Applied Linguistics, Volume 31 [PDF]

Zitiervorschau

APPLIN-31(1)Cover.qxd

1/27/10

3:06 PM

Page 1

APPLI ED LI NGUISTICS

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

Applied Linguistics

Volume 31 Number 1 February 2010

Volume 31 Number 1 February 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

EDITORS 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

REVIEWS AND FORUM EDITOR Stef Slembrouck, Professor of English Linguistics and Discourse Analysis, Universiteit Gent, Vakgroep Engels, Rozier 44, B-9000 Gent, Belgium. 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 Catherine Kneafsey, 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 Zolta´n 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, City University 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 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.

APPLIN-31(1)Cover.qxd

1/27/10

3:06 PM

Page 2

APPLIN-31(1)Cover.qxd

1/27/10

3:06 PM

Page 1

APPLI ED LI NGUISTICS

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

Applied Linguistics

Volume 31 Number 1 February 2010

Volume 31 Number 1 February 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 1 February 2010 CONTENTS Articles Textual Appropriation and Citing Behaviors of University Undergraduates LING SHI Practices of Other-Initiated Repair in the Classrooms of Children with Specific Speech and Language Difficulties JULIE RADFORD Style Shifts among Japanese Learners before and after Study Abroad in Japan: Becoming Active Social Agents in Japanese NORIKO IWASAKI The Relationship between Applied Linguistic Research and Language Policy for Bilingual Education DAVID CASSELS JOHNSON

1

25

45

72

Seizure, Fit or Attack? The Use of Diagnostic Labels by Patients with Epileptic or Non-epileptic Seizures LEENDERT PLUG, BASIL SHARRACK, and MARKUS REUBER English in Advertising: Generic Intertextuality in a Globalizing Media Environment AN H. KUPPENS

115

A Subject–Object Asymmetry in the Comprehension of wh-Questions by Korean Learners of English JIN-HWA LEE

136

REVIEWS F. Christie and J. Martin (eds): Language, Knowledge and Pedagogy: Functional Linguistic and Sociological Perspectives RICHARD BARWELL

156

S. Makoni and A. Pennycook (eds): Disinventing and Reconstituting Languages STEPHEN MAY

159

Alison Wray: Formulaic Language: Pushing the Boundaries PHILIP DURRANT

163

Ruth Wodak, Rudolf de Cillia, Martin Reisigl and Karin Liebhart: The Discursive Construction of National Identity ALEKSANDRA GALASIN´SKA

166

NOTES ON CONTRIBUTORS

169

94

Applied Linguistics 31/1: 1–24 ß Oxford University Press 2008 doi:10.1093/applin/amn045 Advance Access published on 4 December 2008

Textual Appropriation and Citing Behaviors of University Undergraduates

University of British Columbia This article explores the citing behaviors of 16 undergraduates in a North American university. After completing a research paper for their disciplinary courses, each participating student was interviewed to identify in his/her writing words and ideas borrowed from source texts and to explain why and how the relevant texts were appropriated with or without citations. Analysis of students’ writing and comments illustrates how they relied on source texts for various aspects of their essays, some of which they believed required citations while some of which did not. Results showed that they tried to strike a balance between the need to cite published authors to gain credit for the scholarly quality of their writing and the desire to establish their own voice by limiting the extent to which they cited other texts. Some students also reported how they chose between quoting and paraphrasing (though the latter sometimes contained direct copying) on the basis of their ability to rephrase other’s words and their understanding of the different roles played by the two. The study indicates the degree to which citational acts are discursive markings of learning and knowledge construction.

INTERTEXTUALITY One theoretical perspective for research on students’ citing behaviors is the concept of intertextuality which suggests that each academic text is populated with other texts organized to generate new knowledge claims (Fairclough 1992). Researchers have found the concept useful in interpreting how students acquire academic literacy. For example, Starfield states that university students need to learn to interact with various texts which ‘circulate as currency’ (2002: 125) and develop citation skills to accumulate their own ‘textual capital’ (2002: 126). Young and Gaea (1998) suggest that citation skills develop along with one’s disciplinary knowledge and students achieve knowledge transformation by integrating content as interpreted evidence. While working with multiple source texts, many students, however, have produced ‘patchwriting’ by copying from a source text and then deleting or changing a few words and altering the sentence structures, a textual strategy that leads to accusations of plagiarism (Howard 1992, 1995). Such accusations, as Pennycook (1996: 226) states, are ‘inadequate and arrogant’ especially to L2 students from a culture that views borrowing and memorizing others’ words as a legitimate learning strategy. In order to understand issues embedded in students’ citation practices, Chandrasoma et al. (2004: 171)

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

LING SHI

2 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

propose that we should ‘do away with the notion of plagiarism in favour of an understanding of transgressive or nontransgressive intertextuality’.

PREVIOUS RESEARCH A number of researchers compared student-generated texts with their source texts and found that students relied on source texts in their writing (Campbell 1990; Moore 1997; Shi 2004). Among them, Campbell (1990) noted that copying was a major strategy used by both L1 and L2 writers. In two other studies (Moore 1997; Shi 2004), L2 undergraduates were found to rely more than their L1 counterparts on source texts in their summary writing. Close comparisons of students’ texts also revealed that L2 students tended to use implicit attributions, such as ‘it is said’ or ‘the first idea is . . .’, to summarize propositions compared with their L1 peers who tended to mention the individual authors explicitly (Moore 1997; Shi 2004). The varied performances of the participants lead to the question of how students conceptualize textual borrowing or plagiarism.

Inability to identify plagiarized texts To explore students’ perceptions of plagiarism, a second strand of research explored students’ abilities to identify plagiarized texts by using questionnaires that contained examples of edited rewritten versions with their sources (McCormick 1989; Deckert 1993; Roig 1997; Chandrasegaran 2000). The edited rewritten texts contained what researchers believed to be either models of correct use of sources or texts that were copied or paraphrased with or without acknowledgement. For example, McCormick (1989) included three rewritten samples that had direct copying of phrases of 7, 13, and 15 words, respectively, without quotation marks. Together, these questionnaire surveys revealed that many students, either with L1 (McCormick 1989; Roig 1997) or L2 or a non-western background (Deckert 1993; Chandrasegaran 2000), had problems distinguishing the ‘plagiarized’ texts. Gray areas surrounding plagiarism concern the degree of reformulation or paraphrasing. Students, especially those from an L2 culture, are sensitive to the complexities of plagiarism (Pennycook 1994).

Nontransgressive plagiarism A third strand of research used students’ comments to clarify their inappropriate textual borrowing or plagiarized texts identified by researchers. These studies could be further distinguished based on whether the analysis focused more on students’ texts (Pecorari 2003, 2006) or interview data (Hull and Rose 1989; Dong 1996; Spack 1997; Currie 1998; Ange´lil-Carter 2000; Starfield 2002; Chandrasoma et al. 2004; Petric´ 2004; Thompson 2005). Those that

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

Reliance on source texts

L. SHI

3

The present study The above review suggests that researchers that used both textual and interview data have provided more insightful information since citational acts are subjective and private. However, these studies have either provided a limited description of students’ text in terms of quantities and weightings of various types of use of citations or textual appropriation, or solicited students’ comments based on seemingly plagiarized texts identified by the researcher who

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

focused more on students’ voices from interviews were case studies. The rich interview data in these case studies suggest evidence of unintentional plagiarism. Students were reported to defend their unacknowledged use of source information in terms of how they understood common knowledge (Chandrasoma et al. 2004) or relied on memorized and internalized knowledge (Petric´ 2004). There was also evidence that inappropriate use of citations was tied to students’ confusion of how to cite (Thompson 2005), unsuccessful attempts to develop authorial selves (Starfield 2002), underdeveloped skills of reading comprehension as well as critical thinking in relation to the authors’ stance (Spack 1997; Ange´lil-Carter 2000), and limited content knowledge that hindered them from selecting relevant and important references (Dong 1996; Chandrasoma et al. 2004). Being primarily concerned with how their teachers would judge their work, some students actually found patchwriting an efficient strategy for survival (e.g. Hull and Rose 1989; Currie 1998; Chandrasoma et al. 2004; Petric´ 2004). These findings suggest that plagiarism is a problem of academic literacy rather than that of dishonesty (Ange´lil-Carter 2000). The aforementioned case studies, however, provided limited or a less comprehensive description of students’ texts. Recognizing this limitation, Pecorari (2003, 2006) analyzed excerpts of 17 graduate students’ thesis writing in different content areas. Based on comparisons of passages from student texts and the cited sources, Pecorari (2003) found that students’ citations were not transparent as most of them had passages in which 50 per cent or more of the words were from the sources without acknowledgment or quotations. However, like those in the case studies, these students indicated no intention to deceive at the interview. The interview data revealed individual student writers’ understanding of the nature of source use. Citation is, as Pecorari (2003: 324) put it, ‘an occluded feature of academic writing’ because the ‘real nature of source use is only known to the writer’. Further exploring these students’ occluded use of citations, Pecorari (2006) identified that about 18 per cent of the total source mentions in students’ texts were actually from secondary sources without acknowledging the original author. Uncertain about such practice, one student said she relied on her advisor for guidance. However, since such occluded features were a blind spot for readers, Pecorari (2006) expressed a concern about how these students might graduate with the wrong impression that their citation practices were in compliance with the standard practice.

4 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

(i) Why do students appropriate source texts and cite them in their writing? (ii) Why do students appropriate source texts but not cite them in their writing? (iii) How do students apply principles of textual borrowing to quotations, paraphrases or summaries?

METHOD Participants Sixteen undergraduates in a North American university responded to an advertisement and volunteered to participate in the study. Of the 16 participants, 4 were science majors and the rest were in Arts and Social Science (Table 1). The initial letters of the pseudonyms indicate the first language of the students: ‘E’ for English (Elmer, Edward, and Eddy), ‘C’ for Cantonese (Carmen, Carl, Candy, Carol, Cathy, and Cary), ‘M’ for Mandarin (Martin, Mark, and May), ‘J’ for Japanese (Jane), ‘K’ for Korean (Kate), ‘R’ for Romanian (Rose), and ‘P’ for Polish (Polly). Participants who regarded English as their second language, including Carmen and Carl who were born in North America, all spoke their first language at home. Among them, Jane, Kate, and Rose had just arrived as international students and were literate in their home language. The rest were not literate in their first language as they had all or most of their schooling in local North American schools. These students, unlike those who were literate in their first language and might therefore bring different ideas about citation practices from their home cultures, were more like their North American born monolingual peers in terms of learning to cite, a difficult task for all university students.

Writing assignments Participants were asked to bring to the interview a research paper they had just written for a course as well as the source texts they used (see Appendix A

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

examined passages of students’ texts that contained citations. In the latter case, portions of texts were excluded from the analyses when they ‘included no citations’ (Pecorari 2003: 322). What is missing here is a great amount of textual borrowing not indicated by citations. Since textual borrowing is a subjective act guided by, what Pecorari (2006) has noted, students’ own inferred rules, research needs to explore how students identify their own textual borrowing and explain why they make the borrowing and whether each borrowing is cited or not cited. Such research could also symmetrically compare students’ views in terms of how they apply textual borrowing or citations to quotations, paraphrases or summaries. To tap these issues, the present study uses undergraduates’ self-reflections to explore three research questions:

L. SHI

5

Table 1: Participants’ profiles Major

Year of study at the participating university

L1

Age

Elmer

Commerce

1

English

18

Edward

Commerce

1

English

18

Eddy

History

3

English

29

Carmen

Science

1

Cantonese

19

Carl

Arts

2

Cantonese

19

Candy Carol Cathy Cary Martin Mark May Polly Jane Kate Rose

Science Arts Arts Science Commerce Science Arts Economics Law English Literature Political Science

1 1 1 1 3 1 2 2 Exchange Exchange 2

Cantonese Cantonese Cantonese Cantonese Mandarin Mandarin Mandarin Polish Japanese Korean Romanian

18 17 18 18 20 18 19 25 23 21 19

Age arriving

Born in North America Born in North America Born in North America Born in North America Born in North America 7 3 3 10 10 12 9 7 23 21 18

in the supplementary data for the online version of the article). Depending on the courses they were taking at the time, most students brought their research papers for 100 level (first year) courses in English (8), history (1), film studies (1), and biology (2). The rest brought their research papers for 200 or 300 level (second or third year) courses in political science (3) or women’s studies (1). The average length of these papers was 1,805 words. The course handouts that explained the assignments all had a warning against plagiarism and directed students to the university or faculty websites on plagiarism. In other words, the participants were aware of the importance of attributing appropriated materials in their research paper assignments.

Interviews and data analysis I met the 16 students separately and conducted discourse-based interviews (Odell et al. 1983). Each interview lasted for an hour during which the student

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

ID

6 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

was asked to compare his/her own text with the source texts to identify and comment on texts appropriated, whether cited or not cited (see Appendix B in the supplementary data for the online version of the article for interview questions). Since citations without quotation marks signal the borrowed content whereas those with quotation marks signal the borrowed language in which content is presented, I also solicited students’ comments on their decisions of using quotes, paraphrases, or summaries for texts appropriated. The term ‘plagiarism’, though mentioned by some students, was carefully not used by me so as not to bias students’ views. Based on the interview data, a total of 187 units of textual appropriation were identified together with students’ comments. There were occasions when students identified units of textual borrowing but did not make any comments. Such units were excluded as the lack of explanation from students in these cases might be a significant problem worthy of another study. There is no consensus on how much language must be copied to be deemed plagiarism. For example, students have been advised to cite when copying a string with a minimum of three words from a source text (e.g. Hodges 1962). For the present study, a unit of textual appropriation was defined as a sentence or several sentences that contained words or ideas borrowed from source texts. The longest unit in the present data contained seven sentences with a total of 164 words. The boundaries of the units were set by students themselves who connected each unit to one specific borrowing and citation decision. After repeated readings of students’ explanations, a coding scheme was developed based on some key words used by students (e.g. support; new information). The purpose of using students’ own language was to be more accurate about students’ perspectives. This approach would also help produce ‘consensual readings’ of the narrative data (Denzin 1997: 232) and, therefore, catch ‘both the variation and central tendency or typicality’ (Watson-Gegeo 1988: 585). To check intercoder reliability, a research assistant was trained to use the scheme and then coded 10 per cent of the data on her own. The agreement between the researcher and the research assistant reached 84 per cent. The discrepancies were then solved by revising some of the categories. Table 2 presents the revised coding scheme. Fourteen reasons were identified under three major factors that influenced students’ use of source texts and citation decisions. First, the participants were concerned about the functional or rhetorical role of the borrowed texts in terms of whether it could provide support, help develop or form one’s own idea or form the basis of a key point. Such concerns demonstrate that the present students, unlike those in previous research (e.g. Chandrasegaran 2000), did have an understanding of using textual borrowing rhetorically to legitimize their own claims. In addition, the participants were making interpretations of the source information to determine whether it was new information, fact, research finding, background information, common knowledge, or information from a credible source. These reasons indicate that, apart from background information that was identified in previous research (Pecorari 2003), there were various other inferences

Form one’s own idea

Key point

2

3

New information

Fact

Research finding

Background information

5

6

7

Background information or established theory in the field

Previous research findings

Facts or events

New information for the writer

To form the basis of one’s key point

To build up one’s own idea

To provide support for one’s point

Definitions

I just took all the information and put it in my own words. . . . I just cite the author because I didn’t know any of the stuff really before. (Edward) This is just like facts . . . so I put the reference here. . . . I am summarizing her points. (Edward) I cite that, saying that who did the research and what they found. (Cary) I got that from this article. . . . It’s just like the background information. (Edward) I rephrased these from the article. . . . I mean it’s . . . fairly well established theory . . . That’s why I kind of cited the textbook. (Polly)

Quote

These are all quotes from their websites. I used them to support what I wrote here. (Polly) It’s part of my own argument . . . I mean, I took it from their argument but incorporated it into my own . . . overarching description of the situation. (Eddy) I used that [citation] because it forms the basis of my argument. (Elmer)

(continued)

Paraphrase

Summary

Summary

Summary

Formatb

Examplesa

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

4

Interpretations of source text

Support

1

Functions

Explanations

Table 2: Coding of students’ explanations and examples

L. SHI 7

Common knowledge or term

9

Result of learning

Reference cited earlier in the text or in the reference list

No need to cite everything

Teachers’ preference

11

12

13

14

Teacher prefers citations

Not everything needs to be cited

A source cited earlier or in the reference list

Knowledge accumulated by learning

Words and ideas directly taken from the source

Quote

These are the actual words [from the source]. You have to quote them, right? (Edward) And this is kind of I just knew from my mum . . . and from previous school . . . I don’t think I really have to cite it. (Cary) I used my own words and it’s a summary. I didn’t cite because I already mentioned it in the first paragraph. (Carol) I don’t think I was supposed to cite this. We just indicate the book that we used at the end. (Cary) Or else I would have to cite every single sentence. I think it’s kind of distracting. (Candy) They (teachers) prefer you rephrase and cite it. (Cary)

b

Paraphrase

Summary

Paraphrase

Formatb

This is a citation from a book written by a concert artist. . . . I think he’s very credible. . . . It’s paraphrased here. (Cathy) I didn’t cite it because that’s common sense. (Mark) I didn’t cite it because it’s an actual term that people use. (Martin)

Examplesa

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

Keywords that result in relevant coding are highlighted. Coding of the example comments in terms of quote, paraphrase, and summary when relevant information is available.

a

Other’s words/ideas

10

Reasons related to learning

Credible or reliable source

Credible source

8

Knowledge or words commonly known

Definitions

Explanations

Table 2: Continued

8 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

L. SHI

9

FINDINGS AND DISCUSSION Since students’ decisions were based on more than one reason for some units, the 187 units of textual appropriation generated a total of 236 explanations of why and how source texts were used (Table 3). Over one-third of the explanations given (85) were related to the functions served by using the source texts, about one-third (77) were related to the interpretations of previous works, and slightly under one-third (74) were related to reasons of learning to cite. Of the 236 explanations provided, 159 (67 per cent) illustrate why students ended up citing the work. Students also explained how they applied the principle of textual borrowing by quoting, paraphrasing, or summarizing the source texts a total of 122 (out of 236) times. I will analyze how participants used source texts, focusing on their understanding of whether citing was required or not.

Reasons mentioned more frequently for citing texts borrowed The present students, as a group, seemed to mention some reasons for taking something from the source text more frequently for citing and others more frequently for not citing. The only reason that did not favor either citing or not citing was background information that was mentioned three times for citing and three times for not citing. Figure 1 presents the eight reasons mentioned more frequently by students to explain why a unit of textual borrowing was acknowledged.

Securing support for one’s writing Support was the most frequently mentioned reason (48) by students for using source texts with citations (Figure 1). The two occasions when it was mentioned for not citing were double coded with reasons of no need to cite everything and result of learning that typically led to no citations. Among those referring to support as the main reason for citing, many commented on how they searched for materials to cite from various published authors to support their arguments.

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

that the present students made about which type of source information needed to be borrowed with or without citations. Finally, the participants described how their learning experiences influenced their use of source texts and citations. Such experiences included how they identified words of others, wrote from knowledge accumulated as result of learning, followed teachers’ preference, and understood that there was no need to cite everything including references that were cited earlier in the text or in the reference list. Table 2 shows that the relevant mentions were also coded in terms of a quote, paraphrase or summary1 when information was available. The fact that the relevant information was not available for some units will be discussed in the findings section.

10

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

Table 3: Classification of reasons students gave for using source text, by whether source is cited or not Explanation

Cited

Not cited

Total

No. %

No.

No.

%

%

Functions 1 Support 2 Form one’s own point 3 Key point Subtotal and percentage of 236

48 96 2 7 30 16 12 100 0 67 79 18

4 70 0 21

50 23 12 85

100 100 100 100

33 6 10 49

18 82 4 9 64 5 5 100 0 3 50 3 15 100 0 0 0 15 50 65 27

18 36 0 50 0 100 35

22 14 5 6 15 15 77

100 100 100 100 100 100 100

12 7 1 4 7 2 33

37 100 0 0 0 18 0 0 3

0 100 100

37 100 18 100 3 100

32 1 1

0 0 11 5 100 0 42 57 32

100 0 43

11 100 5 100 74 100

3 3 40

Interpretations of source text 4 New information 5 Fact 6 Research finding 7 Background information 8 Credible source 9 Common knowledge or term Subtotal and percentage of 236 Reasons related to learning 10 Other’s words/ideas 11 Result of learning 12 Reference cited earlier or in the reference list 13 No need to cite everything 14 Teachers’ preference Subtotal and percentage of 236 Total

159

67 77

33

236 100 122

a

Students mentioned in 236 instances how they relied on source texts. Among them, 122 were accompanied with explanations of whether they used source texts by quoting, paraphrasing, or summarizing.

The following is a typical example of how Jane took pains to find something to cite.2 Example 1 A unit of textual appropriation in Jane’s paper titled ‘Domination of the Liberal Democratic Party in Japanese Politics’: Keeping the dominance in the diet [parliament] made LDP [Liberal Democratic Party] able to influence the policies dramatically.

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

No. of mentions of format of textual borrowinga

L. SHI 11

40 30

With citing

20

Without citing

10 0 rt

o pp

Su

y Ke

e as nc de i / re s e fi f rd le re ch wo ar 'p dib s e s e r ' s er Cr he Re th ac O e T

t in po

w Ne

ct Fa

ion at m r fo in

e

ng

c ur so

i nd

Figure 1: Leading reasons given for use of source text, with citing

Frequency of mentions

50 40 30

With citing Without citing

20 10

g

t fe r

ev er

en

yt

ce

hi n

lis

g in rn

ci te

re e

to

th

ed ne o N

R

ef er

en

ce

ci

te

d

ea

rli er

or

C

in

R

om

es

m

ul

on

to

kn

fl

ow

n ow s e' on m Fo r

ea

le

po

dg

in

t

e

0

Figure 2: Leading reasons given for use of source text, when not citing

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

Frequency of mentions

50

12

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

The dominant party is generally able to get its own way in policymaking issues in the parliament. (Stockwin 1998: 121) Jane’s reason for citing, which has been classified as an instance of support:

The fatigue Jane expressed at the end shows the enormous time and energy she had spent searching for citations that would support her basic intuition about parliamentary influence. Jane’s comment suggests that she had learned to both trust and check her insights, even as she realized that further evidence beyond her own hunches needed to be found and included in her work. This contrasts markedly with the participant in Thompson (2005) who felt that it was necessary to always find a support because, as a student, he had nothing original to contribute, Jane seemed to have a scholarly sense of following one’s hunches, with enough confidence that one did not give up until either finding the evidence or realizing it was not there. The students’ interest in securing support from source texts was also evident when in 12 instances participants commented on how they presented a key point through the voice of a cited author. For example, in explaining how she argued for the importance of parents’ encouragement for kids learning to play the piano, Cathy said that she decided to use the citation ‘to show it was an important factor’. The reliance on source texts to frame important points was also shown in five instances of students citing research findings and 15 instances of drawing information from a credible source. Martin believed that citations ‘gave some authenticity or authority over [his] essay’. Like Martin, Carol explained that she cited from Milroy and Milroy (1999) about how English was an essential tool for working-class children because ‘quotes or citations . . . [had] been proven and recognized’. By using such a citation, Carol said her message was that ‘[i]t is not just what I am saying, it is from a published author’. These comments reveal how students draw on others to show that they are not alone in thinking and thus to gain credit for their own writing.

Learning to cite Students were also motivated to cite materials for two additional reasons related to what I am terming learning (Figure 1). As learners, the participants chose to cite when they recalled teachers’ preference for extensive citation (5) or

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

This is something else that I got from the textbook . . . I have to find what was said previously to back up my writing. So that’s why I keep putting in references. . . . I had this idea that if you have dominance in the Diet, you must have big influence in the parliament. I just assume that it should be the case but I was just worried because it might not be the case. So I just went through all the books trying to find if somebody was actually saying the same thing. . . . So those are the little things that I went through and made me so tired.

L. SHI 13

when they saw source texts as others’ words and ideas that were worth quoting directly (37). In the following example, Edward explained how he attributed the ownership of the word ‘localization’ to an author (the co-authors of the original text were missing in both his writing and comment): Example 2

In his essay ‘‘Drug Dealers,’’ Caulkins describes the ‘‘localization’’ of certain areas of the city for drug use as a major problem leading to the augmented organization of drug dealers in Vancouver (Caulkins 326). Edward’s reason for quoting, which has been classified as an instance of other’s words: I used quotations for ‘‘localization’’ because it’s a word Caulkins used. It is his word. . . . He describes the ‘‘localization’’ of certain areas of the city for drug use . . . I usually cite if it’s something that most people wouldn’t use. . . . Just like if I was to hear ‘‘localization’’ before, I probably wouldn’t know it means that much before reading. It is interesting to note that Edward felt that the original author owned the word because any average person like himself would not have used the word in such a context. Also regarding themselves as average people, Martin and Mark said that citing words from published authors would save them the trouble of providing supporting details. As a student, Martin said that instructors always cast doubt on ideas he claimed as his own. For example, in explaining why he cited Hirsch (1990) when writing about Woody Allen’s family, Martin said ‘If I didn’t cite that, the instructor would say, ‘‘Where did you get this? Who said this? How did you know who said this?’’’ In comparison, Mark acknowledged the sources when ‘the language is powerful’ because, as he explained, ‘If you just present something really stunning people won’t believe you’. Mark’s lack of confidence in using his own words echoed that of L2 students in previous studies (e.g. Ange´lil-Carter 2000; Chandrasoma et al. 2004).

Learning ‘facts’ and ‘new information’ in various academic disciplines Compared with the above reasons, students were divided on whether fact and new information required citing, although they favored citing (Figure 1). The participants distinguished facts (things exist or performed) from ideas and opinions. For example, Edward interpreted chemical characteristics of drugs as facts that had to be cited, whereas Eddy regarded current or historical events as facts that needed no acknowledgement. The different decisions made by students imply that the debt owed to authors for facts differs across

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

A unit of textual appropriation in Edward’s paper titled ‘Nowhere to Turn: Drug Abuse as a Cause of Homelessness and Poverty in Vancouver’:

14

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

 ‘Influences from early in our lives such as childhood trauma . . . or alcohol abuse can contribute to initial experimentation with a drug’ [a citation given by Edward and the material to which the citation applied came from Jones (1998)];  ‘The most significant side effects of antibiotics, however, is [sic] the depression of the immune system’ [a citation given by Mark and the material to which the citation applied came from Hauser and Reminngton (1982)]; and  that the reformed Indian Constitution ‘articulated the principle of equality of all citizens irrespective of caste, community, race or sex’ [a citation given by Rose and the material to which the citation applied came from Desai (1973)]. One can easily imagine that, on the one hand, some of these students’ peers might not consider such information as new knowledge and, on the other hand, the same students might choose not to cite later when they become acquainted with such information. By comparing students’ citing behaviors, the present study suggests a dynamic process of learning and claiming ownership of knowledge. Judgements on textual appropriation, whether appreciated or illegitimate, are grounded in the context of epistemologically and socially constructed academic literacies.

