterance by animatedly answering for Ako—Chittomo I have never been busy desho (‘‘‘Never ever, I have never been busy’’, right?’ line 18)—probably teasing Ako about her lazy reputation. She then pauses, again waiting for Ako to respond. This is made clear in Tomo’s subsequent actions: She restates the question while poking Ako’s arm with her pen, and then follows with Ako-san, a form of address notable for its distance-indexing formality given their close relationship. As such, it serves as the verbal counterpart of the pen-poke—a strong bid for Ako to respond. Finally, after a longish pause and still looking at the worksheet, Ako replies I havu been busy for two/i/ weeks (line 19), gazes into the middle distance, and then shifts her gaze to Tomo with a flip of her hair while laughing energetically. Tomo responds by leaning away from Ako, laughing, and covering her face with one hand. As their laughter subsides, she queries Nande nani de, for whatto (‘Why? From what? For what?’ line 20), pauses, and finishes with How come why as she brings herself vertical and leans back toward Ako. Ako overlaps with I am sleep, Tomo responds Sleep de, hai. (‘Ok, from sleeping,’ line 22), and Ako laughingly confirms, Yes yes, just as Tomo shifts the participation framework by sitting up straight, overlapping with a quiet and super-polite Kekkoo de gozaimasu doozo (‘That is fine, please proceed,’ line 24), and tapping her pen on the worksheet’s next item.

D. ATKINSON

617

Implications of example for extended, embodied SLA Analyzing interaction in detail yields a radically different picture of learning activity than the mainstream/cognitivist version. Here, I introduce four implications of the foregoing analysis for SLA.

Engagement

Interaction From a cognitivist perspective, L2 learners are viewed in terms of what they lack—the full and fully internalized L2 system. A sociocognitive approach, by contrast, sees learners as already possessing enormous interactive capabilities—interaction engines, in Levinson’s (2006) terms. In this view, language is just the apex of the interaction pyramid, though no less important for that. Yet the mainstream cognitivist perspective, by focusing solely on language, makes SLA appear more mysterious and difficult than it may actually be. The interaction engine is fully displayed in the example—both the grammar-exercise answer and ensuing conversation are highly co-constructed. Certainly, Tomo scaffolds the interaction in various ways, including by using Japanese, yet without pre-existing interactive abilities there would be little to scaffold. I am not suggesting that cognitivists deny such capacities, but they


In cognitivism, interaction merely conditions input for the information-processing at the heart of SLA (Gass 1998). Following Chomsky, language is thus seen as incidentally interactive and social, but fundamentally internal and mental. The present approach, by contrast, sees interaction as SLA’s basic purpose and ground: Instead of isolating language in cognitive space, we wear it on our sleeve, so to speak, because it helps us live in the world. Engagement is therefore a better metaphor than internalization for conceptualizing SLA sociocognitively. There is ample evidence of engagement in the example. Consider lines 16–24, where Tomo refocuses the interaction on their own lives. Although Ako is slow to participate fully, Tomo nonetheless engages her in multiple ways: She introduces her own plight in line 16, then leans toward Ako, checking to see if she has noticed the participation framework shift. On receiving an ambiguous response, Tomo teasingly answers for Ako (line 18), gives her a pen-poke, and uses an honorific suffix that momentarily alters their relationship. Ako responds, apparently using the grammar exercise to help her, and a brief conversation ensues, with Ako sharing a detail from her own life (I am sleep) and/or joining in the game of being called lazy. Then, in line 24, Tomo shifts the participation framework—and the role/identity construals within it—back to grammar-exercise completion, using the Japanese politeness system again in concert with body orientation and gesture.


receive scant attention regarding SLA. This is doubtless because interaction is viewed in a limited, instrumental way.

Visibility

Alignment The preceding implications are central aspects of the more general concept of alignment. As argued previously, alignment is the basis of social life, and beyond that, of adaptivity to the environment. Views of SLA which ignore alignment neglect a fundamental aspect of humanity, as well as reasons for which and actions by which a fundamental human activity—learning—must occur.

CONCLUSION Bakhtin’s dramatic vision is close to the one assumed in this article: To be means to be for the other, and through him [sic], for oneself. Man has no internal sovereign territory, he is all and always on the boundary; looking within himself, he looks in the eyes of the other or through the eyes of the other. . . I cannot do without the other; I cannot become myself without the other; I must find myself in the other, finding the other in me (in mutual reflection and perception). (1979, quoted in Wertsch 1998: 116) For all its literariness, this passage captures the fundamental extendedness of human existence. For almost 400 years now, Descartes’ separation of mind from everything else has mesmerized Western thought, and cognitive science was one result. A self-professed Cartesian—Chomsky—is the foremost modern exponent of this view, but he was also part of a larger movement. Cognitive scientists have now begun to question their field’s founding premises, and a


For cognitivists, language learning is invisible (Doughty 2003). Yet if cognition occurs not just in but between people, and between people and their sociocognitive environments, then it is also in the world. People learn, from a sociocognitive perspective, by participating in extended cognition. When, in lines 3–11, Ako guesses How longu?, Tomo replies Ok ok. Saeteru . . . Sakki to onaji (‘Ok ok. You’re sharp . . . Same as last one.’), Ako ventures Havu, Tomo repeats it, Ako guesses I, Tomo counters with have you how long have you, and Ako responds Have you:, busy?, we see a worldly process of co-cognition, irreducible to happenings in individual heads. Although impossible to prove from a cognitivist perspective, where evidence is represented by post-hoc artifacts, especially on tests of conscious knowledge (Doughty 2003), what we see in such sequences is learning-in-flight—the learner engaging with the object of learning in the public, sociocognitive world.10 This view has novel implications for second language learning and teaching (Atkinson et al. 2007: 183–5).

D. ATKINSON

619

richer, more complex interdiscipline is emerging. They increasingly find that understanding the mind/brain means studying it in the body, and understanding the embodied mind means studying it in the world; and this is simply because the mind is in the body and the world. If SLA studies is a cognitive science—or seriously desires to become one—shouldn’t it follow suit?

ACKNOWLEDGEMENTS

NOTES 1 There is currently debate over whether all forms of cognition are extended and embodied, or just some (e.g. Wilson 2002). In one sense, at least, the former is obviously true—all cognition takes place in human bodies embedded in sociomaterial worlds. If it is further argued, as here, that cognition is fundamentally adaptive, then even off-line cognition must be influenced by that fact—i.e. if cognition exists to help us adapt to/align with our worldly environments, then it is essentially extended and embodied. At the same time, cognitive extendedness and embodiment obviously vary in degree depending on the situation. Such flexibility is a hallmark of human cognition. 2 This summary (and associated terminology) is based on standard historical accounts of cognitive science (e.g. Boden 2006), embodied cognition (e.g. Barsalou 2008), and studies of embodied language use (e.g. Glenberg 1997). 3 The presentation of pictures in this subsection is inspired by Rogoff (2003). 4 In the version of these pictures reproduced for publication it is virtually impossible to see Ako’s pen, except perhaps in Picture 4. 5 I regard cognition and learning as broadly continuous, in the same way

that connectionists do: Each act of understanding updates/augments the organism’s (socio)cognitive state so that there is continuous ecological adaptation/alignment (Atkinson et al. 2007). 6 This is one sense in which the present approach differs from not only cognitivist SLA studies but also sociocultural theory (e.g. Lantolf and Thorne 2006), with which it also has affinities. See Atkinson (2002: 537–38) for further discussion. 7 Although this approach shares an emphasis on the public nature of cognition with conversation analysis-forSLA (Mori and Markee 2009), it differs in that the latter tends to reduce cognition to public performance, whereas I view cognition as operating in integrated sociocognitive space (Atkinson 2002; Atkinson et al. 2007). 8 The example is transcribed using the system developed by Gail Jefferson (Ochs et al. 1996: 461–5). 9 Tomo’s comment is marked as ‘emotion-filled’ by: (i) a ‘catch’ or sudden huskiness in her voice; (ii) the long turn-internal pauses bracketing this comment; (iii) slowness of speech; and (iv) quietness. 10 Different approaches to language acquisition have different ways of


I would like to thank Paul Kei Matsuda, Jinju Nishino, and Yumi Takamiya for expert technical help, Takako Nishino, Hanako Okada, and Yumi Takamiya for expert language advice, and Paul Bruthiaux, Ulla Connor, Paul Kei Matsuda, Takako Nishino, Hanako Okada, Tetyana Smotrova, and Beril Tezeller Arik for thoughtful comments on earlier drafts of this article.


defining ‘acquisition.’ Here, Bickerton’s (1981) criterion of first appearance of a linguistic feature in

the learner’s speech might be more relevant.

REFERENCES Clark, A. and D. Chalmers. 1998. ‘The extended mind,’ Analysis 58: 7–19. Corder, S. P. 1967/1981. ‘The significance of learners’ errors’ in Corder S. P. (ed.): Error and Interlanguage. Oxford University Press. Damasio, A. 1994. Descartes’ Error. Avon Books. Descartes, R. 1637/1960. Discourse on Method and Meditations. Bobbs-Merrill. Dikjstra, K., M. Kaschak, and R. Zwaan. 2007. ‘Body posture facilitates retrieval of autobiographical memories,’ Cognition 102: 139–49. Doughty, C. 2003. ‘Instructed SLA: Constraints, compensations, and enhancement’ in Doughty C. and M. Long (eds): Handbook of Second Language Acquisition. Blackwell, pp. 256–310. Doughty, C. and M. Long. 2003. ‘The scope of inquiry and goals of SLA’ in Doughty C. and M. Long (eds): Handbook of Second Language Acquisition. Blackwell, pp. 3–16. Equinox Films. 1995. The Human Language, Part II: Acquiring the Human Language: Playing the Language Game. (Available through Transit Media at http://www.transmit media.com.) Foster, S. 1990. The Communicative Competence of Young Children. Longman. Gass, S. 1998. ‘Apples and oranges: Or, why apples are not orange and don’t need to be: A response to Firth and Wagner,’ Modern Language Journal 82: 83–90. Gee, J. P. 1995. ‘First language acquisition as a guide for theories of learning and pedagogy,’ Linguistics and Education 6: 331–54. Gibbs, R. 2006. Embodiment and Cognitive Science. Cambridge University Press. Glenberg, A. 1997. ‘What memory is for,’ Behavioral and Brain Sciences 20: 1–55. Glenberg, A., M. Meyer, and K. Lindem. 1987. ‘Mental models contribute to foregrounding during text comprehension,’ Journal of Memory and Learning 26: 69–83.


Atkinson, D. 2002. ‘Toward a sociocognitive approach to second language acquisition,’ Modern Language Journal 86: 525–45. Atkinson, D., E. Churchill, T. Nishino, and H. Okada. 2007. ‘Alignment and interaction in a sociocognitive approach to second language acquisition,’ Modern Language Journal 91: 169–88. Churchill, E., T. Nishino, H. Okada, and D. Atkinson. 2010. ‘Symbiotic gesture and the sociocognitive visibility of grammar,’ Modern Language Journal 94: 2. Barselou, L. 2008. ‘Grounded cognition,’ Annual Review of Psychology 59: 617–45. Bateson, G. 1972. Steps to an Ecology of Mind. University of Chicago Press. Bickerton, D. 1981. ‘Discussion of ’Two perspectives on pidginization as second language acquisition’’’ in Andersen R. (ed.): New Dimensions in Second Language Acquisition Research. Newbury House. Boden, M. 2006. Mind as Machine: A History of Cognitive Science. Clarendon Press. Bonda, E., M. Petrides, S. Frey, and A. Evans. 1994. ‘Frontal cortex involvement in organized sequences of hand movements: Evidence from a positron emission topography study,’ Society for Neurosciences Abstracts 20: 353. Brooks, R. 1990. ‘Elephants don’t play chess,’ Robotics and Autonomous Systems 6: 3–15. Carlson, L. and R. Kenny. 2005. ‘Constraints on spatial language comprehension: function and geometry’ in Pecher D. and R. Zwaan (eds): Grounding Cognition. Cambridge University Press. Clark, A. 1989. Microcognition: Philosophy, Cognitive Science, and Parallel Distributed Processing. MIT Press. Clark, A. 1998. ‘Embodied, distributed, and situated cognition’ in Bechtel W. and G. Graham (eds): Companion to Cognitive Science. Blackwell. Clark, A. 2001. Mindware: An Introduction to the Philosophy of Cognitive Science. Oxford University Press.

D. ATKINSON

Mori, J. and N. Markee. 2009. ‘Language learning, cognition, and interactional practices: An introduction,’ International Review of Applied Linguistics 47: 1–9. Norton, B. 2000. Identity and Language Learning. Pearson. Reiser, J., A. Garing, and M. Young. 1994. ‘Imagery, action and young children’s spatial orientation: It’s not being there that counts, it’s what one has in mind,’ Child Development 45: 1043–56. Rizzolatti, G. and L. Craighero. 2004. ‘The mirror-neuron system,’ Annual Review of Neuroscience 27: 169–92. Rogoff, B. 2003. The Cultural Nature of Human Development. Oxford University Press. Sacks, H., E. Schegloff, and G. Jefferson. 1974. ‘A simplest systematics for the organization of turn-taking in conversation,’ Language 50: 696–735. Saikaluk, P., P. Pexman, L. Aguilera, W. Owen, and C. Sears. 2008. ‘Evidence for the activation of sensorimotor information during visual word recognition: The bodyobject interaction effect,’ Cognition 106: 433–43. Schegloff, E. 2007. Sequence Organization in Interaction. Cambridge University Press. Schegloff, E., E. Ochs, and S. Thompson. 1996. ‘Introduction’ in Ochs E., E. Schegloff, and S. Thompson (eds): Interaction and Grammar. Cambridge University Press, pp. 1–51. Schmidt, R. 2001. ‘Attention’ in Robinson P. (ed.): Cognition and Second Language Instruction. Cambridge University Press, pp. 3–32. Semin, G. 2007. ‘Grounding communication: Synchrony’ in Kruglanski A. and E. T. Higgins (eds): Social Psychology: Handbook of Basic Principles. Guilford Press, pp. 630–49. Semin, G. and J. Cacioppo. 2008. ‘Grounding social cognition: Synchronization, coordination, and co-regulation’ in Semin G. and E. Smith (eds): Embodied Grounding. Cambridge University Press, pp. 119–47. Sharwood Smith, M. 1991. Second language acquisition and the cognitive enterprise.’ Plenary presentation, Second Language Research Forum. Los Angeles, March 2. Smith, L. 2005. ‘Cognition as a dynamic system: Principles from embodiment,’ Developmental Review 25: 278–98.


Goffman, E. 1981. Forms of Talk. University of Pennsylvania Press. Goodwin, C. 2003. ‘The body in action’ in Coupland J. and R. Gwin (eds): Discourse, the Body, and Identity. Palgrave-MacMillan. Harnish, R. 2002. Minds, Brains, Computers: An Historical Introduction to the Foundations of Cognitive Science. Blackwell. Hauk, O., I. Johnsrude, and F. Pulvermu¨ller. 2004. ‘Somatotopic representation of action words in human motor and premotor cortex,’ Neuron 41: 301–7. Hutchins, E. 2005. ‘Material anchors for conceptual blends,’ Journal of Pragmatics 37: 1555–77. Ingold, T. 2000. Perception of the Environment. Routledge. Iverson, J. and E. Thelen. 1999. ‘Hand, mouth, and brain: The dynamic emergence of speech and gesture’ in Nunez R. and W. Freeman (eds): Reclaiming Cognition: The Primacy of Action, Intention, and Emotion. Imprint Academic. Johnson, M. and T. Rohrer. 2007. ‘We are live creatures: Embodiment, American Pragmatism, and the cognitive organism’ in Ziemke T. et al. (eds): Body, Language, and Mind. Mouton de Gruyter. Kinsbourne, M. and J. S. Jordan. 2009. ‘Embodied anticipation: A neurodevelopmental interpretation,’ Discourse Processes 46: 103–26. Larsen-Freeman, D. 2007. ‘Reflecting on the cognitive-social debate in second language acquisition,’ Modern Language Journal 91: 773–87. Lantolf, J. and S. Thorne. 2006. Sociocultural Theory and the Genesis of Second Language Development. Oxford University Press. Lempert, H. and M. Kinsbourne. 1982. ‘Effect of laterality of orientation on verbal memory,’ Neurophysiologia 20: 211–4. Levinson, S. 2006. ‘On the human ‘‘interaction engine’’ ’ in Enfield N. J. and S. Levinson (eds): Roots of Human Sociality. Berg, pp. 39–69. Long, M. and C. Doughty. 2003. ‘SLA and cognitive science’ in Doughty C. and M. Long (eds): Handbook of Second Language Acquisition. Blackwell, pp. 866–70. Maynard, S. 1993. Discourse Modality: Subjectivity, Emotion, and Voice in the Japanese language. Benjamins.

621


Sperber, D. and D. Wilson. 1986. Relevance: Communication and Cognition. Harvard University Press. Tomasello, M. 2008. Origins of Human Communication. MIT Press. van Lier, L. 2002. ‘An ecological-semiotic perspective on language and linguistics’ in Kramsch C. (ed.): Language Acquisition and Language Socialization. Continuum. Weber, M. 1922/1978. ‘The nature of social activity’ in Runciman W. C. (ed.): Weber: Selections in Translation. Cambridge University Press, pp. 7–32.

Wertsch, J. 1998. Mind as Action. Harvard University Press. Wheeler, M. 2005. Reconstructing the Cognitive World. MIT Press. Wilson, M. 2002. ‘Six views of embodied cognition,’ Psychonomic Bulletin & Review 9: 625–36. Wilson, R. and A. Clark. 2009. ‘Situated cognition: Letting nature take its course’ in Aydede M. and P. Robbins (eds): Cambridge Handbook of Situated Cognition. Cambridge University Press, pp. 55–77. Wittgenstein, L. 1958. Philosophical Investigations. Prentice-Hall.


Applied Linguistics: 31/5: 623–650 ß Oxford University Press 2010 doi:10.1093/applin/amq017 Advance Access published on 4 June 2010

Partial Word Form Learning in the Written Mode in L2 German and Spanish 1,

*JOE BARCROFT and

2,y

SUSANNE ROTT

1

Washington University in St. Louis and 2University of Illinois at Chicago *E-mail: [email protected]; [email protected]

INTRODUCTION Studies on the second language (L2) lexicon have demonstrated repeatedly that form-meaning connections following initial lexical form encoding are incomplete, regardless of whether words are learned in an explicit learning condition (e.g. Nation 1982; Meara and Ingle 1986; Hulstijn 1991; Ellis and Beaton 1993) or through contextualized input (e.g. Wesche and Paribakht 1996; Schmitt 1998; Rott 1999). Consequently, researchers have outlined how word knowledge develops across multiple continua involving different aspects of word knowledge, such as the development from receptive to productive, partial to precise, and shallow to deep (e.g. Wesche and Paribakht 1996; Melka 1997; Henriksen 1999; Qian 1999). VanPatten et al. (2004) expanded the description of continua by addressing cognitive processes, such as overgeneralization of word use on the way to native-like lexical use, the relationship between the strength of initial encoding and long-term retention, and the accommodation of multiple word meanings in one lexical entry. While these continua describe the development of the semantic component of word knowledge (word meaning) only, VanPatten et al. also mentioned the dimensions of phonological and orthographic word knowledge, which develop from partial to complete levels of productive ability. Finally, Boogards (2001) found that multiword expressions containing known words were more easily learned than


Previous research on second language (L2) vocabulary learning has examined the relationship between word properties and learnability (e.g. Ellis and Beaton 1993). Few studies, however, have examined patterns in partial word form learning as a method of assessing learnability and improving our understanding of allocation of processing resources during word-level input processing. Expanding on the research of Barcroft (2000b, 2008), the present study examined partial word form learning in L2 German and Spanish while testing for effects of number of syllables in target words. Production data from an L1-to-L2 translation task (administered after a learning phase) were analyzed to determine percentage of partial versus fully produced words; amount of word produced in partial words; fragment length in partial words; and word-internal location of target letters. Results indicated production of approximately 49 percent more partial words than whole words, a high percentage of one-letter fragments, and privileging for word-initial position for both languages.

624 PARTIAL WORD FORM LEARNING IN GERMAN AND SPANISH

REVIEW OF LITERATURE Numerous studies on L2 lexical learning have identified aspects of words that may affect the ease or difficulty of learning a new word form, including word


completely new words, once again emphasizing the critical role of developing word form knowledge within the larger process of L2 vocabulary acquisition. With few exceptions (Meara and Ingle 1986; Schmitt 1998; Barcroft 2000b, 2008), little research has addressed the issue of patterns in partial word form learning. This shortfall in the data may be attributable, at least in part, to an assumption or an unquestioned implied default position that one’s knowledge of complete or whole word forms develops ‘automatically’ as one’s L2 knowledge increases. Theoretical work and empirical evidence suggest otherwise. Studies have demonstrated, for example, that in L1 and L2, words are often retrieved and can be produced only partially (in both the written and spoken modalities). This phenomenon is sometimes connected to the tip-of-the-tongue (TOT) state (e.g. Brown and McNeill 1966; Aitchinson and Straf 1981; Meyer and Bock 1992), which is generally explained as the result of partial word encoding or incomplete activation in memory during usage. These two explanations correspond to two distinct categories of the TOT state, the classic TOT state being for words that are fully encoded but infrequently used. The present study, on the other hand, concerns partial word encoding, which has, as one of its possible consequences, a situation in which learners may be able to retrieve word-initial and word-final segments of words because other (word-medial) segments have not been encoded to a sufficient degree so as to be able to be retrieved. Likewise, research on the learnability of L2 word forms has identified word length, degree of sound-script correspondence, and sequential letter probabilities as key factors during lexical encoding and predictors of subsequent lexical retrieval (e.g. Ellis and Beaton 1993; Laufer 1997). In light of this background, new research on partial word form learning may help to advance our understanding of the process of partial word form encoding during L2 lexical learning. Identifying characteristics of L2 word fragments that learners produce when they produce only part of a target word can help to explain why some parts of words are processed differently than others during the learning process and why word learning results in only partial word form knowledge. The present investigation is an exploratory study aimed to gain new insights into the allocation of processing resources, the relative perceptual salience of word parts, and the potential privileging of different word parts during the word-level form-meaning encoding process. The study focused on word form learning in the written mode. By comparing partial L2 word learning in L2 learners of German and L2 learners of Spanish in a translation-based learn-and-test paradigm, the present study also sought to explore the extent to which identifiable patterns in word-level input processing may be specific to one language as opposed to being more universal.

J. BARCROFT AND S. ROTT 625


length (e.g. Ellis and Beaton 1993); pronounceability (e.g. Rodgers 1969); phonotactic regularity (Ellis and Beaton 1993); sound-script incongruence and difference in writing systems (e.g. Koda 1999); word frequency (e.g. De Groot and Keijzer 2000); acoustic similarity between words (e.g. Henning 1973; Meara 1982); part of speech (Rodgers 1969; Ellis and Beaton 1993); concept imageability (Ellis and Beaton 1993); word concreteness and cognate status (e.g. De Groot and Keijzer 2000); and pronunciation time (Ellis and Beaton 1993). Other studies on sentence-level input processing have provided information about the interrelated nature of perceptual saliency and the allocation of processing resources that may (or may not) be applicable to word form learning. In consideration of a general principle that learners’ attentional resources are limited (see also Skehan 1998; Skehan and Foster 1999, 2001), for example, VanPatten (e.g. 1996, supported by Barcroft and VanPatten 1997; Rosa and O’Neill 1998) has included perceptual salience in his model of sentencelevel L2 processing. Specifically, VanPatten posits that items in sentence-initial position are more perceptually salient and have precedence over items in other positions. This sentence-level primacy effect appears to extend to firstlanguage (L1) word-level processing as well. Brown and McNeill (1966), for example, tested subjects in the tip-of-the-tongue state about words that came to their mind as they were attempting to retrieve L1 words. The researchers found that other words that the subjects reported as being similar in sound to the target word tended to share sounds at the beginnings and ends of the words. This finding provided evidence for what has been referred to as the bathtub effect, or the phenomenon that people tend to remember word parts at the beginnings and ends of words more than the middle parts of words. In a subsequent study on memory for the correct parts of target words in malapropisms, Aitchinson and Straf (1981) found that the bathtub effect was influenced by word length. The researchers found that the beginnings of short words (one to two syllables) were remembered marginally better than the beginnings of long words (three or more syllables) and that the ends of long words were remembered substantially more than the ends of short words. Meyer and Bock (1992) also examined mechanisms of incomplete word form retrieval (tip-of-the-tongue state). Assessing the effect of phonological and semantic clues during retrieval, they found that incomplete word production was due to insufficient word-form activation. Few lexical studies have assessed word form learning in L2 specifically, however. Lexical learning assessment measures have focused largely on the learning of word meaning, such as the comprehension of word meaning (e.g. Meara and Buxton 1987), the production of word meaning (e.g. Laufer and Nation 1995), or word associations (e.g. Riegel and Zivian 1972; Read 1993), although a few studies have attempted to measure multiple aspects of L2 word knowledge (e.g. Schmitt 1998) and how learners progress across multiple stages of knowledge from not having seen the word before to being



able to use it in a sentence as measured with the Vocabulary Knowledge Scale (Wesche and Paribakht 1996). Addressing the limitations of previous word knowledge assessment tools, Laufer and Goldstein (2004) developed a computer adaptive test that measures size and strength of lexical knowledge. Although this test includes the active recall of word form, the scoring procedure does not account for partial word form knowledge. Instead, the procedure includes dichotomous scoring of ‘correct’ or ‘incorrect’. Two other studies have established componential scales that address different degrees of word form knowledge on the partial-to-complete continuum. In a longitudinal study, Schmitt (1998) used a 4-point scale that measured the ability to spell target words correctly. A score of 0 indicated no word-form knowledge; a score of 1 was given for production of initial but not ‘later letters’ (p. 290) of the target word; a score of 2 was given for a phonologically correct but orthographically incorrect version of the target word; and a score of 3 was given if the word was spelled correctly. This scale did not allow Schmitt to detect any increase in the ability to spell words over the course of one year because the three advanced learners in Schmitt’s study demonstrated knowledge Levels 2 and 3 at the first time of testing. Meara and Ingle (1986) isolated the learning of consonants by learners of French as a foreign language. Oral production data were scored for which consonants were produced correctly, in initial, medial or final position. The data revealed that these secondyear students recalled more consonants correctly when they occurred at the beginning of the word than in medial or final position. Barcroft (2000a) developed a lexical production scoring protocol (LPSP) that assesses partial word-form knowledge based on the number and the location of correctly produced letters. This scoring procedure further discriminates between letters produced and letters produced in the correct position. Following this protocol, a score of 1 is assigned to any completely produced word. Other scores are assigned as follows: .25 if approximately one-quarter of the word (or one correct letter) was produced; .50 if approximately half of the word was produced; .75 if approximately three-quarters of the word is produced; and 0 in all other cases. Barcroft (2000b, 2008) has demonstrated the usefulness of scoring with the LPSP in two studies that assessed the nature of partial word form learning for L2 Spanish. In each of the two studies, English-speaking learners of L2 Spanish were asked to do their best to learn 24 new words in Spanish during a word-picture pair learning paradigm with each word-picture pair being repeated two times. After the learning phase in each study, the participants attempted to produce as much of each target word as they could, either in spoken or written form, when they were presented with pictures only. Barcroft analyzed all of the items that learners produced in order to identify (i) percentages of partially versus fully produced target words, (ii) the approximate amount of each word that they produced, (iii) lengths of segments that they tended to produce, and (iv) the location of the segments that they produced. The results of both studies revealed greater production of partial


words than whole words and other patterns in partial word form learning. Barcroft (2008), for example, found production of 69 percent partial words versus 31 percent of whole words, a high percentage of one-letter fragments, and heavy privileging for word-initial position. These findings indicate that partial word form learning, as opposed to whole word learning, is the norm and not the exception. The study also demonstrated a word-internal primacy effect for initial word form learning but did not confirm the recency (word-final) component of the bathtub effect.

PURPOSE OF THE PRESENT STUDY Downloaded from applij.oxfordjournals.org by guest on December 31, 2010

To summarize, the findings of previous research suggest that lexical learning depends heavily on different properties of word-internal fragments, such as their location within a word. As outlined above, studies on sentence-level input processing (VanPatten 1996; Barcroft and VanPatten 1997; Rosa and O’Neill 1998) have demonstrated that (i) not all parts of a sentence are processed equally and that (ii) perceptual salience and allocation of cognitive resources affect which parts of sentences are privileged during processing. Although sentence- and word-level input processing are substantively different cognitive processes, the present investigation examined at the word level parallels of some of the issues previously examined at the sentence level by assessing how different properties of word-internal fragments are affected by allocation of processing resources during the initial encoding of new word forms. By examining partial word form productions, the current study explored whether primacy and recency effects that characterize sentence-level processing also appear L2 vocabulary learning and whether the limited processing resources of beginning L2 learners affect the length and the location of the fragments produced. In this study, word form production was defined as the ability to recall and spell target word forms correctly. We make the general assumption that one’s ability to produce target word form on a posttest, be it partial or full target word form, is a reflection, at least in large part, of the extent to which a learner has attended to and processed the word form, partial or full, during a learning phase. While previous studies have used primarily dichotomous correct/incorrect scoring procedures, the current study used the LPSP-written (Barcroft 2000a), a scoring protocol that is more sensitive to partial word form learning and that allows one to gain insights into the allocation of processing resources during the gradual, bit-by-bit process of word form learning. The current investigation also expanded on previous studies on lexical word form learning (Barcroft 2000b, 2008) by comparing partial word form production by learners of two different L2s: German, a stress language; and Spanish, a syllable language. Syllables in Spanish tend to be relatively short. Seventy percent of syllables follow simple consonant–vowel (CV) or consonant– vowel–consonant (CVC) structure. In contrast, in German, only 30 percent


RESEARCH QUESTIONS The current investigation addressed the following research questions about word learnability by examining patterns in word production after intentional vocabulary learning: 1 Is there a difference between the number of partial words versus whole words produced? If the answer is ‘yes’: (1a) What is the nature of the difference? (1b) Is the difference affected by target language (German or Spanish)? (1c) Is the difference affected by number of syllables (two or three) in target words? 2 Is there a difference between the general amount of the word (onequarter, one-half, three-quarters) produced in partial words? If the answer is ‘yes’: (2a) What is the nature of the difference? (2b) Is the difference affected by target language (German or Spanish)? (2c) Is the difference affected by number of syllables (two or three) in target words?


of the syllables belong to CV and CVC classes. Therefore, in German, syllables generally have more letters, and in particular, German has more multi-syllabic words (Conrad and Jacobs 2004). As such, the present study sought to test for potential language-specific relationships between word-internal properties and allocation of processing resources and to explore the possibility that some patterns in word-form encoding may be shared and possibly universal. Additionally, although German is considered to be largely a phonemic (based on degree of transparency when going from grapheme to phoneme) language, Spanish is even more so, and this difference in degree of grapheme-to-phoneme transparency may impact upon successful and unsuccessful encoding of target word forms and the nature of partial word forms that learners are able to learn and subsequently produce. Finally, in this investigation, we also explored the relationship between word length and learning fragments in different word-internal locations. In order to do so, we assessed word fragment production in three different word-internal locations (initial, medial, final) for both two- and three-syllable words. This approach allowed us to examine the extent to which the bathtub effect (Aitchinson and Straf 1981) may be observable in learner-based productions of L2 vocabulary across words of different numbers of syllables. Would word-initial and word-final fragments be learned more readily, as the bathtub effect would predict, or would only word-initial fragments be learned markedly more than other locations, as previous research (Barcroft 2000b, 2008) indicates?


3 Is there a difference between number of word fragments of a certain length (one, two, three, or four or more letters) produced in partial words? If the answer is ‘yes’: (3a) What is the nature of the difference? (3b) Is the difference affected by target language (German or Spanish)? (3c) Is the difference affected by number of syllables (two or three) in target words?

(4a) What is the nature of the difference? (4b) Is the difference affected by target language (German or Spanish)? (4c) Is the difference affected by number of syllables (two or three) in target words? Answers to each of these questions helps to provide new information about the general nature of word form learning as well as salience or privileging with regard to attention to word fragments during word-level input processing. Based on previous research findings, we hypothesized that there would be a large number of partial words and more so than full words, that there would be preponderance of partial words in the one-quarter-word range, that more oneand two-letter fragments would be retrieved than larger three or more letter segments, and that word-initial segments of words would be privileged and appear more frequently than segments in word-medial and word-final positions. We also anticipated that some of the answers to the four primary research questions would be moderated by target language (German, Spanish), number of syllables (two or three), or both, but given that these aspects of the present study were largely exploratory in nature, we sought to allow the data obtained and the results of the study speak to these issues directly without overspeculating at the hypothesis stage.

METHOD Participants The original pool of participants eligible for this study were 54 first-semester learners of German and 87 first-semester learners of Spanish from two large universities in the Midwest of the USA. First semester courses are for complete beginners who have never taken any language classes and students who have placed into the course after taking a placement test. Among the L2 German students, data provided by seven students in the German group were excluded because five of these participants correctly translated one or more of the


4 Is there a difference between the percentage of target letters produced in some locations (word-initial, word-medial, or word-final)? If the answer is ‘yes’:


target words on the pretest and two of these participants did not complete the production test. After these reductions, the remaining sample of L2 German participants was 47. Among the L2 Spanish participants, data provided by 11 participants were excluded because these participants correctly translated one or more of the target words on the pretest; data provided by seven participants were excluded because English was not the native language of these participants; and data from one participant were excluded because the participant did not answer required questions on the language background questionnaire. After these reductions, the remaining sample of L2 Spanish participants was 68.

The following materials were used to conduct the study: (i) a consent form and a language background questionnaire; (ii) instructions for the experiment; (iii) a pretest with the 24 target words for translating any known words from L2 to L1; (iv) two posttests: a productive test for L1-to-L2 translation and a receptive test for L2-to-L1 translation (data related to the receptive test were not the focus of the analyses of the present study); and (v) a vocabulary learning strategies questionnaire (data from the questionnaire also were not the focus of the analyses of the present study).

Target words Twenty-four English–German and English–Spanish word pairs were compiled. Each target word was a concrete noun and had a counterpart in the other L2 (Spanish or German) that was equal in number of syllables and number of letters. For example, the Spanish word borla ‘tassel’ had the German word Kro¨te ‘frog’ as its counterpart. These two words constituted a GermanSpanish word pair. Both words had two syllables and five letters. All of the target words appear in Table 1. Each German–Spanish word pair (with the same number of syllables and letters) appears in each row. Some of the target nouns were regional variants, such as the German word Bulette ‘hamburger’ and the Spanish word chiringa ‘kite’. The length of the target words ranged between five and nine letters (Table 1).

Instrumentation Pretest In order to ensure that participants were unfamiliar with the chosen target words, an L2-to-L1 translation pretest was administered. At this point participants were not told that they would be asked to learn these words during the treatment phase that immediately followed the pretest. Participants were provided with a sheet that contained the 24 target words and were asked to


Materials


Table 1: Experimental words categorized according to language (German and Spanish), number of syllables (three and four) and number of letters German

Spanish

Two-syllable words

5 5 5 5 5 5 6 6 6 6 6 7

Kro¨te (frog) Erbse (pea) Ziege (goat) Kelle (soup spoon) Eimer (bucket) Nelke (carnation) Wurzel (root) Felsen (rock) Bu¨chse (can) Glocke (bell) Quaste (tassel) Rutsche (slide)

borla (tassel) balde (bucket) pinza (clothespin) trompo (spinning top) naipe (playing card) gaita (bagpipe) flecha (arrow) gancho (hook) clavel (carnation) choclo (corn) tuerca (nut) plancha (iron)

Three-syllable words

6 7 7 7 7 7 7 7 7 7 8 9

Kasuar (ostrich) Matrose (sailor) Gebaüde (building) Bulette (hamburger) Gebirge (mountains) Kolibri (hummingbird) Gema¨lde (painting) Peru¨cke (wig) Patrone (cartridge) Forelle (trout) Radierer (eraser) Bedienung (waiter)

arroyo (stream) colibrı´ (hummingbird) taladro (drill) gorrioń (sparrow) volante (steering wheel) tenazas (pliers) candado (lock) clavija (plug) Serrate (saw) toronja (grapefruit) chiringa (kite) churrusco (caterpillar)

provide an L1 translation of any word that they knew or thought that they might know. The L2-to-L1 translation task was selected to determine whether learners had a form-meaning link stored in their mental lexicon. The researchers chose to not use an L1-to-L2 production task because one can not exclude the possibility that a participant knew a target word but gave an alternate L2 translation. For example, if a participant had translated ‘corn’ as maiz (‘corn’ in many Spanish dialects), we could not have been sure that the word form choclo (‘corn’ in an Andean Spanish dialect) was unknown to the participant. Similarly, if a participant had translated ‘stream’ as rıó (‘river’ in Spanish), we could not have been sure that the word form arroyo ‘stream’ was unknown to the participant. The words on the pretest appeared in a different randomized


Letters


order than the order used in the exposure phase to avoid habituating the participants to the exposure phase order.

Posttest For the L1-to-L2 production posttest, the participants received a sheet with 24 lines with one line for each target word. Instructions on the posttest encouraged participants to write as much of each L2 word as they could, even if they could not recall the entire word.

Procedure

1 Each participant completed the consent form and language background questionnaire. 2 All of the participants completed the pretest. 3 Each participant received general instructions about the procedure. All of the participants were instructed to do their best to learn the 24 German or Spanish words and were informed that there would be a vocabulary test after the learning phase. 4 The 24 target word pairs were presented on a screen at the front of the classroom using a computer presentation program. All word pairs were presented in the center of the screen. In contrast to the Spanish–English pairs where all letters were in lower case, the German nouns in the German–English pairs were capitalized in accordance with the convention of the language. The 24 target word pairs were divided in three groups of eight words. Each group of eight words was presented twice in the same order for 6 s each. The second and the third set of eight words were presented in the same manner. The presentation of each set of eight target pairs was announced by a separate screen display stating that the first second or third group of eight words would be presented next. This display lasted 6 s. 5 One minute after the exposure phase, the L1-to-L2 posttest began. The L1 translations of individual target words were displayed on the screen and participants were asked to write as much of each target word as they could in a line space provided. Two incorrect references in the Spanish version of the presentation program (one case of ‘pictures’ instead of ‘translations’ and an incorrect title for the L1-to-L2 production-oriented posttest) were resolved and clarified as needed by the experimenter. The experimenter for the Spanish students was a speaker of both English and Spanish. The experimenter for the German students was a speaker of both English and German. (Note again that a second posttest, an L2-to-L1 translation posttest and a brief learning strategies


All data were collected in one approximately 50-min class session in the participants’ regular German and Spanish classrooms. The following procedures were followed:


questionnaire were also administered after the L1-to-L2 production posttest, but these tasks, all performed after the tasks of interest in the present study, were related to a exploratory investigation, were not related to the present study, and could not have impacted upon the results of the present study based on the L1-to-L2 production posttest, which was completed in full prior to the other tasks.)

