APPLI ED LI NGUISTICS
ISSN 1477-450X (ONLINE)
Applied
Linguistics
Volume 31 Number 1 February 2010
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
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, Linköping University
David Block, London University Institute of Education
Jan Blommaert, University of Jyväskylä
Deborah Cameron, University of Oxford
Lynne Cameron, Open University (BAAL Representative)
Tracey Derwing, University of Alberta
Zoltán Dörnyei, 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
ADVERTISING
Inquiries about advertising should be sent to Linda Hann, Oxford Journals Advertising,
60 Upper Broadmoor Road, Crowthorne, RG45 7DE, UK. Email: lhann@talktalk.net.
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 1477-450X (ONLINE)
Applied
Linguistics
Volume 31 Number 1 February 2010
www.applij.oxfordjournals.org
APPLIED LINGUISTICS
Volume 31 Number 1 February 2010
CONTENTS
Articles
Textual Appropriation and Citing Behaviors of University
Undergraduates
LING SHI 1
Practices of Other-Initiated Repair in the Classrooms of Children with
Specific Speech and Language Difficulties
JULIE RADFORD 25
Style Shifts among Japanese Learners before and after Study Abroad
in Japan: Becoming Active Social Agents in Japanese
NORIKO IWASAKI 45
The Relationship between Applied Linguistic Research and Language
Policy for Bilingual Education
DAVID CASSELS JOHNSON 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 94
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 GALASIŃSKA 166
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
LING SHI
University of British Columbia
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 stu-
dents 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)
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
Reliance on source texts
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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 copy-
ing 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 compar-
isons 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 proposi-
tions 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 bor-
rowing or plagiarism.
Nontransgressive plagiarism
A third strand of research used students’ comments to clarify their inappro-
priate 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; Angélil-Carter 2000; Starfield
2002; Chandrasoma et al. 2004; Petrić 2004; Thompson 2005). Those that
L. SHI 3
focused more on students’ voices from interviews were case studies. The rich
interview data in these case studies suggest evidence of unintentional pla-
giarism. 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
(Petrić 2004). There was also evidence that inappropriate use of citations was
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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; Angélil-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 effi-
cient strategy for survival (e.g. Hull and Rose 1989; Currie 1998; Chandrasoma
et al. 2004; Petrić 2004). These findings suggest that plagiarism is a problem of
academic literacy rather than that of dishonesty (Angélil-Carter 2000).
The aforementioned case studies, however, provided limited or a less com-
prehensive 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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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:
(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 partici-
pants, 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
L. SHI 5
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
America
Edward Commerce 1 English 18 Born in North
America
Eddy History 3 English 29 Born in North
America
Carmen Science 1 Cantonese 19 Born in North
America
Carl Arts 2 Cantonese 19 Born in North
America
Candy Science 1 Cantonese 18 7
Carol Arts 1 Cantonese 17 3
Cathy Arts 1 Cantonese 18 3
Cary Science 1 Cantonese 18 10
Martin Commerce 3 Mandarin 20 10
Mark Science 1 Mandarin 18 12
May Arts 2 Mandarin 19 9
Polly Economics 2 Polish 25 7
Jane Law Exchange Japanese 23 23
Kate English Literature Exchange Korean 21 21
Rose Political Science 2 Romanian 19 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 plagia-
rism 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.
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 inter-
view 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
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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 identi-
fied 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 tex-
tual 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 identi-
fied in previous research (Pecorari 2003), there were various other inferences
Table 2: Coding of students’ explanations and examples
Explanations Definitions Examplesa Formatb
Functions
1 Support To provide support for one’s point These are all quotes from their websites. Quote
I used them to support what I wrote
here. (Polly)
2 Form one’s own idea To build up one’s own idea It’s part of my own argument . . . I Summary
mean, I took it from their argument
but incorporated it into my own . . .
overarching description of the
situation. (Eddy)
3 Key point To form the basis of one’s key point I used that [citation] because it forms
the basis of my argument. (Elmer)
Interpretations of source text
4 New information New information for the writer I just took all the information and Summary
put it in my own words. . . . I just
cite the author because I didn’t
know any of the stuff really before.
(Edward)
5 Fact Facts or events This is just like facts . . . so I put the Summary
reference here. . . . I am summarizing
her points. (Edward)
6 Research finding Previous research findings I cite that, saying that who did the
research and what they found. (Cary)
7 Background information Background information or I got that from this article. . . . It’s just
established theory in the field like the background information.
(Edward)
I rephrased these from the article. . . . Paraphrase
I mean it’s . . . fairly well established
L. SHI
(continued)
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
Table 2: Continued
Explanations Definitions Examplesa Formatb
8 Credible source Credible or reliable source This is a citation from a book written by Paraphrase
a concert artist. . . . I think he’s very
credible. . . . It’s paraphrased here.
(Cathy)
9 Common knowledge or term Knowledge or words commonly I didn’t cite it because that’s common
known sense. (Mark)
I didn’t cite it because it’s an actual
term that people use. (Martin)
Reasons related to learning
10 Other’s words/ideas Words and ideas directly taken These are the actual words [from Quote
from the source the source]. You have to quote them,
right? (Edward)
11 Result of learning Knowledge accumulated by And this is kind of I just knew from my
learning mum . . . and from previous school . . . I
don’t think I really have to cite it. (Cary)
12 Reference cited earlier A source cited earlier or in the I used my own words and it’s a summary. Summary
in the text or in the reference list I didn’t cite because I already
reference list 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)
8 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS
13 No need to cite everything Not everything needs to be cited Or else I would have to cite every single
sentence. I think it’s kind of distracting.
(Candy)
14 Teachers’ preference Teacher prefers citations They (teachers) prefer you rephrase Paraphrase
and cite it. (Cary)
a
Keywords that result in relevant coding are highlighted.
b
Coding of the example comments in terms of quote, paraphrase, and summary when relevant information is available.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
L. SHI 9
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 influ-
enced 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
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
borrowinga
Functions
1 Support 48 96 2 4 50 100 33
2 Form one’s own point 7 30 16 70 23 100 6
3 Key point 12 100 0 0 12 100 10
Subtotal and percentage of 236 67 79 18 21 85 100 49
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.
L. SHI 11
50
Frequency of mentions
40
30 With citing
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
20 Without citing
10
0
rt in
t
ion ct ng e as e
o
po at Fa i ur
c
de nc
pp fi nd so s /i
e re
Su Ke
y
fo
r m
ch le rd re
f
in ear dib wo s 'p
w s e 's r
Ne Re Cr er he
th ac
O Te
Figure 1: Leading reasons given for use of source text, with citing
50
Frequency of mentions
40
30
With citing
20 Without citing
10
0
e
t
g
g
t
lis
in
dg
in
n
hi
po
rn
ce
le
yt
ea
ow
en
n
er
ow
fl
ev
kn
r
fe
to
s
re
te
on
e'
ul
ci
on
e
m
es
th
to
om
R
m
in
ed
r
C
Fo
or
ne
er
o
rli
N
ea
d
te
ci
ce
en
er
ef
R
Figure 2: Leading reasons given for use of source text, when not citing
12 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS
The dominant party is generally able to get its own way in policy-
making issues in the parliament. (Stockwin 1998: 121)
Jane’s reason for citing, which has been classified as an instance of support:
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
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
I keep putting in references. . . . I had this idea that if you have
dominance in the Diet, you must have big influence in the par-
liament. 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.
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 cita-
tions . . . [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
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
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’:
In his essay ‘‘Drug Dealers,’’ Caulkins describes the ‘‘localization’’
of certain areas of the city for drug use as a major problem lead-
ing 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 compar-
ison, 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. Angélil-Carter 2000;
Chandrasoma et al. 2004).
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 disci-
plinary communities.
Like the notion of fact, the concept of new information reveals the role
of individual learning experiences. For example, the present students cited
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
such new information as:
‘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 owner-
ship of knowledge. Judgements on textual appropriation, whether appreciated
or illegitimate, are grounded in the context of epistemologically and socially
constructed academic literacies.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
May generalized Starbucks’ way of preparing coffee to a common sense of
how to keep food fresh:
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 knowl-
edge based on their own understandings and interpretations, other students
made similar decisions based on whether the source information was intro-
duced 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 illus-
trate 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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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.
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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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:
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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
Paraphrase Summary Quote Subtotal
Functions
1 Support 3 1 29 33 17 50
2 Form one’s 1 3 2 6 17 23
own point
3 Key point 2 1 7 10 2 12
Subtotal 6 2 3 38 49 36 85
Interpretations of
source text
4 New information 3 3 6 12 10 22
5 Fact 1 2 4 7 7 14
6 Research finding 1 1 4 5
7 Background 1 1 1 1 4 2 6
information
8 Credible source 1 1 5 7 8 15
9 Common 1 1 2 13 15
knowledge
or term
Subtotal 5 2 6 4 16 33 44 77
Reasons related
to learning
10 Other’s 32 32 5 37
words/ideas
11 Result of learning 1 1 17 18
12 Reference cited 1 1 2 3
earlier or in
the reference
list
13 No need to cite 3 3 8 11
everything
14 Teachers’ 3 3 2 5
preference
Subtotal 3 1 4 32 40 34 74
Total 14 3 8 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.
L. SHI 19
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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 rele-
vant knowledge or achievement of academic literacies.
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
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.
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
ent 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 learn-
ing 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.
CONCLUSION
The present study illustrates the sophistication and range of the participating
students’ understanding of the role of textual borrowing and citation by ana-
lyzing 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 tex-
tual 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 pre-
viously. Students’ citational acts are situated in a learning context where cer-
tain 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
22 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS
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
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
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 pres-
ent 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 identi-
fied 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
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 discipline-
related differences that may well speak to distinctions that instructors in
these fields may wish to address directly with students. Together, these stud-
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
ies should build up a theory of citing as a process of learning for novice
academic writers.
