APPLIN-30(1)Cover.qxd
3/10/09
11:47 AM
Page 1
APPLI ED LI NGUISTICS
Applied Linguistics
Volume 30 Number 1 March 2009
Volume 30 Number 1 March 2009
ISSN 0142-6001
Published in cooperation with AAAL American Association for Applied Linguistics AILA International Association of Applied Linguistics BAAL British Association for Applied Linguistics
OXFORD
www.applij.oxfordjournals.org
APPLIN-30(1)Cover.qxd
3/10/09
11:47 AM
Page 2
Applied Linguistics Journal online The full text of Applied Linguistics is available online to journal subscribers. Online access has a number of advantages: . quality PDFs ensure articles look the same as the print original and are easy to print out . access is easy—all you need is your subscription number or institutional IP address (see below) . online access is available ahead of print publication—so view while you await your print version! . access the text wherever you are (or from any part of your institution network if you have a library subscription) . perform searches by word or author across the full text of the articles of any part of the journal . download articles whenever you choose—you will be able to access past online issues as long as you have a current subscription . free sample copy available online . fully searchable abstracts/titles going back to volume 1 . Table of Contents email alerting service. The print version will continue to be available as previously. Institutions may choose to subscribe to the print edition only, online only, or both. Individual subscribers automatically receive both.
CONTRIBUTORS There is no need for contributors to format their articles any differently; online files are produced automatically from the final page proofs of the journal. However, if you know that an item in your list of references is available online, please supply the URL. If you have your own website, you are welcome to include the URL with your contact address in your biodata.
ADVANCE ACCESS Applied Linguistics now has Advance Access articles. These are papers that have been copyedited and typeset but not yet paginated for inclusion in an issue of the journal. More information, including how to cite Advance Access papers, can be found online at http://www.applij.oxfordjournals.org.
Applied Linguistics Subscription Information
A subscription to Applied Linguistics comprises 4 issues. Annual Subscription Rate (Volume 30, 4 issues, 2009) Institutional. Print edition and site-wide online access: £203.00/US$365.00/E305.00; Print edition only: £193.00/US$347.00/E290.00; Site-wide online access only: £193.00/US$347.00/E290.00. Personal. Print edition and individual online access: £68.00/US$136.00/E102.00. Please note: US$ rate applies to US & Canada, Euros applies to Europe, UK£ applies to UK and Rest of World.
Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, UK. Email:
[email protected]. Tel (and answerphone outside normal working hours): +44 (0)1865 353907. Fax: + 44 (0)1865 353485. In Japan, please contact: Journals Customer Services, Oxford Journals, Oxford University Press, Tokyo, 4-5-10-8F Shiba, Minato-ku, Tokyo 108-8386, Japan. Tel: þ81 3 5444 5858. Fax: þ81 3 3454 2929.
subscribe to applied linguistics For new subscriptions and recent single issues only. Current subscribers will automatically receive a renewal form.
Prices include postage by surface mail, or for subscribers in the USA and Canada by airfreight, or in India, Japan, Australia and New Zealand, by Air Speeded Post. Airmail rates are available on request. There are other subscription rates available for members of AAAL, BAAL, AILA, and LSA, for a complete listing please visit www.applij.oxfordjournals.org/subscriptions.
Please complete the form below and return it to: Journal Customer Service Department (please see above).
Full prepayment, in the correct currency, is required for all orders. Orders are regarded as firm and payments are not refundable. Subscriptions are accepted and entered on a complete volume basis. Claims cannot be considered more than FOUR months after publication or date of order, whichever is later. All subscriptions in Canada are subject to GST. Subscriptions in the EU may be subject to European VAT. If registered, please supply details to avoid unnecessary charges. For subscriptions that include online versions, a proportion of the subscription price may be subject to UK VAT. Personal rate subscriptions are only available if payment is made by personal cheque or credit card and delivery is to a private address.
Please send me the following single issue(s) Volume_________ Issue_________
The current year and two previous years’ print issues are available from Oxford University Press. Previous volumes can be obtained from the Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Email:
[email protected]. Tel: +1 (518) 537 4700. Fax: +1 (518) 537 5899. For further information, please contact: Journals Customer Service Department,
Please record my subscription to Applied Linguistics, starting with Volume__________ (Subscriptions start with the March issue and can be accepted for complete volumes only.)
Name (BLOCK CAPITALS please) _________________________________________ Adresss __________________________________ _________________________________________ _________________________________________ City _____________________________________ Country _________________________________ Postcode _________________________________ I enclose the correct payment of (see rates above): £/US/E __________________________________ Please debit my credit card: American Express / Mastercard / Visa (delete as appropriate) Card number: __|__|__|__|__|__|__|__|__|__|__|__|__|__|__|__|__ Expiry date: |__|__|__| Signature ________________________________ œ Please tick this box if you do NOT wish to receive details of related products and services of OUP and other companies that we think may be of interest.
Aims Applied Linguistics publishes research into language with relevance to real-world problems. The journal is keen to help make connections between fields, theories, research methods, and scholarly discourses, and welcomes contributions which critically reflect on current practices in applied linguistic research. It promotes scholarly and scientific discussion of issues that unite or divide scholars in applied linguistics. It is less interested in the ad hoc solution of particular problems and more interested in the handling of problems in a principled way by reference to theoretical studies. Applied linguistics is viewed not only as the relation between theory and practice, but also as the study of language and language-related problems in specific situations in which people use and learn languages. Within this framework the journal welcomes contributions in such areas of current enquiry as: bilingualism and multilingualism; computer-mediated communication; conversation analysis; deaf linguistics; discourse analysis and pragmatics; corpus linguistics; critical discourse analysis; first and additional language learning, teaching, and use; forensic linguistics; language assessment; language planning and policies; language for special purposes; lexicography; literacies; multimodal communication; rhetoric and stylistics; and translation. The journal welcomes both reports of original research and conceptual articles. The Journal’s Forum section is intended to enhance debate between authors and the wider community of applied linguists (see Editorial in 22/1) and affords a quicker turnaround time for short pieces. Forum pieces are typically responses to a published article, a shorter research note or report, or a commentary on research issues or professional practices. The Journal also contains a Reviews section. Applied Linguistics is covered by the following abstracting/indexing services: Bibliographie Linguistique/Linguistic Bibliography, BLonline, British Education Index, Current Index to Journals in Education, ERIC (Education Resources Information Centre), International Bibliography of the Social Sciences, ISI: Social Sciences Citation Index, Research Alert, Current Contents/Social and Behavioral Sciences, Social Scisearch, Sociological Abstracts: Language and Linguistics Behaviour Abstracts, Language Teaching, MLA Directory of Periodicals, MLA International Bibliography, PsycINFO, Sociological Abstracts, Zeitschrift fu¨r Germanistische Linguistik.
ADVERTISING Inquiries about advertising should be sent to Linda Hann, Oxford Journals Advertising, 60 Upper Broadmoor Road, Crowthorne, RG45 7DE, UK. Email:
[email protected]. Tel/Fax: +44 (0)1344 779945.
PERMISSIONS For information on how to request permissions to reproduce articles/information from this journal, please visit www.oxfordjournals.org/permissions.
DISCLAIMER Statements of fact and opinion in the articles in Applied Linguistics are those of the respective authors and contributors and not of Applied Linguistics or Oxford University Press. Neither Oxford University Press nor Applied Linguistics make any representation, express or implied, in respect of the accuracy of the material in this journal and cannot accept any legal responsibility or liability for any errors or omissions that may be made. The reader should make his/her own evaluation as to the appropriateness or otherwise of any experimental technique described.
APPLIED LINGUISTICS Volume 30 Number 1 March 2009 CONTENTS Articles The Effects of Input-Based Tasks on the Development of Learners’ Pragmatic Proficiency MASAHIRO TAKIMOTO Soliciting Teacher Attention in an L2 Classroom: Affect Displays, Classroom Artefacts, and Embodied Action ASTA CEKAITE ‘Lego my keego!’: An Analysis of Language Play in a Beginning Japanese as a Foreign Language Classroom CADE BUSHNELL A Novel Approach to Creating Disambiguated Multilingual Dictionaries IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA and CAROLINA GALLARDO
1
26
49
70
Interpreting Inexplicit Language during Courtroom Examination JIEUN LEE Memorial Article: John Sinclair (1933–2007) The Search for Units of Meaning: Sinclair on Empirical Semantics MICHAEL STUBBS
115
FORUM Ideology in Applied Linguistics for Language Teaching ALAN WATERS
138
REVIEWS Nikolas Coupland: Style: Language Variation and Identity GEOFF HALL Malcolm Coulthard and Alison Johnson: An Introduction to Forensic Linguistics: Language in Evidence FRANCES ROCK D. Block: Second Language Identities SAMANTHA NG
93
144
147 150
Alastair McLauchlan: The Negative L2 Climate: Understanding Attrition Among Second Language Students JIM COLEMAN
154
ERRATUM
158
NOTES ON CONTRIBUTORS
159
Applied Linguistics 30/1: 1–25 doi:10.1093/applin/amm049
ß Oxford University Press 2007
Advance Access published on 12 December 2007
The Effects of Input-Based Tasks on the Development of Learners’ Pragmatic Proficiency MASAHIRO TAKIMOTO The present study evaluates the relative effectiveness of three types of input-based approaches for teaching English polite request forms to sixty Japanese learners of English: (a) structured input tasks with explicit information; (b) problem-solving tasks; and (c) structured input tasks without explicit information. Treatment group performance was compared with control group performance on pre-tests, post-tests, and follow-up tests consisting of a discourse completion test, a role-play test, a listening test, and an acceptability judgement test. The results revealed that the three treatment groups performed significantly better than the control group. However, the group that received the structured input tasks with explicit information did not maintain the positive effects of the treatment between the post-test and follow-up test on the listening test component.
INTRODUCTION Among the issues in second language (L2) pragmatics, a fundamental question is which instructional approaches are most effective for teaching L2 pragmatics. Most studies that compare the effectiveness of different teaching approaches select two types of awareness-oriented instruction, explicit and implicit instruction. Kasper and Rose (2002) argued that some form of awareness-oriented instruction is necessary because pragmalinguistic forms and sociopragmatic rules are often not salient enough for learners and that mere exposure to these rules in action does not help learners notice them. Recent studies within the awareness-oriented instruction framework have lent empirical support to this claim (e.g. Alco´n 2005; Koike and Pearson 2005; Martı´nez-Flor and Fukuya 2005). Alco´n (2005) reported that learners’ awareness of a target pragmatic feature, request strategies, benefited from both types of awareness-oriented instruction. Similarly, findings in Koike and Pearson (2005) indicated that treatment groups with these types of awareness-oriented instruction became aware of not only a greater number of options for expressing suggestions but also the necessity for performing pragmatic mitigation more quickly than the control group. Finally, Martinez-Flor and Fukuya (2005) showed that groups exposed to both types of awareness-oriented instruction improved in their production of pragmatically appropriate suggestions. Two key issues here are the extent to which it is possible to provide learners with opportunities for developing
2
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
their pragmatic proficiency and the level of awareness learners should develop to improve their pragmatic proficiency.
AWARENESS-ORIENTED INPUT-BASED INSTRUCTION The present study bases its definition of awareness-oriented instruction, explicit/implicit instruction, on Jeon and Kaya (2006) which described instruction as a continuum between the absolutely explicit and the absolutely implicit extremes. The extent to which the target of the instruction is made overt to the learners determines the positioning of that kind of instruction between explicit and implicit poles. To test claims about the role of awareness and attention in L2 learning, researchers have conducted a number of interventional studies on grammar and pragmatics teaching. Norris and Ortega’s (2001) meta-analysis of instructional studies, with an emphasis on morphosyntax research, suggested the superiority of explicit instruction (e.g. Fotos and Ellis 1991) over implicit instruction (e.g. DeKeyser 1995; Robinson 1996). Jeon and Kaya’s (2006) meta-analysis on the role of instruction in the development of L2 pragmatics also indicated that explicit instruction is more effective than implicit instruction (for explicit instruction see Lyster 1994; Witten 2000; for implicit instruction see Fukuya and Zhang 2002). However, Jeon and Kaya (2006) noted that due to limited available data, the seemingly superior effects of explicit pragmatic instruction should not be taken as definitive but should be examined in greater detail in future studies. In the existing literature on teaching pragmatics, some interventional studies have shown that pragmatic features can be taught explicitly together with input enhancement activities whereas others have shown that pragmatic features can be taught implicitly with input enhancement activities (for explicit instruction see House 1996; Tateyama et al. 1997; Rose and Ng 2001; Takahashi 2001; for implicit instruction see Takahashi 2005). A review of interventional studies on input-based teaching of L2 pragmatics reveals that the studies were largely motivated by the theories and frameworks built for grammar learning. Ellis (2003) explained that two types of input-based approaches, structured input and consciousness-raising, can be best used in teaching grammar. The adaptability of both approaches to the teaching of L2 pragmatics will be examined thoroughly in the present study.
Structured input Research on the development of grammatical ability offers L2 pragmatics ways of conceptualizing and implementing different instructional approaches, in particular structured input tasks.
MASAHIRO TAKIMOTO
3
The structured input task occupies an important part in processing instruction. According to VanPatten (1996), processing instruction entails three basic features: 1 an explanation of the relationship between a given form and the meaning it can convey; 2 information about processing strategies, showing learners how natural processing strategies may not work to their benefit; and 3 structured input activities in which learners are given the opportunity to process form in the input in a controlled situation so that better formmeaning connections might happen compared with what might happen in less controlled situations (VanPatten 1996: 60). A number of grammar teaching studies have provided empirical evidence that structural input tasks themselves, without explicit information, are effective in improving learners’ grammatical proficiency level (e.g. VanPatten and Oikennon 1996). Ellis (1997) argued that structured input texts must be designed in such a way that the target forms are frequent, the meanings of the target forms are clear, and comprehending the target forms is essential for comprehending the whole text. Although Ellis described two ways of structuring input, input flooding and interpretation tasks, only the latter is of direct relevance to the present study. The following are general principles for the design of interpretation tasks (Ellis 1997). 1 An interpretation activity consists of a stimulus to which learners must make some kind of response. 2 The stimulus can take the form of spoken or written input. 3 The response can take various forms, such as true/false, check a box, select the correct picture, draw a diagram, perform an action, but in each case, the response will be either completely non-verbal or minimally verbal. 4 The activities in the task can be helpfully sequenced to require first attention to meaning, then noticing the form and function of the grammatical structure, and finally error identification. 5 Interpretation tasks should require learners to make a personal response, that is, relate the input to their own lives, as well as a referential response (Ellis 1997: 155–9). The present study draws on the interpretation framework proposed by Ellis (1997) and employs interpretation tasks designed to engage learners in intentional and conscious learning of target pragmatic expressions based on exemplars in the input. A key issue is exploring how such interpretation tasks can be extended to the teaching of L2 pragmatics. To teach pragmatics, instruction must promote learners’ conscious noticing of both the relationship between forms and meanings of target structures and the relationship between strategies for realizing speech intentions, linguistic forms used to
4
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
express these intentions, and social conditions governing language use. Knowledge of the strategies for realizing speech intentions and linguistic items used to express these intentions is called knowledge of pragmalinguistics, whereas knowledge of the social conditions governing language use is called knowledge of sociopragmatics (Leech 1983; Thomas 1983). Therefore, the fourth general principle above needs to be revised so that activities in the task are sequenced to first require attention to sociopragmatic features, then noticing of pragmalinguistic features of target structures, and finally, aiding the learner in error identification.
Problem-solving Another type of input-based approach is consciousness-raising. In the present study the term problem-solving is used in place of the term consciousness-raising. The goal of both problem-solving and structured input tasks is to make better form–meaning connections with different degrees of overtness. Whereas problem-solving tasks lead to more overt instruction, structured input tasks lead to less overt instruction. Furthermore, the content of the problem-solving tasks is the language used in the tasks, while the structured input tasks are built around the content of general topics other than language used in the tasks, such as stories, pictures of objects, or opinions. Empirical evidence from a number of grammar teaching studies has shown that problem-solving tasks are effective in improving learners’ grammatical proficiency level (e.g. Fotos 1994; Fotos and Ellis 1991). Thus, the rationale for using problem-solving tasks in the present study is to compare a more overt way of raising awareness of pragmalinguisticsociopragmatic (problem-solving tasks) with a less overt way (structured input tasks). Ellis (2003) argued that three types of operations, identification of the target structure, judgement of appropriateness of the target structure, and rule provision of the target structure, can be embedded in the problem-solving tasks by: (a) isolating a specific linguistic feature; (b) providing data to illustrate the target feature; (c) encouraging learners to use their intellectual resources to understand the target feature; and (d) requiring learners to verbalize rules describing the grammatical structure in question. Again, the important question here is how the problem-solving approach can be extended to the teaching of L2 pragmatics. As with the structured input task, the problem-solving task needs to be revised for pragmatic teaching purposes. Important revisions should include isolating specific pragmalinguistic features, providing learners with data for the target pragmatic features, engaging learners’ intellectual efforts, and requiring students to understand and verbalize the pragmalinguistic and sociopragmatic features of the target structures.
MASAHIRO TAKIMOTO
5
INPUT-BASED INTERVENTION STUDIES OF L2 PRAGMATICS As noted in the previous section, intervention studies on L2 pragmatics have employed input-based approaches in teaching pragmatics with explicit and implicit instruction. Approaches on the explicit end of the intervention continuum include studies of teacher-fronted explicit explanation treating awareness-raising as a method to promote better pragmatic ability (House 1996; Tateyama et al. 1997; Rose and Ng 2001) and studies of teacher-fronted explicit explanation treating awareness-raising as an objective (Takahashi 2001). For example, House (1996) studied the effect of two types of instruction on high-intermediate to advanced university students of German as a foreign language and examined their improvements in the areas of initiating and responding to speech acts and conversational routines. In the explicit instruction group, students received teacher-fronted explicit metapragmatic information about the sociopragmatic conditions governing the use of routines and their pragmatic functions. Students in the implicit instruction group did not receive explicit metapragmatic information about the target features. After 14 weeks of instruction and listening to tapes of their own language behaviour, sample conversation recordings showed that both groups had improved, but the explicit group’s improvement exceeded that of the implicit group. Tateyama et al. (1997) investigated how beginner learners of Japanese as a foreign language developed Japanese pragmatic proficiency under two types of instructional treatment. Targeted pragmatic features were the three functions of the routine formula sumimasen as an attention-getter, an apology, and an expression of thanks. In the explicit group, students discussed the different functions of sumimasen, received explicit teacherfronted explanations, and watched short video clips of examples of the pragmatic routines. The implicit group watched the same video clips as the explicit group but did not engage in any explicit metapragmatic activities. After only 50 minutes of instruction, the results of quantitative and qualitative instruments, including role-play, multiple-choice test, and selfreports, showed the explicit group’s advantage over the implicit group. In a similarly designed study, Rose and Ng (2001) investigated the effectiveness of explicit and implicit approaches to teaching compliments and compliment responses. Both explicit and implicit instruction groups followed the same procedures with one exception: the implicit group was exposed to film segments and additional examples with a guided questionnaire on the target feature in place of teacher-fronted instruction. After six 30-minute lessons, self-assessment, discourse completion, and metapragmatic questionnaires showed that both groups improved in pragmalinguistic proficiency, but only the explicit instruction group effectively developed sociopragmatic proficiency. In a later study, Rose (2005) explained the similar improvement of both explicit and implicit instruction groups as a result of participants’
6
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
advanced proficiency and the relative easiness of the pragmalinguistic target features. In another study that provided support for explicit instruction, Takahashi (2001) investigated four input enhancement conditions for Japanese learners acquiring request strategies in English: explicit instruction, form-comparison, form-search, and meaning-focused conditions. In the explicit instruction condition, a teacher provided metapragmatic and explicit explanations of the target feature. In the form-comparison condition, learners compared their own request strategies with those provided by native speakers of English. In the form-search condition learners compared request strategies of Japanese learners of English with those provided by native speakers of English. Finally, in the meaning-focused condition, learners simply listened to, read, and answered comprehension questions based on the input. After four 90-minute weeks of instruction, the results of discourse completion tests and self-reports demonstrated that the explicit group learned all of the different request strategies more successfully than the other three groups. Despite general support for explicit instruction in the literature, a number of studies have reported inconclusive findings regarding the effectiveness of explicit instruction on the acquisition of L2 pragmatics. In their study of English pragmatic mitigators in requests, Fukuya and Clark (1999) compared input enhancement with explicit metapragmatic instruction for intermediate and advanced ESL learners. The explicit group watched a video of explicit instruction on English mitigators with 30 example scenarios without subtitles. In contrast, the implicit instruction group watched a video of explicit instruction on listening comprehension strategies with thirty example scenarios where requests were subtitled, and the mitigators were highlighted. After one 48-minute lesson, the results of two assessment measures, a listening comprehension test and a pragmatic multiple-choice test, indicated no differences between the input enhancement group and explicit group. Fukuya and Clark suggested that the brevity of treatment may have contributed to their statistically insignificant results. Although lesson length affects pragmatic learning, it should be noted that Tateyama et al. (1997) produced clear results even within a short, 50-minute lesson. To reexamine previous findings, Tateyama (2001) conducted a follow-up study, increasing the instructional period to four 20-minute sessions and found that there were no significant differences between the explicit and implicit groups. As explanation, Tateyama noted that students in the implicit group had more contact with native speakers of Japanese outside of class, and this threat to internal validity contributed to the inconclusive results. Lastly, regarding the nature of the L2 pragmatic learning, studies show differing levels of acquisition in terms of accuracy and retention. Takahashi (2001) found some of the participants in the explicit teaching condition used non-target pragmalinguistic forms in the discourse completion tests as a result of previous instruction interfering with their restructuring process. House (1996) also found that neither implicit nor explicit instruction
MASAHIRO TAKIMOTO
7
improved learners’ performance in the realization of appropriate routinized responses. Finally, Takahashi (2001) observed that the degree of attainment and lasting effect of L2 pragmatic proficiency under the explicit teaching condition was doubtful. These findings lead to the question of what sort of input-based approach, with or without the teacher-fronted explicit information, is the most appropriate way of allowing learners to access and integrate sociopragmatic and pragmalinguistic knowledge more quickly and efficiently.
THE PRESENT STUDY To date, only a small number of studies have examined input-based methods of teaching L2 pragmatics. For this reason, there is no clear indication in the literature as to what type of input-based task involving pragmalinguistic–sociopragmatic connections is most effective in teaching L2 pragmatics. Moreover, no studies have compared the effectiveness of structured input tasks with and without explicit information for L2 pragmatics. To address these gaps, the following research question is investigated in this study. What are the relative effects of three different input tasks including (1) structured input with explicit information, (2) problem-solving, and (3) structured input without explicit information on the development of Japanese learners’ pragmatic proficiency in English?
METHOD Participants Prospective participants were solicited in Japan through an employment advertisement in a weekly magazine and on the Internet. Both the weekly magazine and the Internet website target students. After checking applicants’ scores for the Test of English for International Communication (TOEIC), only learners with intermediate English proficiency, defined as TOEIC scores between 500 and 700, were selected for inclusion in the study. This decision was made to exclude the extreme ends of learner proficiency levels, low and high, which might obscure the effects of the different types of instruction. The sixty participants were assigned to one of the four groups consisting of the three treatment groups, structured input tasks with explicit information, problem-solving tasks, and structured input tasks without explicit information, and the control group (n ¼ 15 for all four groups). The participants included three high school students, ten vocational training school students, twenty-nine university students, and eighteen non-students. All participants had studied English in Japan for a range of five to twentytwo years without receiving explicit instruction on English pragmatics.
8
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
All had Japanese as their first language and they ranged in age from 18 to 40. The present study was conducted at an English conversation school instead of at a regular EFL school because it was not possible at the instructor’s institution, a regular university, to include the control group necessary for observing and comparing the effects of the different instructional treatments.
Instructional goals The present study draws on past research on EFL Japanese students’ acquisition of downgraders, a pragmatic resource for mitigating the strength of a statement or request (Takahashi 1996; Hill 1997). Takahashi (1996) found that Japanese EFL learners tended to use monoclausal English request forms when downgrader biclausal request forms were more appropriate, for example ‘Would/Could you VP?’ vs. ‘Would it be possible to VP?’ Hill (1997) found that even as the proficiency of Japanese EFL learners increased, they continued to underuse clausal downgraders, lexical downgraders, and syntactic downgraders. Lexical/clausal downgraders soften the difficulty that the speaker experiences when asking the hearer to perform a request by modifying the Head Act lexically or clausally, for example ‘Could you possibly come here?’ (lexical) or ‘Would it be possible to come here?’ (clausal-mitigated preparatory question); ‘I wonder if you could come here’ (clausal-mitigated preparatory statement); ‘I would appreciate it if you could come here’ (clausalmitigated want statement). Syntactic downgraders, on the other hand, modify the Head Act syntactically by mitigating the level of difficulty that the speaker experiences through syntactic choices using tense or aspect, for example ‘I am wondering if you could lend me a book’ (continuous aspect); ‘I was wondering if you would come’ (past tense). Given Takahashi (1996) and Hill’s (1997) findings, the current study focuses on teaching lexical/clausal and syntactic downgraders in English requests. Three social context variables were carefully controlled for in the dialogues in the instructional and testing materials: (a) power, the status of the speaker with respect to the hearer; (b) distance, the relationship between the speaker and the hearer; and (c) speaker difficulty, the difficulty that the speaker experiences when asking the hearer to perform the request. These three variables were selected because in cross-cultural pragmatics, they are considered to be the three independent and culturally sensitive variables that subsume all other variables and play important roles in speech act behaviour. The participants in the three treatment groups were instructed to pay attention to these social context variables as well as the pragmalinguistic features of the target structures. Only participants in the structured input tasks with explicit information group were provided with the explicit information about the pragmalinguistic and sociopragmatic features of the target structures.
MASAHIRO TAKIMOTO
9
Table 1: Treatment features of each group Group
Treatment
Structured input tasks with explicit information (n ¼ 15)
Explicit information (10 minutes) þ Pragmalinguistic-sociopragmatic connection activities (18 minutes) þ Reinforcement activities (12 minutes) Pragmalinguistic-focused activities (10 minutes) þ Sociopragmatic-focused activities (10 minutes) þ Pragmalinguisticsociopragmatic connection activities (10 minutes) þ Metapragmatic discussion (10 minutes) Pragmalinguistic-sociopragmatic connection activities (24 minutes) þ Reinforcement activities (16 minutes) TOEIC reading comprehension exercises (40 minutes)
Problem-solving tasks (n ¼ 15)
Structured input tasks without explicit information (n ¼ 15) Control group (n ¼ 15)
Explicit information Yes
No
No
No
Instructional treatments Each group received four 40-minute treatment sessions in Japanese at a major English conversation school in Osaka, Japan. All groups received instruction from the same instructor, who was also the researcher.1 The three instructional treatments were matched for target structure, and all four groups were matched for instruction time. The first treatment session highlighted lexical/clausal downgraders in English requests, and the second treatment session focused on syntactic downgraders. The third and fourth treatment sessions were reviews of the first and second treatments. Handouts contained highlighted pragmalinguistic and sociopragmatic features to promote participants’ conscious noticing of those features. Specific treatment features of each participant group are presented in Table 1.