Scoring

Analyses Overall word gain (LPSP-written) for German and Spanish In order to assess whether the overall performance level of the L2 German and L2 Spanish participants was significantly different, total LPSP-written scores for each participant were included in the statistical analysis. For this analysis, total LPSP-written scores for each participant were submitted to univariate analysis of variance (ANOVA) with language (German, Spanish) as the independent variable and LPSP-written score as the dependent variable. This analysis was helpful in determining whether additional adjustments would be needed to address the research questions of the study. If overall LPSP-written scores of the two groups were significantly different, for example, percentage scores or covariance analyses might have been needed to address the four main research questions of the study.

Partial versus fully produced words For the analysis of overall partial versus fully produced words (Research Question 1), the number of times each participant scored 1 was compared to the sum of the number of times each participant scored .25, .50, or .75. In order to determine if the number of syllables of a word (two or three syllables) affected full and partial word learning, separate scores were tabulated for two-syllable and three-syllable words. Sums of partial words and fully produced words were submitted to a repeated measures ANOVA. Type of production (partial, full) was treated as a within-subjects independent variable. Language (German, Spanish) and number of syllables (two, three) were treated as between-subjects independent variables. Number of items produced was the dependent variable.


The L1-to-L2 posttest was scored first using the lexical production scoring protocol (LPSP-written), developed by Barcroft (2000a). Following this protocol, a score of 1 was assigned to any completely produced word. Other scores were assigned as follows: .25 if approximately one-quarter of the word was produced, .50 if approximately half of the word was produced, .75 if approximately three-quarters of the word was produced, and 0 in all other cases.


Amount of partial word produced

Fragment length For the analysis on fragment length (Research Question 3), tallies were made from the original posttests to determine the number of times each participant produced one-letter, two-letter, and three-letter fragments as well as fragments with four or more letters, in any location of the target word. An overall score for two- and three-syllable words and separate scores for two- and three-syllable words were calculated. Sums of the number of times each participant produced one-letter, two-letter, and three-letter fragments were submitted to a repeated measures ANOVA. Fragment length (one-letter, two-letter, three-letter, four or more letters) was treated as a within-subjects independent variable. Language (German, Spanish) and number of syllables (two, three) were treated as between-subjects independent variables. Number of items produced was the dependent variable.

Word-internal location For the analysis on word-internal location (Research Question 4), tallies were made from the original posttests to determine the percentage of letters correctly produced in word-initial, word-medial, and word-final segments of the target words. For words with three syllables: (i) word-initial percentages were determined by dividing the number of correctly produced letters in the first syllable by the total number of letters in the first syllable; (ii) word-final percentages were determined by dividing the number of correctly produced letters in the last syllable by the total number of letters in the last syllable; and (iii) word-medial percentages were determined by dividing the number of correctly produced letters in the rest of the word by the total number of the letters in the rest of the word. Scores for two-syllable words were determined in a similar manner; however, word-initial position was based on the first letter of the word only, word-final position based on the last letter only, and word-medial position based on the percentage of the remaining letters. An alternative way to do this analysis would be to allow for the entire first


For the analysis on amount of word produced in partially produced words (Research Question 2), the number of times each participant scored .25 (one-quarter of a word), .50 (one-half of a word), and .75 (three-quarters of a word) was compared. Sums of scores of one-quarter (.25), one-half (.50), and three-quarters (.75) were submitted to a repeated measures analysis of variance. Amount of word (one-quarter, one-half, three-quarters) was treated as a within-subjects independent variable. Language (German, Spanish) and number of syllables (two, three) were treated as between-subjects independent variables. Number of items produced was the dependent variable.


RESULTS Overall word gain (LPSP-written) for German and Spanish Means for LPSP-written scores by language were 10.06 (SE = .63) for L2 German students and 10.50 (SE = .52) for L2 Spanish students. Results of the ANOVA indicated that the overall word gain of the two groups was not significantly different (p = .596). In light of this result, no adjustments were made with regard to the statistical analyses used to address the four main research questions of the study.

Research question 1: Partial versus fully produced words Means and standard errors for production type (partial word, full word) by language (German, Spanish) appear in Table 2. The substantially lower means in Table 2, as compared with the means above 10 reported earlier, are due to the inclusion of syllable (two and three syllables) as an independent variable and, consequently, the LPSP-written maximum mean having changed from 24 to 12 (maximum of 12 per each of the two levels of the syllable variable). Means and standard errors for production type by number of syllables in target words (two-syllable, three-syllable) appear in Table 2. Results of the ANOVA revealed a significant main effect for production type, F(1,113) = 27.24, p < .001, Z2 = .194, and the following significant interactions: (i) Production type Language, F(1,113) = 4.17, p = .036, Z2 = .036; (ii) Production type Number of syllables, F(1,113) = 17.34, p < .001, Z2 = .133; and (iii) Language Number of syllables, F(1,113) = 17.33, p < .001, Z2 = .133. The Production type Language interaction, which can be seen in Figure 1, was due to the greater difference in the production of partial versus full words among L2 German participants as compared with the L2 Spanish participants.


syllable and last syllable of two-syllable words to be used to score word-initial and word-final segments. While this option can be explored in future studies, in this initial exploration we allowed for the assumption that attention can, at least potentially, be paid to initial/final letters without necessarily being paid to initial/final syllables of words. An overall score for two- and three-syllable words as well as separate scores for two- and three-syllable words were calculated. Totals for the percentages of the three locations for all 24 words were submitted to a repeated measures analysis of variance. Location (word-initial, word-medial, word-final) was treated as a within-subjects independent variable. Language (German, Spanish) and number of syllables (two, three) were treated as between-subjects independent variables. Number of items produced was the dependent variable. Means reported are estimated marginal means. For all statistical analyses, alpha was set at .05.


Table 2: Means for production type in target words by language and by number of syllables Production typea Scoring

Full word

a

Total

M

SE

M

SE

M

SE

2.55 3.05 2.80

0.28 0.24 0.18

4.47 3.89 4.18

0.26 0.21 0.17

3.51 3.47 3.49

0.18 0.15 0.12

3.26 2.34 2.80

0.24 0.18 0.18

3.82 4.53 4.18

0.20 0.22 0.17

3.54 3.44 3.49

0.14 0.12 0.12

N = 115 for each cell.

Number of Productions

5

4

3 German Spanish 2

1

0 Full Word

Partial Word Production Type

Figure 1: The production type language interaction The Production type Number of syllables interaction, which can be seen in Figure 2, was due to the greater difference in production of partial words for three-syllable as compared with two-syllable target words. The Language Number of syllables interaction was due to L2 German participants having higher means for overall productions (including both full and partial


Language German Spanish Total Number of syllables Two syllables Three syllables Total

Partial word



6 5 4 2-syllable Words 3-syllable Words

3 2 1

Full Word

Partial Word

Production Type

Figure 2: The production type number of syllables interaction

Table 3: Means for amount of word by language and by number of syllables Language Amount German

0.25 0.50 0.75 Total

Number of syllables Spanish

Total

Twosyllable

Threesyllable

Total

M

SE

M

SE

M

SE

M

SE

M

SE

M

SE

1.16 1.17 2.14 1.49

0.13 0.11 0.17 0.09

0.87 1.15 1.88 1.30

0.11 0.09 0.14 0.07

1.01 1.16 2.01 1.39

0.08 0.07 0.11 0.06

0.87 0.87 2.08 1.28

0.09 0.08 0.15 0.07

1.15 1.45 1.93 1.51

0.12 0.11 0.14 0.07

1.01 1.16 2.01 1.39

0.08 0.07 0.11 0.06

n = 47 for German; n = 68 for Spanish; N = 115 for total.

productions) of three-syllable words, in contrast to the L2 Spanish participants’ higher means for overall productions of two-syllable words. No other significant main effects or interactions were revealed.

Research question 2: Amount of partial word produced Means and standard errors for amount of partial word produced (.25, .50, .75) by language (German, Spanish) appear in Table 3. Means and standard errors for amount of partial word produced by number of syllables in target words (two-syllable, three-syllable) also appear in Table 3. Results of the ANOVA


0


2.5

Number Produced

2

1.5 2-syllable Words 3-syllable Words 1

0.5

0.25

0.5

0.75

Amount of Word

Figure 3: The amount of word number of syllables interaction revealed significant main effects for amount, F(2,226) = 36.66, p < .001, Z2 = .245, and number of syllables, F(1,113) = 8.44, p = .004, Z2 = .070, as well as a significant interaction between amount and number of syllables, F(2,226) = 6.22, p = .002, Z2 = .052. The difference in amount of partial word produced for the two languages (German, Spanish) was not statistically significant (p = .084). Pairwise comparisons indicated that the .75 level of partial word (approximately three-quarters of a word) was produced significantly more than the .25 level, p < .0001, and significantly more than the .50 level. As can be seen in Figure 3, the Amount Number of syllables interaction was due to the markedly higher number of productions of .75 as compared with productions of .25 and .50 for the two-syllable words, a pattern that differed from that of the three-syllable words. No other significant main effects or interactions were revealed.

Research question 3: Fragment length Means and standard errors for fragment length (one-letter, two-letter, three-letter, four or more letters) by language (German, Spanish) appear in Table 4. Means and standard errors for fragment length by number of syllables in target words (two-syllable, three-syllable) appear in Table 4. Results of the ANOVA revealed significant main effects for fragment length, F(3,339) = 48.04, p < .001, Z2 = .298, and syllable, F(1,113) = 11.05, p = .001, Z2 = .089, as well as a significant interaction between fragment length and language, F(3,339) = 10.75, p < .001, Z2 = .087. The difference in amount of fragment lengths produced based on language (German, Spanish) was not statistically significant (p = .088). Pairwise comparisons indicated that the following


0


Table 4: Means for fragment length by language and by number of syllables Language German

Spanish

Total

M

SE

M

SE

M

1.96 1.60 1.19 1.07

0.25 0.15 0.11 0.12

3.13 1.38 1.14 1.12

0.21 0.21 0.09 0.10

2.55 1.49 1.17 1.10

Twosyllable

Threesyllable

Total

SE

M

SE

M

SE

M

SE

0.16 0.10 0.07 0.08

2.39 1.26 1.04 0.86

0.20 0.12 0.09 0.09

2.70 1.71 1.30 1.34

0.21 0.20 0.10 0.12

2.55 1.49 1.17 1.10

0.16 0.10 0.07 0.08

1.46 0.11 1.69 0.09 1.57 0.07 1.39 0.09 1.76 0.09 1.57 0.07



4

3

German

2

Spanish

1

0 1

2

3

4 or more

Fragment Length (Number of Letters)

Figure 4: The fragment length language interaction differences were statistically significant: (i) one-letter versus two-letter, three-letter, and four or more letters (p < .0001 in each case) and (ii) two-letter versus three-letter (p = .002) and four or more letters (p = .0001). The Fragment length Language interaction, which can be seen in Figure 4, was due to the markedly higher mean number of one-letter fragments produced by the L2 Spanish participants as compared with the L2 German participants. No other significant main effects or interactions were revealed.


One-letter Two-letters Three-letters Four or more letters Total

Number of syllables


Table 5: Means for word-internal location by language and by number of syllables Language Location German

Spanish

Total

Twosyllable

Threesyllable

Total

M

SE

M

SE

M

SE

M

SE

M

SE

M

SE

1.90 0.45 1.06 1.14

0.29 0.12 0.13 0.13

2.92 1.87 1.63 2.14

0.16 0.16 0.11 0.10

2.41 1.16 1.35 1.64

0.13 0.08 0.08 0.08

2.28 0.95 0.97 1.40

0.16 0.10 0.10 0.10

2.54 1.36 1.72 1.88

0.16 0.11 0.13 0.12

2.41 1.16 1.35 1.64

0.13 0.08 0.08 0.08


Research question 4: Word-internal location Means and standard errors for word-internal location (initial, medial, final) by language (German, Spanish) appear in Table 5. Means and standard errors for word-internal location by number of syllables in target words (two-syllable, three-syllable) appear in Table 5. Results of the ANOVA revealed significant main effects for location, F(2,226) = 92.50, p < .001, Z2 = .450, and number of syllables, F(1,113) = 11.97, p = .001, Z2 = .096, as well as the following significant interactions: (i) Location Language, F(2,226) = 9.29, p < .001, Z2 = .076; (ii) Location Number of syllables, F(2,226) = 4.87, p = .009, Z2 = .041; and (iii) Language Number of syllables, F(1,113) = 8.94, p = .003, Z2 = .073. The potential Location Language Number of syllables interaction was not statistically significant (p = .060). Pairwise comparisons indicated that overall means were significantly higher for word-initial location means than for word-medial location, p < .001, and word-final location, p < .001 and that means for word-final location were significantly higher than for word-medial location, p = 022, even though actual means for L2 Spanish were lower in word-final than in word-medial position. The Location Language interaction, which can be seen in Figure 5, was due to lower means for medial as compared with final position in L2 German versus lower means for final as compared with medial in L2 Spanish. The Location Number of syllables interaction, which can be seen in Figure 6, was due to markedly higher means in final over medial position for three-syllable words. The Language Number of syllables interaction was due to the much higher means for overall productions (including productions determined for all word-internal locations) for three-syllable over two-syllable words for L2 German but not for L2 Spanish. No other significant differences or interactions were revealed.


Initial Medial Final Total

Number of syllables


4

Total Score

3

German Spanish

2

1

Initial

Medial

Final

Location

Figure 5: The location language interaction


3

2 2-syllable Words 3-syllable Words 1

0 Initial

Medial

Final

Location

Figure 6: The location number of syllables interaction

DISCUSSION The findings of the study can be summarized as follows. (i) Partial words were produced much more often than complete words. Approximately 49 percent more partial words (M = 4.18) were produced than whole words (M = 2.80). (ii) Two-syllable words were fully produced more often than three-syllable words. (iii) The difference in the amount of partial versus full words was


0


The pervasiveness of partial word form learning Learners in both languages produced substantially more partial than complete words after engaging in the intentional vocabulary learning task. This finding illustrates the very large extent to which the formal component of L2 word learning is an incremental process. The study revealed evidence that a majority of the L2 word forms were learned in bits and pieces and were not picked up as single entities. Although this finding may be somewhat intuitive, the study provided evidence that to a large extent partial word learning is the norm and not the exception, which leads to the following question. If L2 words are learned incrementally in bits and pieces, is this bit-by-bit incremental learning process random or, alternatively, are there identifiable patterns in the manner in which the bits and pieces, or word fragments, are learned? To clarify our use of the term incremental here is no way meant to deny the possibility of different states of vocabulary knowledge at different times, including the fact that for a variety of possible reasons, a learner may be able to recall one part of a word form at a given time and then not be able to recall that part at a later time. The findings of the present study strongly suggest that there are a number of identifiable patterns and that some parts of words are processed more readily than others. Whereas a substantial amount of previous research has focused on the gradual development of semantic and syntactic information during L2 word learning, the present findings demonstrate the need to attend more to the gradual development of knowledge of target word form. Word forms that have been encoded only partially will require additional processing resources


larger for learners of German than for learners of Spanish. (iv) Three-quarters of a word was produced more often than one-half or one-quarter of a word, and the privileging for three-quarters of a word was more pronounced for two-syllable than for three-syllable words. (v) One-letter fragments were produced much more often than two-letter fragments, three-letter fragments, and fragments with four or more letters. Single-letter fragments were produced 40 percent of the time; two-letter fragments 24 percent of the time, three-letter fragments 19 percent of the time, and four-or-more-letter fragments 17 percent of the time. (vi) Production of target letters was greater in word-initial locations than in other locations. Target letters were produced 49 percent of the time in word-initial position, 24 percent of the time in word-medial position, and 27 percent of the time in word-final position. (vii) The word-initial advantage was more marked for two-syllable words than for three-syllable words. (viii) Production patterns for medial and final position were different for the two L2s. While learners of German produced markedly fewer fragments in word-medial position than in word-final position, this pattern was not maintained for learners of Spanish for whom actual means were slightly higher for word-medial than for word-final position. The implications of these findings are discussed below.



during subsequent encounters with the word. This process of allocating processing resources over time in order to move, incrementally, from partial to complete word form knowledge needs to be considered a central component of L2 vocabulary learning. As such, this form-oriented learning process needs to be given more attention in L2 vocabulary research. Furthermore, given that L2 word learning is incremental both in terms of word form (partial word form learning over time) and word meaning (learning L2-appropriate semantic space and usage over time), tension may arise with regard to the need to allocate resources to learn new bits of word form, on the one hand, and the need to learn new bits of word meaning, on the other hand. Consistent with this view are the increasing number of demonstrations of the potential tension between processing for word form and word meaning during lexical input processing in the research literature (see, e.g. Barcroft 2002, 2003, among others). The present finding about the pervasiveness of partial word learning also has important implications with regard to testing methods in vocabulary research. The results of the study suggest that productive vocabulary knowledge can be broken down and investigated by analyzing production patterns for both partial and fully produced words that learners have recently attempted to learn. This methodology may be beneficial in future studies with regard to measuring degree of knowledge of the formal component of word knowledge. By encouraging participants to produce partially learned words on their own, one may obtain a more precise measure of the degree to which each target word form has been learned. Additionally, in terms of scoring, a large amount of information can be lost if one scores productions only as 0 or 1 or even as 0, .5, or 1. The present evidence about partial word form learning affirms the need for scoring procedures that go beyond two or three tiers in order to obtain more precise measurements of different degrees of knowledge of individual L2 words. An important issue for future research will be the use of additional analyses to assess the learning of word forms. Using morphemes or syllable boundaries as a unit of analysis may provide additional insights into how L2 learners chunk new word forms. Expanding on the present methodology to examine knowledge beyond base forms, such as systematic knowledge of word inflections and derivations, and vocabulary learning with learners at more advanced levels than that of the participants in the present study, should help to provide new insights as well. The present study also tested the extent to which learning full versus partial words might be affected by number of syllables in a word, the target L2 in question, or both. The interaction observed between language and production type, complete versus partial words (Figure 1), suggests that beginning learners of German may have more difficulty encoding target words as entire units than beginning learners of Spanish or, alternatively, that beginning learners of German are more capable of picking up partial words. Our findings on production type, complete versus partial words, also demonstrated a significant interaction between language and production


Amount of word produced in partial word form learning The findings on the amount of a target word produced indicate that the learners in both L2s often came close to learning a target word completely by producing more often three-quarters of the target word than one-half and one-quarter of the target word. One important methodological implication of this finding is that the high level of nearly but not completely full word form knowledge observed in the study most likely would not be discernible using a receptively oriented measure of vocabulary knowledge in which the target word form is provided for the learner (e.g. L2-to-L1 translation) because the missing formal knowledge when one knows approximately three-quarters of a word would not prevent one from recognizing the word when it is presented. Arguably, the same may be true for the one-half and even the one-quarter levels of productive knowledge, but most likely not to the same extent as when one is at the three-quarters level. We also should note the possibility of increased confusion between different partially known words when one has learned an increasingly smaller amount of an entire word form. The less overall word form one has learned, the more that word may be confused with other partially learned words, at least potentially. Additionally, the larger number of productions at the three-quarters level was more pronounced for two-syllable words than for three-syllable words, which yielded the significant interaction between number of syllables and amount of word (Figure 3). One possible explanation of this pattern is that because shorter words tend to be easier to learn (e.g. Ellis and Beaton 1993), learners tend to reach disproportionately higher levels of partial word form knowledge (i.e. three-quarters instead of one-quarter or one-half) for those words.


type (Figure 2). This interaction suggests that attempting to learn words with more syllables (in this case, only one more syllable) yields a greater amount of partial word learning than whole word learning. One possible explanation of this finding is that words with more syllables have more word fragments available to be learned, allowing for a greater number of potential fragments. Another explanation is that learners tend to attend more to shorter two-syllable words overall and therefore learn all of the word fragments within those words more often than is the case with three-syllable words. A third possible explanation would be to combine the first two explanations as both of these factors may contribute to the pattern observed. What is important is that this interaction suggests something about partial word learning that is predictive in nature: learners are more likely to have only partial knowledge of longer words and more likely to have full knowledge of shorter words. The generalizability of this working hypothesis can be examined in future studies on production type (full or partial word) for words with a larger range of word lengths, such as words with one, two, three, four, and five syllables.


Fragment length in partial word form learning


Although the participants in the study often reached the level of three-quarters of a word in their productions, they also produced a large number of smaller fragments within target words overall, including an especially large number of single-letter and bigram fragments. One explanation of this finding involves focusing on the incremental nature of fragment learning. Like the larger incremental process of word form learning, fragment learning can be incremental. Given that single letters were the smallest possible fragment in each target word in the present study, there were more opportunities for learners to encode single-letter fragments by default when attempting but failing to learn larger fragments within the word. Additionally, the beginning L2 learners may have lacked familiarity with the phonotactic patterns of their respective L2s that would be required to be more successful at chunking the target words via multi-letter units, such as phonemes, morphemes, syllables and prosodic patterns (Ellis and Beaton 1993; Ellis 2001). This lack of familiarity may have contributed to a greater number of failures at multi-letter chunking, leading to more cases of encoding single letters only. An interaction between fragment length and language (Figure 4) also was observed in this study. The particularly high number of one-letter fragments among Spanish learners may be related to how many Spanish words contain the letters ‘a’ or ‘o’ and how learners of Spanish become sensitive to this pattern, which could have given rise to the higher percentage of one-letter fragments observed for the learners of Spanish as compared with the learners of German. Given that English and German are both Germanic languages, the learners of German may have been more familiar with phonotactic patterns in German than the learners of Spanish, which is a Romance language. From this perspective, the greater phonotactic familiarity of the learners of German may have allowed them to be more effective at chunking multi-letter sequences, thereby having fewer instances of single-letter productions than the learners of Spanish. It is also possible that the activation of the L1 word form influenced processing of the L2 word forms (Sunderman and Kroll 2006). The target words used in the current investigation do not allow the identification of specific processing patterns that may be due to the L1. Future research should assess the effect of the L1 on L2 word processing by manipulating the choice of target words. Additionally, studies should further investigate whether one’s level of phonotactic competence develops at the same pace in learners of different L2s and determine the relationship between phonological short-term memory and word form learning. Findings from psychological research suggest that phonological short-term/working memory plays an important role in the acquisition of L2 word forms (e.g. Speciale et al. 2004; Kovaćs and Racsmany 2008). It is assumed that a novice learner comes to the task of L2 acquisition with a particular phonological short-term memory capacity, and it is this that limits their rate of L2 acquisition. Yet, as L2 learners develop phonotactic competence and knowledge of phonological regularities of the L2, phonological


short-term memory capacity increases and subsequently enhances word form learning. Finally, the data on fragment length also provided some evidence against solely linear processing during the initial stages of word form encoding. The findings suggest that learners did not systematically segment off large fragments of a word in some order of succession in order to encode these fragments before attending to other segments. Instead, it seems that the learners allocated processing resources to various fragments in more of a parallel and interactive manner.

The analysis on location provided new evidence in favor of the idea that word-initial position is privileged during the initial stages of L2 word learning, corroborating previous findings (Barcroft 2000b, 2008). The overall percentage of target letters in word-initial positions was greater than percentages in word-medial or word-final positions case for two-syllable and three-syllable words. This result provides new evidence for extending the primacy-effect component of the bathtub effect to the initial stages of L2 word learning (Brown and McNeill 1966). The effect of location in the present study differed from the bathtub effect, however, due to the particularly high means for word-initial location as compared with the other two locations (Figure 5). A more appropriate name for the effects of location in the present study would be recliner effect. Whereas the term ‘bathtub effect’ makes use of the metaphor of someone in a bathtub with their head and feet out of the water, the present findings and those of Barcroft (2000b, 2008) are better depicted by the metaphor of someone in a reclining chair with their head higher than the middle of their body and their feet. One possible reason that word-initial positions are favored is simply that they are more easily perceived. An alternative, however, concerns efficient parsing. Even if word-initial positions are not perceptually more salient, learners may parse word-initial positions to a greater extent as a strategy that allows for the most efficient word-level parsing overall. A third possibility is that word-initial privileging is due to some combination of both of the above and is in part due to perceptual salience and in part due to parsing mechanisms employed by the learner. To further examine the distinction in pattern between the bathtub effect, previously observed in TOT states in the spoken mode, and the recliner pattern observed in the present study, new research on partial word production in the spoken mode is needed in order to assess further whether the recliner pattern is specific to the written mode. The interaction between location and language (Figure 5) suggests that English-speaking learners of German processed word fragments in medial and final position differently than learners of Spanish. Learners of German produced substantially fewer fragments in word-medial position than in final


Fragment location in word form learning



position. Learners of Spanish, however, produced almost equal amounts of word fragments in word-medial and word-final positions. Returning to the recliner metaphor, the learners of German have their feet propped higher than the learners of Spanish in the recliner chair, but note that the head of the recliner remains the highest part of the recliner for both languages. One possible explanation of this finding is, again, that the different learning patterns may be related to the extent to which the English-speaking learners of German were familiar with the phonotactic patterns and other properties of word in German due to English being a Germanic language. Although highly speculative, this higher degree of familiarity among learners of German somehow may have allowed them to attend to patterns in word-final position more readily as compared with the learners of Spanish. The lower degree of familiarity among the learners of Spanish may have decreased their ability to demonstrate a potential word-internal recency effect because processing resources needed to do so were exhausted attending to properties of the Spanish words with which they were less familiar as compared to the case of another Germanic language. In future studies that expand upon the methodology of the present study, one limitation of the present investigation that can be addressed, albeit perhaps not easily, is that of the lexical characteristics, particularly word frequency, of the target word pairs. Although in the present study Spanish and German words were matched for number of letters and number of syllables, they were not matched according to word frequency in the languages in question (i.e. frequency of the word for Kro¨te ‘frog’ in German versus frequency of the word borla ‘tassle’ in Spanish and the relative frequency of the English counterparts of these two words in English, the participants’ L1). Finding word pairs that match on the formal properties for which they matched in the present study as well as lexical characteristics, such as word frequency, poses a challenging methodological task. One viable option would be to use pseudowords with properties of Spanish and German respectively in order to be able to use the same referent (e.g. frog) for both languages. With this option one would no longer be dealing with real words in the target languages, but from a methodological standpoint, it would facilitate matching target words on lexical properties because there would be only one referent for each two varieties of target word form in the respective languages. The issue of creating target pseudowords that are sufficiently ‘German-like’ and ‘Spanish-like’, and equally so for each referent in question, would pose another challenge with this option, however. Another reason that we emphasize the exploratory nature of the present study concerns the potential of how individual-difference variables between the Spanish versus the German learners, such as memory span, may have affected the present results. Future studies might recruit naı¨ve absolutebeginner participants who attempt to learn two sets of vocabulary, one in each language, on two separate occasions, with a rest period in between, and counterbalance learning orders across participants.


REFERENCES Aitchinson, J. and M. Straf. 1981. ‘Lexical storage and retrieval: a developing skill,’ Linguistics 19: 751–95. Baddeley, A. D. and G. J. Hitch. 1974. ‘Working memory’ in G. A. Bower (ed.): Recent Advances in Learning and Motivation. Vol. 8. Academic Press, pp. 47–90. Barcroft, J. 2000a. ‘The effects of sentence writing as semantic elaboration on the allocation of processing resources and second language lexical acquisition’ Unpublished Doctoral Dissertation. University of Illinois at Urbana-Champaign. Barcroft, J. 2000b. ‘The nature of partial word form learning,’ Paper presented at the 4th Hispanic Linguistics Symposium. Bloomington, IN. Barcroft, J. 2002. ‘Semantic and structural elaboration in L2 lexical acquisition,’ Language Learning 52/2: 323–63. Barcroft, J. 2003. ‘Effects of questions about word meaning during L2 lexical learning,’ The Modern Language Journal 87/4: 546–61. Barcroft, J. 2008. ‘Second language partial word form learning in the written mode,’ Estudios de Linguistica Aplicada 26/47, 53–72. Barcroft, J. and B. VanPatten. 1997. ‘Acoustic salience of grammatical forms: the effect of

location, stress, and boundedness on Spanish L2 input processing’ in A. Pe´rez-Leroux and W. Glass (eds): Contemporary Perspectives on The Acquisition of Spanish. Volume 2: Production, Processing, and Comprehension. Cascadilla Press, pp. 109–22. Boogards, P. 2001. ‘Lexical units and the learning of foreign language vocabulary,’ Studies in Second Language Acquisition 23: 321–43. Brown, R. and D. McNeill. 1966. ‘The ‘‘tip of the tongue’’ phenomenon,’ Journal of Verbal Learning and Verbal Behavior 8: 325–37. Conrad, M. and A. M. Jacobs. 2004. ‘Replicating syllable frequency effects in Spanish in German: one more challenge to computational models of visual word recognition,’ Language and Cognitive Processes 19: 369–90. De Groot, A. M. B. and R. Keijzer. 2000. ‘What is hard to learn is easy to forget: the roles of concreteness, cognate status, and word frequency in foreign language vocabulary learning and forgetting,’ Language Learning 50: 1–56. Ellis, N. C. 2001. ‘Memory for language’ in P. Robinson (ed.): Cognition and Second


Overall, our findings on location and those of Barcroft (2000b, 2008) suggest that privileging for word-initial location may be a general principle of L2 word form learning. Further investigation is needed, however, in order to examine the extent to which word-final position may be privileged over word-medial position in different vocabulary learning contexts and for different L2s. Likewise, it would be helpful to assess in future research the extent to which privileging for word-initial location is maintained with words that contain more than three syllables as word length gradually increases and with learners at increasingly higher levels of L2 proficiency. Additionally, given that other aspects of word knowledge and word learning may come into play, such as degree of knowledge of the L1 word and personal preferences for different types of words, future studies can be designed to assess the independent contributions of factors such as these in word form learning. Finally, control tests such as digit span or English word-form recall can be incorporated to make between-group comparisons, such as between the L2 Spanish and L2 German groups, more meaningful (although perhaps more challenging to do so in classroom-based research). Systematic modifications of this nature in future research should help to test the generalizability of increased learnability of word-initial fragments and other presently identified patterns in partial word form learning.


Laufer, B. and P. Nation. 1995. ‘Vocabulary size and use: Lexical richness in L2 written production,’ Applied Linguistics 16: 307–22. Meara, P. 1982. ‘Word association in a foreign language. A report on the Birkbeck vocabulary project,’ Nottingham Linguistics Circular II: 29–38. Meara, P. and P. Buxton. 1987. ‘An alternative to multiple choice vocabulary tests,’ Language Testing 4: 142–51. Meara, P. and S. Ingle. 1986. ‘The formal representation of words in an L2 speaker’s lexicon,’ Second Language Research 2: 160–71. Melka, F. 1997. ‘Receptive vs. productive aspects of vocabulary’ in N. Schmitt and M. McCarthy (eds): Vocabulary: Description, Acquisition and Pedagogy. Cambridge University Press, pp. 84–102.

Meyer, A. and K. Bock. 1992. ‘The tip-of-thetongue phenomenon: Blocking of partial activation,’ Memory and Cognition 20: 715–26. Nation, P. 1982. ‘Beginning to learn foreign vocabulary: A review of the research,’ RELC Journal 13: 14–36. Qian, D. 1999. ‘Assessing the roles of depth and breadth of vocabulary knowledge in reading comprehension,’ Canadian Modern Language Review 56: 282–307. Read, J. 1993. ‘The development of a new measure of vocabulary knowledge,’ Language Testing 10: 355–71. Riegel, K. F. and W.M. Zivian. 1972. ‘A study of inter- and intralingual association in English and German,’ Language Learning 22: 51–63. Rodgers, T. S. 1969. ‘On measuring vocabulary difficulty: An analysis of item variables in learning Russian-English vocabulary pairs,’ IRAL 7: 327–43. Rosa, E. and M. O’Neill. 1998. ‘Effects of stress and location on acoustic salience at the initial stages of Spanish L2 input processing,’ Spanish Applied Linguistics 2: 24–52. Rott, S. 1999. ‘The effect of exposure frequency on intermediate language learners’ incidental vocabulary acquisition and retention through reading,’ Studies in Second Language Acquisition 21: 589–619. Schmitt, N. 1998. ‘Tracking the incremental acquisition of second language vocabulary: A longitudinal study,’ Language Learning 48: 281–317. Skehan, P. 1998. A Cognitive Approach to Language Learning. Oxford University Press. Skehan, P. and D. Foster. 1999. ‘The influence of task structure and processing conditions on narrative retellings,’ Language Learning 49/1: 93–120. Skehan, P. and P. Foster. 2001. ‘Cognition and tasks’ in P. Robinson (ed.): Cognition and Second Language Instruction. Cambridge University Press, pp. 183–205. Speciale, G., N. C. Ellis, and T. Bywater. 2004. ‘Phonological sequence learning and shortterm store capacity determine second language vocabulary acquisition,’ Applied Psycholinguistics 25/2: 293–32. Sunderman, G. and J. K. Kroll. 2006. ‘First language activation during second language lexical processing. An investigation of lexical form, meaning, grammatical class,’ Studies in Second Language Acquisition 28: 387–422.


Language Instruction. Cambridge University Press, pp. 33–68. Ellis, N. C. and A. Beaton. 1993. ‘Psycholinguistic determinants of foreign language vocabulary learning,’ Language Learning 43: 559–617. Henning, G. H. 1973. ‘Remembering foreign language vocabulary: acoustic and semantic parameters,’ Language Learning 23: 185–96. Henriksen, B. 1999. ‘The dimensions of vocabulary development,’ Studies in Second Language Acquisition 21: 303–17. Hulstijn, J. 2001. ‘Intention and incidental second language vocabulary learning: a reappraisal of elaboration, rehearsal, and automaticity’ in P. Robinson (ed.): Cognition and Second Language Instruction. Cambridge University Press, pp. 258–86. Koda, K. 1999. ‘Development of L2 intraword orthographic sensitivity and decoding skills,’ Modern Language Journal 83: 51–64. Kovaćs, G. and M. Racsmańy. 2008. ‘Handling L2 input in phonological STM: the effect of non-L1 phonetic segments and non-L1 phonotactics on nonword repetition,’ Language Learning 58/3: 597–624. Laufer, B. 1997. ‘What is in a word that makes it hard or easy: Some intralexical factors that affect the learning of words’ in N. Schmitt and M. McCarthy (eds): Vocabulary: Description, Acquisition and Pedagogy. Cambridge University Press, pp. 140–55. Laufer, B. and Z. Goldstein. 2004. ‘Testing vocabulary knowledge: Size, strength, and computer adaptiveness,’ Language Learning 54: 399–436.


VanPatten, B. 1996. Input Processing and Grammar Instruction. Theory and Research. Ablex. VanPatten, B., J. Williams, and S. Rott. 2004. ‘Form-meaning connections in second language acquisition’ in B. VanPatten, J. Williams, S. Rott, and M. Overstreet (eds):

Form-meaning Connections in Second Language Acquisition. Erlbaum, pp. 1–28. Wesche, M. and S. Paribakht. 1996. ‘Assessing second language vocabulary knowledge: depth versus breadth,’ Canadian Modern Language Review 53: 13–40.


Applied Linguistics: 31/5: 651–670 ß Oxford University Press 2010 doi:10.1093/applin/amq025 Advance Access published on 25 August 2010

Language Policy and Latina Immigrants: An Analysis of Personal Experience and Identity in Interview Talk 1

KENDALL A. KING and 2ANNA DE FINA

1

University of Minnesota and 2Georgetown University E-mail: [email protected]; [email protected]

Like much of the field of applied linguistics, scholarship within the area of language planning and policy (LPP) dates back to the late 1950s. While LPP scholarship is less than five decades old and has only recently developed into an organized field of study, it has undergone three substantial, paradigmatic shifts (Ricento 2000). Early LPP work (e.g. Fishman et al. 1968; Rubin and Jernudd 1971) tended to focus on efforts to resolve ‘language problems’, typically in post-colonial and economically developing nation-states, through positivist, technocratic, and linguistic solutions based on the effective management of linguistic resources. By the late 1970s, as it became apparent that the development (and language policy) efforts in these nations often failed to achieve their aims (Spolsky 2009) and that core language policy constructs such as ‘diglossia’ and ‘native speaker’ were far from neutral (e.g. Davies 1991; Woolard and Schieffelin 1994), LPP turned its focus to the role of language—and language policy in particular—in reproducing social and economic inequality both within and across nation-states. The current phase of LPP work, rooted in the 1990s, greatly expands these critical approaches as LPP research has focused on personal agency and emphasized the ways in which language policy interacts with individual and collective language ideologies and politics as well as with global forces such as large-scale


This article examines how Spanish-speaking Latina (im)migrants position themselves relative to US language policies. Drawing from interviews with 15 Latin American women in the USA, we illustrate how understandings of language policy are constructed through individuals’ reports of everyday experiences and framed within the constraints of contemporary language politics and ideological debates. Building on recent shifts in language policy research towards the study of how individuals engage with language policy issues, we argue for a close analysis of interview talk as a way of gaining important insights into this issue. Data from the study not only illustrate the existence of multiple frames for individual understandings of language policy, but also suggest how these discussions become terrains for the personal negotiation of identities within a broader socio-political context that is overwhelmingly anti-immigrant.