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 2 In all examples cited in this article,
of her units as translations from a students’ writing and comments are
source in her first language. Another presented verbatim. For easy read-
student (Martin) said he created a ing, keywords in students’ comments
quote himself but attributed it to that result in relevant coding are
a credible author. Since these cases highlighted and alternatives for
raise different issues and also repre- special terms in student writing are
sent a small proportion of the data, provided in brackets. See Appendix C
I decided not to deal with them in in the online version of this article for
the present analyses. cited works in student writing.
REFERENCES
Angélil-Carter, S. 2000. Stolen Language? of plagiarism,’ Asian Journal of English
Plagiarism in Writing. Essex: Pearson Language Teaching 10: 91–113.
Education Limited. Chandrasoma, R., C. Thompson, and
Campbell, C. 1990. ‘Writing with others’ words: A. Pennycook. 2004. ‘Beyond plagiarism:
Using background reading text in academic transgressive and nontransgressive intertex-
compositions’ in B. Kroll (ed.): Second Language tuality,’ Journal of Language, Identity, and
Writing: Research Insights for the Classroom. Education 3: 171–93.
Cambridge: Cambridge University Press. Currie, P. 1998. ‘Staying out of trouble: apparent
Chandrasegaran, A. 2000. ‘Cultures in contact plagiarism and academic survival,’ Journal of
in academic writing: students’ perceptions Second Language Writing 7: 1–18.
24 STUDENT’S TEXTUAL APPROPRIATION AND CITING BEHAVIORS
Deckert, G. D. 1993. ‘Perspectives on plagiarism writing,’ Journal of Second Language Writing 12:
from ESL students in Hong Kong,’ Journal of 317–45.
Second Language Writing 2: 131–48. Pecorari, D. 2006. ‘Visible and occluded cita-
Denzin, N. K. 1997. Interpretive Ethnography: tion features in postgraduate second-language
Ethnographic Practices for the 21st Century. writing,’ English for Specific Purposes 25: 4–29.
London: Sage Publications. Pennycook, A. 1994. ‘The complex contexts of
Dong, Y. R. 1996. ‘Learning how to use cita- plagiarism: a reply to Deckert,’ Journal of Second
Downloaded from applij.oxfordjournals.org at Anhui University of Finance and Economics on December 31, 2010
tions for knowledge transformation: non- Language Writing 3: 277–84.
native doctoral students’ dissertation writing Pennycook, A. 1996. ‘Borrowing other’s words:
in science,’ Research in the Teaching of English text, ownership, memory, and plagiarism,’
30: 428–57. TESOL Quarterly 30: 201–30.
Fairclough, N. 1992. Discourse and Social Change. Petrić, B. 2004. ‘A pedagogical perspective on
Cambridge: Polity Press. plagiarism,’ NovELTy 11: 4–18.
Hodges, J. C. 1962. Harbrace College Handbook. Roig, M. 1997. ‘Can undergraduate students
5th edn. New York: Harcourt, Brace & World determine whether text has been plagiarized?’
Inc. Psychological Record 47: 113–23.
Holstein, J. A. and J. F. Gubrium. 2004. Shi, L. 2004. ‘Textual borrowing in second
‘The active interview’ in D. Silverman (ed.): language writing,’ Written Communication 21:
Qualitative Research: Theory, Method and Practice. 171–200.
London: Sage Publications. Sinclair, J. M. 1986. ‘Fictional worlds’
Howard, R. M. 1992. ‘A plagiarism pentimento,’ in M. Coulthard (ed.): Talking about Text.
Journal of Teaching Writing 11: 233–45. Birmingham: University of Birmingham ELR.
Howard, R. M. 1995. ‘Plagiarism, authorships, Spack, R. 1997. ‘The acquisition of academic
and the academic penalty,’ College English 57: literacy in a second language: a longitudinal
788–806. case study,’ Written Communication 14: 3–62.
Hull, G. and M. Rose. 1989. ‘Rethinking reme- Starfield, S. 2002. ‘ ‘‘I’m a second-language
diation,’ Written Communication 6: 139–54. English speaker’’: negotiating writer identity
McCormick, F. 1989. ‘The Plagiario and the and authority in Sociology One,’ Journal of
professor in our peculiar institution,’ Journal Language, Identity, and Education 1: 121–40.
of Teaching Writing 8: 133–45. Tadros, A. 1993. ‘The pragmatics of text averral
Miller, J. and B. Glassner. 2004. ‘The ‘‘inside’’ and attribution in academic texts’ in M. Hoey
and the ‘‘outside’’: finding realities in inter- (ed.): Data, Description, Discourse. London:
views’ in D. Silverman (ed.): Qualitative HarperCollins.
Research: Theory, Method and Practice. London: Thompson, C. 2005. ‘‘Authority is everything’:
Sage Publications. a study of the politics of textual ownership and
Moore, T. 1997. ‘From test to note: cultural knowledge in the formation of student writer
variation in summarization practices,’ Prospect identities,’ International Journal for Educational
12: 54–63. Integrity 1. Available at http://www.ojs.unisa.
Odell, L., D. Goswami, and A. Herrington. edu.au/journals/index.php/IJEI/article/view/18.
1983. ‘The discourse-based interview: a Accessed 15 December 2007.
procedure for exploring the tacit knowl- Watson-Gegeo, K. A. 1988. ‘Ethnography in
edge of writers in non-academic settings’ ESL: defining the essentials,’ TESOL Quarterly
in P. Mosenthal, L. Tamor, and S. Walmsley 22: 575–92.
(eds): Research on Writing. New York: Longman. Young, K. M. and L. Gaea. 1998. ‘Writing from
Pecorari, D. 2003. ‘Good and original: plagiarism primary documents: a way of knowing in
and patchwriting in academic second-language history,’ Written Communication 15: 25–68.
Applied Linguistics 31/1: 25–44 ß Oxford University Press 2008
doi:10.1093/applin/amn046 Advance Access published on 6 December 2008
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
26 REPAIR PRACTICES IN SSLD
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 char-
acterized by the teacher initiating a ‘test’ question (I) that is followed by a
known answer/response (R) (Macbeth 2004). The third turn position, follow-
ing 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
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.
A 8 years 7 months F 3 2 0 0 1 2 1
REPAIR PRACTICES IN SSLD
B 8 years 3 months F 2 2 0 0 2 2 1
C 4 years 4 months M 2 2 1 0 2 2 1
D 4 years 8 months F 3 3 2 0 3 3 0
E 5 years 10 months F 3 3 1 0 2 2 1
F 5 years 5 months F 3 3 1 0 3 3 3
FINDINGS
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
‘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-spe-
cific 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.
J. RADFORD 33
potential location in line 4. However, as the reported event is the child’s experi-
ence, 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 infor-
mation 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
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
38 REPAIR PRACTICES IN SSLD
At the start of the sequence the setting is being described and an additional
character suggestion is made at line 3.
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 defini-
tion, 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
APPENDIX
Transcription key1
NOTES
1 The transcription conventions used in are based on the system originally
this article (as shown in the Appendix) developed by Gail Jefferson, as
J. RADFORD 43
REFERENCES
types and learner repair in immersion class- Ridley, J., J. Radford, and M. Mahon. 2002.
rooms,’ Language Learning 48: 183–218. ‘How do teachers manage topic and repair?’
Macbeth, D. 2004. ‘The relevance of repair for Child Language Teaching and Therapy 18/1: 43–58.
classroom correction,’ Language in Society 33: Sadler, J. and K. Mogford-Bevan. 1997.
703–36. ‘Teacher talk with children with language dis-
Mehan, H. 1979. ‘ ‘‘What time is it Denise?’’: orders: four case studies 2,’ Child Language
asking known information questions in class- Teaching and Therapy 11: 37–58.
room discourse,’ Theory into Practice 18/4: Schegloff, E. A. 1972. ‘Notes on a conversa-
285–94. tional practice,’ in D. Sudnow (ed.): Studies in
Mercer, N. 1995. The Guided Construction of Social Interaction. New York: Free Press.
Knowledge: Talk Amongst Teachers and Learners. Schegloff, E. A. 1997. ‘Practices and actions:
Clevedon: Multilingual Matters. Boundary cases of other-initiated repair,’
Mercer, N. 2000. Words and Minds: How We Use Discourse Processes 23: 499–545.
Language to Think Together. London: Routledge. Schegloff, E. A., G. Jefferson, and H. Sacks.
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 situa-
tions, 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
46 STYLE SHIFTS AMONG JAPANESE LEARNERS
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
N. IWASAKI 47
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 pro-
fessor and a former student the student occasionally used the plain style for
exclamatory, emotional, or soliloquy-like expressions despite their social status
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
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
meanings that contrast with the polite form depends on the degree of subor-
dination 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
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.