Structured input tasks with explicit information The structured input tasks with explicit instruction treatment consisted of two components: (a) teacher-fronted explanation of the target downgraders, and (b) structured input tasks comprising pragmalinguistic–sociopragmatic connection activities and reinforcement activities of the target downgraders. In each lesson, the group received handouts with a brief summary of the targeted downgraders and examples of the target structures in English. The explicit teacher-fronted component lasted 10 minutes, during which time the teacher read the summary and examples aloud in English and explained the summary and the examples in Japanese with special attention to
10
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
sociopragmatic conditions. In the second part, the group engaged structured input tasks consisting of three pragmalinguistic–sociopragmatic connection activities and three reinforcement activities (see Appendix A available online to subscribers at http://applij.oxfordjournals.org/). In the pragmalinguistic– sociopragmatic connection activities, the participants read dialogues for given situations and chose the more appropriate request form out of the two offered based on their pragmalinguistic and sociopragmatic knowledge. Participants then listened to an oral recording of the dialogue and underlined the correct request form. In the reinforcement activities, participants read each dialogue aloud and listened to the oral recordings again. Finally, they were asked to rate the level of appropriateness of each underlined request on a 5-point scale. The goal of the pragmalinguistic–sociopragmatic connection activities was to ensure that participants focused on understanding the relationship between the request, the relevant social context variables, and the targeted pragmalinguistic resources. In contrast, the purpose of the reinforcement activities was to strengthen the pragmalinguistic–sociopragmatic connections by providing the participants with more opportunities to observe and understand how the different factors, the request, the social context variables, and the targeted pragmalinguistic features, were interrelated.
Problem-solving tasks The problem-solving treatment consisted of four activities highlighting the targeted downgraders in English: pragmalinguistic-focused activities, sociopragmatic-focused activities, pragmalinguistic–sociopragmatic connection activities, and metapragmatic discussion. In each lesson, the participants received handouts with three sets of English dialogues (see Appendix A available online to subscribers at http://applij.oxfordjournals.org/). The participants read each situation and the dialogue and then listened to the dialogue. In the first activity, the pragmalinguistic-focused activity, participants were asked to copy and compare the underlined request forms in two dialogues while looking for the differences between the request forms. In the second activity, the sociopragmatics-focused activity, participants answered two questions regarding the relationship between the two characters and the difficulty of the requests. In the third activity, the pragmalinguistic–sociopragmatic connection activity, participants were asked to rate the level of politeness of the requests in both dialogues and to list the ways in which one character tried to be more polite than the other character when making requests. In the last activity, paired participants discussed the features of the target structures with each other. The aim of the first three activities was to provide participants with step-by-step problem-solving opportunities through which they could develop their own explicit knowledge about the target features. In turn, this explicit knowledge would help participants to reinforce the pragmalinguistic–sociopragmatic
MASAHIRO TAKIMOTO
11
connections during their paired metapragmatic discussions of the features of target structures. The decision to include paired metapragmatic discussions was based on findings in Rose (2005) that suggested metapragmatic discussion about the target request forms in context is more effective for learning sociopragmatic distinctions than the teacher-fronted approach.
Structured input tasks without explicit information The treatment for the structured input tasks without explicit information group was the same as for structured input tasks with explicit information but without the teacher-fronted explicit instruction.
Control group Lessons for the control group were designed to help participants do well on the TOEIC. Participants in this group engaged in reading comprehension exercises for the TOEIC; however, they were not exposed to the target structures at all.
Testing instruments and procedures This study employed a pre-test, post-test, and follow-up test design. The pretest was administered two to three days prior to the instructional treatment, the post-test eight to nine days after the treatment, and the follow-up test in the fourth week following instruction. Each test consisted of two input-based tests, a listening test and an acceptability judgement test, and two outputbased tests, a discourse completion test and a role-play test. Immediately following the treatments, participants completed an evaluation questionnaire. Situations in the four testing instruments comprised the speech act request with the three social context variables, power, distance, and speaker difficulty. In particular, the tests included situations with a high level of speaker difficulty combined with power and distance, which were validated by Hill (1997), Hudson et al. (1992, 1995) and Takahashi (1998, 2001). One high speaker difficulty item is shown below. You are writing a difficult paper for Professor Hill. You need some help with the paper but Professor Hill is away for a month. A friend of yours has suggested you go and see Professor Watson. Although you do not know Professor Watson and Professor Watson is extremely busy, you have decided to ask Professor Watson to look through your long paper before you hand it in the next day. What would you ask Professor Watson? (based on Takahashi 1998, 2001) Situations with a low level of speaker difficulty were added as distractors to increase the reliability of instruments. The discourse completion test, the role-play test, and the acceptability judgement test consisted of twenty situations (ten high speaker difficulty and ten low speaker difficulty items), while the listening test consisted of fifteen situations (nine high speaker difficulty items and six low speaker difficulty items).
12
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
Three versions of the four tests were developed and counterbalanced for order of presentation of the same situations across the pre-tests, post-tests, and follow-up tests. During the testing, these counterbalanced versions were used to minimize the possible effect of test learning. During the pre-tests, post-tests, and follow-up tests, test components were administered in the following order: discourse completion test, role-play test, listening test, and acceptability judgement test. The two input-based tests were administered after the output-based tasks to address the concern that they might provide participants with models for the production tests. Although participants were instructed to complete the four tests in the span of 2 hours, only the listening test was timed. In the listening test, participants had two seconds to judge the appropriateness for each question, which required participants to access their proceduralized knowledge of the target structure.
Discourse completion test The discourse completion test (see Appendix B available online to subscribers at http://applij.oxfordjournals.org/) required the participants to read short descriptions of twenty situations in English and to write what they would say in each situation in English. Although there was no time limit for the discourse completion test, most participants finished in 30–45 minutes. Each response was rated by two native English speakers according to the appropriateness of the request forms on a 5-point scale. The test contained ten target items with a maximum score of 100 based on the two raters’ scores (10 5 2).
Role-play test The role-play test (see Appendix B available online to subscribers at http:// applij.oxfordjournals.org/) consisted of short descriptions of twenty situations written in English and required the participants to play particular roles with an interlocutor. Prior to the role play, participants received role cards that described the situations and their roles. In each role play, the participants were required to initiate the conversation by requesting something from their interlocutor.2 The instructor, a non-native speaker of English, acted as the interlocutor. On average, participants took 2–3 minutes to prepare for each role play. Role plays were tape-recorded and individual performances were rated for appropriateness of request forms on a 5-point scale by two native speakers of English. The test contained 10 target items with a maximum score of 100 based on the two raters’ scores (10 5 2).
Listening test The listening test (see Appendix B available online to subscribers at http://applij.oxfordjournals.org/) required participants to listen to dialogues between a Japanese university student and a native speaker of English in fifteen different situations and to score the appropriateness of the Japanese university student’s request forms on a 5-point scale. The test
MASAHIRO TAKIMOTO
13
involved an audio-recording of the dialogue and had a time constraint of 2 seconds per question for responding to each dialogue. Participant ratings were compared with English native speakers’ baseline data3 on a 5-point scale. The test contained nine target items with a maximum score of 45 (9 5).
Acceptability judgement test The acceptability judgement test (see Appendix B available online to subscribers at http://applij.oxfordjournals.org/) was a computer-based test that required participants to read written English descriptions of 20 situations. After reading the descriptions, participants were presented with a series of isolated requests and instructed to score the first request on an 11-point scale and then to score subsequent responses proportionally higher or lower in accordance with the degree of perceived acceptability. Participant ratings were compared with English native speakers’ baseline data4 on a 5-point scale. The test contained ten target items with a maximum score of 50 (10 5).
Evaluation questionnaire The evaluation questionnaire was administered as a supplement to the present study with the goal of examining whether the aims of the instructional treatments had been achieved and how the instruction could be improved for future use. The questionnaire consisted of three 5-point scale close-ended questions and three open-ended questions.
RELIABILITY Interrater reliability was estimated by calculating the correlation of the two raters’ scores. Correlation coefficients for the discourse completion and role-play tests were .995 and .994 respectively, which were statistically significant (p 5 .001). With regard to internal consistency, the Cronbach Alpha reliability estimates for the tests ranged from .853 for the listening comprehension test5 to .926 for the role-play test with .893 for the acceptability judgement test and .917 for the discourse completion test.
VALIDITY To promote content validity, the present study matched test items to the theoretical framework that outlined the degree of the three social context variables: speaker difficulty, power, and distance. Tables 2 and 3 show the distribution and degree of social context variables across tests.
14
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
Table 2: Distribution of social context variables (Version A for the discourse completion, role-play, and acceptability judgement tests) S4 S6 S10 S18 S2 S8 S12 S14 S16 S20 S1 S3 S5 S11 S13 S7 S9 S15 S17 S19 SD þ þ þ P D þ þ þ
þ þ
þ þ þ þ þ þ
þ þ
þ þ
þ þ
þ þ þ
þ
þ
Note: S ¼ Situation; SD ¼ Speaker difficulty; P ¼ Power; D ¼ Distance; þ ¼ More; ¼ Less; ¼ Equal.
Table 3: Distribution of social context variables (Version A for the listening test)
SD P D
S3
S5
S8
S13
S2
S6
S10
S12
S15
S1
S4
S9
S7
S11
S14
þ þ
þ þ
þ þ
þ þ
þ þ
þ þ
þ þ
þ þ
þ þ
þ
þ
þ
Note: S ¼ Situation; SD ¼ Speaker difficulty; P ¼ Power; D ¼ Distance; þ ¼ More; ¼ Less; ¼ Equal.
RESULTS The results showed that the three treatment groups performed significantly better than the control group. However, the group that received the structured input tasks with explicit information did not retain the positive effects of the treatment between the post-test and follow-up test on the listening test component. In the data analysis a Bonferroni adjustment was employed to maintain an approximate experiment-wide .05 alpha level. In other words, the overall alpha level was set at .05, but with four group comparisons (discourse completion, role-play, acceptability judgement, and listening tests) for one item type (high speaker difficulty). Therefore, .05 was divided by the number of comparisons (four), resulting in a p value of .0125 for individual statistical decisions.
Results from the discourse completion test Results of a two-way repeated-measures ANOVA revealed a significant main effect for Instruction for the three treatment groups (structured input tasks with explicit information, problem-solving tasks, and structured input tasks without explicit information), F (3, 56) ¼ 99.92, p ¼ .000. A significant main effect for Time across the pre-test, post-test, and follow-up test was also
MASAHIRO TAKIMOTO
15
100 90 INSTRUCTION MEAN
80
SP
70 PS 60 SI 50 Control 40 Pre-test
Post-test
Follow-up test
TIME
Figure 1: Interaction plot for the discourse completion test Note: SP ¼ Structured input tasks with explicit information; PS ¼ Problemsolving tasks; SI ¼ Structured input tasks without explicit information
found, F (3, 56) ¼ 583.57, p ¼ .000. Lastly, the results revealed a significant interaction effect between Instruction and Time, F (9, 56) ¼ 48.64, p ¼ .000. Figure 1 illustrates three important characteristics of the discourse completion test results: (1) there were no statistically significant differences among the four groups on the pre-test scores; (2) the three treatment groups made gains from the pre-tests to the post-tests and the follow-up tests; and (3) positive effects for the three treatments were maintained through the post-test to the follow-up tests. These results suggest that the three different types of treatment were effective in promoting learners’ acquisition and retention of English request forms in the context of discourse completion activities. Furthermore the lack of crossover between the treatment and control groups on the post-tests demonstrates the relative superiority of the three treatment groups’ performances over the control group’s performance. Post-hoc Scheffe´ tests conducted on the post-test and the follow-up test scores for the main effect for treatment showed the following two contrasts: (1) the three treatment groups performed significantly better than the control group on the discourse completion test, and (2) there were no significant differences among the three treatment groups.
Results from the role-play test The results of a two-way repeated-measures ANOVA for the role-play test scores revealed the same significant main effects as the discourse completion test: a significant main effect for Instruction, F (3, 56) ¼ 83.93, p ¼ .000, a significant main effect for Time, F (3, 56) ¼ 502.61, p ¼ .000, and a
16
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
100 90 INSTRUCTION MEAN
80 SP 70 PS 60 SI 50 Control 40 Pre-test
Post-test TIME
Follow-up test
Figure 2: Interaction plot for the role-play test Note: SP ¼ Structured input tasks with explicit information; PS ¼ Problemsolving tasks; SI ¼ Structured input tasks without explicit information significant interaction effect between Instruction and Time, F (9, 56) ¼ 33.88, p ¼ .000. The results displayed for the role-play test in Figure 2 follow the same pattern as the discourse completion test: there were no statistically significant differences among the four groups on the pre-test scores, the three treatment groups improved from the pre-test to the post-test and the follow-up test, and positive effects for the three treatments were retained. As in Figure 1 above, the lack of crossover between the treatment and control group scores in Figure 2 shows the superior performance of all three treatment groups on the post-test and follow-up tests. Post-hoc Scheffe´ tests provided further evidence for the following two contrasts: (1) the three treatment groups performed significantly better than the control group on the role-play test, and (2) there were no significant differences among the three treatment groups.
Results from the listening test The results of a two-way repeated-measures ANOVA of the listening test scores showed the same significant main effects as both the discourse completion and the role-play test: a significant main effect for Instruction, F (3, 56) ¼ 27.48, p ¼ .000; a significant main effect for Time, F (3, 56) ¼ 81.27, p ¼ .000; and a significant interaction effect between Instruction and Time, F (9, 56) ¼ 7.97, p ¼ .000. Figure 3 reveals two of the same main results for the listening test as for the discourse completion and role-play tests: there were no statistically significant differences among the four groups on the pre-test scores and the three treatment groups improved from the pre-tests to the post-tests and the
MASAHIRO TAKIMOTO
17
50
INSTRUCTION
MEAN
40
SP 30 PS 20 10 Pre-test
SI Control Post-test TIME
Follow-up test
Figure 3: Interaction plot for the listening test Note: SP ¼ Structured input tasks with explicit information; PS ¼ Problemsolving tasks; SI ¼ Structured input tasks without explicit information follow-up tests. However, Figure 3 also shows that unlike the problemsolving tasks and structured input tasks without explicit information groups, the structured input tasks with explicit information group did not retain the positive effects of the treatment between the post-test and the follow-up test. A separate one-way ANOVA performed on the follow-up test scores showed a significant difference between the three treatment groups’ performance on the listening test. Despite the differences in the follow-up test performance, all three treatment groups outscored the control group without any crossovers between group scores, confirming the superior performance of the treatment groups on the listening test. Post-hoc Scheffe´ tests provided further support for the following four contrasts: (1) all three treatment groups performed significantly better than the control group on the post-test and follow-up test; (2) there were no statistically significant differences among the three treatment groups on the post-test; (3) there were no statistically significant differences between the problemsolving tasks and structured input tasks without explicit information groups on the follow-up test; and (4) problem-solving tasks and structured input tasks without explicit information groups performed significantly better than the structured input tasks with explicit information group on the follow-up test.
Results from the acceptability judgement test The results of a two-way repeated-measures ANOVA for the acceptability judgement test showed two of the same significant main effects as the other tests: a significant main effect for Instruction, F (3, 56) ¼ 7.32, p ¼ .000 and a
18
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
60
MEAN
50
INSTRUCTION SP
40 PS 30 20 Pre-test
SI Control Post-test
Follow-up test
TIME
Figure 4: Interaction plot for the acceptability judgement test Note: SP ¼ Structured input tasks with explicit information; PS ¼ Problem-solving tasks; SI ¼ Structured input tasks without explicit information significant main effect for Time, F (3, 56) ¼ 43.07, p ¼ .000. However, no significant interaction effect between Instruction and Time was found, F (9, 56) ¼ 3.21, p ¼ .006. The results displayed in Figure 4 for the acceptability judgement test follow the same pattern as the other test components: there were no statistically significant differences among the four groups on the pre-test, the three treatment groups’ performance improved from the pre-test to the post-test and follow-up test, and positive effects for the three treatments were maintained. As with the other test components, the acceptability judgement test scores exhibit no crossover between groups, showing the superior effects of the three treatment conditions as compared with the control condition on participants’ post-test performance. Post-hoc Scheffe´ tests provided further evidence for the following contrasts: (1) the three treatment groups performed significantly better than the control group on the acceptability judgement test; and (2) there were no significant differences among the three treatment groups.
Results from the evaluation questionnaire Analysis of responses on the evaluation questionnaire provided insight into the participants’ experience from a first-person, retrospective point of view. Table 4 summarizes responses on the close-ended questions (Q1–Q3) with the mean, standard deviation, degrees of freedom, and p-values for each question. Participants responded on a scale of 1–5 with 1 ¼ not at all, and 5 ¼ very interesting/difficult/clearly.
MASAHIRO TAKIMOTO
19
Table 4: Results for close-ended questionnaire items (Q1, Q2, and Q3) Question
Mean
SD
Q1: Did you find the lessons interesting? Q2: Did you find the lesson difficult to follow? Q3: Did you understand clearly how to make polite requests?
3.78 2.49
.83 1.05
F (2, 42) ¼ .51, p ¼ .60 F (2, 42) ¼ .97, p ¼ .39
3.89
.76
F (2, 42) ¼ 1.89, p ¼ .17
Table 5: Summary of open-ended question items (Q4, Q5, and Q6) Questions and reported contents
Task types Structured input Problem-solving Structured input tasks with explicit tasks without information explicit information
Q4: Write down the main points you learned in lessons Including all main points 15 (100%) 15 (100%) 15 (100%) Q5: Were there things you liked a lot about the lessons? Learning polite requests 12 (80%) 12 (80%) 11 (73%) Learning the same 2 (13%) 1 (7%) 1 (7%) thing over and over Other 1 (7%) 2 (13%) 3 (20%) Q6: Were there things you did not like about the lessons? Monotonousness of lessons 4 (27%) 4 (27%) 8 (54%) No output practices 1 (7%) 1 (7%) 0 (0%) Little feedback 3 (20%) 3 (20%) 2 (13%) Prohibition on taking 4 (27%) 1 (7%) 0 (0%) materials home Other 3 (20%) 6 (39%) 5 (33%)
These results suggest that the lessons were interesting (Q1: M ¼ 3.78), relatively easy to follow (Q2: M ¼ 2.49), and comprehensible (Q3: M ¼ 3.89). The results also show that there were no significant differences among the treatment groups’ responses on Q1, Q2, or Q3. Analysis of the participants’ responses on the open-ended questions (Q4, Q5, and Q6) are reported in Table 5. Responses for Q4 demonstrate that all participants were able to remember the main points they learned in the lessons. In their responses to Q5, 73–80 per cent of the participants reported that the main good point of the lessons was learning polite requests. The fact that all participants remembered the main points of the lessons and a high proportion of the
20
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
participants reported learning polite requests as a highlight of the lessons indicates that the three types of input-based instruction were effective. Responses to Q6 show differing participants’ views on the weaknesses of the lessons including monotony, no chance to produce language, and limited feedback.
DISCUSSION The aim of the present study was to investigate the relative effects of input tasks including structured input tasks with and without explicit information and problem-solving tasks on the development of Japanese learners’ L2 pragmatic proficiency in the area of requests. The results show that participants who received the three different types of input-based instruction outperformed the control group. Furthermore, the results for the two types of input-based tasks, structured input tasks (with or without explicit information) and problem-solving tasks, indicate that both types are equally effective. These results show that the development of L2 pragmatic proficiency can be influenced by manipulating input, lending support to findings in previous studies on the effects of structured input tasks and problem-solving tasks on the acquisition of L2 pragmatics. There are two possible reasons for the effectiveness of the different types of input-based tasks. One possibility is that the different treatments drew the participants’ attention to pragmalinguistic forms in the input that they received. Despite their differences, the treatment conditions may have made the target structures equally salient. The participants in the structured input task condition, with or without explicit information, engaged in tasks that required their attention to the pragmalinguistic forms of target structures. In the pragmalinguistic–sociopragmatic connection activities, the participants chose the more appropriate request form from two options, and in the reinforcement activities, participants rated the level of appropriateness of each bold-faced underlined request. On the other hand, participants in the problem-solving condition had to pay attention to the highlighted requests in two dialogues in order to copy and compare the request forms before discussing the metapragmatic features of target structures in the dialogues. The second possible reason for the effectiveness of the input-based tasks is the deeper processing that arises when pragmalinguistic–sociopragmatic connections are involved. In their discussion of the level of processing involved in meaning, Craik and Lockhart (1972) claim that the quality of a memory trace depends on the level or depth of perceptual and mental processing where meaning plays an important role. When the participants focused on the pragmalinguistic–sociopragmatic connections of the target feature, they may have been inclined to process the meanings at a deeper level, leading to greater retention. The tasks in the present study were designed to require participants to access and integrate their pragmalinguistic and sociopragmatic knowledge through various activities. Moreover, the
MASAHIRO TAKIMOTO
21
participants in the problem-solving tasks had the opportunity to discuss the metapragmatic features of target structures, thereby reinforcing pragmalinguistic–sociopragmatic connections and allowing for processing at a deeper level. The results indicate that the three types of treatments had similar effects on the development of Japanese learners’ pragmatic proficiency as measured by three of the four test components: discourse completion, role-play, and acceptability judgement tests. However, regarding the listening test, although all three treatment conditions showed significant improvement on the posttest, the structured input tasks with explicit information group performed at a significantly lower level than the other two conditions on the follow-up test. Why, then, did the structured input tasks with explicit information group perform as well as the problem-solving and structured input tasks without explicit information groups on the listening post-test but not the follow-up test while all the groups performed similarly on the other post-tests and follow-up tests? Any answer to this question is necessarily speculative as no information on the psycholinguistic processing involved in either the treatments or the test was available. What distinguishes the listening test from the other tests is the requirement for online processing. Online processing tests place demands on working memory, as participants have to process and respond to the stimuli rapidly. Also, all three treatments can be assumed to have provided the participants with some explicit knowledge, but the treatments differed in how this knowledge was achieved. In the case of the first treatment, structured input tasks plus explicit information, the participants were simply given explicit information; they did not have to discover the rules for themselves. In the other two treatments, problem-solving and structured input tasks minus explicit information, participants had to discover the rules for themselves. It is possible, then, that the problem-solving and structured input tasks without explicit information participants attended to the pragmalinguistic and sociopragmatic features of the target structures more deeply. That is, the provision of explicit information did not push the participants in the structured input tasks plus explicit information group to process the target structures deeply. The problem-solving and structured input tasks without explicit information treatments, however, involved greater depth of processing, resulting in knowledge that was more firmly embedded and thus more easily accessed. Immediate post-test results did not reveal this difference because the explicit knowledge was fresh in the participants’ memories. However, on the listening follow-up test, participants in the structured input tasks with explicit information group were less successful in accessing their weakly established explicit knowledge while coping with the test’s demands on their working memory capacities. Participants in problem-solving and structured input tasks without explicit information groups, however, were still able to cope with the demands of the listening test because their explicit knowledge was firmly entrenched. Although the
22
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
explanation provided here is speculative at best, the current study’s results are consistent with results in VanPatten and Oikennon (1996) which showed that in that teacher-fronted explicit information is not important because structured input tasks by themselves are effective. Takahashi (2001) also found that the explicit teaching condition was effective through the post-test stage but expressed doubts regarding the lasting effect of gains in L2 competence due to explicit enhancement.
CONCLUSION The present study examined the relative effects of input-based approaches and the presence or lack of explicit information on teaching polite request forms in a Japanese EFL context. The results indicate that input processing tasks, including structured input and problem-solving tasks, function effectively when they provide learners with an emphasis on the pragmalinguistic and sociopragmatic features of the target structure. An important pedagogical implication of this study is that teachers should be aware that effective learning occurs when the tasks provide learners with opportunities for processing both pragmalinguistic and sociopragmatic features of the target structures. Structured input tasks and problem-solving tasks can be used together and can even complement each other. The findings of this study have practical applications, especially in the Japanese EFL context, where English pragmatics rather than English grammar is an increasingly important area of instruction. In such classrooms, learners are encouraged to have strong pragmatic awareness to make use of the limited class time typically available for learning English. These findings may be generalizable to other EFL countries with similar situations. Limitations of the present study suggest several areas for future research. Regarding the activities within each treatment condition, multiple activities were packaged together, and there is no way of disentangling the contributions of each individual activity to the effect on learning outcomes. Another limitation of the study was the representativeness of the participants and the generalizability of the results. Participants in the present study were recruited on a volunteer-basis and may have differed in their outcomes and responses than students at existing educational institutions or the general population given the same instruction. Conducting the same experiment at an existing educational institution would strengthen the generalizability of the results to other EFL situations. Furthermore, conducting the same experiment with a representative sample of any population would improve the generalizability of the study’s results to broader contexts. Kasper (2001) argued that the benefit of conducting research within an established institution is that the results can be translated into recommendations for pedagogical practice with more plausibility than laboratory studies. Perhaps the relationship between teaching at different types of institutions, for example, language schools vs. universities, and learning outcomes could be a subject for future research.
MASAHIRO TAKIMOTO
23
Despite the shortcomings above, the present study contributes to our understanding of how input-based approaches to teaching English pragmatics lead to positive outcomes in the Japanese EFL context. The author hopes that the results of the present study will show researchers and teachers that successful input-based tasks should involve effective pragmalinguistic– sociopragmatic connection activities and be designed to raise the learner’s interest in acquiring L2 pragmatics. Final version received September 2007
ACKNOWLEDGMENTS I would like to express my sincere gratitude to Dr. Rod Ellis and Dr. Gabriele Kasper for their valuable comments and guidance. I would also like to thank anonymous reviewers for their constructive comments and suggestions. Thanks are also due to Mattew Prior and Castle Sinicrope for proofreading the manuscript
NOTES 1 In behavioural research, researcher expectancy can be a problem when the researcher teaches experimental groups. The researcher followed the instructional guidelines rigidly controlled for the effect with the double-blind technique after the data were collected in order to minimize any researcher expectancy effect during the treatments. 2 Ellis (2003) suggested that a testing situation where the examinees interact with other non-native speakers rather than with native speaker examiners is more likely to elicit the examinees’ best performances. 3 Ten native speakers of English listened to a dialogue between a Japanese university student and a native speaker of English in fifteen different situations and then scored the appropriateness of the Japanese university student’s request forms on a 5-point scale. The native speakers’ data were relatively uniformed and consistent (SD ¼ .00 .53, range ¼ .00 1.00). These data were used as the baseline data.