652 LANGUAGE POLICY AND LATINA IMMIGRANTS

THE WIDER FRAME: LANGUAGE POLICY AND LANGUAGE IDEOLOGIES IN THE USA Discourses about language both among and about immigrants in the USA cannot be meaningfully analyzed without a general understanding of how immigration and language are politically linked and racially charged. Indeed, in the USA, language policy debates are key sites of ideological struggle as they imply the legitimization or de-legitimization of languages and their users (e.g. Ricento 2000; Tollefson 2002; Crawford 2008; Garcıá 2008). A central


migration, minority language endangerment, and linguistic human rights struggles. Within this third paradigm, LPP research recently has focused greater attention on how individuals interpret and engage with language policies, and how they position themselves relative to those policies and to language politics more generally across a wide range of contexts (e.g. Canagarajah 2005; Warriner 2007; Johnson 2009). This greater focus on personal agency and individual experience is essential if we are to begin to answer the central (and still largely unaddressed) LPP question of ‘Why do individuals opt to use (or cease to use) particular languages and varieties for specified functions in different domains’ and how those choices influence—and are influenced by—‘institutional language policy decision-making?’ (Ricento 2000: 208). Such a focus on individual experiences and understandings moves us beyond language policy analysis of official documents as technocratic solutions. Instead, attention shifts towards analysis of reported language practices and discursive understandings of those practices as a ground for analysis of the mechanisms through which individuals and institutions shape—and are shaped by—language policy and the broader political climate. The present article builds on this expanding body of work by examining how individuals who we argue are among the implicit, and at times explicit, targets of US language policy—that is non- or limited English-speaking Latina (im)migrants—experience and position themselves relative to US language policy and politics. We take language policy to be deliberate, state, or institutionally sponsored efforts to influence the behavior of others with respect to the acquisition, structure, or functional allocation of language(s) (Cooper 1989), and language politics to be the political, social, and cultural consequences of such efforts. Our aim is to investigate how these women construct understandings of both the language policies and language politics that surround them, and how they relate these issues to personal identity within interviews. Considerable language policy research seeks to examine the impact of language policies targeting (im)migrant and language minority populations (e.g. Hult 2004; Gańdara and Rumberger 2009). Our approach differs in that we center our analysis on women’s reported experiences and understandings of language policy and politics, not on the merit or effectiveness of specific language policies.

K. A. KING and A. DE FINA 653


aim of LPP work in recent years has been to shed light on the ideological forces that have shaped the formation and implementation of these policies (e.g. Hornberger and Johnson 2007; Hult 2007; Rubdy 2008). Within the USA, many of the debates directly or indirectly concern Spanish and implicate politics and policies of immigration generally, and Mexican and Latin American immigration in particular. Ongoing points of contention, for instance, include whether Spanish is a legitimate language of communication in schools, offices, and public domains and under what circumstances Spanish speakers have the right to use their language in public settings, such as work and healthcare (e.g. Martinez 2008), as well as in private settings such as personal cars and homes. Often, the implicit or explicit targets of these policies are Latina immigrants. For instance, recent high-profile cases include Ernestina Mondragon’s court appeal in Dallas, Texas, for having been ticketed for being a ‘non English-speaking driver’ (Goldstein 2009), and multiple cases in which maternal competency and custody have been challenged on the grounds of limited English proficiency (e.g. Barry 2005; Padgett and Mascarenãs 2009). These and other language policy cases are profoundly shaped by longstanding mainstream language ideologies—widely in evidence in both popular and political discourse—that link migration status and language proficiency and equate learning English with being ‘good immigrants’ (e.g. Brandeis 1915; Griswold 2005; Woutat 2010). English is framed as both a prerequisite for personal success and as an easily accessible tool for learning (Haviland 2003). In this way, English use can be understood as a central component of banal nationalism (Billig 1995), as it is strategically employed to define who is (and who is not) a member of the national community. Since knowing English is connected with both ‘becoming and being American’, English proficiency tends to serve as an index of nationalism and patriotism (Warriner 2007: 345). In turn, lack of English proficiency and maintenance of a non-English language is often taken as an individual deficiency and a sign of lack of personal commitment to the USA or ‘American’ values (Pavlenko 2002). These language ideologies are interwoven with broader discourses on Latin American migration to the USA that are overwhelmingly negative. (Im)migrants conduct their lives within a political and social climate in which immigration is politicized and racialized, and in which the individuality, humanity, and positive contributions of migrant workers are diminished. As Santa Ana (1999), among others, has documented, US public discourse on immigrants is profoundly racist and de-humanizing. As a result, many US citizens regard immigrants as criminals. Recently, a 2008 New York Times/CBS News poll found that 69 per cent of American adults believe illegal immigrants should ‘be prosecuted and deported for being in the US illegally’ (Rasmussen Reports 2007). Data also suggests that 52 per cent of adults believe that most recent immigrants ‘cause problems’ and 62 per cent think that immigrants do not make enough of an effort to learn English (Rasmussen Reports 2007). Such stances do not go unnoticed by US Latinos, and indeed the


THE STUDY In this article we focus on a group that thus far has received little academic attention: transnational Latin American (im)migrant women who have contingent legal status, often no US legal documentation at all, and typically limited contact with English-speakers. As such, the women in our study are among the most marginalized of US workers (Hondagneu-Sotelo 2001), with little job security, few to no legal recourses, and minimal institutional or governmental support (Vellos 1997). We selected members of this group in part because, working in domestic spheres, out of public view, their voices are often excluded from both political and academic discussions. We chose to focus our study on women for several reasons. First, migration studies, and in particular those centered on Latin American (im)migrants, have stressed the important role women play in the key migration decisions (Miles 2003). Other work (e.g. Hondagneu-Sotelo 2001) emphasizes the critical functions of women with respect to transactions at stores, schools, and workplaces. Moreover, an established line of sociolinguistic research points to the significance of women’s participation in processes of linguistic shift and maintenance (e.g. Gal 1978; Holmes 1993; Winter and Pauwels 2005). Yet, despite the prominent role of women in these domains, there has been very little exploration of their language ideologies, their investments in English, and their stances towards language policy (an exception being Relanõ-Pastor and De Fina 2005). Finally, as suggested above, Latina women are often the real and imagined targets of language policies. For instance, all of the legal rulings with respect to parental English language competencies have involved mothers; and, Latina women are frequently the targets of English-only workplace sanctions (Baldas 2007).


majority now perceive themselves to be the targets of bias (MacGregorMendoza 1998; Pew Hispanic Center 2007). These ideologies and debates are a prominent component of US political life and are high-stakes for the individual (im)migrants who are being ‘debated’. Furthermore, recent research suggests that widespread expectations concerning language often contradict the experiences of many (im)migrants, who face great challenges in learning English and, even if successful in acquiring it, often fail to achieve their aspirations in the USA (Warriner 2007). Ideologies of language and discourses of immigration thus shape the personal experiences of individual learners as well as their investment in language programs (Skilton-Sylvester 2002). Nevertheless, little research directly examines how immigrants who are the targets of such language policies and implicit subjects of these language politics view and understand them. Given the tight connections that link language ideologies, language policies, and (im)migration, more attention is warranted to how individual (im)migrants position themselves as (im)migrants, speakers, and residents within this context and how they make sense of the multiple and contradictory discourses that surround them.


Participants

Interviews and analysis Our focus here was on talk surrounding language policy, and on gaining participants’ perspectives relative to the issues outlined above. Given the physical and social isolation of our participants, opportunities for ethnographically grounded, naturalistic interactions were nonexistent. As detailed above, the great majority of the women in our study spent nearly all of their waking hours alone or in the company of young children. Given these restrictions, the talk analyzed here was collected through informal interviews. As our aim was to investigate how individuals understand and talk about language policies, interviews were an appropriate approach to data collection. All interviews touched on a similar range of topics (Appendix 2 available as online supplementary material), focusing on participants’ personal history of migration and experiences with language learning, use, and policy.


Participants for the present study were 15 women from Latin America working in the Washington, DC metropolitan area. Like millions of individuals living in the USA, they are best described as ‘contingent citizens’ (Boehm 2008), as they, or members of their immediate family, lack permanent legal residence. Contingent citizenship ‘is national membership that is shifting, partial, conditional, or relational’, including ‘undocumented residents of the USA who are de facto members by virtue of their employment, education, residence, political participation, and civic engagement’ (Boehm 2008: 3–4). Holders of G-5, A-3, and B-1 visas are also contingent citizens because, for these individuals, legal residence in the USA entirely depends on maintenance of their current employment within one private family home.1 The women were all born outside the USA and their ages ranged from 28 to 55 years, with most in their 40s (see Appendix 1 available as online supplementary material). All had lived in the country for at least 4 years, with some moving back and forth several times between the USA and their home country. Reflecting the Latino diversity of the Washington, DC area, the women represented a wide geographic swath, hailing from Mexico (2), Bolivia (3), Colombia (2), El Salvador (5), Chile (1), Ecuador (1), and Peru (1). With the exception of one secretary and one receptionist, all worked in private domestic spheres, cleaning houses or minding children—jobs that are performed in isolated and dispersed conditions (Hondagneu-Sotelo 2001). Most of our participants had restricted opportunities for formal schooling in the USA, although many had intermittently attended weekly English-as-a-secondlanguage or Spanish-language literacy classes, both of which were typically church-based. All spoke Spanish as their first language and were Spanish dominant or Spanish monolinguals, but their particular varieties differed by place of origin. The majority of the women we interviewed had little to no need to use English in their work or social lives.



Participants were also encouraged to expand on topics, experiences, and sentiments of interest and importance to them. Women were recruited through personal and professional networks, including participants in prior studies. In some cases, the interview and interviewer had known each other personally over multiple years, while in others instances, they were recent acquaintances. Interviews ranged from 30 to 90 min, with most lasting about an hour. Researchers explained to the participants that they were interested in their life as immigrants and in how they had dealt with language issues. All interviews took place at a location of the participants’ choosing, typically either in the participant’s place of work or a nearby cafe´, and were conducted and transcribed in Spanish. Interviews have a long—and at points problematic and contentious—history within applied linguistics research. The interview has mainly been treated as a research instrument (Talmy 2010), that is, as a means for investigating particular content, truths, facts, beliefs, or attitudes. A well-established alternative approach, and the one taken here, treats the research interviews as a type of social practice (Cicourel 1964; Briggs 1986; Rapley 2001) wherein both the content of the talk and the interactional processes between the interviewer and interviewee constructing that content are investigated (Zilles and King 2005; Talmy 2010). More specifically, we approach interviews as sense-making events in which interviewers and interviewees participate with different levels of context being brought to bear. On the one hand, the interaction between the participants serves as a site for the positioning of self and other with respect to identity and relational issues. For instance, in our interviews we tended to align ourselves with participants and to try to build on common ground. As evident in the transcripts below, we mainly listened and often expressed agreement with the points made by interviewees; for their part, interviewees often seemed keen to establish a positive rapport with us given our interest in what they had to say. But involvement or distance between the participants and levels of formality varied according to the relationships between interviewer and interviewee, the topics that questions raised, and the emergent shape of the event. In that sense, the participation framework of each interview is unique and the nuances of individual interactions cannot be adequately captured in a discussion of general trends in discourse about policy. On the other hand, it is important to stress that the discourse between participants presupposes, evokes, and implicates legal and social processes and practices relevant to the material life of immigrants as well as discourses and ideologies about migration and language. Such discourses in turn impact the local construction of identities since interviewers and interviewees deal with circulating images of what it means to be a Latina, or a domestic worker, or alternatively a researcher, and adjust to each other in order to negotiate the self images that they want to project. Our premise is that close examination of how these contexts are entrenched in concrete social interactions potentially yields a deeper understanding of people’s reactions to public policies. Such an approach implies considering


FINDINGS AND DISCUSSION While past research has emphasized the ways in which language policy formation is ideologically driven (e.g. Ricento 2006), our data illustrate how language policy discussions often become terrains for engagement with language ideologies. As we demonstrate below, most women interpreted and understood language policy on the basis of their everyday observations and interactions, but also used it as a tool for self-presentation and self-positioning within the broader anti-immigrant context of public discourse in the USA.

Language policy and immigrant identities As ‘impossible subjects’ (Ngai 2004), that is long-term residents who are fully employed yet are simultaneously constructed through US legal and ideological machinery as less than legitimate members of US society, (im)migrant women are left with few options in presenting themselves relative to US language policy and politics. At the simplest level, their choices are to position themselves as individuals who try to belong, accepting the need and importance of English and highlighting their efforts to learn it. Alternatively, they can position themselves as defiant individuals who reject racism and their subservient


the role of both interviewer and interviewee in the analysis of sequences of talk and determining to what extent such roles are oriented to and made relevant by participants. Thus, our analysis draws from past work illustrating how interviews constitute interactional occasions in which participants share personal knowledge and experiences, position themselves relative to such content through their linguistic and discursive choices and project different identities. As argued in past work, it is indeed through the negotiation of personal experience and of social meanings that people build personhood and identities (De Fina 2003; Triandafyllidou and Wodak 2003; De Fina et al. 2006; Krzy_zanowski and Wodak 2008). Thus, interviews were analyzed along several dimensions. We identified and analyzed narrative fragments, that is, recounting of past experience focused on language difficulty or conflict (see De Fina and King in press). We also coded interview exchanges based on theme and topic according to whether they were descriptions and discussions of language difficulty, language policy, ethnic tensions, or immigration policy, collaboratively refining categories as we proceeded. We engaged in a general analysis of these recurrent topics and themes in order to identify similarities and differences in ways in which participants developed and negotiated them in the interviews. In addition, we conducted analyses of each individual interview, looking closely at identity construction and management. This approach, which allowed us to move back and forth between detailed analysis of each individual interview and comparative analysis of broader themes and trends, provides the background for the analysis that we present here focusing on question–answer sequences concerning language policy.


Excerpt 13 01 02

Jasmina

03 04

Int 2

05

Jasmina

06 01

Jasmina

02 03

Porque en sı´ para mucha- o sea gente que es independientemente de transportes y dema´s yo no les veo mucho la emocioń o las intenciones de querer aprender el espanõl. Uh huh. Pero bueno eso ya es cuestioń de cada quien no? da igual nosotros estamos en su paı´s y tambień es nuestro deber aprenderlo! aprender su idioma. Because really for many- I mean people who are independent of transportation and all the rest I do not see them with much excitement or intention to want to learn Spanish.

04

Int 2

Uh huh.

05

Jasmina

But well that is each person’s choice right? It’s the same we are in their

06

country and it is also our duty to learn it! learn their language.

Note how Jasmina places herself as part of a collectivity (nosotros, ‘we’ / nuestro, ‘our’) who needs to learn English because ‘we’ are in ‘their’ (su) country (lines 05–06). Attitudes towards English become a central means for her


subject position and question the value of investing in English. As women move back and forth between these positions across the interviews to differing degrees, there is evidence of their struggle—often via discussions of language policy—to present themselves as acceptable or to justify their perceived ‘wrongs’ (i.e. being in the USA without legal documentation, being Latina, or speaking limited English). These positionings are significant for them as they are talking to researchers who are outsiders to their world and who present themselves as sympathetic, but who might also be perceived as sharing widely circulating opinions about immigrants and language behavior. The local relationship with the interviewer as an outsider and a member of the dominant culture underlies the need for this kind of identity work on the part of the women. As illustrated below, in this sense, there is an awareness of the broader, dominant anti-immigrant and pro-English ideologies outlined at the start of the article. For instance, Jasmina’s dialog with her interviewer emphasized the former of these identity strategies as she tends to present herself as a ‘good immigrant’. In her direct answers to language policy questions she highlights her identity as a language learner and as one who has positive attitudes towards learning English (Excerpt 1). This point is expanded later in the interview (Excerpt 6); as we see below, in answering a question about whether there is popular support for a trial Spanish-language-education for bus drivers policy, she states that she does not see a great enthusiasm among the drivers for learning Spanish (lines 02–03).2 However, rather than critiquing the drivers’ lack of enthusiasm or the policy itself, she excuses their attitude as understandable and as being well within their rights. Instead, she suggests it is the Spanish speakers’ duty to learn English.


Excerpt 2 01

Int 1

Y has tenido algun conflicto aca´ con entre gente por no hablar ingle´s o entre gente que solo+/. . . alguien te ha dicho una vez tienes que hablar ingle´s o algo asi?

Elisa

En Virginia cuando fui a dejar un dine+/. . . el dinero en la uh hace dos meses atra´s estuve en el bus y habıá una senõra que hablaba no se´ que me dijo en ingle´s no?

02 03 04 06

Y yo no le contestaba porque, y hay una persona que habla+/. . . una chica, que habla espanõl e ingle´s me dijo te esta rin˜endo porque eres latina me dijo y

07 08

estas son todas esas personas que son son de ahı´ del lado de Virginia de Arlington no? De ese sector que estań discriminando a los latinos.

05

09

Int 1

Uhu.

10

Elisa

Porque yo estaba sentada adelante y ella estaba atra´s dice que yo le tapaba con mi

12

Int 1

Uh.

13

Elisa

Yo no le habıá entendido no? y entonces seguıá yo sentada pues que me importa que me estara´ diciendo dije yo.

Int 1 Elisa

Oh!@@@ Yo no le estaba dando motivo y entonces dice y la otra dice esta rin˜eńdote eso es. Y yo le dije ah que ni y+/. . . Que le estoy haciendo a esta senõra yo le dije?

11

cabeza su cara y que no querıá eso era.

14 15 16 17 18

Elisa

En ese bus dice va a todas rin˜endo.

19

Int 1

Uh.

20

Elisa

Son dos viejas que siempre estań yendo en el bus y cuando se topa con alguna

22

Int 1

latina que no le cae y andan rin˜endo no? Y que´ estaba diciendo?

23 24

Elisa

Diciendo que no seas+/. . . que que vaya a otro lado a sentarse que no se siente allı´ a su+/. . . a su delante->

25

Int 1

Oh! ((gasps))

26

Elisa

Sı´ asi.

27

Int 1

Wow!

01

Int 1

And have you had any conflict here with people for not speaking English or

21

02

between that only +/. . . Has anyone told you one time that you had to speak English or something like that?


to distinguish herself from others as she develops the argument across the interview that Latin Americans who live in the USA should learn English and contrasts herself with others who do not make this effort. Although not all were as explicit as Jasmina, many participants positioned themselves in similar ways. In sharp contrast to this tendency to highlight efforts at ‘fitting in’ or ‘doing their part’ are the oppositional stances of participants such as Elisa. While Elisa is also responding to the same set of questions concerning language policy and touching on similar discriminatory or hostile practices she experienced, she uses this story as a backdrop to position herself as actively resisting a subjugated role. Elisa, in Excerpt 2 below, presents herself as actively involved in events and as an evaluator of this implicit policy and racist practice.


03

Elisa

In Virginia when I went to leave some mo+/. . . the money at uh two months ago

04

I was in the bus and there was a lady who spoke I don’t know what she said to

05

me in English and I did not answer her because- and there is a person who

06 07

speaks+/. . . a girl, who speaks Spanish and English she said she is fighting with you because you are Latina she told me and these are all those people in who are

08

from there from that side of Virginia of Arlington, right? From that sector that is discriminating Latinos.

09

Int 1

Uhu.

10

Elisa

Because I was sitting in front and she was behind and she said that I was in her

12

Int 1

Uh.

13 14

Elisa:

And I didn’t understand right? And so I kept on sitting down since what do I care what she is telling me I thought.

15

Int 1

Oh! @@@

16

Elisa

I was giving her no reason and so she says and the other says she is fighting with

11

face with my head and she didn’t want it that was it.

and I told her uh that and+/. . . what am I doing to this lady I told her?

18

Elisa

19

Int 1

Uh.

20 21

Elisa:

It’s two old ladies who are always in the bus and when they encounter some Latina that they don’t like they start fighting, no?

22

Int 1

And what was she saying?

23

Elisa:

Saying that do not be+/. . . that go go somewhere else to sit do not sit there in- in

25

Int 1

Oh! ((gasps))

26

Elisa:

Yes like that.

27

Int 1

Wow!

24

In this bus she says she is fighting with everybody.

front of her ->

Elisa responds positively to the interviewer’s inquiry to whether she has been discriminated against for not speaking English. She presents herself here as resistant in two ways. First, by labeling the sector of Virginia where she was traveling as an area of discrimination (lines 07–08), she suggests that the (Virginian) women who mistreated her on the bus were in the wrong for being discriminatory and for ‘starting fighting’ rather than her being in the wrong for being Latina, being on the bus in that area, or speaking Spanish (lines 20–21). Second, by noting that she did not care at all about what the woman was saying (lines 13–14), she suggests that they (or their actions) were irrelevant. The overall effect conveyed to the interviewer an image of herself not so much as a victim of racist abuse or policy, but as clever and resourceful in resisting this abuse and putting the harassing women in their proper place in this telling. For Elisa, this defiance against the pressure to learn English is linked to broader rejection of subservient status in the USA, a rejection that she explicitly discusses at many points in her interview. For instance, she also told in great detail of a negative encounter with her (former) employer in


you that’s it. 17


Three frames for language policy As the above excerpts highlight, interview questions about language policy at times evoked highly personal accounts of specific language experiences; these accounts, and the participants’ roles within them, were shaped by salient circulating ideologies of language and immigration. At the same time, women’s interpretations of ‘language policy’ (in Spanish, polı´tica lingu¨ı´stica) varied widely and with very few offering policy specifics. Given our participants’ social isolation, it is not surprising that they seemed to have limited knowledge of language policy laws or initiatives (e.g. official English legislation, bilingual education policies, legal rights to medical translation, local initiatives to support access to public services). But this lack of information is also an important indication of how those most impacted by specific policies remain largely unaware of policy particulars but highly cognizant of broader language politics. As suggested above, participants readily described concrete experiences with policies in their daily lives. Questions about polı´tica lingu¨ı´stica evoked a range of responses in part because polı´tica lingu¨ı´stica can be understood as both ‘language policy’ and ‘language politics’. These questions also became occasions for personal sense-making and negotiation, suggesting women’s isolation from public policy discussions and their exposure to conflicting messages in public and private domains. Concomitantly, discussions about Spanish became a terrain for reflections focused on language ideologies.


which she reminded her opponent that ‘here in this country. . .slavery does not exist. . .’ (aquı´ en este paı´s . . . no hay esclavitud). Thus, resistance to general pressure to learn English, at least for Elisa, is linked to resistance to subservient and racialized position in the USA. In making these points, Elisa could be said to be engaging in opposition to ‘everyday racism’ (Essed 1991: 261). Looking more closely at the interactional positioning in the interviews with Jasmina and Elisa, in both excerpts above, although the interviewers’ verbal responses were minimal, they were also supportive and sympathetic. However, it is also significant that there are differences between the two interviews. While the participants in the first interview had very recently met, the participants in the second had had many previous interactions and were very familiar with each other. Elisa and the interviewer clearly shared the idea that language discrimination against immigrants exists in the USA. This past relationship and common views might explain Elisa’s confidence in expressing a more confrontational stance towards mainstream discourses about language use while Jasmina is more cautious. These two interview excerpts thus point to wide potential differences in participants’ positions, experiences, and language ideologies, but also to how these shared positions are co-constructed and variable depending on the relationship between the interview participants.


Language policy as racism All participants were quick to note the importance of English and problems inherent in being a monolingual speaker of Spanish in the USA. They repeatedly described English as essential para mejorarse, defenderse, and superarse (‘to better oneself’, ‘to defend oneself’, and ‘to improve oneself’). Many began by stressing the need for English for basic communication, for ‘survival’, as well as for fitting in and belonging, avoiding humiliation, and gaining social status. While few women reported needing English for their current work or social lives, all women at some point indicated that the learning of English symbolized realized dreams of migration, job and educational opportunities. Yet, for many women, as for Lorena (Excerpt 3), language problems—and language policies undergirding them—were not directly the result of communication difficulties, but of racial discrimination. Language policy was interpreted in many cases as the implementation of discriminatory and racially motivated measures against Spanish speakers.

Excerpt 3 01 Int 2

Bueno usted cree que la gente+/. . . que hay una polı´tica del gobierno,

o de la ciudad del gobierno con con respecto al idioma al uso del espanõl y el ingle´s? 03 Lorena Yo pienso que esa polı´tica siempre ha habido. siempre (.) para+/. . . por el idioma por 04 muchas cosas. por raza discriminacioń ma´s que todo de que (.) 02

05

cuando la gente no habla ingle´s, en los trabajos y en cualquier lado uno tiene

06

problemas siempre en el idioma. por eso el ingles es bien importante.

01 Int 2

Well do you believe that people+/. . . that there is a policy by the government,

02

or by the city with respect to language, to the use of Spanish and English?

03 Lorena I think that this policy has always existed. always (.) for+/. . . for the language for 04

many things. because of race discrimination more than anything else that (.)

05

when people do not speak English, at work and in any other place one has

06

problems, always with the language. that is why English is very important.


Most understandings of and orientations towards polı´tica lingu¨ı´stica were constructed by women within one of three frames, reflecting these conflicting experiences and expectations. These included: (i) language policy as racism and discrimination against Spanish speakers; (ii) language policy as deception used to hide the government’s objective to discourage immigration; and (iii) language policy as support services for Spanish speakers. These different constructions often coexisted in the same interview. Their simultaneous presence reveals the extent to which Latino (im)migrants are confronted with conflicting public and private discourses about the value of Spanish and English as linguistic capital and the real (and imagined) role of these languages in their lives. However, the devaluation of Spanish, due to its perceived link to illegal migration noted above, is reflected in the prevalence of defensive stances observed in discourse sequences within all three frames.


Lorena cites racial discrimination (line 04) as linked with language policy, suggesting that Spanish speakers at work experience problems not because of communication difficulties but due to racial prejudice. Lorena then proposes learning English as one way of mitigating or avoiding racial problems. Other women, such as Manuela below (Excerpt 4), turn to personal experience to elaborate on the links between language policy and discrimination.

Excerpt 4 01

Int 1

02 Manuela

05 07

espanõl pero en otros, en otros no! yo tengo una cunãda que ella habla espanõl, es americana, y yo fui a una clıńica con ella y ella me sirvio´ de inte´rprete y le dijeron allı´ la secretaria por que tu

08

hablas, ese idioma si no es tu idioma, como le dijo no tu no tienes

06

09

que hablar ese idioma, entonces ella- y ella le dijo que por QUE?

10

Int 1

Si claro.

11 12

Manuela

Porque le dijo que no que no era su idioma y que ella no tenıá que aprenderlo yo creo pues uh+/. . .

13

Int 1

Wow.

14

Manuela

Cosas de racismo tal vez de ver a la persona.

01

Int 1

And, and do you think that there is a policy by the government

02

with respect to the use of Spanish and English in political places

03

I mean what do, politicians think about Spanish? What are the attitudes of the government?

04

Manuela

Uh (.) well in some places they say that they like people to speak

05 06

Spanish well but in others, in others no! I have a sister-in-law who speaks Spanish, she is American, and I went to a doctor’s office with her and

07

she acted as an interpreter and they told her there the secretary why do you

08

speak, that language if it is not your language, like she said you should not

09

speak that language so she- and she told her WHY?

10

Int 1

Yes of course.

11

Manuela

Because she told her that it wasn’t her language and that she should not learn it I think uh+/. . .

12 13 14

Int 1 Manuela

Wow. Racist things maybe just by looking at the person.

Here, Manuela interprets language policy as acceptance or rejection of one’s right to speak Spanish in public places and declares that attitudes vary (lines 04–05). She then recounts a personal experience in which language policy is translated into concrete discriminatory action. Manuela describes a hospital visit during which the secretary challenged her sister-in-law’s right to speak Spanish on the grounds that this was not the sister-in-law’s language. For


03 04

Y, y usted cree que hay una polı´tica del gobierno en cuanto al uso del espanõl y el ingle´s en lugares polı´ticos o sea que que´ piensan los, polı´ticos del espanõl? Que´ son las actitudes del gobierno? Uh (.) pues en algunos lugares dicen que le gusta que hablen bien


Manuela, language policy, as expressed in this example of the statement by the clinic secretary, is a stand-in for racism, as she sums up in line 14. Her observation is supported implicitly by the interviewer’s comments (‘yes, of course’ and ‘wow’, lines 10 and 13), both of which suggest outrage at the secretary’s reported translation policy. In naming and describing these practices and policies, both Lorena and Manuela bring to light perceived injustices towards Latinos in the USA. Here and in many instances across the interviews, our questions about language policy evoked problematic experiences with English-speaking non-Latinos. While language often played a prominent role in such accounts, these experiences were widely interpreted as involving their being targeted due to racial, rather than linguistic, differences.

Language policy was also framed in the interviews as a tool of deception employed by the US government to obfuscate the presumed real objective of discouraging immigration and often linked to (long-stalled) immigration reform efforts. As many of the women had no legal authorization to stay in the USA, they tended to pay attention to immigration reform as it was covered on Spanish-language television and radio. As the last decade has seen much public discussion of immigration policy, but very few changes (Bacon 2008), skepticism and cynicism were pronounced among some of our participants, particularly those who needed greater cross-border mobility. As evident in the interview with Elisa below (Excerpt 5), for some, language policy was interwoven with long-held-out immigration reform policy.

Excerpt 5 01

Int 1

02 03

Elisa

04

Int 1

05

Elisa

06 07 08

Int 1

09

Elisa

10

Y tu crees que hay una, ah una polı´tica del gobierno en cuanto al uso, del espanõl y del ingle´s? Sı´. Que´ sera´? Porque la polı´tica que estań haciendo estań poniendo el pretexto de la, del ingle´s no ahora? porque hay muchos que hablan espanõl ingle´s y que se puede uno este+/. . . entender. Uh huh. Bueno en todo caso tendrıámos que estudiar no si es que fue´ramos legales, por ejemplo yo tendrıá que tomar clases o sea esforzarme un poquito para tener un poco de ingle´s,

11 12

Int 1

Uhu.

13

Elisa

Pero si fuera legal no? Por que si no soy legal y me tengo que ir a alguń lado para que necesito gastar el dinero en el ingle´s? @@@

15 16

Int 1 Elisa

@@@ Nunca me @@sale!

17

Int 1

18

Elisa

No te va a ayudar!@@ No me va a ayudar en nada me voy despue´s para que me sirve el ingle´s?

14


English language policy as deception


01

Int 1

And do you think that there is a, uh, a policy of the government with

02 03

Elisa

respect to the use of Spanish and English? Yes.

04

Int 1

What would that be?

05

Elisa

Because the policy that they are doing they are putting the pretext of the,

06

of English, now you see? Because there are many who speak Spanish English and people can uh+/. . . understand each other.

07 08

Int 1

Uh huh.

09

Elisa

Well in any case we should study isn’t it if we were legal,

10 11

for example I would have to take classes that is make some effort to learn a little bit of English, Int 1

Uhu.

13

Elisa

But if I were legal, right? Because if I am not legal, and I have to go somewhere

14

why do I need to spend money on English?@@@

15

Int 1

@@@.

16

Elisa

I can @@never speak it!

17

Int 1

It’s not going to help you!@@

18

Elisa

It’s not going to help me in anything, I leave and then what good does English do me?

For Elisa, language policy, viewed as the imposition of English learning, is framed as a pretext for delaying access to citizenship or legal residence. In light of this, Elisa positions herself as rejecting the vision of English as having general market value. She regards language learning as necessary or beneficial only for those who are granted legal status (lines 09–11). She underscores this point by making sarcastic remarks, shared with the interviewer and underscored with laughter (lines 12–17), on the fact that English is not going to help her if she is not in the country legally (lines 13–14 and 18). Like other interviewees, Elisa rejects English language and maintains that she has little reason to invest in learning it. Furthermore, she experiences the pressure to learn English as punitive. She does not subscribe to mainstream discourses concerning the value of English, and instead sees it as a costly, labor-intensive investment with little reward given her legal status. Her (and others’) interview talk suggests that restrictive language policies do not offer the incentives imagined by policy-makers (Crawford 2006). While this overt and repeated rejection of English learning can be linked to a broader oppositional stance (detailed above), it does not mean total rejection of English in practice. For instance, despite Elisa’s strong stance, she laments that English is not taught more effectively in Bolivia and complains that she must pay 70 dollars a month for her college-age daughter to take English classes at a private academy.

Language policy as support services for Spanish speakers As mentioned above, most participants had limited knowledge of particular language legislation and knew none by name; in a few cases, however, they


12


did indicate awareness of specific legislation in practice. For instance, Jasmina (Excerpt 6) notes that the city government tries to help Latinos by offering Spanish language courses to bus drivers.

Excerpt 6 01

Int 2

02 03 04

Jasmina

05

Y cree que hay una polı´tica del gobierno aquı´ con respecto al uso del espanõl y del ingle´s, uh (.) o no? en los lugares pu´blicos por ejemplo. Bueno lo que yo tengo entendido que el gobierno esta´ tomando

07 09

el espanõl, entonces eso es algo bueno y hay que admirarles eh esa iniciativa aunque en mi opinioń digo bueno para mi no es mucha ayuda porque me subo al autobu´s para practicar el ingle´s y

10 11

me hablan en espanõl! entonces no es ayuda para mi para ellos si pero para mi no.

08

01

Int 2

02

with respect to the use of Spanish and of English, uh (.) or not?

03 04

And do you think that there is a policy by the government here In public places for example.

Jasmina

Well what I understand is that the government is taking

05

measures for the, for example to help Latinos that eh (.)

06 07

for example drivers of public buses are taking courses to speak Spanish, then this is something good and one has to admire them

08

for this initiative although in my opinion well I mean for me it is

09

not great help because I get on the bus to practice English and

10

they talk to me in Spanish! so it does not help me it helps

11

them but not me.

Jasmina is among the few to give an interpretation of language policy as positive governmental intervention with respect to language minorities. She describes this policy tentatively as ‘measures taken by the government to help Latinos’ and orients moderately favorably to it (‘something good and one has to admire them’, line 07). However, she also judges the policy negatively as she frames these attempts to provide services to Spanish speakers as something not desirable to her personally since they limit her opportunities to practice her English (lines 08–10). It is notable that on the one hand, positive interpretations of language policy are rare in the interviews, while on the other, local government provisions that attempt to support Latinos seem to have gone mostly unnoticed, at least within this group.

CONCLUSION The bulk of language policy work begins with the analysis of the texts, ideological assumptions, or political formations of specific language policies and


medidas para los, por ejemplo ayudar a los latinos de que eh (.) por ejemplo los choferes de los autobuses estań tomando cursos ya, para hablar

06



then possibly focuses on analysis of their effectiveness in meeting established goals. In contrast, here we have begun with the explicit or implicit targets of broader US language policies and politics—Latina (im)migrants. We then analyzed these women’s reported experiences with and understandings of language policy and politics in the USA. Our analysis of this interview talk suggests three points. First, our data illustrate how language policy is interpreted through women’s own experiences rather than being influenced by public interventions or public media. These experiences involve, among other things, making sense of contradictory discourses concerning language, politics, and immigration and as a result, the women themselves express conflicting attitudes. For instance, they share the prevailing monoglot norm dictating the need to learn English in order to fit in and advance in the USA, but they also emphasize how they and their peers experience limited opportunities for upward social mobility or change in legal status, regardless of their investment in English. They also often experience attempts to push them to learn English as punitive and deceptive. In short, the data here thus illustrate how concrete life experiences shape individuals’ understanding of language policy and often conflict with public discourses and intended policy outcomes. Second, the discourse analysis of interview talk presented here suggests that greater attention is needed with respect to what language policies mean to the people most impacted by them. Such understanding can best be provided through the use of qualitative instruments such as case studies and interviews and through close analysis of a wide range of interactional talk. Indeed, we saw in our data that women use discussions of language policy as a platform for constructing alternative (positive) public identities within a generally hostile, anti-immigrant public climate. For instance, Jasmina frames her account of a local transportation policy in light of her own efforts to learn English, while Elisa uses a narrative about language conflict to position herself as someone who opposes discrimination. These examples demonstrate how discourse data and analysis can generate richer understanding of the complex and difficult to predict ways in which individuals relate to public decisions and public discourses. Third, these data demonstrate how—at least within this population and context—understandings of language policy are intertwined with, or even subsumed by, immigration policy and politics. Language policy as a field has evolved rapidly over the past five decades (Spolsky 2004; Shohamy 2006), and scholars working in this area have devoted significant attention to the development of frameworks and rubrics for the analysis of language policy sub-areas, types, and goals. While this work has done much to define and advance the field on both practical and theoretical grounds, it has also segregated language policy from other areas of policy-making and analysis, such as immigration reform, economics, and public policy. As language policy became defined as a particular field of study, with its attendant assumptions, methodologies, frameworks, and internal theoretical debates, it also differentiated,


SUPPLEMENTARY DATA Supplementary material is available at Applied Linguistics online.

NOTES 1 These visa holders are, respectively, domestic workers for employees of international agencies such as the World Bank; for employees of embassies; and for foreign nationals and US citizens with permanent residence abroad who are allowed to import nannies, housekeepers, cooks, and other domestic workers from abroad. As many such visa-holders reside in the DC area, our sample includes a significant number of women with this status.

2 The DC Metro implemented in 2007 a pilot project to teach Spanish to bus drivers, station managers, and street supervisors in compliance with the Federal guidelines for language access (http://www.washingtonpost.com/ wp-dyn/content/article/2007/10/24/ AR2007102402523.html). 3 See Appendix 3 (available as online supplementary material) for transcription conventions.

REFERENCES Bacon, D. 2008. Illegal People: How Globalization Creates Migration and Criminalizes Immigrants. Beacon Press. Baldas, T. 2007. June 18. ‘English-only workplace policies trigger law-suits,’ The National Law Journal. Barry, E. 2005. ‘Learn English, Judge tells moms,’ Los Angeles Times. Available at http://

articles.latimes.com/2005/feb/14/nation/ na-english14. Accessed 5 August 2010. Billig, M. 1995. Banal Nationalism. Sage. Boehm, D. A. 2008. June. Here/not Here: Contingent Citizenship and Transnational Mexican Children. Paper presented at the Childhood and Migration Interdisciplinary Conference. Philadelphia, PA.


and to some extent, isolated itself from related social sciences. The present analysis reminds us of the importance of taking into account (non-linguistic) policy developments, not only in terms of thinking about language policy creation, but also—and more to the point here—in the interpretation of language policy by its intended targets. Each of these points underlines how all questions concerning language policy are inextricably linked with local language ideologies and discourses, and at the same time, with inequalities of power. Indeed the data and analysis proposed here illustrate the many different ways in which language policy is interpreted by its targets, and point to the fact that what might impact people’s perceptions and behaviors the most is not official policy itself, but rather the individual experiences and broader discourses that provide frames to make sense of this policy. Thus, the design and implementation of policies focused on language use alone—for example, the local initiatives to promote public access to services such as transportation discussed here—are unlikely to meet their intended aims without efforts to alter broader power relations and surrounding discourses.


Garcıá, O. 2008. ‘Teaching Spanish and Spanish in teaching in the USA: integrating bilingual perspectives’ in C. He´lot and A. de Mejıá (eds): Forging Multilingual Spaces: Integrated Perspectives on Majority and Minority Bilingual Education. Multilingual Matters.