Polite Plain Copula Repair Total Polite Plain Copula Repair Total
n (%) n (%) omission n (%) n (%) n (%) omission n (%) n (%)
n (%) n (%)
Greg 74 (79) 9 (10) 8 (9) 3 (3) 94 13 (10) 96 (73) 23 (17) 0 (0) 132
Henry 73 (72) 16 (16) 8 (8) 5 (5) 102 31 (20) 103 (67) 20 (13) 1 (1) 155
Alan 128 (87) 14 (10) 1 (1) 4 (3) 147 172 (86) 21 (11) 5 (3) 1 (1) 199
Sam 107 (84) 16 (13) 2 (2) 2 (2) 127 106 (85) 17 (14) 2 (2) 0 (0) 125
Peter 134 (80) 27 (16) 4 (2) 2 (1) 167 127 (70) 44 (24) 10 (5) 1 (1) 182
Matrix kara, kedo, node, si Total Matrix kara, kedo, node, si Total
clauses ga clauses clauses clauses ga clauses clauses
Greg 0 0 9 9 83 12 0 95
Henry 5 11 0 16 67 36 0 103
Alan 1 9 4 14 10 9 2 21
Sam 0 8 8 16 3 2 12 17
Peter 3 15 9 27 17 13 14 44
52 STYLE SHIFTS AMONG JAPANESE LEARNERS
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
‘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)
‘Member? Well, [inaudible] people, well, how many, about 15, uh, well, there are
‘Because they are important people, it (the club) is very interesting, interesting, it
is very interesting.’
at-all know-NEG-AH Q
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 exam-
ined below. Though it is likely that the interviewer’s choice of style would
influence the interviewee’s style choice, the current interviewer consis-
tently 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).
5 → 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.’
Greg predominantly used plain forms in this sequence as well, and his moti-
vation 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
‘I asked you a lot of questions about your study abroad, but do you have any
‘Well, because Alan wants to take the level 1 proficiency exam, he is looking for
‘Uh, teacher, I am very sorry. I did not, could not make it to the exam.’
5 → Etoo, soreni, eto, keitai-o motte imasen kara, etto, renraku-dekimasen desita.
PAST
‘Well, besides, uh, because I do not have a cell phone, well, I could not contact
you.’
58 STYLE SHIFTS AMONG JAPANESE LEARNERS
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
‘Then could you select one and tell me what kind of person s/he is?’
‘Well, (she) can speak Spanish. (She) can speak very well.’
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 under-
stood 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.
N. IWASAKI 59
1 I: Dooyatte, sono sumu tokoro-o sagasitari, issyoni sumu hito-o sagasita n desu ka.
NONPAST-AH Q
‘How did you look for the place to live and people to live with?’
3 sono ie-wa kurabu-no miitingu-de, sono ie-ga aiterutte itte masita kara,
‘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
2 H: Aa, un, ma, minna, ongaku su ongaku, uh, gakki-o, gakki, hikemasita, minna.
everyone
everyone.’
60 STYLE SHIFTS AMONG JAPANESE LEARNERS
PAST because
‘Two can play the guitar, and one can play could play the piano and so we often
I: de, uti-no sigoto-wa minna de, ano, kootai de sita n desu ka.
‘What is kootai?’
3 I: Hai.
‘Yes.’
N. IWASAKI 61
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
uh my window-nom break-AH-PAST
ah so COP-NONPAST-AH Q
‘Well, while a child was playing baseball, his ball hit my window, and it broke.’
‘I can’t open it, and the window glass, well, the window glass is bro.., or rather, is
gone now’.
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
well practice-do-NONPAST-AH
‘...well, I do practice.’
N. IWASAKI 63
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
‘Ah, about one, one month, first I wal..walked to the Japan Sea.’
‘That’s amazing.’
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.
64 STYLE SHIFTS AMONG JAPANESE LEARNERS
GER 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
3 S: Ha....hairu ka kimemasu.
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 chal-
lenged 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)
but
N. IWASAKI 65
‘Well, isn’t it okay to layoff bad employees? I think... Otherwise, it might affect
In line 2, he spontaneously states his thought using the plain form. Alterna-
tively, 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
2 I: Mondai?
‘Problem?’
3 S: unto Kyoko-san-no kuruma-o unto, unten-site iru toki, kootuu ziko-ga atta n da.
NONPAST
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 mean-
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 soliloquy-
like 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 instruc-
tion, valued ‘being creative’ or ‘imitating teachers’, as in excerpt (21).
(21) (Peter, Pre-OPI)
‘What kinds of things could you not get used to, at first?’
and
ato-wa, iroina sinpaigoto-ga atta kara, ma, nanka, hakkiri ienai n desu kedo,
NONPAST-AH but
‘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
sandwich make-such-as do
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 pre-
and 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.
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
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
Particles
TOP topic marker
NOM nominative
NOTES
1 Japanese also has ‘referent honorifics’ interactions with their peers where
through which the speaker expresses sounding distant may be negatively
deference towards the referent viewed.
or humbleness (when referring to 3 In highly dependent clauses, the use
his/her own action). These are realized of the polite-form predicates
by verb choice. results in hyper-politeness. For exam-
2 This assumption may be question- ple, ittara ‘if I go’ can be expressed as
able when considering the possibility ikimasitara, which sounds excessively
that students primarily participate in polite.
REFERENCES
Althen, G. 1988. American Ways: A Guide for Atsuzawa-Windley, S. and S. Noguchi. 1995.
Foreigners in the United States. Yarmouth, ‘Effects of in-country experience on the acqui-
Maine: Intercultural Press. sition of oral communication skills in
N. IWASAKI 71
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 ideolog-
ical, and presents a rich picture of language policy development as one aspect
amongst many socio-political processes which may perpetuate social inequal-
ity, 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 edu-
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 ethno-
graphic 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
D. C. JOHNSON 75
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 ethno-
graphic 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
scientifically based research; and (ii) the agencies must believe in the effective-
ness 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
THE SDP
The SDP is an example of a local education agency responsible for interpret-
ing 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 popula-
tion 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
80 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: But it’s really evolved – it’s become an enrichment model
whereas before it was just remediation or a deficit type
Emily asserts that NCLB is simply catching up with what they already knew
in the SDP—developmental bilingual education is an effective and research-
based 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 inter-
prets 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
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 pro-
grams, Lucı́a Sanchez, whose beliefs about applied linguistic research were
different than Emily’s and Eve’s and influenced language policy accordingly.
Emily Dixon-Marquez and Lucı́a Sanchez are an interesting comparative jux-
taposition of how language (education) ideologies influence interpretation and
appropriation of top-down language policies like Title III. Lucı́a started work-
DISCUSSION
At this point, I would like to return to the research question proposed earlier:
how does applied linguistic research shape the interpretation and appropria-
tion 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 class-
rooms. 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 admin-
istrator 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
88 APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY
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 obfus-
cated or discredited (Ball 2006). The production of Title III was characterized
by limiting definitions of bilingual education as necessarily transitional and
90 APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY
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
TRANSCRIPTION NOTES
() transcription doubt
(?) unclear utterance
... ellipsis
- short pause
Italics emphatic stress
D. C. JOHNSON 91
NOTES
1 The study upon which this is based 5 This department is responsible ‘for the
won the 2008 National Association administration of new formula grants
of Bilingual Education dissertation and for providing technical assistance
competition. to State and Local educational
2 Levinson and Sutton (2001) prefer the agencies’, http://www.ed.gov/about/
term appropriation to implementation offices/list/oela/aboutus.html
which they feel implicitly ratifies a 6 All names are pseudonyms.
top-down perspective. I agree and 7 This passage might also be subject to
adopt this term. varying interpretations (Johnson
3 While other ethnographic work deals 2007).
REFERENCES
August, D. and T. Shanahan. (eds) 2006. Crawford, J. 1998. ‘Language politics in the
Developing Literacy in Second-language Learners: U.S.A.: the paradox of bilingual education,’
Report of the National Literacy Panel on Social Justice 25/3: 50–69.
Language-minority Children and Youth. Cummins, J. 1981. ‘The role of primary lan-
Lawrence Erlbaum and Center for Applied guage development in promoting educational
Linguistics. success for language minority students in
Bakhtin, M. 1986. Speech Genres and Other Late California State Department of Education’ in
Essays. Texas University Press. Schooling and Language Minority Students: A
Ball, S. J. 2006. Education Policy and Social Class: Theoretical Framework. Evaluation, Dissemina-
The Selected Works of Stephen J. Ball. Taylor & tion and Assessment Center California State
Francis. University.
Baltodano, M. P. 2004. ‘Latino immigrant par- Davis, K. A. 1994. Language Planning in
ents and the hegemony of Proposition 227,’ Multilingual Contexts: Policies, Communities, and
Latino Studies 2: 246–53. Schools in Luxembourg. John Benjamins
Blommaert, J. 1996. ‘Language planning as a Publishing Company.
discourse on language and society: The linguis- Davis, K. A. 1999. ‘The sociopolitical dynamics
tic ideology of a scholarly tradition,’ Language of indigenous language maintenance and loss:
Problems and Language Planning 20/3, 199–222. a framework for language policy and planning’
Canagarajah, S. 2006. ‘Ethnographic methods in in T. Huebner and K. A. Davis (eds):
language policy’ in T. Ricento (ed.): An Sociopolitical Perspectives on Language Policy and
Introduction to Language Policy: Theory and Planning in the USA. John Benjamins
Method. Malden: Blackwell Publishing. Publishing Company.
Corson, D. 1999. Language Policy in Schools: A Fairclough, N. 2001. ‘The discourse of new
Resource for Teachers and Administrators. labour: Critical discourse analysis’
Lawrence Erlbaum. in M. Wetherell, S. Taylor, and S. J. Yates
92 APPLIED LINGUISTIC RESEARCH AND LANGUAGE POLICY
(eds): Discourse as Data: A Guide for Analysis. Krashen, S. 1996. Under Attack: The Case against
Milton Keynes: The Open University. Bilingual Education. Language Education
Fishman, J. S. 1979. ‘Bilingual education, lan- Associates.
guage planning and English,’ English World- Krashen, S. and G. McField. 2005. ‘What
Wide 1/1: 11–24. works? Reviewing the latest evidence on
Foucault, M. 1978. The History of Sexuality: An bilingual education,’ Language Learner 1/2:
Introduction, Vol. 1. Random House. 7–10, 34.