Reasons mentioned more frequently for not citing texts borrowed Since all of the reasons were offered to explain chunks of texts which the students identified as appropriated, a citation would be expected in every case. It is, therefore, highly significant to examine cases when students did not give a citation. Of the 14 reasons for using source texts, five were mentioned more often for not citing than citing (Figure 2).

Interpreting source information as common knowledge/term In 15 instances, the present students interpreted source texts as common knowledge/term and, therefore, did not cite them in their writing (Figure 2). For example, Cathy said that she did not cite the statement that ‘musical success is an innate gift’ which she read from a book because it was ‘something that [she had] grown up with’. Other appropriated texts that students

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

disciplines. If facts about drugs are derived from scientific research, facts of current or historical events are public knowledge. Students’ citing behaviors are thus connected to their variable learning and practices in different disciplinary communities. Like the notion of fact, the concept of new information reveals the role of individual learning experiences. For example, the present students cited such new information as:

L. SHI 15

Example 3 A unit of textual appropriation in May’s paper titled ‘Starbucks Coffee’: Moreover, Starbucks prepare the coffee beans in a sanitary and safe environment that keeps the coffee fresh. May’s reason for not citing, which has been classified as an instance of common knowledge: I got this from the book. . . . But actually it’s kind of common sense because you have to put food in a bag or something so it doesn’t go stale. I think it’s just kind of related to all the food. Compared with May and those students who distinguished common knowledge based on their own understandings and interpretations, other students made similar decisions based on whether the source information was introduced in a textbook (Eddy) or a lecture (Mark), or frequently mentioned in reading materials (Jane). Jane, for example, did not cite the appropriated text about the dominance of the Liberal Democratic Party in Japan because, in her words, ‘I read them in so many places and textbooks so I thought that was just like a general understanding of this whole issue.’ Such comments illustrate that when certain information is recognized by students as public or owned by many, it becomes base knowledge of which students would claim ownership. Although previous research has also reported how one student did not cite common knowledge acquired through personal experiences (Chandrasoma et al. 2004), the present study provides insights of how a group of students across disciplines arrive at such an interpretation in each specific case.

Learning not to cite Students also attributed some of their decisions not to cite to various learning experiences (Figure 2). Specifically, they chose not to cite when the appropriated texts matched their knowledge accumulated as a result of learning (18) or was something they believed not worth citing otherwise they had to cite everything (11). To avoid using too many citations, three students (Carol, Cary, and Martin) said that they did not cite a reference cited earlier in the paper or included

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

considered common knowledge included statements that ‘the industrial policy [of the government] is the driving force for economic growth [of Japan]’ (Kate); that ‘[t]he baby boom that took place from the mid 1940s throughout France had created, by the middle of the 1960s, many thousands of new undergraduates’ (Carl); and that ‘the human use of antibiotics helped the natural selection of bacteria’ (Mark). The following example illustrates how May generalized Starbucks’ way of preparing coffee to a common sense of how to keep food fresh:

16

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

in the reference list. As an example, Cary referred to her learning experiences when commenting on the following unit of textual appropriation: Example 4

However, spontaneous mutation occurs more frequently than induced because spontaneous mutations can occur simply due to natural radiation, and during replication in DNA. Cary’s reason for not citing, which has been classified as instances of the result of learning and no need to cite everything: Well, I didn’t cite these because I know that there is natural radiation. . . . I got it from lecture and the books I used. . . . I learned since I was in high school. . . . I don’t think we really have to cite everything, right? Although the above comment might be interpreted as common knowledge, it is coded as result of learning based on Cary’s own account of how she learned the information in high school. Like Cary, Eddy explained that he did not cite the term ‘unitary actors’ because it was ‘a part of lingo of political science [he had] been studying for a number of years’. Also commenting on the development of one’s own knowledge based on learning, Carol said, ‘When you read an article, you absorb the ideas then it becomes your own’. The participants believed that they were entitled to claim ownership of words and ideas learnt previously or internalized. As two other students explained explicitly when commenting on certain texts appropriated, ‘I’ve learned this in class so it is considered personal knowledge’ (Carmen); ‘I know it before then I don’t think I really have to cite it’ (Cary). These explanations for not citing illustrate how citation practices go hand in hand with students’ learning and accumulation of knowledge.

Claiming authorship of ideas and learning to construct knowledge Compared with the above reasons, the reason to use course texts to form one’s own point was bidirectional (for both citing and not citing) though it was mentioned more often for not citing than citing (Figure 2). Among those who mentioned how they appropriated texts to form their own points, Carmen, in his Biology research paper, explained how he came up with the statement about ‘a direct relationship between light intensity and accumulation of newrosecretions’ based on the information from Cymborosky (1983) that ‘accumulation of neurosecretions is the highest during the night when crickets are supported to move at the highest levels’. In Carmen’s words, ‘I came up with this based on what this person found. . . . So it was what I inferred from the study. This was what I came up with myself’. Although it was arguable

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

A unit of textual appropriation in Cary’s paper titled ‘One Type of Spontaneous Mutations is Point Mutation’:

L. SHI 17

Example 5 A unit of textual appropriation in Cary’s paper titled ‘One Type of Spontaneous Mutations is Point Mutation’: Point mutation also divides into different categories that include base-pair substitution in the DNA, and insertion of deletion in a base pair. Base pair mutation happens when there is a replacement of one nucleotide in a chain of amino acids. . . . Cary’s reasons for not citing, which have been classified as instances of forming one’s own point and result of learning: I know it from a course last term. I just vaguely remember it. I think it’s mine, the idea and wording. I think it’s just my idea. The above examples illustrate how students might make different citation decisions when appropriating source texts to form their own points. If students’ citation decisions are based on various combinations of reasons related to their complex process of learning and constructing knowledge, the present findings highlight the important role of self-reflections in exploring the subjective act of citing.

Quotations, paraphrases and summaries A total of 122 (out of 236) mentions of reasons were accompanied with students’ explanations of whether the relevant units of textual appropriations were presented as quotes, paraphrases or summaries. Most of these identifications referred to texts cited (Table 4). Of the 122 mentions, only 14 (3 for paraphrases, 11 for summaries) were for texts not cited, which was about 18 per cent of total mentions (77, Table 3) for not citing. In addition, probably because quotations were visible, about 86 (out of 122) mentions were for textual borrowing quoted, leaving only 36 for texts either paraphrased (17) or summarized (19).

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

how different the two statements were, Carmen believed that, as a legitimate process of learning and constructing knowledge, he was entitled to take the ownership of the inference he made. Carmen’s choice also reveals, as one anonymous reviewer of the paper noted by citing from Sinclair (1986) and Tadros (1993), an intention to aver or to put forward a claim on the basis of one’s own authority. Unlike Carmen, other students mentioned how they appropriated texts to form their own points together with other reasons for either citing or not citing. The relevant cases further illustrate how students support or claim their ownership of ideas in the context of learning. For example, Mark cited a reliable or credible source to support his understanding that antibiotics can be harmful. In comparison, Cary, in the following example, claimed ownership of the knowledge of point mutation in DNA as a result of learning and, therefore, did not cite the source:

18

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

Table 4: Citation types, format of textual borrowing, and students’ explanations Citation types

No. of units with explanations of format

Summary

Quote Subtotal

Cited Not Cited Not Cited cited cited Functions 1 Support 2 Form one’s own point 3 Key point Subtotal

3 1

1

2 6

1 2

3

3 1 1

3

29 2

33 6

17 17

50 23

3

7 38

10 49

2 36

12 85

2

6 4

1

1

12 7 1 4

10 7 4 2

22 14 5 6

5

7 2

8 13

15 15

16

33

44

77

32

32

5

37

1

1 1

17 2

18 3

3

3

8

11

3

2

5

Interpretations of source text 4 5 6 7

New information Fact Research finding Background information 8 Credible source 9 Common knowledge or term Subtotal

1

1

1

1 1

5

2

1 6

4

Reasons related to learning 10 Other’s words/ideas 11 Result of learning 12 Reference cited earlier or in the reference list 13 No need to cite everything 14 Teachers’ 3 preference Subtotal 3 Total

14

1

1 3

8

4

32

40

34

74

11

86

122

114

236

a Of the 236 units or instances in which students explained how they relied on source texts, 122 were accompanied with explanations of whether they used source texts by quoting, paraphrasing, or summarizing, whereas 114 were left unexplained.

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

Paraphrase

No. of units Total with no explanations of formata

L. SHI 19

Limited identifications for formats of textual borrowing

Understanding the roles of summaries, quotes and paraphrases In instances when students did identify the format of their textual borrowing, some described how they arrived at such decisions. Students’ accounts suggest that while summaries were used to report a large amount of information concisely, quotes and paraphrases seemed to play different roles to individual students. For example, some students said that quoting was chosen over paraphrasing when they ‘did not know how to paraphrase it’ (Carl) or could not ‘think of another way to say it’ (Cathy). In comparison, when explaining their choices of paraphrasing over quoting, Rose and Carmen said that paraphrasing was the skill preferred at the university compared with quoting which they learned to use at high school. Two other students, Elmer and Carol, said that they chose to use paraphrases when citing secondary sources because they did not know if they could use quotations for such information. In the

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

The limited identifications for the formats of textual borrowing not cited or as paraphrases or summaries are worth noting. During the interviews, when prompted to identify how they appropriated the source texts, some students failed to do so with typical responses, such as ‘I remember reading about that but I don’t remember whether I rephrased it or not’ (Polly), or ‘The idea is from the source but I don’t remember I used the words or not’ (Carol). Martin said he could not specify whether an appropriated text about Martin Scorsese and Woody Allen’s childhood was paraphrased or summarized because he ‘got the idea from different books’. Such responses or excuses could suggest students’ hesitations in talking to the researcher about how they missed the citations at the interview, an indication of the epistemological status of the data. One might also think that these students need to upgrade their note-taking skills so as to record source information accurately (e.g. Pecorari 2003). However, a close reading of students’ comments revealed that their inability to remember the source was frequently mentioned together with reasons for using source texts to present result of learning and common knowledge or term which typically led to no citations. For example, arguing that it was impossible and unnecessary to remember and cite all source information, Eddy said, ‘I guess my vocabulary was applied to IR (International Relationship). It’s part of me now. Obviously it was something that I picked up through a lecturer, through experts, but I couldn’t tell you where I got that because it’s just part of my vocabulary.’ Eddy’s comment implies that while some borrowed texts could be accurately cited by improving the note taking process, others are appropriated without acknowledgement deliberately by students who choose to establish their own voice by claiming ownership of the relevant knowledge or achievement of academic literacies.

20

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

following account, Polly said she decided to paraphrase because she believed that quoting was not appropriate at the beginning of the paper: Example 6 Polly’s reason for paraphrasing a unit of textual borrowing

Concern for plagiarism While explaining their choices between paraphrasing and quoting, some students expressed a concern for plagiarism or an uncertainty about whether they had paraphrased some source texts enough. Although they believed that quoting was irrelevant when they changed the source text slightly, these students were uncomfortable when they noted a close resemblance between the source texts and their writing. As an example, the following is a comparison of the source text and the unit of textual borrowing that Polly commented on in the previous quote (words identical in Polly’s text and source text are highlighted): Example 6 (continued) A unit of textual appropriation in Polly’s paper titled ‘GM Crops or Franken Foods: An analysis of the Rhetoric of Activis’: The first successful cross-species gene transfer to plants took place in the early 1980’s, and by 1998 more than 26 percent of the cotton and 40 percent of the soybean acreage in the US was planted with GM crops. (Shields 2000) The corresponding source text from Shields (2000: 18): The first successful gene transfers in plants took place in the early 1980’s, and by 1998 more than 26 percent of the cotton and 40 percent of the soybean acreage in the US was planted with GM crops containing a gene for herbicide resistance. Noticing the direct copying in their writing such as the above, Polly and other students (e.g. Carmen, Eddy, Mark), wondered whether they should have quoted the source texts. In the following comment, Eddy also expressed uncertainty about whether he should have acknowledged the source when summarizing the information about how an Iraqi citizen would be eliminated if he/she openly criticized Saddam Hussein: I got my idea from The Economist. . . . I mean, I took it from their argument but incorporated it into my own . . . overarching

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

I rephrased these from the article. . . . I think I maybe copied that word for word and then rephrased it. I changed the wording around a bit and then cited it. Because it’s really early in the paper, and I figure it would have looked really silly to start quoting . . . It doesn’t look as nice, I think. It’s really a general introduction and background.

L. SHI 21

description of the situation. . . . And I guess that’s sort of a borderline. If you really want to be strict, then I probably should have quoted or cited that. . . . Use the judgment call. ‘Am I plagiarizing or not?’ Maybe I should have cited that.

CONCLUSION The present study illustrates the sophistication and range of the participating students’ understanding of the role of textual borrowing and citation by analyzing students’ self-reflections on not only units of textual appropriation cited but also those not cited. Results show that the citing behaviors of novice scholarly writers (in the case of undergraduates) are guided by a complex set of factors including functional uses of cited works, citers’ interpretations of source texts, a learning process to accumulate one’s own knowledge and textual capital, as well as a choice between quoting and paraphrasing. The study indicates the extent to which students attempt to maintain a balance between a reliance on source texts for support and an attempt to establish their own voice by choosing not to cite. On the one hand, the participants cited source texts that contained similar ideas so that their opinions were bolstered and secured. On the other hand, the participants understood that not everything needed to be cited and would simply draw on, rather than cite directly, source texts that matched either common knowledge or what they had learnt previously. Students’ citational acts are situated in a learning context where certain information seems supportive or irrelevant, new or learnt, owned by a particular author or shared by many others. As there are no fixed rules, some participants wondered when a citation would be appropriate and whether they should quote or paraphrase texts borrowed. Citing a source text is more than providing a name and a date; it is a subjective process of deciding how to make meaning out of the available resources. Students are still learning the general principles and guidelines. As they learn the parameters of appropriate use, students can easily go across the unmarked borders of appropriate borrowing and lapse into

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

The concern about plagiarism expressed by Eddy and others further explains the limited identifications of the format of textual borrowing among the present students. Participants might be reluctant to identify how they borrowed source texts because they were not sure how much to paraphrase, so the relevant text did not need to be quoted, nor were they sure if it was OK not to cite when certain information was summarized using entirely one’s own words. In addition, they might not want to talk about their plagiarized texts with the researcher who might disagree with them on this sensitive issue. The epistemological status of the data complicates the findings of a dynamic learning process of choosing between not only citing and not citing but also quoting and paraphrasing. As the process is a blind spot for teachers, students have to rely on their own judgement call in making these choices.

22

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

unintentional plagiarism. For example, some people might disagree with Cary who considered spontaneous mutation that occurs during replication in DNA as knowledge learnt previously, and therefore did not cite the source when presenting the information; others would see Polly’s text (Example 6), though cited, as almost a direct copy of the original text and, therefore, a clear example of plagiarism. The present study suggests that appropriate or inappropriate textual borrowing cannot be determined by general context-free criteria. The judgement is negotiated, localized, and contingent. Given the students’ understanding of appropriate citation found among this sample, students’ textual appropriation, appropriate or inappropriate, could be viewed as evidence that students are learning this important academic writing skill of intertextuality and learning about it in ways that relate to the expression of their own ideas and acquisition of academic literacies. Like other studies that focus on participants’ post hoc self-reports, the present data analyses might have not represented students’ citation practices with accuracy. Holstein and Gubrium (2004), among many others, have pointed out that responses from the interviewees represent a dynamic meaning-making process done in collaboration with the interviewer or in the direction designed by the interviewer. For example, participants might have also used ‘familiar narrative constructs’ to comment on their textual appropriation rather than providing the lived experiences or ‘meaningful insights into their subjective view’ (Miller and Glassner 2004: 127). As I pointed out earlier, some students might have hesitated to comment on their textual borrowing for fear that the interviewer would disagree with their citation practices. Readers are thus advised to be aware of the epistemological status of the interview data. Despite the limitations, the study sheds some light on how and why some students borrow texts and, therefore, provides teaching implications. Instructors are advised to use specific examples or cases, such as those identified in the present study, to help students learn how to make value judgements around the use of prior texts based on the degree to which those texts belong to others, or represent either new or common knowledge. Since there is no hard and fast set of rules on citation practices in scholarly texts, students need assistance or direct instruction. I tried out this teaching strategy in a university writing workshop where the discussion on whether to cite or not to cite in each case aroused heated discussions. Many students said it was their first experience to discuss, share, and clarify their individual and subjective acts of textual borrowing. The present study also generates implications for follow-up research. One research focus could be on how citing behaviors mark high-level students’ (juniors, seniors, or grad students) development of disciplinary knowledge. Case studies comparing students’ self-reflections on their use of citations in a longer period of time would identify how students learn to use citations as standard symbols of their discipline and shape their own positions among recognized networks of references. Another research focus could be the

L. SHI 23

SUPPLEMENTARY DATA Supplementary data is available at Applied Linguistics online.

ACKNOWLEDGEMENTS This study is part of a larger project on students’ textual appropriation funded by a Standard Research Grant of Social Sciences and Humanities Research Council of Canada. I thank the participating students, Sin Heng Celine Sze for her help in transcribing and analyzing the data, and John Willinsky, Lynne Earls, and the anonymous reviewers for their valuable feedback on an earlier draft of the article.

NOTES 1

One student (Kate) referred to four of her units as translations from a source in her first language. Another student (Martin) said he created a quote himself but attributed it to a credible author. Since these cases raise different issues and also represent a small proportion of the data, I decided not to deal with them in the present analyses.

2

In all examples cited in this article, students’ writing and comments are presented verbatim. For easy reading, keywords in students’ comments that result in relevant coding are highlighted and alternatives for special terms in student writing are provided in brackets. See Appendix C in the online version of this article for cited works in student writing.

REFERENCES Ange´lil-Carter, S. 2000. Stolen Language? Plagiarism in Writing. Essex: Pearson Education Limited. Campbell, C. 1990. ‘Writing with others’ words: Using background reading text in academic compositions’ in B. Kroll (ed.): Second Language Writing: Research Insights for the Classroom. Cambridge: Cambridge University Press. Chandrasegaran, A. 2000. ‘Cultures in contact in academic writing: students’ perceptions

of plagiarism,’ Asian Journal of English Language Teaching 10: 91–113. Chandrasoma, R., C. Thompson, and A. Pennycook. 2004. ‘Beyond plagiarism: transgressive and nontransgressive intertextuality,’ Journal of Language, Identity, and Education 3: 171–93. Currie, P. 1998. ‘Staying out of trouble: apparent plagiarism and academic survival,’ Journal of Second Language Writing 7: 1–18.

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

teacher’s perceptions of the students’ citing behaviors in terms of, for example, how successful the student writers are in learning the conventional citing forms and whether they are right to omit citations for information they think is common knowledge. Relevant findings would reveal disciplinerelated differences that may well speak to distinctions that instructors in these fields may wish to address directly with students. Together, these studies should build up a theory of citing as a process of learning for novice academic writers.

24

STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS

writing,’ Journal of Second Language Writing 12: 317–45. Pecorari, D. 2006. ‘Visible and occluded citation features in postgraduate second-language writing,’ English for Specific Purposes 25: 4–29. Pennycook, A. 1994. ‘The complex contexts of plagiarism: a reply to Deckert,’ Journal of Second Language Writing 3: 277–84. Pennycook, A. 1996. ‘Borrowing other’s words: text, ownership, memory, and plagiarism,’ TESOL Quarterly 30: 201–30. Petric´, B. 2004. ‘A pedagogical perspective on plagiarism,’ NovELTy 11: 4–18. Roig, M. 1997. ‘Can undergraduate students determine whether text has been plagiarized?’ Psychological Record 47: 113–23. Shi, L. 2004. ‘Textual borrowing in second language writing,’ Written Communication 21: 171–200. Sinclair, J. M. 1986. ‘Fictional worlds’ in M. Coulthard (ed.): Talking about Text. Birmingham: University of Birmingham ELR. Spack, R. 1997. ‘The acquisition of academic literacy in a second language: a longitudinal case study,’ Written Communication 14: 3–62. Starfield, S. 2002. ‘ ‘‘I’m a second-language English speaker’’: negotiating writer identity and authority in Sociology One,’ Journal of Language, Identity, and Education 1: 121–40. Tadros, A. 1993. ‘The pragmatics of text averral and attribution in academic texts’ in M. Hoey (ed.): Data, Description, Discourse. London: HarperCollins. Thompson, C. 2005. ‘‘Authority is everything’: a study of the politics of textual ownership and knowledge in the formation of student writer identities,’ International Journal for Educational Integrity 1. Available at http://www.ojs.unisa. edu.au/journals/index.php/IJEI/article/view/18. Accessed 15 December 2007. Watson-Gegeo, K. A. 1988. ‘Ethnography in ESL: defining the essentials,’ TESOL Quarterly 22: 575–92. Young, K. M. and L. Gaea. 1998. ‘Writing from primary documents: a way of knowing in history,’ Written Communication 15: 25–68.

Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010

Deckert, G. D. 1993. ‘Perspectives on plagiarism from ESL students in Hong Kong,’ Journal of Second Language Writing 2: 131–48. Denzin, N. K. 1997. Interpretive Ethnography: Ethnographic Practices for the 21st Century. London: Sage Publications. Dong, Y. R. 1996. ‘Learning how to use citations for knowledge transformation: nonnative doctoral students’ dissertation writing in science,’ Research in the Teaching of English 30: 428–57. Fairclough, N. 1992. Discourse and Social Change. Cambridge: Polity Press. Hodges, J. C. 1962. Harbrace College Handbook. 5th edn. New York: Harcourt, Brace & World Inc. Holstein, J. A. and J. F. Gubrium. 2004. ‘The active interview’ in D. Silverman (ed.): Qualitative Research: Theory, Method and Practice. London: Sage Publications. Howard, R. M. 1992. ‘A plagiarism pentimento,’ Journal of Teaching Writing 11: 233–45. Howard, R. M. 1995. ‘Plagiarism, authorships, and the academic penalty,’ College English 57: 788–806. Hull, G. and M. Rose. 1989. ‘Rethinking remediation,’ Written Communication 6: 139–54. McCormick, F. 1989. ‘The Plagiario and the professor in our peculiar institution,’ Journal of Teaching Writing 8: 133–45. Miller, J. and B. Glassner. 2004. ‘The ‘‘inside’’ and the ‘‘outside’’: finding realities in interviews’ in D. Silverman (ed.): Qualitative Research: Theory, Method and Practice. London: Sage Publications. Moore, T. 1997. ‘From test to note: cultural variation in summarization practices,’ Prospect 12: 54–63. Odell, L., D. Goswami, and A. Herrington. 1983. ‘The discourse-based interview: a procedure for exploring the tacit knowledge of writers in non-academic settings’ in P. Mosenthal, L. Tamor, and S. Walmsley (eds): Research on Writing. New York: Longman. Pecorari, D. 2003. ‘Good and original: plagiarism and patchwriting in academic second-language

Applied Linguistics 31/1: 25–44 ß Oxford University Press 2008 doi:10.1093/applin/amn046 Advance Access published on 6 December 2008

Practices of Other-Initiated Repair in the Classrooms of Children with Specific Speech and Language Difficulties JULIE RADFORD Institute of Education, University of London

INTRODUCTION Many children in schools experience specific speech and language difficulties (SSLDs) (Dockrell et al. 2006), also widely referred to as children with specific language impairment. These children are characterized, in particular, by lexical and grammatical language difficulties (Leonard 2000) that may impact on interpersonal communication. Education is typically delivered in schools where they receive additional resources from trained language teachers and speech and language therapists/pathologists (Law et al. 2000). The priority of specialised professionals is to develop the children’s understanding and use of grammar and vocabulary, to develop interpersonal communication skills and to facilitate access to the curriculum (Dockrell and Messer 1999). The fact that children with SSLDs spend a large proportion of their time in regular classrooms, where the vehicle of instruction is primarily oral (Edwards and Westgate 1994; Cazden 2001), places them at a significant disadvantage. Their dilemma for accessing the curriculum is exacerbated when teacher–pupil discourse is primarily targeted at whole class groups, such as during

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

Repair practices used by teachers who work with children with specific speech and language difficulties (SSLDs) have hitherto remained largely unexplored. Such classrooms therefore offer a new context for researching repairs and considering how they compare with non-SSLD interactions. Repair trajectories are of interest because they are dialogic sites where the child’s meaning is being negotiated and, therefore, where adults might create opportunities for language learning. The interactions take place during activities, such as story writing, where teachers elicit children’s ideas and orient to their lack of clarity. From a data set of 78 cases, four significant patterns of teacher repair initiation emerged. First, non-specific repair initiators (RIs), such as ‘say that again’, target any aspect of the prior turn and reveal the adult’s lack of grasp of its content. Next, specific RIs (‘she has’) that are constructed with minimal components of the child’s turn, pinpoint the location of the trouble but provide no new lexical information. In contrast, specific RIs that are constructed as ‘wh’ questions (‘down where’), target the nature of the trouble and elicit further information. Finally, offers of candidates (‘do you mean X’) do provide new models of lexis but do not elicit repetition from the child.