4 Ten native speakers of English were required to read written English descriptions of twenty situations with a Japanese supplement. They were then presented with a series of isolated requests and instructed to score the first request on an 11-point scale and then to score subsequent responses proportionally higher or lower in accordance with the degree of perceived acceptability. The native speakers’ data were relatively uniformed and consistent (SD ¼ .82 1.08, range ¼ 2.00 4.00). These data were used as the baseline data. 5 The reliability estimate for the LT was low because of five problematic items. By deleting the five problematic items out of the twenty items, a higher level of reliability was achieved. However, the reliability estimate for the LT was still somewhat lower than the others and this might be related to the narrower rating scales in this test. That is, the LT used a 5-point scale, while the AJT used an 11-point scale. According to Hatch and Lazaraton (1991), a broader scale range encourages more precision in respondents’ judgements.
24
THE DEVELOPMENT OF LEARNER’S PRAGMATIC PROFICIENCY
REFERENCES Alco´n, E. 2005. ‘Does instruction work for learning pragmatics in the EFL context?’ System 33: 417–435. Craik, F. and R. Lockhart. 1972. ‘Levels of processing: A framework for memory research,’ Journal of Verbal Learning and Verbal Behavior 11: 671–84. DeKeyser, R. 1995. ‘Learning second language grammar rules: An experiment with a miniature linguistic system,’ Studies in Second Language Acquisition 46/4: 613–42. Ellis, R. 1997. SLA Research and Language Teaching. Oxford: Oxford University Press. Ellis, R. 2003. Task-based Language Learning and Teaching. Oxford: Oxford University Press. Fotos, S. 1994. ‘Integrating grammar instruction and communicative language use through grammar consciousness-raising tasks,’ TESOL Quarterly 28: 323–51. Fotos, S. and R. Ellis. 1991. ‘Communicating about grammar: A task-based approach,’ TESOL Quarterly 25: 605–28. Fukuya, Y. and M. Clark. 1999. ‘Input enhancement of mitigators’ in L. Bouton (ed.): Pragmatics and Language Learning Monograph Series 10: 111–30. Urbana-Champaign, IL: Division of English as an International Language, University of Illinois, Urbana-Champaign. Fukuya, Y. and Y. Zhang. 2002. ‘Effects of recasts on EFL learners’ acquisition of pragmalinguistic conventions of request,’ Second Language Studies 21/1: 1–47. Hatch, E. and A. Lazaraton. 1991. The Research Manual: Design and Statistics for Applied Linguistics. Boston, MA: Heinle & Heinle. Hill, T. 1997. The Development of Pragmatic Competence in an EFL Context. Unpublished doctoral dissertation. Temple University, Japan. House, J. 1996. ‘Developing pragmatic fluency in English as a foreign language: Routines and metapragmatic awareness,’ Studies in Second Language Acquisition 18/2: 225–52. Hudson, T., E. Detmer, and J. D. Brown. 1992. A Framework for Testing Cross-cultural Pragmatics (Technical Report #2). Honolulu, HI: University of Hawai’i Press. Hudson, T., E. Detmer, and J. D. Brown. 1995. Developing Prototypic Measures of Cross-cultural Pragmatics (Technical Report #7). Honolulu, HI: University of Hawai’i Press. Jeon, E. and T. Kaya. 2006. ‘Effects of L2 instruction on interlanguage pragmatic development: A
meta-analysis’ in J. Norris and L. Ortega (eds): Synthesizing Research on Language Learning and Teaching. Philadelphia: John Benjamins, pp. 165–211. Kasper, G. 2001. ‘Classroom research on interlanguage pragmatics’ in K. Rose and G. Kasper (eds): Pragmatics in Language Teaching. New York: Cambridge University Press, pp. 33–60. Kasper, G. and K. Rose. 2002. ‘The role of instruction in learning second language pragmatics,’ Language Learning 52: 237–73. Koike, D. and L. Pearson. 2005. ‘The effect of instruction and feedback in the development of pragmatic competence,’ System 33: 481–501. Leech, G. 1983. Principles of Pragmatics. London: Longman. Lyster, R. 1994. ‘The effect of functional-analytic teaching on aspects of French immersion students’ sociolinguistic competence,’ Applied Linguistics 15/3: 263–87. Martı´nez-Flor, A. and Y. Fukuya. 2005. ‘The effects of instruction on learners’ production of appropriate and accurate suggestions,’ System 33: 463–80. Norris, J. and L. Ortega. 2001. ‘Does type of instruction make a difference? Substantive findings from a meta-analytic review’ in R. Ellis (ed.): ‘Form-focused instruction and second language learning,’ Language Learning 51 (Suppl. 1): 157–213. Robinson, P. 1996. ‘Learning simple and complex second language rules under implicit, incidental, rule-search, and instructed conditions,’ Studies in Second Language Acquisition 18/1: 27–67. Rose, K. 2005. ‘On the effects of instruction in second language pragmatics,’ System 33: 417–35. Rose, K. and C. Ng. 2001. ‘Inductive and deductive teaching of compliments and compliment responses’ in K. Rose and G. Kasper (eds.): Pragmatics in Language Teaching. New York: Cambridge University Press, pp. 145–70. Takahashi, S. 1996. ‘Pragmatic transferability,’ Studies in Second Language Acquisition 18/2: 225–52. Takahashi, S. 1998. ‘Quantifying requestive imposition: Validation and selection of situation for L2 pragmatic research,’ Studies in Languages and Cultures 9: 135–59. Takahashi, S. 2001. ’The role of input enhancement in developing interlanguage pragmatic competence’ in K. Rose and G. Kasper (eds): Pragmatics in Language Teaching.
MASAHIRO TAKIMOTO
New York: Cambridge University Press, pp. 171–99. Takahashi, S. 2005. ‘Noticing in task performance and learning outcomes: A qualitative analysis of instructional effects in interlanguage pragmatics,’ System 33: 437–61. Tateyama, Y. 2001. ‘Explicit and implicit teaching of pragmatic routines: Japanese sumimasen’ in K. Rose and G. Kasper (eds.): Pragmatics in Language Teaching. New York: Cambridge University Press, pp. 200–22. Tateyama, Y., G. Kasper, L. Mui, H. Tay, and O. Thananart. 1997. ‘Explicit and implicit teaching of pragmatic routines’ in L. Bouton (ed.): Pragmatics and Language Learning, Monograph Series 8. Urbana-Champaign, IL: Division of English as an International
25
Language, University of Illinois, Urbana-Champaign, pp. 163–78. Thomas, J. 1983. ‘Cross-cultural pragmatic failure,’ Applied Linguistics 4: 91–112. VanPatten, B. 1996. Input Processing and Grammar Instruction in Second Language Acquisition. Norwood, NJ: Ablex. VanPatten, B. and S. Oikkenon. 1996. ‘Explanation versus structured input in processing instruction,’ Studies in Second Language Acquisition 18/4: 495–510. Witten, C. 2000. ‘Using video to teach for sociolinguistic competence in the foreign language classroom,’ Texas Papers in Foreign Language Education 5/1: 143–75. (Eric Document Reproduction Service No. ED468314.)
APPENDIX A: EXAMPLES OF INSTRUCTIONAL MATERIALS USED Structured input task Pragmalinguistic-sociopragmatic connection activities: Read the following situation and the dialogue and choose the more appropriate request form out of two offered for each underlined part and indicate your choice by circling ‘(a)’ or ‘(b)’. Then, listen to an oral recording of the dialogue and indicate whether the actual request used in the dialogue is ‘(a)’ or ‘(b)’. Situation: Yuka is about to start her car when she notices that her car battery has gone flat. She needs to go to school now and she does not have any other means but to ask her landlord, Mr. Brown, whom she has never spoken to before, to give her a ride to school. Her landlord is extremely busy, but she decides to ask her landlord to drive her to school.
Brown:
Hello.
Yuka:
Hi, you are Mr. Brown, aren’t you?
Brown:
That’s right.
Yuka:
I’m a tenant next door. My car battery has just gone flat and I can’t start my car. I really need to get to school. 1. (a) I wonder if I could get a lift; (b) Could I get a lift?
Brown:
Well, actually, I am really busy helping other tenants moving into this apartment. So, I can’t really help you.
Yuka:
I understand, but it’s important that I get to school today because I have exams.
Reinforcement activities: Read the following situation and the dialogue and answer the following question. Situation: John is living in an apartment. He is extremely busy working on his assignment, but he needs to send a big parcel to England today. His landlady, Mrs. Taylor, whom he has never spoken to before, is extremely busy, but he decides to ask his landlady to send the big parcel. John sees the landlady.
John:
Hi, you are Mrs. Taylor, aren’t you?
Taylor:
That’s right.
John:
Hello. My name is John.
Taylor:
Oh, you are the tenant.
John:
Yes. I live next door.
Taylor:
How is it going?
John:
Pretty good, thank you. I’m very busy working on my assignment. 1. Could I possibly ask you a favor?
Taylor:
What’s the favor?
Indicate the appropriateness level of the underlined request from your point of view based on the scale below. very unsatisfactory 1—2—3—4—5 completely appropriate
Problem-solving task Read Dialogue A and Dialogue B and answer the following questions.
Dialogue A. Situation: Yuka is about to start her car when she notices that her car battery has gone flat. She needs to go to school now and she does not have any other means but to ask her neighbor, James, whom she knows well, to give her a ride to school. She sees her neighbor go out by car and she decides to ask her neighbor to drive her to school.
James:
Hi, Yuka.
Yuka:
Hi, Jim. My car battery has just gone flat and I can’t start my car. I really need to get to school. (a) Could I have a lift ?
James:
Sure. Get in.
Dialogue B. Situation: Yuka is about to start her car when she notices that her car battery has gone flat. She needs to go to school now and she does not have any other means but to ask her landlord, Mr. Brown, whom she has never spoken to before, to give her a ride to school. Her landlord is extremely busy, but she decides to ask her landlord to drive her to school.
Brown:
Hello.
Yuka:
Hi, you are Mr. Brown, aren’t you?
Brown:
That’s right.
Yuka:
I am a tenant next door. My car battery has just gone flat and I can’t start my car. I really need to get to school. (a)' I wonder if I could get a lift.
Brown:
Well, actually, I am really busy helping other tenants moving into this apartment. So, I can’t really help you.
1. Pragmalinguistic-focused activities: Write up the requests Yuka makes in Dialogue A and Dialogue B in the table below and also comment on the differences between the two. Dialogue A (a)
Dialogue B
Differences
(a)'
2. Sociopragmatic-focused activities: Indicate the level of the relationship between Brown and Yuka in Dialogue A and Dialogue B. Dialogue A:
very distant 1—2—3—4—5 very close
Dialogue B:
very distant 1—2—3—4—5 very close
3. Sociopragmatic-focused activities: In consideration of the respective relationship between Brown and Yuka in Dialogue A and Dialogue B, do you think that Yuka is asking for something simple or difficult in Dialogue A and Dialogue B? Dialogue A:
very simple 1—2—3—4—5 very difficult
Dialogue B:
very simple 1—2—3—4—5 very difficult
4. Pragmalinguistic-sociopragmatic connection activities: Look at the requests you found in #1 and answer the following questions. (1) Do you think Yuka is being polite in Dialogue A? Indicate the politeness level of the request by Yuka in Dialogue A. very impolite 1—2—3—4—5 completely polite (2) Do you think Yuka is being polite in Dialogue B? Indicate the politeness level of the request by Yuka in Dialogue B. very impolite 1—2—3—4—5 completely polite
5. Pragmalinguistic-sociopragmatic connection activities: How does Yuka try to be polite when making requests?
APPENDIX B: EXAMPLES OF TEST MATERIALS USED Discourse completion test Directions: Read each of the situations. After each situation write what you would say in that situation in a normal conversation.
You are writing a difficult paper for Professor Hill. You need some help with the paper but Professor Hill is away for a month. A friend of yours has suggested you go and see Professor Watson. Although you do not know Professor Watson and Professor Watson is extremely busy, you have decided to ask Professor Watson to look through your long paper before you hand it in the next day. What would you ask Professor Watson? (based on Takahashi 1998, 2001) Note: speaker difficulty =＋; power =‐; distance =＋; ＋= more; ‐= less;
= equal
You:__________________________________________________________________________
Role-play test Directions: There are 20 role-plays. You will be given a role-card before each of the role-plays. The role-card will describe the situation and your role (Role B). Read the situation described on the first card. It will require you to ask for something from the native speaker so you will have to start the role-play. The native speaker may or may not respond.
You must write an essay on a topic and turn it in by tomorrow. You have just found a very long interesting article on the topic, but it is written in Chinese. You can’t read the article in Chinese and you need to translate it into Japanese. You have just heard that there is a student in another department who is bilingual in Chinese and Japanese. Although you have never spoken to the student before, you decide to ask the student to translate the article into Japanese. (based on Hill 1997) Note: speaker difficulty =＋; power =
; distance =＋; ＋= more; ‐= less;
= equal
Now: You see the student.
Listening test Directions: Read each of the following situations and after each situation listen to a conversation between Taro (a Japanese university student) and an interlocutor (a native speaker of English) and then score the appropriateness of
Taro’s request on a 5-point scale.
Taro is working in a restaurant. The owner has asked Taro to get each customer to complete a very long questionnaire about the quality of the food and the service in the restaurant. Taro has given the questionnaire to a customer. Taro notices that the customer has not filled it in but is about to leave in a hurry. Taro needs to have the questionnaire filled by the customer. (based on Hudson et al. 1992, 1995) Note: speaker difficulty =＋; power =‐; distance =＋; ＋= more; ‐= less;
= equal
Taro:
Excuse me. I can see you are in a hurry. But please fill in this questionnaire.
Brown:
Oh, look. I’m sorry I really haven’t got the time.
not appropriate at all 1—2—3—4—5 completely appropriate
Acceptability judgment test Directions: Read each of the situations. After each situation you will be presented with three possible responses. Score the first possible response on an 11-point scale and score subsequent responses with a proportionally higher or lower number in accordance with the response’s degree of acceptability.
You overslept and missed the final exam for Professor Jackson’s course. You are not so familiar with Professor Jackson and you know that Professor Jackson has to hand in students’ grades in a few days and does not like to offer students a make-up exam. However, you need to pass the final exam to graduate and you have decided to go and ask Professor Jackson to give you a make-up exam. What would you ask Professor Jackson? (based on Takahashi 1998, 2001) Note: speaker difficulty =＋; power =‐; distance =＋; ＋= more; ‐= less;
a:
= equal
I was wondering if it would be possible for me to have a make-up exam. not appropriate at all 0—1—2—3—4—5—6—7—8—9—10 completely appropriate
b:
I want you to give me a make-up exam. not appropriate at all 0—1—2—3—4—5—6—7—8—9—10 completely appropriate
c:
Could you possibly give me a make-up exam? not appropriate at all 0—1—2—3—4—5—6—7—8—9—10 completely appropriate
Applied Linguistics 30/1: 26–48 ß Oxford University Press 2008 doi:10.1093/applin/amm057 Advance Access published on 14 February 2008
Soliciting Teacher Attention in an L2 Classroom: Affect Displays, Classroom Artefacts, and Embodied Action ASTA CEKAITE Linko¨ping University This paper explores L2 novices’ ways of soliciting teacher attention, more specifically, their summonses. The data are based on detailed analyses of video recordings in a Swedish language immersion classroom. The analyses illuminate the lexical shape of summonses in conjunction with prosody, body posture, gestures, and classroom artefacts. As demonstrated, a simple structure of summoning provided a handy method for soliciting and establishing the teacher’s attention, and facilitated the novices’ participation in classroom activities from early on. Importantly, however, the local design of the summonses was influenced by the competitive multiparty classroom setting. The analyses illustrate how the novices upgraded their summonses by displaying a range of affective stances. Different aspects of the students’ embodied actions were employed as ways of indexing affective stances, for example ‘tired’, ‘resigned’, or ‘playful’, that in the local educational order created methods that invited the teacher’s attention and conversational uptake. These locally available resources allowed children to upgrade their summonses and to indicate their communicative projects, in spite of their limited Swedish (L2) resources. The findings are discussed in terms of their implications for understanding participation in L2 classroom interactions as being a matter of delicately calibrated collaborative accomplishments. Keywords: L2 novices; summonses; embodied action; affective stances; classroom artefacts.
INTRODUCTION The present paper explores how primary school students solicit and secure teachers’ attention and participation in an immersion class context, where a second language (L2) constitutes the primary mode of classroom life. Recently, ethnographic discourse-oriented studies have illustrated how the micropolitics of classroom life shape children’s access to ‘linguaculture’ and resources in L2 learning (Willett 1995; Toohey 2000; Day 2002). To underscore the relevance of active participation in language practices, a growing number of microanalytical studies highlight how language learning opportunities are collaboratively constructed in learners’ interactions with teachers in mundane classroom discourse (Hall and Verplaetse 2000; Mondada and Pekarek Doehler 2004; Seedhouse 2004). As indicated by
ASTA CEKAITE
27
prior research, to be reviewed below, the ability to recruit participation of expert others is crucial for language learners. Managing to secure the teacher’s attention forms a part of a student’s interactional competence in the social ecology of the classroom (cf. Mehan 1979; Markee 2004). However so far, little is known about how children at an early stage of L2 learning are able to bring about the teacher’s attention and conversational involvement. Attention-getting strategies are important interactional devices that children employ to bring about the teacher’s involvement. The present study focuses on children’s (L2 beginners’) summonses as well as the responses they occasion in the unfolding classroom interaction. As an initiating speaker’s part in a sequence of actions, a summons (a vocative address, physical action, e.g. raising a hand) is typically designed to solicit the recipient’s attention (Schegloff 1968: 1080). The recipient’s answer to the summons (e.g. ‘yes’, gaze, nod) displays her/his expectation for further action on the first speaker’s part (C. Goodwin et al. 2002). Thereby, the recipient commits him-/herself to staying within the encounter (cf. Schegloff 1968). This type of sequence provides the most rudimentary form of attention-getting activity in L2 novice’s talk. However, in everyday life in the classroom, L2 novices attempting to secure the teacher’s attention often face a much more elaborate interactional task. Classroom life takes place in an interactional environment, where multiple simultaneous activities are pursued by various constellations of participants (van Dam 2002). Thus, an overwhelmingly pressing concern for students is to secure the teacher’s attention and interactional uptake in the context of many competing voices. By investigating ways in which children solicit and establish the teacher’s attention, the present study seeks to address the issue of learners’ access to participation in educational activities in a complex interactional setting. More generally, by examining the complex communicative abilities required when learners attempt to get conversational access to classroom activities, the present study seeks to enhance our understanding of a multilingual classroom as a social site for participation and language learning (Willett 1995; Breen 2001).
EARLIER RESEARCH ON ATTENTION-GETTING IN CHILDREN’S L2 INTERACTIONS Focusing on child L2 learners, several studies have illustrated the importance of attention-getting within L2 conversations. In her early discourse-oriented study, Hatch (1978) has shown that summonses are a crucial interactional resource in language learning. Attention-getting formats facilitate language acquisition in that they allow for collaborative topic nominations, establishment, and negotiations together with a more competent speaker (Hatch 1978; on L1 see Bruner 1981). Summoning is reported to be an initial and recurrent interactional device in children’s early L2 use (Hatch 1978; Linnakyla¨ 1980; Cathcart 1986; Cathcart-Strong 1986; Willett 1995; Pallotti 1996). For instance, in her study
28
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
of a 5-year-old Finish boy learning English, Linnakyla¨ (1980) documented that attention-getting strategies (vocatives, directives to look at something) were crucial initial means of achieving participation in informal conversations. Moreover, at an early stage of learning, such actions were often the only lexical contributions the child could make to the interaction (see also Krupa-Kwiatkowski 1998). Educational settings, including language classrooms, present institutional environments in which students’ and teachers’ actions are guided by the local institutional order that defines participants’ mutual rights and responsibilities and makes them accountable for specific actions (Freebody and Freiberg 2000). They also delineate what provides access to the teacher’s interactional space. Classroom discourse entails a variety of speech-exchange systems that define what counts as a legitimate way of gaining the conversational floor and the teacher’s attention (Mehan 1979; Merritt 1982; Seedhouse 2004). Merrit’s (1982) analysis of children’s solicitation of the teacher’s attention (during individual seatwork) in an L1 primary classroom setting revealed that getting the teacher’s attention was by no means unproblematic: the students often needed to recycle their soliciting moves several times. They used numerous lexical reformulations of their summons turns, at times verbally indicating their reason for wanting the teacher’s attention and assistance (e.g. ‘Ms. C. I haven’t done this one yet’). Clearly then, being accepted as a ratified conversational participant presents a basic communicative challenge that the student faces in multiparty classrooms and requires complex interactional skills (Merrit 1982). Studies on formal L2 learning settings have also provided some evidence that securing a teacher’s attention involves considerable interactional work. For instance, Cathcart (1986), in her study of Spanish children’s interactions in a bilingual English–Spanish immersion class, demonstrated that calling the teacher’s attention (‘Lookit’, ‘Mrs P.’) and asking for compliments on their classroom work constituted a substantial part of novices’ language use during classroom seatwork. Such interactional moves also entailed a nonverbal element in that the children held out their papers when summoning the teacher. Further, in a study of a 5-year-old Moroccan girl’s L2 learning during her first year in an Italian kindergarten, Pallotti (1996) demonstrated that the novice initially learned and used words (‘Look’, ‘teacher’, proper names) that allowed her to become accepted as a communicative partner in this multiparty kindergarten setting. Yet although studies on children’s early L2 learning show that summoning forms a crucial interactional resource, we know surprisingly little about the interactional organization of such practices as they are produced within the dynamic flow of classroom encounters.
EMBODIMENT IN L2 SPEAKER TALK C. Goodwin argues that in order to properly understand how people manage their face-to-face encounters, we must take into account the multiple
ASTA CEKAITE
29
semiotic systems on the basis of which interaction is designed (C. Goodwin 2003). Multimodality is fundamental for the organization of social interactions, and talk gains its meaning as a social action in conjunction with embodied features, encompassing activities, sequential structures, and the material structure in the surround (Kendon 1990; C. Goodwin 2003). Although van Lier (2000) suggested that nonverbal aspects of action comprise significant interactional resources for L2 novices at an early stage of learning, it is fair to say that work on multimodal aspects of L2 interactions is still an emergent area of research. In general, second language acquisition (SLA) research has been preoccupied with verbal aspects of language learning. Lately, however, several microanalytic studies have focused on teacher’s gestures and other nonverbal behaviour as significant forms of input in L2 classrooms (e.g. Allen 2000; Lazaraton 2004). Several studies have explored the role of nonverbal aspects, primarily gestures, in L2 learners’ speech. However, they have mostly dealt with dyadic interactions between native and non-native speakers (McCafferty 2002) and story retelling tasks (Gullberg 1998; McCafferty 1998) rather than complex multiparty classroom environments. Some of the few studies on L2 novices’ small group interactions demonstrate the intricate ways in which adult L2 learners employ verbal and nonverbal resources in interaction (Carroll 2004; Olsher 2004). For instance, Olsher (2004) explored adult learners’ embodied practices evolving during small group project work. He demonstrated how the novices completed the (initially verbal) turn with gestures and embodied displays. In conversations between adult L2 and L1 speakers, such embodied completions provide opportunities for language learning because they allow L1 speakers to reformulate what the L2 speaker has said with a more elaborate linguistic expression (Mori and Hayashi 2006). As language use is inextricably interwoven with embodied aspects of action, it is important to consider the multimodal resources deployed by the participants in relation to talk. That is, in addition to examining L2 novices’ summons sequences as segments of talk alone, this study also places the locus of interest on embodied accomplishment of these communicative practices (cf. Hatch 1978 on the importance of visual aspects in the analysis of learner–expert interactions). Such an approach may provide insights into how L2 novices deploy a broad range of resources in order to facilitate interaction and elucidate how learners and teachers organize and coordinate their participation in the dynamic classroom setting. In line with studies that investigate ‘the social construction’ of language classroom talk (cf. Markee 2004: 583; see also Hall and Verplaetse 2000), the present study presents an in-depth analysis of L2 novices’ interactional work when summoning the teacher during individual seatwork. By attending to the sequential organization of the participants’ actions, I will explore how L2 novices deploy summons–answer sequences in pursuit of their teachers’ attention. More specifically, the lexical shape of the summons will be
30
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
examined in conjunction with the multiple semiotic resources exploited by the participants, including body posture, gestures, and the artefacts provided by the surroundings. Importantly, when located in the interactional ecology of the classroom, summoning sequences can be defined as a type of interactional routine involving language use as an enactment of specific social relations (Peters and Boggs 1986: 81). By highlighting the recipient design (Sacks et al. 1974) of children’s summons, I will explicate how children’s ways of designing such calls display their sensitivity to the social ecological demands of classroom discourse. By demonstrating how complex the ability to recruit the teacher’s attention is, the present paper tentatively suggests that summoning (in and of itself) may provide a context for the development of interactional skills. In the dynamic multiparty classroom setting, L2 novices face several tasks when summoning the teacher, including (i) establishing attention and (ii) indicating what they want to interact about, that is, achieving a shared understanding of specific actions. Moreover, students have restricted rights to initiate talk in classroom settings, as well as restricted rights to choose topics; they also face the task of securing the teacher’s uptake. So, how do they solicit and establish the teacher’s attention?
METHOD Methodologically, the present study adopts conversation analysis’ (CA) detailed attention to interactional processes explored through participants’ sensemaking orientations on a turn-by-turn basis (Sacks et al. 1974; Lerner 2004). By attending to the sequential organization of talk, it explores the resources and methods participants rely on in the accomplishment of talk-in-interaction. The analysis also incorporates from language socialization a concern with participants’ use of indexical resources for the enactment of specific communicative roles and social relations (cf. Ochs 1988, 1996; Garrett and Baquedano-Lopez 2002). Integrating microanalyses with ethnographic observations allows us to more fully analyse children’s interactional contributions, which in many cases are comprehensible only in the light of the classroomspecific interactional arrangements (e.g. Kasper and Rose 2002).
Setting The analyses presented in this paper are based on video recordings and ethnographic observations of everyday interactions in a Swedish immersion class for refugee and immigrant children—a so-called ‘mottagningsgrupp’ (‘reception classroom’) in a Swedish school. ‘Reception’ classes have several aims, such as introducing the children to the Swedish language, and preparing them for transfer to regular classes, that is, introducing them to the Swedish educational system. The present group included children in grades 1–3 (7–10 years). They represented several languages: Arabic,
ASTA CEKAITE
31
Kurdish, and Thai. All children in this class were beginner learners who had recently arrived in Sweden. However, the children’s time spent in Sweden and the time spent in this class varied, as did the children’s L2 proficiency. Swedish was the language of instruction as well as the language taught, in that the teachers Vera (a native Swede, not present in the current episodes) and Fare (a Swedish-Arabic bilingual) taught Swedish vocabulary and reading, as well as maths and other school subjects. Fare used Arabic (as a language of instruction) only when the Arabic-speaking children experienced repeated problems in understanding classroom activities.