Freeman, R. D. 1998. Bilingual Education and Lemke, J. L. 1995. Textual Politics: Discourse and
Social Change. Multilingual Matters. Social Dynamics. London: Taylor & Francis.
Gale, T. and K. Densmore. 2003. Engaging Levinson, B. A. U. and M. Sutton. (eds) 2001.
Teachers: Towards a Radical Democratic Agenda Policy as Practice: Toward a Sociocultural Analysis
for Schooling. Milton Keynes: Open University of Educational Policy. Ablex Publishing.
Press. Lindholm-Leary, K. J. 2001. Dual Language
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; Norén 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 repre-
sentations’ (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
L. PLUG, B. SHARRACK, and M. REUBER 95
likelihood of response to such treatment’ (Manber et al. 2003: 335). For exam-
ple, 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,
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
METHODOLOGY
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
N Npatients N Nconsultations
patient, the number of instances in which patients used the terms fit and attack
were also rather similar.
In the cases of fit and attack, there were interesting differences in usage
between doctor and patients. Patients used fit much more commonly than
the doctor. The doctor on the other hand showed a marked preference for
the label attack. Whereas he used attack 99 times in 17 consultations, he only
used fit 6 times in 3 consultations. This means that there were several
consultations in which a patient used the label fit repeatedly, but the doctor
chose not to adopt the label. Conversely, there were a number of encounters
in which the doctor used attack repeatedly, but the patient persisted in their
use of alternative labels. The differences between patients’ and doctor’s usage
The clearest evidence that (like fit) blackout is a more specific diagnostic label
than seizure is provided by Chris; in particular in the fragment given in (2).
(2) Doctor: well can I (.) take you back to your first (1.0) seizure (1.6)
Chris: which one
Doctor: well you know you’ve come here because of these blackouts (0.3)
what about the first one you can remember (0.3) what can you tell
me about that (1.6)
Chris: the blackout or the (.) fit cos i’m having like (.) two different types at
the moment
Chris’ subsequent usage of blackout suggests that both ‘fits’ and ‘blackouts’
In (3a), Chris indicates that what he calls ‘blackouts’ are seizures that
involve him ‘just dropping’—but not ‘jerking’. While Alastair also describes
seizures in which he loses control over his body, he suggests that in some
‘nothing happens’, and he remains stationary while his mind has ‘gone’ (3b
and c). Sue’s use of blackouts in (3d) highlights a possible mismatch between
the diagnostic label that the patient may assign to a seizure-related event
on the basis of his/her own experience, and that of onlookers. Sue indicates
that she started experiencing brief periods of unconsciousness and loss
of memory. She was not aware of her physical state during these periods;
therefore she classified as ‘blackouts’ what to those around her looked
like ‘fits’.
The usage of attack by the patients (as well as the doctor) suggests that it is
synonymous with seizure; that is, it is a diagnostic label with a more general
denotation than fit and blackout. Relevant examples are given in (4).
(4) a. ‘it’s been ver- very – they’ve been very disruptive to my life, the, the er, the
seizures, the attacks I’ve been having’ (Zack)
b. ‘I had another, erm, I had another, er, seizure, but there again, I d- I don’t
know what happened with these attacks’ (Jack)
c. ‘I was having a fit at the time [. . .] everything was going off-scale in that
attack’ (Betty)
d. ‘it must be difficult to remember individual seizures when it’s – they’re such
long way back, but maybe, maybe you can tell me about the last attack, what
– just explain exactly what, what, what happened in, in the last seizure that
you had’ (Doctor to Sandra)
102 SEIZURE, FIT OR ATTACK?
e. ‘so you, you said you’re scared after seizures [. . .] and how do you feel in the
attack?’ (Doctor to Laura)
Both Zack and Jack use seizure and attack within one speaking turn, with no
indication that the two labels differ in meaning (4a and b). Betty’s use of both
fit and attack in (4c) is consistent with the idea that the former is a sub-type
of the latter: she describes one particular ‘attack’ as involving her ‘having a
fit’. The doctor uses attack and seizure interchangeably on several occasions,
as seen in (4d and e).
Most of the less frequent diagnostic labels, in particular the epilepsy-specific
terms such as absence, partial seizure, grand mal, and petit mal refer to particular
types of seizures, but we have too few occurrences of these labels in our data
Alastair: erm (0.3) I was just hoping that we could find (1.0) what it was (0.3)
that was causing them?
Doctor: mmm
Alastair: hopefully lead to something that would stop me having them so I
could continue having a normal life
Doctor: mmm
Alastair: you know cos er (.) it’s been very disrupted since it started
Doctor: it’s been very disrupted
Alastair: yeah (.) well cos I’ve been off (.) of work on (.) at work off work
(several lines omitted; no diagnostic label)
they come and go so I can’t really (2.2) risk being in a car
Doctor: mmm
We have already mentioned patients’ resistance to the use of the label attack,
which the doctor used frequently as a synonym of seizure: only 11 out of our
21 patients used it. While one of these eleven patients used the label more
than 40 times during the consultation, most others used it sparingly. As seen
above, six patients chose not to adopt the label despite the doctor’s repeated
use. Illustrative fragments are given in (12) and (13). In both, the doctor’s
inquiry contains attack, but in their responses the patients choose to use an
alternative label.
(12) Doctor: so you (.) came in here to learn about the small
Ken: yeah
Doctor: attacks
Here, Tallulah’s response to the doctor’s inquiry takes the form of a repetition
of part of the inquiry: ‘not waking up from a seizure’. However, Tallulah
hesitates before ‘a seizure’ and subsequently cuts off its production in favour
of ‘having a fit’. Her self-repair, which constitutes a marked deviation from
the part-turn repetition format and a marked refusal to align with the doctor
in terms of lexical choice, makes it very clear that the adoption of the diag-
nostic label seizure is problematic for her.
In addition, the hesitations and hedging remarks illustrated in (11) above
most frequently involve the use of the label seizure. Some more examples are
given in (15).
and 56 by patients with NES. This means that the mean number of instances
per patient was 9.5 for the epilepsy sub-group and 4.3 for the NES sub-group.
First, we can therefore state that patients with epilepsy used seizure more
frequently than patients with NES. When we do the same comparison for fit
and blackout we find the reverse pattern. Of the 41 instances of fit, 31 were
produced by patients with NES (2.4 instances per patient) and 10 by patients
with epilepsy (1.3 instances per patient). All 22 instances of blackout were
produced by patients with NES. That is, patients with epilepsy preferred seizure
over fit and blackout, and patients with NES used fit and blackout but appeared
to disprefer the use of seizure. The pattern is summarized in Figure 1. A Mann–
Whitney U-test reveals that the difference for seizure is statistically significant
10
Epilepsy
9
NES
8
7
Instances per patient
0
Seizure Fit Blackout
Diagnostic label
Figure 1: Bar chart showing mean instances per patient for seizure, fit, and
blackout in epilepsy and NES sub-groups
patient population may reveal patterns for which our proposals cannot
account, and will provide a stronger empirical basis for the statement of the
lexical meanings of the various diagnostic labels used by doctors and patients.
Moreover, our analysis somewhat undervalues the fact that parameters of
lexical choice may shift during the course of an interaction (Hakulinen and
Selting 2005; Norén and Linell 2007). While we have discussed in some detail
several contexts in which individual labels are recurrently used and have
commented on the extent to which the doctor’s usage of terminology influ-
ences the patient’s and vice versa, our observations only scratch the surface
of the sequential-interactional dimension of lexical choice. We must leave this
dimension as an area for further research.
with the doctor, one would not expect patients with NES to use these
terms more frequently than patients with epilepsy: it is patients with NES
who tend to stress the negative impact of the seizures on their lives. Patients
with epilepsy are more likely to communicate that they are actively engaged
in minimizing the impact of the seizures on their daily lives, and that they
are coping as well as they can. We suggest that it may be this higher level of
engagement with the process of diagnosis and treatment which makes patients
with epilepsy more likely to adopt vocabulary which we have characterized as
‘medical’ in consultations than patients with NES. The latter patients’ general
resistance to engaging in a detailed discussion of their seizure experience
would seem compatible with a resistance to the use of this type of vocabulary,
REFERENCES
Biber, D., S. Conrad, and R. Reppen. 1994. Gerhardt, U. 1989. Ideas About Illness. An
‘Corpus-based approaches to issues in Intellectual and Political History of Medical
applied linguistics,’ Applied Linguistics 15: Sociology. Macmillan.
169–89. Gill, V. T. and D. W. Maynard. 1995. ‘On
Bishop, D. V. M. 2003. The Test for Reception ‘‘labelling’’ in actual interaction: delivering
of Grammar, version 2 (TROG-2). Psychological and receiving diagnoses of developmental dis-
Corporation. abilities,’ Social Problems 42: 11–37.
Drew, P., J. Chatwin, and S. Collins. 2001. Goldstein, L. H., A. Deale, S. Mitchell-
‘Conversation analysis: A method for research O’Malley, B. K. Toone, and J. D. C. Mellers.
into interactions between patients and health- 2004. ‘An evaluation of cognitive behavioral
care professionals,’ Health Expectations 4: 58–70. therapy as a treatment for dissociative seizures,’
Edwards, R., R. Suresh, S. Lynch, Cognitive Behavioral Neurology 17: 41–49.
P. Clarkson, and P. Stanley. 2001. ‘Illness Green, A., S. Payne, and R. Barnitt. 2004.
perceptions and mood in chronic fatigue ‘Illness representations among people with
syndrome,’ Journal of Psychosomatic Research nonepileptic seizures attending a neuro-
50: 65–68. psychiatry clinic: A qualitative study based
Fischer, K. 1998. From Cognitive Semantics on the selfregulation model,’ Seizure 13:
to Lexical Pragmatics. Mouton de Gruyter. 331–39.