26

REPAIR PRACTICES IN SSLD

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

mathematics and literacy lessons (Mroz et al. 2000; Smith et al. 2004). Arguably the most important factor concerns the quality of the interaction itself. International research continues to report the prevalence of interrogative and evaluative styles of talk in classrooms (Alexander 2001), despite many efforts to foster the increased participation of children during interactions. A well-documented system, called the ‘IRE/F’, constrains the participation of teachers and students (Sinclair and Coulthard 1975; Mehan 1979). It is characterized by the teacher initiating a ‘test’ question (I) that is followed by a known answer/response (R) (Macbeth 2004). The third turn position, following the child’s response, is where evaluation (E) typically occurs, although there is potential for other devices that build upon the child’s turn, such as follow-up (F) or extension moves (Wells 1993) and uptake questions (Nystrand 1997). In IRE/F discourse contexts, however, children with SSLDs run the risk of becoming passive and lacking engagement in their learning (Radford and Ireson 2006). Minimal participation may lead to reduced opportunities for language learning, because topic-related exchanges are shortened and trouble sources may remain unresolved (Sadler and MogfordBevan 1997). An alternative to interrogative and evaluative talk is dialogic discourse that is evident during some types of classroom activities, for example in France, England and India (Skidmore 2000; Alexander 2001). Dialogic interactions provide higher quality learning experiences for children than talk that is teacher-led (Mercer 1995, 2000; Mercer and Littleton 2007). Dialogic talk is characterized by (at least some) questions that provoke thoughtful answers, the sharing of ideas between teachers and students, children’s free articulation of ideas and the chaining of ideas for a common purpose (Mercer 2000; Alexander 2006). Children with SSLDs could benefit from dialogic discourse provided that the language is accessible and allowance is made for their expressive difficulties. There is preliminary evidence that specialist language teachers engage in dialogic teaching with children with specific language difficulties during small group activities, such as story writing and news telling (Radford et al. 2006). Drawing on studies into the co-construction of topic using conversation analysis (CA) (Button and Casey 1984, 1985; Radford and Tarplee 2000; Ridley et al. 2002), the authors demonstrate how the children and their teachers co-construct topic on a turn-by-turn basis. Two examples are now provided that are relevant to the data analysed in the current study because children’s own ideas are generated in these contexts. Owing to the children’s language difficulties novel ideas, as we shall see in the later analyses, frequently necessitate the initiation of repair. The resulting trajectory of talk that deals with the repair is, arguably, an example of dialogic discourse. The first example, an ‘invitation’ sequence, has been found when teachers construct a story with a group of children [Extract (A), taken from Radford et al. 2006]. Invitation sequences differ from question-with-knownanswer sequences because, although they are suited to constraining the child’s response in terms of adherence to a story-line, they accomplish the generation

J. RADFORD

27

of the child’s novel story ideas. As such, the teacher is not predicting precisely what ideas the child will offer, and therefore what s/he will subsequently receive as a relevant contribution.

The second example is found during a one-to-one speaking-book activity where the teacher and child are discussing pictures and photographs that the child has chosen. These pictures have been brought from the child’s home and the teacher has not seen them before. Extract (B) shows the teacher using a topic initial elicitor (TIE) (Button and Casey 1984) which is heard by the child as a request to offer ideas. The TIE sequence differs from a question-with-known-answer sequence because the child treats it as an opportunity to choose his/her own topical material, albeit within the constraints of relevancy to the picture. In the case of photographs, the child may offer personal news ideas whereas, in the case of pictures, the ideas may concern a description of any detail of the picture (Radford et al. 2006).

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

Extract (A) begins with the teacher’s (T) story starter line (‘there was a’), that is designed to be incomplete and, therefore, completed by a nominated child. The child (line 2) treats T’s turn as a request to contribute an idea concerning the story’s character. Other examples of next turns include children’s ideas relevant to the story’s title, the setting or the plot. The teacher’s response to the child’s ideas is not an evaluation, but an ‘acceptance’ which can take a variety of forms. In line 3 it is a repeat of the child’s idea and, in some cases, the verbal idea is drawn onto a flipchart to make a visual representation of the story constructed so far (lines 5 and 6).

28

REPAIR PRACTICES IN SSLD

PARTICIPANTS, DATA SET AND METHOD The adult participants were three specialized teachers who had all worked in language resource provision for at least 10 years. Whilst they were not

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

In line 2, the teacher’s TIE at first appears to resemble an instruction, although it could take the form of a request for a telling ‘can you tell me. . .’. However, the child treats it as an open inquiry and offers her own ideas to describe the picture (line 2). The teacher responds to the ideas with topicalization of the material generated so far (Button and Casey 1984; Radford and Tarplee 2000). Here we find ‘mhm’, a continuer that does the job of topicalizing in this context. The child orients to the topicalizer and the silence with continuation of her description (line 6). If, instead of topicalization, the teacher had elected to evaluate the child’s response and asked a test-like question, the child would have been denied the opportunity to pursue the topic that she had nominated herself. As invitation and TIE sequences are suited to the generation of ideas by the children, they provide potentially richer resources for language learning than known-answer questions that generate short, unexpanded responses. Sequences that have further potential, but have not yet been investigated, are where acceptance or topicalization of the children’s ideas is withheld. Instead, the participants suspend the usual sequence in order to deal with a trouble (see Schegloff 1972, on insertion sequences). Children with SSLDs, owing to the very nature of their linguistic problems, produce turns that contain grammatical or lexical errors or risk being unclear to the listener. When teachers initiate repair (usually) in the next turn, the insertion sequence deals contingently with the child’s trouble. During the insertion, the teacher might provide correction (Jefferson 1987), which in language teaching would take the form of a more competent model of grammar or lexis, as compared with the child’s version. Another possibility is that the teacher might initiate repair such that the child is placed to revise her/his original move. This article aims to explicate, in detail, a series of practices where teachers and children are dealing with repair, when acceptance or topicalization of the child’s idea is withheld. More specifically, the article will pay particular attention to the design of the teachers’ initiator techniques and the actions accomplished by them. Owing to the pedagogical interests of the researcher, the analysis will highlight the relative strength of the repair initiator (RI) to locate the problem with speaking/hearing/understanding (Schegloff et al. 1977; Schegloff 1997). Techniques that provide the least help in locating the trouble, such as open class RIs (e.g. ‘pardon’) (Drew 1997) will be compared with those that locate the repairable more specifically and/or supply a candidate offer to the child. Some tentative suggestions will be made concerning the ordering of these devices in oral language lessons. It is of further empirical interest to compare the findings to studies conducted by conversation analysts in first and second language discourse contexts since there is no parallel work in the classrooms of children with SSLD.

J. RADFORD

29

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

qualified as speech and language therapists/pathologists, they did have post-training qualifications in the area of language development and SSLDs. There were six children, aged 4 years 4 months to 8 years 7 months at the time of the first visit, who all had ‘statements’ of special educational need that indicated the need for language resource provision. Information about the nature of their language difficulties was collected by questionnaire that was completed by the teacher and speech and language therapist/pathologist together. Table 1 shows that the children had mainly receptive and expressive language difficulties,2 characterized also by semantic problems. Children with significant speech difficulties, such as verbal dyspraxia, phonological difficulties and pragmatic difficulties3 were excluded. Story writing, book-sharing and circle-time activities were selected by the teachers as best examples of oral language teaching. Story writing and circletime activities were conducted in small groups of five to six children, although two target children in each activity were the focus for the analysis. Booksharing was conducted on a one-to-one basis with the teacher. Video-recordings were made with a digital camera on four separate occasions, yielding a data set of 24 lessons (8–45 min long). Radio-microphones were positioned on the teacher and the quietest of the two target children in the group activities to help increase the accuracy of the recordings. Repeated viewings were made followed by detailed transcriptions of selected segments. Given that, apart from aforementioned work, there are no detailed studies of interaction in the classrooms of children with specific language difficulties, one possible analytical approach would be to borrow or adapt a coding scheme used in second language classroom research. This was rejected for several reasons. In terms of children’s moves, coding single turns, for instance as phonological, lexical or grammatical ‘errors’ (Lyster 1998), would be inappropriate because (i) many of the turns contain multiple sources of trouble and (ii) unclear content is a common source of trouble in specific language difficulties data, but it only becomes apparent from the teacher’s display of difficulty in the subsequent turn or turns. Similarly, coding single teacher moves as, for example, repetitions, recasts or requests for clarification (Loewen 2004) would be problematic as shown in the analysis. This study thus adopts CA perspective in order to represent ‘how participants themselves produce and interpret each other’s actions’ (Pomerantz 1988: 361). First, the author made a collection of 245 instances of repair of which 78 cases appeared, superficially at least, to be dealing with repair initiated in response to the child’s idea. The analysis next proceeded by repeated viewings of the video sequences, which uncovered how apparently similar practices, such as repetitions, accomplish completely different actions (Schegloff 1997). It also showed how actions can co-occur, notably checking/correction/clarification. Furthermore, a rich and detailed analysis of participants’ moves was made, including linguistic detail, non-verbal action and prosody, which is essential to fully document the competences of children with language difficulties and their teachers.

8 8 4 4 5 5

A B C D E F

7 months 3 months 4 months 8 months 10 months 5 months

F F M F F F

Gender

3 2 2 3 3 3

Receptive

2 2 2 3 3 3

Expressive syntax 0 0 1 2 1 1

Phonology

0 0 0 0 0 0

Verbal dyspraxia 1 2 2 3 2 3

Naming

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

Note: 0 = none, 1 = mild, 2 = moderate, 3 = severe.

years years years years years years

Age

Child

Table 1: Children’s SSLDs Pragmatics

1 1 1 0 1 3

Word meaning 2 2 2 3 2 3

30 REPAIR PRACTICES IN SSLD

J. RADFORD

31

FINDINGS

Non-specific (‘general’) RIs What is distinctive about non-specific RIs is that when teachers signal trouble with the child’s response turn, they do not, by any specific design indicator, locate the item to be repaired. As such they target potentially any aspect of the child’s prior utterance. Children have different options in response: they can either repeat or reformulate their own previous material in the following turn or turns. Extract (C) illustrates three examples of ‘general’ RIs: a repetition request at line 3 and two statements of uncertainty (lines 5 and 7). The context of the extract is story writing with a small group of six children. The teacher (T) has already written and drawn several of the children’s ideas onto a flipchart regarding the characters and story-line and now asks the children a question.

T’s question (line 1) would typically be heard by the child as a request for an idea concerning the story’s title. As the child (Ch) offers a verbal contribution at line 2, T’s relevant next would be to accept the idea (often in the form of a repeat). Instead, an insertion sequence suspends production of T’s acceptance to deal with a trouble. T’s use of ‘say that again’ gives

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

A series of other-initiated repair (OIR) practices are described, of interest for language learning because they are dialogic sites where the child’s meaning is potentially being negotiated. First, there is an account of OIRs that are ‘non-specific’ in the sense that each RI has an open design that orients to the teacher’s problems hearing/understanding the child’s turn(s). Non-specific OIRs are, for the most part, treated as prompts to repeat. In contrast, the next examples are classed as ‘specific’ OIRs since the teacher pinpoints, in various ways, the location of the trouble, either because she leads the child into its location with a non-completed utterance, or because she uses a ‘wh’ question to specify its nature. Finally, examples where candidate offers are made will be explored, including requests for definition.

32

REPAIR PRACTICES IN SSLD

‘Pardon’ could indicate that the trouble-source entails an issue related to T’s hearing and thus be heard as a request for a repeat. Yet, the child’s subsequent move is designed differently: while she does repeat some lexical components of her idea, she also reformulates her original word order and verb form. An explanation for the reformulation might be that open class RIs are claimed to treat ‘the whole of the prior turn as in some way problematic’ (Drew 1997). The child thus orients to T’s signalling of trouble with potentially any element of her turn at line 2. As in Extract (C), the open class RI is not targeting a specific component of the child’s message for repair. Features of all non-specific designs are that T’s options for repair initiation are constrained: if the child’s turn is not heard or understood clearly, T is unable to incorporate elements of the turn in a subsequent move. In this sense, T’s response is a ‘general’ repair initiation, as opposed to the more specific formats that will be presented next.

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

the child an unambiguous message that a repeat is needed. In fact, line 3 has the appearance of an instruction, given the absence of lexical or syntactic markers of politeness. Although the child offers a response in line 4, T indicates continued trouble with a further non-specific marker of uncertainty (line 5). Ch next produces a non lexical item (‘ee::’), so T responds with a third RI, this time clearly displaying her difficulty with understanding. This also fails to solve the trouble so it is left to T to suggest a story name idea. A notable feature is that each of T’s RIs target the whole of Ch’s prior as opposed to marking out specific elements for Ch’s attention. Extract (C) showed three examples of non-specific RIs: a statement that a repeat is needed, and two statements that display general trouble with hearing or understanding. As the child’s responses were unintelligible, further data are needed to clarify the sequential consequences of such actions. Extract (D) has a clearer outcome because the teacher indicates trouble with an open class RI (‘pardon’, line 3) and this is taken up by the child as an opportunity to reformulate her grammar. The sequence takes place where the children are invited to suggest a plot line for the group’s story. The child’s subsequent ideas include both a novel character and an event. That the name ‘Dina’ that she supplies is searched-for is indicated by the stretched sound at the end of ‘com:::’ and the brief silence. The relevant next, a teacher acceptance, is once again withheld and an insertion sequence begins.

J. RADFORD

33

Specific RIs (designedly incomplete utterances) that pinpoint the location of the trouble source

Although Tom offers the name of an animal, T withholds acceptance of this idea. Instead, she initiates an insertion sequence in line 3 with a DIU. It is a single word that is formed in a similar fashion to T’s original version and the child’s production: it repeats the element that preceded the lexical item, preserving the lengthening of the vowel and continuing intonation. Her trouble could have entailed hearing and is confirmed by her body language: a shake of the head and leaning posture. The child’s response shows her interpretation of the DIU, thus she supplies a repeat of the noun phrase (line 4), rather than a reformulation. The repeat supplies T with information and she offers her candidate hearing, proffered for confirmation. Although the hearing is confirmed by the child, T pursues a definition, as if to find out if the child knows what a cheetah is. One notable feature of the DIU is that it targets the trouble source

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

When teacher’s acceptance of a child’s idea is withheld, a common design for initiation of an insertion sequence, is a designedly incomplete utterance (DIU). Whilst acknowledging that there is a small data set and more research is needed, the DIUs found in the classrooms of children with SSLDs share the following characteristics to those reported in second language writing conferences (Koshik 2002). First, they are constructed with minimal components, either a phrase or an individual word, that borrow fragments of the child’s prior turn. Next, they are used to perform different actions: on the one hand, through partial repeat of the child’s turn, a DIU may elicit repetition of prior talk [Extract (E)]; on the other hand, it may be used to target an error and prompt the child to self-correct [Extract (F)]. Extract (E) takes place during the early phase of the group story writing session where characters are being decided. The teacher targets Tom, using an oral cloze technique (incomplete utterance and vowel stretch), to select a character. In his turn, Tom begins with a repeat of T’s modelled story starter ‘there was a:::’. Prolongation of the vowel sound, constructed with continuing intonation, and the brief silence indicate that he is searching for a novel idea to offer the teacher.

34

REPAIR PRACTICES IN SSLD

very precisely by leading the child, at least syntactically, up to the location of the trouble. Furthermore, the DIU has pedagogical value in so far as it generates both confirmation and definition sequences that afford further opportunities for language learning. A similar design of DIU is found in Extract (F) but the work that it does is different from Extract (E) because the child hears it as a request for revision rather than for repetition. The teacher and child have just embarked on the speaking-book activity so, as they turn the page to reveal a new picture to talk about, they embark on a TIE sequence. Downloaded from applij.oxfordjournals.org by guest on December 31, 2010

The child responds to T’s topic elicitor with a description of a girl (line 2), although the turn is partially unclear to both T and the researcher. That Ch’s gaze remains on the picture suggests that she is engaged in solitary searching during the silence. At line 3 T withholds topicalization and, instead, produces a DIU. It is formed by using two words of Ch’s turn up to the point at which there was potentially a missing adjective, as seen in Ch’s subsequent turn (line 4). Continuing intonation and gaze directed at the child, rather than the picture, suggests that the turn may be actioning a prompt. Ch hears it as a prompt, perhaps owing to the syntactic incompleteness of ‘she has’, not to produce the noun but a relevant next adjective ‘long’. T hears Ch’s turn and displays acceptance of the child’s idea with a repeat of both the child’s adjective and noun from line 2. In terms of pedagogical value, by generating further information from the child, the DIU sets up an opportunity for T to expand the material and supply further contingent models of language. Like Koshik’s (2002) DIUs, these two examples are not complete turn constructional units, but are recognizably complete actions that are designed, with the child’s own words, to be incomplete and thus completed by the child. In response, the child is not ‘interrupting’, even though T’s turn construction unit is not completed. The child orients to T’s design (where it stops) and continues with a syntactically fitted contribution. The first example (‘a::’) led the child to re-produce her noun and therefore complete the noun phrase. The second, a phrase, was not only an oral invitation to complete the sentence, but also taken as an opportunity by the child to revise the next item of her original phrase. DIUs are prosodically marked here, by sound stretches and by continuing intonation which position the child to continue the phrase or sentence. DIUs, thus, target very specifically the location of a trouble. Yet, even when they are heard as invitations to self-correct, they do not give the child semantic

J. RADFORD

35

Specific RIs (‘wh’ questions) that pinpoint the nature of the trouble source The next sequences show three types of RIs that are formatted as ‘wh’ questions: a request to provide specific information and two requests for definition. Instead of being heard as requests for repetition, the ‘wh’ questions are treated as requests to supply further information. The analysis will reveal how their designs serve to pinpoint precisely, not so much the location of the trouble source, as in DIUs, but the nature of the repairable. The analysis will also show how the child orients to the RI by making a revision that provides additional information in terms of novel aspects to the story-line. Extract (G) takes place in the context of genuine news enquiry and reporting. As the child’s reports are potentially ‘news’ to the teacher, any new ideas are susceptible to lack of clarity. The turn of interest is at line 6, and is part of the emerging newstelling sequence. The child and teacher are looking at a photograph in the ‘speaking-book’ and Ch is reporting a personal news account about taking her dog, Penny, for a walk.

The teacher responds to Ch by asking a genuine question since she has not experienced the walk-with-the-dog. The enquiry is itemized in so far as she targets the location of the walk as a topic for further talk. The child’s pause and searching behaviour (‘uh::m’) show that a response is being searched-for. To assist the child’s search T offers a candidate answer of a

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

information that could be usable in the subsequent correction. Both examples have pedagogical value in so far as T indicates the trouble source very specifically and generate further opportunities for targeted language practice. Finally, it is worth mentioning how the current examples are different from the writing conference examples reported by Koshik (2002). In most of the writing conference data, a written text was an additional available resource that was visible (at least on occasion) to the participants. In contrast, the language difficulty data are taken from oral language lessons, so the teachers are dealing with online discourse problems. That the children have language difficulties places an additional burden on the listener to process the children’s turns accurately, especially when novel ideas are being generated in an activity like story writing.

36

REPAIR PRACTICES IN SSLD

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

potential location in line 4. However, as the reported event is the child’s experience, Ch holds the expert knowledge about what happened and rejects T’s version (‘no’). Following the rejection, Ch does supply some information that answers the itemized enquiry. At line 6, given T’s puzzled look, she seems to treat this turn as providing partial information regarding the location of the walk. She thus appears to be making an itemized enquiry to elicit specific information relevant to the news report. Although the request has a minimal format, recycling just one lexical item of the child’s turn (‘down’), it is informative to the child. There are two important dimensions: lexically, ‘whe:re’ indicates that a location is the source of trouble; prosodically, loudness on ‘whe:re’ marks out the item that lacks clarity. In response, (line 7), Ch answers the question by supplying potentially clarifying information. Even though the analyst found it impossible to tell, because the ambiguity of the homophone ‘see/sea’ in English, whether or not the child was using a verb (see) or was self-repairing use of the determiner ‘the’ before the noun ‘sea’, T displays an understanding which is confirmed by the child. T’s repair initiation at line 6, then, is heard by the child as an invitation to offer further information that has the additional effect of clarifying her news report. Furthermore, at line 8, when T pursues the information generated by the RI, the child is afforded another language model that is contingent. Extracts (H) and (J) illustrate typical definition-type requests that were widely used by the specialist teachers. Whilst the author acknowledges that more examples are needed to be confident about claims regarding their systematicity, both examples adopt a similar grammatical format (‘what’s X’) and initiate repair on a lexical item just used by the child which displays the T’s lack of understanding of Ch’s usage in that context. In the first example, Ch supplies different information that revises the earlier version of her lexical item. In the second example, there is no revision but, instead, it is heard as a request to provide additional information concerning the trouble source. In Extract (H), the exchange takes place in the context of picture description where, earlier in the sequence, the child had contributed a range of ideas. At line 20, Ch is continuing her description and is currently talking about the activities of a frog.

J. RADFORD

37

The teacher’s question at line 5 signals her on-going trouble with Ch’s idea, also confirmed by T’s simultaneous non-verbal behaviour (shaking her head). The question has the appearance of a request for definition. How Ch heard T’s request is unclear. Perhaps, owing to her language difficulty, she was unable to supply a standard definition (e.g. a small wooden building). The action accomplished here, interestingly, is that she supplies a novel idea that is relevant to the story-line. T’s subsequent reformulation (line 7) incorporates the ideas from both line 2 as well as the new material produced

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

In UK English, ‘neighbourhood’ (line 20) is a rare item of vocabulary and its use with respect to frogs is particularly unusual. The teacher in line 21, drawing presumably on this world knowledge, offers a candidate hearing by repeating the final two words of Ch’s turn. The candidate hearing is confirmed by Ch in line 22 (‘yeah’). Despite Ch’s confirmation, T continues to remain confused. She displays this confusion with a specific request in line 23 to define the problematic phrase (‘his neighbourhood’). The grammatical format ‘what’s X?’ supplies Ch with information that a word definition is needed. In line 24, the child does not, however, supply a definition but responds by reformulating the trouble source. The child’s self-repair is prefaced by ‘I mean’ that appears to indicate a degree of metalinguistic awareness on her part about self-correction. That Ch produces the lexical item (‘neighbour’) at line 24 gives T opportunity for a repeat and, therefore, a further module of the correct lexis. Extract (J) illustrates an apparently comparable request at line 5. It also emerges from prior talk that is dealing with checking mutual understanding and has the same grammatical form (‘what’s X’). What is different here is that, instead of being heard as a request for revision, the child is able to provide a partial definition of item ‘X’. At the beginning of Extract (J) we find a TIE sequence that generates a plot idea from the child. Although Ch offers a novel idea (given that ‘sheds’ have not been talked about so far), T withholds acceptance to initiate an insertion sequence. She repeats the last two words that the child said with additional loudness on ‘SHED’ and final rising intonation, as if to check for confirmation. The child offers non-verbal confirmation rather than a revision of her turn (line 4).

38

REPAIR PRACTICES IN SSLD

Offers of candidates that supply a linguistic model The examples shown in Extract (K) are distinctive because there is material in the teacher’s turns that has not been mutually shared in prior talk, according to the available data. There are two teacher moves of particular interest: at line 5 there is a candidate understanding and, at line 10, an offer of a candidate that does the job of correction. Offers of candidates can be useful when a speaker anticipates that a hearer will experience difficulty supplying the relevant information and the speaker wants to offer help (Pomerantz 1988). The example at line 10 is formed as a modulated other-correction ‘Do you mean X?’ (Schegloff et al. 1977), although at line 5 there is a variation of form. T and Ch are talking about a picture of the seaside that Ch has drawn to represent her story setting.

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

in 6. The work done by the repair initiation is therefore to achieve a definition that has the additional effect of contributing to the ongoing story-line (T later displays acceptance by drawing both ideas onto the flipchart). The RI also creates a sequential opportunity for the teacher (line 7) to supply an expanded model of the child’s ideas, which has value in terms of language learning. A further observation is that a candidate hearing/understanding (proffered for confirmation) precedes clarification in each of these sequences and appeared with some regularity elsewhere in the data set. As an illustration, Extracts (H) and (J) involve two teachers working with very different children of different ages and yet the pattern is similar. First of all, there is the first part of an adjacency pair, formed with a repeat of the child’s noun phrase [Extract (H), line 21, ‘his neighbourhood’; Extract (J), line 3, ‘a SHED?’]. This provides a candidate hearing or understanding that makes relevant the child’s confirmation or correction. What appears to be happening is that, in the context of content troubles, a first move is to check that the trouble does not reside with the teacher herself: could it have entailed her own problem with hearing or understanding? In both cases confirmation is given, so a hearing/understanding problem is ruled out. Yet, T’s pursuit of elaboration of the item persists, which has the effect of initiating a pedagogical side sequence that targets the child’s current understanding of the lexical item. A design like ‘what’s X’ does two jobs. First, it invites the child to provide further information and, as we see here, the information is forthcoming, either as a revision or as material relevant to the topic. Next, since X is a component of the child’s prior turn, it marks out very specifically the item to be talked about, which renders it pedagogically useful for displaying the child’s understanding. This distinguishes the practice from the single item ‘what?’ technique that targets the whole of the prior (Schegloff 1997). A final, somewhat tentative observation, is that there appears to a ‘natural ordering’ at work, based on the relative strength or power of the teachers’ techniques to locate the repairable (Schegloff et al. 1977). The teachers are employing their moves in order of increasing strength.

J. RADFORD

39

At the start of the sequence the setting is being described and an additional character suggestion is made at line 3.

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

Ch’s idea, ‘a nice snake’ is treated as problematic by T who withholds acceptance. An insertion sequence is initiated in line 4 with a repeat of the lexical item which is formed with high pitch rise, as if to signal surprise. T makes two specific queries concerning the snake’s whereabouts in lines 4 and 5. The first, beginning with ‘wha d’y’mea:n’, at first appears as a challenge to the credibility of Ch’s account (Schegloff 1997). A more likely interpretation is that T is targeting ‘in the water’ for clarification of the child’s meaning. Instead of waiting for a response, T directly offers a candidate understanding ‘in the sea?’ (line 5) of the Ch’s turn at line 3. Ch does not respond pragmatically to T’s queries or take up the offer of the candidate. Instead, she treats the turn as a hearing or understanding check, with a repeat of her phrase (line 6). No new topical material is produced by Ch, so mutual understanding has yet to be reached which is indicated by T re-visiting the problem at line 9. This turn signals T’s confusion (‘I’ve never see:n a snake in the sea.’) now making it explicit that her own world experience does not fit Ch’s claim. In line 10, T offers a further candidate with the modulated other correction format ‘do you mean X.’ In contrast with line 5, it has falling intonation which confirms its status as a correction. Use of the verb ‘mean’ indicates clearly to Ch that the issue concerns T’s understanding rather than hearing. In order to correct ‘snake’ to ‘eel’ T needed to guess by drawing on a range of contextual information. How are candidate offers significant for a discussion of language learning opportunities? In lengthy repair sequences, given the new material that they include, they occasion further opportunities for language practice. When offering a candidate item, the teacher is providing a model to the child of what is relevant and appropriate to produce as a response (Pomerantz 1988). In the case of ‘d’y’mean an eel?’ the model is a single lexical item, although grammatical units could equally be modelled in this way. The teacher leads the child to focus on the item, and further exposes it, by placing it in the final position of the turn and marking it with additional prosodic emphasis. Yet, in the example presented here and others in the data set, it is not repeated, even following

40

REPAIR PRACTICES IN SSLD

teacher repetition at line 12. An explanation is that a relevant response to this type of yes/no question is either for the child to confirm T’s guess as the correct target, or to give a different version if the guess is incorrect.