Goldstein, S. 2009, October 26. ‘Dallas police ticketed 39 drivers in 3 years for not speaking English,’ Dallas News Available at http://www .dallasnews.com/sharedcontent/dws/news/ localnews/stories/DN-citations_23met.ART.State. Edition2.4bac015.html. Accessed 8 January 2010. Griswold, D. 2005. ‘Mexican migration, legalization, and assimilation’ Cato Institute. http:// www.cato.org/pub_display.php?pub_id=10922. Accessed 5 August 2010. Haviland, J. B. 2003. ‘Ideologies of language: some reflections on language and U.S. law,’ American Anthropologist 105/4: 764–74. Holmes, J. 1993. ‘Women’s role in language maintenance and shift,’ Working Papers in Language Gender and Sexism 3: 5–24. Hondagneu-Sotelo, P. 2001. Dome´stica. Immigrant Workers Cleaning and Caring in the Shadow of Affluence. University of California Press. Hornberger, N. H. and D. C. Johnson. 2007. ‘Slicing the onion ethnographically: layers and spaces in educational language policy,’ TESOL Quarterly 41/3: 509–32. Hult, F. 2004. ‘Planning for multilingualism and minority language rights in Sweden,’ Language Policy 3/2: 181–201. Hult, F. 2007. ‘Multilingual language policy and English language teaching in Sweden’. Unpublished doctoral dissertation, University of Pennsylvania. Johnson, D. C. 2009. ‘Ethnography of language policy,’ Language Policy 8: 139–59. Krzy_zanowski, M. and R. Wodak. 2008. ‘Multiple identities, migration and belonging: voices of migrants’ in R. Caldas-Coulthard and R. Iedema (eds): Identity Trouble: Critical Discourse and Contested Identities. Palgrave. MacGregor-Mendoza, P. 1998. ‘The criminalization of Spanish in the United States’ in D. Kibbee (ed.): Language Legislation and Linguistic Rights. John Benjamins. Martinez, G. 2008. ‘Language-in-healthcare policy, interaction patterns, and unequal care on the U.S.-Mexico border,’ Language Policy 7/4: 345–63. Miles, A. 2003. From Cuenca to Queens: An Anthropological Study of International Migration. University of Texas Press. Ngai, M. M. 2004. Impossible Subjects: Illegal Aliens and the Making of Modern America. Princeton University Press.


Brandeis, L. 1915. To ‘possess the national consciousness of an American’. Center for Immigration Studies. Available at http://www.cis .org/articles/cantigny/fonte.html. Accessed 5 August 2010. Briggs, C. L. 1986. Learning How to Ask: A Sociolinquistic Appraisal of the Role of the Interview in Social Science Research, Studies in the Social and Cultural Foundations of Language. Cambridge University Press. Canagarajah A. S. (ed.). 2005. Reclaiming the Local in Language Policy and Practice. Lawrence Erlbaum Publishers. Cicourel, A. 1964. Method and Measurement in Sociology. Coller Macmillan. Cooper, R. L. 1989. Language Planning and Social Change. Cambridge University Press. Crawford, J. 2006. Official English Legislation: Bad for Civil Rights, Bad for America’s Interests, and Even Bad for English. (Testimony before the House Subcommittee on Education Reform.) Crawford J. (ed.). 2008. Advocating for English Language Learners: Selected Essays. Multilingual Matters Press. Davies, A. 1991. The Native Speaker in Applied Linguistics. Edinburgh University Press. De Fina, A. 2003. Identity in Narrative. A Study of Immigrant Discourse. John Benjamins. De Fina, A. and K.A. King. In press. ‘Language problem or language conflict? Narratives of immigrant women’s experiences in the U.S,’ Discourse Studies 13/1. De Fina, A., D. Schiffrin, and M. Bamberg (eds). 2006. Discourse and Identity. Cambridge University Press. Essed, P. 1991. Understanding Everyday Racism. Sage. Fishman, J. A., C. A. Ferguson, and J. Das Gupta. (eds). 1968. Language Problems of Developing Nations. John Wiley & Sons. Gal, S. 1978. ‘Peasant men can’t get wives: language change and sex roles in a bilingual community,’ Language in Society 7: 1–16. Gańdara, P. and R.W. Rumberger. 2009. ‘Immigration, language, and education: How does language policy structure opportunity?’ Teachers College Record 111/3: 750–82.


Shohamy, E. 2006. Language Policy: Hidden Agendas and New Approaches. Routledge. Skilton-Sylvester, E. 2002. ‘Should I stay or should I go? Investigating Cambodian women’s participation and investment in adult ESL programs,’ Adult Education Quarterly 53/1: 9–26. Spolsky, B. 2004. Language Policy. Cambridge University Press. Spolsky, B. 2009. Language Management. Cambridge University Press. Talmy, S. 2010. ‘Qualitative interviews in applied linguistics: from research instrument to social practice,’ Annual Review of Applied Linguistics 30: 128–48. Tollefson, J. W. 2002. Language Policies in Education: Critical Issues. Lawrence Erlbaum Associates. Triandafyllidou, A. and R. Wodak. 2003. ‘Conceptual and methodological questions in the study of collective identities: an introduction,’ Journal of Language and Politics 2/2: 202–25. Vellos, D. 1997. ‘Immigrant Latina domestic workers and sexual harassment,’ American University Journal of Gender and the Law 5: 407–32. Warriner, D. S. 2007. ‘Language learning and the politics of belonging: Sudanese women refugees becoming and being ‘‘American’’,’ Anthropology and Education Quarterly 38/4: 343–61. Winter, J. and A. Pauwels. 2005. ‘Gender in the construction and transmission of ethnolinguistic identities and language maintenance,’ Australian Journal of Linguistics 25/1: 153–68. Woolard, K. A. and B. Schieffelin. 1994. ‘Language ideology,’ Annual Review of Anthropology 23: 55–82. Woutat, R. 2010, 21 May. ‘Assimilation and acceptance of immigrants,’ Kitsap Sun. Available at http://www.kitsapsun.com/ news/2010/may/21/rob-woutat-assimilationand-acceptance-of/. Zilles, A. and K. A. King. 2005. ‘Self-presentation in sociolinguistic interviews: Identities and language variation in Panambi, Brazil,’ Journal of Sociolinguistics 9/1: 74–94.


Padgett, T. and D. Mascarenãs. 2009, 17 August. ‘Can a mother lose her child because she doesn’t speak English,’ Time Com. Available at http://www.time.com/time/nation/article/ 0,8599,1918941,00.html. Accessed 5 August 2010. Pavlenko, A. 2002. ‘ ‘‘We have room for but one language here’’: language and national identity in the US at the turn of the 20th century,’ Multilingua 21/2–3: 163–96. Pew Hispanic Center. 2007, November 29. English Usage among Hispanics in the United States. Available at http://pewhispanic.org/ files/reports/82.pdf. Accessed 5 August 2010. Rapley, T. J. 2001. ‘The art(fulness) of openended interviewing: some considerations on analysing interviews,’ Qualitative Research 1/3: 303–23. Rasmussen Reports. 2007, May 25. NY Times/ CBS Poll Finds That 69% Believe Illegal Immigrants Should Be Prosecuted. Available at http://www.rasmussenreports.com/public_ content/politics/current_events/immigration/ ny_times_cbs_poll_finds_that_69_believe_ illegal_immigrants_should_be_prosecuted. Accessed 5 August 2010. Relanõ-Pastor, M. and A. De Fina. 2005. ‘Contesting Social place: Narratives of Language Conflict’ in M. Baynham and A. De Fina (eds): Dislocations, Relocations, Narratives of Displacement. St Jerome Publishing. Ricento, T. (ed.). 2000. Ideology, Politics, and Language Policies: Focus on English. John Benjamins. Ricento, T. (ed.). 2006. An Introduction to Language Policy: Theory and Method. Blackwell. Rubin, J. and B. Jernudd (eds). 1971. Can Language Be Planned? Sociolinguistic Theory and Practice for Developing Nations. East-West Center and University of Hawaii Press. Rubdy, R. 2008. ‘Language planning ideologies, communicative practices and their consequences’ in M. Martin-Jones, A. M. de Mejıá, and N. Hornberger (eds): Encyclopedia of Language and Education, 2nd edn. Springer. Santa Ana, O. 1999. ‘ ‘‘Like an animal I was treated’’: anti-immigrant metaphor in US public discourse,’ Discourse & Society 10: 191–224.

Applied Linguistics: 31/5: 671–692 ß Oxford University Press 2010 doi:10.1093/applin/amq026 Advance Access published on 16 September 2010

NNES Doctoral Students in English-Speaking Academe: The Nexus between Language and Discipline 1,

*YU-JUNG CHANG and 2YASUKO KANNO

1

National Tsing Hua University and 2Temple University *E-mail: [email protected]

Since the mid-20th century, US graduate schools have seen a growing participation of international students. In the 2008/2009 academic year, for instance, the USA accommodated over 283,000 international graduate students, including more than 108,000 doctoral students (Institute of International Education 2009). Over 30 per cent of new doctorates were awarded to international students in 2007 (Welch 2008). In light of the ever growing representation of foreign students in US graduate schools, a substantial body of research has been devoted to the issues and challenges faced by international students whose first language is not English. Since English is the means to access, produce, and disseminate knowledge in US research communities, linguistic challenges have been prominently featured in studies of non-native-English-speaking (NNES) graduate students’ academic and social integration. Problems reported by NNES students include the large amounts of required reading and writing, their lack of confidence, and fluency in using English, and their insufficient familiarity with English


Since the mid-20th century, graduate schools in the USA have witnessed a growing participation of international students, many of whom do not speak English as their first language. Previous research has often portrayed non-native-English-speaking (NNES) students in US doctoral programs as disadvantaged because of the presumed primacy of English in their academic pursuits. This study examines NNES doctoral students’ participation in US academia, and in doing so, challenges this assumption. Drawing on the concepts of communities of practice (Lave and Wenger 1991; Lave 1996; Wenger 1998) and capital (Bourdieu 1977, 1991), the study analyzes how linguistic competence plays out in NNES students’ participation in three different disciplines. We argue that language competence as cultural capital does not have the same value across different disciplines and may not always be critical to NNES students’academic success. Furthermore, despite their keen awareness of their differences from their native-speaking counterparts, the NNES doctoral students in this study had other forms of cultural capital with which they claimed legitimacy and recognition in their disciplinary communities.

672 ASSESSED LEVELS OF SECOND LANGUAGE SPEAKING PROFICIENCY


academic vocabulary, discourse, colloquialism, and body language (Shaw 1991; Morita 2000; Lacina 2002; Parks and Raymond 2004; Tardy 2004; Kim 2006; Poyrazli and Kavanaugh 2006; Kuwahara 2008). Other scholars have also highlighted the importance of sociolinguistic competence in NNES international students’ academic success. They argue that compared with native-English-speaking (NES) American graduate students, NNES international graduate students are likely to have less access to rhetorical, pragmatic, and other sociolinguistic means to express themselves appropriately and establish positive relationships with peers and professors (Schneider and Fujishima 1995; Ballard 1996; Dong 1998; Angelova and Riazantseva 1999; Morita 2004). Although previous literature has generated significant insights into the critical challenges faced by NNES international graduate students (henceforth NNES students for the sake of brevity), in our opinion, it has also presented an incomplete and biased picture of their experiences in English-speaking academic communities. First, NNES students are often assumed to encounter disadvantages because they ‘often lack the cultural and linguistic capital of their native speaker peers’ (Kuwahara 2008: 187; also see Zimmerman 1995; Cho 2004). Even Casanave (2008; Casanave and Li 2008), who points out that learning to ‘do’ graduate school does not come naturally to most people, comments that if NES American students have to go through a double socialization (Golde 1998) in graduate school, what lies ahead for NNES students is the process of triple socialization: namely, socialization into (i) the role of graduate students, (ii) a profession, and (iii) ‘a language and culture that their mainstream peers have been immersed in for a life time‘ (Casanave and Li 2008: 3). With the underlying assumptions of NNES students’ linguistic and cultural disadvantages, the possibility that these students may have other kinds of competences and resources that can facilitate their participation in English-speaking disciplinary communities has rarely been pursued. Second, studies examining NNES students’ experiences are often set within one discipline (Casanave 2002; Parks and Raymond 2004; Li 2006) or a group of related disciplines (Angelova and Riazantseva 1999; Morita 2004; Tardy 2005). The different roles that competence in English may play in NNES students’ academic experiences across various disciplines have been obscured by this intra-disciplinary focus. Researchers have long acknowledged crossdisciplinary variations in discursive requirements (Casanave and Hubbard 1992; Parry 1998; Hyland and Tse 2007). For instance, Casanave and Hubbard’s (1992) survey study found that writing skills are critical to students in the humanities and social sciences throughout their doctoral education, whereas in science and technology, writing practices become prominent only when students approach the end of their education. However, such insights have not yet informed inquiry into NNES students’ participation in various disciplinary communities. Little research has been conducted to examine whether some disciplines are more heavily reliant on linguistic expressions to communicate and disseminate ideas than others—what we call

Y.-J. CHANG AND Y. KANNO

673

THEORETICAL FRAMEWORK The theoretical framework of this study draws on Lave and Wenger’s (1991; Wenger 1998) conceptualization of communities of practice and Bourdieu’s (1977, 1986, 1991; Bourdieu and Passeron 1990) notions of capital. In line with the sociocultural traditions that focus on the social and situated nature of learning, Lave and Wenger (1991; Lave 1996; Wenger 1998) propose that learning is a process of growing participation in a community of practice. Newcomers start at the periphery with limited participation in the community practices. Such limited participation is legitimate in the sense that it gives newcomers direct access to community practices, thereby allowing them to start learning. By interacting with other community members and apprenticing with more experienced old-timers, newcomers gradually develop their competence. What is important in such an apprenticeship model of learning is that explicit teaching may or may not happen (Lave 1996). Rather, a more crucial condition for learning than explicit teaching is the learner’s engagement in shared practices (Wenger 1998). The experiences of doctoral students in US graduate schools closely mirror this apprenticeship model of learning. Although they take courses and receive explicit guidance on their dissertations from their advisors, much of what students must learn, if they are to become full-fledged members of academia, is learned through their engagement with various members of the community. If their engagement increases over time, doctoral students become less of a novice and more of a full member, for learning is ‘itself an evolving form of membership’ (Lave and Wenger 1991: 53).


language dependence—and how different degrees of language dependence in various disciplines affect NNES students’ challenges in academia. The conclusions that have been drawn without attention to possible disciplinary variations risk overgeneralizing the primacy of English in NNES students’ success in academia. This qualitative study attempts to address these gaps in knowledge by examining the role of English in NNES students’ learning and participation across different disciplines at a major US university. Drawing on the concept of communities of practice (Lave and Wenger 1991; Lave 1996; Wenger 1998) and Bourdieu’s forms of capital (1977 and 1991), the study aims to analyze how language competence influences NNES students’ participation and selfperception in different disciplinary communities of practice. Our analysis of the firsthand accounts of NNES students and their doctoral advisors suggests that language competence as a form of cultural capital may not have the same value across different disciplines and may not always be critical to NNES students’ academic success. We further argue that the NNES students’ sense of legitimacy and confidence in the English-speaking research community is not always threatened because they may possess other strengths that can be utilized to claim legitimate membership within that community.


1 To what extent do NNES students regard competence in English as critical to their academic participation and performance in their disciplinary communities?


Thus, Lave and Wenger’s conceptualization of learning is closely tied to the notion of identity development. However, not every newcomer becomes a legitimate and full member by virtue of sustained engagement in shared practices; some, in fact, become marginalized over time (Wenger 1998). For a newcomer to attain the community’s recognition and become a legitimate member, therefore, she needs to develop and display the kinds of competence that are valued in the community. As Wenger (1998: 152) notes, ‘Identity . . . is an experience and a display of competence’. Our inquiry in this article is fundamentally about what forms of competence doctoral students need to develop and display in different disciplinary communities. To that end, we find Bourdieu’s notions of capital useful. Capital refers to material and immaterial resources that confer power to people. Of the various forms of capital that Bourdieu (1977, 1986, 1991; Bourdieu and Passeron 1990) has introduced, two that are particularly important in this inquiry are cultural capital and linguistic capital. Bourdieu’s conceptualization of cultural capital has shifted from one work to another over time. In some studies, Bourdieu has focused more on class-based dispositions and attitudes, while in others he has referred more to formal knowledge, previous academic experiences, and educational qualifications (Lamont and Lareau 1988). In its broadest sense, however, cultural capital refers to cultural resources that have high market value in a particular ‘field,’ or social milieu (Horvat 2001). In this study, then, we define cultural capital broadly as attitudes, dispositions, skills to navigate the system, academic preparation, disciplinary knowledge, and educational qualifications that are useful in earning legitimacy in a community. Linguistic capital, on the other hand, is sometimes subsumed under cultural capital and at other times used as a separate construct in Bourdieu’s work (1977, 1991). Since Bourdieu was primarily concerned with class-based reproduction of social inequality, he used the concept to refer to one’s facility with the privileged variety of a language (e.g. having the standard accent). But in this study, we use this concept to refer to competence in a language of high market value in a community. Cultural and linguistic capital are related to other forms of capital such as economic capital (money and property), symbolic capital (status and legitimacy), and social capital (networks and connections) (Bourdieu 1977, 1986, 1991). Naming these different forms of capital enables us to recognize that doctoral students’ participation in their disciplinary community may not depend solely on one form of capital, such as disciplinary knowledge or English proficiency, but rather on various forms of capital at their disposal and how they make use of them. Developed from our theoretical framework, then, the research questions that guided our study were:


675

2 To what extent do NNES students’ perceptions match those of more senior members of their disciplinary communities, i.e. their doctoral advisors? 3 How does the students’ ownership of different types of capital shape their participation and identities as legitimate members in English-speaking academic communities?

METHODS

Participants The participating students were Burnerman, Hou, Julius, and Kathy (Table 1). Burnerman, in his late 20s, and Hou, in his early 30s, were from the department of aeronautics and astronautics engineering and both were in their third year of study. The two students had majored in mechanical engineering in Taiwan’s top universities for their bachelor’s and master’s degrees. Before coming to the USA, Burnerman worked as a researcher for an industrial


The study was conducted at Padelford University,1 a major public research university in the USA. Four NNES students from Taiwan and their doctoral advisors were recruited for this study. We recruited PhD students from a single place of origin, Taiwan, in order to eliminate nationality-related variability. Taiwanese students occupy a large segment of the US higher education demography, ranking as the fourth leading place of origin for international students earning US doctorates (Welch 2008). Despite Taiwanese students’ large representation, only a few published studies have analyzed their experiences in US graduate schools (Schneider and Fujishima 1995; Swagler and Ellis 2003). Since Chang is Taiwanese, we decided that it would make sense to focus on Taiwanese students. To recruit participants for this study, we set the following selection criteria: (i) that they are full-time, matriculated Taiwanese students who came to study in the USA at the master’s or doctoral levels; (ii) that they have completed at least two years of their doctoral programs at the outset of the study; (iii) that their advisors agree to participate in the study. Our original goal was to recruit two students each from the sciences, social sciences, and humanities. The students interested in participating in the study were asked to respond to a recruitment e-mail we sent to academic advisors2 in the sciences, humanities, and social sciences departments. However, finding doctoral students who met all of the selection criteria under the limited range of recruitment strategies allowed by Padelford’s IRB3 turned out to be a challenge. In the end, we found one female and three male students along with their advisors. Since we are not making claims about the role of gender in NNES students’ success in US graduate schools, we determined that the gender imbalance was not critical to our inquiry.


Table 1: Four participants’ profiles Participants

Hou

Julius

Kathy

Gender

Male

Male

Male

Female

Doctoral program

Aeronautic and astronautics engineering 3

Economics

Chinese

Years in the program

Aeronautic and astronautics Engineering 3

5

5

Age

Late 20s

Early 30s

Mid-30s

Early 30s

TOEFL score

243

Around 270

253

Around 580 (paper-based)

Undergraduate education

Mechanical engineering (Taiwan)


Chinese literature (Taiwan)

Graduate education



English & industrial and information management (Taiwan) Financial engineering (USA)

Chinese literature (Taiwan)

research institute for nine months and Hou worked for a well-known IT company as an engineer for three years. In their doctoral program, Burnerman and Hou had different advisors and worked in different labs. At the time of the research, Burnerman had passed the qualifying exams but had not yet presented at conferences or published a journal article. He received a departmental scholarship in his first year of study and a teaching assistantship (TAship) in his second year. In his third year, he was paying his own tuition. After obtaining his PhD, Burnerman planned to work as a researcher in the US engineering industry. Burnerman’s advisor was a NES full professor with a long list of publications in the area of aerothermofluids. He oversaw two labs in the department. Hou had published one article in Chinese in a local academic journal in Taiwan before coming to the USA. Three years into the doctoral program, Hou had passed the qualifying exams, presented at two major international conferences, and submitted a journal article in English that was in the process of being revised. After supporting himself for the first two years of study, Hou received a research assistantship in his third year. Hou was planning to return to Taiwan and find a faculty position after finishing his PhD program. Hou’s doctoral advisor was a tenure-track assistant professor in the department and the only NNES advisor in this study. Although still a junior faculty member, Hou’s advisor already had numerous publications in aerodynamics and digital particle image thermometry. Hou’s advisor had received his undergraduate education and graduate education in the USA. He oversaw one lab in the department. Julius, in his mid 30s, was in his fifth year of PhD study. Unlike the other three participants, he studied in several different disciplines before he finally landed in economics. Back in Taiwan, Julius majored in computer science at a junior college and transferred to a university double-majoring in English and


Burnerman


677

Data collection Data collection took place during the 2007/08 academic year. Multiple forms of data were collected for this study, including interviews, shadowing observations, and written documents. Chang conducted all of the data collection while Kanno contributed to the theoretical framing, data analysis, and the writing of this article. Each student was interviewed four to six times, depending on the length of their responses. An interview protocol was prepared in advance to implement some consistency across the participants, but questions were adjusted during the interviews to accommodate each participant’s unique situation (see Supplementary Appendix A1 for sample interview questions). In addition to the four students, their doctoral advisors were also interviewed once individually. Each interview lasted between 60 and 90 min and was audio-taped and later fully transcribed. Questions asked in the interviews with advisors focused on their experiences working with NNES international students and their mentoring and teaching philosophies in scaffolding students to become


industrial and information management. He worked in non-engineering departments at two IT companies before coming to the USA. He studied for a master’s degree in financial engineering at a state university in the Midwestern US Julius subsequently enrolled in the doctoral program in mathematics at the same university only to discover that the content of study was too theoretical for him. He dropped out of the program after a year and transferred to the doctoral program in economics at Padelford University. He had been a TA in the department since his second year in the program. During the data collection, he was preparing for general exams. Although he was undecided about his future plans, he hoped to find a well paid job that would also allow him to continue doing research. Julius’ doctoral advisor was a NES full professor and also the director of the graduate program in the department of economics. He published in the area of mathematic economics. Julius was his only advisee. In her late 20s, Kathy was in her fifth year of study in the Chinese program in the department of Asian languages and literature. Kathy came to the USA for a graduate degree because she wanted to receive the best academic training and eventually become a Chinese modern literature scholar either in Taiwan or in the USA. Although she had already received a master’s degree in Chinese literature in Taiwan, she had to start the graduate program in the USA as a master’s student to fulfill the course requirements. She spent three years completing the master’s degree before she started the doctoral program. During the time of data collection, she was in the process of preparing for field exams, which test one’s knowledge of one’s chosen areas of research interest. Kathy worked as a TA teaching Chinese language classes on campus. Kathy’s advisor, a NES, was an associate professor specializing in modern Chinese literature. He has published one book and several journal articles.


Data analysis In analyzing the data, themes and patterns were extracted and clustered to reconstruct each student’s experiences as well as to make cross-participant and cross-disciplinary comparisons. The analysis of the students’ data focused on their participation and interaction in the disciplinary communities, their views on the role of language in their academic endeavors, and their self-positioning as NNES students. The analysis of the advisors’ data focused on their perceptions of the role of English in their disciplines and their observations of NNES students’ strengths and weaknesses.

NNES INTERNATIONAL STUDENTS’ PARTICIPATION IN THREE GRADUATE PROGRAMS In this section, the results are presented in three parts, corresponding to the three disciplines in which the four students belonged. In each part, we present (i) the practices and interactions in which the students engaged, (ii) the challenges the students experienced as non-native speakers, (iii) their perceptions of language dependence in their disciplinary communities of practice, (iv) the kinds of capital each student believed they possessed and (v) the doctoral advisors’ perspectives.


future professionals, or their fellow colleagues. As Chang is a native speaker of Mandarin Chinese, the participant students were given the choice to be interviewed in Chinese or English. All students chose to be interviewed in Mandarin Chinese with occasional English code-switching. Student interview excerpts in this study are Chang’s translations. The interviews with professors were in English. Each student was also shadowed for six full school days in conjunction with the interviews: each person was observed for two non-consecutive days followed by two or three interviews. They were then observed again for another four non-consecutive days and finally interviewed another two to three times. During the shadowing, the students were observed for the types of academic activities and interactions in which they were engaged. Detailed field notes were taken during the observations and follow-up questions were asked either at the end of the observation or during the following interview. In addition to interviews and observations, we also collected written documents, such as information on the graduate school website, departmental websites, graduate student handbooks, and students’ writing samples, in order to achieve a more contextualized understanding. We gathered writing samples that the students considered crucial to their participation in their disciplinary communities and asked them to explain the process of production as well as the academic practices and interaction related to each writing activity. We also examined departmental brochures and website information in order to situate participants’ experiences within the larger institution.


679

Engineering Students’ views

How fast and to what extent [the mechanics] can help you depends very much on how clearly you can articulate your needs. The more discussion you have with them, the more ideas you might be able to get from them. If your English is not good and you are not able to provide enough information, then the help will be limited. Nonetheless, Burnerman and Hou sometimes avoided interacting with NES lab mates, peers, or lab technicians and mechanics for fear of potential embarrassment, lack of understanding, and fatigue. Hou, a self-proclaimed introvert, said he was afraid that he might waste other people’s time and cause possible awkwardness: ‘I have a hard time understanding people when they speak [English] too quickly and I feel rather embarrassed to have to ask people to


For engineering majors like Burnerman and Hou, lab activities occupied almost all of their on-campus time. However, contrary to the stereotypical image of science and engineering laboratories where many graduate and postdoctoral students engage in intense collaborations, Burnerman and Hou’s lab lives were rather solitary due to the physical segregation of lab spaces and research project assignments. Burnerman was the only person running his lab and except for occasional visitors and weekly group meetings, he worked and studied in the lab alone. Hou, though sharing the same lab space with two white American male students, had a different project assignment so that opportunities for collaboration were limited. It was not uncommon to see Hou doing an experiment or reading quietly in one corner of the lab while his lab mates were engaged in a heated discussion in another corner. Despite the long, relatively solitary lab hours, Burnerman and Hou saw frequent communication with others in the disciplinary community as critical to accessing necessary support and generating new ideas for one’s research. Burnerman said, ‘After doing an experiment, you need to discuss and exchange ideas with others . . . through the communication, you gain valuable experiences and obtain ideas that you did not learn from doing the experiment alone’. Although they had already met the language requirements for admission, Burnerman and Hou felt that their English communicative skills for social interaction were still limited. According to Hou, mechanics in engineering departments were valuable members of the community to students because they often helped build experiment equipment as well as providing expert advice when students encounter problems in running experiments. However, Hou’s chances of gaining valuable hands-on knowledge from the seasoned mechanics were smaller than his American lab mates, he said, because he was not able to build as close a relationship with the mechanics as did his lab mates:


repeat’. Even for an extrovert like Burnerman, socializing with native speakers was still taxing: If I want to talk about my personal life or to ask about other’s personal lives, there is so much slang that I can’t possibly know without having a lot of native English speaking friends . . . if people start using [slang], I will be totally at a loss.

Shock location In this problem, assume it’s isentropic everywhere except at the location of the shock. At first, we should find the total pressure across the shock. P01 1 2 =1 M ¼ ½1 þ ¼ 2:625 2 P1 Since the process is isentropic, we get At =A = 1.05 when Mt ¼ 1:26 Then, Ae Ae At ¼ ¼ 1:74 ) Me ¼ 0:36 A At A P02 1 2 =1 Me ¼ 1þ ¼ 1:09 2 Pe Despite Burnerman’s extra effort to add more verbal explanation, we can see that the meaning of the content still depends largely on mathematical formulae and discipline-specific terminologies. In addition to such discursive practices in the discipline, the existence of many successful NNES scholars in engineering also contributed to Hou and Burnerman’s belief that English competence did not influence one’s academic achievement in their discipline. For these NNES novices who were trying to work their way up the academic ladder, it was reassuring to find models of active NNES scholars making significant contributions in the English-speaking academe. NNES scholars’ strong presence in the field demonstrated to them


On the other hand, neither Burnerman nor Hou thought that their academic performance itself was particularly hampered by their English. The representation and evaluation of disciplinary knowledge depended largely on the use of mathematic language and discipline-specific terminologies. As Hou put it, ‘If you get it, then you get it. A good command of language or English is not going to give you much of an advantage’. What follows is an excerpt from one of Burnerman’s papers for a computational fluid dynamics course he took in his first year in the program. Burnerman said he did not receive good grades for the first few assignments in this course because he filled his papers with mathematic formulae without much textual explanation. This example, then, illustrates his conscious effort to add more text, and he received good feedback from the instructor.


681

Advisors’ views According to Hou’s and Burnerman’s doctoral advisors, the ability to think and research independently and make original contributions were among the most important qualities doctoral students should develop in the discipline. However, countering their students’ views, the two professors believed that their discipline was relatively language-dependent. They stressed that even though engineering students often did not seem to use language as often, language competence was no less critical to them than to students in other disciplines. Hou’s professor, a NNES scholar, argued: I would say maybe for the [students in] social sciences, they would have to talk in English more often than in the science because in the science [students] are probably in the lab, doing their experiments. They don’t need to talk that much . . . But . . . the level of importance that I give to be able to communicate what you do is absolutely critical. Like I said, if you can’t communicate what you’ve discovered, it’s worthless. At the same time, the advisors did not consider the linguistic standards required for effective communication to be a problem for NNES students.


that in their field, the value of disciplinary knowledge and the ability to communicate such knowledge through mathematical codes and specialized terminologies outweighed that of general English academic proficiency. Hou’s previous conference presentation and journal publication experiences also assisted him in participating in similar academic activities in his present doctoral program in the USA. Since he was already familiar with the genres, ‘it became a matter of filling in the contents in English’. Hou also found that Taiwanese students generally have higher computer literacy and skills, which gave them a competitive edge in the aeronautics and astronautics engineering program. He felt that even though American students’ often seemed more expressive and more aggressive, they did not necessarily know more. Burnerman was also aware of the different kinds of cultural capital that Taiwanese and American students possessed. He pointed out that his past education in Taiwan had trained him well in doing mathematic calculations and in exam-taking skills, but less well in the areas of hands-on exercises and critical and creative thinking skills. Burnerman emphasized that the most rewarding and empowering part of his overseas studies was the opportunity to experience different orientations to academic training. Although there were challenges, Burnerman chose to see them as a blessing disguised as an ordeal and believed that if he could succeed in his doctoral studies, he would be stronger mentally and intellectually. ‘Is it worthwhile’? asked Burnerman rhetorically. By the time I leave this place, I am sure I am going to answer in the affirmative’.


Economics Student’s view Unlike the engineering students, Julius preferred to work at home when he was not teaching or attending classes. Although he shared an office with three other TAs in the department, he hardly knew his officemates because he rarely used the space except during his office hours. However, Julius’ academic undertaking was no less intense than that of the engineering students. He seemed absorbed in his academic work and rarely talked about anything else in his interaction with Chang. Despite having a B.A. degree in English and previous graduate school experience in the USA, Julius felt that he could only ‘talk about really shallow stuff’ with native-English speakers and did not have much interaction with his NES peers in the department. Fortunately, Julius said his lack of communicative competence in English did not really interfere with his academic work. According to Julius, although the required levels of English varied across different branches of economics research, Julius argued that his field of research, econometrics and microeconomics, relied heavily on mathematical and visual representations in expressing ideas. He therefore had not encountered many problems caused by English per se. Rather, he attributed the challenges he encountered to his unfamiliarity with disciplinary knowledge: ‘In fields like finances, math, or economics, the definitions of terminologies depend very much on the [disciplinary] contexts. It is not something you may find in the dictionary’. The following excerpt is from a literature critique Julius wrote for a labor econometric class. Julius noted that the course was in the area of labor economics, which is less mathematics-centric and closer to sociology and humanities. Even so, Julius’ paper shows that information and argumentation still relies mainly on mathematic representations and would be challenging for a non-specialist to understand.


Although Burnerman’s advisor thought that some international students needed more help with writing, he had not experienced many problems with NNES students’ English overall as long as they had met the admission language requirements. Hou’s advisor also argued that the abilities to think independently and to solve problems transcended culture. By the time NNES students made it to the graduate program, he said, ‘they have sufficient understanding of English that I am fully capable of communicating with them’. In short, although both advisors placed more importance on the role of English in their disciplinary practices than did their advisees, they also believed that NNES students who met the admission language requirements already possessed sufficient English proficiency to meet the communicative needs in their discipline.


683

II Model specification A regression, such as Equation 1, controls for measured personal and background characteristics: Oi ¼Xi þ Pi þ e

I find that their ability to express in English was not necessarily good, some even worse than mine. However, they were able to come up with unique and valuable academic ideas and express them through mathematics or graphics . . . their bad grammar or inaccurate word choices did not matter at all. In terms of his teaching as a TA, Julius said his English did not hinder his communication with students. While he still worried about whether his misuse of words or expressions might impede students’ understanding, he rehearsed his lectures in advance and used many examples to present his ideas clearly in class: ‘In other words, knowing that I might not be able to offer crystal clear explanations for economic concepts, I am using examples to help me do the job’. However, all this preparation took time. A student of limited economic means, Julius needed the TAship to continue with his program: International students in the USA have limited eligibility to apply for student loans and internships and cannot legally work outside of the university. Although Julius did not think his overall teaching quality was hampered by his English proficiency, he thought that the criteria for the renewal of the TAship appointment worked against NNES TAs: One of the key determinants for [TAship] funding is teaching evaluation. But the problem is that it is very hard to teach better than your American peers . . . They can probably prepare a perfect lecture with 20 hours of preparation. But for me, I will have to spend a lot [more] time trying to figure out the way to express myself clearly in English. Overall, however, having studied and worked in multiple fields, Julius strongly believed that his interdisciplinary background not only provided a niche for


where Oi indicates the outcome for respondent i; Xi: a function of personal and background characteristics; Pi: a vector of variables measuring family or parental composition Julius also found that except in top economics journals, the English level required to write publishable journal articles was not particularly high. He believed that, unless one’s writing skills in English were severely limited, the content of the research was what counted. ‘Even if there are obvious language problems, they might not be the reason that they reject you for’, said Julius. Furthermore, similar to Burnerman and Hou in aeronautics and astronautics engineering, Julius’ perception of the role of English in his discipline was also influenced by the many NNES scholars in the field of economics:


him in the academic community, but also gave him self-assurance and resilience in navigating his current disciplinary community: After changing so many disciplines, I realized that tireless and repetitive practice would eventually get me through challenges . . . I know I probably won’t be able to receive a Nobel Prize with tireless and repetitive practice, but it will get me somewhere.

Advisor’s view Echoing Julius’ view, Julius’ advisor also noted that their area of research was relatively language-independent: work in econometrics and microeconomics involved developing mathematic models and analyzing data. ‘The extent that you are doing one of those two things, you are not going to use much English,’ he said. According to Julius’s advisor, a competent doctoral student in his research area should have up-to-date technical proficiency with numbers and model building as well as the ability to think rationally and produce cutting-edge and publishable research. However, he cautioned that the language independence of the field did not mean that English was unimportant. He stressed that students needed to be proficient enough to be able to read publications and write in a ‘comprehensible’ manner. Based on his experience working with NNES students, Julius’s advisor said that English competence was regarded as more critical only for students who aspired to teach in English-speaking contexts. Otherwise, he felt that NNES students’ linguistic problems generally remained on the superficial level, grammar, and word choice, which did not hamper communication and mutual understanding. In reference to the types of academic competence valued, Julius’ advisor said that while US students typically had the linguistic advantage, many international students had a stronger technical background, an important form of cultural capital in this field that gave international students a competitive edge. Even when learning disciplinary jargon in English, Julius’s advisor observed, many NNES students’ seemed to be able to acquire the terms as quickly as US students. Therefore, ‘as far as plain economics goes, American students’ advantage might not be that great after all’.

Chinese Student’s view As a TA, Kathy shared an office space on campus with another international TA in the department. Depending on her teaching and class schedule, Kathy


Julius said that linguistic challenges would compel NNES students to strengthen their non-linguistic competences (i.e. math, programming, etc.) so that once their English proficiency improved; they would be equipped with multiple skills. ‘It might be frustrating at the beginning,’ he said, ‘but I think God is going to reward fairly’.


685

usually spent about half of her day teaching, attending classes, and working in her office on campus, and the other half doing her academic work at home. Even though there were relatively more international and Asian students in her department, Kathy still found herself rather isolated due to both linguistic and cultural barriers as well as the solitary research environment in literature disciplines. The need to cope with academic demands also forced her to give up social and extracurricular activities:

However, since Kathy observed that most of the students and scholars in her discipline lived a solitary academic life, she did not feel disadvantaged because of her lack of social life. Although Kathy was in the Chinese program, she considered high proficiency in English to be critical to her performance. She noted that the vocabulary used in her discipline were ‘GRE words’ and the standards for English proficiency in her area of study were beyond fluent and grammatical: ‘It is not a matter of whether you are literate in English, but whether your English is good’. For Kathy, the need to use advanced academic English to express her thoughts aesthetically and persuasively was one of the greatest challenges in her studies. She made a concerted effort to improve her language skills by frequently consulting dictionaries, having weekly Chinese-English language-exchange meetings with a native-English speaker, and seeking writing tutorials from her doctoral advisor. It was also true, however, that Kathy enjoyed the advantage of being a native Chinese speaker, a highly coveted form of linguistic capital in her department. First of all, Kathy had the privilege of receiving a TAship, which brought her steady financial support from her second year of study. In teaching, her authority as a native Chinese speaker more than compensated for her non-native English. Kathy reported that her students tolerated the imperfections in her English without undermining her knowledge: Although my grammar is poor, my mistakes are not preventing students from understanding me . . . They probably just think that I just said something ungrammatical but . . . it is not like ‘I don’t understand what you are talking about’. Moreover, Kathy’s native Chinese-speaker status also gave her an advantage in the speed of reading and the depth of cultural understanding when working with original Chinese texts: ‘[For non-Chinese students,] the Chinese culture has to be learned, it is not in their bones’. While American students may be able to write linguistically more sophisticated prose in English, Kathy believed that people from Chinese culture were in a better position to understand


If I had to read every book assigned by my professors . . . and write [assignments] which would only take native-English-speaking students two days but take me five . . . I would need a lot of time [to study,] . . . As for having a good time [with friends], well, just forget it! So . . . I actually feel like a loner.