L. PLUG, B. SHARRACK, and M. REUBER 113
Hagan, J. 1973. ‘Labeling and deviance: A case in the doctor–patient encounter,’ Journal of
study in the sociology of the interesting,’ Social the Royal Society of Medicine 96: 223–27.
Problems 20: 447–58. Peräkylä, A. 1998. ‘Authority and accountabil-
Hagger, M. S. and S. Orbell. 2003. ‘A meta- ity: The delivery of diagnosis in primary health
analytic review of the common-sense model care,’ Social Psychology Quarterly 61: 301–20.
of illness representations,’ Psychological Health Petrie, K. J., L. D. Cameron, C. J. Ellis,
18: 141–84. D. Buick, and J. Weinman. 2002. ‘Changing
Hakulinen, A. and M. Selting. (eds) 2005. illness perception after myocardial infarction:
Syntax and Lexis in Conversation. John An early intervention randomized controlled
Benjamins. trial,’ Psychosomatic Medicine 64: 580–86.
Halkowski, T. 2006. ‘Realising the illness: Plug, L. and M. Reuber. 2009. ‘Making the
Patients’ narratives of symptom discovery’ diagnosis in patients with blackouts: It’s all in
in J. Heritage and D. Maynard (eds): the history,’ Practical Neurology 9: 4–15.
applied linguistics,’ Annual Review of Applied 2003. ‘What should we call pseudoseizures?
Linguistics 22: 3–31. The patient’s perspective,’ Seizure 12: 568–72.
Schwabe, M., S. J. Howell, and M. Reuber. Stone, J., W. Wojcik, D. Durrance, A. Carson,
2007. ‘Differential diagnosis of seizure disor- S. Lewis, L. MacKenzie, C. P. Warlow, and
ders: A conversation analytic approach,’ Social M. Sharpe. 2002. ‘What should we say
Science and Medicine 65: 712–24. to patients with symptoms unexplained by
Schwabe, M., M. Reuber, M. Schöndienst, disease? The ‘‘number needed to offend’’,’
and E. Gülich. 2008. ‘Listening to people British Medical Journal 325: 1449–50.
with seizures: How can Conversation analysis Warrington, E. K. 1997. ‘The graded naming
help in the differential diagnosis of seizure dis- test: A restandardisation,’ Neuropsychological
orders?’ Communication and Medicine 5: 59–72. Rehabilitation 7: 143–46.
Scull, D. A. 1997. ‘Pseudoseizures or non- Xiao, R. and T. McEnery. 2006. ‘Collocation,
epileptic seizures (NES): 15 synonyms,’ semantic prosody, and near synonymy: A
Journal of Neurology, Neurosurgery and
AN H. KUPPENS
University of Antwerp, Belgium
INTERTEXTUALITY IN ADVERTISING
In a sense, every media text is intertextual (Barthes 1975; Bakhtin 1986;
Fiske 1987). Producers do not create their texts in a vacuum: other media
texts inevitably permeate their work. Similarly, when readers consume
media texts, their decoding is influenced by other texts they are familiar
with. In this view, intertextuality refers to ‘the fundamental and inescapable
interdependence of all textual meaning upon the structures of meaning
proposed by other texts’ (Gray 2005: 3–4).
However, one should distinguish this type of ‘inescapable intertextuality’
(Fiske 1987: 115) from the ‘intertextual intent’ that some texts have. Such
texts ‘aim themselves at other texts and genres, and [. . .] want us to read
them through other texts or genres’ (Gray 2005: 4). In this type of intertex-
tuality, which Hitchon and Jura (1997) have called ‘postmodern intertextual-
ity’, the meaning of a new text is almost entirely dependent on the reader’s
knowledge of the ‘referenced’ text(s). Withalm (2003), for instance, discusses
a number of advertisements that intertextually refer to movies. The meaning
and/or humor of these ads can only be understood and appreciated if viewers
are familiar with movies such as Hitchcock’s classics Psycho (1960) or The Birds
A. H. KUPPENS 119
(1963), or box office hit Titanic (1997). Besides such intergeneric intertextual
references (i.e. references to media genres other than advertising), advertise-
ments can also contain intrageneric references, such as references to other
particular advertisements, or to more general advertising clichés, codes, or
routinized strategies (Cook 2001). In both inter- and intrageneric intertextual-
ity, pastiche, parody, and humor often feature prominently.
The advent of postmodern intertextual advertising in the 1980s attempted
to appeal to a generation of critical, media-literate, and skeptical viewers
(Goldman and Papson 1994; Leiss et al. 2005). ‘After nearly forty years
of watching TV ads, viewers had grown too acclimatized to advertising’s
routinized messages and reading rules. [. . .] Savvy, media-literate viewers
Transcultural intertextuality
The process of cultural and media globalization has dramatically increased
the resources for intertextual references. As Hesmondhalgh (2002: 178)
120 ENGLISH IN ADVERTISING
points out: ‘Cultural texts originated in one country are increasingly seen,
heard, and so on in other countries. Because of this increasing flow of cultural
texts, audiences and symbol creators can, in many places, draw on texts
from many other different places. Texts, genres and even technologies (such
as musical instruments) will be often reinterpreted and adapted by symbol
creators in other contexts.’ Similarly, Regev (2007: 126) considers the ‘world
culture’ that globalization has created as ‘a bank of visual, sonic and textual
stylistic elements and techniques of expression, from which every local pro-
ducer at the national level can draw materials for her own use’. In Appadurai’s
(1996: 35) terminology, mediascapes provide ‘large and complex repertoires
of meanings, narratives, and ethnoscapes to viewers throughout the world’.
visual, auditory, and narrative characteristics of the ads will be taken into
account. This holistic approach is required in order to fully grasp the inter-
textual references made in the advertisements. As advertisers’ possibilities
of making ‘global’ generic intertextual references are greatly affected by
their audiences’ genre literacy, and as this literacy is primarily dependent on
these audiences familiarity with foreign media, the next section will give a
short description of Flemish and Dutch audiences’ access to English-language
media.
(including, but not limited to: American, British, and/or Australian soaps,
drama series, sitcoms, documentaries, reality television, talk shows, whodun-
its, animated series, and science fiction).
When it comes to the access to, and popularity of English-language media,
the situation in the Netherlands is comparable with that in Flanders. However,
in the Netherlands, English is the first foreign language children learn in
school (at about the age of 10 years); French and German are only taught in
secondary education.
of the frog (Figure 1C and D). Meanwhile, the voice-over continues, ‘This
particular specimen hasn’t moved an inch for almost half a day now.’ Next,
we see a long-shot image of the frog jumping onto the road, ending up under-
neath the wheels of a passing car (Figure 1E). The voice-over says, ‘Oh’,
and the advertisement shows the image of a flattened out frog, who
croaks ‘Bollocks’. ‘Well’, a voice-over says, in Dutch, ‘Sometimes it’s better
to stay where you are. With Belgacom, for instance’ (my translation).
Simultaneously, the slogan ‘Sometimes it’s better to stay where you are’
appears on screen in Dutch (Figure 1F). Next, the Belgacom logo is shown.
This advertisement is a textbook example of intergeneric intertextuality, as
it intertextually refers to the genre of British wildlife documentary. Several
Sweden.’ As the Swedish ‘athlete’ starts spinning on his back on the ice
(Figure 2C), the second voice continues, ‘And he’s off! Look at him spin,
burning up that ice . . . Sweep, com’on sweep it up! Sweep!’ The third voice
adds, ‘This really looks good Steve.’ As the Swedish athlete nears the target
(Figure 2D), the second voice says, ‘Look out, India, look out!’ The third voice
yells, ‘And it’s good! Sweden wins. What a game!’ Finally, both sportscasters
(i.e. the ‘second voice’ and ‘third voice’) are shown (Figure 2E), with the left
one chuckling, ‘Now look who’s going home with the gold. But now first a
word from our sponsor.’ While the Studio Brussel logo appears on the screen
(Figure 2F), the sportscaster on the right-hand side—seemingly unaware that
he is still on air—tells his colleague on the left-hand side, ‘I’m freezing my ass
off’; the left sportscaster responds with ‘I can’t feel my toes’.
This ad, too, is an example of intergeneric intertextuality. More specifically,
the advertisement presents a parody of American sports commentary, thereby
adopting several features of the genre. For instance, the advertisement follows
the general structure of a sports event broadcast: Starting with the announce-
ment of the event and an image of the venue (Figure 2A), then switching
to the presentation of the performing team (Figure 2B), followed by the
actual performance (Figure 2C and D), and finishing with the commentators’
wrap-up (Figure 2E and F). The advertisement also adopts the typical ‘dialogic
presentation’, with one announcer who narrates the play-by-play (e.g. ‘Okay,
India leading. Here comes Sweden’, ‘And it’s good! Sweden wins. What
a game’) and one commentator providing the color commentary (e.g. ‘Look
at him spin, burning up that ice . . . Sweep, com’on sweep it up! Sweep!’)