SUMMARY AND CONCLUSIONS

(i) Non-specific (‘general’) RIs:    

constructed as an open class RI (‘pardon’) or as a message that the trouble source entails hearing (‘say that again’) or understanding (‘I don’t know what that means’); target the whole, or potentially any aspect, of the child’s prior turn; reveal a ‘substantial lack of grasp of what was said’ (Schegloff 1997; 524); used to elicit a repeat or reformulation of any (or all) of the prior.

(ii) Specific RIs: DIUs:    

constructed with minimal components of the child’s prior turn (‘a:::’; ‘she has’); used as a clue that pinpoints the location of the trouble source; provide no new lexical information; used to elicit a repeat of the child’s prior or to target an error and prompt self-correction.

(iii) Specific RIs: ‘Wh’ questions:   

constructed as ‘wh’ questions (‘down whe:re.’; ‘what’s his neighbourhood?’); used as a clue that pinpoints the nature of the trouble source; used to elicit further information about the trouble.

(iv) Offers of candidates:   

constructed as modulated other correction ‘do you mean X’; provide a new model of the type of information that is needed (Pomerantz 1988), such as a lexical item, but could be grammatical, for example morphology: not suited to eliciting a repeat (uptake) of the model since formed as a yes/no question.

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

This analysis has detailed, for the first time, a variety of techniques adopted by specialist teachers to initiate repair in response to troubles hearing or understanding children’s novel ideas. Given that children with SSLD typically display problems with their production of phonology, syntax and semantics, their original ideas commonly occasion insertion sequences that deal with the business of correction, other-initiation and self-repair. Such sequences afford contingent opportunities for dialogic discourse and, therefore, provide potential opportunities for language learning. Four distinct types of RI practices emerged as prevalent patterns in a substantial database of 78 cases. It is now worth summarizing the key characteristics of these techniques.

J. RADFORD

41

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

Non-specific RIs are distinguished from specific RIs principally by the degree of information afforded to the child. The former target the trouble source more generally, so the child has a wider domain of prior utterance(s) from which to select in order to formulate a repeat or revision. Specific formats are more focused: they either target the location of the trouble (DIUs) or the nature of the trouble (‘wh’ questions). DIUs are used to both elicit repeats and to prompt self-correction. Furthermore, ‘wh’ questions include requests for word definition, and these are suited to eliciting the child’s display of understanding of lexical items. Whereas offers of candidates provide a linguistic model, they are not taken as opportunities for repeat (or ‘uptake’, as in Loewen 2004). General’ RIs are preceded by less clear child contributions, where the only option is to target the whole/any of the prior. In contrast, specific RIs are possible in contexts where the child has been, at least partially, heard or understood. This affords the opportunity to repeat part of the prior, as in DIUs, or to target a specific component such as in offers of candidates (‘Do you mean X?’) or ‘wh’ questions. Detailed descriptions of repair practices have important implications for practitioners. Studies of repair trajectories in classrooms where children are learning a second language have identified distinctive patterns that are associated with different communicative tasks (Seedhouse 2004). As far as the field of speech and language therapy is concerned, comparable work has provided preliminary insights into interactions between adults and children with specific language impairment (Gardner 2005) and adults with acquired speech disorders (Bloch 2005). In terms of the current study, both teachers and speech and language therapists will gain a richer understanding of adults’ responses to children’s ideas during language interventions, especially when dealing with sources of unclear meaning. From a methodological perspective, using CA has afforded fresh insights into the participants’ behaviours. Whereas Radford et al. (2006) broadened the notions of teacher initiation (I) and child response (R), this study provides fresh insight into feedback (F) practices that initiate repair and the actions accomplished by a range of designs. Although not a primary aim, the study also demonstrates the rich interactional competencies of the children to engage in repair sequences, despite their evident language difficulties. Furthermore, it is evident that in analysing the data of children with speech and language difficulties, external judgements of ‘errors’ would be misleading. The broader construct of trouble source may be better suited to ‘messier’ data, whilst the participants’ perspectives are key to interpreting the actions accomplished. Yet, as CA has rarely been adopted in studies of such children, there is a considerable amount of work still to do. Future research aims to explore a wider range of lesson and participant contexts and examine in more detail other design features, such as the non-verbal and prosodic features of the interaction.

42

REPAIR PRACTICES IN SSLD

APPENDIX Transcription key1 (0.5) (.) = []

hh ((points)) ::: ! () (guess) . , ? "# under CAPITALS 88 >< !

NOTES 1

The transcription conventions used in this article (as shown in the Appendix)

are based on the system originally developed by Gail Jefferson, as

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

.hh

The number in brackets indicates silence by tenths of seconds. A dot enclosed in a bracket indicates a gap in the talk of less than twotenths of a second. The equals sign indicates ‘latching’ between utterances. Square brackets between adjacent lines of concurrent speech indicate the onset and end of a spate of overlapping talk. A dot before an ‘h’ indicates speaker in-breath. The more h’s the longer the breath. An ‘h’ indicates an out-breath. The more h’s the longer the breath. A description enclosed in a bracket, and written in italics, indicates a non-verbal activity. For example ((points at picture)). A dash indicates a sharp cut-off of the prior word or sound. Colons indicate that the speaker has stretched the preceding sound or letter. The more colons the greater the extent of the stretching. Exclamation marks indicate an animated or emphatic tone. Empty parentheses indicate the presence of an unclear fragment of tape. The words within a single bracket indicate the transcriber’s best guess at an unclear utterance. A full stop indicates a stopping fall in tone. It does not necessarily indicate the end of a sentence. A comma indicates a ‘continuing’ intonation. A question mark indicates a rising inflection. It does not necessarily indicate a question. Pointed arrows indicate a marked falling or rising intonational shift. They are placed immediately before the onset of the shift. Underlined fragments indicate speaker emphasis. Words in capitals mark a section of speech noticeably louder than that surrounding it. Degree signs are used to indicate that the talk they encompass is spoken noticeably quieter than the surrounding talk. ‘More than’ and ‘less than’ signs indicate that the talk they encompass was produced noticeably quicker than the surrounding talk. Arrows in the left margin point to specific parts of an extract discussed in the text.

J. RADFORD

2

reported in Atkinson and Heritage (1984: ix–xvi). ‘Expressive’ language difficulties may include some or all of the following: problems with expressive syntax, morphology, articulation, phonology and word finding difficulties. ‘Receptive’ language difficulties entail problems

3

43

with the comprehension of syntax, morphology and word meanings (Conti-Ramsden and Botting 1999). ‘Pragmatic’ language difficulties are defined as problems in using language appropriately in a conversational context despite well-formed syntax (Bishop 2000).

REFERENCES Dockrell, J., G. Lindsay, B. Letchford, and C. Mackie. 2006. ‘Educational provision for children with specific speech and language difficulties: perspectives of speech and language therapy service managers,’ International Journal of Language and Communication Disorders 41/4: 423–40. Drew, P. 1997. ‘ ‘‘Open’’ class repair initiators in response to sequential sources of trouble in conversation,’ Journal of Pragmatics 28/1: 69–101. Edwards, A. D. and D. P. G. Westgate. 1994. Investigating Classroom Talk. 2nd edn. London: Falmer Press. Gardner, H. 2005. ‘A comparison of a mother and a therapist working on child speech’ in K. Richards and P. Seedhouse (eds): Applying Conversation Analysis. Basingstoke: Palgrave Macmillan. Jefferson, G. 1987. ‘On exposed and embedded correction in conversation’ in G. Button and J. Lee (eds): Talk and Social Organization. Clevedon: Multilingual Matters. Koshik, I. 2002. ‘Designedly incomplete utterances: a pedagogical practice for eliciting knowledge displays in error correction sequences,’ Research on Language and Social Interaction 35/3: 277–309. Law, J., G. Lindsay, N. Peacey, M. Gascoigne, N. Soloff, J. Radford, S. Band, and L. Fitzgerald. 2000. Provision for Children with Speech and Language Needs in England and Wales. London: DfEE. Leonard, L. 2000. Children with Specific Language Impairment. Cambridge, MA: MIT Press. Loewen, S. 2004. ‘Uptake in incidental focus on form in meaning-focused ESL lessons,’ Language Learning 54/1: 153–88. Lyster, R. 1998. ‘Negotiation of form, recasts and explicit correction in relation to error

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

Alexander, R. 2001. Culture and Pedagogy: International Comparisons in Primary Education. Oxford, UK/Malden, USA: Blackwell. Alexander, R. 2006. Towards Dialogic Teaching: Rethinking Classroom Talk. 3rd edn. Dialogos, UK: University of Cambridge. Atkinson, J. and J. Heritage (eds). 1984. Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press. Bishop, D. V. M. 2000. ‘Pragmatic language impairment: a correlate of SLI, a distinct subgroup or part of the autistic continuum?’ in D. V. M. Bishop and L. B. Leonard (eds): Speech and Language Impairments in Children: Causes, Characteristics and Outcomes. East Sussex, UK: Psychology Press. Bloch, S. 2005. ‘Co-constructing meaning in acquired speech disorders: word and letter repetition in the construction of turns’ in K. Richards and P. Seedhouse (eds): Applying Conversation Analysis. Basingstoke: Palgrave Macmillan. Button, G. and N. Casey. 1984. ‘Generating topic: the use of topic initial elicitors’ in J. M. Atkinson and J. Heritage (eds): Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press. Button, G. and N. Casey. 1985. ‘Topic nomination and topic pursuit,’ Human Studies 8: 3–55. Cazden, C. B. 2001. Classroom Discourse: The Language of Teaching and Learning. 2nd edn. Portsmouth: Heinemann. Conti-Ramsden, G. and N. Botting. 1999. ‘Classification of children with specific language impairment,’ Journal of Speech, Language and Hearing Research 42: 1195–204. Dockrell, J. and D. Messer. 1999. Children’s Language and Communication Difficulties: Understanding, Identification and Intervention. London: Cassell.

44

REPAIR PRACTICES IN SSLD

Ridley, J., J. Radford, and M. Mahon. 2002. ‘How do teachers manage topic and repair?’ Child Language Teaching and Therapy 18/1: 43–58. Sadler, J. and K. Mogford-Bevan. 1997. ‘Teacher talk with children with language disorders: four case studies 2,’ Child Language Teaching and Therapy 11: 37–58. Schegloff, E. A. 1972. ‘Notes on a conversational practice,’ in D. Sudnow (ed.): Studies in Social Interaction. New York: Free Press. Schegloff, E. A. 1997. ‘Practices and actions: Boundary cases of other-initiated repair,’ Discourse Processes 23: 499–545. Schegloff, E. A., G. Jefferson, and H. Sacks. 1977. ‘The preference for self-correction in the organization of repair in conversation,’ Language 53: 361–82. Seedhouse, P. 2004. The Interactional Architecture of the Language Classroom. Malden, MA: Blackwell. Sinclair, J. M. and R. M. Coulthard. 1975. Towards an Analysis of Discourse: The English Used by Teachers and Pupils. London: Oxford University Press. Smith, F., F. Hardman, K. Wall, and M. Mroz. 2004. ‘Interactive whole class teaching in the national Literacy and Numeracy Strategies,’ British Educational Research Journal 30/3: 395–411. Skidmore, D. 2000. ‘From pedagogical dialogue to dialogical pedagogy,’ Language and Education 14/4: 283–96. Wells, G. 1993. ‘Re-evaluating the IRF Sequence: a proposal for the articulation of theories of activity and discourse for the analysis of teaching and learning in the classroom,’ Linguistics and Education 5: 1–37.

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

types and learner repair in immersion classrooms,’ Language Learning 48: 183–218. Macbeth, D. 2004. ‘The relevance of repair for classroom correction,’ Language in Society 33: 703–36. Mehan, H. 1979. ‘ ‘‘What time is it Denise?’’: asking known information questions in classroom discourse,’ Theory into Practice 18/4: 285–94. Mercer, N. 1995. The Guided Construction of Knowledge: Talk Amongst Teachers and Learners. Clevedon: Multilingual Matters. Mercer, N. 2000. Words and Minds: How We Use Language to Think Together. London: Routledge. Mercer, N. and K. Littleton. 2007. Dialogue and the Development of Children’s Thinking: A Sociocultural Approach. London: Routledge. Mroz, M., F. Smith, and F. Hardman. 2000. ‘The discourse of the literacy hour,’ Cambridge Journal of Education 30/3: 379–90. Nystrand, M. 1997. Opening Dialogue: Understanding the Dynamics of Language and Learning in the English Classroom. New York: Teachers College Press. Pomerantz, A. 1988. ‘Offering a candidate answer,’ Communication Monographs 55: 360–73. Radford, J. and C. Tarplee. 2000. ‘The management of conversational topic by a ten-year-old child with pragmatic difficulty,’ Clinical Linguistics and Phonetics 14/5: 387–403. Radford, J. and J. Ireson. 2006. ‘Developing language skills through dialogic discourse,’ AFASIC Abstract Spring 1: 3. Radford, J., J. Ireson, and M. Mahon. 2006. ‘Triadic dialogue in oral communication tasks: what are the implications for language learning?,’ Language and Education 20/3: 191–210.

Applied Linguistics 31/1: 45–71 ß Oxford University Press 2008 doi:10.1093/applin/amn047 Advance Access published on 9 December 2008

Style Shifts among Japanese Learners before and after Study Abroad in Japan: Becoming Active Social Agents in Japanese NORIKO IWASAKI School of Oriental and African Studies, University of London

INTRODUCTION Societal interactions are structured by sociocultural norms of communities and are not easy to replicate in foreign language classrooms. Study abroad experiences provide second language (L2) learners with opportunities to observe first-hand how the language is used, to use the language themselves, and to observe how their utterances and performances unfold in subsequent interactions. Hence, studying abroad may facilitate acquisition of the aspects of language that are the most intimately associated with social norms and situations, such as the use of formal and informal styles. Yet, previous studies show that, though learners gain sociolinguistic knowledge while studying abroad, their knowledge and performance diverge from target norms. Barron (2006) found that Irish learners of German switched between the two address forms of ‘you’ (‘intimate/simple’ du and ‘polite/distant’ Sie) haphazardly within a turn after a 10-month stay in Germany. Regan (1995) also found that French learners overuse informal ne deletion in negatives after one academic year in France, which she

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

Previous studies on L2 Japanese sojourners often reported that learners overuse the plain style or haphazardly mix the plain and polite styles upon return. These styles, which are often associated with formal or informal contexts, also index complex social and situational meanings, and native speakers are reported to shift their styles to create desired contexts. In order to better understand L2 development of the use of the plain and polite styles during study abroad, the current study examined the use of the polite/plain styles and style shifts among five English-speaking male students who studied in Japan for one academic year by comparing their performances both quantitatively and qualitatively in oral proficiency interviews before and after they studied abroad. Upon return, three predominantly used the polite style talking to the interviewer (their former teacher), while two primarily used the plain style. Though the quantitative analysis may lead one to conclude that these two students regressed in their pragmatic competence, the qualitative analysis revealed that all five learners gained some understanding of social meanings of the plain and polite styles and became more active social agents who make decisions to shift the styles.

46

STYLE SHIFTS AMONG JAPANESE LEARNERS

(1)

a. Iku. ‘(I) will go.’ b. Iki-masu. ‘(I) will go’.

(2)

a. Kiree da. ‘(it) is pretty.’ b. Kiree desu. ‘(it) is pretty.’

The choice between polite and plain forms needs to be made constantly with the consideration of various social and situational factors. Japanese textbooks generally first introduce the polite verb forms –mas(u) and copula des(u). This is presumably due to the ease of acquisition of polite forms and to the presumed desirability to err on the side of formality rather than informality. While the use of plain forms requires the learning of rather cumbersome morphological rules for negative and past formations [e.g. iku (non-past), ika-nai (negative), itta (past)] for each verb, the polite form formation is consistent across all verbs

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

attributes to insufficient input in formal contexts. L2 Japanese speakers are also reported to overuse the informal style after studying abroad, presumably due to the influence of informal interactions (Marriott 1993, 1995). Therefore, ‘the recurrent finding was that students learn a more informal style during study abroad than at home in the classroom’ (Kasper and Rose 2002: 230). Interestingly, Kinginger and Farrell (2004) found that some advanced learners of French deliberately choose not to conform to the norms of address forms of ‘you’ (intimate tu and polite/distant vous) despite their knowledge. Learners sometimes feel that they do not need to conform to the norm (Barron 2003), or that the norm conflicts with their identity (Dufon 2000; Siegal 1995). Thus, the learning of formal/informal styles is not merely a matter of acquiring the forms and associating them with certain contextual features (e.g. the addressee’s social status). Rather, L2 learners learn to make their own choices as to which forms to use based on their understanding of the forms’ social meanings. Acquiring ways to convey an appropriate demeanor is vital in learning Japanese as an L2. Japanese has no neutral forms of politeness, and speakers cannot easily opt out of choosing the level of politeness—though there are some avoidance strategies speakers employ, such as the use of incomplete sentences (Cook 2006). Finite verbs and copulas are marked to indicate some acknowledgment of social and contextual factors as examples (1a and b) and (2a and b) indicate. The forms in (1a) and (2a) are generally used in informal situations and/or with an addressee with whom the speaker feels close, while (1b) and (2b) are used in more formal situations, most likely with a socially and psychologically distant addressee. The latter is the polite form, which is also referred to as des(u)/mas(u) form, or ‘addressee honorifics’ because it acknowledges distance and formality towards the addressee.1

N. IWASAKI

47

Polite and plain style and style shift What has thus far been called polite style is the style that utilizes polite forms of verbs –mas(u) or copula des(u), which express deference to addressees. The distinction between the polite and plain styles is traditionally viewed as a formal versus informal distinction based on social and situational factors (e.g. social status, presence/absence of intimacy) and is thus explained to L2 Japanese learners. However, the choice of these two styles is not simply a matter of observing social norms based on discernment of the contextual feature. Instead, polite and plain expressions index social meanings (i.e. the nature of relations and identities), such as in-group/out-group orientation (Makino 1983, 2002) and distance/proximity to the self and social role (Cook 1996). The choice between the two depends on ‘beliefs about who should speak deferentially and formally

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

[e.g. iki-masu (non-past), iki-masen (negative), iki-masita (past)]. Furthermore, the consequence of using the polite form in informal situations is generally assumed to be not as adverse as the opposite in which case the speaker may sound rude.2 Learning the appropriate use of the styles was found to be the major obstacle for Australian secondary school students even after a year in Japan. Many overused the plain form (Hashimoto 1993; Marriott and Enomoto 1995; Marriott 1993, 1995). Most of the students in Marriott’s (1995) study used the polite style before going to Japan, but upon return they predominantly used the plain style or haphazardly mixed them. Interestingly, some of the students used a greater proportion of polite forms during role plays and picture descriptions than in interviews with a university instructor, which Marriot attributed to their higher awareness of language forms. Difficulty of acquiring appropriate styles is also reported for L2 Japanese learners who are college/university students or older. Atsuzawa-Windley and Noguchi (1995) found that university students who had previously studied in Japan for various lengths of time did not perform well in aspects related to politeness, overusing the informal plain style. Siegal’s (1995, 1996) in-depth study of four adult women studying in Japan demonstrated the complexity involved in the acquisition of sociolinguistic competence—involving the interplay of race, gender, and social status—which sometimes led to their resistance of using polite expressions. These previous studies show that studying abroad in Japan often leads to learners’ overuse of plain forms. However, these studies might have neglected the social/situational meanings of the styles by assuming that language style is bound to certain contextual features, such as the social status of the interlocutor. Recent work on style shifts among native Japanese speakers demonstrates that they do not always observe socially agreed-upon rules. Rather, they actively create speech styles and shift between them during interactions with the same addressee.

48

STYLE SHIFTS AMONG JAPANESE LEARNERS

THE CURRENT STUDY This study examines the use of polite and plain styles among five male university students from the United States, comparing their use of these styles before and after they studied abroad for a year. This may represent the circumstances in which politeness may not be highly valued by the participants because of the emphasis on informality in North American values (e.g. Althen 1988) and of a general cross-cultural expectation for women to be more polite than men (e.g. Brown and Levinson 1987; Okamoto 2002). By employing the oral proficiency interview (OPI), which complies with the guidelines of the American Council on Teaching of Foreign Languages (ACTFL), this study first assesses the participants’ gains in overall proficiency,

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

to whom, and under what circumstances’ (Okamoto 1999: 58), and the two styles are often mixed. In predominantly polite discourse, a speaker may switch to the plain style to state summaries, facts, convictions, feelings (e.g. Makino 1983, 2002), or to provide background information that is subordinate to the main idea (Ikuta 1983; Maynard 1991). Japanese speakers also express empathy (Ikuta 1983), spontaneous assertion of the speaker’s thought (Cook 1996), soliloquy, emotional, exclamatory reactions or closeness/friendliness (Okamoto 1999) by switching to the plain style. Okamoto (1999), for example, found that in an interaction between a professor and a former student the student occasionally used the plain style for exclamatory, emotional, or soliloquy-like expressions despite their social status and age difference. The professor also mixed the styles to maintain a degree of formality and to show friendliness. Likewise, Cook (2006) found that, in academic consultation sessions, a student shifted to the plain style to index his conviction to himself or to co-construct an idea with the professor by supplying the continuation of the professor’s utterance. The current study shares Okamoto’s (1999) and Cook’s (2006) theoretical perspective of ‘indexicality’ that the plain/polite forms (indirectly) index social meanings (and thus its use is a social action) and that Japanese speakers manipulate speech styles ‘in order to create a desired context, in particular, preferred interpersonal relations and identities’ (Okamoto 1999: 70), and ‘every move that the speaker makes is his or her choice’ (Cook 2006: 288). Native speakers vary in their beliefs and understanding of politeness and in their preferred modes of presenting their social roles, and they use this knowledge to actively create speech styles. Hence, the development of the use of the plain and polite forms involves not only an understanding of their respective social meanings but also the making of spontaneous choices of styles with the consideration of various contextual and situational factors. Study abroad experiences give L2 learners opportunities to interact with various interlocutors, and they may acquire speech styles through the process of language socialization by becoming active members in the society (e.g. Schieffelin and Ochs 1986).

N. IWASAKI

49

and then investigates how their use of the plain/polite styles changed as a result of studying abroad. In particular, it focuses on how L2 Japanese learners switch between the two styles before and after studying abroad.

Participants

Procedures and analyses Before studying abroad, the students participated in OPIs conducted by a certified OPI tester, who was the primary Japanese teacher of Henry, Peter, Sam, and Alan in the two semesters preceding their study abroad. In the fall after they returned, they participated in OPIs again. The tapes and the interviewer’s ratings of the post-study-abroad OPIs were sent to ACTFL to receive official ratings to avoid potential biases of the tester who had knowledge of the participants’ pre-study-abroad performances. The learners’ gains in overall oral proficiency were assessed by these OPI ratings. Because the criteria for OPI ratings do not include the use of appropriate styles (except for Superior), gains in these ratings mostly indicate the learners’ gains in other aspects of their performance (functions/tasks, content, text types, accuracy). The OPIs before and after study abroad are referred to as pre-OPI and post-OPI below. The OPIs were transcribed and analyzed both quantitatively and qualitatively. The appropriate base style for an interviewee to use in such interviews is the polite style, especially because the interviewer is a former teacher, who is clearly older than the participants. Role plays are part of the OPI protocol, but the types of role plays varied depending on the proficiency levels of the participants. For interviewees approaching Superior-level, the Japanese OPI requires formal and informal role plays to elicit both styles. Thus, the most proficient speakers of the five, Peter and Sam, were given role plays for both styles. In the quantitative analysis, predicates produced in OPIs, excluding the role play segments, were coded for the forms used: the polite form, plain form, omission of predicate/copula (in the case of nominal predicates). The proportions of each type were computed. The sentence-final matrix predicates and coordinate/subordinate predicates in which polite forms can be used without sounding excessively polite were considered.3 Further, these non-matrix clauses were broken down into two types to aid in the analysis. This is because whether and to what extent the plain form in non-matrix clauses has social

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

Five male students, referred to by pseudonyms as Alan, Henry, Peter, Sam, and Greg, from a northeastern US state university voluntarily participated. Their ages at the time of departure ranged from 19 to 21. One participant, Greg, had studied Japanese for 1 year (6 hours of instruction per week for two semesters) before studying abroad, and the other four had completed 2 years of Japanese instruction (6 hours of instruction per week for four semesters). They attended four different universities in Japan. Henry studied in Aichi, Peter in Ibaraki, Sam in Tokyo, and Alan and Greg studied at the same institution in Hyogo.

50

STYLE SHIFTS AMONG JAPANESE LEARNERS

FINDINGS Gains in overall oral proficiency The participants’ OPI ratings were rather high even before studying abroad, most likely because they were highly motivated, enough to opt to study abroad. Four improved their proficiency by 1 or 2 sub-levels as indicated in Table 1. Henry, whose proficiency was at the Intermediate-Mid in pre-OPI, remained at the same level. Interestingly, despite the fact that OPI criteria do not include the appropriate speech styles for Novice to Advanced levels, those who rated higher demonstrated more appropriate and/or active use of styles and shifts. For example, as discussed below in the quantitative analyses, those who were rated higher (Alan, Sam, and Peter) appropriately chose the polite style as the base style. Moreover, as discussed in the qualitative analysis, Peter, who was rated the highest, actively shifted styles in a target-like manner rather than simply conforming to the appropriate polite base style.