Data and recordings The present study is part of a larger investigation of children’s participation and L2 learning in an immersion classroom, where the children’s classroom interactions were video-recorded during three periods, covering an early (autumn), middle (winter), and late phase (spring) of the school year (90 hours of recordings). The data for this study are recordings from the middle phase (winter period) of the school year.1 The present study focuses on two girls, Fusi (a Kurdish 7-year-old) and Nok (a 7-year-old from Thailand), who were some of the latest arrivals in the class. During the mid-period of data collection, they had spent between 4 and 5 months in this class and were the least proficient in Swedish. Fusi also mastered some basic Arabic and used it occasionally when addressing Fare or Arabic-speaking classmates (girls). At the time, the other children had spent up to one year in the classroom and were more proficient in Swedish. Although Fusi’s and Nok’s Swedish skills were (still) very limited, they were actively engaged in individual work on assignments (instrumental classroom activities where talk was just one mode of participating, e.g. Goffman 1963).2
Interactional organization of individual work on assignments This study presents analyses of the girls’ participation during individual work on an assignment. During these classroom activities, the children usually worked on their own. They were seated in a half-circle, while the teacher walked around assisting them. Hand-raising was rare (Figure 1). The students were usually involved in many separate activities (doing exercises at their own skill level in, for instance, maths, writing, drawing), and the teacher was engaged in multiple tasks with multiple students. The students needed continuous supervision from the teacher in order to get instructions on a new task, to get an evaluation of completed work or to get help on work in progress. Because the teacher was continuously assisting the children, there were generally no ‘empty’ conversational slots during which the teacher was ‘free’ from involvement with other students. Therefore, it was often the case that several students were soliciting the teacher’s attention simultaneously, and they were often facing the task of either getting the teacher’s attention
32
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
Layla
Miran
Ali
Ahmed
Rana Whiteboard
Nok Hiwa Sawan
Fusi
Figure 1: Classroom layout by disrupting the teacher’s current activity or having to wait for his/her postponed involvement.
Analytic unit The analytic units are students’ summonses of the teacher. They have been analysed in relation to the children’s communicative projects. The term communicative project is meant to include not only the verbally invoked topic related to the summons, but also a range of visible displays (e.g. of classroom artefacts) that invoked the reason for the call. A communicative project is defined as ‘the task it is designed to solve, and/or in fact actually solves’ (Linell 1998: 220). Communicative projects are inherently dialogical in the sense of being jointly produced in an interactional context by the participants (Linell 1998). In the present analyses, the communicative projects were identified through post-hoc analysis of the students’ summonses. Thus, in the context of individual seatwork, the students’ summoning can be related to distinct communicative projects, such as calling for assistance or inviting evaluations.
SOLICITING TEACHER ATTENTION Summons sequences and verbal contributions among more advanced learners In order to outline a more complete picture of the interactional ecology of the present classroom, I will initially illustrate the methods that the more advanced students in the classroom relied upon when calling for assistance from the teacher. At the time, they displayed a broad repertoire of
ASTA CEKAITE
33
interactional resources in that they could produce rather elaborate linguistically formed turns; they also participated in conversation-based classroom activities. When soliciting the teacher’s attention, they recurrently introduced their communicative project verbally, as a request for action on the teacher’s part.
Excerpt 1. Individual work on maths. Teacher Fare, class, including the girls Layla and Nok. Fare is standing close to Nok and is helping her.
Layla (an Arabic speaker) calls for the teacher, who is involved in talking to a nearby child, Nok (line 3). Layla employs lexical means to perform several actions within a single turn. She (i) specifies the recipient of the utterance, the teacher; (ii) calls for his attention; (iii) indicates that she is experiencing difficulties in her work on the assignment, identifying why she needs the teacher’s assistance, that is, providing a reason for the summons (cf. C. Goodwin et al. 2002). Note that the teacher does not stop talking to Nok (moreover, he does not even shift his gaze from Nok to Layla). Thus, despite the overlapping talk and the teacher’s gaze towards another student, Layla succeeds in securing his assistance, in that he (i) moves to Layla and (ii) answers her request by immediately providing instructions on the task (‘take the first’ in line 7). In this way, Layla and other relatively advanced learners performed several actions within the same turn using verbal means, indicating the communicative project at issue in their initiating move; they did not need to first secure the teacher’s visual orientation. Thus, similar to the L1 primary grade students in Merrit’s study (1982), the advanced students in the present study could verbally introduce the reason for their summons, thereby making the teacher accountable for providing assistance. Naturally, this is not to say that their calls for the teacher’s
34
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
attention lacked embodied characteristics (see line 3, Layla’s gaze at the book)3 but rather that the analysis above aimed at highlighting the verbal means that the students used to solicit the teacher’s attention.
Visual displays of classroom artefacts in novices’ summons–answer sequences In contrast to the more advanced students in Excerpt 1, the L2 novices in this classroom did not have elaborate verbal resources at their disposal. In the following, I will present the ways in which they solicited the teacher’s attention through their summons turns. Excerpt 2 illustrates how an L2 novice deploys a summons when addressing the teacher during individual maths work. In particular, I will focus on how a visual display of a classroom artefact, the book, makes relevant a specific communicative project, that of securing the teacher’s assessment.
Excerpt 2. Individual work on maths. Teacher Fare, class, including Fusi and the boy Sawan. Fare assists Sawan at his desk.
The girl Fusi summons the teacher Fare, using a vocative (line 1). His response ‘jepp’ (‘yep’) indicates that he is waiting for Fusi’s forthcoming action, but he is still looking at the book of another student. Fusi’s directive to look ‘shuffi haie’ (‘look at this’ in Arabic) invites him to scrutinize her book, which she now holds raised towards him (line 3). Because of her limited skills in Arabic, Fusi wrongly uses a feminine verb form ‘shuffi’ to address the male teacher. The correct (masculine) form is ‘shuff’. Although grammatically incorrect, Fusi’s summons combined with her visual display of the book provides information on her communicative project and guides the teacher’s interpretation of her actions (lines 4 and 5). By looking at her book and producing an evaluation of her work (line 4), the teacher is
ASTA CEKAITE
35
orienting to Fusi’s communicative project. This interpretation is supported in that Fusi in fact packs her books, after obtaining his positive assessment. The (deictic) display of the specific page with the completed exercise constitutes an efficient ‘meaning producing system’ (C. Goodwin 2003: 228) and provides a way of precisely indicating the communicative project, that is, inviting an evaluation of work completed. That is, Fusi’s actions make relevant the contextually bound routines that together enable the teacher to ‘read’ and act upon her communicative project (asking for evaluation). Although the novices’ L2 vocabulary was still very limited, the teachers recurrently easily recognized what was apparently their communicative project. The participants’ shared knowledge of the encompassing activity (individual work on task) constituted the broad frame of interpretation for the visual displays of the classroom artefacts. Thus, a summons–answer sequence allowed the novices to recruit the teacher as a conversational partner in that (i) it committed the teacher to subsequent interaction, and (ii) it allowed for embodied demonstrations of the (significant) classroom artefacts. Such a summons sequence involves a multiparty performance of actions, organized and sustained within the ‘frameworks of attentiveness’ (M. Goodwin 1998: 39).4 Instead of directly telling the teacher why they need his/her assistance, the L2 novices first need to establish the teacher’s visual orientation (so as to be able to indicate their interactional concerns at hand).
Upgrading summonses During individual seatwork, the teacher was multitasking and several students would simultaneously call for assistance, thus summons turns were often unavoidably ignored. Although the classroom activities in this study were based on a student-centred approach, the student’s conversational rights were limited in terms of the topics to be brought up, the manner of speaking, and the timing of their contributions. Recruiting the teacher’s attention within the multiple voices of a classroom was usually a time-consuming activity that involved a considerable amount of interactional work. Although the L2 novices did not yet have elaborate lexical resources, the analysis demonstrated that through prosody, gesture, body posture and position, and the lexico-semantic features, their summons turns were highlighted, upgraded, and differentiated. The children upgraded their summonses in a number of ways, for example by moving towards the teacher (ambulatory design) and by combining their summons with displays of affect.
‘Ambulatory’ design of a summons Because the participants needed access to each other’s visual field, the novices could make their summons attempts visible by using what I will call
36
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
an ambulatory design of a summons, which simply entailed moving into the physical proximity of the teacher.
Excerpt 3. Individual writing exercises. Teacher Fare, temporary teacher Martin, class, including the girls Fusi and Layla. Fare assists Layla at her desk.
When Fusi has finished her exercise, she issues an announcement concerning her progress with the work and directs it to Fare (line 1). Fare, however, is involved with some other children at the other end of the classroom and does not respond to her. Fusi then leaves her desk, grabs the teacher Martin (a temporary replacement) who is passing by, stops him and puts her writing book in front of him, displaying her finished exercise. By approaching the teacher, she actively constructs a state of ‘copresence’ (Goffman 1963: 17). Although she does not provide any verbal indication as to what is expected from the teacher when she approaches him, she successfully obtains his evaluation of her exercise.5
Affective design of a summons In producing and upgrading a summons, a student calling the teacher took up different affective stances (cf. Ochs 1996). Affective stances are important aspects of language use and language socialization, in that they are part of what constitutes interactional competence, in this case, students’ ‘appropriate’ classroom behaviour. Affect permeates different layers of human interaction, and even ‘novices are expected to recognize and to display emotions in culturally defined ways and according to local norms and preferences’ (Garrett and Baquedano-Lopez 2002: 352).6 While affect can be indexed at all levels of language—grammar, prosody, lexical, and conversational structures (e.g. Besnier 1990; Caffi and Janney 1994; Gu¨nthner 1997; Ochs and Schieffelin 1989)—it is also located within embodied sequences of action (Goodwin and Goodwin 2000).
ASTA CEKAITE
37
In the present classroom, the students recurrently employed gestures, body posture, and body position in arranging affective displays. Such bodily displays (e.g. half-lying on the desk, turned to the teacher, leaning away from the book, ambulatory design of actions) became meaningful in relation to situated classroom artefacts (books, desk). Repeatedly, novices and the more advanced students alike emphatically marked their summoning turns. In the conversational context of summoning, increased volume, accentuated and markedly prolonged vowels, and shifting the position of stress to the final prolonged vowel were recurrent features of prosody displaying the emphatically marked claims for the teacher’s co-orientation. While there are no prosodic features that can be straightforwardly associated with specific affective stances, an interactional approach to prosody has shown how some specific features tend to be recurrently employed in particular sequential and situational environments (e.g. Selting 1994; Local 1996) and can be seen to display ‘iconicity with the affective referents’ (Freese and Maynard 1998: 198)7. The present students’ emphatically marked summons turns carried a distinctive prosodic shape, marked by abrupt pitch leaps (high pitch) and a rising–falling pitch movement on a lengthened syllable (employed on key words, e.g. vocatives). Also, shifts in prosodic cues in relation to the prior turns as well as the surrounding talk in the classroom provided critical devices in the prosodic indexing of affectively charged summons turns. Such actions (prosodic indexing of affective stances) were significant interactional resources in that they audibly indexed an affective stance and could alert the teacher even before the teacher’s visual orientation was established (e.g. Excerpt 4). In the following example, I will discuss how multimodal resources are configured so as to bear on the interaction, by allowing the novice to display an affective stance as a ‘public and witnessable’ method for highlighting her claims for the teacher’s attention and assistance.
Excerpt 4. Individual work on maths. Teacher Fare, class, including Nok and the boy Hiwa. Fare is assisting Hiwa.
Nok summons Fare with an elongated (singing) vocative, Fare:::, in an intonation that can be glossed as ‘resignation’. She is also lying on her desk,
38
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
looking at him, thereby ‘doing being ignored’ (e.g. Sacks 1992, vol. I). The prosodic indexing of ‘resignation’ (‘creaky’ voice and elongated final vowel) can be seen to indicate that this student is in trouble and needs help. The affective display is audible before the teacher has turned towards Nok, highlighting her summons. It is embellished by the ‘reclining’ body posture that can also be seen to indicate resignation. Nok’s affective display is coordinated with her simultaneous orientation to the classroom artefact, more specifically, her task. Through her body posture, Nok foregrounds the page with the problematic exercise while covering the irrelevant (left) page of the book, thus indicating that the teacher’s help is needed on the ongoing work (on the right page). Fare is turned away from Nok, involved in assisting Hiwa. While he turns to Nok, he produces a summons response (‘mh’), directing his gaze at Nok and her exercise, which is displayed on her desk. He then provides instructions and encourages her to continue to work on a particular ‘problematic’ exercise (line 4). Thereby, Nok’s visual display of the book provides for an interactionally economic performance in that it directly ‘catches’ the teacher’s eye and, together with the affective display, invokes the reason for the teacher’s assistance.8 As can be seen in Excerpt 4, resources such as prosodic cues, in coordination with body posture, and orientation to classroom artefacts were employed in constructing affective stance. The summons turn was tailored as affectively charged action that provided cues to the teacher concerning the reason for the student’s call for the teacher. Thus, when situated in the encompassing activity, the affective stances co-created the interpretative framework of the action. In this case, it was successful, in that the teacher, Fare, responded to Nok’s communicative project (line 4), even though she had not stated it in words. (Excerpt 5 and analysis can be found in the online version of this article, available to online subscribers at http://applij. oxfordjournals.org/.)
ASTA CEKAITE
39
Let us consider the participants’ interactional work in lines 7 and 8 in more detail:
Line .......... indicates Fusi’s pointing at the book; line ______ indicates Fare’s shifts in gaze towards Sawan (S) and Fusi (F).
Employing contrastive affective stances in pursuit of teacher’s attention As demonstrated in Excerpt 4, the affectively indexed summonses were not random displays of emotion. Instead, they provided the students with resources to take up different types of footing, or projected selves (Goffman 1981: 128), displaying their alignment with respect to the current action (summons). The way in which the present students shaped their summonses as emotionally charged action indicates their thorough work on recipient design (Sacks et al. 1974) and their delicate attunement to the institutional order of the classroom.9 Hence, the girls were able to recognize and deploy institutionally relevant ways for indexing social identities and relations (Ochs 1996; Antaki and Widdicombe 1998; Aronsson 1998). In the present classroom, affectively charged summons turns appealed to the teacher as a person responsible for the students’ classroom work, conduct, and emotional status. As Sacks (1992, vol. I: 256) has pointed out, an announcement of a relevant trouble to the other can serve as a proper beginning to a conversation when a conversationalist’s rights to talk to the other are somehow restricted. Affectively charged summonses provided interactional procedures for constructing locally relevant student identities interpretable as ‘needy’, ‘irritated’, ‘demanding’, ‘frustrated’, or ‘resigned’. They oriented to the teacher’s responsibility to supervise classroom work, thereby making him accountable for responding.
40
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
Excerpt 6. Individual work on maths. Teacher Fare, researcher R, class, including the girls Nok, Layla,10 and the boy Hiwa. Fare is helping Nok at her desk.
ASTA CEKAITE
41
Nok has been doing a mathematics exercise (adding 5 to a preceding number). The teacher Fare, who has been assisting Nok, moves to another student, Layla (line 2). Nok comes up with an answer and writes it down (line 3), and she then summons Fare (lines 4–5). Her summons is embellished with a prosodic marking (rising tone), emphatically marked loudness (‘Fa"RE:’) (Selting 1994), a prolonged vowel, an additional loud outbreath, and a nonlexical response cry ‘o::hhh’ (Goffman 1981: 97). Nok’s actions, gliding down from the chair and raising herself with difficulty, are finely coordinated with her facial expression of a ‘tired’, ‘exhausted’ person, and with her response cry (line 5). In that Nok’s affectively indexed summons appears in the sequential context of her writing an answer to an adding operation, her multimodal performance makes her actions readable as those of a somewhat frustrated student who experiences trouble in her work on the current task and needs assistance. Upon returning to a seated position, Nok laughs and looks at the researcher. Her laughter and collusive gaze towards the researcher (line 6) indicate Nok’s somewhat playful attitude towards her immediately previous displays of the ‘exhausted student’ (in lines 4–5). Note that there is no uptake from Fare, who is still engaged in work with another student, Layla (line 7). Nok addresses the researcher in a rather loud collusive side-play ‘och tjugi tie’ (‘and twenty ten’) (line 8). Although this is not directed at the teacher, it is interpretable as a humorous comment on her task that flouts the task conditions, adding five to a preceding number (20þ5; 20þ10), and it is still formatted as ‘on-task’ action. Generally, the teachers responded to and corrected the students’ erroneous contributions. The ‘incorrectness’ of Nok’s loud comment can be seen to work on the conditions that the teacher will correct, that is, respond to such an action. Nok then turns to the teacher and produces her second recycled summons ‘Fare:!’, followed by nonsense talk (‘pepepe’) in a low volume (line 10). Her summons is marked by a prolonged final syllable and exclamative intonation, and presents a shift in affective stance (now to a serious summons mode). Thus, once again, while calling the teacher, Nok indexes her summons as an indignant request for the teacher’s attention, on the part of a student who is rightfully entitled to the teacher’s assistance. Nok also starts moving closer to the teacher using an ambulatory design in her actions, thereby upgrading her attempts to summon the teacher. Her summons (line 11) presents a contrast in affective stance in that it is marked
42
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
by a playful mode, evident in the singing intonation and the metric pattern of the repetitive use of ‘little Fale’. In fact, the lexical design ‘lilla Fale’ (‘little Fale’) can be interpreted as an expression of inverted social relations, a tease, as the diminutive formula ‘little þ name’ was quite commonly used by the teachers to address the children. Clearly, Nok’s address is packaged as a humorous and teasing sort of action. The playful affective stance co-occurs with Nok leaving her desk and can be seen to modulate the ‘inappropriate’ character of her actions, that is, leaving her desk without permission. Simultaneously, the lexical design can be seen to highlight the student’s summons attempts in that it makes the teacher accountable for responding to her teasing. Nok again summons Fare (line 13). She elaborates her performance, playfully positioning her upper body and head on the desk, close to the teacher, and speaking in a playful key. She also points at him with a pencil. At this moment, he turns to Nok, their gazes meet, and the teacher finally leaves Layla’s desk. Nok moves back to her desk and waits for him (line 15). Finally, Fare looks at her book and instantly addresses the problematic issue in her exercise, that is, the adding operation (25þ5; line 16). The prosodic characteristics of the teacher’s turn (emphatic marking and voice quality) display an affective stance that makes his utterance ‘witnessable’ as a disciplining move, aimed at re-establishing the classroom order. Thereby he does not simply respond to Nok’s summons as a request for assistance but also to her playful affective displays. An important point demonstrated here is that whereas most of Nok’s utterances were summonses of relatively similar semantic content, they were differentiated in a progressive fashion through a range of contrasting affective stances when initial summoning attempts did not achieve success. While she employed a range of paralinguistic cues when highlighting her summonses, she was also actively rearranging her bodily position in the classroom space, thereby making her ‘non-working’ identity publicly witnessable. The ‘nonworking’ display was invoked by arranging body posture (and position) with respect to the workbook: while leaning away from the desk, gaze averted from the workbook, gliding down from the chair; or turned in body torque towards the teacher, moving closer to the teacher in an ambulatory design of summons (see also Excerpt 2). I want to argue that such multimodally structured summonses are based on some crucial knowledge of the classroom, namely, that if the student does not work, the teacher will— or at least will be expected to—deal with that. Thus, the different keyings of Nok’s affective stances are not unmotivated; rather they accomplish different interactional tasks. Nok shifts the affective indexing of her actions with regard to (i) the addressee (‘frustrated’, ‘resigned’ affective stance towards the teacher, whose help she is entitled to) and ‘playful’ towards the researcher (an observer, who has no responsibilities or rights to assist the student in classroom work), and (ii) the student’s own appropriate or inappropriate classroom conduct: calling the teacher from her own desk (‘resigned’ stance),
ASTA CEKAITE
43
or while approaching the teacher (playful keying of summons). Hence, Nok’s differentiated summonses located the teacher in a web of different institutional responsibilities: answering a ‘resigned’ student in need of assistance, or answering a playful action teasing the teacher’s authority. Such embodied performances served as powerful devices in pursuit of the teachers’ uptake in the socially and materially sedimented space of the local classroom setting (cf. M. Goodwin 1998) and were crucial for the local constitution and coordination of participants’ actions.
CONCLUDING DISCUSSION As demonstrated, a simple structure of summoning provided a handy method for soliciting and establishing the teacher’s attention and facilitated the novices’ participation in classroom activities from early on. Importantly, though, the local design of the summonses was influenced by the competitive multiparty classroom setting. A detailed analysis of multiple modalities in the summons turns demonstrates the intricate work in which the L2 novices engaged in order to achieve the communicatively loaded content of the summons. Thus the classroom was not a neutral setting, unaffected by (tacit) norms guiding the participants’ actions. As can be seen, both novices and more advanced students oriented to similar requirements concerning what constituted an appropriate reason for calling the teacher. In the local institutional order, they designed their initiating moves as displays of ‘trouble relevant for the other’ (Sacks 1992, vol. I: 256), orienting to the teacher’s responsibilities to help and assist the children. Such moves served as relevant warrants to initiate interaction and created additional incentives for the teacher to assist the student. The use of summonses as a method for engaging others in talk relied greatly on novices’ deployment of affective stances as well as their use of classroom artefacts. When highlighting and upgrading their summons, they deployed a multimodally structured performance. In conjunction with the visual displays of classroom artefacts, affective stances were employed as parallel ways of spelling out the communicative project (for instance, appealing to the teacher’s responsibilities to assist students who were experiencing difficulties with their work on an assignment). Also, summons turns could be designed as affectively charged action that invoked and exploited these institutional responsibilities. For instance, Nok’s publicly recognizable display of ‘not working’ (Excerpt 6) traded on the teacher’s responsibility to keep the students at work. That is to say, it spelled out the summons turn as ‘trouble for the other’ and invited the teacher’s response. Within the local classroom community, affective stances were accomplished through prosody, simple forms of grammatical structure, and embodied action, rather than ‘emotive vocabulary’ (Wierzbicka 1999). The nonverbal devices included prosody (pitch, loudness, tempo, etc.),
44
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
paralinguistic characteristics, such as voice quality (smiley, creaky, tense voice), and body posture, movements and body positioning with respect to classroom artefacts (e.g. half-lying on the desk, gaze averted from the book). Such nonverbal resources were artfully used in conveying a range of affective stances, casting students as ostensibly ‘resigned’, ‘eager’, ‘bored’, ‘needy’, ‘challenging’, ‘indignant’, or ‘witty’. These affect displays simultaneously revealed the children’s orientation to the local classroom ethos and their intricate interactional work on recipient design in pursuing the teacher’s attention in the often fragmented and busy classroom interaction. As such, displays of affect comprised an integral feature of classroom interactions. For the L2 novices, the performative displays of affect supplemented the children’s limited lexical repertoire by establishing a framework for interpreting their interactional moves and provided embodied cues as to how one might construe a subsequent interactional move. Here, it is important to underscore the teachers’ readiness to make sense of minimal verbal or visual cues, to work out the communicative project invoked by the student, and to draw on the shared interactional history of the classroom. Achievement of involvement and participation in classroom activities, therefore, relies on participants’ employment of socially distributed methods for the production of accountable actions, including embodied action. Although the present study did not explicitly focus on L2 development, it may provide ground for considerations regarding potential implications of summons for language learning, more specifically, learners’ development of interactional competencies. As noted earlier, being able to recruit the participation of the ‘expert’ and direct the teacher towards specific interactional tasks is one of the basic conditions for gaining access to the ‘linguaculture’ of the classroom (e.g. Willet 1995; Toohey 2000). As the close scrutiny of children’s summonses demonstrated, these seemingly simple discursive structures involve rather complex communicative abilities. While summoning can be accomplished by employing resources that are already available even for language novices, these resources need to be configured so as to fit into the interactional ecology of the institutional setting (e.g. Ochs 1996; Garrett and Baquedano Lopez 2002). Language learners, thereby, need to display affective stances and identities in institutionally ratified ways. These interactional procedures, however, are not taught explicitly. The novices therefore (may) need to discover and appropriate these local procedures through observation and participation in the everyday classroom activities (e.g. Rogoff 1990; Ohta 2000). Summoning, thereby, may in and of itself provide a learning context. If we see language (including interactional competencies) as evolving from specialized ways of solving interactional problems (Hatch 1978; Bruner 1981), summons sequences may provide a fundamental framework for students’ initial L2 learning. Final version received June 2007
ASTA CEKAITE
45
SUPPLEMENTARY DATA Supplementary material mentioned in the article is available to online subscribers at the journal website www.applij.oxfordjournals.org
ACKNOWLEDGEMENTS Thanks are due to Karin Aronsson, Jakob Cromdal, Ann-Carita Evaldsson, Gabriele Kasper, and three anonymous reviewers for their helpful comments on an earlier draft of this article. Financial support from the Swedish Research Council and the Bank of Sweden Tercentenary Foundation is gratefully acknowledged.
TRANSCRIPTION KEY : [] (.) (2) YES x (xx) what jala
(()) 45 54 ? . ¼ Fare heh "#
prolonged syllable demarcates overlapping utterances micropause, i.e. shorter than (0.5) numbers in single parentheses represent pauses in seconds relatively high amplitude inaudible word unsure transcription translation into English word in Arabic denotes speech in low volume further comments of the transcriber quicker pace than surrounding talk slower pace than surrounding talk denotes rising terminal intonation indicates falling terminal intonation denotes latching between utterances sounds marked by emphatic stress are underlined indicates laughter rising, falling intonation
NOTES 1 Thus, this study does not entail a developmental focus. On the results of the longitudinal study, see Cekaite (2007). 2 Individual work on assignments provided a locus for the girls’ very early active participation in the classroom, documented already during the first period of recordings, see Cekaite (2007). 3 Szymanski (1999), for instance, has demonstrated how, during group work, the students relied on both
verbal and visual characteristics of actions to alert the others to the upcoming talk. 4 From a somewhat different setting, C. Goodwin et al. (2002) demonstrated that a person with aphasia employed the summons as a systematic procedure to establish the recipient’s visual orientation. When visual orientation was established, pointing, gestures, and embodied demonstrations of artefacts were deployed as meaningful interactional resources in collaborative meaning making.
46
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
5 Although in this case Fusi achieved an evaluation of her work without being disciplined, the children were usually not allowed to leave their desks to approach the teacher. Instead, they had to comply with the norms of appropriate classroom behaviour and solicit the teacher’s assistance from their own desks. It is notable that Martin was a substitute teacher present in the classroom just for a couple of days, which is probably why he did not discipline Fusi. Simultaneously, his ‘reading’ of why his attention was needed demonstrates that even a novice teacher in the classroom orients to the book display as a locally meaningful artefact. 6 Different language communities and communities of practice may have certain commonalities ‘in the linguistic means used to constitute’ certain situational meanings, including affective stances (Ochs 1996: 425). These common features in the linguistic structuring of social activities may present a common ground for socialization for different categories of novices (Ochs 1996: 428). However, ‘these commonalities do not necessarily imply that the full social meanings of particular stances or particular acts are shared across communities’ (Ochs 1996: 426). Novices (including L2 learners) need to recognize, develop, and learn to deploy the distinctly local ways of indexing social situations (according to the expectations of the local community of practice).