(Ferguson 1983: 156; Hansen 1999). The fact that the two presenters
are actually shown to the viewer further marks the advertisement as an
A. H. KUPPENS 125
American sports event broadcast (as this is rarely the case in Flemish broad-
casts). Examples of visual features that refer to a sports event broadcast are
the event logo (Figure 2A–F), the display with information on the competitors
(Figure 2B), and the sportscasters’ sartorial choices (big headphones, wearing
a warm casual coat over a tuxedo; Figure 2E). Also very significant is the use
of typical elements of the language of sports commentary such as updates
of the course of the game (‘India leading’), the use of specific routines
(e.g. ‘What a game!’, ‘And he’s off!’, ‘And it’s good!’), some syntactic features
of sports commentary (copula deletion in ‘India [is] leading’; the use of and
for starting a new statement (as in ‘And he’s off!’, ‘And it’s good!’)), and the
sportscasters’ rising intonation as the game nears its climax (Ferguson 1983;
sighs (Figure 3H). Finally, white letters against the black background read
(originally in Dutch): ‘Being a teenage mum is not a game. Use contraception’
(Figure 3I) and ‘The pill is now cheaper if you are under 21. Info on laura.be’
(my translation).
This advertisement combines both intrageneric and intergeneric intertex-
tuality. Intergenerically, the commercial appropriates the animation style of
the popular The Sims video games. The setting of the ‘Teenage Mum’ game
(everyday life, ‘real world’ rather than fantasy world), the visual look of
the characters and their environment, the erratic physical movements of the
characters, and the several head-up displays that are shown, are all highly
similar to the The Sims video games. However, the style of the advertisements
for The Sims games is generally not as hyperactive as the Teenage Mum
advertisement. On the contrary, such advertisements adopt a rather ‘laidback’
style, with minimal use of voice-over. The Teenage Mum advertisement,
however, adopts the style of a different type of commercial, namely that
of energetic commercials, in which a voice-over tries to mention as many
of the products’ exciting features as is possible in the constrained duration of
an advertisement. This short, hard sell, informational and rather irritating style
A. H. KUPPENS 127
need to be a punishment. For Cif introduces Cif Power Cream Sprays. The
power of Cif, in a handy spray’ (my translation).
This fourth advertisement can be interpreted as intertextually referring to
the genre of American television and Hollywood drama, and more specifically
to movie scenes of prisoners being visited by family members. Again, several
visual, auditory, and narrative cues are used to temporarily make viewers
believe that they are watching a scene from a Hollywood movie or television
drama, rather than a commercial for cleaning products. An important audi-
tory reference to the genre is the melancholic background music. The inter-
textual nature of the advertisement is further visually enhanced by its
grayish look, by the orange jumpsuit the mother is wearing (typical prison
dress), and by the editing choices that where made (e.g. close-ups of emo-
tion-ridden faces, shot-reverse-shots of the interactants). The most conspicu-
ous intertextual references are, however, the several clichés of Hollywood
scenes of prison visits that are used in the ad. The most prominent
examples of such references are the hands of the mother/prisoner and
the child/visitor touching through the glass (see especially the close-up in
Figure 4B), the mother’s/prisoner’s utterance ‘I gotta get back’ (in movies
generally preceded by a prison ward’s request to leave the visiting room),
and the over-the-top emotions that lie in the facial expressions of both inter-
actants (e.g. Figure 4E) and in utterances such as ‘I love you momma!’ and
‘I love you too, baby!’
A. H. KUPPENS 129
DISCUSSION
In this article, I discussed four television advertisements in which English is
the dominant language, despite the fact that they target Dutch-speaking
audiences. In these advertisements, the choice for English cannot be explained
by the motivations that have thus far been described in the scholarly literature:
the advertisements are not part of an international campaign, and English
is not used for purely creative-linguistic reasons (e.g. for linguistic puns,
to fill a lexical gap), nor for the cultural connotations it potentially reflects
on the product or its potential users. Rather, in these ads, specific varieties and
registers of English seem to be used because of their association with the genre
to which the advertisement intertextually refers. Advertisers thereby draw on
viewers’ familiarity with a range of American and British media genres. This
observation has interesting implications for our views of foreign English in
130 ENGLISH IN ADVERTISING
advertising and of the global spread of English through popular culture more
generally.
First of all, the analysis shows that, when English is used in advertisements,
this does not necessarily entail an uncritical celebration of that language and
the cultural values it is associated with. The current scholarly studies have
demonstrated that English is often used because it is supposed to imbue the
advertised product with the cultural connotations it carries, or because it is
expected to appeal to a specific target audience that values those connotations.
These connotations can be culturally linked to a specific English-speaking
country (e.g. ‘British class’), or they can be more general (e.g. ‘modernity’).
In the four advertisements discussed here, however, while the connotations
a ‘subordinated’ position in the ads: it is used for parodying and mocking the
genres it is associated with, it is used for entertaining the viewer, but when
the ads come to revealing their ‘real’ message, English is quickly disposed of.
Secondly, the advertisements discussed here demand a different, more
active role from their readers than advertisements in which English is used
because of the connotational values it evokes. The latter type of advertisements
assumes an essentially passive role for their readers, as they draw on readers’
more or less subconscious associations between English and certain cultural
connotations or stereotypes. In this sense, consumers are regarded as gullible
and uncritical: it does not really matter what an advertiser says about the
product; as long as he says it in English, consumers will rise to the bait (as
ACKNOWLEDGEMENTS
The author would like to thank Philippe Meers, Annick De Houwer, Jelle Mast and the
anonymous reviewers for their insightful feedback on an earlier version of this article. Many
thanks also to Steven Malliet for his suggestions with respect to gaming jargon, and to
Manoëlle Ballaux (Belgacom), Christ Lannoy and Karel Vinck (Duval Guillaume Antwerp),
Erik Liebe (Unilever Rotterdam), Bruno Ruebens (NVSM) and Geert Van Hoeymissen (VRT) for
their kind permission to publish images from their commercials.
Conflict of interest statement. None declared.
A. H. KUPPENS 133
REFERENCES
Appadurai, A. 1996. Modernity at Large. Cultural Gerritsen, M., H. Korzilius, F. van Meurs, and
Dimensions of Globalization. University of I. Gijsbers. 2000. ‘English in Dutch commer-
Minnesota Press. cials: Not understood and not appreciated,’
Baumgardner, R. J. 2006. ‘The appeal Journal of Advertising Research 40/4: 17–31.
of English in Mexican commerce,’ World Gerritsen, M., C. Nickerson, A. Van Hooft,
Englishes 25/2: 251–66. F. van Meurs, U. Nederstigt, M. Starren,
Bakhtin, M. M. 1986. ‘The Problem of Speech and R. Crijns. 2007. ‘English in product
Genres.’ In C. Emerson and M. Holquist (eds): advertisements in Belgium, France, Germany,
Speech Genres and Other Late Essays. University the Netherlands and Spain,’ World Englishes
of Texas Press, pp. 60–102. 26/3: 291–315.
Barthes, R. 1975. S/Z. Cape. Goldman, R. and S. Papson. 1994. ‘Advertising
Kelly-Holmes, H. 2000. ‘Bier, Parfum, Kaas. and implications,’ Journal of Advertising 21/4:
Language Fetish in European advertising,’ 61–71.
European Journal of Cultural Studies 3/1: 67–82. Nutt, D. and D. Railton. 2003. ‘The Sims: Real
Kelly-Holmes, H. 2005. Advertising as Multilingual life as genre,’ Information, Communication &
Communication. Palgrave Macmillan. Society 6/4: 577–92.
Krishnasamy, K. 2007. ‘English in Tamil: The O’Donohoe, S. 1997a. ‘Leaky boundaries.
language of advertising,’ English Today 23/3–4: Intertextuality and young adult experiences
40–49. of advertising’ In M. Nava, A. Blake, I.
Lee, J. S. 2006. ‘Linguistic constructions of MacRury, and B. Richards (eds): Buy
modernity: English mixing in Korean televi- This Book. Studies in Advertising and
sion commercials,’ Language in Society 35/1: Consumption. Routledge, pp. 257–75.
59–91. O’Donohoe, S. 1997b. ‘Raiding the postmodern
Leiss, W., S. Kline, S. Jhally, and J. Botterill. pantry. Advertising intertextuality and the
A Subject–Object Asymmetry in
the Comprehension of wh-Questions
by Korean Learners of English
JIN-HWA LEE
Chung-Ang University, Seoul, Korea
INTRODUCTION
Wh-questions have been extensively investigated in the second language (L2)
acquisition literature, mostly within the Chomsky’s (1981) Principles and
Parameters framework (Bley-Vroman et al. 1988; Schachter 1989, 1990;
Johnson and Newport 1991; White 1992). Under the assumption that
wh-movement is constrained by Universal Grammar (specifically, Subjacency
(Chomsky 1977)), and that this constraint is instantiated in some languages but
not in others, the central question of the studies was whether L2 learners are
able to access the constraint when their first language (L1) does not instantiate
it. L2 learners’ ability to accept grammatical questions and reject ungrammatical
ones in L2 has been interpreted as evidence for UG availability in L2 acquisition.
More recently, however, a different perspective has been brought to
research on wh-questions. Researchers tried to explain L2 learners’ perfor-
mance in terms of sentence processing or its relationship with the grammar
(Schachter and Yip 1990; Juffs and Harrington 1995; White and Juffs 1998;
Juffs 2005). It was Schachter and Yip who first drew researchers’ attention to
the processing of wh-movement. They found that Chinese and Korean learners
J.-H. LEE 137
of English, and even native speakers of English had more difficulty in accepting
grammatical wh-questions when the subject of an embedded clause was
extracted as in (1) than when the object was extracted as in (2).
(1) Whoi did Cathy believe ei met her friend? (subject extraction)
(2) Whoi did Cathy believe Tom met ei? (object extraction)
grammatical object extraction. While their mean scores for object wh-questions
ranged from 5.1 to 6.9 out of 9 depending on the number of clauses
over which the wh-word had been extracted, their mean scores for subject
wh-questions ranged from 2.8 to 5.1. More interestingly, native English
speakers showed a similar pattern.