Quantitative analysis: proportions of use of the polite and plain styles Table 2 shows the types of predicates each participant used. Table 3 shows breakdowns of the plain forms into three types of clauses. In terms of the proportion of the polite forms, the five participants did not differ greatly in

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

meanings that contrast with the polite form depends on the degree of subordination of the clauses in question (Makino 1983; Cook 2001). The plain/polite forms used in the sentence-final matrix predicates primarily determine the level of politeness. The plain forms in highly independent clauses (ending with ga ‘but’, kedo ‘but’, and kara ‘because’) index some informality, but they may not necessarily express informality in clauses possessing marginal independence (ending with node ‘because’ and si ‘and’). For the sentence-final nominal predicates, such as yo-nensei da ‘I am a senior’, the omission of the copula da as in yo-nensei 6 0 is common in spoken Japanese. There are also instances of omission of entire predicates such as in zenzen ‘not at all’. These omissions are tallied as predicate/copula omissions; they are equivalent to the plain forms in degree of politeness/formality. In the qualitative analysis, the students’ style shifts in sentence-final matrix predicates were examined (leaving the examination of non-matrix clauses to future studies due to space limitations). The current analysis primarily focuses on the interaction with the interviewer taking her authentic role (university faculty), but performances in the role plays are also discussed. Note, however, these styles may index multiple social meanings, and their meanings may be occasionally ambiguous (Okamoto 1999). This is because the functions discussed above are not mutually exclusive. Thus, multiple plausible functions realized by the form are noted, though they may not be the only possibilities.

N. IWASAKI

51

Table 1: Gains in OPI ratings Pseudonyms

Length of prior instruction (years)

Before study abroad

After study abroad

Alan Henry Peter Sam Greg

2 2 2 2 1

Intermediate-High Intermediate-Mid Advanced-Low Intermediate-High Intermediate-Low

Advanced-Low Intermediate-Mid Advanced-High Advanced-Mid Intermediate-Mid

Pre-study abroad OPI Polite n (%)

Plain n (%)

Post-study abroad OPI

Copula Repair Total Polite omission n (%) n (%) n (%)

Plain n (%)

Greg

74 (79)

9 (10) 8 (9)

3 (3)

94

Henry

73 (72) 16 (16) 8 (8)

5 (5)

102

13 (10)

Alan

128 (87) 14 (10) 1 (1)

4 (3)

147

172 (86)

21 (11)

Sam

107 (84) 16 (13) 2 (2)

2 (2)

127

106 (85)

17 (14)

Peter

134 (80) 27 (16) 4 (2)

2 (1)

167

127 (70)

44 (24)

Copula Repair Total omission n (%) n (%) n (%)

96 (73) 23 (17)

0 (0)

132

31 (20) 103 (67) 20 (13)

1 (1)

155

5 (3)

1 (1)

199

2 (2)

0 (0)

125

10 (5)

1 (1)

182

Table 3: Clauses where the plain forms were used Pre-study abroad OPI

Greg Henry Alan Sam Peter

Post-study abroad OPI

Matrix clauses

kara, kedo, ga clauses

node, si clauses

Total

Matrix clauses

kara, kedo, ga clauses

node, si clauses

Total

0 5 1 0 3

0 11 9 8 15

9 0 4 8 9

9 16 14 16 27

83 67 10 3 17

12 36 9 2 13

0 0 2 12 14

95 103 21 17 44

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

Table 2: Types of predicates produced

52

STYLE SHIFTS AMONG JAPANESE LEARNERS

(3)

(Greg, Pre-OPI)

1

I:

De, ano, hima-na zikan-mo arimasu ka. and well free time-also have-NONPAST-AH Q ‘And do you also have free time?’

2→

G:

Hima-no toki, aa, video geemu-o suru, aa, o suru n desu. free when uh video game-ACC play uh ACC N COP ‘When I have free time, uh I play ... uh it is that I play video games.’

Upon return, Greg used only 13 polite forms in 132 predicates, including eight uses of the epistemic modal marker desyoo with rising intonation, three uses of the polite form of the verb omoimasu ‘(I) think’. Greg did not use daroo, the informal equivalent of desyoo. In fact, Cook (2007) does not consider these two forms as polite and plain counterparts because desyo(o) is used in most informal situations as well. Taking this into consideration, Greg used only five polite forms including three instances of omoimasu in post-OPI. Likewise, Henry also primarily spoke in the polite style in pre-OPI, with a slightly higher tendency to use plain forms than Greg. He used 73 polite forms in 102 predicates. He made five repairs, suggesting a conscious effort to speak in the polite style while resisting the inclination to use the plain style as shown in an exchange about a student organization he belongs to in excerpt (4) line 3. (4)

(Henry, Pre-OPI)

1

I:

Membaa-wa takusan iru n desu ka. member-TOP many exist-NONPAST N COP-AH Q ‘Are there many members?’

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

pre-OPIs: Greg 79 per cent, Henry 72 per cent, Alan 87 per cent, Sam 84 per cent, and Peter 80 per cent. All participants used the polite forms as their base style. Upon return, Alan, Sam, and Peter primarily maintained the polite style (86 per cent, 85 per cent, and 70 per cent), but Henry and Greg used the plain forms as their base style. Their proportions of polite forms were only 10 per cent and 20 per cent in their post-OPIs. In pre-OPI, Greg mostly used polite forms (74/94), and his plain forms were limited to nine uses in node and si clauses (which does not necessarily index informality), eight copula/predicate omissions (five of which were moo iti-do? ‘once again?’ requesting repetition), and three repairs. He repaired his use of the plain form to the polite form as shown in excerpt (3) line 2. In what follows, plain forms are underlined and polite forms are bolded. (‘I’ refers to Interviewer; pseudonym initials refer to participants.)

N. IWASAKI

2

H:

53

Membaa? Ano, [inaudible]-no hito-ni, ano, nan-nin-kurai, a, zyuu-go-nin-gurai member well

-GEN person-DAT well how-many-CL 15-CL-about

ano, takusan aru, ano takusan hito zyanai to omou kedo, well many exist well many people COP-NEG QT think but ‘Member? Well, [inaudible] people, well, how many, about 15, uh, well, there are many...well, I don’t think there are many, but,’ taisetu-na hito da kara,

3

→

totemo omosiroi, omosiroi yo, omosiroi desu. very interesting very interesting-SFP(assert) interesting COP-AH

‘Because they are important people, it (the club) is very interesting, interesting, it is very interesting.’

In post-OPI, his polite forms dramatically decreased. He used only 31 polite forms in 155 predicates, including one desyo, and he no longer corrected himself to be polite. Instead, he once adjusted a polite form to the plain form (his base style in post-OPI) as in excerpt (5) line 4. (5)

(Henry, Post-OPI)

1

I:

Sotugyoo-site kara donna sigoto-ga sitai n desu ka. graduate-GER after what-kind work-NOM want-to-do N COP Q ‘After graduation, what kind of job do you want to do?’

2

H:

(laugh) zense, zensen, wakarimasen. [fragment] at-all know-NEG-AH ‘I don’t know at all.’

3

I:

zenzen wakarimasen ka. at-all know-NEG-AH Q ‘You don’t know at all?’

4→

H:

Honto ni, ano, kangaete imasen, kangaete nai. really uh think-GER has-NEG-AH think-GER NEG ‘Really, well, I haven’t thought, haven’t thought about it.’

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

important people COP-NONPAST because

54

STYLE SHIFTS AMONG JAPANESE LEARNERS

Qualitative analysis The quantitative analysis above showed that Greg and Henry’s choice of base style was no longer the polite style in post-OPIs, while Alan and Sam predominantly maintained the polite style. Peter increased his use of plain forms in his predominantly polite style. Their style shifts are examined below. Though it is likely that the interviewer’s choice of style would influence the interviewee’s style choice, the current interviewer consistently used polite forms and only used one-word utterances for clarification/ confirmation and plain forms in role plays (when playing a friend or a child role).

Greg and Henry In pre-OPI, Greg consistently used polite forms, and all plain forms were in –si/node clauses. The copula was occasionally omitted when he requested or provided clarification. However, there was one shift in a role play to invite a friend to go to a movie. He spoke in the polite style, but switched to the plain style in a soliloquy-like, exclamatory utterance in excerpt (6) line 3. (6)

(Greg, Pre-OPI)

1

G:

A, kyoo-no yoru-wa hima desu ka. ah today-GEN night-TOP free COP Q ‘Uh, are you free tonight?’

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

Henry’s mix of the styles appeared to be haphazard, but the qualitative analysis below reveals that his style shift is not entirely random. Like Greg, he had a slight tendency to use the polite forms for expressions of epistemic stance (omou ‘think’, wakaru ‘understand’, and kanziru ‘feel’). These verbs accounted for 6 of 19 verbs expressed in the polite forms out of 31 polite forms, though he also used these verbs in the plain forms. Unlike Greg and Henry, Alan and Sam spoke in the polite style both in pre- and post-OPIs. Their plain forms were mostly in non-matrix clauses: Alan used plain forms 13/14 (non-matrix/of all plain forms) in pre-OPI and 11/21 in post-OPI, and Sam 16/16 in pre-OPI and 14/17 in post-OPI. Peter, who received the highest OPI ratings both in pre- and post-OPIs, also mostly spoke in the polite style in both OPIs. Interestingly, however, his use of polite forms decreased in post-OPI from 80 per cent to 70 per cent, and his plain forms in matrix predicates increased from 3 to 17 instances. The question, then, is whether he shifted the styles in ways similar to how native speakers do in ways that reflect the social meanings of the forms.

N. IWASAKI

2

I:

55

Hai, zikan arimasu yo. Kyoo-no yoru dattara. yes time have-NONPAST-AH SFP(assert) today-GEN night COP-CONDITIONAL ‘Yes, I have time—tonight.’

3→

G:

um. yokatta. um Supaidaa-man-o mitai desu ka. um good-PAST um Spiderman-ACC watch-want COP-AH Q

Though this is an effective shift, it is unclear whether he had acquired the meaning of an in-group self-directed shift for soliloquy/exclamatory utterance because this was the only instance of shift. Possibly, he learned the sequence yokatta! as an equivalent to the English ‘Great!’. Upon return, his base style was the plain style, and his switches from the plain to polite style appeared to be rather haphazard. He used only two polite forms, other than eight instances of desyo(o) and three omoimasu ‘I think’. The use of the polite form for the verb omou is not lexically fixed for Greg as he also used the plain form omou six times in post-OPI. The remaining two instances used the copula desu and -masu as shown in excerpts (7) and (8), lines 5 and 4, respectively. (7)

(Greg, Post-OPI)

1

I:

Gureggu-san-wa, ano, ima daigaku-no nan-nen-sei na n desu ka. name-TOP

well now university-GEN what-year-student COP N COP-NONPAST Q

‘Greg-san, uh, which year are you at the university now?’

2

G:

Etto, ima, yo-nensei-gurai da to omou. well now 4th-year-about COP QT think ‘Well, now, I think I am about a senior.’

3

Hontoni roku-nensei-gurai kedo... really 6th-year-about but ‘....Actually, about the sixth year, but...’

4

I:

Yo-nensei-gurai na n desu ka. 4th-year-about COP N COP-NONPAST Q ‘You are about the fourth year?’

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

‘Great! Uh, do you want to see Spiderman?’

56

5 →

STYLE SHIFTS AMONG JAPANESE LEARNERS

G:

Un. Roku-nen-gurai benkyoo-site ita ga...ima, etto, sono, ima yo-nensei desu. yeah 6-year-about study-GER AUX but now uh well now 4th-year COP-NONPAST AH ‘Yeah. I studied about six years, but... Now, uh, well, now I am a senior.’

(8)

(Greg, Post-OPI)

1

I:

Etto, ano, ryuugaku-no koto-toka takusan Gureggu-san-ni situmon-simasita kedo, uh well study-abroad-GEN about-such many name-DAT question-PAST-AH but Gureggu-san, watasi-ni situmon-wa arimasu ka. name me-DAT question-TOP have-NONPAST-AH Q SFP ‘I asked you a lot of questions about your study abroad, but do you have any questions for me?’

2

Nan-demo ii desu yo. what-ever good COP-NONPAST AH SFP(assert) ‘Any questions will be alright.’

3

G:

Etto, Alan-wa ikkyuu-o uketai kara, etto, kyooryoo*-o sagasite iru. well Alan-TOP level-one-ACC take-want because uh textbook-ACC look-for AUX ‘Well, because Alan wants to take the level 1 proficiency exam, he is looking for [textbooks].’(*Greg probably meant ‘textbook’ [kyookasyo])

4 →

Etto, ikkyuu-no nooryoku-siken motte imasu ka. well level-one-GEN proficiency-exam have Q ‘Do you have a level 1 exam?’

5

I:

Hai, mae-no hurui no-wa motte imasu kedo. yes past-GEN old one-TOP possess-NONPAST-AH but ‘Yes, I have the old versions.’

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

Greg predominantly used plain forms in this sequence as well, and his motivation for the abrupt switch to the polite form in line 5 does not seem to utilize the social meanings of the polite form, such as deference or presentation of self. However, when a situation clearly calls for expressions of deference, he uses the polite form. In excerpt (8), he asks the interviewer a question to request something from her, when she gave Greg an opportunity to ask questions as part of OPI protocol for Novice-High- to Intermediate-level candidates.

N. IWASAKI

6

G:

57

Un, sore-demo ii. yeah that-COP-GER-also good-NONPAST ‘That will be fine.’

(9)

(Greg, Post-OPI)

1→

G:

Etto, sensei, moosiwake arimasen ga, uh teacher excuse have-NEG-AH but tyotto siken-ni maniawanakatta, aemasen desita. little exam for make-it-on-time-NEG-PAST POTENTIAL-NEG-PAST ‘Uh, teacher, I am very sorry. I did not, could not make it to the exam.’

2

→

Tesuto-ni iku toki, watasi-no kuruma-wa kowarete simaimasita. test-to go when my car-TOP break-GER end-up-PAST-AH ‘When I was going to the exam, my car broke down.’

3

I:

Demo siken-no mae-ni renraku-ga arimasen desita nee... but exam-GEN before-at contact-NOM exist-PAST-AH SFP ‘But you did not notify me in advance...’

4 →

G:

Unn, etto, zikan-ga arimasen desita. Umm well time-NOM have-NEG-PAST-AH ‘Umm, uh, I did not have time.’

5 →

Etoo, soreni, eto, keitai-o motte imasen kara, etto, renraku-dekimasen desita. well also uh cell-phone-ACC possess-NEG because well contact-POTENTIAL-NEGPAST

‘Well, besides, uh, because I do not have a cell phone, well, I could not contact you.’

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

Unlike his other utterances, in which he primarily talks about himself (his hobby, experiences in Japan, and his future goal), the agent of the verb motu ‘possess’ in line 4 is the interviewer. Moreover, when he performed a role play to ask his teacher to arrange a make-up exam, he consistently used the polite style as in excerpt (9). First, he carefully chose a very polite expression for apology and quickly corrected his unintended use of a plain form to the polite form, and continued to use the polite style in line 2.

58

STYLE SHIFTS AMONG JAPANESE LEARNERS

(10)

(Henry, Pre-OPI).

1

I:

Hitori erande, donna hito ka osiete kudasai. one-person select-GER what-kind person Q tell-imperative-AH ‘Then could you select one and tell me what kind of person s/he is?’

2

H:

Eto, supeingo-ga hanasemasu. Yoku hanasemasu. well Spanish-NOM speak-POTENTIAL-AH well speak-POTENTIAL-AH ‘Well, (she) can speak Spanish. (She) can speak very well.’

3 →

Sorekara, kao-ga kawa, kawaii. also

face-NOM [fragment] cute-NONPAST

‘Also, (her) face is cut...cute.’

Henry’s use of the plain form kawaii ‘(her face) is cute’ in excerpt (10) line 3 is an effective shift to express an emphatic, emotional reaction to what is being described. His pre-OPI suggests that, before studying abroad, he understood that the appropriate base style to address a teacher/interviewer is the polite style and that some style mixes are possible. In post-OPI, however, he used the plain style as the base style. He shifted to the polite styles 31 times, including one instance of desyo and six instances of expressions of epistemic stances. Some of his shifts to the polite style appeared to be rather haphazard as shown in his descriptions of how he found a house to rent in excerpt (11). Henry first corrected his plain style utterance to polite in line 3, but abruptly switched back to the plain form in line 4. It is unclear what prompted this switch.

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

His utterances in the role play clearly indicate that he can use the polite style consistently in both matrix and kara clauses if he chooses to do so. He uses the polite style in situations that call for expressing deference. This is in contrast to his use of polite forms in pre-OPI in which he may have been passively observing the social norms that he learned in classes. Though Greg’s base style no longer conforms to the norm, he was actively making choices as to which style to use. Like Greg, Henry primarily used the polite style in pre-OPI, and changed his base style to the plain style in post-OPI. In pre-OPI, he used only five plain forms in matrix predicates, most of which can be considered similar to native speakers’ shifts:  situmoi-nikui by which he probably meant setumei-sinikui ‘it is difficult to explain’ as a soliloquy-like utterance and empathic expressions, as shown in excerpt (10) line 3. Prior to this sequence, Henry said that in his club there are many people who are important to him.

N. IWASAKI

(11)

(Henry, Post-OPI)

1

I:

59

Dooyatte, sono sumu tokoro-o sagasitari, issyoni sumu hito-o sagasita n desu ka. how uh live place-ACC look-for-such together live people-ACC look-for N COPNONPAST-AH Q

‘How did you look for the place to live and people to live with?’ 2→

H:

ano staffu membaa-ga sono uti-ga aiteru tte, i, itte ta kara,

3

sono ie-wa kurabu-no miitingu-de, sono ie-ga aiterutte itte masita kara, that house-TOP club-NOM meeting-at that house-NOM is-vacant-AH because

4

sagasu hituyoo-ga nakatta. search need-NOM exist-NEG-PAST

‘Because that staff member was saying that the house was vacant, because they were saying that the house is vacant at the club meeting, there was no need to look for it.’

However, closer examination reveals that some shifts can be explained. In excerpt (12), where Henry describes his roommates, the information provided by the plain forms of hiku ‘play’ in line 3 is background information subordinate to the main idea (Makino 1983; Maynard 1991). (12)

(Henry, Post-OPI).

1

I:

Yuniiku-na hitotati desita ka. unique people COP-NONPAST-AH Q ‘Were they (roommates) unique people?’

2

H:

Aa, un, ma, minna, ongaku su ongaku, uh, gakki-o, gakki, hikemasita, minna. well uh everyone music..uh musical-instrument-ACC play-POTENTIAL-PAST-AH everyone ‘Yes, uh, well, everyone lik..played music, music, musical instruments, everyone.’

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

that staff member-NOM that house-NOM vacant-that said because,

60

STYLE SHIFTS AMONG JAPANESE LEARNERS

3→

Hutari-ga gitaa hikeru... Hitori-ga piano hikeru, hiketa kara, two-NOM guitar play-POTENTIAL one-NOM piano play-POTENTIAL play-POTENTIALPAST because

minna issyo-ni yoku onngaku-o kiitari, ongaku-o tuku, tukuttari simasita. everyone together often music-ACC listen, music-ACC compose did-AH

did such activities as listening to music together and comp, composing music.’

Moreover, in post-OPI, he consistently switched to the polite style when asking the interviewer questions, such as in excerpts (13) and (14), lines 2 and 4. (13)

(Henry, Post-OPI) I:

de, uti-no sigoto-wa minna de, ano, kootai de sita n desu ka. and house-GEN work-TOP all by well turn do-PAST N COP-NONPAST-AH Q ‘Did you take turns doing house chores?’

H:

Kootai tte nan desu ka. (Taking-turns) that what COP-AH Q ‘What is kootai?’

(14)

(Henry, Post-OPI)

1

I:

Nanika Henrii-san-kara watasi-ni situmon-sitai koto-ga arimasu ka. something name-from me-DAT want-to-ask thing-NOM have-NONPAST-AH Q ‘Are there anything that you would like to ask me about?

2 →

H:

Ima kariforunia ni sunde imasu ka. now California in live-GER AUX-AH Q ‘Are you living in California now?’

3

I:

Hai. ‘Yes.’

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

‘Two can play the guitar, and one can play could play the piano and so we often

N. IWASAKI

4→

H:

61

Sono daigaku-wa doo desu ka. that university-TOP how COP-AH Q ‘How is the university?’

(15)

(Henry, Post-OPI)

1→

H:

Ano, watasi-no mado-ga kowaremasita. uh my window-nom break-AH-PAST ‘Uh..my window broke.’

2

I:

Aa, soo desu ka. ah so COP-NONPAST-AH Q ‘Is that so?’

3→

H:

Ano, kodomo-ga yakyuu-o yarinagara, ano, sono beesubooru-ga, well child-NOM baseball-ACC play-while, well that baseball-NOM ano, watasi-no mado-o butukete, kowaremasita. uh my

window-ACC hit-GER break-AH-PAST

‘Well, while a child was playing baseball, his ball hit my window, and it broke.’ 4

Doo sureba ii desu ka. what do-CONDITIONAL good COP-AH-NONPAST Q

‘What would be good to do?’ 5

I:

Anoo, donnahuu-ni kowarete iru n desu ka. well what-way-in break-GER AUX N COP-NONPAST-AH Q

‘Well, how is it broken?’ 6

H:

Akerarenai si, mado-no gurasu-ga, moo, nanka, un, open-POTENTIAL-NEG and window-GEN glass-NOM already somehow um mado-no gurasu-ga, ko. to-iu-ka, uun, nai.

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

Though some of Henry’s shifts seem haphazard, he speaks in the polite style when he asked the interviewer questions. Indeed, if he had produced the questions as in excerpts (13) and (14) in the plain style, he would have sounded severely inappropriate. Moreover, like Greg, Henry appropriately switched to the polite style when he performed a role play. He needed to call an apartment manager to request a window repair, and he mostly maintained the polite style as in excerpt (15), lines 1 and 3.

62

STYLE SHIFTS AMONG JAPANESE LEARNERS

Window-GEN glass-NOM [fragment] rather umm exist-NEG-NONPAST Gurasu-ga nai, ima. glass-NOM exist-NEG-NONPAST now

‘I can’t open it, and the window glass, well, the window glass is bro.., or rather, is gone now’.

Alan and Sam The quantitative analyses revealed that Alan and Sam were the most consistent in their use of polite forms. In his pre-OPI Alan used the plain form only once in a matrix clause as in excerpt (16) line 3, but the predicate in question is more likely an embedded predicate in a disfluent sequence. Prior to excerpt (16), Alan stated that after practice one can do amazing things in juggling. (16)

(Alan, Pre-OPI)

1

I:

huun, sugoi koto tte donna koto desu ka. hmm amazing thing QT what-kind thing COP-NONPAST-AH Q

‘Hmmm, what sort of things are those ‘amazing things’? 2

A:

Anoo..aa...booru-o, koo, takusan, aa, jaguru-suru toka, well ah ball-ACC this-way many ah juggle such-as

‘Well..uh, like..juggling many balls, and..’ 3 →

aa, tatoeba, ima, watasi-wa, oo, itu..itutu? itutu-no booru-o jaguru-suru ah for-example now I-top [fragment]...five five -GEN ball-ACC juggle

‘Uh, for example, now, I juggle fiv..five? five balls’ 4

A:

ano, rensyuu simasu. well practice-do-NONPAST-AH

‘...well, I do practice.’

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

In excerpt (15) line 6, Henry returned to the plain form in the -si clause, akerarenai, and in the matrix predicate nai. Otherwise, he mostly used the polite style in this role play. Thus far, we have qualitatively examined the choice of styles by Greg and Henry, whose post-OPI base style was the plain style, diverging from the native norm. Striking similarities were found between the two. They both spoke in the polite style when expressing deference is highly needed—in asking for information and making requests.

N. IWASAKI

63

(17)

(Alan, Post-OPI)

1

I:

De, donogurai-no aida ryokoo-sita n desu ka. and how-long-GEN duration travel-PAST N COP-NONPAST-AH Q ‘So, how long did you travel?’

2

A:

A, ikka..kurai, ikkagetu-gurai, toriaezu, ano, nihonkai-made, aru..arukimasita. ah [fragment]..about one-month-about first well Japan-Sea-to walk-PAST-AH ‘Ah, about one, one month, first I wal..walked to the Japan Sea.’

3

Are-wa sugokatta desu. that-TOP amazing-PAST COP-AH ‘That was amazing.’

4

I:

Sugoi desu nee. amazing COP-NONPAST-AH SFP(exclamation) ‘That’s amazing.’

5→

A:

un, de, tyotto tookatta. yeah and a bit far-PAST ‘Yeah, and that was a bit far.’

Like Alan, in pre-OPI Sam predominantly used the polite style and did not shift to plain forms in matrix clauses. Though there was one instance of the plain form in a matrix-like clause shown in excerpt (18) line 2, this sequence seems to be incomplete. He is describing his audition to become a music major at his university.

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

His disfluency made the plain form suru at the end of line 3 look like a sentence-final verb; however, it is likely that instead he was saying itutu-no booru-o jaguru-suru rensyuu simasu ‘I practice juggling five balls’. Put simply, Alan probably did not shift styles in matrix predicates at all in pre-OPI. In post-OPI, Alan did use plain forms in matrix predicates (10 instances). His shifts resemble native speakers’ shifts; many of them are soliloquy-like. He is thinking to himself, using such expressions as doo ieba ii ‘how shall I say it’, and . . . tte ieba ii kana ‘that’s probably how I should describe it’, or evaluating past events while recalling them. In excerpt (17), he first informed the interviewer that his travel was amazing in the polite style in line 3, and recalling his experience, he makes an exclamatory remark as to how far it was in the plain style in line 5.

64

STYLE SHIFTS AMONG JAPANESE LEARNERS

(18)

(Sam, Pre-OPI)

1

I:

Oodisyon-to iu no-wa donna koto-o sita n desu ka. audition-QT say N-QT what-kind thing-ACC do-PAST N COP-NONPAST-AH Q ‘What kinds of things did you do in the audition?’

2 →

S:

Eto, senkoo, unn, ongaku-no senkoo-o saseru tame-ni, eto

sensei-no mae-ni, zibun-no, zibun-no, a, itiban tokui-na, aa, gakki-o hiite, aa, teacher-GEN front-in self-GEN self-GEN best good musical-instrument-ACC playGER ah

sono ato-de, sensei-ga saseru, aa, saseru, aa... that later-at teacher-NOM do-CAUSATIVE-NONPAST ah do-CAUSATIVE-NONPAST AH ‘Well, in order to major...let students major, (they make them) to play their own, own, favorite musical instrument in front of the teacher, and after that, the teacher let.. let.. ah...’ 3

S:

Ha....hairu ka kimemasu. [fragment] enter whether decide-NONPAST-AH ‘... decides whether the student gets in.’