Chinese texts in greater depth and offer insider perspectives in their analysis. She was proud that even her NES advisor sometimes had to consult her on rarely used Chinese characters. Immersed in an English-speaking environment, Kathy felt that her chances of becoming more competent in English were far better than her American peers’ chances of improving their Chinese. Furthermore, Kathy pointed out that even though NNES students often need to pursue their degree without either familial or financial support, the lack of support actually gave them the pressure to work harder. As a result, they ‘are able to finish the degree faster than local students,’ she claimed.

Sharing Kathy’s view, Kathy’s doctoral advisor also stressed that English was very much the tool of the trade in the study of Chinese literature in the English-speaking research community: Students need to present themselves convincingly with accuracy and sensitivity and even a sense of humor in English speaking and writing . . . [If not] it’s not likely that they are going to get very far in the American academy. According to Kathy’s advisor, the participation of Chinese-speaking international students had raised the standards of Chinese language competence and familiarity with Chinese literature in the program. However, since he upheld critical thinking skills as the most important skills for graduate students, Kathy’s advisor also pointed out that NNES students tended to be underprepared in this area. This weakness, together with their unfamiliarity with academic genre conventions, he noted, could prevent NNES students from producing satisfactory work in US graduate school. As a doctoral advisor, he felt responsible for providing extra instruction on academic writing to his NNES international advisees, including Kathy: ‘You know . . . they are already here as a graduate student. They are not going to get this training [elsewhere] and they need it. And therefore it falls to me to try to at least make a little bit of an effort’. Without any prior training on the argument and logic structure of advanced academic writing, Kathy was especially grateful to a quarter-long, one-on-one tutorial on academic writing offered by her advisor: ‘The practice was really helpful . . . Although the issue with accurate word choice and usage can’t be fixed right away, at least I have had fewer problems with the development of an argument’. The following excerpt is from one of the writing exercises—a book review—Kathy wrote under the guidance of her advisor. The difference between this excerpt and those from Julius and Burnerman is clear. The sentences are longer and more complex. Words play the central role in conveying ideas: Hung’s book is indeed inspiring because Hung well organizes abundant materials which cover at least four forms of popular literature


Advisor’s view


687

in the Sino-Japanese War period and points out the significance of the war-time period to the development of popular culture. But I still have a reservation about the prevalence of popular culture in the war-time period. Hung’s main argument in this book is that, in the Sino-Japanese War period, popular culture was not just ‘urban popular culture’ any more but a culture that penetrated to ‘interior provinces’.

DISCUSSION In examining NNES students’ disciplinary enculturation in English-speaking graduate schools, researchers have tended to assume the primacy of English in their academic endeavors and have analyzed students’ experiences through their efforts to acquire the linguistic and rhetoric knowledge required to participate in academic practices. This study, however, has explored NNES students’ participation in disciplinary communities from the students’ and their advisors’ emic perspectives. Our findings can be divided into two major themes. In what follows we discuss each in turn. The first central finding is that the importance of language competence varied across disciplinary communities, community practices, and different community members. We began this study with relatively simplistic predictions of what we would find: Disciplines in the humanities would be the most language dependent, while the sciences and math would be the least language dependent, with the social sciences being somewhere in between. What we found instead was a much more complex picture of when and how language competence mattered in NNES students’ learning and becoming legitimate members of the academic community. More specifically, the results of this study suggest the complexity of language issues at four different levels. First, the degree of language dependence varied not only among disciplines, but also among subdisciplines. For instance, in the department of economics, Julius and his advisor maintained that their research area was less language-dependent than other branches of economics, such as labor economics. Similarly, Kathy’s advisor commented that even within the Chinese program, the work of those in Chinese linguistics was more ‘technical’ compared with those in the literature track: ‘For us, the [English] requirements are steeper’, he said. Second, as far as the NNES students were concerned, language competence was not always critical to their academic participation and performance. Passing the minimal language requirements at an American university


According to Kathy, in writing book reviews, she learned to write a ‘strong first sentence’ that provides an overall evaluation of a piece of work. Kathy said that while training her to become a better writer, her advisor was also training her in the American way of doing research, ‘so it is really a good practice to the academic ways of thinking and writing . . . and expressing myself in English’.



meant very different things to NNES students in different disciplines. To Kathy, passing the institutional language requirements simply meant that a student had the minimal linguistic capital to get started with the program; a much higher level of competence was required to become a full member of her community. In contrast, for Burnerman, Hou and Julius, passing the same minimal language requirements meant that the students already had a sufficient level of linguistic capital to meet the academic demands in their disciplines; NNES students’ effort thus was better spent on accumulating more highly-valued cultural capital, such as scientific or mathematic expertise, than on the further improvement of their English. Third, the importance of English, and therefore the disadvantage of being a non-native speaker, varied across different community practices within a disciplinary community. Hou, Burnerman, and Julius saw their fields of research as largely language-independent, and as far as their academic performance was concerned, they did not feel any setback because of their English proficiency. On the other hand, some aspects of their community participation and interaction were clearly affected by their non-native English proficiency, such as their ability to socialize easily with other community members. Thus, even within a single disciplinary community, NNES students may find themselves participating competently in some shared practices while they may find it harder to negotiate an identity of competence in other practices. Fourth, different members of the same disciplinary community may perceive the discipline’s linguistic demands and students’ linguistic needs differently. While the two engineering professors regarded their discipline to be no less language dependent than other disciplines, their NNES apprentices saw linguistic competence playing a much less critical role in their academic pursuits. This discrepancy may be explained by the different levels of community practices in which junior and senior members engage. As third year doctoral students, Burnerman and Hou were newcomers to the disciplinary community. They were not yet on the front line of knowledge production and dissemination as were their advisors. They therefore may not have perceived a sophisticated and exact use of English as part of the discipline’s necessary repertoire. Their advisors, in contrast, were the ones who had to publish journal articles, present at conferences, and write grant applications. As such, they were more keenly aware of the role of English in disseminating and marketing their research. In other words, as one’s position in a discipline community changes over time, one’s perception of the importance of language and rhetoric may also change (Casanave and Hubbard 1992). A second central finding of the study is that the NNES students were able to make use of the cultural capital they possessed in order to claim legitimate membership in their disciplinary communities. In some areas, they readily acknowledged their ‘otherness’ because of their linguistic and cultural differences. In other areas, however, they spoke of their differences in positive terms. The NNES students did not enter the English-speaking academic community as blank slates (Hirvela and Belcher 2001). Having attended college,


689

CONCLUSION In this study, we have provided an analysis of the emic perspectives of NNES students and their advisors in three different disciplines. The findings of the study challenge the often negative portrayals of NNES students’ positions in the English-speaking academe and the presumed centrality of English in most previous studies. Although the international students in this study were aware of their disadvantages as non-native speakers, they felt that they were nonetheless legitimate members of their disciplinary communities because they have other cultural and linguistic capital that enabled them to display enough competence in their communities. In showing these results, our intent was not to gloss over NNES students’ linguistic and cultural challenges. Rather, the findings of the study underscore (i) the importance of examining NNES students’ experiences within the specific cultural, linguistic, and disciplinary contexts in which they are situated, and (ii) the ways in which the students make use of the resources they do possess to vie for legitimate membership in their disciplinary communities. This study only examined the experiences of four NNES students in three disciplines, and within those disciplines the students’ overall experiences were


graduate school, and, for some, worked in the industry, they were most likely seeking advancement in, rather than entry into, their disciplinary communities. With the knowledge and skills Burnerman, Hou, Julius, and Kathy had accumulated through their past academic and professional experiences, they had confidence in themselves. They knew that they possessed the kinds of cultural capital that had high market value in their disciplines: be that knowledge in mathematics and physics, proficiency in Chinese, publication experiences, computer programming skills, or interdisciplinary education backgrounds. In other words, compared with their NES peers, the NNES students in this study viewed themselves as no less legitimate, albeit relatively new, members in the English-speaking academic community. Their cultural capital certified these foreign students’ membership so that in their still peripheral but legitimate positions, they were able to regard the challenges they encountered as part of the ‘doctoral package’. Likewise, their doctoral advisors also believed that NNES students brought positive influences to enrich the English-speaking academic community. The advisors did not think that the four students suffered marked disadvantages because of their NNES status. In short, rather than participating as marginalized members, the students related their experiences in their programs with a strong sense of pride and hope. They saw themselves as competent and legitimate members of the academic communities despite the extra linguistic and cultural barriers they encountered. By going through the extra work and challenges, they believed that their cross-cultural investment would yield high returns: They were going to be better and stronger academically and mentally compared with not only local NES students, but also doctoral students trained in their home country.



ACKNOWLEDGEMENTS We would like to thank the participants in this study, who generously shared their experiences and insights with us. We are also grateful for the helpful suggestions of Dr Jane Zuengler and three AL anonymous reviewers on earlier versions of this study. Finally, Ms Sarah Grosik enhanced the quality of this work with her thorough copyediting.

NOTES 1 All the names of institutions and individuals in this study are pseudonyms. The students’ pseudonyms were chosen by themselves. 2 Academic advisors are staff members who provide advice to students about

course taking and program requirements. They are different from doctoral advisors, who are faculty members who supervise doctoral students’ research and dissertation work.


positive. However, are such positive experiences shared by NNES students in other disciplines—in education, English literature, or political science, for instance? In Morita’s (2004) study, Lisa, a NNES student in an MATESOL program in a Canadian university had difficulty negotiating her sense of competence and legitimacy because she did not speak perfect English amid her NES, English-teacher colleagues. In a discipline like TESOL, in which native or near-native English proficiency is often assumed, do NNES students have the same kind of leverage to claim their legitimate membership as those in this study? According to Kiang (1992), Asian immigrant students in the USA tend to concentrate in a narrow band of majors such as business, engineering, and computer science for reasons that include these disciplines’ lighter English demands, potential financial returns in the job market, and the more objective (and therefore perceived to be less culturally and linguistically biased) evaluation criteria. Cho’s (2004) study also suggests that NNES students tend to gravitate towards quantitative rather than qualitative research because quantitative research is perceived to be less language-dependent. When 37 per cent of international students (including undergraduate and graduate students) choose business or engineering while only 2.9 per cent choose humanities (Institute of international Education 2009), there must be a reason. The results of the study, then, raise the question of whether NNES students intentionally and strategically choose the disciplines in which their cultural capital would enjoy a higher market value. In future research, it would be important to examine not only how well NNES students fare in a variety of fields, but also on what criteria they choose their area of study and what role English plays in their selection.


3 Padelford University has an extremely stringent IRB process, and we were limited in how we were allowed to recruit students. For instance, we were not allowed to recruit participants

691

through our personal contacts or use ‘snowballing’ strategies (Patton 1990; Heckathorn 1997; Atkinson and Flint 2001) commonly used in social science research.

REFERENCES Casanave, C. P. and X. Li. 2008. ‘Introduction’ in C. P. Casanave and X. Li (eds): Learning the Literacy Practices of Graduate School: Insiders’ Reflections on Academic Enculturation. The University of Michigan Press. Cho, S. 2004. ‘Challenges of entering discourse communities through publishing in English: Perspectives of nonnative-speaking doctoral students in the United States of America,’ Journal of Language, Identity, and Education 3/1: 47–72. Dong, Y. R. 1998. ‘Non-native graduate students’ thesis/dissertation writing in science: self-reports by students and their advisors from two U.S. institutions,’ English for Specific Purposes 17/4: 369–90. Golde, C. M. 1998. ‘Beginning graduate school: explaining first-year doctoral attrition,’ New directions for higher education 101: 55–64. Heckathorn, D. D. 1997. ‘Respondent-driven sampling: A new approach to the study of hidden populations,’ Social Problems 44: 174–99. Hirvela, A. and D. Belcher. 2001. ‘Coming back to voice: The multiple choices and identities of mature multilingual writers,’ Journal of Second Language Writing 10/1/2: 83–106. Horvat, E. 2001. ‘Understanding equity and access in higher education; The potential contribution of Pierre Bourdieu,’ Higher Education; Handbook of Theory and Research 16: 195–238. Hyland, K. and P. Tse. 2007. ‘Is there an ‘‘academic vocabulary’’?’ TESOL Quarterly 41/2: 235–53. Institute of International Education. 2009. ‘Open Doors 2009 data tables,’ available at http://opendoors.iienetwork.org/?p=150827. Accessed January 2010. Kiang, P. N.-C. 1992. ‘Issues of curriculum and community for first-generation Asian


Angelova, M. and A. Riazantseva. 1999. ‘If you don’t tell me, how can I know?’: A case study of four international students learning to write the U.S. way,’ Written Communication 16/ 4: 491–525. Atkinson, R. and J. Flint. 2001. ‘Accessing hidden and hard to reach populations: snowball research strategies,’ Social Research Update, 33. Ballard, B. 1996. ‘Through language to learning: preparing overseas students for study in Western universities’ in H. Coleman (ed.): Society and the Language Classroom. Cambridge University Press. Bourdieu, P. 1977. ‘The economics of linguistic exchanges,’ Social Science Information 16/6: 645–68. Bourdieu, P. 1986. ‘The forms of capital’ in J. G. Richardson (ed.): Handbook of Theory and Research for the Sociology of Education. Greenwood Press. Bourdieu, P. 1991. Language and Symbolic Power. Harvard University Press. Bourdieu, P. and J.-C. Passeron. 1990. Reproduction in Education, Society and Culture. 2nd edn. (R. Nice, Trans.). Sage. Casanave, C. P. 2002. Writing Games: Multicultural Case Studies of Academic Literacy Practices in Higher Education. Lawrence Erlbaum Associates. Casanave, C. P. 2008. ‘Learning participatory practices in graduate school: Some perspective-taking by a mainstream educator’ in C. P. Casanave and X. Li (eds): Learning the Literacy Practices of Graduate School: Insiders’ Reflections on Academic Enculturation. The University of Michigan Press. Casanave, C. P. and P. Hubbard. 1992. ‘The writing assignments and writing problems of doctoral students: Faculty perceptions, pedagogical issues, and needed research,’ English for Specific Purposes 11/1: 33–49.


Parry, S. 1998. ‘Disciplinary discourses in doctoral theses,’ Higher Education 36: 273–99. Patton, M. Q. 1990. Qualitative Evaluation and Research Methods. 2nd edn. Sage Publications, Inc. Poyrazli, S. and P. R. Kavanaugh. 2006. ‘marital status, ethnicity, academic achievement, and adjustment strains: the case of graduate international students,’ College Student Journal 40/4: 767–80. Schneider, M. and N. K. Fujishima. 1995. ‘When practice doesn’t make perfect: The case of a graduate ESL student’ in D. Belcher and G. Braine (eds): Academic Writing in a Second Language: Essay on Research and Pedagogy. Ablex. Shaw, P. 1991. ‘Science research students’ composing processes,’ English for Specific Purposes 10: 189–206. Swagler, M. A. and M. V. Ellis. 2003. ‘Crossing the Distance: Adjustment of Taiwanese Graduate Students in the United States,’ Journal of Counseling Psychology 50: 420–37. Tardy, C. M. 2004. ‘The role of English in scientific communication: Lingua franca or tyrannosaurus rex?,’ Journal of English for Academic Purposes 3: 247–69. Tardy, C. M. 2005. ‘‘‘It’s like a story’’: Rhetorical knowledge development in advanced academic literacy,’ Journal of English for Academic Purposes 4/4: 325–38. Welch, V. Jr 2008. Doctorate recipients from United States Universities: Selected Tables 2007. National Opinion Research Center, available at http://www.norc.org/NR/rdonlyres/2D5FD7 C8-4AE0-4932-B777-0BC8EA7965EF/0/2007_ selectedtabs.pdf. Accessed 20 May 2009. Wenger, E. 1998. Communities Cambridge University Press.

of

Practice.

Zimmermann, S. 1995. ‘Communication competence and international student adaptation to an American campus,’ Communication Education 44: 321–35.


Americans in college’ in L. S. Zwerling and H. B. London (eds): First-generation Students: Confronting the Cultural Issues. Jossey-Bass. Kim, S. 2006. ‘Academic oral communication needs of East Asian international graduate students in non-science and nonengineering fields,’ English for Specific Purpose, 25: 479–89. Kuwahara, N. 2008. ‘It’s not in the orientation manual: how a first-year doctoral student learned to survive in graduate school’ in C. P. Casanave and X. Li (eds): Learning the Literacy Practices of Graduate School: Insiders’ Reflections on Academic Enculturation. The University of Michigan Press. Lacina, J. 2002. ‘Preparing international students for a successful social experience in higher education,’ New Directions for Higher Education 117: 321–7. Lamont, M. and A. Lareau. 1988. ‘Cultural capital; Allusions, gaps and glissandos in recent theoretical developments,’ Sociological Theory 6: 153–68. Lave, J. 1996. ‘Teaching, as learning, in practice,’ Mind, Culture, and Activity 3/3: 149–64. Lave, J. and E. Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge University Press. Li, Y. 2006. ‘A doctoral student of physics writing for publication: A sociopolitically-oriented case study,’ English for Specific Purposes 25: 456–78. Morita, N. 2000. ‘Discourse socialization through oral classroom activities in a TESL graduate program,’ TESOL Quarterly 34/2: 279–310. Morita, N. 2004. ‘Negotiating participation and identity in second language academic communities,’ TESOL Quarterly 38/4: 573–603. Parks, S. and P. M. Raymond. 2004. ‘Strategy used by nonnative-English speaking students in an MBA program: Not business as usual,’ The Modern Language Journal 88/3: 374–89.

Applied Linguistics: 31/5: 693–713 ß Oxford University Press 2010 doi:10.1093/applin/amq031 Advance Access published on 16 September 2010

The Effects of Repetition and L1 Lexicalization on Incidental Vocabulary Acquisition CHUNTIEN CHEN and *JOHN TRUSCOTT National Tsing Hua University *E-mail: [email protected]

INTRODUCTION A common, if not universal, assumption in vocabulary learning research is that words are learned incidentally in reading and that this learning is central (e.g. Nagy et al. 1985, 1987; Krashen 1989; Nagy 1997; Paribakht and Wesche 1999; Paribakht 2005). But in many respects this incidental vocabulary learning is still poorly understood. The present study focuses on two key questions. One is the effect of number of exposures on the development of various aspects of word knowledge. This line is based on the work of Webb (2007) but differs in crucial respects and thus complements that work. The second issue is the influence of L1 lexicalization on the acquisition of word meaning (see Paribakht 2005), a little-explored but potentially crucial aspect. Each area is discussed in the following sections.


This study of incidental vocabulary acquisition through reading, involving 72 freshmen at two Asian universities, investigated (i) the effect of repeated encounters with target words on the development of seven aspects of word knowledge, and (ii) the effect of L1 lexicalization on the acquisition of meaning—does the absence of an L1 translation equivalent make acquisition of a word’s meaning especially difficult? Seven measures were used, immediately after the treatment and again after a two-week delay. The design was based on that of Webb (2007) but emphasized ecological validity over control, presenting genuine words in meaningful reading passages and thus complementing Webb’s more tightly controlled experiment. Results for repetition largely supported his findings while suggesting that the nature of his study led to an overestimate of learning. Knowledge of orthography, part of speech, and meaning showed different patterns of development with increasing encounters. On the immediate posttest, repetition affected productive knowledge somewhat more than receptive, consistent with Webb’s findings, but this relation reversed on the delayed posttest. For L1 lexicalization, non-lexicalized words caused great difficulty.

694 REPETITION AND L1 LEXICALIZATION

Effects of repeated word exposures on incidental vocabulary learning


Past research has shown that the number of times an unknown word is met in reading affects whether its meaning will be acquired and retained (e.g. Saragi et al. 1978; Jenkins et al. 1984; Rott 1999; Webb 2007). There is no clear conclusion, though, regarding the number of encounters needed (e.g. Saragi et al. 1978; Nation 1982; Jenkins et al. 1984; Rott 1999; Zahar et al. 2001). This uncertainty is not surprising, as results are presumably influenced by a number of mediating variables, including learners’ proficiency level (Zahar et al. 2001; Tekmen and Daloglu 2006), context informativeness (Nagy et al. 1987; Shu et al. 1995; Webb 2007, 2008), and word properties (Nagy et al. 1987; Shu et al. 1995). Thus, the goal of research should be not to identify a definitive number of exposures needed but rather to understand a complex process involving multiple, interacting variables. Another crucial element is the type of word knowledge being investigated. Previous studies have usually focused on meaning, disregarding the many other types. In Nation’s (2001) classification, these include form (sound, spelling, and word structure), meaning (including associations, referents, and the concept expressed), and use (the patterns a word appears in, its collocations, and constraints on its use). But most have received little attention in studies of incidental learning. Webb (2007) is the main exception. He tested 10 different aspects of word knowledge, including receptive and productive. All aspects improved with number of repetitions (1, 3, 7, 10), consistent with previous findings that repetition has a positive impact. He also found that knowledge of word meaning developed more slowly than other aspects, such as syntax and word association, conforming to previous findings that some properties of a word are mastered before others (Schmitt 2000). An additional finding was that scores on productive measures were significantly greater after seven and ten exposures than after fewer exposures, suggesting that seven may be a threshold for productive knowledge. Webb’s (2007) study emphasized stringent control, giving it important advantages. But the price was a reduction in ecological validity. Specifically, invented words were used as targets, each precisely corresponding in meaning to an L2 word the learners already knew, and the words were presented in isolated sentences. The virtue of this approach is that it allowed control of the difficulty of individual words and the helpfulness of the contexts in which they appeared and therefore made possible a relatively reliable inference that the independent variable, repetition, did indeed produce the observed results. But the downside is that it made the learning task rather removed from actual second language learning through reading. Thus, research that explores the issues with more concern for authenticity is a necessary complement to Webb’s work.

C. CHEN AND J. TRUSCOTT

695

The impact of L1 lexicalization on vocabulary learning

The present study Webb (2007) made substantial contributions to our understanding of the effects of repetition, by examining several aspects of knowledge and by imposing stringent controls. But, again, the emphasis on control necessarily limited the ecological validity of the work. A necessary complement to Webb’s study is one that favors authenticity (ecological validity) over control. This complementary approach also has disadvantages, in the form of looser controls and therefore limits in the ability to determine causal relations. The hope is that the


An issue that has potentially profound implications for incidental word learning, but has received very limited attention, is L1 lexicalization. An unlimited number of concepts can be expressed in the vocabulary of a language, and languages naturally vary in which they lexicalize. The Mandarin word xuejie, for example, means ‘fellow student who is female and is older’, a concept not lexicalized in English. Will such words pose special difficulties for learners? The artificial words used by Webb (2007) were all synonyms of words the learners already knew. Thus, participants simply needed to map a new label onto a familiar concept. However, learning new words often involves new concepts as well (Nagy et al. 1985, 1987). When all the target words have meanings directly corresponding to those of known words, a study is looking only at the easiest cases of learning. The importance of the issue has long been recognized (e.g. Swan 1997; Jiang 2000), but to date, research on the impact of L1 lexicalization patterns on L2 learning is remarkably limited. Blum and Levenston (1979: 52) found that ‘learners avoid quite systematically words that have no equivalent in their mother tongue’. Similar findings were reported by Sjo¨holm (1998). Yu (1996) found that learning of L2 verbs was affected by their semantic similarity with L1 verbs. But these studies did not directly address the issue of how L1 lexicalization affects learning, particularly in relation to vocabulary acquisition through reading. Paribakht (2005), working with Farsi-speaking English learners, had participants read passages containing both lexicalized (L) and non-lexicalized (NL) words, using think-aloud to study their inference processes. She found substantially greater success for L words, but little difference was found in learning. Because her focus was on inferencing, she did not explore learning in any depth, but the lack of a meaningful difference can be readily attributed to the fact that very little learning occurred in general. One finding did suggest a difference in learning difficulty—before (and after) the study, learners had greater knowledge of L than NL words. Thus, NL words might be expected to pose special problems for learning, as Paribakht (2005) suggested, but little evidence exists either to support this idea or to disconfirm it. The extent of any such problems, if they do exist, is also an open question. Further research is clearly needed in this area.


1 How does repetition of words in reading passages affect L2 learners’ acquisition and retention of orthography, part of speech, associations (receptive and productive in each case), and meaning and form? 2 Are the meanings of L2 words that are not lexicalized in learners’ L1 especially difficult to learn and/or retain, as measured by an L2–L1 translation task and receptive and productive association tests? How does number of exposures contribute to the learning and retention of L and NL words?

METHOD Participants The participants came from two classes of Mandarin-speaking freshmen (n = 72) at two universities in Taiwan, all majoring in foreign languages, with an emphasis on English. All were intermediate-level learners with at least six years of English instruction. Their mean score on the 2000 word level of Version 1 of the Vocabulary Levels Test (Schmitt 2000), given to them before the experiment, was 28.1 out of a possible 30, indicating a mastery of this level. Participants were randomly divided into three groups, each receiving a different number of exposures to the target words.

Target words Target words were 10 English words (Table 1) not known to any of the participants, based on responses to a checklist containing 100 potential targets.


use of both types of research will allow a more complete picture to be formed. The first goal of the present study is thus to provide a complement to Webb’s (2007) research. The second goal is to further explore the influence of L1 lexicalization on incidental vocabulary learning, based on the potential importance of this factor and the surprisingly limited empirical work so far done on the subject. This exploration differs from past work in that it involves a different L1 (Mandarin Chinese) and a different set of concepts and also examines the role of repetition in relation to the lexicalization variable. It also tests the effects of lexicalization after a delay period (two weeks), which Paribakht (2005) was unable to do. The experiment, involving university freshmen majoring in English, was methodologically based on Webb’s (2007) but differed from it in several respects. First, actual English words served as the targets of learning. The ten target words included five that are lexicalized in the participants’ L1 (Mandarin) and five that are not. These words were presented in the context of L2 reading passages rather than isolated sentences. Finally, a delayed posttest was carried out two weeks after the treatment. The experiment was thus guided by the following research questions:


697

Table 1: The 10 target words Definition Lexicalized words Rivulet (n.) Repartee (n.) Jettison (v.) Convalesce (v.) Obdurate (adj.) Non-lexicalized words Zeitgeist (n.) Troubadour (n.) Manumit (v.) Bedizen (v.) Picaresque (adj.)

A very small stream Quick and usually amusing answers and remarks in conversation To get rid of something or someone that is not wanted or needed To rest in order to get better after an illness Extremely determined to act in a particular way and not to change despite argument or persuasion

The general set of ideas, beliefs, feelings, etc. which is typical of a particular period in history An itinerant composer and performer of songs (in medieval Europe) To release from slavery To dress or adorn gaudily Involving clever rogues or adventurers especially as in a type of fiction


They were evenly categorized into L and NL, with two nouns, two verbs, and one adjective of each type. Lexicalized words were defined as those that have a conventional Chinese translation equivalent or correspond to a fixed item in Chinese, based on various online dictionaries (http://dictionary.cambridge.org, http://www.merriam-webster.com, http://en.wiktionary.org/wiki/ Wiktionary:Main_Page, http://dictionary.babylon.com, http://www.onlinedict.com) and the judgements of several native speakers of Chinese. If a target word does not have a Chinese translation that is a fixed expression, it was defined as NL. All target words were tri-syllabic except for the NL zeitgeist, and an effort was made to avoid morphological clues to meaning and grammatical category. Ideally, relative difficulty of the words should have been controlled (as it was by Webb 2007), but the goal of ecological validity severely limited the possibilities for such control. The study required 10 English words, all unknown to all the participants. An inevitable consequence was that the words were uncommon and naturally tended to be ‘odd’ in some respects, especially the five not lexicalized in Mandarin—the unavoidable negative side of the tradeoff between control and ecological validity.


Reading materials

Instruments Prior to the experiment, participants were given a checklist of 100 potential target words, L and NL, and instructed to check (3) words they knew. A total of 14 words were found to be unknown to all participants; 10 were chosen as the target words. The checklist was given a month before the experiment, but in this EFL context the participants were very unlikely to encounter the target words during this time, because these words were quite uncommon and the intervening period was a school break. Each posttest measured receptive and productive knowledge of orthography, part of speech, and association. Knowledge of meaning and form was tested only receptively because there was no apparent way to test productive knowledge in this case. These tests were based on Webb’s (2007) 10 tests, though we made some adjustments (see below) and omitted his two tests of syntagmatic association (Webb’s ‘syntax’ tests) and his second, multiple-choice, measure of receptive knowledge of meaning and form. Before the immediate posttest, a reading comprehension test was given to verify that participants read and understood the passages. It included five multiple choice questions, based on what each group had read. The delayed posttest consisted of the same seven


Reading materials consisted of 13 passages, all stories, in two groups. Main readings included seven stories, composed by several advanced Chinese bilingual learners and edited by an English native speaker. Each included exactly one occurrence of each of the 10 target words, each context providing helpful clues for learners to infer their meanings. For example, in one story (see appendix available as online supplementary material), the word convalesce was introduced by reference to a ‘man who was convalescing in the hospital after he had been hurt in an automobile accident’. In the same reading, manumit was offered in the following context—‘Slave owners still don’t want to manumit their slaves. It’s because our society doesn’t accept the idea of freedom’. These cases are typical in providing strong clues to meaning but stopping well short of giving a definition. The remaining six passages served as distracter readings, the purpose being to have each group read the same number of passages despite the differences in number of exposures to target words. Distracter readings were selected from a general English as a Foreign Language (EFL) reading textbook (Sinclair et al. 2005) designed for beginning-level EFL learners and therefore easily understandable for the participants. Because all 13 readings were designed to avoid difficult vocabulary, we could reasonably assume that apart from the 10 target words, participants knew all or very nearly all the words in them. Each passage contained 250–300 words, so unknown words made up no more than 5% in any of them, in keeping with findings that learners must know at least 95% of the running words to infer the meaning of unknown words.


699

Table 2: Description and sequence of the seven dependent measures (based on Webb 2007) Knowledge measured

Test type

1 2 3 4 5 6 7

Productive knowledge of orthographic form (PO) Receptive knowledge of orthographic form (RO) Receptive knowledge of meaning and form (RMF) Productive knowledge of parts of speech (PP) Productive knowledge of associations (PA) Receptive knowledge of parts of speech (RP) Receptive knowledge of associations (RA)

Dictation Multiple choice Translation (L2–L1) Sentence construction Paradigmatic association Multiple choice Multiple choice

vocabulary tests used in the immediate posttest, without the reading comprehension test. The vocabulary tests had seven pages, each measuring one aspect of knowledge of the target words. The tests were sequenced following Webb’s (2007) example to avoid any possible learning effect. For example, productive knowledge of orthography was placed before its receptive counterpart because the latter included the correct form of each target word as one of the possible answers and thus provided information participants might use on the productive test. Answers for the non-objective tests were scored by two advanced English learners (after consultation with an English native speaker), with occasional discussion between raters on disagreements. Inter-rater reliability was a near perfect 0.9997, disagreements occurring only in two instances. Table 2 briefly describes the tests and shows the sequence in which they were given. Test 1. Productive knowledge of orthography (PO) The first measure, productive knowledge of orthographic form, used a dictation test in which participants heard each target word pronounced twice, from a recording, and then had 10 seconds to write it down. Any wrong spelling, even minor, resulted in the item being scored as incorrect, because partial success could be due to the phonological prompt rather than the treatment (Webb 2007). Test 2. Receptive knowledge of orthography (RO) This test used a four-option multiple choice format. Distracters were made to resemble the target item. The following examples are for the target words manumit and jettison: (a) manumit (a) jeterson

(b) munamit (b) jettison

(c) manemit (c) jetison

(d) manomet (d) jettision

Test 3. Receptive knowledge of meaning and form (RMF) Receptive knowledge of meaning and form was measured by asking participants to provide a Chinese translation for each target word. The answer was


Order


scored as correct only if it contained all the essential semantic features of the target word. Taking the word troubadour as an example, answers translating as ‘wandering musician’ were scored as correct while ‘musician’ was marked as incorrect, as was ‘wanderer’.

Test 5. Productive knowledge of associations (PA) On this test, participants were asked to write one word paradigmatically associated with each target word. For manumit, for example, release was accepted while sell was not. The directions included several examples intended to push participants toward paradigmatic associations. Thus, syntagmatic associations were uncommon and were marked incorrect (along as an answer for rivulet, for example). Test 6. Receptive knowledge of parts of speech (RP) The sixth test corresponded to Webb’s (2007) test of receptive knowledge of grammatical function. Each question presented three sentences as options, each containing the target word in a different part of speech, only one of them correct. Sentences were designed to avoid contextual clues to the word meaning. The following example is for obdurate: Obdurate

(a) It is an obdurate. (b) She obdurated. (c) It is so obdurate.

Test 7. Receptive knowledge of associations (RA) This last test again used a multiple-choice format. In each question, five words were presented as options beside each target word, and participants were required to choose the options paradigmatically related to the target word. This test was a modified version of Webb’s (2007) receptive test of associations, in which learners had to select only one correct option. In the present study, it was necessary to have more than one correct answer per question in order to determine whether learners had acquired the meanings of the NL words. A learner who mistakenly believed that troubadour corresponded to an L1 word meaning ‘singer’ or ‘musician’ could easily select an option that captured this aspect of its meaning but would miss the feature that distinguishes it from these mistaken translations. The inclusion of the option wandering allowed such misunderstandings to be detected. One point was


Test 4. Productive knowledge of parts of speech (PP) On this test participants were asked to use each target word to make a correct sentence. Responses were scored as correct if the word was used as the correct part of speech. Other errors were disregarded. Examples of correct responses are ‘Her father is very obdurate’ and ‘He became a troubadour when he was 15’. Incorrect answers included ‘I don’t like those who look bedizen’, and ‘He reparteed his song’. This test corresponded to Webb’s (2007) test of productive knowledge of grammatical functions. We have avoided the term ‘grammatical function’ because it is commonly used in linguistics to refer to the distinction between, for example, subjects and objects.


701

Table 3: The distribution of readings Group

E1 E3 E7

The distribution of reading materials

D1 M1 M1

D2 D2 M2

D3 D3 M3

D4 M4 M4

D5 D5 M5

D6 D6 M6

Target-word Exposures M7 M7 M7

1 3 7

M: main reading; D: distracter reading.

‘jettison’

(a) obtain (b) look for (c) abandon (d) steal

(e) buy

Directions for this test included an example with correct answers based on similar meaning, opposite meaning, and other relatedness (sudden-surprising), as well as two unrelated options.

Procedure The experiment included three phases—reading comprehension task, immediate posttest, and delayed posttest two weeks later. The 72 participants were randomly assigned to three experimental groups based on number of target-word exposures (E1, E3, E7) in the texts. Each group read seven passages, each appearing on a single page. The seven pages were stapled together and participants were told not to flip back to previous pages. For the E1 group the first six readings were distracters and only the final reading contained the target words. E3 read three main and four distracter passages, and E7 read all seven main readings. For E1 and E3 a main reading was placed in the final position so that at the time of the testing the three groups would not differ in how recently they had seen the target words. The other two main readings for E3 were placed in the first and fourth positions. The distribution of readings for each group is shown in Table 3. Participants were told beforehand only that there would be a reading comprehension test, to encourage them to attend to the readings and not to pay extra attention to the target words. Immediately after the reading task, all three groups were given this test and the first posttest, consisting of the seven vocabulary tests described above. They were allowed as much time as needed to finish the questions. Each test was presented on a separate page and participants were not allowed to turn back to a previous page. A delayed posttest was administered two weeks later, using the same vocabulary tests as the immediate posttest, but without any readings.


awarded for each correct answer and scores were converted to percentages for statistical analysis. The example of jettison is given below, the correct answers being (a), based on opposite meaning, and (c), based on similar meaning.


Pilot study

Data analysis To answer the first research question, regarding number of exposures, scores of the three groups on the seven vocabulary tests were submitted to one-way analysis of variance (ANOVA) tests, or to Kruskal–Wallis tests when results did not satisfy the normality requirement for ANOVA. Post hoc Tukey and least significant difference (LSD) tests were performed to locate the site of significant effects for each vocabulary test. Effect sizes (Cohen’s d) were also calculated for each of the contrasts. Regarding the second research question, whether NL words would create additional difficulty and whether this factor would interact with repetition, a comparison was made between gains on L and NL words, using repeated-measures ANOVA for each test of semantic knowledge (RMF, PA, RA). For each research question, the analyses were repeated for results from the delayed posttest.

RESULTS AND DISCUSSION On the five reading comprehension questions, the groups’ average correct answers ranged from 4.75 to 4.88, and no participant scored below 4, indicating that the reading materials were comprehensible to all the learners and that they focused on the content of the stories. These are by definition the conditions for incidental learning, so we conclude that any vocabulary learning that occurred was indeed incidental.

Effects of repetition: Immediate For Research Question 1, the independent variable was number of exposures (1, 3, 7). The dependent variables were the seven aspects of word knowledge. Prior to quantitative analyses, the results for four of the tests (PO, RMF, PA, and RP) were found not to be normally distributed, thus failing to meet the assumption of the one-way ANOVA. Hence, these results were subjected to the non-parametric Kruskal–Wallis test (p = .05), while those for the remaining


Prior to the experiment, a pilot study was conducted with nine learners, all comparable with the participants in the main study. The goals were to test whether all participants could understand the texts and would attend to them, to determine the amount of time to be allotted to the reading task, and to ensure the feasibility of the measurements. It was determined that 120 seconds per reading would be sufficient for all the participants to complete the reading but not so much as to allow them to use mnemonic strategies to remember the target words (Webb 2007), as some might given that these were probably the only words unknown to them in the readings. Based on the pilot study, small adjustments were made in some tests, apart from which no need was found for changes in the instruments or materials.