7 Participants’ heightened emotive involvement is, for instance, indexed through an emphatic speech style, which is constituted through shifts in prosodic markers in relation to prior turns of talk (Selting 1994). Affective intensity may be indexed through modulation of volume, vowel lengthening, and code-switching between registers (Ochs 1996: 427). 8 Here, Nok pre-arranges the prospective locus of mutual attention during the ongoing summons, before the teacher indicates he has committed himself to participating in the encounter. Simultaneous displays of classroom artefacts can be seen as a method that provides early (visual) indications of the communicative project. 9 Ochs (1988), in her study on Samoan children’s primary socialization, demonstrated that even young children are sensitive to the social organization of the group with respect to participants’ mutual rights and responsibilities, and, consequently, communicative roles in child–adult interactions. The children designed their requests to caregivers by employing affective markers rhetorically, for instance, they used pronouns expressing sympathy for oneself in order to elicit sympathy from the adult (1988: 187). 10 Rana, who was usually sitting between Nok and Layla, is not present in the current episode.
REFERENCES Allen, L. Q. 2000. ‘Nonverbal accommodations in foreign language teacher talk,’ Applied Language Learning 11: 155–76. Antaki, C. and S. Widdicombe (eds). 1998. Identities in Talk. London: Sage. Aronsson, K. 1998. ‘Identity-in-interaction and social ethnography,’ Research on Language and Social Interaction 31/1: 75–89.
Besnier, N. 1990. ‘Language and affect,’ Annual Review of Anthropology 19: 419–51. Breen, M. 2001. ‘Overt participation and covert acquisition in the language classroom’ in M. Breen (ed.): Learner Contributions to Language Learning. Longman: Pearson Education. Bruner, J. 1981. ‘The social context of language acquisition,’ Language and Communication 1/2–3: 155–78.
ASTA CEKAITE
Caffi, C. and R. W. Janney. 1994. ‘Toward a pragmatics of emotive communication,’ Journal of Pragmatics 22/3–4: 325–73. Carroll, D. 2004. ‘Restarts in novice turn beginnings: Disfluencies or interactional achievements?’ in R. Gardner and J. Wagner (eds): Second Language Conversations. London: Continuum. Cathcart, R. 1986. ‘Situational differences and the sampling of young children’s school language’ in R. Day (ed.): Talking to Learn: Conversation in Second Language Acquisition. Rowley: Newbury House. Cathcart-Strong, R. 1986. ‘Input generation by young second language learners,’ TESOL Quarterly 20/3: 515–30. Cekaite, A. 2007. ‘A child’s development of an interactional competence in a Swedish L2 classroom,’ The Modern Language Journal 91/1: 45–62. Day, E. 2002. Identity and the Young English Language Learner. Clevedon: Multilingual Matters. Freebody, P. and J. Freiberg. 2000. ‘Public and pedagogic morality: The local orders of instructional and regulatory talk in classrooms’ in S. Hester and D. Francis (eds): Local Educational Order. Amsterdam: John Benjamins. Freese, J. and W. D. Maynard. 1998. ‘Prosodic features of bad news and good news in conversation,’ Language in Society 27/2: 195–219. Garrett, P. and P. Baquedano-Lo´pez. 2002. ‘Language socialization: Reproduction and continuity, transformation and change,’ Annual Review of Anthropology 31: 339–61. Goffman, E. 1963. Behaviour in Public Places: Notes on the Social Organization of Gatherings. New York: Free Press. Goffman, E. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Goodwin, C. 2003. ‘Pointing as situated practice’ in S. Kita (ed.): Pointing: Where Language, Culture, and Cognition Meet. Mahwah, NJ: Lawrence Erlbaum. Goodwin, C., M. H. Goodwin, and D. Olsher 2002. ‘Producing sense with nonsense syllables: Turn and sequence in conversations with a man with severe aphasia’ in C. Ford, B. Fox, and S. Thompson (eds): The Language of Turn and Sequence. Oxford: Oxford University Press. Goodwin, M. H. 1998. ‘Games of stance: Conflict and footing in hopscotch’ in S. Hoyle and S. Temple Adger (eds): Kids’ Talk: Strategic
47
Language Use in Later Childhood. New York: Oxford University Press. Goodwin, M. H. and C. Goodwin. 2000. ‘Emotion within situated activity’ in N. Budwig, I. C. Uzgiris, and J. V. Wertsch (eds): Communication: An Arena of Development. Stamford: Ablex Publishing. Goodwin, M. H., C. Goodwin, and M. Yaeger-Dror. 2002. ‘Multi-modality in girls’ game disputes,’ Journal of Pragmatics 34/10–11: 1621–49. Gullberg, M. 1998. Gesture as Communication Strategy in Second Language Discourse. Lund, Sweden: Lund University Press (Unpublished PhD Dissertation). Gu¨nthner, S. 1997. ‘The contextualization of affect in reported dialogues’ in S. Niemeier and R. Dirven (eds): The Language of Emotions: Conceptualization, Expression, and Theoretical Foundation. Amsterdam: John Benjamins. Hall, J. K. and L. S. Verplaetse (eds). 2000. Second and Foreign Language Learning through Classroom Interaction. Mahwah, NJ: Lawrence Erlbaum. Hatch, E. 1978. ‘Discourse analysis and second language acquisition’ in E. Hatch (ed.): Second Language Acquisition. A Book of Readings. Rowley: Newbury House. Kasper, G. and K. R. Rose. 2002. Pragmatic Development in a Second Language. A supplement to Language Learning. Volume 52, supplement 1. Kendon, A. 1990. Conducting Interaction: Patterns of Behavior in Focused Encounters. Cambridge: Cambridge University Press. Krupa-Kwiatkowski, M. 1998. ‘ ‘‘You shouldn’t have brought me here!’’ Interaction strategies in the silent period of an inner-directed second language learner,’ Research on Language and Social Interaction 31: 133–75. Lazaraton, A. 2004. ‘Gesture and speech in the vocabulary explanations of one ESL Teacher: A microanalytic inquiry,’ Language Learning 54/1: 79–117. Lerner, G. (ed.). 2004. Conversation Analysis Studies from the First Generation. Amsterdam: John Benjamins. Linell, P. 1998. Approaching Dialogue: Talk, Interaction and Contexts in Dialogical Perspectives. Amsterdam: Benjamins. Linnakyla¨, P. 1980. ‘Hi superman: What is most functional English for a Finnish five-year-old,’ Journal of Pragmatics 4/4: 367–92.
48
SOLICITING TEACHER ATTENTION IN AN L2 CLASSROOM
Local, J. 1996. ‘Conversational phonetics: Some aspects of news receipts in everyday talk’ in E. Couper-Kuhlen and M. Selting (eds): Prosody in Conversation: Interactional Studies. Cambridge: Cambridge University Press. McCafferty, S. G. 1998. ‘Nonverbal expression and L2 private speech,’ Applied Linguistics 19/1: 73–96. McCafferty, S. G. 2002. ‘Gesture and creating zones of proximal development for second language learning,’ Modern Language Journal 2/2: 192–203. Markee, N. 2004. ‘Zones of interactional transition in ESL classes,’ Modern Language Journal 88/4: 583–96. Mehan, H. 1979. Learning Lessons: The Social Organization of Classroom Behavior. Cambridge, MA: Harvard University Press. Merritt, M. 1982. ‘Repeats and reformulations in primary classrooms as windows of the nature of talk engagement,’ Discourse Processes 5: 127–45. Mondada, L. and S. Pekarek Doehler. 2004. ‘Second language acquisition as situated practice: Task accomplishment in the French second language classroom,’ Modern Language Journal 88/4: 501–18. Mori, J. and M. Hayashi. 2006. ‘The achievement of intersubjectivity through embodied completions,’ Applied Linguistics 27/4: 195–219. Ochs, E. 1988. Culture and Language Development: Language Acquisition and Language Socialization in a Samoan Village. Cambridge: Cambridge University Press. Ochs, E. 1996. ‘Linguistic resources for socializing humanity’ in J. Gumperz and S. Levinson (eds): Rethinking Linguistic Relativity. Cambridge: Cambridge University Press. Ochs, E. and B. Schieffelin. 1989. ‘Language has a heart,’ Text 9/1: 7–25. Ohta, A. 2000. Second Language Acquisition Processes in the Classroom. Mahwah, NJ: Lawrence Erlbaum. Olsher, D. 2004. ‘Talk and gesture: The embodied completion of sequential actions in spoken interaction’ in R. Gardner and J. Wagner (eds): Second Language Conversations. London: Continuum. Pallotti, G. 1996. ‘Towards an ecology of second language acquisition: SLA as a socialization process’ in E. Kellerman, B. Weltens, and
T. Bongaerts (eds): Proceedings of EUROSLA 6, Toegepaste Taalwetenshap in Artikelen, No. 55. Peters, A. M. and S.T. Boggs. 1986. ‘Interactional routines as cultural influences upon language acquisition’ in B. Schieffelin and E. Ochs (eds): Language Socialization across Cultures. New York: Cambridge University Press. Rogoff, B. (1990). Apprenticeship in Thinking: Cognitive Development in Social Context. New York: Oxford University Press. Sacks, H. 1992. Lectures on Conversation: Volumes I and II (ed. G. Jefferson). Cambridge, MA: Blackwell. Sacks, H., E. A. Schegloff, and G. Jefferson. 1974. ‘A simplest systematics for the organization of turn-taking for conversation,’ Language 50: 696–735. Schegloff, E. A. 1968. ‘Sequencing in conversational openings,’ American Anthropologist 70/6: 1075–95. Schegloff, E. A. 1998. ‘Body torque,’ Social Research 65/3: 535–96. Seedhouse, P. 2004. The Interactional Architecture of the Language Classroom: A Conversation Analysis Perspective. Malden, MA: Blackwell. Selting, M. 1994. ‘Emphatic speech stylewith special focus on the prosodic signalling of heightened emotive involvement in conversation,’ Journal of Pragmatics 22/3–4: 375–408. Szymanski, M. H. 1999. ‘Re-engaging and dis-engaging talk in activity,’ Language in Society 28/1: 1–23. Toohey, K. 2000. Learning English at School: Identity, Social Relations and Classroom Practice. Clevedon: Multilingual Matters. van Dam, J. 2002. ‘Ritual, face, and play in a first English lesson: Bootstrapping a classroom culture’ in C. Kramsch (ed.): Language Acquisition and Language Socialization. London: Continuum. van Lier, L. 2000. ‘From input to affordance: Social-interactive learning from an ecological perspective’ in J. P. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press. Wierzbicka, A. 1999. ‘Defining emotion concepts,’ Cognitive Science 16/4: 539–81. Willett, J. 1995. ‘Becoming first graders in an L2: An ethnographic study of L2 socialization,’ TESOL Quarterly 29/3: 473–503.
Applied Linguistics 30/1: 49–69 ß Oxford University Press 2008 doi:10.1093/applin/amn033 Advance Access published on 29 September 2008
‘Lego my keego!’: An Analysis of Language Play in a Beginning Japanese as a Foreign Language Classroom CADE BUSHNELL University of Hawai’i at Ma¯noa In this article, I present an analysis of talk-in-interaction from an introductory Japanese as a foreign language classroom at an American university. An examination of the data revealed language play (LP) to be a highly salient feature of the participants’ interactions. LP has come into increasing focus in the second language acquisition research of the last decade. Research in L1 has long shown the prevalence of LP in both the language data available to the learner and learner language production (e.g. Garvey 1984, [1977] 1990), and recent research in L2 has shown that LP is also a prominent characteristic of the language production of both child and adult L2 learners (Kramsch and Sullivan 1996; Cook 1997, 2000, 2001; Lantolf 1997; Sullivan 2000; Tarone 2000; Broner and Tarone 2001; Belz 2002a, 2002b; Bell 2005; Cekaite and Aronsson 2005; Kim and Kellog 2007). Adopting Cook’s (2000) definition of LP, I use conversation analysis to examine instances of LP in the participants’ interactions. Analysis focuses specifically on the ways in which LP functions within the context of the language learning classroom to provide ‘affordances’ (van Lier 2000, 2004) for language learning, and to become a resource for sequenceorganization. The analysis shows that by and through the fictional world of LP, the participants were able to engage in the teacher-assigned pedagogical activities on their own terms. In the discussion, I argue that LP is potentially of great benefit to the linguistic development of second language learners— echoing Cekaite and Aronsson’s argument in favor of a ludic model of language learning, in which they contend that ‘we need to take non-serious language more seriously’ (2005: 169).
INTRODUCTION Cook (2000: 5) argues that play is highly beneficial to human development, and that language play (henceforth LP) in particular is important not only in child language acquisition, but in adult language learning as well. However, as Cook noted, a serious examination of LP had at that time ‘on the whole been neglected, or at least sidelined, in the study of language and language learning’ (2000: 4) and until the 2000s, second language researchers had paid relatively little attention to LP, a notable exception being Kramsch and Sullivan (1996). In the field of second language acquisition (SLA), where the dominant theoretical frameworks have tended to emphasize interaction
50
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
focusing on referential meaning (e.g. ‘task-based’ instruction; Long 1983, 1996; Doughty and Pica 1986; Skehan 1998; Pica 2005; but see, e.g., Block 2003, 2007a, 2007b on the recent ‘social turn’ in SLA research), this is not a surprising fact. However, research in L1 has long shown the prevalence of LP in both language data available to the child language learner, as well as learner language production (e.g. Garvey 1984, [1977] 1990). LP is also a salient feature of child L2 acquisition (Tarone 2000; Broner and Tarone 2001; Cekaite and Aronsson 2005; Kim and Kellogg 2007) and recent research on adult L2 acquisition has unequivocally shown it to be in no way a ‘childish’ activity (Kramsch and Sullivan 1996; Cook 1997, 2000, 2001; Lantolf 1997; Belz 2000a, 2000b; Sullivan 2000; Bell 2005). While previous research has often focused on the intrinsic features of LP (e.g. Cook 1997, 2000), or the possible effects of LP on interlanguage structures and second language acquisition (e.g. Tarone 2000), the present study will examine several of the socially situated functions of LP. Using conversation analysis, I examine naturally occurring linguistic data collected from a beginning Japanese as a foreign language classroom. In the analysis, I will show that the participants co-construct and use the fictional worlds of LP as a resource by which to organize the pedagogical-task-as-social-interaction. Moreover, I will argue that LP functions to offer ‘affordances’ (van Lier 2000, 2004) for the development of sociolinguistic competence (Tarone 2000), and for encoding the target language in a highly internalizable and ‘deeply processed’ (Craik and Lockhart 1972; Craik and Tulving 1975) fashion.
Functions of language play in second language acquisition Tarone (2000) argues that LP may be an important facilitator to SLA in at least the following ways. First, LP may lower affective barriers to SLA by providing a means of assuaging anxiety, thus allowing linguistic data to pass through the ‘affective filter’ (Krashen 1981) and become ‘intake’ (Chaudron 1988; Schmidt 1990). Second, LP may increase the memorizability of the discourse engaged in by the learner. Craik and Lockhart (1972), and Craik and Tulving (1975) suggest that retrieval from the long-term memory is facilitated by the creation of a ‘trace’, or triggering association. Craik and Lockhart (1972) also argue that ‘depth of analysis’ or ‘elaborative encoding’ will promote the creation of a strong trace. They define ‘depth’ as ‘involving a greater degree of semantic or cognitive analysis (1972: 675)’ and argue that this semantic enrichment can be achieved through an accumulation of ‘associations, images or stories on the basis of the subject’s past experience with the word’ (ibid.). In this vein, Cook (2001: 381–3) provides concrete examples of the mnemonic efficacy of LP, and suggests that LP may produce such lasting impressions on the memories of learners that, even after years of disuse, they will still be able to vividly recall the language encoded via LP.
CADE BUSHNELL
51
Third, LP may provide the learner with opportunities for incorporating other ‘voices’ (see Bakhtin 1981) into their L2—possibly fostering sociolinguistic competence. Sociolinguistic competence requires the appropriation of not just one register, but of ‘several voices or varieties appropriate to the speech communities to which the learner belongs, or wishes to belong’ (Tarone 2000: 46). Bakhtin (1981, 1984) emphasizes that we do not learn our languages from a dictionary, but rather that we weave together a patchwork of voices appropriated from the speech of others. LP gives learners an opportunity to experiment with other voices without concern for adverse social consequences. This process may enable learners to gain ownership of the voices and construct their own complex identities, allowing them to participate in their speech communities with a greater range of resources for and freedom of self-expression (Tarone 2000). Fourth, because LP may entail production of alternative linguistic forms, it could play a crucial role in the destabilization and restructuring of the learner’s interlanguage (IL) (Tarone 2000; Broner and Tarone 2001; Bell 2005; Kim and Kellogg 2007). Tarone (2000) argues that IL development requires both centripetal and centrifugal (see Bakhtin 1981) forces, which are manifest in the push of the demand for accuracy and the pull of creativity and innovation, respectively. She suggests that LP creates just such a situation for learners as they engage in the act of noticing linguistic forms in the course of LP and gradually replace incorrect productions with correct L2 forms. Cekaite and Aronsson (2005: 170) touch upon two additional functions of LP. The first is that of a ‘face-saving device’ which allows participants to commit face-threatening acts (FTAs) (Brown and Levinson 1987) while effectively avoiding social repercussions by remaining ‘off-record’ in the context of play (see also Zajdman 1995; van Dam 2002). Second, they suggest that LP may function as a venue for extended multiparty interaction. They note that ‘[i]n ordinary conversations speech errors and overt corrections frequently trigger play episodes’ (Cekaite and Aronsson 2005: 176). In the case of L2 learners, such instances may provide an opportunity to engage in a ‘language related episode’ (LRE) where the focus of the interaction shifts from conveying a message to attending to the linguistic form of the message itself (Swain 2000). LP may thus contribute to the creation of a space for continued collaborative attention to form. It is arguable that all of the functions mentioned above (i.e. affective, mnemonic, sociolinguistic, IL destabilizing, FTA mitigating, and interactional) may have considerable impact on SLA. In the analysis presented below, however, special attention is given to the ways in which LP functions to provide affordances for learners to internalize interactional episodes, and to develop greater levels of sociolinguistic competence. Additionally, I will also emphasize another important function of LP, which, to the best of my knowledge, has not been directly touched upon by previous research: LP as a resource for organizing and engaging in social interaction.
52
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
Research questions and methodology I will employ the methodology of conversation analysis (CA) in considering the following research questions:
In what ways do the participants of this study use LP? How does this use of LP function as a resource for engaging in social interaction? What affordances (van Lier 2000, 2004) for language development are made available through the LP?
CA is an extremely robust tool by which to examine social interaction. It is especially useful in helping the researcher to understand, from the participants’ perspective, how they (the participants) co-achieve social order and intersubjectivity within their interaction (Sacks et al. 1974; see also Tanaka 1999 for Japanese). Much of the previous research has tended to view LP through the lens of the individual learner. LP has been seen as a social phenomenon only insofar as it ‘occur[s] as part of a process in which learners appropriate the L2 speech of others in interaction and internalize it’ (Broner and Tarone 2001: 497). Furthermore, the relationship between LP and social action has often been de-emphasized in order to focus on its intrinsic features (see, e.g., Cook 1997). By using CA, however, the present study seeks to give careful consideration to the ways in which the participants orient to and use LP as a resource for engaging in and organizing their social interaction in the context of the pedagogical task.
SETTING AND DATA Classroom setting Data were collected from two second semester Japanese as a foreign language (JFL) classrooms at an American university. Signed consent was obtained from all participants prior to data collection. The data used in this study come from a subset of the data collected from one of these two classes. Ten of the fifteen students in the JFL class examined here had been together in the same class taught by the same teacher from the previous semester. None of the five new students were known to the other ten prior to the beginning of the data collection period. This fact served to create a classroom in which the intra-group interactions of one group of students were characterized by a familiarity and solidarity that reflected their shared social history from the previous semester. Such familiarity/solidarity was generally absent in the other group of students in both inter- and intra-group interactions (though some of the students from the ‘new group’ had developed associations within their own group prior to data collection). Class met four days a week for 50 minutes per session. The typical pedagogical flow began with a teacher-fronted presentation followed by whole class, teacher-fronted practice. The students would then be asked to
CADE BUSHNELL
53
form pairs or groups and be given various tasks related to the pedagogical focus (often involving some variation of role play) as the teacher circled the room providing assistance. The teacher then usually led the students in discussing any highlights, problems, etc. they experienced while engaging in the task. Occasionally, several pairs or groups of students would be asked to perform the task in front of the whole class.
Participants and data collection procedures Approximately 25 hours of audio data were collected from the class in question. In the present study, I analyze data from a ‘whole-class’ interaction (Excerpt 1), and data from the interactions of two individuals (Excerpts 2 and 3) from the JFL class described above. These two participants, Sal and Hal, were representative of the average to above average students in the class. Data collection was accomplished by having one participant, Sal, wear a lapel microphone for the entire class period during every day of the data collection period. This procedure was followed from the week prior to collection of the data used in this study. Additional data were also gathered via several informal interviews with Sal. According to Sal, at the time of data collection he and Hal were already well acquainted with each other as a result of extensive prior interaction.
Types of language play: ‘rehearsal’ vs. ‘fun’ Generally speaking, two distinct categories of LP have been discussed in the literature (Broner and Tarone 2001). The first type has been referred to as ‘rehearsal’ (Bell 2005; Broner and Tarone 2001) and is marked by such characteristics as a lower vocal volume, the absence of laughter, manipulation of phonological and morphosyntactic elements new to the learner, and lack of overt reference to a fictional world. Additionally, this type of LP is typically addressed to the self in the form of private speech (Broner and Tarone 2001; see also Lantolf 1997, 2000; Ohta 2001). The second type of LP, which most resembles the type examined in the current study, has been referred to as ‘fun’ LP (Broner and Tarone 2001). According to Broner and Tarone (2001), this type of LP contrasts with LP as rehearsal in that it may typically feature smiles or laughter, marked shifts in vocal pitch and quality, use of linguistic forms already known to the learner, reference to fictional worlds, and unlike the typically private rehearsal LP, fun LP often appears to be addressed to an other. In this article, I shall assume that the type of LP discussed by Cook (2000) falls under the heading of ‘fun’ rather than ‘rehearsal’ LP, though he does not use these terms himself. Cook notes that fun LP is ubiquitous in everyday interactions and that it may take on different functions according to the contexts within which its various features are deployed. Furthermore, he analyzes fun LP into three levels (formal, semantic, and pragmatic) and identifies several defining features for each level
54
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
(e.g., repetition and patterning [formal]; separation from ‘real-world’ reference [semantic]; and social inclusion and or exclusion [pragmatic]). Importantly, however, Cook also maintains that it is not necessary that all features exhibit equal prominence, nor that all features be present in every instance. Thus, for the purposes of this study, the presence of one or more of Cook’s features will be acknowledged as an instantiation of LP.
ANALYSIS Playing together: features of language play in ‘whole-class’ interaction Excerpt 1 shows a segment of talk involving ‘whole-class’ interaction where the teacher (T) is working to create a transition to a new pedagogical focus by asking the students (Ss) to recall the topic of the previous week’s lesson: keego ‘honorific language’. While consideration is also given to the function(s) (offering affordances for language learning, and sequence-organizational) of the LP, for illustrative purposes, the analysis will foreground the various features of the LP as they are evident in the excerpt. In this regard, Excerpt 1 is an especially perspicuous example because features from all three of Cook’s (2000) levels are identifiable within this one excerpt. (See the Appendix for a list of transcription conventions and grammatical terms.) Excerpt 1: ‘Lego my keego’1
1
T:
ii good
2
desu ka? minasan. keego C
simasita did
Q
everyone
(.) o benkyoo
honorifics
O
study
ne.= IP
Ok? Everyone, we studied (.) honorifics right? 3
S1:
=keego. honorifics
Honorifics. 4
T:
((nodding to S1))keego. keego wa nan desu ka? honorifics
honorifics T
what
C
Q
((nodding to S1)) Honorifics. What are honorifics? 5 6
(2) S2:
keego? honorifics
kego (.) kaimasita? KEG. [KEEGO.= kego
bought
keg
honorifics
=Honorifics? (You) bought a kego? Keg! Honorifics! 7
Ss:
[KEG. KEGO.=
CADE BUSHNELL
KEG
KEGO
Keg! Kego! 8
S2:
=KE::GO:::hehehehehehehe KE::GO:::hehehehehehehe
Kego! hehehehehehe 9
Ss:
=hehehehehehehe hehehehehehehe
=hehehehehehe 10 T:
hai, DAME::. [hehehehehehe yes
no good
hehehehehehe
Yeah right! hehehehehehe 11 S2:
[AH::: ((disappointed tone)) AH:::
12
AH. KE:GO [hehehehehe OO]KI BIIRU. AH
KE:GO
hehehehehe
big
beer
Aaah! ((disappointed tone)) Ah! Kego hehehehehe big beer! 13 S3:
[OOKII BIIRU. hehehe] big
beer
hehehe
Big beer! hehehe 14 Ss: 15 T:
HEHEHEHEHEHEHE sore wa ke:ggu. ke:ggu. (.4) [ke:ggu kore that
T
keg
keg
keg
this
That’s keg. Keg. Keg this 16 S2:
[kegu? keg
Keg? 17 T:
keego. [hehehehehe honorifics
hehehehehe
is keego. hehehehehe 18 S2:
[KE:GO KEGU KE:GO. (1) KE:GO KEGU KE:GO.