In contrast, in post-OPI, Sam did shift his style three times when he was talking to himself with exclamation or emotion as in Oo sore-wa muzukasii ‘oh, that’s difficult (to explain)’, itiban tanosikatta ‘it was the most enjoyable’, and logical consequence that reflects his thought process when stating his opinion as in excerpt (19) line 2. Prior to excerpt (19), the interviewer mildly challenged Sam’s opinion in favor of a recent change from seniority-based to merit-based employment in Japan, stating that while some people get pay raises, some are laid off due to this merit system. (19)

(Sam, Post-OPI)

1

S:

Eeto, warui syoku-nin-wa kaiko-site ii n zyanai desu ka.. to omoimasu kedo. well bad employee-TOP layoff-GER good COP-NEG-AH Q COMP think-NONPAST-AH but

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

well major um music-GEN major-ACC do-CAUSATVE-NONPAST purpose-of uh

N. IWASAKI

2 →

65

Eto, kaisya-ni warui eikyoo-o oyobosu kamo-sirenai. well company-on bad influence-ACC give may-NONPAST

‘Well, isn’t it okay to layoff bad employees? I think... Otherwise, it might affect the company negatively.’

(20)

(Sam, Pre-OPI)

1

S:

A, konnitiwa, aa, watasi-wa tyoto..mondai-ga aru n de..da. ah hello ah

I-TOP little problem-NOM exist COP-GER..COP-NONPAST

‘Ah, hello, uh, I hav, have a problem.’ 2

I:

Mondai? ‘Problem?’

3

S:

unto Kyoko-san-no kuruma-o unto, unten-site iru toki, kootuu ziko-ga atta n da. um [name]-GEN car-ACC um drive-GER AUX when traffic accident-NOM occur COPNONPAST

‘Um, when I was driving your car, I had a car accident.’ 4

Watasi-no sei dewa nakute, hoka-no untensyu-wa yopparatte ita to omoimasu. I-GEN fault COP-NEG

other-GEN driver drunk was COMP think-NONPAST-AH

‘It wasn’t my fault, I think the other driver was drunk.’ 5

Eto, aa, bampaa-ga aa bampaa-wa tyotto kowarete ita. Uh ah bumper-NOM ah bumper-TOP a little broken was ‘Uh..uh, the bumper, bumper was a little broken’

6

Sore igai, aa, sore igai wa daizyoobu de, da. that other ah other-TOP all-right COP-GER COP-NONPAST ‘But other, other than that, all right, it is all right.

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

In line 2, he spontaneously states his thought using the plain form. Alternatively, the segment expressed in the plain form may be an increment. In other words, background reasoning for his stance (i.e. it is alright to layoff those who are incapable), which could have been stated earlier (e.g. before kaiko-site mo ii n zya nai desu ka), may have been added after the final predicate, omoimasu kedo. Sam was given a role play that called for the plain style in pre- and postOPIs. His performances in the role plays revealed his understanding of the social meanings of both styles. In pre-OPI, he was instructed to talk to a friend about a car accident he had in his friend’s car. Though he primarily used the plain forms, he shifted the plain forms as in excerpt (20).

66

STYLE SHIFTS AMONG JAPANESE LEARNERS

7

Eto, aa, watasi-ga naosu, aa, naosu okane-o aa harawasete kudasai. uh ah I-NOM repair ah repair money-ACC ah pay-CAUSATIVE IMPERATIVE-AH ‘Uh..please let me pay the money to repair it.’

Peter Peter increased the use of plain forms in matrix predicates from 2 to 17 instances. The qualitative analysis revealed that most of his shifts resemble reported native speakers’ strategic shifts. In pre-OPI, he largely maintained the polite style and shifted to the plain style only twice. One is a soliloquylike expression sore-wa muzukasii ‘It’s difficult (to explain)’, and the other is to state his stance by choosing from nominalized propositions that the interviewer listed—whether his teacher, who adopted the Suzuki method of piano instruction, valued ‘being creative’ or ‘imitating teachers’, as in excerpt (21). (21) 1

(Peter, Pre-OPI) I:

De, sono Suzuki-sensei-no hoohoo, toiuka, Suzuki Program-dewa, and that Suzuki-teacher-GEN method rather Suzuki Program-in-TOP dotira-o daizi-ni suru n desu ka, kurieitibu-na koto to, which-ACC value-COP do N COP-NONPAST-AH Q, creative N sensei-to onazi yoo-ni, ma, sensei-no mane-o suru koto to. and teacher-with same like well teacher-GEN imitation-ACC do N and ‘So, in Suzuki-sensei’s method, or Suzuki Program, which one do they value more, being creative or doing the same.. imitating the teacher?’

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

Sam was using both the plain forms atta n da and daizyoobu da and the polite forms omoimasu and harawasete kudasai in this role play. Though not all native speakers necessarily switch styles in the same way, this is the kind of occasion in which native speakers manipulate the styles to sound apologetic. Sam’s performance in pre-OPI indicates that he understood the general social meanings of both the polite and plain styles prior to studying abroad, as he appropriately switched from the plain style to the polite in his role play. In post-OPI, he also switched appropriately from the polite to plain as discussed above, and in a role play where he talked to a visitor’s 4-year-old child in the plain style to be friendly and approachable. In pre-OPIs, Alan and Sam (except for Sam’s role play) were passive observers of the alleged social norms and only used the polite style. Upon return, they voluntarily switched from the polite to the plain styles, which cautious speakers would choose to do only if they have the knowledge of social meanings of the shifts.

N. IWASAKI

2

P:

67

Sensei-no mane-o suru, motiron. Teacher-GEN imitation-ACC do-NONPAST of-course ‘Imitate the teacher, of course.’

(22)

(Peter, Post-OPI)

1

I:

Dooiu koto-ni nareru koto-ga dekinakatta n desu ka, hazime wa. what-kind things-to accustom N-NOM can-NEG-PAST COP-NONPAST-AH Q, first-TOP ‘What kinds of things could you not get used to, at first?’

2

P:

hazime-wa, ma, sugoi, sonnani karada-o tukau-no-wa narete inakatta si, first-TOP uh really that-much body-ACC use-N-TOP accustom-GER AUX-NEG-PAST and ato-wa, iroina sinpaigoto-ga atta kara, ma, nanka, hakkiri ienai n desu kedo, rest-TOP various worry-NOM have-PAST because, well clearly say-NEG N COPNONPAST-AH

but

zentaitekini zikan-ga kakatta ki-ga suru. overall time-NOM take-PAST feeling-nom do-NONPAST ‘At first, I hadn’t been used to using the body that much, and I had various things to worry about. I cannot clearly say..., but overall, I feel that it took a long time (for me to get used to the club).’

In three instances, information provided in plain form predicates offer subordinate information (Ikuta 1983; Maynard 1991) as in excerpt (23), in which he is explaining his part-time job.

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

Peter more actively used style shifts in post-OPI; 12 of 17 plain forms in matrix predicates were in self-directed soliloquy-like utterances—wondering, recalling, or expressing emotions, such as soo kana ‘I wonder’, a musume-san-ga ita ‘ah, he had a daughter’, and omosirokatta ‘that was fun’. He often concluded his turn with plain forms, which served as summary statements, often recalling his experiences with emotion or internal thoughts, such as in excerpt (22). His uses of the plain forms are congruent with the functions of style shifts discussed in the previous studies (e.g. Maynard 1991; Okamoto 1999; Makino 2002; Cook 2006). Peter, in response to the interviewer’s follow-up explanation concerning his kendo club activities, concluded his turn with a plain-form predicate.

68

STYLE SHIFTS AMONG JAPANESE LEARNERS

(23)

(Peter, Post-OPI)

1

I:

Huun, donna baito site iru n desu ka. Hmm what-kind part-time-job do-GER AUX-NONPAST N COP-NONPAST-AH Q ‘Hmm, what kinds of part-time jobs are you doing?’

2

P:

Mazu-wa kyampasu-no [name of cafeteria]-de hataraite iru n... first campus-GEN

-LOC work-GER AUX that...

sandwich make-such-as do ‘First, I work in the campus cafeteria... like, making sandwiches.’

He also performed a role play in which he talked to a little child using the plain style in a friendly approachable way. Overall, Peter’s shifts both in preand post-OPIs are quite appropriate. Importantly, he became more active in shifting styles upon return, demonstrating his capacity to play a more active role in interactions in Japanese society.

SUMMARY AND CONCLUSION The quantitative analysis showed that even though all five participants spoke in the polite style prior to studying abroad, upon return, Greg and Henry’s styles diverged from the norm. They clearly overused the plain forms, confirming the reported tendency for some L2 sojourners to use the informal style upon return. The other three continued to use the polite style as their base style, though Peter increased the use of plain forms. If the quantitative analysis alone were considered, one might conclude that Greg and Henry (and possibly Peter) did not gain pragmatic competence and regressed in their use of the polite/plain styles while studying abroad. However, the qualitative analyses revealed that all participants gained some understanding of social meanings of the styles and became more active language users by choosing and shifting the styles. Rather than merely observing the assumed native speakers’ norms presumably taught to them, they appear to be actively making choices as to which style to use in different contexts. Alan, Sam, and Peter, who appropriately chose the polite style as their base style in post-OPIs, occasionally switched to the plain style. Peter, whose OPI rating was the highest, most actively shifted the styles. Their switches to the plain form index utterances that are close to their ‘selves’ (their self-directed thoughts, recalls, and emotions) and/or organize information (subordinate information, summary statements). Their moderate and appropriate style shifts also have the effect of not sounding too distant, resulting in a degree of friendliness.

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

sandoicchi tukuttari suru.

N. IWASAKI

69

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

On the other hand, both Greg and Henry chose to use the plain style as the base style in post-OPIs—though this choice may be considered inappropriate. Crucially, both Greg and Henry’s choice is not due to their inability to use the polite style. When the situations called for an expression of deference, they demonstrated their ability to consistently use the polite style, which reflects their understanding of the social/situational meanings of the polite style. Moreover, only some of their shifts to the polite style seem haphazard. It is evident from the qualitative analyses that all participants in this study learned through their study-abroad experiences that Japanese speakers make choices, and they themselves also learned to choose between styles—though there is considerable variability among individuals as to whether and to what extent their choices resemble those of native speakers. The conclusion above is not without limitation. In this study, the ways in which the participants use the plain/polite styles in post-OPIs, as compared with pre-OPIs, demonstrate that they became more active users of the style shifts that reflect their social meanings, but they do not tell us the processes by which they learned to become such active social agents. We can only infer that they learned the social meanings of these forms through the active process of language socialization while in Japan. We need to await future studies that examine the sojourners’ socialization during their stay abroad or their retrospective views about their experiences and language use while abroad in order to uncover the intricate relationship between their language choice, experiences, and perceptions, which is no doubt related to their sense of identities. L2 Japanese learners are generally introduced to the polite forms first and are given abundant practice using them in the classroom. Moreover, both linguists and textbook writers traditionally assert that Japanese speakers observe the socially agreed-upon rules. As a result, they may simplistically associate the polite forms with formal contexts and the plain forms with informal contexts, which are often only equated with interactions with close friends, without truly understanding the social meanings of the styles. During their stay in Japan, however, as they socialize and interact with Japanese speakers, they are likely to be encouraged or pressured to use the plain style by their Japanese peers and/or host families. They then may realize that the plain style is not bound to certain contextual features (e.g. talk with close friends), but that the form has social meanings which one employs moment-to-moment to index intimacy/friendliness and self-directed utterances to create a desired context. If socialization in the target language community is the key for L2 learners to acquire social meanings associated with different styles or address forms, then the question arises as to whether formal instruction in the home country can play any role in L2 learning of such social meanings. Though normative association between the form and context may be useful for the sake of simplicity, some discussion of the social meanings of the forms and their flexibility and variation among Japanese speakers that reflect the meanings may prove helpful for students.

70

STYLE SHIFTS AMONG JAPANESE LEARNERS

ACKNOWLEDGEMENTS I would like to express my sincere gratitude to Janet S. Shibamoto Smith, Gabriele Kasper, Jane Zuengler, and the anonymous reviewers for their helpful and constructive comments, and to Tammy Gales for thorough and thoughtful proofreading/editing.

APPENDIX Abbreviations used in the excerpts topic marker nominative accusative genitive dative instrumental quotative sentence final particle

Others AH AUX COP GER NEG Q N CL

address honorific auxiliary verb copula gerund negative question marker nominalizer classifier

NOTES 1

2

Japanese also has ‘referent honorifics’ through which the speaker expresses deference towards the referent or humbleness (when referring to his/her own action). These are realized by verb choice. This assumption may be questionable when considering the possibility that students primarily participate in

3

interactions with their peers where sounding distant may be negatively viewed. In highly dependent clauses, the use of the polite-form predicates results in hyper-politeness. For example, ittara ‘if I go’ can be expressed as ikimasitara, which sounds excessively polite.

REFERENCES Althen, G. 1988. American Ways: A Guide for Foreigners in the United States. Yarmouth, Maine: Intercultural Press.

Atsuzawa-Windley, S. and S. Noguchi. 1995. ‘Effects of in-country experience on the acquisition of oral communication skills in

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

Particles TOP NOM ACC GEN DAT INSTR QT SFP

N. IWASAKI

Makino, S. 1983. ‘Speaker/listener-orientation and formality marking in Japanese,’ Gengo Kenkyuu 84: 126–45. Makino, S. 2002. ‘When does communication turn mentally inward? A case study of Japanese formal-to-informal switching,’ in N. M. Akatsuka and S. Strauss (eds): Japanese/Korean Linguistics 10. Stanford, CA: CSLI, pp. 121–35. Marriott, H. 1993. ‘Acquiring sociolinguistic competence: Australian secondary students in Japan,’ Journal of Asian Pacific Communication 4/4: 167–92. Marriott, H. 1995. ‘The acquisition of politeness patterns by exchange students in Japan,’ in B. Freed (ed.): Second Language Acquisition in a Study Abroad Context. Amsterdam: Benjamins, pp. 197–224. Marriott, H. and S. Enomoto. 1995. ‘Secondary exchanges with Japan: Exploring students’ exchanges and gains,’ Australian Review of Applied Linguistics 12: 64–82. Maynard, S. K. 1991. ‘Pragmatics of discourse modality: a case of ‘da’ and ‘desu/masu’ forms in Japanese,’ Journal of Pragmatics 15: 551–82. Okamoto, S. 1999. ‘Situated politeness: manipulating honorific and non-honorific expressions in Japanese conversations,’ Pragmatics 9/1: 51–74. Okamoto, S. 2002. ‘Ideology and social meanings: rethinking the relationship between language, politeness, and gender,’ in S. Benor, M. Rose, D. Sharma, J. Sweetland, and Q. Zhang (eds): Gender Practices in Language. Stanford: Center for the Study of Language and Information, pp. 91–113. Regan, V. 1995. ‘The acquisition of sociolinguistic native speech norms: effects of a year abroad on second language learners of French,’ in B. Freed (ed.): Second Language Acquisition in a Study Abroad Context. Amsterdam: John Benjamins, pp. 245–67. Schieffelin, B. B. and E. Ochs. 1986. ‘Language socialization,’ Annual Review of Anthropology 15: 163–91. Siegal, M. 1995. ‘Individual differences and study abroad: women learning Japanese in Japan,’ in B. Freed (ed.): Second Language Acquisition in a Study Abroad Context. Amsterdam: John Benjamins, pp. 225–44. Siegal, M. 1996. ‘The role of learner subjectivity in second language sociolinguistic competency: western woman learning Japanese,’ Applied Linguistics 17/3: 356–82.

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

Japanese,’ Australian Review of Applied Linguistics 12: 83–98. Barron, A. 2003. Acquisition in Interlanguage Pragmatics: Learning How to do Things with Words in a Study Abroad Context. Amsterdam: John Benjamins. Barron, A. 2006. ‘Learning to say ‘you’ in German: the acquisition of sociolinguistic competence in a study abroad context’ in M. A. Dufon and E. Churchill (eds): Language Learners in Study Abroad Contexts. Clevedon: Multilingual Matters. Brown, P. and S. C. Levinson. 1987. Politeness: Some Universals in Language Usage. vol. 4. Cambridge, UK: Cambridge University Press. Cook, H. M. 1996. ‘The Japanese verbal suffixes as indicators of distance and proximity’ in R. Dirven, R. W. Langacker, and J. R. Taylor (eds): The Construal of Space in Language and Thought. Berlin: Mouton de Gruyter, pp. 3–27. Cook, H. M. 2001. ‘Why can’t learners of JFL distinguish polite from impolite speech styles?’ in K. R. Rose and G. Kasper (eds): Pragmatics in Language Teaching. Cambridge: Cambridge University Press, pp. 80–102. Cook, H. M. 2006. ‘Japanese politeness as an interactional achievement: Academic consultation sessions in Japanese universities,’ Multilingua 25: 269–91. Cook, H. M. 2007. ‘The Role of Desho(o) in JFL Learners’ Language Socialization’. Paper presented at the CLIC Symposium entitled, ‘Language Socialization, Interaction and Culture,’ UCLA, February 23–24. Dufon, M. A. 2000. ‘The acquisition of linguistic politeness in Indonesian by sojourners in naturalistic interactions. PhD thesis. University of Hawai’i Manoa, 1999,’ Dissertation Abstracts International-A 60/11: 3985. Hashimoto, H. 1993. ‘Language acquisition of an exchange student within the homestay environment,’ Journal of Asian Pacific Communication 4: 209–24. Ikuta, S. 1983. ‘Speech level shift and conversational strategy in Japanese discourses,’ Language Science 5: 37–53. Kasper, G. and K. R. Rose. 2002. Pragmatic Development in a Second Language. Oxford, UK: Blackwell Publishing. Kinginger, C. and K. Farrell. 2004. ‘Assessing development of meta-pragmatic awareness in study abroad,’ Frontiers: The Interdisciplinary Journal of Study Abroad 10: 19–42.

71

Applied Linguistics 31/1: 72–93 ß Oxford University Press 2009 doi:10.1093/applin/amp011 Advance Access published on 9 April 2009

The Relationship between Applied Linguistic Research and Language Policy for Bilingual Education1 DAVID CASSELS JOHNSON Washington State University

TOWARDS MULTI-LAYERED RESEARCH IN LANGUAGE POLICY The field of language planning and policy (LPP) has progressed through a number of phases [see review in Ricento (2000)], engendering multiple frameworks which attempt to describe the process of national language planning/ policy development (Fishman 1979; Haugen 1983). While enumerating some of the possible (and perhaps preferable) steps in and goals of language planning, these earlier frameworks have been criticized for focusing on top-down policy making and ignoring the sociopolitical contexts in which language planning occurs [see review in Ricento (2000)]. Reacting to these so-called positivistic approaches, critical research emphasizes how the state can use language policy to perpetuate systems of social inequality (Tollefson 1991, 2002, 2006; Ricento 1998; Wiley 2002; Pennycook 2006). Tollefson (2006) articulates the aims of critical language policy (CLP): (i) it is critical of traditional apolitical LPP approaches and instead ‘acknowledges that policies often create and sustain various forms of social inequality, and that policy-makers usually promote the interests of dominant

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

Currently, restrictive-language policies seem to threaten bilingual education throughout the USA. Anti-bilingual education initiatives have passed easily in California, Arizona, and Massachusetts, while one was closely defeated in Colorado, and federal education policy has re-invigorated the focus on English education for English language learners, while concomitantly obfuscating the possibility of native language maintenance and developmental bilingual education. This is the educational landscape within which bilingual education researchers, educators, and students must face the formidable challenge of preserving educational choice and bilingual education. Thus, substantive research is needed on how bilingual educators navigate this challenging ideological and policy landscape. Based on an ethnographic study of bilingual education language policy, this article takes up this challenge by focusing on how beliefs about Applied Linguistics research influence the interpretation and appropriation of federal language policy in one US school district. The results have implications for the relationship between the Applied Linguistic research community and language policy processes.

D. C. JOHNSON 73

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

social groups’ (Tollefson 2006: 42); (ii) it seeks to develop more democratic policies which reduce inequality and promote the maintenance of minority languages; and (iii) it is influenced by critical theory. CLP scholarship has helped illuminate how language policies can be ideological, and presents a rich picture of language policy development as one aspect amongst many socio-political processes which may perpetuate social inequality, but it has also been criticized for underestimating the power of human agency (Ricento and Hornberger 1996) and not capturing the processes of language policy development (Davis 1999). As others have noted (Pennycook 2002; Hornberger and Johnson 2007), an (over)emphasis on the hegemonic power of language policies delimits the agentive role of local educators during implementation. Ricento and Hornberger (1996) introduce the metaphor of an onion to evoke the multiple layers through which language policy develops and emphasize the language policy power of the teacher exercised through pedagogical decisions. They argue that LPP research has unsuccessfully accounted for activity in all layers. Similarly, Ricento (2000) points out that language policy research has tended to fall short of fully accounting for precisely how micro-level interaction relates to the macro-levels of social organization. Still, more recent work has examined implementation. Proposition 227 was a voter-approved measure that restricted the development and maintenance of bilingual education programs in California. After Proposition 227 was enacted, bilingual programs were reduced (Ga´ndara 2000), but Garcı´a and CurryRodriguez (2000) note that teachers responded to the law in different ways—from defiance to acceptance. Reflecting this complexity, ethnographic portraits of California education, post-Proposition 227, reveal different findings. Baltodano (2004) finds that the formerly pro-bilingual education parents internalized the English-only ideology in Proposition 227, thus succumbing to its hegemonic influence. Valdez (2001) and Stritikus (2002), on the other hand, discuss the agentive role that teachers played in implementation, sometimes shaping the English-only focus of Proposition 227 to meet the needs of their linguistically diverse students. Some anthropological (Levinson and Sutton 2001; Gale and Densmore 2003) and sociological work (Ball 2006) on educational policy foregrounds the agentive role of local educators in policy processes. For example, Levinson and Sutton (2001) propose a sociocultural approach to educational policy which recognizes the power in authorized policy and involves researching official policy formation, but also emphasizes the need to consider policy appropriation2 when the ‘temporarily reified text is circulated across the various institutional contexts, where it may be applied, interpreted, and/or contested by a multiplicity of actors’ (Levinson and Sutton 2001: 2). They hope that traditional and clear-cut divisions between policy formation and implementation will be replaced with conceptualizations of policy making as a process which stretches across time and contexts (Ball 2006).

74

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

METHOD The ethnography of language policy This study is an ethnography of language policy which ‘include[s] textual and historical analyses of policy texts but must be based in an ethnographic understanding of some local context’ (Hornberger and Johnson 2007). The results reported herein are part of a larger 3-year (2002–5) multi-sited ethnographic study of bilingual education policy and program development in the SDP (Johnson 2007). Ethnographic data collection emerged out of a series of action-oriented research projects on language policy and bilingual program development with bilingual education teachers, administrators, and outside researchers. These projects engendered participant observation and field note collection in many different contexts within the SDP, including the following, which are of interest for this article: (i) a series of meetings attended by teachers and administrators during the development of the SDP language policy, entitled ‘Policy for English Language Learners’ (2005); (ii) the central administrative office in charge of language policy which operated under three different names during data collection but is currently called the Office of Language, Culture, and the Arts (OLCA); and (iii) bilingual teacher

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

The fate of bilingual education relies on language policy at multiple levels from national policies [like No Child Left Behind (NCLB)] to classroom policies, and everything in between. Yet, most LPP research has traditionally adopted a ‘top-down’ approach in that it examines either policy language and/or the historical and sociopolitical development of policies without looking at how such language is interpreted and appropriated. While ethnographic studies of educational policy (Walford 2003) and educational language policy (Stritikus 2002; Skilton-Sylvester 2003; Canagarajah 2006) are emerging, Levinson and Sutton (2001) note that there is still a paucity of ethnographically informed data on educational policy and, similarly, Blommaert (1996) and others (Davis 1999; Hornberger and Johnson 2007) argue that the field of language policy needs more ethnographic studies which illuminate local language policy processes.3 As well, little work has been done which includes both critical analyses of policy discourse and ethnographically captures the multiple layers of language policy creation, interpretation, and appropriation, or, in Ricento and Hornberger’s (1996) terms, illuminates the various layers of the LPP onion. Thus, this article focuses on the appropriation of one policy—Title III of NCLB—in the School District of Philadelphia (SDP) and, more specifically, I focus on a theme which emerged during ethnographic field work: the role of Applied Linguistics research and researchers in the creation, interpretation, and appropriation of language policy. The following research question guides this article: how does Applied Linguistics research shape the interpretation and appropriation of Title III of NCLB?

D. C. JOHNSON 75

Language policy discourse analysis Along with ethnography, I use intertextual discourse analysis to analyze the connections between the various layers of policy discourse. Bakhtin (1986) proposes that both the texts we write and the speech we create (discourse) are filled with the echoes of previous speakers and writers. These echoes, or intertextual connections, imbue texts and discourse with dialogic overtones and multiple meanings and any interpretation of a (potentially multi-voiced) text or discourse requires an understanding of these intertextual connections. Lemke notes that an intertextual approach requires any given text to be analyzed ‘in the context of and against the background of other texts and discourses’ (Lemke 1995: 10). For this study, the object of analysis is policy discourse which includes spoken interaction (e.g. policy meetings, congressional debate, and interviews) and writing (e.g. language policy language). Contained within this policy discourse are policy texts which are a part and product of the discourse and typically take the form of a language policy in the traditional sense, i.e. the language policy text. For example, Title III is the product of congressional debate and, therefore, a piece of Title III text is both a policy text and a product/part of policy discourse. What is referred to as a text, then, is simply the reproduction of some policy discourse and text for the purpose of analysis in this article. Following Gee’s (1990) distinction of little d and big D discourse, (policy) Discourse refers to larger discourses [after Foucault (1978)] which influence behavior and interaction and tend to normalize particular ways of speaking and being. The purpose of intertextual discourse analysis of language policy is to analyze policy text and discourse ‘in the context of and against the background of other texts

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

professional development meetings. For the sake of data triangulation, I conducted multiple formal and informal interviews with key teachers and administrators, including recorded interviews with four administrators and four bilingual teachers and I recorded naturally occurring conversation at language policy meetings. I also analyzed (and contributed to) multiple drafts of the formal SDP language policy as well as informal policy texts (SDP language education literature and web site texts). In order to contextualize what was happening in the SDP, the ethnographic data were then compared with: (i) ‘top-down’ policy texts, including Title VII of the Elementary and Secondary Education Act (ESEA), commonly known as the Bilingual Education Act (BEA), and the multiple drafts of the policy which took its place, Title III of NCLB; (ii) the political discourse which accompanied the creation, interpretation, and appropriation of Title III, including Senate and House debate over NCLB (found in the legislative record), political literature released by politicians, and web site information from the USA and Pennsylvania Departments of Education; and (iii) interviews with one Federal Department of Education administrator and one Pennsylvania Department of Education Administrator.