Juffs and Harrington (1995) not only replicated Schachter and Yip’s (1990)
findings but also observed that advanced adult Chinese L2 learners of English
were slower in judging subject extraction than object extraction from infiniti-
val clauses. In addition, using a moving window technique which measures
word-by-word reading time, they found a greater slowdown at the point
where the learners encountered the subject extraction site than the object
Schachter and Yip (1990) attributed the preference for object extraction over
subject extraction to the general nature of left-to-right processing in human
sentence processing and the Minimal Attachment principle (Frazier and Fodor
1978). Minimal Attachment states that the parser attempts to integrate an
incoming item into the currently constructed phrase structure using the min-
imal number of syntactic nodes. To apply the Minimal Attachment principle
to the sentences above, the parser initially posits an empty category (linked to
the wh-word) as the object of believe for both sentences because this leads to
a simpler structure than projecting an embedded clause as its complement.
J.-H. LEE 139
As parsing proceeds in sentence (1), the verb met calls for a subject, which
leads the parser to ‘backtrack’ to posit a subject gap that is linked to the
wh-word that had previously been associated with the potential object gap
in the matrix clause. According to Schachter and Yip, it is this backtracking
that makes subject extraction more difficult to process because it is against
the left-to-right nature of human processing and therefore demands an
extra processing effort. On the other hand, no backtracking occurs in object
extraction in (2). When the parser encounters Tom, it only has to embed the
NP node inside the following clause. As parsing proceeds, the verb met calls for
an object, and the initial object gap in the matrix clause falls in the postverbal
position. This process goes from left to right without any backtracking.
The parser initially interprets who as the object of believe and thus posits
a matrix object trace in both sentences. However, as it encounters met in
the subject wh-question in (1), the initial matrix object trace must be reana-
lyzed as a subject trace in the embedded clause. This reanalysis gives rise
to concomitant changes in theta role assigner (believe ! met), theta role
(internal ! external), and Case assigned (accusative ! nominative). On the
other hand, in the object wh-question in (2), the parser has to reanalyze
the trace as an embedded object rather than a matrix object. This process
requires a change only in theta role assigner (believe ! met). The type of
theta role and Case that are assigned remains the same. According to Juffs
and Harrington (1995), since subject extraction involves more changes in
theta role and Case, it is harder to parse than object extraction for not only
L2 learners but also native speakers of English.
The review of the aforementioned studies suggests that subject wh-questions
are more difficult to process than object wh-questions and that the locus of
the difficulty is the extraction site. However, this conclusion is perhaps pre-
mature since the previous studies in particular suffer from nontrivial design
flaws in terms of test sentences. For instance, Schachter and Yip (1990) used
different verbs and lexical items for subject and object questions (e.g. Who
did you say John suspects fell in love with Sue? versus Who did the President say
he thinks he’ll appoint as ambassador?), thereby leaving open the possibility
that lexical knowledge plays a role in the observed asymmetry pattern. The
animacy of wh-words and NPs is another hidden factor which may have
affected the result in the study. L1 studies suggest that animate NPs are
more likely to be associated with an agent or subject, whereas inanimate
ones are more likely to be associated with the object (Ervin-Tripp 1970;
Tyack and Ingram 1977; Traxler et al. 2002). In Schachter and Yip’s study,
two-thirds of stimuli included inanimate or at least nonhuman wh-words
140 SUBJECT–OBJECT ASYMMETRY IN L2
(e.g. what or which), which are more favorable to object extractions. Therefore,
it is possible that L2 learners’ preference of object wh-questions to subject wh-
questions in the study reflects the effect of animacy rather than the effect of
syntactic configuration.
Unlike Schachter and Yip (1990), Juffs and Harrington (1995) matched
lexical items, but their stimuli are still biased either against subject
wh-questions or in favor of object wh-questions for other reasons. Referential
properties of NPs are one of the concerns. According to Warren and Gibson
(2002), the less accessible the referent of an NP is in the discourse, the more
resources are required to find or construct it in sentence processing. In this
regard, they argue that pronouns whose referents are already present in
Method
The experimental study made use of an elicited comprehension task.
Participants were required to listen to recorded questions and choose answers,
based on accompanying pictures. This task was chosen for two reasons: to
explore whether the asymmetry previously observed in grammaticality judge-
ment tasks is replicable in other types of tasks, and to make the present study
Items were randomized so that the subject and object questions in a pair
did not occur consecutively. The questions were recorded by a female native
English speaker with a pause of 8 s between questions.
Procedure
Data from 9th graders, 11th graders, and college students were collected
in Seoul, Korea during regular English classes. For the 11th graders and the
college group, the researcher visited their classes and collected data with the
help of the teachers of the classes. In the case of the 9th graders, the English
teacher alone conducted the data collection. This difference should not have
had a significant influence on the results because the role of data-gatherer
in this study was limited to management of the session, e.g. distributing test
materials and playing the previously recorded tape.
Each student was given a booklet of pictures and an answer sheet. When
the teacher or the researcher played the tape, students were required to listen
to each wh-question and choose an answer based on an accompanying
picture. Each sentence was repeated twice. The task took approximately
10 min to complete. After the test, the students were given small gifts in appre-
ciation for their participation.
The graduate students were tested in Hawaii, either individually or in a small
group. The procedure was identical to the one that was employed for the EFL
groups except that the graduate group listened to each sentence only once and
J.-H. LEE 145
RESULTS
Subject–object asymmetry in groups
The scores for the four groups in the form of mean correct responses to subject
and object questions are presented in Table 2.
In general, the participants performed better on subject wh-questions than
Table 2: Mean accuracy scores for subject and object wh-questions by group
Subject wh-Q (k = 5) Object wh-Q (k = 5)
p <.05
146 SUBJECT–OBJECT ASYMMETRY IN L2
high effect sizes (2 = .379 for Group, 2 = .371 for Question Type). However,
the interaction of Group and Question Type was also significant
(F (3, 107) = 5.129, p < .05). This means that the effect of Question Type
was moderated by that of Group, as illustrated in Figure 2.
While all the groups were more accurate on subject wh-questions
than on object wh-questions, the difference was greater for 11th graders and
college students than the other two groups.
Since a significant Group effect was identified, two separate one-way
ANOVAs (Analysis of Variance) and post hoc analyses were conducted on
each of the comprehension of subject and object questions in order to locate
the differences among the four groups. Again, the Group effect was significant
9th
3
11th
College
2
Graduate
0
Su b -Q Ob j-Q
DISCUSSION
This study has addressed the question of whether the asymmetry previously
observed in judgements of wh-question grammaticality is replicable in the L2
comprehension of English long-distance wh-questions by Korean learners of
English. Contrary to the previous finding, the results of the present study clearly
showed that subject extraction is easier to comprehend than object extraction.
Evidence for the relative ease of subject extraction was threefold. First, within-
group comparisons of mean accuracy scores for subject and object questions
revealed that the learners were more accurate on subject extraction than on
object extraction. Second, the majority of individual learners obtained higher
scores on subject questions than on object questions, while only a few learners
showed the opposite pattern. Finally, the learners consistently made far more
reversal errors in object wh-questions than in subject wh-questions, indicating
a strong tendency to interpret object wh-questions as subject wh-questions.
This study also showed that this asymmetry in favor of subject questions
was not restricted to advanced learners. The learners at earlier stages of English
acquisition also showed the better performance on subject wh-questions
than on object wh-questions, indicating that they all rely on the same sentence
processing strategies.
Overall, the learners’ performance improved along with the number of years
of English instruction, but at different stages depending on the question type.
While a significant change occurred on subject wh-questions in 11th graders,
a significant change on object wh-questions was observed only in the graduate
group. This suggests that subject wh-questions are easier to improve on than
are object wh-questions, adding another piece of evidence for the relative ease
of subject wh-questions.
Then, why do L2 learners find long-distance subject wh-questions easier to
process than object question counterparts? Given that Korean has no
wh-movement, at least at the surface level, there is no reason to expect
Korean-speaking learners of English to prefer one type of extraction pattern
to the other in English. Besides, since the same pattern of asymmetry was
150 SUBJECT–OBJECT ASYMMETRY IN L2
observed in all the groups, despite differences in the length of instruction and
the type of learning environment (i.e. EFL versus ESL), the preference for the
subject pattern can apparently not be attributed to instruction effects.
Another possibility worth explaining is that the learners in this study have
been exposed to more subject questions than object questions. No corpus
data are available specifically for L2 learning, but a possible indication of the
general practice of native English speakers comes from Stromswold’s (1995)
study of mono-clausal wh-questions in parental speech to children. Based on
an analysis of the CHILDES transcripts of 12 children, Stromswold reports
that adult English speakers asked more subject questions (M = 63.1 per cent,
SD = 21.4 per cent) than object questions. Interestingly, however, there was
CONCLUSION
ACKNOWLEDGEMENTS
I am grateful to William O’Grady and three anonymous reviewers for their comments and
suggestions on this paper. My thanks also go to the participants and to their teachers for
agreeing to participate in this experiment.
NOTES
1 While it can be argued that two difficulty in answering object
adjacent tensed verbs in subject wh-questions.
wh-questions also may serve as a cue 3 There were no fillers in the task. As
to the extraction site, their sentence- one reviewer points out, it is possible
medial position is less salient than the that the participants used metalinguis-
REFERENCES
Baek, S. 2007. ‘Is subject extraction more Behavioral Determinants of Language
difficult than object extraction for second lan- Development. Erlbaum, pp. 321–37.
guage learners of English?’ Unpublished Crain, S. and R. Thornton. 1998. Investigating
manuscript, University of Illinois, Urbana- Universal Grammar: A Guide to Experiments on
Champaign. the Acquisition of Syntax and Semantics. MIT
Bley-Vroman, R., S. Felix, and G. Ioup. 1988. Press.