19
20 21 22 23
DON'T PLAY WITH MY KE:GO= Kego kegu kego! (1) Don’t play with my ke:go!= S3: S2: T: Ss:
=LEGO MY KEGO. [HEHEHEHEHEHE] [LEGO MY KE:GO. HEHEHEHEHE] [HEHEHEHEHEHE] [HEHEHEHEHEHE]
55
56
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
S2’s utterance in line 6 corresponds with the second feature of LP noted above: a separation from ‘real-world’ reference. Here, after a 2 second pause following T’s redirection to the previously studied topic of keego ‘honorifics’ (lines 1–5), S2 repeats the word keego with a rising intonation and immediately follows with the question ‘kego (.) kaimasita?’ ((you) bought a keg?) (line 6). By using the verb kaimasita ‘bought’, S2 ‘re-semiotizes’ (Belz 2002a) the Japanese word keego based on its phonological similarity with the English word ‘keg’ (i.e. keg of beer). This re-semiotization signifies a segment boundary between talk oriented to the real world and a new orientation to a ‘non-real’ world in which T has purchased a keg of beer. In lines 12 and 13, S2 and S3 co-clarify this re-semiotization by the addition of ookii biiru ‘big beer’. The fact that the interactional frame has been shifted to one of LP is further evidenced by the extended loud laughter from the other Ss in line 14: they are now in on the joke, having entered into an intersubjective state with the authors of this LP segment. The second instance of LP in Excerpt 1 corresponds with the linguistic patterning and/or repetition feature of LP. The Ss foreground the phonological features of their play via manipulations and repetitions (lines 6, 7, 8, 12, and especially 18–21). In lines 15 and 17, we see T making a repair of S2’s utterance by emphasizing the phonological differences between keggu ‘keg’ (foreign loan word: cf. native Japanese sakadaru ‘sake cask; keg’) and keego ‘honorifics’. T’s contribution, notably accompanied by laughter (line 17), triggers a further expansion of the LP by juxtaposing the two pronunciations, which S2 immediately incorporates into the play in a ‘sing-songy’ manner (lines 16 and 18). S2 also incorporates the phonological material ‘ke:go’ into the L1 clause ‘don’t play with my ke:go’ (line 19). The third feature of LP mentioned above—the pragmatic function of social inclusion and or exclusion—is especially salient in line 20 where S3 latches onto S2’s substitution play utterance with ‘lego my kego’ (echoing the catch phrase ‘lego my EggoÕ ’ from a well-known television commercial for Kellogg’s ‘EggoÕ waffles’). By couching ‘kego’ in a highly idiomatic chunk of culturally charged language, this move again results in the re-semiotization of the Japanese word keego. This triggers widespread and extended laughter— evidence of the socially inclusive nature of the collaboratively constructed LP frame (note the references to shared social experience among American college students, i.e. beer that comes in a keg and instant breakfast food).2 LP in the case of whole-classroom interaction makes salient a complex web of orientations to talk. We see instances of T constructing his default identity (Richards 2006) as teacher through his use of organizational talk (line 4) and his repair (lines 15 and 17). However, while the initial exchange between T and S1 orients towards classroom management as an interactionally accomplished activity, the orientation of the talk quickly changes as S2 reorients to T’s talk not as a student, but as a speaker (Shimazu 2000; Richards 2006). Other Ss subsequently orient to S2’s move as a potential initiation of an LP round and use the LP as a resource by which to extend and engage in the interaction via collaborative participation and laughter (lines 7, 9, 13, 14, 20, and 23).
CADE BUSHNELL
57
Playing being ‘sensee’: Engaging in the task as play In Excerpts 2 and 3 below, I turn to a consideration of LP in dyadic interactions between learners. An examination of the data revealed many instances of language play. LP activities such as joking, story telling and verbal dueling were common—in spite of the fact that the participants of this study were beginning learners of Japanese. Though I continue to make note of instances in the data that correspond to the features of LP as identified by Cook (2000), I shift the analytic emphasis to the function(s) (sociolinguistic, mnemonic and sequence-organizational) of the LP. In Excerpt 2, Sal and Hal are negotiating a task in which they are required to talk about what they did over the weekend. Just prior to this sequence, whole class practice had focused on talking about weekend activities using the question shuumatu wa nani o simasita ka (what did you do during the weekend?) and the response pattern X o simasita (I did X), or X o 5verb4-ta (I 5verb4-ed). Excerpt 2: ‘repooto o kakimasita’
1
H:
S-san, nani o (.) er shuma:tu wa (.) nani o S-title
2
what
O
er
weekend
T
what
O
simasita ka? did
Q
Mr. S, what did you do over the weekend? 3
S:
AI::::::uh:: (5) watasi wa (2) ukagai (2) ai
4
uh
I
T
MASU.
i ask you dude=
DS marker
I ask
you
HU-ask
dude
Ai, uuh I’ll ask. I ask you dude 5
H:
=uh:::
6
S:
shumatu ni nani o simasita ka, (.)H-san. weekend
DA
what
O
did
Q
H-title
What did you do on the weekend, Mr.H. 7 8
(3) H:
{repooto o}kaki:masit [a, it's right there= report
O
wrote
it's
right
(I) wrote (my) report, it’s right there 9
S:
10 H:
[hehehehe =you know. {repooto o
kakimasita},
there
58
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
you
know
report
O
wrote
you know. (I) wrote my report 11 S:
AH. so AH
that
desu ka:: hehe C
Q
hehe
Ah. Is that right? hehe ((lines 8 and 10: {words}= ‘regal’ tone))
The interaction begins with both Sal and Hal orienting to the task-at-hand as directed by the teacher. Hal asks Sal about his weekend using the model provided by T in the prior activity, forming the first pair part of a Question/ Answer adjacency pair. However, Sal hesitates to respond to Hal’s question by deploying a greatly lengthened first word followed by the hesitation token ‘uh’, followed by a 5 second pause (line 3). At this point, Sal informs Hal that he (Sal) will do the asking. Though Sal’s turns in lines 3, 4, and 6 ignore the adjacency pair initiated by Hal, Sal formulates his utterance in line 4 by using highly colloquial language accompanied by contextualization cues (Gumperz 1982; Ostermann 2003) that are subsequently oriented to by Hal as a potential initiation of LP: the alignment marker, ‘dude’, a code-switch into English, and finally, use of the voice of Teacher by referring to Hal in the same manner that T does (e.g. by attaching of ‘Hal-san’ to the end of his question in line 6; a possible ‘inversion of reality’, one feature of LP (Cook 2000: 123)).3 In line 8, following a 3 second pause, Hal orients to Sal’s actions by deploying a highly marked tone of voice (an apparent imitation of a British accent), which serves the function of aligning his interactional frame with the ‘non-real world’ orientation (one of Cook’s (2000) features) initiated by Sal. Hal’s actions simultaneously become a preferred response to Sal’s invitation to engage in LP and the first pair part of a Joke/Laughter adjacency pair (Schegloff 1987; Sacks 1989). Sal unhesitatingly responds (note the overlap in lines 8 and 9) with the second pair part of laughter (line 9). Upon experiencing favorable reception from Sal, Hal recycles his laughter-evoking utterance to which Sal responds in line 11 with further laughter and an acknowledgment token, ‘so desu ka::’, with an affected elongation on the final syllable. In Excerpt 2, although Sal and Hal chose to orient to the task as friends-at-play rather than students-at-work, they have been able to skillfully merge the requirements of the task with their LP. They are collaboratively using and creating with the target language. Additionally, line 6 shows Sal experimenting with the use of a different voice (i.e. the voice of ‘teacher’) as he initiates this round of LP. Such experimentation has been argued to be beneficial to the development of sociolinguistic competence of both child and adult L1 and L2 learners (Tarone 2000) and is a common feature of the LP in my data. Finally, Excerpt 2 shows Hal and Sal collaboratively co-constructing their LP; I suggest that this co-construction and use of ludic activity becomes a resource by which learners may organize the deployment of ‘on-task’ target language forms.
CADE BUSHNELL
59
Making ‘sensee’ sick: Interactionally (re)organizing the task through play During the portion of class directly preceding Excerpt 3, T has modeled the interaction for the Ss by calling on a volunteer and working through the pattern of interaction with him, and has also offered a brief explanation in English and Japanese regarding the proper use of keego ‘honorific language’ for referring to the actions of one’s superiors. Excerpt 3 shows Sal and Hal enacting a role play in which one of them is to play the part of a sick student seeking audience with his busy teacher in order to reschedule an appointment, and the other to play the part of the teacher. In their interaction, however, Hal and Sal (re)organize and transform the task by, through, and for their play, while simultaneously displaying an orientation to the use of ‘on-task’ language. Excerpt 3: ‘siroi bi:nzu’
1
S:
so, you’re gonna be the sensee. so
you’re
gonna
be the
professor
alright? alright
So, you’re gonna be the professor. Alright? 2
[at least on this one.(.) so, a, sensee.= at
least
on
this
one
so
a
professor
At least on this one. (.) So, a, professor.= 3
H:
[yeah, i can live with that Yeah, I can live with that
4
S:
=konnichi wa. today
T
=hello. 5
H:
{a:::. S-sa:n}.= a
S-title
{Ah. Mr. S}.= 6
S:
7
H:
=heheh koni(h)ti
wa(h).
today
T
He(h)llo(h). 8
S:
ano::, uh:::(.) um
uh
Um. uh 9
H:
{S-sensee. [S-sensee} S-teacher
S-teacher.
{Professor S. Professor S} ((lines 5 and 9: {words}=‘East Asian sage’ tone)) 10 S:
[° is this kon? (is this) kon?° =
60
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
is this
this
is
this
this
Is this this’ ? (is this) this’? 11 H:
=it’s ima. it’s now
It’s ‘now’. ((lines 10 and 11: talking about how to read a certain Chinese character)) 12 S: 13
ima, ah::. ima wa. ano:, ima:: wa:
chotto
now
little
ah
now
yorosii desu good
T
um
now
T
ka?
C
Q
Now, ah. Do you. uum, do (you) have a minute now? 14 H:
hai. yes
mo:: mochiron. of c-
of course
nan
deshoo?
what
C
Yes. of c-, of course. What seems to be the matter? 15 S:
ano:: (1) uh (2) ki- ki:noo:, ki:noo:, uh um
16
uh
ye-
yesterday
uh
(.) ki:noo, nani o (.) mesiagari (.) yesterday
17
ye:sterday
masita
what
O
H-eat
ka?
DS marker
Q
Um uh ye- yester, yester, uh what did you eat yesterday? 18
ah, hiru go:han. hiru. (1) ((to T)) uh, how Ah
19
noon
food
noon.
uh
how
would you say eat something for lunch? would
you
say
eat
something
for
lunch
Ah, lunch. noonÕ. ((to T)) Uh, how would you say eat something for lunch? 20 T:
hiru gohan de. noon
food
DA
For lunch. 21 S:
ok. we're good. uh, thank yo- uh, domo ok
22
we're
good
uh
thank
yo-
uh
very
arigatto:: gozaimasita, a: thank you
HU-exist
a
Ok. we’re good. uh, thank yo- uh, ‘thank you’. 23
((reorienting to H)) ok, so ano(H)O(H)O:, ok
24
kino(.)
so
um
hiru- hiru gohan DE(.)
yesterday
noon
noon
food
DA
((reorienting to H)) Ok, so u:(h)m(h), yesterday lun- for lunch (.) 25
uh, nani o(.) uh, uh
what
O
uh,
mesiagarimasita ka? H-eat
Q
CADE BUSHNELL
uh, what (.) did you eat? 26 27 H:
(2) uh::, Wahoo’s sando (1) o tabemasita, uh
28
Wahoo’s
sandwich
O
ate
sosite, (3.2) uh (.) {siroi BI:NZU} o (4) and then
uh
white
bean(s)
O
Uh, (I) ate a Wahoo’s sandwich and then, (3.2) uh white beans (4) 29 S:
>SENSEE SENSEE.< {HANBAAGAA O professor professor
30
hamburger
O
[MESIAGARIMAS-}MASEN >MAS [EN DESITA KA
DS-NG
C
Q
Professor, professor! Didn’t >didn’t you eat a hamburger
[siroi whoa o:::o:::o:: white
whoa
White whoa ooo 32 H: 33
no no::, SIROI BI::NZU::. siroi
bi::nzu o
no
bean(s )
no
white
bean(s )
white
O
tabemasita kedo (.) uh (1) ge:ri o su- heh ate
but
uh
diarrhea
O d-
heh
No no. White beans! I ate white beans, but (.) uh (1) diarrhea (I) ge- heh 34 S:
GEri
o::,=
diarrhea
O
Diarrhea,= 35 H:
=geri
o
simasit(h)a. [heheheheh
diarrhea
O
did
heheheheh
=(I) got diarrhe(h)a. heheheheheh 36 S:
[A::::::H ah
37
HEHEHGERHEHEH ↓A::: hehehedihehehe
ah
Aaah. hehehehdiarhehehe ! aaa. 38
GE- [SENSEE(.) GERI
O SIMA- SIMASITA KA?
di-
O
professor
diarrhea
di-
Di- professor (you) go- got diarrhea? 39 H:
[geri diarrhea
heheh heheh
Diarrhea heheh 40
HAI. yes.
did
Q
61
62
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
In lines 1 and 2, Sal initiates the interaction via his managerial use of L1. Hal produces a turn overlapping the last half of Sal’s utterance—a weak endorsement of Sal’s proposal: ‘yeah, I can live with that’. In line 5, Hal deploys a marked tone of voice in his production of an exaggeratedly elongated change of state token: ‘ah’ (Heritage 1984; see also Mori 2004 for Japanese). The participants subsequently orient to this move as an initiation of a round of LP in which they collaboratively co-construct a double-framed interaction, that is simultaneously both a ‘non-real world’ and ‘on-task’ (i.e. real world) frame. Hal’s marked tone of voice and word choice suggest that this is an instantiation of ‘double-voicing’ (Bakhtin 1981; Tarone 2000), and that Hal is assuming the role of a stereotypical East Asian sage/sensee. Several things are happening here. First, T has assigned the Ss a role play wherein they must assume either the role of T or S. By engaging in this task, Sal and Hal are doing being students. Second, within this T-imposed frame, Sal has assigned the role of ‘sensee’ to Hal. Third, Hal simultaneously plays the role of T (an orientation to the real world demands of the task), and the role of ‘sage/ sensee’ (an orientation to the non-real world feature of LP). This is evidence of Hal’s developing awareness of the interactional effects made available through assuming different ‘voices’. It also highlights the way in which LP provides a venue for further development of such sociolinguistic competency through experimentation. In line 6, Sal responds to Hal’s LP with laughter, a sign that he acknowledges Hal’s attempt to double-frame the interactional sequence as an opportunity to initiate a round of LP nested within an orientation to task accomplishment. In the subsequent interaction, Sal and Hal weave LP into the task-at-hand in a complex way, eventually reorganizing and transforming the task dramatically. The pair temporarily puts the Hal-initiated LP on hold while they engage in a brief side sequence (Jefferson 1972) concerning the correct reading of a Chinese character printed on the cue sheet for the role play (lines 10 and 11), and pursue several moves with a practical orientation to the task-at-hand (lines 12–17). After asking T for some grammatical assistance in lines 18 to 22, Sal produces the first pair-part of a Question/ Answer adjacency pair in lines 23–25. Hal’s line 27 begins by first providing the second pair-part to Sal’s adjacency pair and then continues using sosite ‘and’ (line 28), which maintains the floor by creating an addition-relevant slot. Hal fills this slot with the considerably loud and emphasized ‘siroi BI:NZU’ (i.e. ‘white bean(s)’; Hal subsequently uses this object to produce an accounting of his intestinal distress in lines 31–35). In lines 29 and 30, Sal continues to orient to Hal as ‘sensee’ on the one hand (both by reference to the title and by using honorific language), while resonating with Hal’s ‘siroi BI:NZU’ by producing a prosodically similar ‘HANBAAGAA’, a word recycled (note that such recycling may work to heighten a sense of social inclusion— a feature of the pragmatic level of LP) from the LP of a previous interactional sequence (not shown) on the other. This move shows that LP need not be
CADE BUSHNELL
63
sustained in an uninterrupted manner. Rather, it can be put on hold within the course of the interaction (as in lines 18–22), or even across interactions (as with Sal’s recycling of ‘HANBAAGAA’), as it is collaboratively constructed by the participants. Thus LP has arguably provided affordances (van Lier 2000, 2004) to the participants for elaborative encoding (Craik and Lockhart 1972; Tarone 2000)—making the language of their interactions highly memorizable. In lines 31–40, Hal uses his line 28 ‘siroi BI:NZU’ in a subtle reversal of the discursive roles of the task4—an action which makes relevant a sickness telling by Hal-as-sensee rather than by Sal-as-student.5 Hal does this by producing an accounting of his ‘sickness’, asserting that the ‘siroi BI:NZU’ were a contributing factor to his geri ‘diarrhea’ (lines 32, 33, and 35). In line 34, Sal produces ‘GEri o::,¼’, with an elongated last syllable, a slightly rising intonation, and a volume increase as well as a heavy emphasis on the first syllable of geri. In line 35, Hal orients to this action by Sal as being a possible initiation of repair by latching on to Sal’s line 34 with a redeployment of the potential repairable ‘geri o simasit(h)a.’, with a falling final intonation. Hal’s line 35 also features laughter, which begins during the last syllable of the final word in his utterance. This laughter token touches off a round of partially overlapped laughter, by which Hal and Sal collaboratively orient to Hal’s line 35 as a joke (Schegloff 1987; Sacks 1989). In lines 38 and 39 the participants produce repetitions of the laughable item, geri. In addition to the absurdity of the notion of ‘sensee’ confiding in his student in this manner about his intestinal distress, the joke here seems also to be related to a sort of covert6 social inclusion (one of Cook’s (2000) LP features) based on the shared cultural knowledge that this type of lexical item might normally be considered taboo in a classroom setting. Thus, the participants are able to use LP to organize a co-display of their orientations to social norms, and the potential humorousness of flouting such norms. Of additional significance is the fact that geri was a new vocabulary word for this lesson. Thus, the participants’ use of geri as they engage in LP further evidences the doubleframed nature of this interaction; Sal and Hal are simultaneously orienting to the task as both friends-at-play and students-at-work. Also, though the participants have effectively transformed the task by reversing the discourse roles by and through their LP, they have still displayed an orientation to (most of) the requirements of the pedagogical task by ‘gaining audience to the teacher’ and then producing a ‘sickness telling’ (although it was the teacher that was sick and not the student). Excerpt 3 shows that Sal and Hal have oriented to the T assigned role play7 in an LP frame. They have used their LP as a resource by which to organize their interaction, into which they creatively and seamlessly incorporated ‘ontask’ linguistic elements such as newly introduced vocabulary like geri, and even keego ‘honorifics’. It must also be emphasized that not only have Sal and Hal managed to include such language in their LP, but that the LP has provided affordances (van Lier 2000, 2004) for language learning:
64
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
(i) Hal’s multi-voiced orientation to the task (i.e. his simultaneous assumption of both the T assigned role of ‘teacher’ and the Sal assigned role of ‘sensee’, along with his own deployment of his East Asian sage ‘voice’), Sal’s incorporation of keego, and the participants’ flouting of social norms in co-constructing their ‘joke’ have provided them with affordances for further development of their sociolinguistic competence; and (ii) the ways in which the participants were seamlessly able to put the LP on hold in their interaction, recycle items from previous interactions featuring LP, and the fact that Sal subsequently went on to recycle items from the LP in Excerpt 3 in later interactions8 is indicative of how LP has functioned to provide affordances for elaborative encoding.
CONCLUSION Cekaite and Aronsson (2005: 169) argue in favor of a ludic model of language learning, contending that ‘we need to take non-serious language more seriously’. The goal of this research has been to give serious consideration to instantiations of LP in the interactions of beginning students of JFL. To do so, the following research questions were considered:
In what ways do the participants of this study use LP? How does this use of LP function as a resource for engaging in social interaction? What affordances for language development are made available through the LP?
The present study has clearly illustrated the complexity and depth with which adult L2 learners may engage in LP. First, the learners have been shown to use LP as a resource through which to organize their co-engagement in pedagogical tasks. Within the contexts of the data considered in this study, I have argued that LP used in this manner functions to provide affordances (van Lier 2000, 2004) for encoding the target language in a highly memorable fashion, and for developing greater sociolinguistic competence by, for example, experimenting with different voices. The findings of this study contribute to SLA research by accounting for these under-considered functions of LP, which may be of great benefit to classroom language learning. A growing body of research has shown that not only do learners tend not to engage in negotiation when performing ‘(referential) meaning-focused’ interactional tasks (Foster 1998; Roebuck 2000; Foster and Ohta 2005), but they often fail even to do the expected task (DiNitto 2000; Seedhouse and Richards 2005). In my data, however, a joint orientation to an LP frame seems to have provided a shared space in which the participants were able to reorganize the task as play and then effectively engage in the task-as-play.9 Importantly, the participants have been shown to be using ‘on-task’ language forms as they engage in LP. This fact forces us to re-conceptualize LP as a possible motivator and facilitator rather than as disruptive, ‘off-task’ behavior.
CADE BUSHNELL
65
In this vein, Cook (2000: 204) argues that ‘Play . . . does not entail a rejection of order or authority, though it does at least imply more voluntary and creative reasons for embracing them’. Furthermore, Foster and Ohta (2005) suggest that one possible reason for the paucity of ‘negotiation for meaning’ in classroom interaction may be that such negotiation actually constitutes a face-threatening act (Brown and Levinson 1987). By providing a non-real world frame, however, LP may create a low anxiety (Tarone 2000) space for learners to freely experiment with and use L2 free from any concerns of ‘losing face’ (see Zajdman 1995; van Dam 2002; Cekaite and Aronsson 2005). Future research should be done to examine the mechanism by which LP creates such a space. If, as many SLA researchers have reasoned, it is true that negotiated interaction and engagement in tasks are an important or essential ingredient to SLA, it thus becomes arguable that LP is indeed worthy of serious consideration as a contributing factor to language development. Finally, longitudinal research should be done to track the ways in which participants act on affordances provided by LP, and how these affordances contribute to language development.
ACKNOWLEDGEMENTS An earlier version of this paper was presented at the 10th Annual International Conference of the Japanese Society for Language Sciences in Shizuoka, Japan, 12–13 July 2008. I wish to express my heartfelt appreciation to Dina Yoshimi for her valuable comments on earlier versions of this paper. I am also deeply grateful to the Applied Linguistics editors and the anonymous reviewers for their valuable suggestions, and to the University of Hawai’i at Ma¯noa CA data session participants for their many helpful insights. However, any errors or misinterpretations of the data are my own.
APPENDIX Transcription conventions ^ glottal stop hehe laughter "# high or low pitch (placed prior to affected element) 4words5 quicker than surrounding talk 5words4 slower than the surrounding talk [ beginning of overlapped speech ] end of overlapped speech ¼ latching (i.e. no pause after the completion of one utterance and the beginning of another) (3.3) length of pause (measured in seconds and tenths of seconds) (.) unmeasured pause (words) unclear utterance ((words)) commentary by transcriptionist wo:::rd geminate WORDS louder than surrounding talk
66
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
words softer than surrounding talk words more emphasis than surrounding talk wocut-off ’ continuing intonation . final intonation ? question intonation
Interlinear grammatical notation key C: Copula CT: Continuer D: Double particle (kamo, toka, etc.) DA: Dative particle (he, ni) F: Speech filler IP: Interactional particle (yo, ne, etc.) L: Linking device (-te, si, kedo, etc.) M: Noun modification particle (no, na, etc.) N: Nominalizer NG: Negative O: Object marker P: Past tense PA: Passive Q: Question marker QT: Quotation marker S: Subject marker T: Topic marker Stylistic indicators (when necessary): DS-: Distal style FS-: Formal style H-: Honorific HU-: Humble PS-: Plain style
NOTES 1 Transcriptions appear with the first line in Romanized Japanese followed by a literal translation with grammatical elements in all capital letters. An italicized gloss in natural English is supplied in the third line. 2 Although Excerpt 1 embodies all three features of LP, it will be recalled that, according to Cook (2000), this need not necessarily be the case. 3 One of the anonymous reviewers questioned whether Hal’s use of ‘S-san’ in line 1 should not also be
seen as an instantiation of doublevoicing. While we cannot know Hal’s intentions in framing his utterance in such a way, three features of the data must be noted. First, and most importantly, Sal does not display an understanding of Hal’s line 1 as being LP initiation relevant. Second, Sal’s line 6 differs from Hal’s line 1 in that it affixes ‘H-san’ in an utterance final position (which imitates T’s use of this resource for classroom management). Finally, it is both accompanied by
CADE BUSHNELL
other contextualization cues and contextualized as being ‘teacher’s voice’ via prosodic features such as intonation and stress (features which are contrastingly absent in Hal’s line 1). 4 I am indebted to both one of the anonymous reviewers and the participants in a fall semester 2007 data session at the University of Hawai’i at Ma¯noa for their insights on this point. 5 Note that, Sal’s action of questioning the sensee (i.e. Hal) about what he had for lunch does not accomplish the role reversal in and of itself. Rather, it is Hal’s subsequent orientation to the ensuing adjacency pair (especially his own line 28 second pair-part) as a resource for initiating his sickness telling that does this. 6 Some ethnographic data suggest that Sal and Hal may have perceived their LP as an illicit activity. In one recording, Sal made the following comment (addressed apparently to the researcher, whom Sal knew would eventually be listening to the recording) about the LP he and Hal had been engaging in: Ok. Hal’s not answerin’ my questions! I ask him where he’s going and he says hamburger. I ask him what he ate and he says nemasita (i.e.‘slept’) (laugh). You give’im the grade you want; I just wanna set the record straight (laugh). This comment by Sal not only evidences Sal’s awareness of having engaged in LP, but also suggests that Sal views LP as a somehow ‘substandard’ or ‘illicit’ activity which could rightly be dealt with in a disciplinary manner (i.e. you give’im the grade you want) 7 One of the anonymous reviewers questioned the value of examining LP within a role play task since it would seem to be a given that play would be a salient part of such an interaction.
67
However, the simple fact that the word ‘play’ is included in ‘role play’ does not a priori ensure that LP will feature in the interaction (in fact, in my data, role plays were also often characterized by a ‘serious’ orientation to the task—likely stemming from the possibility of being requested to present the role play in front of the class afterwards). Furthermore, a major part of the analytic focus in Excerpt 3 is on the ways in which, by and through their LP, Sal and Hal reorganized and transformed the task in a way which provided them with affordances for language learning that almost certainly would not have surfaced otherwise. 8 In addition to the evidence for elaborative encoding provided by Sal’s recycling of ‘HANBAAGAA’ from a previous interaction with Hal, the LP of this interaction apparently provided (at least) Sal with affordances for internalizing the new lexical item geri ‘diarrhea’ as well: he was observed to use it in LP during an interaction with a different partner two days later. 9 Orienting to work as play is a common human phenomenon. Cook (2000: 203; emphasis mine) notes that people often play while working, and that imbuing ‘work’ with a playful sense may increase productivity by ‘co-ordinating actions and making [the work] seem lighter and more co-operative.’ In a study of interactions in various New Zealand white collar workplaces, Holmes (2007) offers additional empirical support for the work-related benefits of humor. In particular, she notes that humor (i) potentially contributes to the construction of effective relationships in the workplace, and (ii) may stimulate intellectual activity relevant to the achievement of work-related objectives.