703

Table 4: Means and standard deviations on the immediate posttest Group n

E1 24

E3 24

E7 24

M

SD

M

SD

M

SD

RO PO RP PP RMF RA PA

4.29 1.08 4.79 1.46 0.13 3.97 0.62

1.97 0.72 2.08 1.91 0.45 1.34 1.17

6.25 2.12 5.54 2.54 0.58 3.88 1.21

1.87 1.12 1.35 1.67 0.93 1.4 1.06

6.5 2.71 6.58 3.92 1.5 5.09 3.17

2 1.60 1.79 2.67 1.35 1.28 2.24

Maximum score = 10.

Table 5: Results of the post-hoc Tukey HSD tests for the immediate posttest Test/group

E1 vs. E3

E3 vs. E7

E1 vs. E7

RO PP RA

0.008* NS NS

NS NS 0.007*

0.000* 0.000* 0.014*

*p < .05.

three tests (RO, RA, and PP) were analyzed by one-way ANOVA (p = .05). For the location of significant effects in each case, a post hoc LSD test was performed with the four non-parametric cases and a post hoc Tukey HSD for the three parametric cases. Table 4 shows means and standard deviations for the three groups on each vocabulary test of the immediate posttest. Scores consistently increased with number of exposures, the only exception being the slight drop on RA from E1 (3.97) to E3 (3.88). The ANOVA and Kruskal–Wallis tests revealed significant effects on each of the seven measures. Results of the post hoc Tukey and LSD tests are shown in Tables 5 and 6. Significant differences were found for all E1–E7 comparisons. E1 and E3 differed significantly on the two orthography tests, E3 and E7 on the three semantic tests. This pattern of results offers a relatively clear picture of the effect of exposures on the different types of vocabulary knowledge. For orthographic knowledge, the benefits come with the first few exposures. For semantic knowledge


Test


Table 6: Results of the post hoc LSD tests for the immediate posttest Test/group

E1 vs. E3

E3 vs. E7

E1 vs. E7

PO RP RMF PA

* NS NS NS

NS NS * *

* * * *

*p < .05.

E1–E3 E3–E7 E1–E7

RO

PO

RP

PP

RMF

1.021 0.129 1.113

1.130 0.434 1.405

0.437 0.662 0.925

0.603 0.636 1.074

0.652 0.807 1.522

RA 0.066 0.903 0.855

PA

Average

0.529 1.188 1.496

0.615 0.680 1.199

more exposures are required for notable progress, gains coming mainly between three and seven. Part of speech shows a more smooth development. Some complications will be added to this picture below. Additional insights can be obtained by looking at the results in terms of effect sizes (see Norris and Ortega 2000; Lipsey and Wilson 2001), which directly measure the effect that the independent variable had on the dependent variable. The effect size measure used here is Cohen’s d, which is the number of standard deviations by which one group’s mean is greater than another’s. The numbers are interpreted as follows—0.20–0.50 = small effect; 0.50–0.80 = medium effect; >0.80 = large effect (Cohen 1992). Table 7 gives the effect sizes for each comparison between groups on the immediate posttest. For the overall contrast, E1–E7, all effect sizes were large, showing that repetition does in general have a large, positive impact on learning. Productive tests yielded somewhat higher values on each knowledge type (but see the discussion of the delayed posttest below). Note that this does not imply the counterintuitive conclusion that productive knowledge was acquired more quickly or successfully than receptive, but rather that repetition had a greater effect on the former than on the latter, matching Webb’s (2007) finding that scores on productive tests correlated more strongly with number of repetitions than did scores on receptive tests (but, again, see below). The size of the effect varied not only with the productive/receptive distinction but also with knowledge type—orthographic, part of speech, and semantic. For orthographic knowledge, the difference between one exposure and three yielded large effects, while the addition of four more exposures made only a small contribution. The apparent explanation is that three exposures


Table 7: Effect sizes for the immediate posttest


705

Effects of repetition on retention after two weeks Two weeks after the treatment, the same seven tests were given, without any additional target-word exposure, to investigate retention of the previously acquired word knowledge. Only 57 of the original 72 participants were available, an unavoidable attrition given that the experiment was conducted outside of class and participation was entirely voluntary. We know of no reason to think the 15 missing students differed from the others in any way other than the times at which they were available. Data on RO, PO, RP, PP, and RA met the assumption of the one-way ANOVA test, data normality, whereas the data for RMF and PA were not normally distributed, dictating use of the non-parametric Kruskal–Wallis test.


were often sufficient for words that were relatively easy to spell (repartee, manumit) and that seven were not enough for the more difficult words (convalesce, zeitgeist). The pattern was quite different for part of speech. The overall effect (E1–E7) was again large, but the two component comparisons (E1–E3 and E3–E7) contributed about equally to this effect, presenting a picture of gradual, cumulative gains with increasing exposure. Each of the effect sizes for semantic knowledge (RMF, RA, and PA) was also large in the overall (E1–E7) comparison, with the E3–E7 contrast making the larger contribution (for RA, the only contribution). Thus, a relatively large number of exposures appear to be especially important for acquisition of meaning. Our overall findings match Webb’s (2007) reasonably well. Both studies showed, on each test, a pattern of rising scores as the number of exposures increased. Both found that learning of meaning requires more exposures than are needed for other knowledge types. Contrasts appear on some specific points, most of which are readily attributable to the differences in target words and contexts. On the orthography tests, each of Webb’s groups scored much better than ours, as expected because his words were especially easy to spell and some of ours were quite difficult. A similar but somewhat weaker pattern appeared on the semantic tests, with a small variation on productive association, for which Webb’s findings showed the largest contrast from E1 to E3 while for ours it occurred between E3 and E7. This again can be readily explained by the need for a greater number of exposures for learning more difficult words. The two part-of-speech tests taken together did not yield any clear contrast with Webb’s (grammatical function) results. This non-contrast is not surprising, as the relevant information for learners was, in both studies, the positions in which the word appeared in a sentence and the affixes it included, which did not systematically differ in the two studies. Altogether, our findings suggest that Webb’s results do accurately portray the relation between repetition and learning, while suggesting that they overestimate learners’ success in orthography and semantics in more authentic reading situations.


Table 8: Means of learning conditions on the delayed posttest Group n

E1 19

E3 19

E7 19

M

SD

M

SD

M

SD

RO PO RP PP RMF PA RA

4.68 1.11 4.05 1.21 0.11 0.32 3.14

1.734 0.737 1.747 1.903 0.315 0.671 0.867

5.89 1.74 5.89 2.05 0.37 0.73 4.15

1.524 1.261 1.822 1.81 0.761 1.046 0.953

5.63 1.53 5.89 2.84 0.95 2.05 5.07

1.64 0.841 1.853 2.192 0.78 1.393 0.974

Maximum score = 10.

Table 9: Location of significant effects through the post-hoc Tukey HSD test Tests

E1 vs. E3

E3 vs. E7

E1 vs. E7

RO PO RP PP RA

NS (0.067) NS (0.108) 0.008* NS (0.394) 0.004*

NS (0.873) NS (0.772) NS (0.621) NS (0.440) 0.010*

NS (0.184) NS (0.362) 0.012* 0.036* 0.000*

*p < .05.

Table 8 gives descriptive statistics for the delayed posttest. Most of the means show a decline from the immediate posttest (Table 4), with some small exceptions, especially in orthography, which can be explained at least in part by a practice effect. In almost all cases, scores still showed a tendency to rise with increasing word encounters (E1 < E3 < E7). The exceptions were small drops from E3 to E7 on orthography and no change on RP from E3 to E7. The ANOVA and Kruskal–Wallis tests revealed that the results of RP, PP, RMF, PA, and RA reached significance, only the orthography results falling short. Not surprisingly, the same pattern was found for the E1–E7 comparison. Tables 9 and 10 present the location of significant effects. E3 learners’ scores were significantly higher than E1’s on the tests of RP and RA. Significant differences between E3 and E7 were identical to those on the immediate posttest, appearing in the knowledge of RMF, RA, and PA, that is, the semantic tests.


Test


707

Table 10: Location of significant effects through the post hoc LSD test Tests

E1 vs. E3

E3 vs. E7

E1 vs. E7

RMF PA

NS NS

* *

* *

*p < .05.

Table 11: Effect sizes for the delayed posttest RO 0.743 0.164 0.563

0.631 0.200 0.532

RP

PP

RMF

RA

PA

Average

1.031 0 1.022

0.452 0.395 0.796

0.483 0.753 1.534

1.110 0.955 2.097

0.478 1.082 1.676

0.704 0.403 1.174

Effect sizes also reveal important information from the delayed posttest. The numbers are presented in Table 11. The average effect size for the delayed posttest (1.174) was only slightly below that of the immediate posttest (1.199), indicating that the effect of repetition on word knowledge is lasting. With the exception of orthography, all effect sizes were large, and those for the semantic tests were very large. Recall that on the immediate posttest, productive tests yielded somewhat higher effects on each knowledge type. On the delayed posttest this relation largely reversed. In other words, repetition had larger initial benefits for productive knowledge but larger long-term benefits for receptive. Thus, Webb’s (2007) finding that productive scores correlated more highly with number of repetitions than did receptive scores may be the result of his reliance on an immediate posttest without follow-up testing. The reversal found after two weeks might be explained as reflecting the greater vulnerability of productive knowledge to forgetting (e.g. Rott 1999). Perhaps the most interesting feature of Table 11 is the way that the size of the effect varies with knowledge type, the three types each clustering in a neat way. For the three semantic tests, the smallest effect size is 1.534, far higher than that for any of the other four tests. At the other extreme, the two orthography tests showed only modest, and nearly identical values, clearly below all the other five tests. The two part of speech measures fell, together, between orthography and semantics. The unimpressive results for the orthography tests suggest that while repetition has strong immediate effects, these effects are not very durable, at least not for the number of repetitions examined here and the difficulty of the target words. The part of speech measures, in contrast, yielded large effect sizes on each posttest, the effect of the delay period being primarily to reverse the position of productive knowledge (higher on the immediate


E1–E3 E3–E7 E1–E7

PO


Table 12: Results for L and NL words on the immediate posttest Test

RMF RA PA

L

NL

M

SD

M

SD

0.46 2.22 0.85

0.82 0.84 1.11

0.28 2.11 0.82

0.48 0.69 1.03

Significance

Effect size

0.035* NS NS

0.278 0.144 0.028

*p < .05; maximum score = 10.

Effects of L1 lexicalization: Immediate The second issue explored in this study was the effect of L1 lexicalization on learning and retention—Do learners have special difficulty learning the meanings of words that are not lexicalized in their L1? Table 12 compares the results for the five L and five NL words. On the RMF test, a significant difference was found, but it was quite small (effect size = 0.278). The association tests yielded much smaller, non-significant results. The apparent implication is that lexicalization was unimportant. However, a look at the results for individual words produces a very different conclusion. Results for RMF are given in Table 13. As can be seen, learning of NL words occurred almost entirely with one word, troubadour. If this one word is removed, the contrast between the two sets is quite striking; it would be little exaggeration to say that no learning occurred with the NL words. An explanation for the exceptional status of troubadour readily suggests itself. The seventh main reading, given to all three groups, was a story about the adventures of a man who was explicitly labeled a troubadour early in the story. Thus, the concept that the word represents was central to the passage. Its role


posttest) and receptive knowledge (higher on the delayed). Productive knowledge showed the same pattern found on the immediate posttest, the E1–E3 and E3–E7 contrasts contributing about equally to the overall effect. For RP, though, the large overall effect came entirely from E1 to E3. Turning to semantic knowledge, for RMF the effect of repetition was very strong and showed essentially no change from immediate (1.522) to delayed posttest (1.534). Results for PA were similar—1.496 on the immediate posttest and 1.676 on the delayed. Results for RA were also strong in both cases but showed an unanticipated contrast between the two tests, the effect size rising from .855 to 2.097 because E7 suffered no decline from immediate to delayed posttest. Overall, the findings indicate that repetition has a very strong and durable effect on the acquisition of word meaning. Three exposures are clearly better than one, and seven are clearly better than three.


709

Table 13: Correct answers for L and NL words on RMF test Group

E3

1

2 2 2 1 2

1

E7

5 2 7 1 7 1

1 5

1 12

in some of the other readings was also quite substantial; no other target word had such a prominent role. What this suggests is that successful learning is partly determined by the importance of the word in the reading and that this contextual variable can even outweigh a word property as influential as lexicalization. This idea of the importance of the word in the text might be connected to saliency (see Brown 1993). Without this additional variable, the contrast between L and NL words is clear, providing a positive answer to the second research question. This clear positive result contrasts with the relatively limited differences in learning that Paribakht (2005) found. But this lack of contrast is expected, because only minimal learning occurred in her study, for both L and NL words, and it was mainly in participants’ familiarity with the form of the words. Regarding the relation between repetition and lexicalization, little can be said, as essentially no learning of NL words occurred among any of the groups (when troubadour is removed from consideration), eliminating the possibility of an interaction.

Effects of lexicalization on retention after two weeks The second issue regarding lexicalization is whether any contrasts found on the immediate posttest would remain on the delayed posttest two weeks later. As can be seen in Table 14, no significant differences were found on any of the tests, and all three effect sizes were negligible, apparently indicating that lexicalization had no lasting effect. But a look at the individual words yields the same conclusion reached for the immediate posttest. For L words in general, a substantial amount of retention occurred, but gains for NL words were limited almost entirely to troubadour. Apart from this one word, essentially no retention was found for NL


L words Convalesce Jettison Repartee Rivulet Obdurate NL words Manumit Zeitgeist Bedizen Troubadour Picaresque

E1


Table 14: Results for L and NL words on the delayed posttest Test

RMF RA PA

L

NL

Significance

M

SD

M

SD

0.26 2.16 0.47

0.52 0.67 0.93

0.21 2.04 0.56

0.41 0.71 0.71

NS NS NS

Effect size

0.108 0.174 0.110

Maximum score = 10.

CONCLUSION This study investigated how word repetition in reading affects EFL learners’ vocabulary acquisition and whether L2 words that are not lexicalized in learners’ L1 cause special difficulty. The overall findings fit well with Webb’s (2007) findings for repetition, though we conclude that in regard to orthographic and semantic knowledge his methodology resulted in an overestimate of the learning that occurs in more natural situations. Findings for lexicalization supported the hypothesis that NL words are especially difficult (Paribakht 2005). For repetition, the main finding was that increasing exposures does help, consistent with the common understanding that vocabulary learning is a gradual process (Nagy et al. 1985; Schmitt 2000), and that the effect remains for at least two weeks. Immediately after the treatment, productive knowledge was found to benefit from repetition somewhat more than receptive knowledge, a relation that had reversed by the time of the delayed posttest, perhaps due to the greater vulnerability of productive skills. Interesting contrasts were found in the pattern of development for each knowledge type. Orthographic knowledge benefited greatly from three exposures but not at all from an additional four, reflecting the presence of both relatively easy words (learnable in three exposures) and difficult words (not learnable even from seven exposures). Part of speech showed a more steady growth with exposure. For semantic knowledge, most gains occurred between three and seven exposures. The overall strength of the effect of repetition also varied with knowledge types. For orthography, it was strong immediately after the treatment but greatly reduced two weeks later. For part of speech, it was strong both immediately and after the delay period, to similar degrees. The effect on semantic knowledge was strong immediately and even stronger two weeks later. For the second research question, we found that L2 words not lexicalized in learners’ L1 do cause learning difficulty, both immediately and after a


words. We conclude again that such words present special difficulties for incidental learning, difficulties which might be overcome by a strong focus in the readings on a particular word.


711


two-week delay. This study thus provides empirical evidence for Paribakht’s (2005) conclusions about the difficulty of inferring the meanings of NL words. The study also suggests that increasing the number of exposures up to seven makes little contribution to the acquisition of meaning for NL words, because these words are too difficult to learn from even seven exposures. Again, Webb’s (2007) use of invented words may underestimate vocabulary learning difficulties, as all of his invented words precisely corresponded to already-known words, ruling out the kinds of difficulties that arose with the NL words in this study. An additional contrast with Webb’s (2007) findings suggests the same conclusion. Gains did not increase as rapidly with number of exposures as in Webb’s study, especially for semantic knowledge and orthographic knowledge of difficult words. This difference is most readily attributable to the use of real words and the way in which reading materials were presented. The use of invented words in isolated sentences may thus overestimate what learners gain under more authentic learning conditions. More generally, our findings reinforce the point that repetition is crucial but there cannot be any simple answer to the question of how many exposures are needed because many factors interact with repetition. Based on our results, these include knowledge type, importance/saliency in the text, and the character of the particular words being learned, including difficulty in various respects and, most notably, L1 lexicalization. The latter suggests that the particular L1-L2 pair involved can be of fundamental importance. Regarding pedagogical implications, the obvious conclusion from this study and others is that teachers should provide learners with opportunities to encounter a given word repeatedly in reading. A likely approach is the narrow reading recommended by Krashen (1981) and Schmitt and Carter (2000), in which a number of readings in a single topic area give learners repeated exposure to particular words. Most aspects of word knowledge, for most words, appear to develop successfully through such exposure. This is most clearly the case for part of speech, for which we find no evidence of a need for intervention, as it showed gradual, substantial improvement with increasing exposures. But our findings suggest other cases in which intervention may be appropriate. In regard to meaning, NL words like zeitgeist or manumit may require special treatment (Paribakht 2005), such as explicit instruction, possibly using direct contrasts with L1 words, or marginal glosses and/or dictionary use (Hulstijn et al. 1996). A possible alternative for an important NL word is to make it the theme of a short reading passage, like troubadour in the present study. An intriguing possibility is for publishers to develop readers with NL words in mind, focusing such words in the way that troubadour was focused here while giving L words more of a background role. We also found that spelling did not develop well with the more difficult words in our study, suggesting that while in most cases it can be successfully acquired through incidental exposure, instruction might be appropriate for difficult words like zeitgeist or convalesce. Such instruction might involve simply drawing learners’



ACKNOWLEDGEMENTS The authors wish to thank Iru Su and Hao-Jan Chen for their valuable comments, Chih-chun Tsai for help with the statistical analysis, and of course all the students who participated in the study. Conflict of interest statement. None declared.

REFERENCES Blum, S. and E. Levenston. 1979. ‘Lexical simplification in second-language acquisition,’ Studies in Second Language Acquisition 2: 43–63. Brown, C. 1993. ‘Factors affecting the acquisition of vocabulary: frequency and saliency of words’ in T. Huckin, M. Haynes, and J. Coady (eds): Second Language Reading and Vocabulary Learning. Ablex, pp. 263–86. Cohen, J. 1992. ‘A power primer,’ Psychological Bulletin 112: 155–9. Hulstijn, J. H., M. Hollander, and T. Greidanus. 1996. ‘Incidental vocabulary learning by advanced foreign language students: the influence of marginal glosses,

dictionary use, and reoccurrence of unknown words,’ Modern Language Journal 80: 327–39. Jenkins, J. R., M. L. Stein, and K. Wysocki. 1984. ‘Learning vocabulary through reading,’ American Educational Research Journal 21: 767–87. Jiang, N. 2000. ‘Lexical representation and development in a second language,’ Applied Linguistics 21: 47–77. Krashen, S. 1981. ‘A case for narrow reading,’ TESOL Newsletter 15: 23. Krashen, S. 1989. ‘We acquire vocabulary and spelling by reading: additional evidence for the


attention to the spelling. Alternatively, it could consist of lessons on word structure, output practice, or any of a variety of other exercises. There are no guarantees, however, that intervention will produce the desired results, and it is quite possible that with a larger number of exposures all aspects of word knowledge will develop as well as, or better than, they would as a result of pedagogical intervention. The limitations of the study include the relative lack of control over the characteristics of the individual target words and the contexts for each, particularly the informativeness of the contexts (see Webb 2008). The varying difficulty of the words was an issue especially in spelling. For effects of context, the clearest example is the finding that troubadour stood out from all the other NL words (and even from L words), presumably because of its importance in the stories. One could also argue that the contexts were too informative to reflect the typical conditions readers face, though they might provide a reasonably good match in this respect with materials designed for learners. Another limitation is the subjectivity in distinguishing L from NL words. Future research could benefit from the development of a more reliable means of making the distinction. Finally, the small number of target words represents another limitation, which might be improved upon in future research.


Schmitt, N. 2000. Vocabulary in Language Teaching. Cambridge University Press. Schmitt, N. and R. Carter. 2000. ‘The lexical advantages of narrow reading for second language learners,’ TESOL Journal 9: 4–9. Shu, H., R. C. Anderson, and H. Zhang. 1995. ‘Incidental learning of word meanings while reading: a Chinese and American cross-cultural study,’ Reading Research Quarterly 30: 76–95. Sinclair, F. M., F. Yasar, and R. Ayden. 2005. Get the Point 2. Caves. Sjo¨holm, K. 1998. ‘A reappraisal of the role of cross-linguistic and environmental factors in lexical L2 acquisition’ in K. Haastrup and A. Viberg (eds): Perspectives on Lexical Acquisition in a Second Language. Lund University Press, pp. 135–47. Swan, M. 1997. ‘The influence of the mother tongue on second language vocabulary acquisition and use’ in N. Schmitt and M. McCarthy (eds): Vocabulary: Description, Acquisition and Pedagogy. Cambridge University Press, pp. 156–80. Tekmen, E. A. F. and A. Daloglu. 2006. ‘An investigation of incidental vocabulary acquisition in relation to learner proficiency level and word frequency,’ Foreign Language Annals 39: 220–43. Webb, S. 2007. ‘The effects of repetition on vocabulary knowledge,’ Applied Linguistics 28: 46–65. Webb, S. 2008. ‘The effects of context on incidental vocabulary learning,’ Reading in a Foreign Language 20: 232–45. Yu, L. 1996. ‘The role of L1 in the acquisition of motion verb in English by Chinese and Japanese learners,’ Canadian Modern Language Review 53: 190–218. Zahar, R., T. Cobb, and N. Spada. 2001. ‘Acquiring vocabulary through reading: effects of frequency and contextual richness,’ Canadian Modern Language Review 57: 541–72.


input hypothesis,’ Modern Language Journal 73: 440–63. Lipsey, M. W. and D. B. Wilson. 2001. Practical Meta-analysis. Sage. Nagy, W. 1997. ‘On the role of context in firstand second-language vocabulary learning’ in N. Schmitt and M. McCarthy (eds): Vocabulary: Description, Acquisition and Pedagogy. Cambridge University Press, pp. 64–83. Nagy, W. E., R. C. Anderson, and P. A. Herman. 1987. ‘Learning word meanings from context during normal reading,’ American Educational Research Journal 24: 237–70. Nagy, W. E., P. A. Herman, and R. C. Anderson. 1985. ‘Learning words from context,’ Reading Research Quarterly 20: 233–53. Nation, I. S. P. 1982. ‘Beginning to learn foreign vocabulary: a review of the research,’ RELC Journal 13: 14–36. Nation, I. S. P. 2001. Learning Vocabulary in Another Language. Cambridge University Press. Norris, J. M. and L. Ortega. 2000. ‘Effectiveness of L2 instruction: a research synthesis and quantitative meta-analysis,’ Language Learning 50: 417–528. Paribakht, T. S. 2005. ‘The influence of first language lexicalization on second language lexical inferencing: a study of Farsi-speaking learners of English as a foreign language,’ Language Learning 55: 701–48. Paribakht, T. S. and M. Wesche. 1999. ‘Reading and incidental L2 vocabulary acquisition: an introspective study of lexical inferencing,’ Studies in Second Language Acquisition 21: 195–224. Rott, S. 1999. ‘The effect of exposure frequency on intermediate language learners’ incidental vocabulary acquisition through reading,’ Studies in Second language Acquisition 21: 589–619. Saragi, T., I. S. P. Nation, and G. F. Meister. 1978. ‘Vocabulary learning and reading,’ System 6: 72–8.

713

Applied Linguistics: 31/5: 714–733 ß Oxford University Press 2010 doi:10.1093/applin/amq034 Advance Access published on 6 October 2010

The Branding of English and The Culture of the New Capitalism: Representations of the World of Work in English Language Textbooks JOHN GRAY

Coinciding with the global boom in commercial English language teaching is the development of a sizeable publishing industry in which UK-produced textbooks for the teaching of English as an international or foreign language are core products. This article takes the view that these ‘curriculum artefacts’ can also be understood as ‘cultural artefacts’ in which English is made to mean in highly selective ways. The article focuses specifically on representations of the world of work in textbooks from the late 1970s until the present and shows how they have drawn consistently on evolving discourses of the new capitalism. It argues that students are repeatedly interpellated in these materials to the subject position of white-collar individualism in which the world of work is overwhelmingly seen as a privileged means for the full and intense realization of the self along lines determined largely by personal choice. The article concludes by suggesting that such materials have increasingly constructed English as a branded commodity along lines which are entirely congruent with the values and practices of the new capitalism.

INTRODUCTION The global explosion of commercial English language teaching (hereafter ELT) is largely coterminous with the arrival of the so-called ‘new capitalism’ (Gee et al. 1996; Fairclough 2002; Sennett 2006) ushered in during the Thatcher and Reagan periods. Central to the exponential rise in commercial ELT is the development of a sizeable and financially lucrative publishing industry in which textbooks aimed at the global market are core products. As has been argued elsewhere (van Dijk 2008), such artefacts can be seen not only as mediating tools of subject knowledge, but also as organs for the ideological reproduction and legitimation of ‘particular constructions of reality’ (Apple and ChristianSmith 1991: 3). In 1990, as the countries of the ex-Soviet Union prepared themselves for a ‘shock therapy’ approach to the introduction of market economies (Klein 2007; Stuckler et al. 2009; Zˇizˇek 2009), The Economist magazine’s Intelligence Unit published a detailed report on the state of ELT throughout the region (McCallan 1990). The report noted with approval that the state


TESOL/Applied Linguistics, University of East London E-mail: [email protected]

J. GRAY

715

TEXTBOOK ANALYSIS My approach to the study of textbooks has its roots in cultural studies and in particular the work of Du Gay et al. (1997) and their seminal study of the construction and circulation of meanings associated with the Sony Walkman. Du Gay et al. (1997) argue that culture is an endlessly recursive process of meaning making and meaning taking. Their theoretical model for conducting a cultural study known as the ‘circuit of culture’ (ibid.: 3) analyses five key moments in the life of a cultural artefact: representation; identity; production; regulation; and consumption. From this perspective, textbooks can be seen not simply as ‘curriculum artefacts’ (Apple and Christian-Smith 1991: 4), but also as cultural artefacts or ‘communicative acts’ (Singapore Wala 2003a) which serve to make English mean in particular ways—although the meanings which are inscribed on the page cannot ultimately be guaranteed. As Hall (1980) has argued, the dominant-hegemonic position in which the reader accepts the preferred meaning is only one of several reading positions which are possible. And indeed research confirms that oppositional readings do occur and that teachers and students can seek to subvert specific ideological content (Canagarajah 1999; Gray 2010). However, in this article my focus is on the inscription of meanings rather than on the ways in which such meanings may be challenged in the classroom. Here I am concerned primarily with identifying what Du Gay et al. (1997) refer to as the ‘representational repertoires’ related to the world of work—namely the stock of ideas, images, and linguistic choices which are deployed in the creation of meanings, and the identifications that these seek to create in readers. The rationale for this choice is that work is a recurrent theme in ELT textbooks aimed at the global market (Gray 2010), and also because the specific discourse of new capitalism (see below) represents the


school sector in most countries was ill-prepared to meet the anticipated demand for English and urged the rapid development of private sector provision. In this way, it was suggested the state sector would find it difficult to catch up, while at the same time British ELT would find itself well-placed to promote further opportunities for British trade. In paving the ground for the development of markets favourable to the UK, the timely provision of ELT textbooks was identified as a strategic initial move (Gray 2002). Such a perspective, underlining the way in which ELT is imbricated in the extension of a globalized capitalist economy, resonates with analyses both within applied linguistics (Phillipson 1992; Pennycook 1994) and beyond. Thus Ives (2006: 136–7), writing from the perspective of political science, has suggested that the teaching of English in the global private sector is a ‘crucial element of an international business class structure’ which ‘facilitates the growth and spread of multinational corporations and trade’. Although Ives does not mention textbooks, the specific role that the content of such mediating tools might play in this process is one that merits attention.

716 BRANDING OF ENGLISH AND THE CULTURE OF THE NEW CAPITALISM

world of work along ideological lines which have been transformed since the 1980s (Boltanski and Chiapello 2007).

THE GLOBAL COURSEBOOK


The textbooks analysed here are all examples of ‘global coursebooks’—that is, they each form part of an incremental general English course aimed at the global market. These commodities, generally produced by prestigious UK academic publishers and frequently marketed aggressively to compete with locally produced materials (Thomas 1999), are increasingly accompanied by a range of expensive technological supplements (Masuhara et al. 2008). Consumed primarily by the global commercial sector, they are also used in colleges, university language centres and in some state schools. In general, they differ from materials produced to meet the curriculum requirements of state education or many of those designed to enable students to develop a particular skill such as reading or writing. Indeed textbooks such as these may deploy very different representational repertoires—thus the former are frequently more educational in orientation (e.g. the Ministry of Education requirement of a culture syllabus in textbooks produced for the Italian state school market). The latter, on the other hand, can be more serious in tone and aim to produce greater engagement with specific topics as a way of developing a particular skill [e.g. Wallace’s (2004) reading and writing textbook which contains readings drawn from sources such as The British Journal of Psychology]. Given the need to maximize sales in the greatest number of markets, the thematic content of global courses is highly regulated by ELT publishers, whether through guidelines for authors which may list supposedly controversial topics to avoid (e.g. politics, religion, sex) or through more informal editorial advice (see Gray 2010 for detailed discussion). One of the effects of this regulation is that generations of successful ELT global courses rely on a narrow range of topics and tend to resemble each other—although, as we shall see, there are exceptions [cf. Singapore Wala’s (2003b) case study of commercial considerations in a local market]. Not surprisingly, many of these materials have been criticized for their pedagogical shortcomings and the highly selective representational repertoires they deploy in the construction of mainly benign versions of a globalizing world (Wajnryb 1996; Tomlinson et al. 2001; Tomlinson 2003; Masuhara et al. 2008; Gray 2010).1 At the same time, Wajnryb (1996) and Leung (2005: 137) have also been critical of their representation of language, with the latter arguing that their ‘reductionist and static idealizations . . . are at best partial representations of social reality’—a feature which he suggests stems from the absence of any serious ethnographic orientation to language-in-use in textbook production and which Wajnryb (ibid.) suggests ‘effectively turns language into a manageable, indeed a marketable product . . . more like a discrete item on a shop-shelf’. An appropriate (if pedagogically questionable) strategy, one might conclude, for the commodification of English as a global lingua franca. Before addressing

J. GRAY

717

the representational repertoires related specifically to the world of work, I now consider briefly some key features of the new capitalism.

THE NEW CAPITALISM


Following Doogan (2009) I take the view that what is ‘new’ about contemporary capitalism is the extent to which it is characterized by the ideology and the practices of neoliberalism. Neoliberalism refers to the political and philosophical ideas originating in the work of Friedrich von Hayek who argued that an unfettered market economy was the only means of preserving ‘a free political order’ and that ‘the whole conception of social or distributive justice’ (Hayek 1978: 110) being pursued by many post-war social democratic European governments was the enemy of this version of freedom—what future generations of media commentators in the UK would refer to as the ‘nanny state’. According to the neoliberal view, the role of government is primarily to guarantee and extend the reach of the market. Hence the deregulation of financial markets; the privatization of state assets; and the marketization of areas of life which were previously seen as the preserve of the state (to list but a few of the interventions associated with neoliberal government). The enthusiastic uptake of these ideas in Britain and the US entailed a reconfiguration of the public sphere with implications both for language and the world of work (discussed by Fairclough (2006) under the heading of ‘globalism’). Thus we see the rise in ‘customer care’ culture, the accompanying communication skills training and the imposition of regimes of scripted and stylized talk in the work place (Cameron 2000; Bourdieu and Wacquant 2001). More recently Holborow (2007) has referred to the re-semanticization of the lexical field of business to include more and more areas of life, such that in modern Ireland even asylum seekers are positioned as customers of the Irish state in official documentation. As Cameron (2000) has pointed out, such changes fulfill the important ideological function of establishing the market as the model for all interaction in the public sphere. At the same time, with regard to foreign language learning, Heller (2003) argues that the ‘new economy’ (Castells 2000) results in the commodification of language, as languages are learnt increasingly in terms of their perceived usefulness for employment (see also Block 2008; Tan and Rubdy 2008). One of the many consequences of these changes is that the world of work becomes both highly insecure and stressful (Bourdieu 1998; Sennett 2006; Stuckler et al. 2009). Castells (2000: 12) also sees the emergence of two kinds of labour in this economy—what he calls ‘self-programmable’ and ‘generic’ labour. The former is described as labour which is ‘equipped with the ability to retrain itself, and adapt to new tasks, new processes and new sources of information, as technology, demand and management speed up their rate of change’ (ibid.: 12). Such labour is referred to by Bauman (2007: 9) as having ‘zero drag’, a term which originated in Silicon Valley in the late 1990s and refers positively to the kind of employee who is able to switch jobs and


countries with a minimum of effort. Generic labour, on the other hand ‘is exchangeable and disposable, and co-exists in the same circuits with machines and with unskilled labour from around the world’ (ibid.: 12). It is those in this second group who are most disadvantaged in the neoliberal economy. While this state of affairs has provoked varieties of critique (Hertz 2002; Harvey 2005; Fairclough 2006; Doogan 2009), others have seen and celebrated a brave new world of risk and opportunity. Nowhere is this more apparent than in the self-help literature aimed at white-collar workers, which argues that the reconfiguration of the world of work requires nothing short of a reconfiguration of the self—and it is in this respect that I will argue that parallels with ELT textbooks become most apparent.

With regard to the reconfiguration of the self in the new capitalism, Klein (2000), Cameron (2000) and Lury (2004) all refer to the work of the self-styled marketing guru Tom Peters, who argues that the way for individuals to survive in a neoliberal climate is effectively to brand themselves in order to stand out from the growing army of generic labour. In the same way that commodities are branded to give them a distinct market identity, it is suggested that individuals too need to image and promote themselves along similar lines (cf. Bauman 2007). For those with sufficient wherewithal, the challenge of the new capitalism is enthusiastically presented as ‘a matchless opportunity for liberation’ (Peters 2008: x). The blurb on Peters’ (2008) self-help manual ‘The Brand You 50: Fifty Ways to Transform Yourself from an ‘‘Employee’’ into a Brand that Shouts Distinction, Commitment, and Passion!’ addresses the reader in a hectoring style which sets the tone for the entire volume: The fundamental unit in today’s economy is the individual, a.k.a. YOU! Jobs are performed by temporary networks that disband when the project is done. So to succeed you have to think of yourself as a freelance contractor-Brand You! (bold in original). Peters’ assessment is entirely congruent (although from an altogether different perspective) with that of Castells (2000) and Bauman (2007)—namely, that self-programmable labour with zero drag is what the new capitalism requires and values. To become a brand, Peters recommends what can be identified as six broad steps. First, there is the need to be distinct—individuals, he suggests, need to break out of the pack and give full rein to their individuality and their quirkiness if they are to be noticed. Second, there is the need to be committed. Individuals have to be prepared to put in the work that is necessary to create a brand identity. Third, there is the need for passion. Here Peters points out that the successfully branded individuals to whom he frequently refers, such as Oprah Winfrey, Michael Jordan, and Steven Spielberg, are consistently passionate about the work they do. Fourth, he argues for the need to think and act strategically and he suggests that people who seek success can get to the top


RECONFIGURATION OF THE SELF

J. GRAY

719

through being prepared to move sideways within an organization or even to move downwards, as this can be an opportunity to learn new skills which may be of use later on. And of course, being strategic can also mean being prepared to relocate. Fifth, he suggests that individuals should exercise choice and only do work that is ‘cool’ and ‘fun’; and finally, he urges his readers to begin branding themselves immediately if they wish to succeed in a rapidly changing world. With this in mind, I turn in the following sections to the textbooks themselves.

CRITERIA FOR SELECTION AND METHODOLOGY Downloaded from applij.oxfordjournals.org by guest on December 31, 2010

For this article I have revisited a small set of textbooks analysed in another study (Gray 2010): Streamline Connections (Hartley and Viney 1979); Building Strategies (Abbs and Freebairn 1984); The New Cambridge English Course 2 (Swan and Walter 1990); and The New Edition New Headway Intermediate (Soars and Soars 2003a). The selection of these courses was based on the fact that they are recognized within the industry as major global bestsellers (OUP, personal communication, 2000; UK booksellers Waterstones, personal communication, 2004). However, given the spectacular success of the Headway course since its first publication in 1986, I have augmented this sample by including an additional set of textbooks from the series. This course has been described by Holliday (2005: 41) as one of the ‘cultural icons’ of ‘Western TESOL’—that is as a revered and commercially successful artefact which exerts a powerful influence on textbook design and pedagogic practice globally. Sales figures for the Headway course are not available, but one editor I spoke to admitted that sales from the course alone were sufficient to fund an entire publishing house (OUP, personal communication, 2000). I have also included a further two textbooks from the Cutting Edge course as examples of what another contemporary bestselling course contains. Altogether the additional textbooks are: New Headway Intermediate (Soars and Soars 1996); New Headway Upper-Intermediate (Soars and Soars 1998); New Headway Elementary (Soars and Soars 2000a); New Headway Pre-Intermediate (Soars and Soars 2000b); New Headway Advanced (Soars and Soars 2003b); New Headway Upper-Intermediate (Soars and Soars 2005); New Cutting Edge Intermediate (Cunningham and Moor 2005a); New Cutting Edge Upper-Intermediate (Cunningham and Moor 2005b). In identifying the representational repertoires related to the world of work, I initially adopted a content analysis approach and began by counting the number of units in each textbook which used the topic of work for a skills-based or grammar-based activity. In this way (following scholars such as Littlejohn 1992; Sercu 2000), I was able to make quantitative statements about the extent of work as a theme across each textbook (see Table 1 where these are represented proportionally). I then followed this up with a more qualitative analysis in which I sought to identify the correlations between representations of the world of work and the specific values associated with the new


capitalism. My approach here was to code any work-related text which explicitly featured self-programmable labour, zero drag, or individuals who displayed any of the Brand You characteristics as indexing new capitalist values. Take, for example, the following profile of a young businessman:

New capitalist values are evident in Roberto’s relocation to New York in pursuit of professional success, his commitment to hard work (confirmed by the authorial voice), and his endorsement of the US as a place where individual choice can be freely exercised. It is important to point out that this is a positive correlation with new capitalism (reinforced by a smiling photograph of the handsome young Roberto). Negative correlations, in which aspects of the new capitalism are seen as problematic, are rare in the data base and the few that do exist are referred to in the analysis below. At the same time, the following profile of a shopkeeper, although coded as being about work, could not be correlated with any of the values outlined above: My uncle is a shopkeeper. He has a shop in an old village by the River Thames near Oxford. The shop sells a lot of things—bread, milk, fruit, vegetables, newspapers—almost everything! It is also the village post office. The children in the village always stop to spend a few pence on sweets or ice-cream on their way home from school. My uncle doesn’t often leave the village. He hasn’t got a car, so once a month he goes by bus to Oxford and has lunch at the Grand Hotel with some friends. He is one of the happiest people I know (ibid.: 33). I now turn in the following section to the nature of the representational repertoires identified and discuss these in greater detail.