‘The accessibility of Universal Grammar in Ervin-Tripp, S. 1970. ‘Discourse agreement:
adult language learning,’ Second Language How children answer questions’ in J. Hayes
Research 4: 1–32. (ed.): Cognition and the Development of
Carpenter, P., M. Miyake, and M. Just. 1994. Language. Wiley, pp. 79–107.
‘Working memory constraints in com- Frazier, L. 1998. ‘Getting there (slowly),’ Journal
prehension: Evidence from individual differ- of Psycholinguistic Research 27: 123–46.
ences, aphasia, and aging’ in M. Gernsbacher Frazier, L. and J. D. Fodor. 1978. ‘The sausage
(ed.): Handbook of Psycholinguistics. Academic machine: A new two-stage parsing model,’
Press, pp. 1075–122. Cognition 6: 291–325.
Chomsky, N. 1977. ‘On wh-movement’ Gibson, E. 2000. ‘The dependency locality
in P. Culicover, T. Wasow, and A. Akmajian theory: A distance-based theory of linguistic
(eds): Formal Syntax. Academic Press, complexity’ in Y. Miyashita, A. Marantz, and
pp. 71–132. W. O’Neil (eds): Image, Language, Brain. MIT
Chomsky, N. 1981. Lectures on Government and Press, pp. 95–126.
Binding. Foris. Grodner, D. and E. Gibson. 2005.
Collins, C. 1994. ‘Economy of derivation and ‘Consequences of the serial nature of linguis-
the generalized proper binding condition,’ tics input for sentential complexity,’ Cognitive
Linguistic Inquiry 25: 45–61. Science 29: 261–90.
Crain, S. and R. Thornton. 1991. ‘Recharting Hawkins, R. 1999. ‘Processing complexity and
the course of language acquisition: Studies in filler-gap dependencies across grammars,’
elicited production’ in N. Krasnegor, Language 75: 244–85.
D. Rumbaugh, R. Schiefelbusch, and Johnson, J. and E. Newport. 1991. ‘Critical
M. Studdert-Kennedy (eds): Biological and period effects on universal properties of
J.-H. LEE 155
language: The status of subjacency in the Schachter, J. and V. Yip. 1990. ‘Grammaticality
acquisition of English as a second language,’ judgments: Why does anyone object to subject
Cognition 39: 215–58. extraction?,’ Studies in Second Language
Juffs, A. 2005. ‘The influence of first language Acquisition 12: 379–92.
on the processing of wh-movement in English Seidl, A. and G. Hollich. 2002. ‘Infants’ and
as a second language,’ Second Language Research toddlers’ comprehension of subject and object
21/2: 121–51. wh-questions,’ BULCD Proceedings 26: 596–607.
Juffs, A. and M. Harrington. 1995. ‘Parsing Stromswold, K. 1995. ‘The acquisition of sub-
effects in second language processing: Subject ject and object wh-questions,’ Language
and object asymmetries in wh-extraction,’ Acquisition 4: 5–48.
Studies in Second Language Acquisition 17: Tyack, D. and D. Ingram. 1977. ‘Children’s pro-
483–516. duction and comprehension of questions,’
Kim, S. 1999. ‘The subject-object asymmetry Journal of Child Language 4: 211–224.
in the acquisition of wh-questions by Korean
REVIEWS
literacy, including cultural studies and new literacy studies. To address the
inequity that results from these invisible pedagogies, they propose the use of
functional linguistics as a ‘meta-language’ with which to compare different
approaches to subject English, and with which to construct alternative curri-
cula. Wignell’s chapter presents an analysis of the development of social
science as a vertical discourse through examination of texts by Hobbes,
Smith, Ricardo and Marx. In her chapter, O’Halloran presents a similar ana-
lysis of the development of mathematical and scientific discourse, highlighting
the different grammars of symbols and images and the process through which
these grammars, combined with that of natural language, combine to create
specialist forms of meaning. She brings her analysis up to date with a discus-
dictionaries’ (p. 2), and one that is still widely employed in contemporary
national and/or post-colonial contexts. In the process, the much more complex
and fluid reality of multilingual speakers ‘on the ground’—heteroglossia, in a
Bakhtinian sense—is overlooked, even actively disguised/distorted.
Situating the construction/invention of languages within this wider political
context points to the second key concern of Makoni and Pennycook’s volume:
the very real material, often highly deleterious, consequences of this process of
language invention for individual multilingual speakers. Returning to
Bourdieu, such processes are clearly not a game devoid of consequences.
Rather, the construction of discrete, standardized languages, and related lan-
guage hierarchies, actively delimits the choices of speakers and undermines
Pennycook examines the myths that pervade the role and influence of (stan-
dard) English as the current ‘international language’. Rather than achieving
social and economic mobility via learning English, as is often claimed: ‘[t]his
thing called English colludes with many of the pernicious processes of globa-
lisation, deludes many learners through the false promises it holds out for
social and material gain, and excludes many people by operating as an exclu-
sionary class dialect, favouring particular people, countries, cultures and forms
of knowledge’ (p. 101).
A similar scepticism is apparent in the chapters by Makoni and Mashiri, on
Africa, and de Souza, on Brazil, although their focus is primarily on so-called
indigenous languages. Both argue that the construction of standardized indig-
REFERENCES
REFERENCES
The book is divided into four sections. The first gives a general characteriza-
tion of formulaic language and summarizes the main claims from Wray (2002).
It also includes a review of the literature on the psycholinguistic processing of
formulae, a chapter discussing the role of formulaicity in oral epic poetry and
in the historical transition to literacy, and another discussing the differences
between speech and writing in terms of their use of formulaic language. These
discussions lead to a model of the ways in which the situational autonomy of a
text influences its level of formulaicity. The book’s second section starts by
setting out desiderata for a theory of formulaic language. Wray argues that a
comprehensive model needs to encompass and show the relationships
between three facets of formulaicity: description of what patterns exist; sys-
Ruth Wodak, Rudolf de Cillia, Martin Reisigl and Karin Liebhart: THE
DISCURSIVE CONSTRUCTION OF NATIONAL IDENTITY. SECOND
EDITION. Transl. by Angelika Hirsch, Richard Mitten and J. W. Unger.
Edinburgh University Press, 2009.
Ruth Wodak, Rudolf de Cillia, Martin Reisigl and Karin Liebhart: THE
DISCURSIVE CONSTRUCTION OF NATIONAL IDENTITY. SECOND
EDITION. Transl. by Angelika Hirsch, Richard Mitten and J. W. Unger.
Edinburgh University Press, 2009.
2005. And finally an extended and revised 2009 second edition shows how
important this book was, is and almost certainly will remain for academics and
students of many disciplines within the broadly defined social sciences.
It is after all a book that inspires many. Its ever interesting topic of nation
and identity, focus on sameness and uniqueness in contrast to other work
within this field which prefers to concentrate on otherness, its stimulating
deployment of other important concepts from related disciplines, such as ‘ima-
gined community’ and habitus, to the study of discourse makes this book
widely appreciated and well regarded. By the same token, vigorous method-
ology of a discourse-historical approach developed in the Vienna School of
Discourse Analysis, immaculate examination of multi-genred data (i.e. com-
attention to political processes that influence and shape the discourses under
investigation. This seems to be slightly narrowing the method of ‘revisiting’ the
field. After all, the Vienna School sees discourse as inter alia ‘integrating var-
ious different positions and voices’ (Wodak 2009: 39); they postulate the need
to look closely into different contexts, one of which is ‘the extra-linguistic
social/sociological variable’ (Wodak 2009: 38). One can wonder then why
other types of data were not taken into consideration. I think here mainly of
the broad range of data from semi-public/semi-private settings such as internet
discussions. This is in my opinion an extremely interesting and broadening
source of texts which may reveal interesting aspects of the researched issue of
national identity. It is the expansion of Internet which has already transformed
REFERENCES
Burawoy, M. 2003. ‘Revisits: an outline of a Weiss, G. and R. Wodak (eds). 2003. Critical
theory of reflexive ethnography,’ American Discourse Analysis. Theory and Interdisciplinarity.
Sociological Review 68/5: 645–79. Palgrave Macmillan.
Burawoy, M. 2009. The Extended Case Wodak, R. 2009. The Discourse of Politics in Action.
MethodUniversity of California Press. Politics as Usual. Palgrave Macmillan.
Heer, H., W. Manoschek, A. Pollak and Wodak, R. and M. Krzy_z anowski (eds). 2008.
R. Wodak (eds). 2008. The Discursive Construction Qualitative Discourse Analysis in the Social Sciences.
of History: Remembering the Wehrmacht’s War of Palgrave Macmillan.
Annihilation. Palgrave Macmillan. Wodak, R. and M. Meyer (eds). 2001 Methods
Martin, J. R. and R. Wodak (eds). 2003. of Critical Discourse Analysis. Sage.
Re/Reading the Past: Critical and Functional
Perspectives on Time and Value. John Benjamins.
Applied Linguistics 31/1: 169–171 ß Oxford University Press 2010
doi:10.1093/applin/amq002
NOTES ON CONTRIBUTORS
Jin-Hwa Lee earned a PhD in second language acquisition from the University
of Hawai’i in 2006 and is currently working in the Department of English
Education at Chung-Ang University, Seoul, Korea. Her main research interests
include second language acquisition and instruction, corpus-based L2 analysis
and teaching, task-based language teaching, and L2 programme development.
Address for correspondence: Department of English Education, 221 Heukseok-
dong, Dongjak-gu, Seoul 156-756, Korea. <jinhlee@cau.ac.kr>