68
LANGUAGE PLAY IN A BEGINNING JAPANESE FL CLASSROOM
REFERENCES Attardo, S. 1994. Linguistic Theories of Humor. New York: Mouton de Gruyter. Bakhtin, M. [1934] 1981. The Dialogic Imagination, Four Essays by M. M. Bakhtin. Ed. M. Holoquist, trans C. Emerson. Austin, TX: University of Texas Press. Bakhtin, M. [1929] 1984. Problems in Dostoevsky’s Poetics. Ed. and trans. C. Emerson. Minneapolis: University of Minnesota Press. Bell, N. 2005. ‘Exploring L2 language play as an aid to SLL: A case study of humour in NS–NNS interaction,’ Applied Linguistics 26/2: 192–218. Belz, J. 2002a. ‘The myth of the deficient communicator,’ Language Teaching Research 6/1: 59–82. Belz, J. 2002b. ‘Second language play as a representation of the multicompetent self in foreign language study,’ Journal of Language, Identity, and Education 1: 13–39. Block, D. 2003. The Social Turn in Second Language Acquisition. Edinburgh: Edinburgh Univesity Press. Block, D. 2007a. Second Language Identities. London: Continuum. Block, D. 2007b. ‘Socializing second language acquisition’ in Z. Hua, P. Seedhouse, L. Wei, and V. Cook (eds): Language Learning and Teaching as Social Interaction. New York: Palgrave Macmillan. Broner, M. and E. Tarone. 2001. ‘Is it fun?: Language play in a fifth-grade Spanish immersion classroom,’ The Modern Language Journal 85/ 3: 363–79. Brown, P. and S. Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Cekaite, A. and K. Aronsson. 2005. ‘Language play: A collaborative resource in children’s L2 learning,’ Applied Linguistics 26/2: 169–91. Chaudron, C. (1988). Second Language Classrooms: Research on Teaching and Learning. Cambridge: Cambridge University Press. Cook, G. 1997. ‘Language play, language learning,’ English Language Teaching Journal 51/3: 224–31. Cook, G. 2000. Language Play, Language Learning. Oxford: Oxford University Press. Cook, G. 2001. ‘ ‘‘The philosopher pulled the lower jaw of the hen’’: Ludicrous invented sentences in language teaching,’ Applied Linguistics 22/3: 366–87. Craik, F. and R. Lockhart. 1972. ‘Levels of processing: A framework for memory research,’ Journal of Verbal Learning and Verbal Behavior 11: 671–84.
Craik, F. and E. Tulving. 1975. ‘Depth of processing and the retention of words in episodic memory,’ Journal of Experimental Psychology: General 104/3: 268–94. DiNitto, R. 2000. ‘Can collaboration be unsuccessful?: A sociolinguistic analysis of classroom setting and Japanese L2 performance in group tasks,’ Journal of the Association of Teachers of Japanese 34/2: 179–210. Doughty, C. and T. Pica. 1986. ‘Information gap tasks: Do they facilitate second language acquisition?’ TESOL Quarterly 20: 305–25. Foster, P. 1998. ‘A classroom perspective on the negotiation of meaning,’ Applied Linguistics 19: 1–23. Foster, P. and A. Ohta. 2005. ‘Negotiation for meaning and peer assistance in second language classrooms,’ Applied Linguistics 26: 402–30. Gafaranga, J. 2005. ‘Demythologising language alternation studies: Conversational structure vs. social structure in bilingual interaction,’ Journal of Pragmatics 37: 281–300. Garvey, C. 1984. Children’s Talk. Cambridge, MA: Harvard University Press. Garvey, C. [1977] 1990. Play. Cambridge, MA: Harvard University Press. Gumperz, J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Hauser, E. 2005. ‘Coding ‘‘corrective recasts’’: The maintenance of meaning and more fundamental problems,’ Applied Linguistics 26/3: 293–316. Heritage, J. 1984. ‘A change-of-state token and aspects of its sequential placement’ in J. M. Atkinson and J. Heritage (eds): Structure of Social Action. Cambridge: Cambridge University Press. Holmes, J. 2007. ‘Making humour work: Creativity on the job,’ Applied Linguistics 28/4: 518–37. Jefferson, G. 1972. ‘Side sequences’ in D.N. Sudnow (ed.): Studies in social interaction. Glencoe, IL: Free Press, pp. 294–333. Kim, Y. and D. Kellogg. 2007. ‘Rules out of roles: Differences in play language and their developmental significance,’ Applied Linguistics 28/1: 25–45. Kramsch, C. (ed.) 2002. Language Acquisition and Language Socialization: Ecological Perspectives. New York, NY: Continuum. Kramsch, C. and P. Sullivan. 1996. ‘Appropriate pedagogy,’ English Language Teaching Journal 50: 199–212. Krashen, S. 1981. Second Language Acquisition and Second Language Learning. Oxford: Pergamon Press.
CADE BUSHNELL
Lantolf, J. 1997. ‘The function of language play in the acquisition of L2 Spanish’ in A. Perez-Leroux and W. Glass (eds): Contemporary Perspectives on the Acquisition of Spanish. Vol. 2: Production, Processing, and Comprehension. Sommerville, MA: Cascadilla Press. Lantolf, J. (ed.) 2000. Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. Long, M. 1983. ‘Native speaker/non-native speaker conversation and the negotiation of comprehensible input,’ Applied Linguistics 4: 126–41. Long, M. 1996. ‘The role of the linguistic environment in second language acquisition’ in W. C. Ritchie and T. K. Bhatia (eds): Handbook of Research on Language Acquisition: Second Language Acquisition. Vol. 2. New York: Academic Press, pp. 413–68. Mori, J. 2004. ‘Negotiating sequential boundaries and learning opportunities: A case from a Japanese language classroom,’ The Modern Language Journal 88/4: 536–49. Ohta, A. 2000. ‘Rethinking interaction in SLA: Developmentally appropriate assistance in the zone of proximal development and the acquisition of L2 grammar’ in J. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. Ohta, A. 2001. Second Language Acquisition Processes in the Classroom: Learning Japanese. Mahwah, NJ: LEA. Ostermann, C. 2003. ‘Localizing power and solidarity: Pronoun alternation at an all-female police station and a feminist crisis intervention center in Brazil,’ Language in Society 32: 357–81. Pica, T. 2005. ‘Classroom learning, teaching and research: A task-based perspective,’ The Modern Language Journal 89/3: 339–52. Richards, K. 2006. ‘ ‘‘Being a teacher’’: Identity and classroom conversation,’ Applied Linguistics 27/1: 51–77. Roebuck, R. 2000. ‘Subjects speak out: How learners position themselves in a psycholinguistic task’ in J. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. Sacks, H. 1989. ‘An analysis of the course of a joke’s telling in conversation’ in R. B. a. J. Sherzer (ed.): Explorations in the Ethnography of Speaking. Cambridge: Cambridge University Press, pp. 337–53. Sacks, H., E. A. Schegloff, and G. Jefferson. 1974. ‘A simplest systematics for the
69
organization of turn-taking for conversation,’ Language 50: 696–735. Schegloff, E. A. 1987. ‘Some sources of misunderstanding in talk-in-interaction,’ Linguistics 25: 201–18. Schmidt, R. 1990. ‘The role of consciousness in second language learning,’ Applied Linguistics 11/2: 129–58. Seedhouse, P. and K. Richards. (eds) 2005. Applying Conversation Analysis. New York: Palgrave Macmillan. Shimazu, M. 2000. ‘Co-construction of classroom identity in Japanese as a foreign language classrooms,’ Unpublished master’s thesis. University of Hawai’i at Ma¯noa. Skehan, P. 1998. A Cognitive Approach to Language Learning. Oxford: Oxford University Press. Sullivan, P. 2000. ‘Playfulness as mediation in communicative language teaching in a Vietnamese classroom’ in J. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. Swain, M. 2000. ‘The output hypothesis and beyond: Mediating acquisition through collaborative dialogue’ in J. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. Tanaka, H. 1999. Turn-taking in Japanese Conversation: A Study in Grammar and Interaction. Amsterdam: J. Benjamins. Tarone, E. 2000. ‘Getting serious about language play: Language play, interlanguage variation and second language acquisition’ in B. Swierzbin et al. (eds): Social and Cognitive Factors in Second Language Acquisiton. Sommerville, MA: Cascadilla Press, pp. 31–54. van Dam, J. 2002. ‘Ritual, face, and play in a first English lesson: Bootstrapping a classroom culture’ in C. Kramsch (ed.): Language Acquisition and Language Socialization: Ecological Perspectives. New York, NY: Continuum. van Lier, L. 2000. ‘From input to affordance: Social-interactive learning from an ecological perspective’ in J. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Cambridge: Cambridge University Press. van Lier, L. 2004. The Ecology and Semiotics of Language Learning: A Sociocultural Perspective. Boston: Kluwer. Zajdman, A. 1995. ‘Humorous face-threatening acts: Humor as strategy,’ Journal of Pragmatics 23: 325–39.
Applied Linguistics 30/1: 70–92 ß Oxford University Press 2008 doi:10.1093/applin/amn036 Advance Access published on 11 September 2008
A Novel Approach to Creating Disambiguated Multilingual Dictionaries IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA and CAROLINA GALLARDO Universidad Polite´cnica de Madrid Multilingual lexicons are needed in various applications, such as cross-lingual information retrieval, machine translation, and some others. Often, these applications suffer from the ambiguity of dictionary items, especially when an intermediate natural language is involved in the process of the dictionary construction, since this language adds its ambiguity to the ambiguity of working languages. This paper aims to propose a new method for producing multilingual dictionaries without the risk of introducing additional ambiguity. As a disambiguated intermediate language we use the so-called Universal Words. A set of more than 200,000 unambiguous Universal Words have been constructed automatically on the basis of the well-known English lexical database WordNet. This approach is being used for the construction of a five languagedictionary in the field of cultural heritage within the framework of the PATRILEX project sponsored by the Spanish Research Council.
INTRODUCTION Multilingualism in the modern world requires extensive multilingual lexicographic resources both for human translation and for diverse computer applications, such as multilingual information retrieval, information extraction, localization of industrial products, technical manuals, machine translation, or multilingual chats. In this paper, a novel approach to building multilingual dictionaries is proposed based on an artificial interlingua. The plan of the paper is as follows. Next, we will discuss the problem of ambiguity as one of the obstacles standing in the way of compiling multilingual dictionaries. We come to the conclusion that introducing an interlingual representation is a convenient way to cope with this problem. We will conclude the introduction with a brief overview of existing approaches to building interlingual representations for different purposes. One of these—the Universal Words (UWs) approach—is our choice for multilingual terminological lexicography. In ‘Foundations of Universal words’, theoretical foundations of UWs are given and ‘Building the General UW Dictionary’ describes a method for the automatic generation of a UW dictionary on the basis of existing lexicographic resources. The obtained UWs are fully disambiguated and serve as the basis for creating multilingual dictionaries. ‘Method for constructing Multilingual Dictionaries’ focuses on
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
71
the method for constructing multilingual dictionaries based on UWs, where adding a language to a multilingual dictionary based on UWs consists in searching the equivalent term for a given UW and does not require looking for equivalents in the other working languages. This means that to build a dictionary containing, say, Arabic and Russian data, one does not need bilingual Arabic–Russian lexicographers but only monolingual experts in Arabic and in Russian. Finally, ‘Method for constructing Multilingual Dictionaries’ concludes with experimental testing of the viability as well as the quality of the produced dictionaries. Before we start, we would like to make two remarks in order to prevent any possible misunderstanding. First, the approach proposed in this paper does not pretend to solve all the difficult problems encountered in translation and dictionary making. Nor does it guarantee that the lexicographer will easily find an equivalent to any term in another language. It does not relieve him or her of the creative part of lexicographic work. It only tries to take up the tedious and repetitive parts of the work, in which computers are better than humans. It does not go beyond two tasks both related to making the source term more easily understandable by a lexicographer working in a multilingual setting: delimitation of the meaning of the source term and finding a comprehensible intermediate representation for it. By delimitation of the meaning of the term we understand two complementary aspects— internal and translational disambiguation. On the one hand, a choice is made among several interpretations which the term may have in the source language (internal disambiguation). On the other hand, lexical mismatches between the source and the intermediate languages are resolved (translational disambiguation). Both tasks are solved by means of semantic restrictions. Second, the method we propose is intended in the first place for constructing terminological dictionaries for restricted subject domains. General-vocabulary items rich in idiomatic language- and culture-specific meanings, collocations, and shadings of meaning are much less eligible for this type of processing, although computational techniques for dealing with collocations and idiomatic expressions are being actively investigated now.
Multilingual dictionaries and ambiguity Usually, the need for lexicographic resources is served by bilingual dictionaries that have only two working languages. However, existing bilingual dictionaries do not come close to covering all relevant language pairs. It would hardly be possible, for example, to find a good Urdu–Estonian or Somali–Russian dictionary. The problem of multilingualism does not boil down to the existence of several major languages. The real problem is what might be called ‘massive multilingualism’, that is, the need to link dozens of languages among themselves. Most of these languages are poorly equipped by even traditional printed dictionaries, to say nothing of electronic ones or machine translation systems. For many pairs of languages, even human
72
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
translators do not exist. The relevance of this problem is clearly manifested in recent publications (cf. Tsakona 2007). The direct consequence of this lack of resources is that in many cases the link between two languages can only be carried out in two steps: from the source language into one of the major world languages (usually, English) and from that language into the target language. This approach is followed for bilingual purposes in the FrameNet project context (Boas 2002) where English is added to the six languages originally considered for this initiative (Baker et al. 1998, 2003). This strategy is feasible but has a serious drawback: the ambiguity of the intermediate language adds to the ambiguity of the source language which dramatically increases the error rate of the resulting translation. However, in many cases, the only alternative to such translation is no translation at all. Obviously, when creating lexicographic resources for a multilingual setting, building one multilingual dictionary which would embrace all working languages seems to be a much more practical approach than constructing a large series of bilingual dictionaries, each relating only two languages. Many multilingual dictionaries, printed or electronic, have been built, most of them covering a narrow subject domain. Multilingual dictionaries usually select one language as the leading one (or vedette). Data in all other working languages are translated into this one and in this way are connected to each other. Very often, English serves as the reference point. A typical example is the recently published ‘Glossary of art: Conservation, materials and techniques, museology’ (Xarrie´ 2006), which contains terminology pertaining to art in five languages. The author of the glossary has researched various bibliographical English, French, Spanish, Italian, and German sources and translated all the data into English. Thus, even in multilingual dictionaries the correspondence between the working languages is mostly established through an intermediate language— an interlingua—very much in the same way as it is done when connecting two languages by means of a couple of bilingual dictionaries. The problem of ambiguity is usually not taken into account in multilingual dictionaries, since most of them are terminological and the terminology is mostly unambiguous within the given subject domain (cf., e.g., Janssen 2004). However, the absence of ambiguity in closed subject domains should not be exaggerated. To a certain extent, terminology is also ambiguous, although the level of ambiguity is much lower than in the general lexicon. For example, the Dictionary of Agriculture (Haensch and Haberkamp 1986) covers the applicable terminology in six languages: English, French, German, Italian, Russian, and Spanish. In each of them one can easily find ambiguous terms, that is the terms that have at least two translations in at least one language. For example, French noix and Spanish nuez are translated into German as Nuss and Walnuss, and into English as nut and walnut; Russian lopata corresponds to German Schaufel and Spaten, French pelle and beˆche, Russian maslo is equivalent to English butter and oil, German Butter and O¨l,
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
73
French beurre and huile, Spanish mantequilla and aceite, Italian burro and olio. The list can easily be lengthened. As soon as one leaves the ‘one word–one sense’ area and tries to construct a multilingual dictionary for a more extensive fragment of language, the ambiguity problem stands out much more clearly. The failure of multilingual dictionaries to cope with the ambiguity problem seriously reduces their value. Let us illustrate this point with a simple example. Suppose we wish to build a multilingual dictionary which includes Spanish and Russian, and intend to use English as an intermediate language. If we take a moderately ambiguous English noun party, we will have to deal with the following typical situation. On the one hand, there are English–Spanish pairs party–partido (as in socialist party–partido socialista) and party–fiesta (as in to go to a party–ir a una fiesta). On the other hand, we will also find similar English–Russian pairs party–partija (as in socialist party–socialisticheskaja partija) and party–vecherinka (as in to go to a party–pojti na vecherinku). Obviously, one should see to it that English– Spanish pairs correctly match English–Russian ones in order to avoid wrong triples party–partido–vecherinka and party–fiesta–partija. When constructing a multilingual dictionary entry, one should make sure that all its members share a common meaning. This is only possible in two instances: (a) if the lexicographer knows all working languages equally well and can guarantee the synonymy of all members of the entry (a variant of this option is to have a team of lexicographers working together and sharing their competence) or (b) if the meaning to be conveyed is clearly represented in a way that makes it possible to find a correct equivalent in one language without necessarily understanding all the other languages. The second alternative requires that an interlingual representation should be drawn up which guarantees that all members of the dictionary entry have a common meaning.
Interlingual representation An interlingual representation can be thought of in different ways. One of the most obvious methods is to refer to a common meaning by means of a picture. In Germany and in France there is a long tradition of publishing picture dictionaries (Bildwo¨rterbu¨cher, dictionnaires visuels) (cf., e.g., Duden 2005; Corbeil and Archambault 2006). These dictionaries give detailed pictures of various complex objects (e.g. a human body, a car, a house, a farm, an electric plant, etc.) and provide equivalents in several languages for every part of this object. The language of pictures is an excellent interlingua. It is easy to use, convenient both for the dictionary developers and the users, and very efficient. Its only–but crucial!–defect is that it has very limited applicability and cannot be used for denoting abstract concepts that are difficult to draw. Some dictionaries of this type attempt to cover actions, but with poor results. For example, in the six-language picture dictionary Yiddish–Hebrew–English–German–Russian–French Picture Dictionary (Vaisman 1996), we encountered a picture of a dog and a duckling and would have
74
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
never guessed that the picture was intended to represent the concept of ‘chase’. One interesting approximation to a solution to this problem is the EuroWordNet initiative (Vossen 1998, 2001, 2004) (cf. the website of the project at www.illc.uva.nl/EuroWordNet/). It is a multilingual database for several European languages (Dutch, Italian, Spanish, German, French, Czech, and Estonian). The data for each language—a wordnet—are a collection of synsets (sets of synonymous words) with basic semantic relations between them. Each synset represents a concept that corresponds to one of the meanings of the members of this synset. Each wordnet, in its turn, represents a unique language-internal system of lexicalizations. In addition, all the wordnets are linked to an Inter-Lingual-Index, based on the WordNet for English (Fellbaum 1998). Via this index, the languages are interconnected so that it is possible to go from synsets of one language to similar synsets in any other language. The index also gives access to a shared top-ontology of sixtythree semantic distinctions. The shortcoming of this useful initiative is the lack of version control, which leads to the incompatibility of subsequent versions of WordNet with the one used as a pivot in EuroWordNet. This greatly hampers any attempt to use WordNet as a reference resource for the construction of multilingual dictionaries. However, some recent work is based on the use of Wordnet and LDOCE to define a kind of semantic interlingua focused in the definition of semantic frames (Green et al. 2004). Another attempt to construct an interlingua for semantic annotation has been undertaken in the IAMTC 2003 project (cf. the website of the project at http://aitc.aitcnet.org/nsf/iamtc/). The interlingua is built on the basis of a series of bilingual corpora. The source languages are Japanese, Korean, Hindi, Arabic, French, and Spanish, and the target language is English. More details on this interesting approach can be found in (Farwell et al. 2004; Mitamura et al. 2004). Thus, it seems that the most promising direction in which one can look for the solution to the multilingualism problem is the development of artificial interlinguas, see, for example, the LISA report (Dillinger and Lommel 2004), which testifies to a growing tendency to consider interlinguas to be the only solution to the massive multilingualism problem. Some interlinguas were produced and implemented in the 1980s for machine translation, although without any great success. The reason for the low efficiency of these attempts lay not so much in the interlingua concept as such, but rather in its application and in the computation capacities of the then existing systems. It is worthwhile to review the evolution of interlinguas in its simplest approach, that is, not as a means for contents representation but as a means for the creation of multilingual dictionaries. The vocabulary component of interlinguas is by far the most difficult part to model, especially for general vocabulary. In fact, different interlinguabased projects are built along two main lines: open vocabulary and a
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
75
domain-restricted one. As such, the concept of an interlingual dictionary does not exist, since any interlingual dictionary is an ancillary support for a Natural Language Processing (NLP) system. Let us give a brief overview of several dictionaries in interlingua-based systems. Machine Translation systems like ATLAS (Uchida 1989) and PIVOT (Muraki 1989) fall into the category of interlingua-based machine translation. Both can be considered as general-purpose machine translation systems. They were developed in Japan during the early 1990s. The ATLAS interlingua is composed of two types of units: concepts and a finite inventory of conceptual relations based on thematic roles along with a small number of relations that pertain to the descriptive dictionary (part_of, element_of, similar, equivalent, super_sub). The conceptual domain of the interlingua is not restricted beforehand, since any concept appearing in a text can be included into the interlingual dictionary. The interlingual dictionary in PIVOT is based on the so-called ‘Conceptual Primitives’, which correspond to concepts. However, the nature of conceptual primitives is left undefined and corresponds to lexicalized concepts in English rather than really primitive (elementary, indivisible) meanings. Interlingua-based machine translation evolved into the so called knowledge-based machine translation, exemplified by the KANT (Nyberg and Mitamura 1992) and Mikrokosmos (Beale et al. 1995) systems. Although knowledge-based machine translation systems rely on an interlingua, they manifest some important differences with those based on classical interlinguas. First, knowledge-based systems are restricted to specific domains. Second, interlinguas in knowledge-based machine translation systems rely on more sophisticated linguistic and knowledge representation theories, which can be implemented due to the limitations in the domain and vocabulary. In KANT, the interlingua is composed of a shared set of domain concepts. A concept is represented as a frame where semantic and argument framerelated information is included. Concepts of the interlingua are chosen by means of domain analysis techniques. On the other hand, Mikrokosmos is supported by an ontology (a set of domain language-independent concepts) that serves as the interlingua. Concepts in the ontology are hierarchically organized and interrelated by means of semantic relations. In addition, the representation of concepts is provided with its argument frame, selectional restrictions for arguments, world knowledge, among others. As can be seen, knowledge-based machine translation highlights the knowledge representation dimension of the interlingua, adopting an ontological and frame-based approach for the definition of the concepts. The burden of such an intense and detailed knowledge-based conceptual modelling can only be afforded in specific domains and for a limited number of language pairs. Current interlinguas are illustrated by the IAMTC already mentioned (Reeder et al. 2004). The IAMTC interlingua resorts to a progressive definition of the interlingua as long as it is needed. However IAMTC uses an underlying
76
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
ontology (OMEGA v3), which makes the expansion of the interlingua dependent on the expansion of a general-domain ontology. The work described in this paper is closely related to another currently used interlingua—UNL (Universal Networking Language), a language designed to represent informational content conveyed by natural languages, and to be processed by computers (Uchida 2006). It was introduced in 1996 by the United Nations University in the framework of the UNL Programme (UNDL Foundation, see www.undl.org) aimed at supporting massive multilingualism on the Internet. UNL is composed of three main elements: a set of concepts called Universal Words (hereafter referred to as UWs), relations, and attributes. UWs form the vocabulary of the interlingua, relations express thematic roles, and attributes represent the context and speaker-dependent information. The UWs model was the one we selected as a starting point for an interlingual representation to be used for the construction of domain-specific multilingual dictionaries. From UNL we borrowed the idea that unambiguous interlingual concepts could be conceived as restricted meanings of natural language words. In the present paper, we apply this idea to the interlingual representation of terms for multilingual dictionaries. The salient features of our approach are its relative simplicity and lower cost, as compared with traditional methods of multilingual lexicography, obtained without detriment to quality.
FOUNDATIONS OF UNIVERSAL WORDS (UWs) In the UNL System, the vocabulary of the language is constituted by the set of UWs. Below, we will focus our presentation on this vocabulary and not on the rest of the UNL language. The set of UWs does not require a subjacent ontology to build its lexical resources, thus avoiding the time-consuming burden of creating and maintaining an ontology. Besides, UWs are not as undefined as the vocabulary of early interlinguas such as ATLAS or PIVOT. More concretely, a UW is composed of an English word along with a list of semantic restrictions. The main purpose of semantic restrictions is to eliminate lexical ambiguity as well as overcome discrepancies between lexical meanings of different languages. Besides that, they establish major lexical relations with other words and specify the argument frame of the predicates. In this way, UWs allow disambiguated lexical meaning to be represented. For example, English state has several meanings. Among them, (a) ‘the group of people comprising the government of a sovereign state’ as in The state has lowered income tax, (b) ‘the territory occupied by one of the constituent administrative districts of a nation’ as in His state is in the deep south, (c) ‘to express in words’ as in State your opinion and some others. The apparatus of UWs ensures clear differentiation between these meanings: the UW for (a) is state(icl4government4thing), for (b)—state(icl4administrative_ district4thing) and for (c)—state(icl4express4do, agt4thing, obj4thing). The UWs are divided in two parts: the headword and the list of
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
77
state (icl>express>do, agt>thing, obj>thing Headword
List of semantic restrictions separated by commas
Figure 1: Different parts of a UW semantic restrictions enclosed in parentheses and separated by commas, as shown in Figure 1 for UW (c). The semantic restrictions have the following meaning:
icl4express4do: (where icl stands for included) establishes the class/ subclass relation: state is included in the class of express-situations, which in its turn is included in the class of actions. agt4thing, obj4thing: (where agt stands for agent and obj for object) establish the obligatory semantic arguments of the predicate to state.
Restrictions in UWs express lexical relations between the terms, namely, class/ subclass relation (cf. icl relation in Figure 1), synonymy (equ relation), antonymy (ant relation), and meronymy, or part/whole relation (pof). Besides, the semantic arguments of predicates (that is, verbs, some nouns, adjectives, and adverbs) must be specified. Since UWs are described by means of relations between terms, the result is a connected net of UWs, constituting the UW system. A more comprehensive view of the UW system is described in Boguslavsky et al. (2005) and also in the specifications of the UNL system (Uchida 2006).
Local dictionaries The general UW Dictionary constitutes the complete repository of UWs. The link between UWs and the words of local languages is done in the so-called local dictionaries. These dictionaries contain pairs of the type 5local word, UW4 along with grammatical information necessary for language analysis and generation. The relation between local and the general UW dictionary is shown in Figure 2.
Subjacent aspects of the UW system The semantic restrictions of UWs organize them into a hierarchy of concepts, linked by the hyperonymy, instance-of and synonymy relations. The resulting hierarchy cannot be considered an ontology in its strict sense, since UWs do not attempt to be completely language-neutral, independent, self-explaining concepts. They are just a label for a concept that corresponds to a given word sense in a given natural language. Upper levels of the hierarchy are languageneutral. Labels such as ‘thing’ (standing for any nominal entity), ‘do’ (verbal concept denoting an action or an activity), ‘occur’ (any verbal concept denoting a process) or ‘be’ (verbal concept denoting a state or a property) can equally well subsume concepts of any language.
78
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
Spanish mesa NOUN, COUNT, FEMENINE, REGULAR MORPH,… UW: table(icl>furniture>thing)
UW DICTIONARY
armario NOUN, COUNT, MASCULINE, REGULAR MORPH… UW: wardrobe(icl>furniture>thing) . .