76

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

FEDERAL LANGUAGE POLICY This section deals with the role of research in the development of Title III and in the resulting policy language. An analysis is offered of early drafts of Title III and the discourse (legislative debate) which led to its enactment. I look at how ‘bilingual education’ is defined and the requirement that all language education programs should have scientifically based research support.

Defining bilingual education Beginning in 1968 when it was first enacted, Title VII of the ESEA, otherwise known as the BEA, was the preeminent federal educational language policy which governed how federal money was administered to language education programs in the USA. At the time of its inception, the goals of bilingual education were vague and definitions and typologies were scarce (Crawford 1998; Ricento 1998; Wiley and Wright 2004); those on the political left and right in the US supported what seemed to be an effective way for transitioning ELLs into English language classrooms. This ambiguity in goals affected each installment of the act when it was revised four more times, but in 1994 there was a definitive shift in the policy language toward valuing bilingualism as a resource (Wiese and Garcia 2001) and promoting both transitional and developmental4 bilingual education (in which native languages are developed). Yet, in 2002, Title VII was effectively replaced by Title III of the newly named ‘No Child Left Behind Act’ (NCLB). With its vigorous attention to English language acquisition for ELLs, Title III has fomented concern that developmental bilingual education will be phased out and transitional or English-only pedagogical approaches phased in (Wiley and Wright 2004). While critics of Title III decry the dominant focus on English language education and concomitant obfuscation of bilingual education, early drafts of the policy were even more restrictive. House Resolution 1 (HR 1) was introduced

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

and discourses’ to illuminate policy Discourse and establish links between the various layers of language policy. Specifically for this article, I focus on circulating ideas about ‘research’ as they influence bilingual education policy. Each text, collected during ethnographic fieldwork, is analyzed in light of features within the layers of context which may be intertextually linked. Explanation of the context in which each text was collected is supported by field notes and participant observation. Intertextual connections are then analyzed between micro-level language policy discourse and ‘top-down’ or macro-level language policy Discourse. The goal here is to trace the intertextual links between federal and local language policy discourse by focusing on a particular theme: the role of Applied Linguistics research and/or researchers in language policy creation, interpretation, and appropriation.

D. C. JOHNSON 77

in the House of Representatives by John Boehner (Republican-Ohio) and began with ‘findings’ which make a conspicuous shift away from the language in the BEA: (a) FINDINGS – The congress finds as follows: (1) English is the common language of the United States and every citizen and other person residing in the United States should have a command of the English language in order to develop their full potential [HR 1, Title III, Sec. 3102]

[Purpose:] (3) to assist local educational agencies to . . . prepare limited English proficient students, including recent immigrant students, to enter all-English instructional settings within 3 years [emphasis added, HR1, Title III, Sec. 3102]. HR 1 might have restricted bilingual education to three-year transitional programs and concomitantly outlawed developmental bilingual education. However, when HR 1 was sent to the Senate for debate, Senators such as Jack Reed (Democrat-Rhode Island) argued that 3 years was an arbitrary time limit and students may need more time before they transition into all English instructional settings (22 June 2001, Congressional Record). Therefore, in the Senate, the ‘findings’ and 3-year time limit were struck from the policy and, when it was sent back to the House, legislators like Silvestre Reyes (Democrat-Texas), who championed late-exit bilingual education throughout the legislative process, celebrated the new version of the bill as a victory for bilingual education. On the House floor, Reyes exclaimed: I am proud to support the conference report on HR 1 . . . bilingual programs are important to limited English proficiency children because they build on native language proficiency to make the transition to all-English academic instruction . . . Opponents of bilingual education favored placing a three year time limit on how long students can be enrolled in bilingual education regardless of what level of English proficiency they reach . . . The compromise bill gives students the flexibility to remain enrolled in bilingual education as long as is appropriate. [Congressional Record, 20 December 2001, emphasis added] Reyes identifies ‘bilingual education’ as a pedagogical method which builds on native language proficiency to transition students to all-English academic instruction. He does not believe that time restrictions should be placed on this transition but he considers bilingual education to be, by definition, transitional, and not a method for native language maintenance and, thus, his definition does not include developmental programs.

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

HR 1 places English at the fore, both in education and in society, because it is necessary for developing one’s full potential. This nationalistically monolingual sentiment is accompanied by a three-year time limit on ELL participation in bilingual education:

78

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

While bilingual education supporters, like Reyes (and Ted Kennedy, Democrat-Massachusetts), saw Title III as a victory for bilingual education, opponents of bilingual education also felt victorious. In a press release, Boehner, who had introduced HR 1, proclaimed: As a result of the No Child Left Behind Act (H.R. 1), the bipartisan education reform legislation signed in January by President Bush, bilingual education programs across the country are being transformed to give new tools to parents and to focus on helping limited English proficient (LEP) children learn English. [News from the committee on Education and the Workforce, 17 October 2002]

‘Scientifically based research’ Along with the focus on English, Title III includes the provision that any education program must be based on ‘scientifically based research’, repeating the phrase 119 times. For example, in the beginning of Title III, one of the ‘Purposes’ sets the tone for the rest of Title III: The purposes of [Part A of Title III] are to . . . provide State agencies and local agencies with the flexibility to implement language instructional educational programs, based on scientifically-based research on teaching limited English proficient children, that the agencies believe to be the most effective for teaching English [Title III, Part A, Sec. 3102, 9]. This excerpt suggests that local educators must choose language programs which are supported by scientifically based research; yet it also poses an interpretive dilemma by simultaneously insisting on local flexibility (another salient theme coursing through NCLB) while placing at least one, and perhaps two, restrictions on local choice: (i) any program must be based on

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

Boehner’s press release suggests that Title III will transform bilingual education programs such that they focus on English and, presumably, not mother tongue education. This, according to HR 1 (which Boehner introduced to Congress), is necessary for ELLs to develop their full potential, and the quicker they can transition to English, the better. Both political allies (like Reyes and Kennedy) and political opponents (like Boehner) celebrated Title III as a victory; as well, even political allies of bilingual education perpetuated the Discourse that bilingual education was necessarily transitional, a notion that is reflected in the final policy text. This policy Discourse ignores applied linguistic research which (i) compares multiple bilingual program typologies (Hornberger 1991; Garcı´a and Baker 2007) and (ii) demonstrates the effectiveness of developmental bilingual education relative to transitional and ESL-focused programs (Thomas and Collier 1997, 2002; Lindholm-Leary 2001; Krashen and McField 2005; Rolstad et al. 2005; August and Shanahan 2006; Genesee et al. 2006).

D. C. JOHNSON 79

THE SDP The SDP is an example of a local education agency responsible for interpreting and appropriating Title III. The fifth largest school district in the USA, the SDP has around 200,000 students, 14,000 of whom receive ESOL/bilingual program support services (representing more than a third of the ELL population for the entire state of Pennsylvania). While almost 50% of the ELLs are Spanish speaking, the SDP also serves large populations of Khmer, Vietnamese, Chinese, and Russian speakers. The SDP was a forerunner in bilingual education policy and has a history of supporting bilingual education programs—shortly after Title VII competitive grant funding became available under the BEA in 1968, pilot programs were implemented. Since then, bilingual programs have included Russian, Chinese, Khmer, and Spanish programs. Today, one central administrative office

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

scientifically based research; and (ii) the agencies must believe in the effectiveness of their chosen program. What is believe referring to—beliefs about the most effective programs or beliefs about scientifically based research? Both interpretations are plausible based on the policy text alone. In order to ascertain how the US Department of Education interprets these policies, I interviewed the director of the State Consolidated Grant Division5 (Brinda Sea6) in the Office of English Language Acquisition to discover what, if any, educational programs were supported or promoted by the Department of Education. She immediately and unequivocally responded that they do not promote or prefer any particular method and are, in fact, prohibited from doing so: ‘We stay completely out of it’ (telephone conversation, 24 May 2006). She stressed that it was up to the states and schools to choose particular pedagogical programs for ELLs as long as the chosen programs are ‘research-based’ and added that as long as they provide support for their programs in the form of research, they are eligible to receive Title III funds. While NCLB does outline the parameters of scientifically based research (see Title IX, A-B7), the US Department of Education provides little guidance for local school districts8 regarding which programs the research supports or what even counts as scientifically based research. Brinda Sea neither advocates particular bodies of research nor language programs which such research might support, and asserts that the US Department of Education cannot prescribe particular pedagogical approaches.9 Not only does federal policy discourse offer no clear answers, but also federal policy texts, like Title III, can be confusing since they necessarily contain multiple voices and are an intertextual mixture of new and old policy language. Of the multiple authors of Title III, whose intentions carry the most power? Further, how are such policies—whose authors have conflicting intentions— interpreted and appropriated in school districts around the USA? How do educators make sense of these multi-voiced texts?

80

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

Reflexive appropriation of applied linguistics research In the fall of 2002, OLCA recruited Eve Island, a Title VII consultant to the SDP and well-known and respected applied linguistic researcher, to begin developing an official language policy for the entire school district. Beginning in the spring of 2003, Eve organized and facilitated a series of policy meetings which culminated in a 2-day language policy retreat at which around 35 school district employees, teachers, and principals gathered to flesh out what they would like to see in a language policy. Eve was a human conduit and intertextual bridge between the Applied Linguistics research community and the SDP practitioners. Eve would portray the bilingual teachers’ struggle to maintain their bilingual programs as part of a larger bilingual education movement going on throughout the USA. For example, before the language policy retreat, Eve sent an inquiry to the TESOL Advocacy Listserv, requesting advice for writing a school district language policy; while she did not receive specific suggestions, she did receive words of encouragement which she then announced at the beginning of the language policy retreat. She reported that Donna Christian said, ‘I’ve never seen anything like this!’ and Richard Ruiz said that he ‘loved this!’ (field notes, 13 March 2003). In this way, well-known applied linguists were positioned as supporters of bilingual program development and the language policy initiative within the SDP. Eve not only brought words of encouragement to the language policy retreat, but she also relied heavily on David Corson’s book Language Policy in

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

oversees all ESOL/bilingual programs. OLCA is the primary interpretative conduit for federal language policy and the administrators therein are the primary arbiters for interpretation and appropriation of educational language policies— including Titles VII and III. The data reported in this study arose from my close work with OLCA on various projects including the SDP language policy which began with what was called a ‘language policy retreat’ (March, 2003) and ended with its ratification at a school board meeting (June, 2004). I acted as a participant– observer during the production of the policy and I also helped edit the policy document at various stages in its development. However, while this gave me an insider’s perspective of how a local policy might be created, the minor additions I made to the document were largely ignored in the end. OLCA, in a sense, facilitates the intertextual links between federal and local language policy, yet the nature of this facilitation varies depending on who is doing the interpreting and, here, I focus on the various ways that applied linguistic research and researchers influenced the interpretation and appropriation of Title III. Specifically, I look at how two OLCA administrators and one outside applied linguistic researcher (and SDP consultant) influenced language policy in the SDP during the development of the SDP language policy.

D. C. JOHNSON 81

Eve refers to creating spaces or the emancipation potential within a language policy for challenging dominant ideologies and questioning how linguistic capital is allotted and this is really important in Corson’s book. She hopes his theories about the power of a school language policy to be emancipatory manifests in the SDP [field notes, 14 March 2003]. Eve herself anthropomorphized the philosophy espoused in Corson’s book by facilitating negotiation between teachers and administrators and empowering teachers in the language policy development process. Eve challenged traditional divisions of language policy ‘creators’ and ‘implementers’. Eve cited a community of applied linguistic researchers to promote and maintain local Discourses about the benefits of developmental bilingual education and she portrayed larger societal Discourses outside the SDP—as espoused by Donna Christian and Richard Ruiz—as supportive of local policy discourse. In this way, Eve challenged monolingual Discourses, so prevalent within and without US educational language policy, and instead portrayed applied linguistic research and researchers which support developmental bilingual education as a dominant Discourse.

Applied linguistics research as support During the development of the SDP language policy, I interviewed another champion of developmental bilingual programs—the Chief Officer of OLCA, Emily Dixon-Marquez. Emily helped author both the Title III application and Title VII grants and was also an integral member of the community that developed and drafted the SDP language policy. Her influence, however, transcends the time limits of this study as she first started working in the district in 1973, shortly after Title VII monies were made available under the BEA. In a recorded interview, she reflects on the history of US bilingual education language policy, beginning with the BEA: David: How do you interpret the parameters that were set in the Bilingual Education Act?

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

Schools: A Resource for Teachers and Administrators (1999) (which proposes theoretical models of school district language policy development) as a foundation for the development of the SDP policy. She used the book as a stepby-step manual for constructing the policy and she perpetrated the modus operandi of Corson’s policy theories. For example, Corson adamantly argues for school language policy creation which draws upon the knowledge and resources of many different actors, not just a few key administrators, and in doing so can resist ‘unfair aspects of reproduction and . . . help soften social injustices’ (Corson 1999: 25). At the language policy retreat, Eve and I discussed the ramifications of Corson’s perspective on the SDP language policy:

82

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

Emily: Well, I think that whoever set them probably wasn’t basing it on research because there was none – so in terms of federal grants 3 and 5 years were always convenient numbers. David: What are the 3 and 5 years? Emily: Well, transitional bilingual programs were initially 3 years and then out in the (mainstream) – and that’s how the legislation read. David: So in that title VII grant has continued until NCLB?

Emily characterizes the growth of language policy in the USA as developing and incorporating a better sense of the research because, in part, there was no body of bilingual education research when Title VII was developed. According to Emily, US education policy has grown alongside and been buttressed by applied linguistic research—as the research trends have grown away from remediation or deficit approaches, and increasingly supported enrichment models of language education, so has policy, leading right up to NCLB. I then asked Emily about NCLB’s requirement that all educational programs be supported by scientifically based research and how this influenced their Title III application. Emily said that they referred to the results reported in Thomas and Collier (1997, 2002) suggesting that two-way (or dual language) and one-way developmental bilingual education are the two most effective program models for facilitating content knowledge, acquisition of English, and maintenance of students’ mother tongues. Based on these studies, OLCA chose to develop ‘dual language programs’. I asked her why these programs were acceptable under Title III: Emily: [W]e preceded NCLB in terms of models and so forth – we’ve been doing dual language, that’s not new to us, so it hasn’t been enlightening in that sense. We’re glad that it’s an acceptable model that they subscribe to. David: Why is [dual language] an acceptable model? Emily: I think probably because of the Thomas and Collier study – which the government is comfortable with and they think has some integrity to it – and it’s research based . . . plus I think this country is finally coming to the realization – and I’m probably being really optimistic, and maybe naı¨ve here – but coming to the realization that we need to learn more about languages and cultures [recorded interview, 11 April 2003].

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

Emily: But it’s really evolved – it’s become an enrichment model whereas before it was just remediation or a deficit type approach . . . it’s much much much better . . . I think it started it was like three million – and it’s now into the hundreds of millions [recorded interview, 11 April 2003].

D. C. JOHNSON 83

Referencing research in local language policy Within the discourse community responsible for the SDP language policy, Emily and Eve helped perpetuate the belief that developmental bilingual education was superior to transitional programs. Inspired by this belief and buttressed by applied linguistic research (Thomas and Collier 1997, 2002) and researchers (e.g. Donna Christian, Richard Ruiz, and David Corson), it appeared as though a policy which outright resisted and/or ignored Title III would develop. However, initial policy language filled with idealism eventually gave way to pragmatism, and, while the initial community of policy developers included a more egalitarian mix of teachers and administrators from multiple levels of institutional authority, OLCA eventually took control of the drafting process. The developers of the policy, including Eve, and especially Emily, were cognizant of the language in Title III and drafted a policy which appropriates, ignores, and/or alters Title III text. The first shift is a conspicuous snub of the ‘scientifically-based’ language in Title III in favor of: High quality programs for ELL’s provide optimum conditions for English Language acquisition, and therefore, the District commits to . . . programs for ELLs using sound instructional practices

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

Emily asserts that NCLB is simply catching up with what they already knew in the SDP—developmental bilingual education is an effective and researchbased language education model. Further, Emily senses a shift in the language ideological landscape in the USA—we are coming to the realization that we need to promote learning about various languages and cultures. For Emily, the advancement of knowledge about languages and bilingual education, supported by applied linguistic research, has engendered bilingual educational policies which are increasingly sensitive to these shifts. She interprets Title III as being at the forefront of this national ideological shift and thus it embodies an enrichment model of bilingual education. Yet, as she argues, the SDP preceded this national trend toward enrichment models of bilingual education and thus NCLB has not been enlightening in that sense. In our interview, Emily was generally very optimistic about Title III and its effects on bilingual education in the SDP. Along with the policy text support for dual language education, she expressed gratitude that under Title III, unlike Title VII, money was no longer distributed according to a competitive grant and the per pupil entitlement was more per pupil. For Emily, this was ‘wonderful news’ (recorded interview, 11 April 2003). Emily’s interpretation of Title III stands in contrast to its creators, like Reyes, who conceptualize ‘bilingual education’ as necessarily transitional. This is not because Emily was unaware of the English-focused flavor of NCLB (as the author of the Title III application, she was well aware of its requirements), but she still interpreted it as supporting the dual language programs she intended to implement with Title III money.

84

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

as shown by research in the field of second language acquisition [III, 1a]. Note the immediate nod to English language acquisition and the district’s assertion that they are committed to programs which ensure this, thus appropriating Title III’s English-focused discourse. However, using second language acquisition research to engender sound instructional practices is conspicuously different than Title III’s assertion that programs need to be based on ‘scientifically based research’. Then, unlike Title III, the policy defines bilingual education:

This text defines ‘bilingual education’ as a method which develops and maintains first language literacy which is not a goal of transitional programs and stands in contrast to Title III’s assertion that the goal of bilingual education is for ‘limited English speakers to enter all English-instructional settings’. Thus, definitions of bilingual education as necessarily transitional in federal policy text and discourse are resisted (or ignored) in favor of language which allows for developmental programs. Then, unlike Title III, the SDP policy makes explicit claims about what language programs the ‘research’ supports: The district recognizes that . . . research indicates that strong first language development serves as an effective foundation for second language acquisition [III, 3b]. This allusion to the interdependence hypothesis (Cummins 1981) asserts that English acquisition is aided by first language development, and presumably, pedagogy in which first languages are strongly developed. The SDP language policy text makes a few important discursive shifts away from, while concomitantly maintaining, some of Title III’s English-only focus. While SDP language policy text appropriates some of the salient Englishfocused language from Title III, it also creates implementational space for developing various types of bilingual programs because it makes specific claims about what the research supports—strong first language development—and defines bilingual education as a program which develops first languages.

Applied linguistic research as doctrine The remainder of this article focuses on shifts in bilingual education policy that took place between the Fall of 2003 and Spring of 2005, illustrated by changing definitions of ‘good’ bilingual education and shifting interpretations of Applied Linguistics research. What is missing in this study, and what might be expected at this point, is a description of the implementation of

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

A bilingual education program develops and maintains first language literacy as well as literacy in the second language [Section III, Part 2b].

D. C. JOHNSON 85

When you talk about bilingual education, in many research based models, it is uh – considered a way to help children, or assist students who are second language learners, to acquire English – The problem we have seen historically with bilingual education is that – somehow even students who are born . . . in the United States have been placed in bilingual education programs just . . . because, even though they’re native speakers of English, they come in with some gaps with their linguistic development due to maybe speaking another language at home or even hearing another language at home [recorded interview, 13 June 2005]. Like Silvestre Reyes, what Lucı´a refers to as ‘bilingual education’ is a programmatic model, the primary goal of which is acquisition of English for ELLs. Bilingual education in her usage is specifically intended for non-native English speakers. Speaking or hearing a non-English language at home may induce gaps in some English-speaking students’ English language development, but those students, as English speakers, are still not good candidates for bilingual education in her view. Lucı´a’s beliefs about research influence her conceptualization of what ‘bilingual education’ is and who it is for; it does not include developmental programs and is not for English speakers born in the USA.

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

the SDP language policy. In large part, there was no implementation and thus it existed, and continues to exist, as an interesting remnant on the OLCA web site, but has no real power. Its lack of power is, in part, due to a shift in pedagogical philosophy and bilingual education policy at OLCA. In the fall of 2003, OLCA acquired a new director of ESOL/bilingual programs, Lucı´a Sanchez, whose beliefs about applied linguistic research were different than Emily’s and Eve’s and influenced language policy accordingly. Emily Dixon-Marquez and Lucı´a Sanchez are an interesting comparative juxtaposition of how language (education) ideologies influence interpretation and appropriation of top-down language policies like Title III. Lucı´a started working at OLCA as Emily, who quit OLCA in 2004, was leaving, and as the current director of ESOL/bilingual programs, Lucı´a has been the most influential interpreter and appropriator of federal and state language policy since 2004. Before coming to Philadelphia, Lucı´a worked at the Pennsylvania Department of Education where she reviewed and approved the Title III plan submitted by Emily. Thus, when Lucı´a started working at OLCA, she adopted the responsibility of overseeing implementation of Title III monies which she herself had approved! Yet, Lucia’s beliefs about bilingual education differ from Emily’s and stand in contrast to the additive bilingual beliefs circulating through OLCA and the development of the SDP language policy. I had a chance to interview Lucı´a and ask her about programmatic changes that were underway:

86

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

Lucı´a has promoted transitional bilingual programs which exit ELLs to English-medium classrooms by middle school. Her interpretation of the research accommodates (or vice versa) her interpretation of the intentions of Title III: Title III was created to improve English language acquisition programs by increasing the services or creating situations where the students would be getting supplemental services to move them into English language acquisition situations [recorded interview, 13 June 2005].

Teacher: Who or where did the decision make – come from to [transition students]? Lucia: Because, because, number one, we looked at all the programs that are effective based on Krashen’s research – and the beginning of Title III of the No Child Left Behind Act, which is long, and there’s nothing we can do to change that [tape recorded, 1 December 2005]. Lucı´a’s beliefs about the research, and here she alludes to Stephen Krashen’s research, are used to justify her interpretation of Title III as restrictively focused on English language development. By conferring language policy decisions to Stephen Krashen and Title III, Lucı´a deflects responsibility for her decision to shift SDP language policy to outside experts and outside language policy, both of which rigidly dictate a transitional bilingual language policy and ‘there’s nothing they can do to change that’. Later in the meeting, she emphasizes this point by saying: Everyone knows about Stephen Krashen – he’s a linguist that has devoted most of his research to education, but he’s a linguist. He’s a scientist that studies different linguistic patterns but he really – we heard about the silent period through Krashen, we heard about comprehensible input, that’s Krashen. We heard about the lowering the affective filter, that’s Krashen, error correction, that’s

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

Lucia interprets the goals of ‘bilingual education’ and Title III as the same— eventual transition of ELLs into mainstream classrooms. She does not feel that the ELL students in Philadelphia are well served by maintenance programs and for these students, the preferred model is transitional (with all bilingual education students entering mainstream classrooms by middle school) with the option of heritage language classes to continue biliteracy development. Throughout ethnographic data collection, bilingual education meetings were held which involved policy and program development and teacher training. During one such meeting, led by Lucı´a, a couple of the teachers expressed concern that the programmatic changes being implemented were a clandestine attempt to get rid of bilingual education and mainstream ELLs, a claim which Lucı´a adamantly rejected. After Lucı´a introduced the transitional program, one of the teachers challenged her decision:

D. C. JOHNSON 87

Krashen, so all of that is good research that we all as language teachers need to know. And he said, he is the expert, and he said that, yes, you can introduce English right away – Yes it is important that we know what the research says [tape recorded, 1 December 2005].

DISCUSSION At this point, I would like to return to the research question proposed earlier: how does applied linguistic research shape the interpretation and appropriation of Title III in the SDP? 1 Ignoring the research: despite the preponderance of research supporting the relative ‘effectiveness’ of bilingual programs, administrators in the US Department of Education and federal lawmakers seem largely oblivious, or at least do not subscribe, to particular studies or bodies of research. Federal policy discourse perpetuates the limiting definition of ‘bilingual education’ as a means for transitioning ELLs into English-medium classrooms. Obscured in this federal discourse about bilingual education is the notion that it can be a means for fostering bilingualism and biliteracy for both native and non-native English speakers alike; that it is a means for utilizing languages as (cultural, educational, or economic) resources; or that ELLs have a right to literacy in their mother tongues. Thus, federal policy discourse and the resultant policy text (Title III) help perpetuate the popular Discourse that bilingual education is, by definition, transitional. 2 Vague demands for research support: still, in the US Department of Education, the views about what the research supports are not clear and/or not enforced when administering Title III monies, and the administrator interviewed for this study insisted that they would ‘stay out of it’ as long as the Title III grant applicants provide (any) research support for their chosen programs. While the definition of ‘scientifically based

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

In these two texts, Lucı´a clarifies her beliefs about what role ‘research’ should play in language policy decision making. While Eve refers outside research and researchers in an attempt to enhance teacher agency, and Emily interprets the research as supportive of local choice and developmental bilingual education, Lucı´a positions the research, here embodied by Krashen, as setting rigid standards to which SDP language policy, and the teachers, must adapt. She enhances Krashen’s standing as the expert by emphasizing his position as a linguist and a scientist; in other words, Lucı´a is relaying Krashen’s expertise to the teachers who should not blame the messenger for the message. The teachers are stripped of their expertise and agency in making language policy decisions—since they are neither linguists nor scientists (nor experts according to Lucı´a)—and Lucı´a deflects the responsibility of SDP language policy decision making to Krashen and Title III.