THE WORLD OF WORK IN ELT TEXTBOOKS As can be seen in Table 1, the world of work features in all the textbooks analysed but varies considerably in terms of the actual numbers of units which explicitly address the topic. That said, although a textbook such as Cambridge English 2 has only three units in which the world of work is salient


Roberto came from Acapulco to New York ten years ago. At first he missed everything—the sunshine, the food, his girlfriend. But now he has a successful business with his three brothers and his sister. They run a soccer store in New Brunswick. Roberto’s girlfriend is now his wife, and they have two children who go to American schools. When asked why he came to the US, Roberto says without hesitation, ‘Because I wanted to work hard and be successful.’ He certainly works hard. He’s at the store all day, then works as a driver in the evening. ‘That’s why I like America,’ he says. ‘You can be what you want’ (Soars and Soars 2000b: 19).

J. GRAY

721

Table 1: Work-related and new capitalist-related units in the textbook sample Number Number Proportion of units of units of workin which related work is content a major theme

Number of work-related units which correlate with new capitalist values

Proportion of new capitalist-related content

Streamline Connections 1979 Building Strategies 1984 The New Cambridge English Course 1990 The New Edition New Headway 2003 New Headway 1996 New Headway Upper-Intermediate 1998 New Headway Elementary 2000 New Headway Pre-Intermediate 2000 New Headway Advanced 2003 New Headway Upper-Intermediate 2005 New Cutting Edge Intermediate 2005 New Cutting Edge Upper-Intermediate 2005

80

12

0.15

3

0.04

16

9

0.56

9

0.56

36

3

0.08

0

0

12

4

0.33

4

0.33

12 12

4 4

0.33 0.33

3 4

0.33 0.33

14

3

0.21

3

0.21

14

6

0.43

3

0.02

12

3

0.25

3

0.25

12

4

0.33

4

0.33

12

2

0.17

2

0.17

12

2

0.17

2

0.17

as a topic, none of which can be correlated with new capitalism, its cast of female characters which feature in other units are often marked as being distinct by virtue of the less traditional jobs many of them do (e.g. assault course leader, taxi driver, pilot and explorer)—although in many cases these jobs are mentioned only in passing. As I will show below, the qualitative analysis of the textbooks suggests an evolution over the period towards an increased focus on individualism, concomitant with a ‘Brand You’ perspective and, in general, a celebratory view of the world of work as a means to personal fulfilment. Streamline Connections (Hartley and Viney 1979), which appeared as the neoliberal era was beginning to take shape, is interesting in this respect. Here the world of work is often represented negatively and textbook characters


Coursebook


regularly complain about work or are represented as workshy. The following extract typifies this approach:

This is similar to representations of work identified by Leahy (2004) in her study of contemporary German grammar books—namely an association with deontic modality (‘I need the money’); a negative impact on health (‘you look tired’; ‘You’re going to kill yourself’); and feelings of powerlessness (‘we’re just too busy; ‘What can I do?’; ‘I can’t’). Elsewhere in the textbook the recurring figure of ‘the boss’ (the aptly named Mr Power) is represented variously as paternalistic and authoritarian (also congruent with Leahy 2004), and work itself is frequently seen as repetitive or as something which infringes on the pursuit of happiness or personal autonomy. However, from the 1980s onwards the world of work is represented in a more consistently positive light and characters repeatedly display distinction, commitment and passion in relation to their chosen careers—in which they begin to achieve increasingly spectacular success. With the exception of two titles (other than Streamline Connections), Table 1 shows a close correlation overall between representations of work and new capitalist values. It should also be said that there is considerable overlap between the Brand You characteristics themselves—thus successful characters may achieve distinction through their exercise of personal choice or as a result of their commitment and passion. In many cases those who are distinct are also women—a feature which reflects the profound and enduring impact of feminism on all post-Streamline courses produced in the UK. Thus in all the textbooks surveyed we find women who stand out on account of their success in the world of business, science, politics, sport, the arts, and the media, alongside those who are distinct through their choice of non-traditional or quirkier occupations (e.g. clown, racing driver, plumber). Such characters, when profiled in detail, are invariably presented for student approval. This can happen in a variety of ways: characters, their jobs or their lifestyles may be positively evaluated by the authorial voice of the text; they may be positively evaluated by others mentioned in the text; they may positively evaluate their own jobs and the rewards they bring; and/or the accompanying artwork may be designed so as to elicit a positive emotional response in the viewer—what in multimodal terms is referred to as ‘sensory orientation’ (Kress and


Wendy: Hello Charles . . . you look tired today. Charles: Yes, I’m working too hard. Wendy: You should take a holiday. Charles: Yes, I know I should . . . but we’re just too busy. I’m working twelve hours a day. Wendy: Twelve hours! You’re going to kill yourself! Charles: What can I do? Wendy: Perhaps you should change your job. Charles: I can’t . . . I need the money! (ibid.: Unit 38, no pagination).

J. GRAY

723

Our aim in life is not to own a house and a car and to bring up children. It is to do something different, to achieve something special (ibid.). Zero drag and individual choice are also evident in the same textbook. Thus, Rod Nelson, a young electrical engineer and one of the central characters, relocates from Canada to Bristol in the first unit. Profiled in the local newspaper as someone ‘making a new start’ (ibid.: 18), Rod answers the question ‘Why did you leave Canada?’ as follows: I was bored. I worked in the same office and saw the same people and did the same thing every day. I needed a change. I wanted adventure (ibid.). In fact all the key characters in Building Strategies relocate, plan to relocate or have relocated as part of their working lives—invariably impelled by individual choice, and in flight from boredom or in pursuit of ‘fun’, challenge, or improved working conditions. In the case of production manager Jack Cooper, Rod’s colleague at work, the decision to relocate to France is linked to his frustration at industrial unrest in the Bristol factory. He eventually opts for the personal and professional challenge of remaking his life abroad away from the threat of strikes. Thus, while the post-Connections textbooks analysed do not ignore that work can be ‘problematic’, the treatment and individual responses to work-related problems tends to resonate with the ‘Brand You’ imperative to take control and to look out for oneself.


van Leewuen 1996), whereby the image is softened, coloured or otherwise manipulated to enhance its impact or its attractiveness to the viewer. An early example combining several such representational modes is found in the profile of ice skaters Jayne Torvill and Christopher Dean (Abbs and Freebairn 1984). The text, which features a colour photograph of the skaters bathed in a warm golden light, is printed in white against a black background with the heading ‘The Perfect Pair’ in a font which is suggestive of neon lighting. In addition to the evaluative adjective in the title, the authorial voice also mentions their ‘incredible’ scores and quotes from journalists’ evaluations—‘ ‘‘Unbelievably brilliant!’’ [. . .] ‘‘They are not just the best, they’re the best ever!’’ ’ (ibid.: 25). Public evaluation is also included through the listing of character adjectives, thus, ‘Everyone agrees that Jayne and Christopher are two nice people. They are quiet, shy, polite and not at all spoilt by their success’ (ibid.). In terms of their commitment and passion, the text describes how those who wish to succeed as ice skaters ‘must give up their whole lives to their training’ (ibid.), an assessment echoed by a direct quotation from Jayne: ‘ ‘‘you have to practise all the time’’, says Jayne simply’ (ibid.). In this case the association with deontic modality indexes their commitment to distinction, rather than any infringement of personal agency. As Jayne explains:


Textbook characters do on occasion lose their jobs—so, for example, Roger Dromard, a fictional office worker (Soars and Soars 1996) tells his story in a unit entitled ‘Happiness!’ In what could be an anecdote from Peters’ (2008) manual on how to deal with company downsizing, he begins: D’you know the best thing that ever happened to me? D’you know what it was? It was when I lost my job. Yes, really! I never liked it—hated it in fact—stuck in an office all day with computers and a telephone. Now my hobby is my full-time job! (Soars and Soars 1996: 127). Downloaded from applij.oxfordjournals.org by guest on December 31, 2010

Losing his job is presented as an opportunity to take control and set himself up in business as a gardener. This attitude, both to the old job and to the opportunity presented by its loss, is similar to that expressed by Peters (2008: 13), who concludes that in the new capitalism: ‘Our lives are more precarious. But they’ve been given back to us. The challenge: What are we going to make of them?’ Self-employment (or becoming CEO of your own life, as Peters puts it), certainly for some textbook characters, is presented as an unproblematic option. In fact Roger Dromard is replaced in the 2003 edition of the same textbook by college graduate Jeff Norman who has chosen to do what could be seen as a low-status job working as a ‘paperboy’ so that, as he says, he can do what he wants in his free time. In this way the strategic move downwards advocated by Peters is given a somewhat different emphasis. However, as the text clarifies, Jeff runs his own business and earns $60,000 a year, with an additional $50 a week in tips. This profile is in strong contrast to that of the highly successful, globe-trotting lawyer Sidney Fisk featured on the previous page of the same textbook. His commitment to hard work [‘He’s paid very well, but he usually has to work long hours’ (Soars and Soars 2003: 15)] and his zero drag [‘At the moment he’s working in Mexico, and next week he’s travelling to France’ (ibid.)] clearly resonate with new capitalist values. However, the text suggests that he may not be happy. As elsewhere in the data base (see below), such negative correlations are counter balanced by more celebratory accounts— such as that of Jeff Norman who may be said to have risen to the challenge of the new capitalism more effectively. Individuals such as Jeff may be said to openly reject Fordist working practices in favour of the more insecure but more (supposedly) satisfying rewards of post-Fordist flexibility and autonomy (see Harvey 1989; Ritzer 1992). In the Headway course, the ‘rat race’ is either something to be pursued enthusiastically or unproblematically abandoned. Those who voluntarily downsize invariably demonstrate the wherewithal to reinvent themselves along lines more in keeping with their own personal desires and inclinations. Thus, in a section on ‘dream jobs’ (Soars and Soars 2003a: 58–9), the fictional Linda Spelman explains how she gave up her job as a lawyer to pursue a more fulfilling life as a trapeze artist. When asked for advice she says, ‘I’d say to anyone with a dream ‘Go for it! You only live once, so why stay in a boring job’ (ibid.: 59).

J. GRAY

725

There’s no such thing as a job for life these days. Employers can make you redundant as soon as there’s a downturn, so people don’t feel the same loyalty. A lot of my friends are changing their jobs to boost their career prospects. I expect I’ll have several jobs before I’m 30, and I hope I’ll have several careers. I don’t want to do the same thing for ever. I’m going for an interview next week. More money, more responsibility (Soars and Soars 2005: 50). This perspective on the world of work is essentially that of self-programmable labour and resonates with Peters’ (2008: 197) advice to ‘Just say ‘‘No’’ to Loyalty!’. So although the employment market is seen as precarious (‘There’s no such thing as a job for life these days’), this state of affairs accords with Alex’s own inclinations (‘I don’t want to do the same thing for ever’). That said, two of the seven characters interviewed provide bleaker assessments of the economic climate—and two rare negative correlations with new capitalist values. Peter Jamieson, a trainee manager and Ellie Green, a corporate lawyer say they are worried about job security and not being able to afford to buy their own house. But as Doogan (2007: 3) has pointed out, new capitalist discourse is also one of ‘precariousness and insecurity’—in fact it is presented as the price of freedom in Peters’ (2008) narrative. However, the majority of the characters interviewed are enthusiastic about the challenge this presents. Thus Bob West, a fictional plumber from London who is keen to relocate, says: I’m saving money, and as soon as my application has been processed, I’m going to leave the country and live in Canada. Now there’s a country that encourages young people and enterprise! (ibid.).


Throughout these representations of the world of work there is a clear ideology of individualism in operation. These characters are what Probyn (1990) in another context has referred to as members of a ‘choiceoisie’—that is members of a group whose lifestyle choices are largely unaffected by personal, financial or social constraints of any kind. As Bauman (2005) explains, it is the freedom to exercise choice within consumer society which indexes social status, and in such a society, typified by the aestheticization of ever more aspects of human activity and experience (Bourdieu 1984; Harvey 1989; Featherstone 1991), work too is increasingly evaluated in terms of its glamour, sensation or fulfilment potential. And it is precisely this kind of work which is celebrated repeatedly in these textbooks. At the same time, there are two further points to make with regard to the representation of the world of work—the first has to do with specific reference to the instability and stressfulness of life in the neoliberal climate and the second with the increasing importance of celebrity. In the following example we find a group of young people talking specifically about employment. Alex Williams, a fictional 24-year-old ‘marketing account manager’ says:



As with ‘choice’, ‘enterprise’ can be seen as a key term in neoliberal discourse. Cameron (2000: 7), referring to ‘enterprise culture’ (rather than to neoliberalism per se), suggests that ‘enterprise’ connotes values such as the ‘resourcefulness, self-discipline, openness to risk and change—that enable people to succeed in bold and difficult undertakings’. Bob’s enthusiasm to relocate to a country which encourages such values may be said to signal his more general subscription to the rules of the neoliberal game. In fact, enterprise is celebrated throughout the Headway course. In New Headway Advanced (Soars and Soars 2003b: 14) Vijay and Bhikhu Patel, the pharmaceutical company owners and joint winners of the 2001 Ernst & Young Entrepreneur of the Year award, are interviewed at length about how they built their multimillion-pound business after arriving in the UK with just five pounds between them. The reading activity which introduces a series of listening and speaking exercises about the brothers carries an evaluative subheading ‘The inspiring tale of two Asian brothers who fled to Britain from East Africa and made a fortune’ (ibid.: 14), and in the speaking exercise which follows, students are asked to say in what ways the Patels are good role models for young people. The Patel brothers and the late Anita Roddick, the founder of the Body Shop who is profiled in another unit, are examples of what Zˇizˇek (2008: 14) refers to ironically as ‘liberal communists’—that is philanthropic capitalists who provide a much-needed human face for the new capitalism. Thus Vijay Patel states that he would ‘like to be a role model, for anybody who wants to be somebody tomorrow’, adding that, ‘if I can touch one life, then my job in this life’s done’ (Soars and Soars 2003b: 133). Anita Roddick, talking about the entrepreneurial mindset, explains that ‘the money we make is of no interest to us’, and aligns herself with those who are ‘outside the system’ and for whom wealth simply ‘allows you to be generous’ (ibid.: 135). Of course, as Zˇizˇek (2008: 14) argues, the claim that ‘we can have the global capitalist cake, that is thrive as profitable entrepreneurs, and eat it, too, that is endorse the anti-capitalist causes of social responsibility and ecological concern’ is open to question. From the perspective of the new capitalism, such philanthropy (however well-intentioned) serves to offset potential appeals to the need for structural change. At the same time, there is recognition that such a benign view of the new capitalism is not shared by everyone (Soars and Soars 2003b). The same textbook reproduces an article from the left-leaning New Internationalist magazine which is openly critical of economic globalization and consumer culture in particular. However, in the speaking exercise which follows the reading, the students are invited to question the authorial stance as follows: ‘The writer holds strong views on these issues. Can you present some counterarguments?’, and then to get the critique started, the textbook provides the following example: ‘Multinational corporations keep prices down’ (ibid.: 28). This is one of the few instances of students being encouraged to read against the text that I have encountered and it is significant that it is on this particular topic. More common is the way in which students, elsewhere in the course, are taught to talk the very specific talk of the new capitalism, as in the

J. GRAY

727

following grammar exercise which asks them to select the appropriate expression in bold and then to practise the dialogue: A I don’t know how you can afford to buy all those fabulous clothes! B Still/Hopefully, I’m going to get a bonus this month. I should do. My boss promised it to me. After all/Presumably, I did earn the company over £100,000 last year. Basically/Actually, it was nearer £150,000. I do deserve it, don’t you think. B Of course/In fact, you do (Soars and Soars 1996: 127).

If you want to make it really big, don’t take any established, familiar path to celebrity and don’t follow in anyone else’s footprints. Create your own unique route [. . .] People like Oprah Winfrey, the Queen of Talk Shows, or Bill Gates, the Chairman of Microsoft have reshaped and redefined an occupation and even an industry in their own image. Their fame is assured (ibid.: 85). Although the unit does include a reading exercise in which it is suggested that excessive interest in celebrity can be dangerous, students are also asked


Here the subject matter is not available for discussion (unless the teacher draws attention to it)—rather, it provides the unquestioned and un-remarked-upon content in which language is contextualized. However, more pervasive than such specific references to new capitalist practices is the appearance and treatment of celebrity in all the textbooks analysed. This merits attention given that the celebrity is the most recognizable type of branded individual in consumer culture, but also because the neoliberal period is characterized by a proliferation of celebrities and because celebrity is increasingly associated with a wide range of jobs—for example, celebrity chefs, celebrity historians, celebrity doctors, celebrity entrepreneurs, etc. It is worth noting that a textbook such as Corder’s (1960) An Intermediate English Practice Book contains no reference to any celebrity of the period and that O’Neill’s (1970) English in Situations refers only indirectly to celebrity. However, by the late 1970s explicit references to celebrity are much more common. Table 2 lists the celebrities featured in Streamline Connections (Hartley and Viney 1979) and New Cutting Edge Upper-Intermediate (Cunningham and Moor 2005b). This illustrates not only the new salience of celebrity in textbooks aimed at the global market, but also the increase in the number of celebrities included over the period. In this respect, The New Cutting Edge Upper-Intermediate is typical of the textbooks in the sample dating from the mid 1990s. In fact, New Headway Advanced (Soars and Soars 2003b) and New Cutting Edge Upper Intermediate (Cunningham and Moor 2005b) both include a whole unit on the topic. The main reading text in the latter is entitled ‘How to be a celebrity’ and offers readers seven routes to achieving celebrity status. The seventh is ‘Create your own formula for success’ and includes the following advice:


Table 2: Celebrities in Streamline Connections (Hartley and Viney 1979) and New Cutting Edge Upper Intermediate (Cunningham and Moor 2005b) New Cutting Edge Upper Intermediate

Miklos Nemeth; Alberto Juantorena; Bob Beamon; Annegret Richter; Rosemarie Ackermann; David Wilkie; Vasily Alexeev; Paul McCartney; James Hunt; UK Queen; J.F. Kennedy; Elvis Presley

Jackson 5; Osmonds; Bee Gees; Corrs; Oasis; Boom Kat; Britney Spears; Eminem; Ozzy Osborne; Kelly Osborne; Brian May; Freddie Mercury; Rowan Atkinson; Halle Berry; Paul McCartney; Ringo Star; Steve Redgrave; Ghandi; Martin Luther King; Evana Trump; Bill Gates; John Kennedy Jnr; Jade Jagger; Princess Diana; Liz Hurley; Oprah Winfrey; Tom Cruise; Robbie Williams; Woody Allen; Marilyn Munroe; Arnold Schwarzenegger; Alan Alda; Fred Allen; Harrison Ford; Brad Pitt; David Beckham; George Cluney; Jennifer Lopez; Tony Blair; Madonna; Nicole Kidman; Steven Spielberg; Ricky Martin; John Lennon; Jennifer Aniston; Ralph Fiennes; Hillary Clinton; Elvis Presley; Kiera Knightley; Parminda Nagra; Gurinder Chadha; Roman Polanski; Nirvana

to invent a celebrity persona to role play and to find out if their partner ‘has got what it takes to be a celebrity’ (ibid.: 88). New Headway Advanced (Soars and Soars 2003b) adopts a similar approach and includes a reading in which our (assumed) interest in celebrity is seen as inevitable, if not altogether healthy, and at the same time asks students to work in small groups and decide on ways of becoming an A-list celebrity. Other exercises such as role playing and speculating about the lives of celebrities occur throughout the textbook sample. Furthermore, celebrities’ lives, whether fictional or non-fictional, are also repeatedly treated as examples of successful careers—as we saw earlier in the case of Torvill and Dean. Overall, the way in which celebrity operates in consumer societies is largely unaddressed in the textbooks analysed—rather celebrity tends to function as an index of professional success and, especially when accompanied by high-quality photographic artwork, provides a veneer of glamour.

DISCUSSION What then are we to make of such representations? Analysis of publishers’ guidelines for authors and interviews with senior ELT publishers confirm that there is nothing accidental about the nature of textbook content (Gray 2010). Taken together, I would suggest they constitute what Althusser (1971) refers to as an ‘interpellation’—that is, a hailing to the ideological and subject position of white-collar individualism in which the world of work is overwhelmingly seen as a privileged means for the full and intense realization of the self,


Streamline Connections

J. GRAY

729

It has pervasive effects on ways of thought to the point where it has become incorporated into the common-sense way many of us interpret, live in, and understand the world. Neoliberalism and the values associated with it have in fact become naturalized in many spheres. This is particularly so in commercial ELT, a lucrative global service industry which places a high value on self-programmable teaching labour with zero drag. The British Council’s 2003 recruitment campaign for teachers which ran under the heading ‘Teach English and Individualism’ is entirely congruent with the culture of the new capitalism (EL Gazette 2003: 282–7)—as indeed is the concomitant casualization of minimally-trained labour which typifies many pockets of the industry. A second answer is to


along lines determined largely by personal choice. As constructed in the majority of these materials, the world of work (and indeed the world in general) is seen in highly idealized and aestheticized terms. But why should this be the case? As suggested earlier, some languages are commodified in the globalized economy in the sense that they are marketed primarily in terms of their perceived economic usefulness. However, commodification in consumer culture also entails branding—indeed the branding (and re-branding) of people, places, institutions or languages is a pervasive feature of the new capitalism (see Wernick 1991; Lury 2004; Ritzer 2007). In an economy which is increasingly ‘organized around attention’ (Schroeder 2002: 3), branding becomes necessary if commodities are to be noticed and if they are to be considered worth having. Essentially branding is about the construction of a set of associations and a recognizable identity for products (whether material or symbolic). Ultimately the aim is to create identifications through which consumers can insert themselves into the ‘world’ of the brand. In this respect the representational repertoires deployed in these materials can be seen as part of the branding of English as if it were a commodity like any other in the marketplace—if branding is understood to operate through association and hoped for identification on the part of students, in this case with certain characters (e.g. celebrities) and certain characteristics (e.g. distinction, commitment, passion, success, enterprise, and zero drag). Elsewhere (Gray 2010) I have argued that the textual construction (and imaging) of English in ELT coursebooks parallels the processes of commodity promotion more generally. In the same way that the Sony Walkman was branded through repeated association with mobility, leisure and youth (Du Gay et al. 1997) and Benetton clothes through association with a particular discourse of multiculturalism (Tinic 1997), so too is English, as represented in these materials, given a specific identity. But the question remains as to why the branding takes the form it does—that is, why English is repeatedly endowed with a range of associations, one set of which is congruent with the values of the new capitalism rather than any other set of values. One answer I think is to be found in the fact that neoliberalism has become, in Harvey’s (2005: 3) words, ‘hegemonic as mode of discourse’. He continues:


ACKNOWLEDGEMENTS I would like to thank Angela Leahy, Marnie Holborow, John Trushell, and the peer reviewers for their advice and comments on earlier versions of this article.


be found in the publishers’ perception of the market. In the interviews with ELT publishers mentioned above, textbook content in which choice, individualism and spectacular professional success were central was repeatedly referred to by interviewees as ‘aspirational’. That is, it was seen as representing the kind of lifestyle students might aspire to and which (it was suggested) would motivate them in their language learning. Although Modiano (2001: 164) has speculated about the danger of ‘ontological imperialism’ implicit in ELT whereby ‘the learner’s mind is colonized through the acquisition of a foreign tongue’, and Phillipson (2008: 36) has argued that English as currently taught ‘contributes to the imperial production of subjectivities’, very little is actually known about the impact of such ideologically motivated content on students. Certainly it was the belief of scholars such as I.A. Richards, a key figure in the development of Anglo-American ELT, that the teaching of English had such potential (see Phillipson 1992, 2008). At the same time, Canagarajah (1999) suggests that students do on occasion recognize the ideologically motivated nature of certain content and seek to challenge it through, for example, defacement or annotation of materials. Clearly there is a need for more detailed research into how such content is perceived, the meanings students attribute to it and any motivational appeal it may or may not have. Research into teachers’ thinking, on the other hand, suggests that many teachers are indeed aware of the ideological dimension of much global textbook content and that some find the uncritical celebration of the lives of philanthropic capitalists and new capitalist values problematic (Gray 2002, 2010). As one teacher put it, ‘there is no criticism [. . .] no questioning [. . .] no discussion or anything, whether this is like morally right or not’ (Gray 2010: 157). Rather, another said, global textbooks engage in ‘glorifying a kind of middleclass’ for its eccentric individualism and conspicuous wealth which he felt ‘might be considered shallow’ in certain educational settings (ibid.: 156). The same teacher also made the case for critical pedagogy as a means of counteracting the possibility of ‘transferring a dominant culture to my students and to, even to myself’ (ibid.), a remark which suggests that Modiano’s speculations may be justified in some cases. As with all cultural artefacts, textbooks are products of the cultures which produce them—in this case the commercial culture of a powerful global industry. As currently branded along lines which I have suggested are congruent with the values and the culture of the new capitalism, it remains to be seen if the ongoing crisis in neoliberal ideology occasioned by the 2008 financial crisis will necessitate a re-branding of commercial ELT along new ideological lines.

J. GRAY

731

NOTE 1 See Dor (2006) for a nuanced discussion of the role of other languages in the global economy.

REFERENCES and the Political Economy of the Linguistic Code,’ Public Culture 16.1: 98–118. Du Gay, P., S. Hall, L. Janes, H. Mackay, and K. Negus. 1997. Doing Cultural Studies/The Story of the Sony Walkman. Sage/The Open University. EL Gazette 2003. Issue 282: 7. London. Fairclough, N. 2002. ‘Language in new capitalism,’ Discourse and Society 13/2: 163–6. Fairclough, N. 2006. Language and Globalization. Routledge. Featherstone, M. 1991. Consumer Culture and Postmodernism. Sage. Gee, J. P., G. Hull, and C. Lankshear. 1996. The New Work Order: Behind the Language of the New Capitalism. Polity Press. Gray, J. 2002. ‘The Global Coursebook in English Language Teaching’ in D. Block and D. Cameron (eds.): Globalization and Language Teaching. Routledge, pp. 151–67. Gray, J. 2010. The Construction of English: Culture, Consumerism and Promotion in the ELT Global Coursebook. Palgrave Macmillan. Hall, S. 1980. ‘Encoding/decoding’ in S. Hall (ed.): Culture, Media, Language. Hutchinson, pp. 128–38. Hartley, B. and P. Viney. 1979. Streamline English Connections. Oxford University Press. Harvey, D. 1989. The Condition of Postmodernity. Basil Blackwell. Harvey, D. 2005. A Brief History of Neoliberalism. Oxford University Press. Hayek, F. A. 1978. New Studies in Philosophy, Politics, Economics and the History of Ideas. Routledge and Kegan Paul. Heller, M. 2003. ‘Globalization, the new economy, and the commodification of language and identity,’ Journal of Sociolinguistics 7/4: 473–92. Hertz, N. 2002. The Silent Takeover: Global Capitalism and the Death of Democracy. Arrow Books. Holborow, M. 2007. ‘Language, ideology and neoliberalism,’ Journal of Language and Politics 6/1: 51–73.


Abbs, B. and I. Freebairn. 1984. Building Strategies/Student’s Book. Longman. Althusser, L. 1971. Lenin and Philosophy and other Essays. New Left Books, Monthly Review Press. Apple, M. W. and L. K. Christian-Smith. 1991. The Politics of the Textbook. Routledge. Bauman, Z. 2005. Work, Consumerism and the New Poor. Oxford University Press. Bauman, Z. 2007. Consuming Life. Polity Press. Block, D. 2008. ‘Language Education and Globalization’ in S. May and N. H. Hornberger (eds.): Encyclopedia of Language and Education, vol. 1. Springer, pp. 1–13. Boltanski, L. and E. Chiapello. 2007. The New Spirit of Capitalism. Verso. Bourdieu, P. 1984. Distinction: A Social Critique of the Judgement of Taste. Routledge and Keegan Paul. Bourdieu, P. 1998. ‘Utopia of endless exploitation: the essence of neoliberalism,’ Le Monde Diplomatique. Available at: http://mondediplo .com/1998/12/08bourdieu. Accessed 15 September 2010. Bourdieu, P. and L. Wacquant. 2001. ‘NewLiberalSpeak: notes on the new planetary vulgate,’ Radical Philosophy 105: 2–5. Cameron, D. 2000. Good to Talk? Routledge. Canagarajah, S. 1999. Resisting Linguistic Imperialism. Oxford University Press. Castells, M. 2000. ‘Materials for an exploratory theory of the network society,’ British Journal of Sociology 51.1: 5–24. Corder, S. P. 1960. An Intermediate English Practice Book. Longman. Cunningham, S. and P. Moor. 2005a. Cutting Edge/Intermediate. Pearson Longman. Cunningham, S. and P. Moor. 2005b. New Cutting Edge/Upper Intermediate. Pearson Longman. Doogan, K. 2009. New Capitalism? The Transformation of Work. Polity Press. Dor, D. 2004. ‘From Englishization to Imposed Multilingualism: Globalization, the Internet


Sercu, L. 2000. Acquiring Intercultural Communicative Competence from Textbooks. Leuven University Press. Singapore Wala, D. A. 2003a. ‘A coursebook is what it is because of what it has to do: an editor’s perspective’ in B. Tomlinson (ed.): Developing Materials for Language Teaching. Continuum, pp. 58–71. Singapore Wala, D. A. 2003b. ‘Publishing a coursebook: completing the materials development circle’ in B. Tomlinson (ed.): Developing Materials for Language Teaching. Continuum, pp. 141–61. Soars, L. and J. Soars. 1996. New Headway. Oxford University Press. Soars, L. and J. Soars. 1998. New Headway/Upper Intermediate. Oxford University Press. Soars, L. and J. Soars. 2000a. New Headway/ Elementary. Oxford University Press. Soars, L. and J. Soars. 2000b. New Headway/ Pre-Intermediate. Oxford University Press. Soars, L. and J. Soars. 2003a. New Edition New Headway/Intermediate. Oxford University Press. Soars, L. and J. Soars. 2003b. New Headway/ Advanced. Oxford University Press. Soars, L. and J. Soars. 2005. New Headway/ Upper-Intermediate. Oxford University Press. Stuckler, D., L. King, and M. McKee. 2009. ‘Mass privatisation and the post-communist mortality crisis: a cross national analysis,’ The Lancet 373: 399–407. Swan, M. and C. Walter. 1990. The New Cambridge English Course 2. Cambridge University Press. Tan, P. K. W. and R. Rubdy. 2008. Language as Commodity: Global Structures, Local Marketplaces. Continuum. Thomas, D. 1999. ‘Culture, ideology and educational change: the case of English language teachers

in

Slovakia,’

Unpublished

PhD

thesis, Institute of Education, University of London. Tinic, S. A. 1997. ‘United colors and untied meanings: Benetton and the commodification of social issues,’ Journal of Communication 47/3: 3–25. Tomlinson, B. 2003. ‘Humanizing the Coursebook’ in B. Tomlinson (ed.): Developing Materials for Language Teaching. Continuum, pp. 162–73. Tomlinson, B., B. Dat, H. Masuhara, and R. Rubdy. 2001. ‘EFL Courses for Adults,’ ELT Journal 55/1: 80–101.


Holliday, A. 2005. The Struggle to Teach English as an International Language. Oxford University Press. Ives, P. 2006. ‘‘Global English’: Linguistic Imperialism or Practical Lingua Franca?,’ Studies in Language and Capitalism 1: 121–41. Klein, N. 2000. No Logo. Flamingo. Klein, N. 2007. The Shock Doctrine. Penguin. Kress, G. and T. van Leeuwen. 1996. Reading Images: The Grammar of Visual Design. Routledge. Leahy, A. 2004. ‘The representation of work in German grammar books’. Unpublished PhD thesis, Dublin City University. Leung, C. 2005. ‘Convivial communication: recontextualizing communicative competence,’ International Journal of Applied Linguistics 15/2: 119–44. Littlejohn, A. 1992. ‘Why are English Language Teaching materials the way they are?’ Unpublished PhD thesis, Lancaster University. Lury, C. 2004. Brands: The Logos of the Global Economy. Routledge. Masuhara, H., N. Hann, Y. Yi, and B. Tomlinson. 2008. ‘Adult EFL courses,’ ELT Journal 62/3: 294–312. McCallan, B. 1990. English in Eastern Europe. Special Report No. 2057. The Economist Intelligence Unit. Modiano, M. 2001. ‘Ideology and the ELT practitioner,’ International Journal of Applied Linguistics 11/2: 159–73. O’Neill, R. 1970. English in Situations. Oxford University Press. Pennycook, A. 1994. The Cultural Politics of English as an International Language. Longman. Peters, T. 2008. The Brand You 50: Fifty Ways to Transform Yourself from an ‘‘Employee’’ into a Brand that Shouts Distinction, Commitment, and Passion! Alfred A. Knopf. Phillipson, R. 1992. Linguistic Imperialism. Oxford University Press. Phillipson, R. 2008. ‘The Linguistic Imperialism of Neoliberal Empire,’ Critical Inquiry in Language Studies 5/1: 1–43. Probyn, E. 1990. ‘New traditionalism and post-feminism: TV does the home,’ Screen 31/2: 147–59. Ritzer, G. 1992. Sociological Theory. McGraw-Hill. Ritzer, G. 2007. The Globalization of Nothing 2. Pine Forge Press. Schroeder, J. E. 2002. Visual Communication. London. Sennett, R. 2006. The Culture of the New Capitalism. Yale University Press.

J. GRAY

Van Dijk, T. A. 2008. Discourse and Power. Palgrave Macmillan. Wallace, M. J. 2004. Study Skills in English. Cambridge University Press. Wajnryb, R. 1996. ‘Death, taxes, and jeopardy: systematic omissions in EFL texts, or life was

733

never meant to be an adjacency pair,’ Ninth Educational Conference, ELICOS, Sydney, Australia. Wernick, A. 1991. Promotional Culture. Sage. Zˇizˇek, S. 2008. Violence. Profile Books. Zˇizˇek, S. 2009. First as Tragedy, Then as Farce. Verso.


Applied Linguistics: 31/5: 734–747

ß Oxford University Press 2010

REVIEWS Roy Harris and Christopher Hutton: DEFINITION IN THEORY AND PRACTICE: LANGUAGE, LEXICOGRAPHY AND THE LAW. Continuum, 2007.