88

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

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

research’ in Title IX is (somewhat) clear, this administrator did not give a strict definition nor make claims about which programs are best, and she is directly responsible for administering Title III monies, thus leaving implementational space for creative interpretation of what ‘scientificallybased research’ is and which programs it supports. 3 Including researchers in the local discourse community: applied linguistic research is appropriated in the SDP in varying ways. Eve Island, the Title VII consultant and applied linguistic researcher, appropriated research (including her own) and voices from an outside community of bilingual education advocates and applied linguistic scholars (like Richard Ruiz) to contextualize and support educator efforts to develop developmental bilingual policy and pedagogy. Positioning the research(ers) as allies, and the SDP practitioners as experts, empowered teachers to become active agents in policy making. Eve served as a human conduit and intertextual link between a community of Applied Linguistics researchers and SDP educators; she positioned the policy developer’s initiative ‘within the context of and against the background of’ (Lemke 1995: 10) academic discourse which supports both developmental bilingual education and developing school district policy in an egalitarian way that appropriates a diversity of teacher and administrator voices. Concomitantly, she challenged traditional divisions between policy ‘creators’ and ‘implementers’ and exemplified how policies which reduce inequality and promote the maintenance of minority languages can be developed more democratically (Tollefson 2006). 4 Applied linguistic research as support: the Titles VII and III grant writer, Emily Dixon-Marquez, viewed the requirement that programs must be based on scientifically based research as beneficial because the research she knew—especially the Thomas and Collier studies (1997, 2002)—supported the programs she wanted to promote in the SDP. Within the ideological space created at the language policy retreat and maintained by key players, developmental bilingual programs were positioned as superior to transitional programs and the Thomas and Collier studies (1997, 2002) were appropriated as allies. Further, Emily interprets Title III as being supportive, in its language and funding, of dual language programs. 5 (Mis)appropriating Applied Linguistics research as a mandate: while Emily (and Eve) referred outside research and researchers in an attempt to enhance teacher agency, the new head of ESOL/bilingual programs, Lucı´a Sanchez, positioned the research, here embodied by linguists like Stephen Krashen, as setting rigid standards (i.e. transition into English) to which SDP language policy, and teachers, must adapt. Lucı´a’s interpretation of Title III was filtered through her definition of ‘bilingual education’ as necessarily transitional and she cited Krashen (in her words, a linguist and a scientist) as an expert who supports this transition. The expertise of the teachers was diminished in favor of outside experts whose research

D. C. JOHNSON 89

placed restrictions on SDP language policy by demanding transitional programs, which is an inaccurate interpretation of Krashen’s research (Krashen 1996; Krashen and McField 2005). Lucı´a reified popular definitions of bilingual education (as promoted by federal policy discourse) as necessarily transitional and, further, limited the role of teachers in language policy processes to that of ‘implementer’.

CONCLUSION It is difficult enough to pinpoint intentions in a single-authored text but a language policy has multiple authors with varied intentions. As Bakhtin (1986) argues, there are dialogic overtones and multiple meanings within any given text. Title III is no different. Still, while a plurality of readings and interpretations of national language policies like Title III of NCLB are possible (and a reality in this study), policy Discourse constrains interpretation and implementation possibilities—some interpretations will be privileged, especially those sanctioned in federal discourse, while others may be obfuscated or discredited (Ball 2006). The production of Title III was characterized by limiting definitions of bilingual education as necessarily transitional and

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

Cutting edge theoretical developments portray language policy as a potentially hegemonic mechanism through which dominant ideologies about language (education) are promoted (Pennycook 2006; Tollefson 2006). The results of this paper suggest that emphasis should be placed on the word ‘potentially’ since it is shown that while language policies can have this effect, the effects of policies are channeled through human agents in local educational contexts and, thus, policy appropriation can be varied and unpredictable. While Emily and Eve resisted the English-focused discourse in Title III, Lucı´a appropriated it and promoted state-sponsored definitions of ‘bilingual education’ over definitions cited in applied linguistic research. Both Emily and Lucı´a advocate bilingual education but their beliefs about the best type of bilingual education, and what the research says, colored their interpretations of Title III. Using the same Title III policy language and money as support, they implemented ESOL/bilingual programs in very different ways. Emily’s interpretation that Title III is as flexible as it claims, and her beliefs about bilingual education research, created and supported ideological and implementational spaces for additive bilingualism. She and Eve fostered an ideological space for bilingual language policy and program development in which developmental programs were favored and an egalitarian discourse community of administrators and teachers was encouraged to engage with language policy processes. The educational transformation relies on Lucia’s interpretation of Title III as rigidly English-dominant and her own beliefs about language education research. In order for the effects of Title III to be truly monolingual, at least in Philadelphia, administrators must allow themselves to be conscripted by its monolingual discourse.

90

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

TRANSCRIPTION NOTES () (?) ... Italics

transcription doubt unclear utterance ellipsis short pause emphatic stress

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

this influenced the resulting policy text and some interpretations in the SDP. One administrator’s appropriation of Title III suggests that her interpretation was constrained by the English-focused and transitional bilingual education discourse; therefore, Title III put limits on what was educationally feasible. Still, other bilingual educators in the SDP were neither swayed by the English-focused discourse, nor the limiting definitions of bilingual education so prevalent in Title III, and used applied linguistic research to support their interpretation. It is up to local education agents to either expand or restrict their bilingual programs using Title III monies and it has gone both ways in the SDP. During the first half of ethnographic data collection, Title III was interpreted as flexible. In this ideological space of bilingual language policy and program development, Title III was portrayed as a real windfall for SDP bilingual programs and its flexibility was emphasized to teachers and other stakeholders. A shift in personnel, however, led to different interpretations of how Title III monies should be used, marked by beliefs that Title III’s focus on English was a mandate for transitional programs. As well, it is up to applied linguists to become engaged in local language policy processes. This article shows that bilingual language policy implementation is an agentive process which can be influenced by applied linguistic research and researchers. Because they are positioned as experts, at least in Philadelphia, researchers can become members of the discourse community which shapes the appropriation of language policy. In order to resist the pendulum swing away from native language maintenance, researchers will need to team with local educators—and especially administrators—to foster local ideological spaces in which developmental bilingual education is championed. And this is possible. SDP educators make choices and these choices influence the implementation of Title III. Such choices are constrained by the language policies which tend to set boundaries on what is allowed and/or what is considered ‘normal’, but the line of power does not flow linearly from the pen of the policy’s signer to the choices of the teacher. The negotiation at each institutional level creates the opportunity for reinterpretations and policy manipulation. Local educators are not helplessly caught in the ebb and flow of shifting ideologies in language policies—they help develop, maintain, and change that flow.

D. C. JOHNSON 91

NOTES 1

2

3

5

6 7

8

9

This department is responsible ‘for the administration of new formula grants and for providing technical assistance to State and Local educational agencies’, http://www.ed.gov/about/ offices/list/oela/aboutus.html All names are pseudonyms. This passage might also be subject to varying interpretations (Johnson 2007). A school district is an administrative unit which serves the schools in one or more cities. There is one school district which serves Philadelphia but it is divided into 11 regions. Similar assertions were made by representatives from the Pennsylvania Department of Education.

REFERENCES August, D. and T. Shanahan. (eds) 2006. Developing Literacy in Second-language Learners: Report of the National Literacy Panel on Language-minority Children and Youth. Lawrence Erlbaum and Center for Applied Linguistics. Bakhtin, M. 1986. Speech Genres and Other Late Essays. Texas University Press. Ball, S. J. 2006. Education Policy and Social Class: The Selected Works of Stephen J. Ball. Taylor & Francis. Baltodano, M. P. 2004. ‘Latino immigrant parents and the hegemony of Proposition 227,’ Latino Studies 2: 246–53. Blommaert, J. 1996. ‘Language planning as a discourse on language and society: The linguistic ideology of a scholarly tradition,’ Language Problems and Language Planning 20/3, 199–222. Canagarajah, S. 2006. ‘Ethnographic methods in language policy’ in T. Ricento (ed.): An Introduction to Language Policy: Theory and Method. Malden: Blackwell Publishing. Corson, D. 1999. Language Policy in Schools: A Resource for Teachers and Administrators. Lawrence Erlbaum.

Crawford, J. 1998. ‘Language politics in the U.S.A.: the paradox of bilingual education,’ Social Justice 25/3: 50–69. Cummins, J. 1981. ‘The role of primary language development in promoting educational success for language minority students in California State Department of Education’ in Schooling and Language Minority Students: A Theoretical Framework. Evaluation, Dissemination and Assessment Center California State University. Davis, K. A. 1994. Language Planning in Multilingual Contexts: Policies, Communities, and Schools in Luxembourg. John Benjamins Publishing Company. Davis, K. A. 1999. ‘The sociopolitical dynamics of indigenous language maintenance and loss: a framework for language policy and planning’ in T. Huebner and K. A. Davis (eds): Sociopolitical Perspectives on Language Policy and Planning in the USA. John Benjamins Publishing Company. Fairclough, N. 2001. ‘The discourse of new labour: Critical discourse analysis’ in M. Wetherell, S. Taylor, and S. J. Yates

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

4

The study upon which this is based won the 2008 National Association of Bilingual Education dissertation competition. Levinson and Sutton (2001) prefer the term appropriation to implementation which they feel implicitly ratifies a top-down perspective. I agree and adopt this term. While other ethnographic work deals with language policy (Hornberger 1988; Davis 1998; Freeman 1998), the object of analysis in this body of work is not policy itself. Developmental or additive bilingual education can be contrasted with subtractive bilingual education and can include one- and two-way/dual language programs.

92

APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY

Krashen, S. 1996. Under Attack: The Case against Bilingual Education. Language Education Associates. Krashen, S. and G. McField. 2005. ‘What works? Reviewing the latest evidence on bilingual education,’ Language Learner 1/2: 7–10, 34. Lemke, J. L. 1995. Textual Politics: Discourse and Social Dynamics. London: Taylor & Francis. Levinson, B. A. U. and M. Sutton. (eds) 2001. Policy as Practice: Toward a Sociocultural Analysis of Educational Policy. Ablex Publishing. Lindholm-Leary, K. J. 2001. Dual Language Education. Clevedon: Multilingual Matters. Pennycook, A. 2002. ‘Language policy and docile bodies: Hong Kong and governmentality’ in J. W. Tollefson (ed.): Language Policy in Education: Critical Issues. Lawrence Erlbaum. Pennycook, A. 2006. ‘Postmodernism in language policy’ in T. Ricento (ed.): An Introduction to Language Policy: Theory and Method. Blackwell Publishing. Ricento, T. 1998. ‘National language policy in the United States’ in T. Ricento and B. Burnaby (eds): Language and Politics in the United States and Canada: Myths and Realities. Lawrence Erlbaum. Ricento, T. 2000. ‘Historical and theoretical perspectives in language policy and planning,’ Journal of Sociolinguistics 4/2: 196–213. Ricento, T. and N. H. Hornberger. 1996. ‘Unpeeling the onion: language planning and policy and the ELT professional,’ TESOL Quarterly 30/3: 401–27. Rolstad, K., K. Mahoney, and G. V. Glass. 2005. ‘The big picture: a meta-analysis of program effectiveness research on English Language Learners,’ Educational Policy 19/4, 572–94. Skilton-Sylvester, E. 2003. ‘Legal discourse and decisions, teacher policymaking and the multilingual classroom: constraining and supporting Khmer/English biliteracy in the United States,’ International Journal of Bilingual Education and Bilingualism 6/3,4: 168–84. Stritikus, T. 2002. Immigrant Children and the Politics of English-only. LFB Scholarly Publishing LLC. Thomas, W. P. and V. P. Collier. 1997. School Effectiveness for Language Minority Children. National Clearinghouse for Bilingual Education.

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

(eds): Discourse as Data: A Guide for Analysis. Milton Keynes: The Open University. Fishman, J. S. 1979. ‘Bilingual education, language planning and English,’ English WorldWide 1/1: 11–24. Foucault, M. 1978. The History of Sexuality: An Introduction, Vol. 1. Random House. Freeman, R. D. 1998. Bilingual Education and Social Change. Multilingual Matters. Gale, T. and K. Densmore. 2003. Engaging Teachers: Towards a Radical Democratic Agenda for Schooling. Milton Keynes: Open University Press. Ga´ndara, P. 2000. ‘In the aftermath of the storm: English learners in the post-227 era,’ Bilingual Research Journal 24/1: 2. Garcı´a, E. E. and J. E. Curry-Rodriguez. 2000. ‘The education of limited English proficient students in California schools: an assessment of the influence of proposition 227 in selected districts and schools,’ Bilingual Research Journal 24/1: 2. Garcia, O. and C. Baker. (eds) 2007. Bilingual Education: An Introductory Reader. Multilingual Matters. Gee, J. P. 1990. Social Linguistics and Literacies. Falmer. Genesee, F., K. Lindholm-Leary, W. M. Saunders, and D. Christian. 2006. Education English Language Learners: A Synthesis of Research Evidence. Cambridge: Cambridge University Press. Haugen, E. 1983. ‘The implementation of corpus planning: theory and practice’ in J. Cobarrubias and J. A. Fishman (eds): Progress in Language Planning. Mouton de Gruyter. Hornberger, N. 1988. Bilingual Education and Language Maintenance. Foris Publications. Hornberger, N. H. 1991. ‘Extending enrichment bilingual education: revisiting typologies and redirecting policy I’ in O. Garcı´a (ed.): Bilingual Education: Focusschrift in Honor of Joshua A. Fishman. Volume 1. John Benjamins. Hornberger, N. H. and D. C. Johnson. 2007. ‘Slicing the onion ethnographically: layers and spaces in multilingual language education policy and practice,’ TESOL Quarterly (Special Issue: Language Policies and TESOL: Perspectives from Practice) 41/3: 509–532. Johnson, D. C. 2007. Language Policy within and without the School District of Philadelphia. Unpublished dissertation. University of Pennsylvannia.

D. C. JOHNSON 93

Walford, G. (ed.) 2003. Investigating Educational Policy through Ethnography. JAI. Wiese, A. M. and E. E. Garcia. 2001. ‘The bilingual education act: language minority students and US Federal Educational Policy,’ International Journal of Bilingual Education and Bilingualism 4/4, 229–48. Wiley, T. G. 2002. ‘Assessing language rights in education: a brief history of the U.S. context’ in J. W. Tollefson (ed.): Language Policies in Education: Critical Issues. Lawrence Erlbaum. Wiley, T. G. and W. E. Wright. 2004. ‘Against the undertow: language-minority education policy and politics in the ‘‘age of accountability’’,’ Educational Policy 18/2: 142–68.

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

Thomas, W. P. and V. P. Collier. 2002. A National Study of School Effectiveness for Language Minority Students’ Long-term Academic Achievement. Center for Research on Education, Diversity, and Excellence. Tollefson, J. W. 1991. Planning Language, Planning Inequality: Language Policy in the Community. Longman. Tollefson, J. W. (ed.) 2002. Language Policies in Education: Critical Issues. Lawrence Erlbaum. Tollefson, J. W. 2006. ‘Critical theory in language policy’ in T. Ricento (ed.): An Introduction to Language Policy: Theory and Method. Malden: Blackwell Publishing. Valdez, E. O. 2001. ‘Winning the battle, losing the war: bilingual teachers and postProposition 227,’ The Urban Review 33: 237–53.

Applied Linguistics 31/1: 94–114 ß Oxford University Press 2009 doi:10.1093/applin/amp012 Advance Access published on 17 April 2009

Seizure, Fit or Attack? The Use of Diagnostic Labels by Patients with Epileptic or Non-epileptic Seizures 1

LEENDERT PLUG, 2BASIL SHARRACK, and MARKUS REUBER

2 1

We present an analysis of the use of diagnostic labels such as seizure, attack, fit, and blackout by patients who experience seizures. While previous research on patients’ preferences for diagnostic terminology has relied on questionnaires, we assess patients’ own preferences and their responses to a doctor’s use of different labels through the qualitative and quantitative analysis of doctor–patient interactions in a realistic clinical setting. We also examine whether two sub-groups of patients—those with epileptic seizures and those with (psychogenic) nonepileptic seizures—show different behaviours in this respect. Our findings suggest first that patients make fine lexical distinctions between the various diagnostic labels they use to describe their seizure experiences; secondly, that patients play an active role in the development and application of labels for their medical complaint; and thirdly, that attention to patients’ lexical choices and interactive use or avoidance of labels can be relevant for the differential diagnosis of seizures.

INTRODUCTION It is widely acknowledged that while lexical choice plays a crucial role in human verbal interaction (Hakulinen and Selting 2005), establishing the meaning of alternative lexical items is far from straightforward, especially if contextual factors are taken into account (Fischer 1998; Xiao and McEnery 2006; Nore´n and Linell 2007). In this article, we explore the function of lexical choice and the meanings of a number of related lexical items in the context of a specific type of medical interaction. We focus on the use of diagnostic labels—that is, lexical items that refer to an illness or the symptom of an illness—in interactions between a doctor and patients with recurrent seizures. The choice of diagnostic labels is significant for many reasons. For patients, the labels which describe their illness are a core aspect of their ‘illness representations’ (Leventhal et al. 1992; Horne 1999; Hagger and Orbell 2003). A range of studies have confirmed that ‘the manner in which individuals perceive their illnesses is likely to impact on many aspects of their experience, including the likelihood of seeking help, the particular nature of the help being sought, the degree of adherence to the treatment prescribed, and the

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

Department of Linguistics and Phonetics, University of Leeds, UK and Academic Neurology Unit, University of Sheffield, UK

2

L. PLUG, B. SHARRACK, and M. REUBER

95

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

likelihood of response to such treatment’ (Manber et al. 2003: 335). For example, patients who view their depression as caused by a ‘chemical imbalance’ are unlikely to engage in or respond to psychotherapy, even if this treatment is the most suitable from the physician’s point of view (Manber et al. 2003: 336). Associations between patients’ illness representations and treatment outcomes have been found in diverse clinical scenarios including Irritable Bowel Syndrome and Chronic Fatigue Syndrome (Edwards et al. 2001; Rutter and Rutter 2007), cardiac events (Petrie et al. 2002; Lau-Walker 2004), and epileptic and non-epileptic seizures (NES) (Kemp et al. 1999; Goldstein et al. 2004; Green et al. 2004). For the doctor, it is important to adopt diagnostic labels which are clear, but which do not carry unwanted connotations. Misunderstandings are particularly likely in the area of medically unexplained symptoms, or ‘psychosomatic disorders’, which include NES. NES superficially resemble epileptic seizures, but are not caused by abnormal electrical activity in the brain. NES are an involuntary response to distressing situations, physical, or emotional stimuli (Reuber and Elger 2003). Most patients with NES have a previous history of traumatic experiences or face difficult dilemmas in their current lives. However, they typically fail to recognize the relevance of psychological or emotional factors for their seizures (Reuber and Gru¨newald 2007). They may therefore resist labels that associate their disorder more closely with a psychological problem and prefer labels which imply a physical aetiology, such as epilepsy. However, the clear differentiation of epilepsy and NES is very important to doctors because the two seizure disorders are treated differently. Whereas the first line of treatment for epileptic seizures involves the use of antiepileptic drugs, the treatment of choice for NES is psychotherapy (Reuber and Elger 2003; Reuber et al. 2005). Patients with NES who are inappropriately labelled as having epilepsy, or who fail to understand that they do not have epilepsy, are at risk of receiving unnecessary and potentially harmful medical treatments, and are unlikely to improve. Unfortunately, there is considerable uncertainty amongst doctors about the most appropriate diagnostic label for NES. One paper identified 19 different terms in the recent medical literature (Scull 1997). Whereas some terms give patients the impression that the doctor does not understand the cause of the attacks, others suggest that seizures are ‘put on’ or ‘all in the mind’: in fact, it has been shown that terms such as ‘hysterical seizure’ and ‘pseudoseizure’ are likely to offend patients (Stone et al. 2003). The implications for compliance with proposed treatment are obvious. For the doctor–patient relationship to work effectively, ‘diagnostic labels have to be not only helpful to doctors but also acceptable to patients’ (Stone et al. 2002: 1449; see also Page and Wessely 2003). Whereas the medical literature on diagnostic labels tends to discuss the advantages and drawbacks of particular terms from the doctor’s point of view, this article will focus on patients’ use of labels for their condition or their main symptom. Our study is based on one-to-one conversations between

96

SEIZURE, FIT OR ATTACK?

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

a doctor and 21 patients with epileptic or NES. The study is a part of the project Listening to people with seizures at the University of Sheffield, UK, which set out to improve the differential diagnosis of seizure disorders by analysing the communicative behaviour of patients with epileptic and NES (Schwabe et al. 2007). This project was inspired by a range of sociolinguistic studies carried out at the Bethel Epilepsy Centre and the University of Bielefeld in Germany (Schwabe et al. 2008). Our article seeks to answer two questions. First, we wanted to determine how patients use diagnostic labels for their seizures and which labels they prefer. Stone et al. (2003) employed a questionnaire to address the second half of this question. They asked neurology outpatients a range of variants of the question ‘If you had blackouts, your tests were normal and the doctor said you had pseudoseizures, would that be suggesting that you were putting it on?’ In contrast, our study takes a corpus-based approach (Biber et al. 1994), relying solely on the analysis of actual doctor–patient interactions. This methodology allows us to assess patients’ preferences directly: that is, we can gain insight into patients’ actual preferred use of diagnostic labels, rather than their reported preference in hypothetical scenarios. The sociolinguistic methods applied in Listening to people with seizures are strongly informed by work in Conversation Analysis (Drew et al. 2001; Schegloff et al. 2002), which has a long history of investigating issues of lexical choice and labelling in realistic interactional settings: see for example the work on membership categorization by Sacks (1972, 1992) and Schegloff (2007a, 2007b). Secondly, we set out to explore whether patients with epilepsy and patients with NES differ in their preferences or use of diagnostic labels. Although epileptic and NES look similar to an external observer, the results of the German studies as well as the preliminary findings of the Listening to people with seizures project show that patients with these aetiologically distinct seizure disorders describe their seizures very differently (Schwabe et al. 2007; Plug et al. in press). Broadly speaking, patients with epilepsy readily focus on the seizure experience and on the description of individual episodes, and volunteer a lot of information about how they feel during their seizures. Patients with NES, on the other hand, are more likely to talk about the impact of the seizures on their lives and about the failure of previous treatment; they need to be prompted to focus on how they feel during seizures, and generally provide less detailed information about individual seizure episodes than patients with epilepsy. In view of the superficial similarities between the manifestations of epileptic and NES it is perhaps not surprising that the differential diagnosis represents a serious challenge for doctors, who have to base their diagnosis on seizure descriptions in most cases. Tests carried out in between seizures are only modestly useful in this setting. Seizure recordings are impossible if seizures are infrequent. These difficulties may explain why most patients eventually diagnosed with NES on the basis of the ‘gold standard’ investigation (the synchronous recording of a typical seizure with video and EEG

L. PLUG, B. SHARRACK, and M. REUBER

97

(electroencephalography)) have carried an inaccurate diagnosis of epilepsy for several years (Reuber et al. 2002). Any additional differential diagnostic pointers, for instance from patients’ interactive use or selection of labels, could therefore be very useful in clinical practice.

METHODOLOGY

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

This study is based on the analysis of 21 first clinical encounters between a doctor (M. R., a consultant neurologist) and patients with seizures. The clinical interviews were conducted between August 2005 and July 2007. All patients had been admitted to the neurology ward at the Royal Hallamshire Hospital in Sheffield, UK, for 48 hours of video-EEG monitoring because their admitting neurologist was uncertain whether they had epileptic or NES. All patients had seizures with impairment of consciousness while being monitored. All diagnoses were confirmed by the video-EEG recording of a seizure which was considered typical of the habitual attacks by the patient and a seizure witness. Eight patients were found to have epilepsy, 13 NES. Pseudonyms were used to protect participants’ identity. The study was approved by the South Sheffield Ethics Review Committee and all patients gave written informed consent for their consultations to be recorded and analysed. The consultations were audio- and video-recorded using pre-installed monitoring equipment. Each consultation lasted between 20 and 35 minutes and followed a semi-standardized interview procedure which encourages the doctor to adopt an unusually passive but very attentive role, and which allows patients to develop their own communication agenda (Schwabe et al. 2007; Plug and Reuber in press). One aspect of the interview procedure of particular importance for this study is that the doctor opens the consultation by asking the patient about his/her expectations of the current hospital visit— with no mention of seizures or any other diagnostic label. At the beginning of the interview, it is therefore left to the patient how he/she refers to the seizure episodes. Moreover, the doctor does not present a diagnosis during the consultation. Therefore, while the doctor’s lexical choices may influence the patient’s, the doctor cannot be said to impose a terminological frame of reference: this is largely left to the patient. All 21 consultations were transcribed following standard conversationanalytic conventions. All nouns referring to the patients’ seizures—such as seizure, fit, attack, blackout—were identified and subjected to further qualitative and quantitative analysis. In particular, we were interested in establishing whether different labels were used synonymously or whether differences in meaning could be observed. We were also interested in how the doctor’s use of a particular label affects the patient’s lexical choice in the immediately subsequent speaking turns. Quantitative differences in usage between the doctor, patients with epilepsy and patients with NES were analysed using the Mann–Whitney U-test for independent groups (ordinal variables) and

98

SEIZURE, FIT OR ATTACK?

DIAGNOSTIC LABELS AND OVERALL USAGE PATTERNS Table 1 provides an overview of the diagnostic labels identified in the 21 consultations. The diagnostic label seizure was used most frequently, both by patients and by the doctor (255 instances in total). The term seizure was also used by the greatest number of patients (16 out of 21) and it was used by the doctor in most interviews (16 of 21). The most commonly used alternative labels were fit, attack, and blackout; together, these occurred 240 times in our data. A similar number of patients used the terms fit and attack (12 versus 11, difference not significant); and considering that 40 of the 66 usages of attack were attributable to a single

Table 1: Overview of diagnostic labels used by patients and doctor Diagnostic label

Seizure Fit Attack Blackout Absence, Grand mal, Petit mal, Partial seizure, Tonic clonic (seizure), Dizzy do, Funny do, (Funny) turn, Reaction, Chin thing, (Chin) episode, Blank spell, Collapse

Patients’ usage

Doctor’s usage

N

Npatients

N

Nconsultations

132 41 66 22 10

16 12 11 4 3

123 6 99 6 2

16 3 17 5 1

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

Pearson’s Chi-square test (categorical variables). Two-tailed p-values