The essence of definition and its applicability in various spheres of human life has been the subject of scholarly inquiry for centuries. This book is an attempt to reframe long-existing issues through the lens of modern conceptualization and, in particular, through the advances of linguistic theory. As mirrored in the title, the authors address three areas—theory, lexicography, and the law— where definition has played a particularly important role and has undergone multiple conceptualizations and interpretations over the ages. The structure of the book includes three main parts, each dealing with one of the three main topics, and a conclusion. The first chapter, ‘On Stipulative Definition’ tackles the controversial issue, previously addressed in one way or the other by many language philosophers, including Euclid, Aristotle, Stebbing, and Chomsky, of whether using a stipulative definition, or stating the correct or intended definition in advance, has any conceptual value. Drawing on Saussure’s (1922) work on the nature of sign, the authors suggest that an important prerequisite for the validity of stipulative definition concerns the generally acknowledged arbitrariness of linguistic signs and general communicative effects of efforts directed at creating a stipulative definition. In Chapter 2, ‘On Definition and Common Usage’, the authors address the aspect of common usage, also referred to as ‘lexical’ (Hurley 1988), linguistic, or verbal in previous research. They analyze multiple prior approaches to what was considered a ‘common usage’ definition, differentiating between approaches which are reocentric (centered around the belief that words connect to the independently existing reality; e.g. Socrates/Plato and Mill) and psychocentric (focused more on connections between ideas within a human mind; Locke 1706), and in particular zooming in on Locke’s (1706) argument undermining the existence of common usage. The way around such a conclusion is to construe common usage as something ‘other than the sum total of conflicting individual usages’ (p. 35), or to treat a variety of meanings as part of the overall organization of language, rather than as creative decisions of its users. In Chapter 3, ‘On Real Definition’, the authors explore how one can address the essence of a phenomenon or thing without resorting to using its name. The ‘real definition’ is ‘in a sense a summary of what is currently believed to be uniquely distinctive about a certain type of thing found in Nature’ (p. 43). The authors proceed to analyze problems associated with ‘real definition’ in philosophy, logic, and some branches of modern linguistics, also focusing on the key problem of definition in logic—the generally unrealistic orientation toward a perfectly orderly world with a static

REVIEWS 735


language and sharp boundaries. In Chapter 4, ‘Ostensive Definition and Linguistic Theory’, the authors explore repercussions of linguistic theory reflected in the issue of definition, in particular tapping into language acquisition research. They make a case for an integrationist view and outline key differences from Locke’s, Bloomfield’s, and Quine’s views, arguing instead that definition is a dynamic construct and thus by default allows room for semantic ambiguity and divergence. The second part of the book, ‘Definition and the Dictionary’, outlines the issues of definition in the processes of designing and creating a dictionary. One of the key ideas within Chapter 1, ‘The Lexicographer’s Task’, concerns the stance of previous ‘general’ dictionary authors and possible errors (often originating from the so-called ‘historical principles’ and lexicographers’ prescriptivist assumptions) that may be lingering in present-day dictionaries. Analyzing the quotes from influential lexicographers such as Johnson and Murray, the authors argue that they did not just describe independently existing meanings, but in fact constructed and allocated such meanings on the basis of their respective viewpoints and accordingly selected materials. Chapter 6, ‘Definitions and History’, is dedicated to discussing the consequences of some ‘historical principles’ and factors that contributed to construction of a historical definition. One example of such is Murray’s stance that every meaning should be tracked back to the word’s etymology, thus denying the value of knowing the word without knowing its etymology. The discussion of etymology also uncovers issues of synonymy and antonymy, and how they may be viewed in philosophy and in some linguistic theory. In Chapter 7, ‘Types and Problems of Definition’, the authors examine a relatively standard typology of dictionary definitions [selected from Robinson (1954)] and discuss its multiple shortcomings, the key one being its failure to ground the distinguishing features of various definition types in any theory. Another key problem of modern lexicography is the lack of clear differentiation between connotation and denotation, as well as the absence of ‘communicational’ definition (p. 130) from modern dictionaries. The next three chapters constitute the part titled ‘Definition and the Law’. Despite the acknowledged ‘need to balance legal certainty with flexibility and adaptability’ in light of specific ‘texts, intentions, and factual events’ (p. 133), the issue of definition so far received less attention from legal scholars than it deserves. A significant part of Chapter 8, ‘The Definition of Law and Legal Definition’ is dedicated to the discussion of conceptualization of the law and the role of definition within it (specifically, Stone’s (1964) discussion of the nature of law, Hart’s (1954) lecture on ‘Definition and theory in jurisprudence’, and Tiersma’s (2000) discussion of definition in law). The authors indicate the integrational view of law as ‘being at stake in the legal conflicts and processes themselves’ (p. 141). Chapter 9, ‘Strategies of Construction’, addresses the problems of definition as they arise in common law. Generally, definitionally related disputes are either solved on a case-by-case basis or aligned with a particular approach to interpretation (e.g. ‘new textualism’ of

736 REVIEWS


Justice Antonin Scalia in the USA). The authors proceed with discussing various approaches to the interpretation of legal texts and the role of dictionaries in those approaches, zooming in on specific examples that demonstrate the extent of the blurriness of borders between ordinary language and legal language and, correspondingly, between ‘literal’ and ‘purposive’ interpretation. The authors suggest that their approach, integrationism, assumes that each legal case and/or context should dictate its own definitional characteristics and rules of interpretation. Chapter 10, concluding the trilogy of law chapters in the book, is titled ‘Linguistics, Science and Meaning’, and accordingly addresses the status of law as a field of study and its potential range of interaction with other spheres of intellectual and social life. The authors address applicability of various linguistic theories and even specific interpretative tools to contentious legal issues. The final, Chapter 11, ‘Definition, Indeterminacy, and Reference’, addresses whether trying to establish definitional determinacy is even a wise endeavor in the vast and mostly indeterminate chaos of human thought. Integrationists, overall, embrace the view of ‘radical indeterminacy’ (p. 201), which implies that linguistic signs are ‘ ‘‘made to mean’’ whatever the circumstances require’ (p. 202), and the issue of seeking determinacy thus becomes superfluous and unnecessary by default. The authors also discuss how indeterminacy may be related to the issue of reference, introducing an integrationist concept of ‘referential focus’, applicable to ‘referring expressions in general, and indeed to signs of all kinds’ (p. 211). In general, the book offers a deep and wide-ranging discussion on how the roles of definition have manifested themselves in theory, lexicography, and law. It is a powerful overview of prior thought on the topic, but one minor criticism concerns the general introduction of integrationism to the reader. While the authors frequently refer to the integrationist theory and periodically include brief characterizations of the integrationist stance toward one or the other aspect discussed in a given chapter, such brief references to integrationism might appear a bit choppy and somewhat confusing to readers who are new to integrationism. In the book’s concluding paragraphs, the authors aptly characterize their theory as a whole: ‘Integrationism is committed to a ‘‘lay-oriented’’ analysis of communication in the following sense: that, where language is concerned, there is no basis of knowledge or expertise other than that available to any lay member of the community’ (p. 222). This kind of characterization would have been more appropriate earlier on in the book, perhaps in a separate chapter with a more focused overview of the theory (supplemented by a summary of main theoretical tenets and works). That said, the book’s layout is truly integrational in nature, as no discussion of integrational stance is provided outside of the context in question. Rhetorically, each chapter contains a multitude of opinions, summaries of and references to prior research on a given topic. Within such polyphony, it is sometimes difficult to hear the analytic voices of the authors themselves; or, rather, the voices are often manifested indirectly through choice of arguments

REVIEWS 737

and issues selected for consideration, rather than through direct statements. Finally, while integrationism is positioned as a theory of language grounded in real-world linguistic usage, it would be helpful to gain theoretical insights regarding the origins of integrationism, its methodology, and its conceptual stance against other usage-based approaches to language analysis, for example, cognitive linguistics. Overall, the book is thoroughly engaging and thought-provoking. It is a must-read for anyone interested in the issues of definition.

REFERENCES Hart, H. L. A. 1954. ‘Definition and theory in jurisprudence,’ Law Quarterly Review 70: 37–60. Hurley, P. J. 1988. A Concise Introduction to Logic, 3rd edn. Wadsworth. Locke, J. 1969 (originally 1706). An Essay Concerning Human Understanding, 6th edn. (Reprinted in Fraser, A. C. 1894.) Dover.

Robinson, R. 1954. Definition. Clarendon. Saussure, F. de. 1922. Cours de Linguistique Geńe´rale, 2nd edn. Payot. Stone, J. 1964. Legal System and Lawyer’s Reasoning. Stevens. Tiersma, P. M. 2000. Legal Language. University of Chicago Press.

Philip Seargeant: THE IDEA OF ENGLISH IN JAPAN: IDEOLOGY AND THE EVOLUTION OF A GLOBAL LANGUAGE. Multilingual Matters, 2009. There is a lot more to language than language, a lot more to English than English. This will not come as news to teachers, policy-makers, applied linguists, sociolinguists, or indeed to anyone who has taken a moment to think seriously about the stuff that flows from our mouths and pens. We all know that, as Philip Seargeant writes, ‘our use of language is always influenced by the ideas we form of language’ (p. 1), and we all know, too, that these ideas are inextricably enmeshed with a variety of non-linguistic factors. In The Idea of English in Japan, Seargeant gives substance and rigor to these intuitions, exploring, for example, why the Japanese, a well-educated people, instructed in generously funded programs by cadres of education professionals both foreign and native, have so little apparent success in mastering English. The trouble, one is certain, is neither in the language itself nor in the learners and teachers. Rather, it is to be found in the ideology that has grown up around English—in, that is, ‘language ideology’, a concept that Wolfram and Schilling-Estes (1998: 9) have defined as the ‘ingrained,


Reviewed by Natalia D. Jacobsen Georgetown University, USA E-mail: [email protected] doi:10.1093/applin/amq029 Advance Access published on 13 September 2010

REVIEWS 737

and issues selected for consideration, rather than through direct statements. Finally, while integrationism is positioned as a theory of language grounded in real-world linguistic usage, it would be helpful to gain theoretical insights regarding the origins of integrationism, its methodology, and its conceptual stance against other usage-based approaches to language analysis, for example, cognitive linguistics. Overall, the book is thoroughly engaging and thought-provoking. It is a must-read for anyone interested in the issues of definition.

REFERENCES Hart, H. L. A. 1954. ‘Definition and theory in jurisprudence,’ Law Quarterly Review 70: 37–60. Hurley, P. J. 1988. A Concise Introduction to Logic, 3rd edn. Wadsworth. Locke, J. 1969 (originally 1706). An Essay Concerning Human Understanding, 6th edn. (Reprinted in Fraser, A. C. 1894.) Dover.

Robinson, R. 1954. Definition. Clarendon. Saussure, F. de. 1922. Cours de Linguistique Geńe´rale, 2nd edn. Payot. Stone, J. 1964. Legal System and Lawyer’s Reasoning. Stevens. Tiersma, P. M. 2000. Legal Language. University of Chicago Press.

Philip Seargeant: THE IDEA OF ENGLISH IN JAPAN: IDEOLOGY AND THE EVOLUTION OF A GLOBAL LANGUAGE. Multilingual Matters, 2009. There is a lot more to language than language, a lot more to English than English. This will not come as news to teachers, policy-makers, applied linguists, sociolinguists, or indeed to anyone who has taken a moment to think seriously about the stuff that flows from our mouths and pens. We all know that, as Philip Seargeant writes, ‘our use of language is always influenced by the ideas we form of language’ (p. 1), and we all know, too, that these ideas are inextricably enmeshed with a variety of non-linguistic factors. In The Idea of English in Japan, Seargeant gives substance and rigor to these intuitions, exploring, for example, why the Japanese, a well-educated people, instructed in generously funded programs by cadres of education professionals both foreign and native, have so little apparent success in mastering English. The trouble, one is certain, is neither in the language itself nor in the learners and teachers. Rather, it is to be found in the ideology that has grown up around English—in, that is, ‘language ideology’, a concept that Wolfram and Schilling-Estes (1998: 9) have defined as the ‘ingrained,


Reviewed by Natalia D. Jacobsen Georgetown University, USA E-mail: [email protected] doi:10.1093/applin/amq029 Advance Access published on 13 September 2010

738 REVIEWS


unquestioned beliefs about the way the world is, the way it should be, and the way it has to be with respect to language’. Language ideology is the lens through which Philip Seargeant examines English in Japan, and his scrutiny is fruitful. He begins by dissecting the lip service that is often paid in Japan to the importance of English as a global language, quoting as an example (in pp. 6–7) the Japanese Ministry of Education’s assertion that, ‘for children living in the 21st century, it is essential for them to acquire communication abilities in English as a common international language’. One need not have recourse to language ideology or to any other recondite concept to see that this is, on its face, absurd. While it is certainly true that some small percentage of Japanese will need English to succeed, and that the existence of that small group may be essential for Japan to interact in necessary ways with the world, most Japanese barbers, chefs, mathematics teachers, and domestic business people, for example, will have no problem at all living lives in which English plays little part. In a country such as Japan, which shares borders with no other countries, English will be necessary for a few, helpful and life-enhancing for some, but irrelevant to the vast majority. Their pronouncement notwithstanding, the mandarins at the Ministry of Education surely know this. It is clear, therefore, that there is something other than a rational belief in the necessity of English at play here. As Seargeant demonstrates, the missing piece that is essential to understanding the status of English in Japan is language ideology. Ideology arises from aspirations and ideals just as often as from what those not in thrall to the ideology might call ‘the facts’, and it is clear that the Japanese educational bureaucracy is, at least when drafting these proclamations, ensnared by an ideology that positions English as a language which everyone must know. Having established, with this example and others, the significant influence that language ideologies can have, Seargeant moves, in his fourth chapter, to ‘the state and status of English language teaching in Japan’ (p. 43). More precisely, he returns to the way the state of English teaching in Japan is perceived through the ideological filters that all who are involved with it— applied linguists not excluded—employ. Examining several academic accounts, and also governmental pronouncements and ‘action plans’, Seargeant focuses, in this chapter, on ‘[. . .] the discourse of institutional regulators of the education system rather than that of its participants’ (p. 44). Taken for granted in this discourse, he finds, is the notion that English education in Japan is, for reasons traceable to events in Japanese history, broken and therefore needs to be fixed. This assumption, Seargeant believes, affects research into English education in significant ways: ‘By positing that there is a ‘‘problem’’ in need of solution—in using this as the narrative arc of the research presentation—the research needs to establish the existence of a problem before being able to move on to the possibility of a solution’ (p. 61). Though Seargeant allows that ‘the problem frame’ (p. 47) has generated work that can be ‘illuminating and instructive’ (p. 61), it is, he believes, constraining.

REVIEWS 739


Foremost among the solutions offered to the problems assumed to be afflicting English education in Japan are attempts to replace or augment the traditional grammar translation method of language teaching with more communicative approaches. In his discussion of these attempts Seargeant is able, by attending to the ideologies surrounding English, to shed light on the equivocal effects of these methodologies. A significant impediment, for example, is that ‘communication’ is not an unambiguous concept. Since Japanese society is traditionally more hierarchical than many Western societies, what is meant by ‘communication’ in Japan may not be identical with what is meant by ‘communication’ in, say, the USA. Thus, when ‘communication’ is taught in an English class in Japan, students may be learning not just the English language, but also a style of communication foreign to them. The danger, as Seargeant notes, in pursuing an approach whereby the Western-style communication that goes on in the classroom is cordoned off from the communication that goes on in Japanese society at large is the implication that before the ability to communicate in English can be effectively taught or learned, Japanese society will have to change significantly in areas as fundamental as communication style, something that is unlikely to happen any time soon. Further, when ‘[. . .] the ‘‘foreignness’’ of English, both as code and cultural practice is foregrounded’ (p. 60) the inclusiveness implied by notions of English as an international language is negated. In chapter five, Seargeant is concerned with globalization, and how Japan is responding to the irruption of foreign cultures and languages into the archipelago. Most relevant to those concerned with language is the odd and unconventional use that Japanese have made of English, use which exemplifies— as is the case with the ‘foreign country theme parks’ (gaikokumura) Seargeant also considers (p. 70)—how Japanese culture takes possession of that which is not Japanese and uses it for its own purposes. Seargeant cites the familiar example of ‘mansion’ (manshon), a word that, in Japan, does not refer to a luxurious residence, but means, instead, ‘apartment’ (p. 83). In other cases, it is not that the English words one encounters in Japan have been redefined; it is that they are entirely ornamental. Indeed, a stroll around Tokyo will suggest that on t-shirts and in advertisements, for example, English has roughly the same status as shapes, colors, and fonts: it is an element of graphic design (and is often, Seargeant finds, not even recognized as English by many Japanese (p. 137)). This sort of appropriation will be alarming to those who want English to be an international language, because, used in this idiosyncratic way, English is anything but international. It does not bring people together. Rather, in its distance from English as used by native speakers and many non-native speakers, it ‘act[s] as a further boundary between Japan and the outside world’ (p. 85). In chapter six, Seargeant discusses the similar segregation that is implicated in venues such as British Hills, a gaikokumura that bills itself as ‘an Education and Cultural Resort’ (p. 87), and also at commercial English schools such as Nova. These sorts of enterprises tend to define (and sell) one sort of English as ‘authentic’, and may, as is the case with British Hills,

740 REVIEWS

Reviewed by David Cozy Showa Women’s University, Tokyo E-mail: [email protected] doi:10.1093/applin/amq028 Advance Access published on 13 September 2010

REFERENCE Wolfram, W. and N. Schilling-Estes. 1998. American English: Dialects and Variation. Blackwell.

Ken Hyland and Giuliana Diani (eds): ACADEMIC EVALUATION: REVIEW GENRES IN UNIVERSITY SETTINGS. Palgrave MacMillan, 2009. Largely in response to the writings of Susan Hunston, Ken Hyland, and Geoff Thompson, evaluation has, over the last decade, become the most examined aspect of academic discourse. Several contributors to this volume invoke Hunston and Thompson’s (2000) depiction of evaluation as a ‘broad cover term for the expression of speaker or writer’s attitude or stance towards, viewpoint on, or feelings about the entities or propositions that he or she is talking about’ (2000: 5), although most go on to develop from this ‘broad, cover term’ narrower and more technical distinctions in evaluation, some of considerable


foster the notion that authentic language use is only possible in exotic environments. The underlying ideology affirms once again that ‘English is [. . .] separate from the mainstream of Japanese society’ (p. 104). If an ideology that separates English from the mainstream of Japanese life is actually dominant, and if very few Japanese actually do need English for instrumental reasons, why then do places such as British Hills, and a whole industry based on teaching something called ‘English Conversation’ exist? Why, in spite of the fact that few will need English, do so many Japanese want English? In chapter seven Seargeant considers this question, and comes to understand that, like learners everywhere, Japanese study language for a variety of reasons; the key is, however, that only some of those reasons have to do with what is traditionally defined as linguistic competence. Thus, Seargeant leads his readers to the inescapable conclusion that a model of English focusing exclusively or excessively on the language’s communicative function is insufficient. Proponents of English as a Global Language and English as a Lingua Franca, English educators, policy-makers, and, indeed, anyone concerned with English in Japan must take into account the ideology that is an inescapable component of English in Japan. Many of us have long suspected this. Thanks to the rigor of Seargeant’s study, we now know it.

740 REVIEWS

Reviewed by David Cozy Showa Women’s University, Tokyo E-mail: [email protected] doi:10.1093/applin/amq028 Advance Access published on 13 September 2010

REFERENCE Wolfram, W. and N. Schilling-Estes. 1998. American English: Dialects and Variation. Blackwell.

Ken Hyland and Giuliana Diani (eds): ACADEMIC EVALUATION: REVIEW GENRES IN UNIVERSITY SETTINGS. Palgrave MacMillan, 2009. Largely in response to the writings of Susan Hunston, Ken Hyland, and Geoff Thompson, evaluation has, over the last decade, become the most examined aspect of academic discourse. Several contributors to this volume invoke Hunston and Thompson’s (2000) depiction of evaluation as a ‘broad cover term for the expression of speaker or writer’s attitude or stance towards, viewpoint on, or feelings about the entities or propositions that he or she is talking about’ (2000: 5), although most go on to develop from this ‘broad, cover term’ narrower and more technical distinctions in evaluation, some of considerable


foster the notion that authentic language use is only possible in exotic environments. The underlying ideology affirms once again that ‘English is [. . .] separate from the mainstream of Japanese society’ (p. 104). If an ideology that separates English from the mainstream of Japanese life is actually dominant, and if very few Japanese actually do need English for instrumental reasons, why then do places such as British Hills, and a whole industry based on teaching something called ‘English Conversation’ exist? Why, in spite of the fact that few will need English, do so many Japanese want English? In chapter seven Seargeant considers this question, and comes to understand that, like learners everywhere, Japanese study language for a variety of reasons; the key is, however, that only some of those reasons have to do with what is traditionally defined as linguistic competence. Thus, Seargeant leads his readers to the inescapable conclusion that a model of English focusing exclusively or excessively on the language’s communicative function is insufficient. Proponents of English as a Global Language and English as a Lingua Franca, English educators, policy-makers, and, indeed, anyone concerned with English in Japan must take into account the ideology that is an inescapable component of English in Japan. Many of us have long suspected this. Thanks to the rigor of Seargeant’s study, we now know it.

REVIEWS 741


complexity. The expression of evaluation in academic prose has thus emerged as an area of distinctive—and fashionable—study. The volume itself consists of a helpful scene-setting opening chapter by the editors, followed by 12 further contributions, as many as eight dealing with book reviews (or book review articles), two on blurbs and single chapters on science review articles and literature reviews in PhD theses. This last, by Paul Thompson, seems a little out of place since his texts are not in the public domain in the same way, and it is questionable how evaluative they really are. As a colleague in the School of Education remarked the other day, the dissertation literature reviews she sees largely consist of ‘inert lumps of ill-digested material’. While Thompson does manage to quote some extracts that indeed show the students’ voices coming through, the list of common noun lemmas found in his corpus would suggest much more description than evaluation. A more general feature of the genres selected in this collection is that, with one partial exception, they deal with what would be called in broader educational terms summative rather than formative evaluations. In effect, almost no attention is directed to those semi-occluded review genres that shape the resultant public texts, such as reviews of book proposals or reviews of articles submitted to journals (Fortanet 2008) or to evaluations that affect academic careers, such as letters of recommendation or those solicited as part of tenure and promotion decision-making processes (Chen and Hyon 2007). Although such genres are doubtless harder to access, it can also be argued that they have greater impact, both on the level of the individual and on the level of disciplinary formation and evolution. As for after-the-event book reviews, it remains unclear what influence they may have, except on those more directly involved such as the author of the book and his or her commissioning editor. The only review that I know of that had a decisive impact on the field was Chomsky’s (1959) long review of Skinner’s Verbal Behavior (1957), which essentially killed off behavioral approaches in language acquisition. At the other end of the spectrum, as it were, I do not think any of the reviews I have written has ever been cited. Undoubtedly, strongly positive or negative reviews will have an impact on sales, such as the classic one liner, ‘This book fills a much needed gap’, but on the whole book reviews are of limited—and likely declining—influence on their respective fields in any epistemological or methodological sense. Doubtless, as indeed shown by a number of journals in our field, they provide an early opportunity for junior scholars to show their stuff, and they do ensure that books are read carefully and critically by at least one person, but, I suspect, they are largely viewed as some kind of ‘service to the profession’. I say this because book reviewing is not these days seen as a particularly ‘points-worthy’ activity in terms of academic credibility, as indeed the papers in this volume dealing with book reviews in Spain make clear. The four papers in Part I offer analyses of a wide variety of review genres. In the first, Giannoni shows the movement from endorsement to criticism

742 REVIEWS


across the blurb, the publisher’s description, the book review, and the commissioned review from a specialist in the field. For example, negative evaluators comprised 14 per cent of the total in expert reviews and 8 per cent in book reviews, while, not unexpectedly there were none in publisher’s descriptions or blurbs. This then is a useful overview paper, even if, as on page 22, it is hard to work out where some of the percentages come from. The next paper by Noguchi is particularly valuable for its focus on the reception histories of texts—a largely neglected aspect of studies of academic discourse. She takes the genre of the review article in science and, indeed is able to show that review articles that identify issues and propose theories can establish a sufficiently rich citational history to indicate their part in the processes of knowledge creation. The paper by Thompson has already been discussed. The final paper in this section is by Basturkmen and is one of two devoted to the ‘mini-genre’ of blurbs. Her careful analysis shows that, fulsome praise notwithstanding, the blurb also needs content that will resonate with the values of its readers. The next section, entitled Disciplinary Variation, consists of three papers covering book reviews in various fields: Diani looks at linguistics, history and economics, Hyland and Tse, philosophy and biology, and Groom, history and literary criticism. Their points of departure and arrival are very different, Diani focusing on reporting verbs, Hyland and Tse on putative gender effects, while Groom is interested in NP+of+NP sequences as providing evidence of a link between phraseology and epistemology because, in his estimation, such sequences most often deal with conceptualizations of phenomena such as, ‘Roberts addresses the issue of voice from a different angle’. Diani’s study confirms the generally accepted view that in the humanities and social sciences, ‘discourse’ reporting verbs will predominate, as evidenced, for example, by the fact that in her three disciplines argue is the most frequent reporting verb. In their well-designed and well-written study, Hyland and Tse tackle the elusive issue of whether men and women adopt different strategies in their reviewing behavior. They conclude, ‘However, while gender is an important component of our lived experience, it is unlikely to be a determining influence on our academic writing’ (p. 119). Rather, they hint that seniority or authority will likely turn out to be more influential—a topic then for future research. Part III also contains three papers, this time dealing with cross-linguistic variation. Lore´s Sanz and Moreno and Sua´rez compare English and Spanish book reviews, while Bondi investigates English and Italian book review articles. Both the Spanish–English papers demonstrate the major contribution that Spanish scholars and researchers have been making to our understanding of academic discourse. Both undertake a move analysis of their reviews and then relate this to sections of the review where evaluation (either positive or negative) is more or less likely to occur, and both make use of interesting

REVIEWS 743

Reviewed by John M. Swales The University of Michigan, USA E-mail: [email protected] doi:10.1093/applin/amq030 Advance Access published on 15 September 2010

REFERENCES Chen, R. and S. Hyon. 2007. ‘Faculty evaluation as a genre system,’ Journal of Applied Linguistics 4: 153–84. Chomsky, N. 1959. ‘A review of B. F. Skinner’s Verbal Behavior,’ Language 35: 28–58. Fortanet, I. 2008. ‘Evaluative language in peer review referee reports,’ Journal of English for

Academic Purposes 7: 27–37. Hunston, S. and G. Thompson. 2000. Evaluation in Text: Authorial Stance and the Construction of Discourse. Oxford University Press. Skinner, B. F. 1957. Verbal Behavior. Copley Publishing Group.


and enlightening e-mail responses from specialist informants. Their conclusions fall basically in line with current perceptions that the Anglo reviews attempt a balance of praise and (mitigated) criticism, while the Spanish reviews are more likely to be, using Lore´s Sanz’s term, ‘rehashes’. Bondi’s study of English and Italian is also strong, even if, for some readers, the fact that the Italian extracts are not translated may present an obstacle to full appreciation. One interesting finding is that Anglo writers apparently often adopt agonistic metaphors in their evaluative reporting, while the Italians prefer those concerned with ‘vision’ through their use of verbs concerned with revealing, illuminating, or bringing into light. The final two papers explore evaluation from a diachronic perspective. Gea-Valor and Inigo Ros trace the development of blurbs over time in Penguin Classics, one notable finding being the increasing use of catchphrases and the increasing typographic enhancement of those phrases to make them more prominent. Shaw examines book review data from an economics journal at two time points, taking a sample from 1913 and another from 1993. This is a complex paper, but one that repays close attention. Of particular interest is his analysis of how evaluation is distributed across parts of speech; in both periods, positive evaluation correlates with adjectives and adjectival phrases, while verbs and verb phrases are the preferred vehicle for negative evaluation. This, obviously enough, is a finding that needs further examination and validation. Overall, this is a useful volume, particularly for its main focus on the genre of the book review; indeed, our knowledge of this genre is considerably enhanced by this collection. It turns out, however, that Academic Evaluation has more of a consolidatory rather than innovative character, and so, in this review, I have tried to bring to the reader’s attention, those findings that are worth further consideration in the future, rather than those, such as praise-blame pairings, that we already know something about.

744 REVIEWS

Brian Richards, Michael H. Daller, David D. Malvern, Paul Meara, James Milton and Jeanine Treffers-Daller: VOCABULARY STUDIES IN FIRST AND SECOND LANGUAGE ACQUISITION. Palgrave Macmillan, 2009.


The importance of vocabulary to both language learning and to the study of language itself has become more widely recognized in recent years. Vocabulary Studies in First and Second Language Acquisition provides further evidence of this, with 12 papers investigating a wide variety of issues. These include the lexical nature of parental speech to pre-school children, vocabulary in L1 schooling, the productive vocabulary of, variously, L2 learners, bilinguals and monolingual native speakers, teachers’ explanations of vocabulary in the L2 classroom, vocabulary and academic success, and vocabulary size and L2 proficiency. What links these topics is the fact that all the papers ‘have implications for educational practice and policy’ (p. xiii). The papers are organized by context, with Chapter 1 looking at vocabulary development in the home context, followed by chapters on L1 vocabulary in the early years of schooling, moving on to bilingual education and adult bilingualism, then second language learning and finally foreign language learning. Thus, as one progresses through the book, one gains a sense of the similarities and differences across contexts, the concerns of researchers working in those contexts, and their chosen methods for exploring those concerns. One also notices two recurrent themes: the importance of vocabulary for educational success and lexical diversity. That vocabulary is important to educational success is the motivation for Henrichs and Schoonen’s study (Chapter 1). They investigate correlations between lexical features of parental speech and the vocabulary growth of pre-school Dutch children, with a view to understanding how some children begin school more ready to deal with academic language than others. Dickinson et al. (Chapter 2) present an overview of classroom-based research into L1 vocabulary development in the early years of schooling and its importance to the development of reading skills. They identify four useful general strategies that teachers can adopt and highlight a number of successful intervention programmes. Sealey (Chapter 3) outlines the thinking behind a project encouraging primary school children to use a corpus of texts written for children to help them explore their understanding of lexical concepts. These three studies all deal with L1 contexts; the relationship between vocabulary and educational success is also the concern of Daller and Xue (Chapter 11), who look at correlations between vocabulary knowledge and the academic success of Chinese students studying at a British university. These four papers do not, however, share the same views. Henrichs and Schoonen, and Dickinson et al. see the range of vocabulary used in the early years as

REVIEWS 745


critical to school success. Sealey, however, after noting that there is a small-scale publishing industry devoted to the idea that a large vocabulary leads to success in life and that educational policy in England also subscribes to this view, suggests that corpus research shows that the ability to deploy the most frequent words in all their many uses may be as important. Daller and Xue, meanwhile, do not find significant correlations between measures of lexical diversity and lexical sophistication and academic success, as they expect to. The second theme identified above is lexical diversity, which no fewer than seven of the twelve papers touch on. In addition to the chapters by Henrichs and Schoonen, and Daller and Xue already mentioned, it also features in the papers by Skehan and David, Myles, Rogers, and Rule discussed below, as well as Treffers-Daller’s, Foster’s, and Malvern and Richard’s papers. Treffers-Daller (Chapter 5) looks at the lexical diversity in French of French-dominant bilinguals, non-French dominant bilinguals, and learners of French as an L2. Foster’s paper (Chapter 6) investigates differences between learners of similar proficiency in their home country and studying abroad. Malvern and Richards (Chapter 10) propose a new method of investigating rare word diversity. In one sense, the prominence of lexical diversity could be seen as unsurprising given that two of the editors developed a prominent measure of lexical diversity: D, which takes multiple samples of different sizes from a text and calculates mean type-token ratios for each sample size, the distribution of which is then related to a mathematical model (Malvern et al. 2004). What is noteworthy, however, is finding lexical diversity being used as a general measure of vocabulary knowledge, whereas previously vocabulary size would have often been used in this way, or perhaps some attempt at measuring vocabulary depth. Many of the authors do not make their reasons for choosing to look at lexical diversity clear; however, Henrichs and Schoonen explain their interest as being in gaining insight into how explicit and clear a message can be made. My own view on the reason for this increasing interest in lexical diversity is that it is a result of researchers wishing to look more at learners’ use of vocabulary. As Meara and Olmos Alcoy (2010) discuss, Nation’s (1990) Vocabulary Levels Test has become firmly established as a test of receptive vocabulary size, but there is no equally well-established measure of productive vocabulary size. While work in this area is continuing, it may be that measures of lexical diversity, being readily computable and reportedly valid and reliable, have become the method of choice. In addition to the two themes just discussed, another aspect of the book that stands out is the diversity of the studies included. The volume embraces expository papers, overviews of previous research, exploratory studies, and quasi-experimental research. It also welcomes authors from a range of specialist areas. Indeed, some of the most interesting papers are by researchers who are better known for their work in areas besides vocabulary. Skehan (Chapter 7) presents a meta-analysis of six previous studies into language

746 REVIEWS

Reviewed by Dale Brown Nanzan University, Japan E-mail: [email protected] doi:10.1093/applin/amq032 Advance Access published on 13 September 2010

REFERENCES Malvern, D. D., B. J. Richards, N. Chipere, and P. Durań. 2004. Lexical Diversity and Language Development. Palgrave Macmillan.

Meara, P. and H. Bell. 2001. ‘P_Lex: a simple and effective way of describing the lexical characteristics of short L2 texts,’ Prospect 16: 5–19.


learning tasks. The performances of both native speaker and non-native speaker participants are re-analysed in terms of lexical diversity, measured with D, and lexical sophistication, measured using P-Lex, which divides a text into 10-word chunks and looks at the number of infrequent words in each (Meara and Bell 2001). Skehan provides a number of insights into the two, each of which he links to Levelt’s model of speaking. The paper is a model of organization and clarity. Macaro et al. (Chapter 8) also provide an overview of several previous studies, looking at the code-switching behaviour of Chinese teachers of English when explaining difficult items of vocabulary to learners. They develop the idea of the teacher as a dictionary; actually a kind of super-dictionary, able to provide a variety of types of information on words in either the target language or the L1, choosing the information and the most appropriate way to provide it in response to the needs of the learners and the difficulty of the items concerned. They then describe their initial attempts to find evidence of links between teachers’ choices and vocabulary acquisition itself. Occasionally, however, the book’s diversity seems to be stretched too far, with papers which seem an odd fit for a volume of ‘vocabulary studies’. David et al. (Chapter 9) look at the relationship between morphosyntactic development and lexical development, yet the focus is very much on the former rather than the latter. The results concentrate on the differences between three groups of participants on five different morphosyntactic measures, with the single measure of vocabulary only involved in terms of correlations with those five. Edwards (Chapter 4) discusses the impact of increasing diversity due to the effects of globalization on mainstream education, highlighting a number of problematic issues and attempts at resolving them. Vocabulary is discussed, but it is very much in the background. Vocabulary Studies in First and Second Language Acquisition is on the whole an interesting and insightful book. The range of issues included gives a sense of the various directions that researchers interested in vocabulary are currently pursuing. The book does not provide an overview of vocabulary research, but it is to be recommended as an enlightening snapshot of the field.

REVIEWS 747

Meara, P. M. and J. C. Olmos Alcoy. 2010. ‘Words as species: an alternative approach to estimating productive vocabulary size,’ Reading in a Foreign Language 22: 222–36.

Nation, I. S. P. 1990. Teaching and Learning Vocabulary. Newbury House.


Applied Linguistics: 31/5: 748 doi:10.1093/applin/amq038

ß Oxford University Press 2010 Advance Access published on 22 September 2010

CORRIGENDUM Investigating L2 Performance in Text Chat Shannon Sauro and Bryan Smith Applied Linguistics, 31/4: pp. 554–77 doi:10.1093/applin/amq007


The authors regret that the wrong publisher has been cited for the book Planning and Task Performance in a Second Language in the reference section. The correct publisher is John Benjamins.

Applied Linguistics: 31/5: 749–751 doi:10.1093/applin/amq042

ß Oxford University Press 2010

NOTES ON CONTRIBUTORS Dwight Atkinson teaches at Purdue University, Indiana, USA. His research interests include second language acquisition, second language writing, qualitative research, and the lives of non-traditional students in English-medium universities in India. He is currently editing a volume (to be published by Routledge) entitled Alternative Approaches to Second Language Acquisition. Address for correspondence: English, Purdue University, 324 Heavilon Hall, 500 Oval Dr West Lafayette 47907-2038, Indiana, USA.

Dale Brown teaches at Nanzan University in Nagoya, Japan. His research interests are in extensive reading, the analysis of language learning materials, and all aspects of vocabulary learning and teaching. Address for correspondence: Nanzan University, 18 Yamazato-cho, Showa-ku, Nagoya 466-8673, Japan. Yu-jung Chang is an Assistant Professor in the Department of Foreign Languages and Literature at National Tsing Hua University, Taiwan. She is interested in issues of globalization, English as an international language, and EFL/ESL students’ language learning resources and identities. Chuntien Chen is a businessman and English teacher, recently graduated from the TEFL program at National Tsing Hua University in Taiwan. Vocabulary research is one of his interests regarding second language teaching and learning. Address for correspondence: No.18, Ln. 328, Zhongzheng Rd., North Dist., Hsinchu City 300, Taiwan. David Cozy is an Associate Professor in the Department of English Communication at Showa Women’s University in Tokyo. He has lived and worked in Japan for more than 25 years, and is interested in poetry, modern literature, and how they can be used effectively in language teaching. Address for correspondence: 4-10-21 Higashi Kaigan Kita, Chigasaki, Kanagawa 253-0053, Japan.


Joe Barcroft is Associate Professor of Spanish and Second Language Acquisition and Affiliate Associate Professor of Psychology at Washington University in St. Louis. He is Director of Basic Language Instruction in Spanish, Co-Director of the Graduate Certificate in Language Instruction, and a faculty member in the Program in Linguistics and the Program in Philosophy, Neuroscience, and Psychology at Washington University in St. Louis. His research focuses on second language vocabulary acquisition, input processing, the bilingual mental lexicon, and psycholinguistic approaches to issues in second language acquisition.

750 NOTES ON CONTRIBUTORS

Anna De Fina is Associate Professor of Italian Language and Linguistics at Georgetown University. Her books include Identity in Narrative: An Analysis of Immigrant Discourse (2003, John Benjamins) and the edited volumes Dislocations, Relocations, Narratives of Migration (2005, St Jerome Publishing, with Mike Baynham) and Discourse and Identity (2006, Cambridge University Press, with Deborah Schiffrin and Michael Bamberg). <definaa@ georgetown.edu> John Gray is Senior Lecturer in TESOL/Applied Linguistics at the University of East London, UK, where he is programme leader for the in-service MA ELT and the pre-service MA TESOL. <[email protected]>

Yasuko Kanno is an Associate Professor of TESOL in the College of Education, Temple University. She is interested in how linguistic minority students employ their agency to negotiate their identities and educational opportunities. Kanno is the author of Negotiating Bilingual and Bicultural Identities (2003) and Language and Education in Japan (2008). Kendall A. King is Associate Professor of Second Languages and Cultures at the University of Minnesota. She is an editor of the journal, Language Policy and author of Language Revitalization Processes and Prospects: Quichua in the Ecuadorian Andes (2001). Her recent work appears in the Modern Language Journal, Annual Review of Applied Linguistics and International Journal of Bilingual Education and Bilingualism. Address for correspondence: 228 Piek Hall, 159 Pillsbury Drive SE Minneapolis, MN 55455-0208, USA. Susanne Rott is Associate Professor German and Second Language Acquisition. She directs the Sandy Port Errant Language and Culture Learning Center as well as the Basic German Language Program at the University of Illinois at Chicago. In her research she focuses on the lexical development of second language learners, in particular the partial acquisition of individual word aspects, collocational units, and lexico-grammatical constructions, as well as the impact of instructional interventions on learning. <[email protected]> John M. Swales is Professor Emeritus of Linguistics at the University of Michigan, where he was also Director of the English Language Institute for many years. Although officially retired, he remains active in research and


Natalia D. Jacobsen is a PhD candidate in Applied Linguistics at Georgetown University in Washington DC. Her research interests include the application of cognitive linguistics to second language teaching, issues of identity in second language acquisition, and legal discourse, in particular legal genre patterns and issues of definition in common law. Address for correspondence: Department of Linguistics, Georgetown University, ICC 479, 37th and O Streets, NW, Washington, DC 20057-1051, USA.

NOTES ON CONTRIBUTORS

751

textbook-writing projects. Address for correspondence: 546 Fifth Street, Ann Arbor, MI 48103-4839, USA. <[email protected]>


John Truscott is a Professor in the Center for Teacher Education at National Tsing Hua University in Taiwan. His research interests include the development of a broad processing-oriented framework for the study of (second) language acquisition (MOGUL), the effects of form-focused instruction and feedback on language learning, and second language vocabulary teaching and learning. He has published extensively in the first two of these areas. Address for correspondence: Center for Teacher Education, Education Hall, National Tsing Hua University, Hsinchu, Taiwan 30013. <truscott@ mx.nthu.edu.tw>

Applied Linguistics, Volume 31, issue 5, 2010

Applied Linguistics, Volume 31, issue 1, 2010

Applied Linguistics, volume 31, issue 2 , 2010

Applied Linguistics, Volume 31, Number 3, 2010

Applied Linguistics, Volum1 , Issue 1, 1980

Behavioral and Brain Sciences, Volume 33, Issue 5, October 2010

sports medicine Volume 40 Issue 5 May 2010

ADVANCES IN APPLIED MICROBIOLOGY VOL 31, Volume 31

Advances in Applied Mechanics, Volume 31