LOCAL DICTIONARIES
Figure 2: Relation between local dictionaries and the UW Dictionary However, as far as terminal leaves of the hierarchy are concerned, UWs do not imply language neutrality. Instead, the UWs situated at the lower levels of the hierarchy are a collection of word senses of all working languages. Thus, UWs do not form a closed set, since new UWs can be defined for new texts. When a concept of the local language differs from the English one, the meaning of the headword is modified by means of restrictions, so that the resulting UW better suits the local concept. To illustrate this important point, let us consider some examples. In Russian, there is no neutral equivalent of the English non-causative verb to marry as represented in sentences such as John married Ann in June. The expression that exactly corresponds to this English verb—vstupat’ v brak (‘to contract a marriage’)—is an official term and is not used in everyday life. Instead, Russian speakers make use of two different expressions: zhenit’sja, if the agent of the action is a male, and vyxodit’ zamuzh, if it is a female. Since the English and the Russian words differ in their meaning, they generate different UWs. The UW for English to marry looks like (1), while Russian expressions have UNL equivalents with a narrower meaning—(2) and (3), respectively (for simplicity’s sake, only the relevant fragments of the UWs are given). 1 marry(icl4do, agt4human) 2 marry(icl4do, agt4male) 3 marry(icl4do, agt4female) In very much the same way, English makes no distinction between two concepts lexicalized differently in Spanish—rinco´n ‘an interior corner’, as in the corner of the room and esquina ‘an exterior corner’, as in the street corner. To represent these specific concepts, the English concept of corner has also to be modified with restrictions: 4 corner(icl4area, equ4interior angle) 5 corner(icl4area, equ4intersection)
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
79
The absence of a fixed set of semantic restrictions ensures the flexibility necessary to overcome lexical mismatches. Thus, the UW Dictionary is considered as a repository of UWs coming from different languages where subsumption relations are established by means of semantic restrictions. This kind of flexibility distinguishes UWs not only from existing ontologies, as mentioned above, but also from various meaning representation languages in which lexical meanings are decomposed by means of a small set of semantic primitives or fully defined and specified a priori.
Requirements for UWs The fact that the UW system is flexible and that restrictions are not limited a priori does not mean that UWs are devoid of certain quality properties, namely: 1 Constraints assigned to the Headword should ensure: – indication of a hypernym or an instance-of UW (ontological function); – effective differentiation from all other senses of the Headword (semantic function); – indication of all semantic arguments of the UW (argument frame function). 2 There should be no incidental UWs. If two UWs have the same headword, they should denote clearly different concepts. 3 UWs should be easy to understand; in non-obvious cases they should be supplied with clarifying examples and/or comments in the UW Dictionary. Of course, the broad labels seen so far are not always sufficient to serve the semantic function of UWs. More elaborate restrictions are sometimes required for UWs to maintain human readability of UWs while effectively differentiating word senses. The flexibility inherent in the UW system serves to handle lexical mismatches well but has a side-effect of the decentralization of the UW general dictionary, resulting in making the common body of UWs difficult to maintain.
BUILDING THE GENERAL UW DICTIONARY When constructing a multilingual dictionary, it is preferable to establish the correspondence between the languages via an intermediate language. In this case, the lexicographers are not required to know all working languages. As we have seen before, the intermediate language is liable to ambiguity, and to ensure synonymy between the terms of all working languages,
80
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
the expressions of this intermediate language must denote a single concept unambiguously. For that, we propose that concepts are represented by means of UWs, since UWs represent concise disambiguated concepts and are easy to understand and to construct. The General UW Dictionary is meant to constitute a large repository of UWs, conceived as a public resource easy to maintain and access. Since UWs are English headwords accompanied by a list of semantic restrictions, we have exploited available free-access resources of English to automatically create a sufficiently large set of UWs. In fact, the main English lexical resource used is WordNet (Fellbaum 1998). WordNet is a large lexical database of English, freely downloadable from the Internet (http:// wordnet.princeton.edu/). We made use of the similarity of WordNet and the UW system and employed WordNet as the main source for creating a UW dictionary. The complete process and the final UW dictionary are described in the following sections.
A view on Wordnet WordNet is ordered conceptually, by means of semantic relations. The main organizing entity in WordNet is the synset, defined as a group of cognitive synonyms that express a single concept. Synsets are interconnected by means of lexico-semantic relations like hyperonymy (hierarchical relation between class and subclass), antonymy (an opposite term), meronymy (part-of) and other relations like pertaining_to, sentence frames for verbs, etc. Figure 3 shows two samples of WordNet that illustrate the relations of hyperonymy and antonymy for the synset ‘male child, boy’. WordNet includes nouns, adjectives, adverbs, and verbs. Other categories such as prepositions, determiners, or conjunctions are spelled out from WordNet, since they do not denote any semantic concept.
Synset composed of two terms. The synset denotes a single concept Sense 1 male child, boy -- (a youthful male person; "…; => male, male person -- (a person … => person, individual, someone, somebody, … Relation of Hyperonymy (=>) between two synsets. Relation of Antonymy between two synsets: Sense 1 male child, boy -- (a youthful male person; .. Antonym of girl (Sense 2) =>female child, girl, little girl )
Figure 3: Example of Lexical Relations in Wordnet
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
81
Equivalences between Wordnet and the UW system The main design issue when considering a UW dictionary and WordNet as the data source is that the structure of lexical relations in WordNet can be used to construct the list of restrictions of UWs. To do that, we must first establish the main similarities between WordNet and the UW system. These similarities are exposed in Table 1, where the first column describes elements of WordNet, the second column states their equivalents in the UW system, and the third column establishes the function (ontological, semantic, or argument frame) that the proposed restriction serves. Table 1 shows how a word included in WordNet can be used to construct a UW. WordNet synsets are taken to represent separate senses of the words they are composed of. Each synset is delimited by means of a set of synonyms, hypernyms, antonyms, and other lexical relations associated with that synset, in the same way as the sense of a headword of a UW is delimited by a list of semantic restrictions. It is important to explain the relationship between WordNet and the dictionary of UWs. We used WordNet as a computerized lexicographic resource that contains disambiguated word senses supplied with several paradigmatic relations with other words and examples of use. These data were extracted from WordNet and remodelled into UWs. Insofar as this information is concerned, UWs may be viewed as simplified synsets. The crucial difference between WordNet and the UW Dictionary is determined by the fact that the former is not comprehensive. There are many classes of lexical units that are not sufficiently represented in WordNet. For the purposes of our topic, the most important gap is the absence of many specialized terms, especially consisting of two or more words. UWs for which no WordNet equivalent can be found are constructed by the expert manually. Another difference which makes UWs more suitable for serving as an interlingua than WordNet is their compactness. They are short and easy to use, which is an important advantage for practical lexicography. They convey all semantic information by means of a headword supplied with several
Table 1: Correspondence between WordNet and the UW system WordNet 2.1
UW system
Function
A member of a synset A word of the same synset Hypernym Hyponym Antonym Pertaining to
Headword Relation equ4 Relation icl4 Relation icl5 Relation ant4 Relation com4
— semantic ontological semantic semantic semantic
82
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
restrictions, while WordNet synsets and a network of paradigmatic relations they are involved in form a much more cumbersome structure (cf. Figure 3). This set of correspondences allows us to devise a method that defines UWs in a systematic way using WordNet. The method can be summarized as follows. 1 Extract a Word from WordNet 2 Find all Synsets where this Word appears 3 For each occurrence of the Word in a Synset of the Word, do the following: – Take the Word as the Headword of a UW – Depending on the syntactic category (noun, adjective, adverb, verb) and on the data obtained from WordNet, generate the semantic restrictions according to the correspondences in Table 1 – Take the Headword and the restrictions, and construct the complete UW – Store the UW in the dictionary There are two aspects that require further explanations in this method. The first regards the number of UWs that are created per word. This method will generate one UW per word sense. For example, the word ‘boy’ as a noun has four senses and thus generates four different UWs. In some cases, when the difference between the senses is too subtle, WordNet relations are not sufficient to differentiate between them (producing duplicate UWs). In these cases, the method will generate identical UWs for different senses. These UWs must be treated in a special way. Thus, the noun ‘boy’ in its first sense produces the following semantic restrictions:
icl4male4thing (by means of Hypernym and ‘icl4’ correspondence) equ4male_child (by means of Synonym and ‘equ4’ correspondence) ant4girl (by means of Antonym and ‘ant4’ correspondence)
The final UW is the concatenation of the generated semantic restrictions: boy(icl4male4thing, equ4male_child, ant4girl) The order of semantic restrictions is conventional: a different ordering will not imply different semantics of a UW. The second question refers to verbal UWs. Whereas all the information required for creating UWs for nouns, adverbs, and adjectives is present in the WordNet, the mapping between verbal UWs and verbs in WordNet is not straightforward. This is due to the following reasons:
Verbal UWs are categorized into three basic types of events: actions (restriction ‘icl4do’), processes (restriction ‘icl4occur’) and states (restriction ‘icl4be’). This categorization is absent in WordNet.
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
83
Verbal UWs should be provided with semantic arguments. Verbs in WordNet are assigned a Sentence Frame, a description of syntactic arguments of the verbs, which is often incomplete.
Since there is no one-to-one relation between verbal UWs and the verbs, it is necessary to infer the type of event and the semantic arguments from the scarce information present in WordNet. For that, we made use of the socalled lexicographic files which define broad ontological categories. Some of these categories are ‘verbs of dressing and bodily care’, ‘cognition verbs’, ‘verbs of being and having’. The combination of the ontological category together with the sentence frame of a verb gives us a hint about its type of event and semantic arguments. Table 2 shows an excerpt of the combinations that have been used to define verbal UWs.
Table 2: Combinations to define verbal UWs from Wordnet Ontological category in WordNet
Sentence frame in WordNet
UW event type
Proposed semantic arguments
Example of UW
verbs of being, having verbs of weather. . .
Somebody — to somebody Somebody —
be occur
aoj4thing, obj4thing obj4thing
verbs of creation. . .
Somebody — something
do
agt4thing, obj4thing
conform (icl4be, aoj4thing, obj4thing) steam (icl4occur, obj4thing) cut (icl4do, agt4thing, obj4thing)
The UWs General Dictionary software application The common set of produced UWs is managed as a public resource. It can be freely accessed on the Internet at: http://www.unl.fi.upm.es/dicweb. Given a headword, the dictionary application returns the set of UWs having this headword. A screenshot with the UWs generated for the noun chair is given in Figure 4. Examples provided in the WordNet are assigned to synsets and not to individual words. These examples illustrate the synset as a whole and, accordingly, the meaning of the whole bunch of UWs constructed from the synset. Therefore, as can be seen in Figure 4, a UW may be illustrated with examples that do not contain its headword but nevertheless help to identify its meaning.
84
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
Figure 4: Screenshot of the dictionary application The software application has the Conversor Module and the Database as main components. The conversor module is the software component in charge of generating the complete set of UWs using the data from WordNet and the rules of Tables 1 and 2. The set of generated UWs together with their corresponding comments and examples constitutes the UW General Dictionary which is stored in a database. In order to facilitate the task of dictionary search and updating, access to the dictionary database is done via the Internet. Thus, the only requirement for using the dictionary is to have access to the Internet (no other installation process is required).
Results and evaluation All the UWs have been created automatically, without human intervention. The very first aspect to evaluate is the existence of duplicated UWs, that is, the cases in which different senses of the word generate identical UWs. The method yields a total amount of 207,016 UWs. Table 3 shows the total number of UWs (divided into unique and duplicated) according to their type (nominal, verbal, adjectival, and adverbal). As can be seen from Table 3, the rate of duplicate UWs for nouns is less than 2 per cent, a good result for the most polysemous syntactic category. Thus, we considered that nouns have a good coverage in the general dictionary and the resulting UWs can be considered as unique and significant. Nouns are by far the most elaborated category both in WordNet and consequently in our general dictionary. This poses a rather satisfactory scenario for the creation of multilingual lexical resources such as thesauri or terminological dictionaries, mainly composed of nouns. On the other hand, both adjectives and adverbs yield quite a high rate of duplicated UWs (about 17 per cent). The reason is that the relations provided in the WordNet for these categories (mostly, synonymy and hyperonymy) are insufficient for disambiguation. We could think of a possible solution to
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
85
Table 3: Obtained results
Unique UWs Duplicate UWs %
Nouns
Adjectives
Adverbs
Verbs
All cat.
142,343 2,761 1.93
26,748 4,518 16.9
4,958 762 15.4
23,716 1,174 4.9
197,765 9,215 4.45
reduce to the minimum duplicate UWs by means of the inclusion of new rules and therefore new semantic restrictions in the definition of a UW. However, an important feature of any UW is human-readability, which may be obscured by an excess of semantic restrictions. Take for instance, the case of hyponymy (icl5), this relation may effectively disambiguate the senses of two different UWs; however, in terms of legibility, its role is rather questionable. The rate of duplicated verbal UWs is less than 5 per cent, which is not bad. On the other hand, as mentioned above, verbal UWs are often devoid of some semantic restrictions (reflecting the event type and the argument frame), since WordNet does not contain this information. This is a serious defect if UWs are to be used within the machine translation framework. However, for the purposes of constructing terminological dictionaries, this is hardly a drawback, because this missing information is irrelevant for finding translation equivalents for terms.
METHOD FOR CONSTRUCTING MULTILINGUAL DICTIONARIES We have already pointed out that the bilingual approach to matching more than two languages creates additional ambiguity, difficult to solve in many cases. The confrontation of terms coming from different languages and the agreements cannot be based on the subjective interpretation of human experts with respect to word senses, especially if the accuracy and standardization of the process are pursued.
The method step-by-step The steps for creating a multilingual dictionary based on UWs are as follows: 1 Terminology extraction. Traditional methodologies for terminology extraction are used, since this step is out of the scope of this method. 2 Terminology identification. Experts in the corresponding domain must identify the terminology of the domain if it is not general. 3 UW construction. For each identified term in a source language L, its equivalent UW must be constructed with the help of the General UW Dictionary. To do that: – for any word, the expert in language L, with command of the English language, consults a bilingual L–English dictionary and arrives at the
86
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
equivalent English word. The English word will be the input headword in the General UW Dictionary. – The complete set of UWs with the inputted headword is retrieved and shown to the user. The UW that better fits the sense of the original word of L is picked up. The General UW Dictionary is intended to be used as a help when defining UWs. The expert is thus relieved of the responsibility of creating new UWs from scratch, he/she is asked to pick up the correct UW from a common repository. 4 Construction of the pairs (WordL, UW). Once the expert has chosen the most adequate UW, he/she creates a table with the pairs (Word of L, UW). The pair is extended with the glosses and examples of use present in the General UW Dictionary. 5 Translation of the UWs into other working languages. The UW will be translated into the other languages. Its translation into English is straightforward, whereas its translation into other languages will have to consider the list of semantic restrictions of the UW. When generating the equivalent terms in Italian, French, or Arabic, the experts will need just a working knowledge of English and of course a good command of the target language. Knowledge of the initial language L (the language where the original term comes from) is not needed. This could be the main difference when confronted with traditional multilingual dictionaries. In essence, what really characterizes the method is the representation of concepts. A concept is not represented by a set of three or more words in working languages, but by a single UW, and terms from target languages are defined as the best equivalent for the UW. This very simple method allows constructing linguistically accurate multilingual dictionaries, permits scalable maintenance and is the optimal solution to define domain-specific multilingual dictionaries, of great utility in the software localization industry, translations of technical documentation, etc.
Evaluation of the process In order to evaluate the method, we have devised an experiment that has as its outcome the multilingual dictionary. The experiment proves that it is possible to construct multilingual dictionaries without knowing all the working languages, just the target languages and English (so as to understand UWs). This possibility is very attractive. It should be stressed here that the proposed method is more important than the actual dictionary of UWs that we used for our experiments. Although very large, this dictionary contains general vocabulary, while our immediate aim is to construct a specialized dictionary for a specific subject domain.
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
87
The experiments we carried out consisted in the application of the proposed method to several data samples and the verification of the results obtained.
Stage one A sample of 100 terms is extracted from the selected subject domain (cultural heritage). The size of 100 terms/sample is more than sufficient to be statistically representative of much larger populations than the terminology of a subject domain (estimated in around 3000–4000 terms as maximum). The cultural heritage corpus we used consists of more than 600 files containing a total of 2,600 specialized terms formed by one, two, or three words. The source language of the texts is Spanish.
Stage two These terms were converted into UWs by an expert linguist working with UWs. We used the UW Dictionary described in the previous section. Each UW was supplemented by the definition and an illustrative example. The data facilitated the task of selecting a UW appropriate for the context and the subject domain. The quality of the output of this stage is essential for the rest of the process, since the adequacy of the target language’s equivalents will depend on the proposed UWs, which is why this phase of UW production was carried out by an expert in UWs. Of course, some terms of the cultural heritage domain, especially the ones composed of two or more words, are not present in the general dictionary and had to be manually constructed by the expert. An example of this is given in Table 4 (cf. cataloguing). At this moment, we are designing a method of producing UWs for multiword terms semi-automatically. The result of this preliminary phase is illustrated in Table 4.
Stage three A table with UW-related data (that is, a UW, the definition, and an example of use) was handed over to two other UW specialists, native speakers of Russian and Arabic. Their task was to translate the UWs provided into Russian and Arabic, respectively. Tables 5 and 6 illustrate the tables produced in this stage. Then, Russian and Arabic equivalents of the UWs were brought together in a new table and sent to an expert in both languages, Russian and Arabic, for evaluation (see Table 7). The task was to assess the quality of translation in the range of 0 to 10. The table for evaluation contains the definition and example field, so as to contextualize the proposed pair. The evaluation results of one sample are presented in Table 8, showing the frequency of the scores. It is easy to see that the most frequent score is 10, and that less than 5 per cent are under score 5.
Campan˜as anuales para la catalogacio´n de los bienes muebles del Patrimonio Histo´rico Espan˜ol de titularidad eclesia´stica. Oferta de venta irrevocable. Ofertas directas de venta.
catalogacio´n
oferta
coordinacio´n entre la Administracio´n del Estado y las Comunidades Auto´nomas en materia de Patrimonio Histo´rico . . .
Context (fragment of the sentence where the word appears)
administracio´n
Spanish
bid (icl4offer4thing)
cataloguing (icl4classification4thing)
administration (icl4body4thing)
Proposed UW
Table 4: Output of the terminology extraction and UW construction phase
the persons (or committees or departments, etc.) who make up a body for the purpose of administering something EX: ‘he claims that the present administration is corrupt’ The action of making an itemized list or a catalogue of something. NOT COMING FROM GENERAL DICTIONARY a formal proposal to buy at a specified price
Definition and example of use
88 CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
89
Table 5: Pairs UW–Russian as produced by the Russian specialist Proposed UW
Definition and example of use
Russian
administration (icl4body4thing)
the persons (or committees or departments, etc.) who . . . The action of making an itemized list or a catalogue . . . a formal proposal to buy at a specified price
‘dkhlhpqo‘uh~
cataloguing (icl4 classification4thing) bid (icl4offer4thing)
i‘q‘jmchg‘uh~
noedj‘c‘ek‘~ uel‘
Table 6: Pairs UW–Arabic as produced by the Arabic specialist Proposed UW
Definition and example of use
administration (icl4body4thing)
the persons (or committees or departments, etc.) who . . . The action of making an itemized list or a catalogue . . . a formal proposal to buy at a specified price
cataloguing (icl4classification4thing) bid (icl4offer4thing)
Arabic
Table 7: Table for evaluation Russian ‘dkhlhpqo‘uh~ i‘q‘jmchg‘uh~ noedj‘c‘ek‘~ uel‘
Arabic
Definition and example
Score
the persons (or committees or departments etc.) who . . . The action of making an itemized list or a catalogue . . . a formal proposal to buy at a specified price
[0–10] [0–10] [0–10]
As can be seen from Table 8, the translation results obtained in this sample, which are representative of other samples as well, are very promising. In 84 per cent of the cases the translation quality has been estimated as ‘perfect’ (perfect is estimated with scores of 10 and 9). ‘Acceptable’ means that it is accepted in principle but maybe it will be revised. The scores needed to reach ‘acceptable’ are 7 and 8. Scores of 5 and 6, entered as ‘need revision’, mean that they must be revised. Finally, scores of 0, 1, 2, 3, and 4 mean that they have to be defined completely by bilingual experts. Just a single term has been rejected as fully inadequate (achieving score 0).
90
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
Table 8: Evaluation results Score Score Score Score Score Score Score Score Score
10 9 8 6 5 3 2 0
Absolute frequency
Qualitative estimation
80 4 4 3 5 1 2 1
Perfect Perfect Acceptable Need revision Need revision Do it again manually Do it again manually Do it again manually
This Arabic–Russian dictionary proved to be highly reliable, which opens good prospects for compiling multilingual dictionaries with more languages. It should be emphasized that we are not referring to unidirectional dictionaries, that is an Arabic–Russian dictionary as opposed to a Russian– Arabic one. Our method provides equivalents in different languages for a single unambiguous concept (represented by means of a UW) and, therefore, these equivalents should be valid in both directions. The approach presented in this paper was supported by UNESCO, who sponsored a small project aiming to generate multilingual dictionaries for six official languages of the United Nations. This project was oriented towards encoding into the UNL interlingua a number of texts related to the safeguarding of world heritage and their generation into the official languages of UNESCO. As a side effect, the project yielded the vocabulary of the six languages linked by UWs. The idea of constructing multilingual dictionaries (mainly, domain dependent) supported by UWs is also being explored in the PATRILEX project (Spanish Council of Research with reference HUM2005–07260). One of the main objectives of the project is the construction of a multilingual dictionary based on the concept of UW restricted to the Cultural Heritage domain. More concretely, the PATRILEX project proposes a method for (i) obtaining equivalences among terms from different languages with relative ease and (ii) the systematization in the construction of multilingual lexicons. The approach described in this article is currently used in this project due to the good results demonstrated and systematically applied to the total contents of about 1,000 web pages of the Spanish Ministry of Culture dealing with cultural heritage. UWs have served as the pivot for producing a multilingual dictionary for five languages (Spanish, Russian, English, Arabic, and German) in the first phase, and other languages in the near future. This dictionary will thus include ten bilingual dictionaries in the cultural heritage domain, hopefully with a very low error rate. Complete results and the
IGOR BOGUSLAVSKY, JESU´S CARDEN˜OSA, and CAROLINA GALLARDO
91
corresponding multilingual dictionaries will be placed on the web in the first quarter of 2009.
CONCLUSIONS AND FUTURE WORK Multilingual dictionaries are of great utility for a large spectrum of multilingual applications. Given the lack of multilingual experts, such dictionaries are usually built via an intermediate language—an interlingua. If it is a natural language that serves as an interlingua, its ambiguity seriously hinders the construction of the dictionary and reduces its quality. In this paper a method is proposed for building multilingual dictionaries on the basis of an artificial unambiguous interlingua—Universal Words—of acceptable quality without the availability of experts that know the equivalences among pairs of target languages. Part of the immediate work is to finish the definition and experimentation for all the materials of this work based on contents with almost 250,000 words. This massive work is based on the preliminary success of the approach described in this article. Now, we will finish the terminological work when the method to create specialized composed UWs has been tested and defined in terms of accuracy and efficiency. Multilingual dictionaries based on bidirectional equivalences have a high added value due to the importance of building multilingual search engines for the web and for specialized documentation centres. The dictionaries built according to the method described below can be used in a wide range of applications including human translation.
ACKNOWLEDGEMENTS We would like to thank Adriana Toni and Tatiana Frolova for their collaboration in the experiment, guaranteeing the quality of the translation of the UWs into Arabic and excellent evaluation of the Russian–Arabic pairs. We also want to express our gratitude to Jorge Vergara who developed the terminology extraction tool from texts and Juan Bekios who developed the database of UWs.
REFERENCES Baker C. F., C. J. Fillmore, and J. B. Lowe. 1998. ‘The Berkeley FrameNet Project’ in C. Boitet and P. Whitelock (eds): Proceedings of the Thirty-Sixth Annual Meeting of the Association for Computational Linguistics and Seventeenth International Conference on Computational Linguistics. San Francisco, CA: Morgan Kaufmann Publishers. Baker, C. F., C. J. Fillmore, and B. Cronin. 2003. ‘The structure of the FrameNet Database,’ International Journal of Lexicography 16/3: 281–96.
Beale, S., S. Nirenburg and G. Mahesh. 1995. ‘Semantic analysis in the Mikrokosmos Machine Translation Project’ in: Proceedings of the Second Symposium on Natural Language Processing (SNLP95). Bangkok, Thailand. Boas, H. C. 2002. ‘Bilingual FrameNet Dictionaries for machine translation’ in M. Gonza´lez and C. P. Sua´rez (eds): Proceedings of the Third International Conference on Language Resources and Evaluation. Las Palmas, Spain. Vol. IV, pp. 1364–71.
92
CREATING DISAMBIGUATED MULTILINGUAL DICTIONARIES
Boguslavsky, I., J. Carden˜osa, C. Gallardo, and L. Iraola. 2005. ‘The UNL initiative: An overview,’ Lecture Notes in Computer Science 3406: 377–87. Corbeil, J. C. and A. Archambault. 2006. Le mini visuel: dictionnaire franc¸ais–anglais. Montre´al: Que´bec Ame´rique. Dillinger, M. and A. Lommel. 2004. ‘LISA, Best practice Guide. Implementing machine translation. Localization Industry Standards Association.’ Available online at www.lisa.org/pro ducts/bestPractice/. DUDEN. 2005. Bildwo¨rterbuch der deutschen Sprache. 6. Auflage. Mannheim, Wien, Zu¨rich. Farwell, D., S. Helmreich, B. Dorr, N. Habash, L. Levin, T. Mitamura, F. Reeder, K. Miller, E. Hovy, O. Rambow, and A. Siddharthan. 2004. ‘Interlingual annotation of multilingual text corpora’ in A. Meyers (ed.): Proceedings of the North American Chapter of the Association for Computational Linguistics, Workshop on Frontiers in Corpus Annotation. Boston, MA: Association for Computational Linguistics. Fellbaum, C. (ed.) 1998. WordNet: An Electronic Lexical Database. Language, Speech, and Communication Series. Cambridge, MA, MIT Press. Green, R., B. Dorr, and P. Resnik. 2004. ‘Inducing frame semantic verb classes from WordNet and LDOCE’ in: Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL’04), Main vol., pp. 375–82. Barcelona, Spain. Haensch, G. and Haberkamp de Anton, G. 1986. Dictionary of Agriculture in Six Languages: German, English, French, Spanish, Italian, Russian. Amsterdam; New York: Elsevier. Janssen, M. 2004. ‘Multilingual lexical databases, lexical gaps, and SIMuLLDA,’ International Journal of Lexicography 17/2: 137–54. Mitamura, T., K. J. Miller, B. J. Dorr, D. Farwell, N. Habash, L. Levin, S. Helmreich, E. Hovy, L. Levin, O. Rambow, F. Florence, and A. Siddharthan. 2004. ‘Semantic annotation of multilingual text corpora’ in: Proceedings of the Workshop on Beyond Named Entity
Recognition: Semantic Labelling for NLP Tasks. LREC. Lisbon, Portugal. Muraki, K. 1989. ‘PIVOT: Two-phase machine translation system’ in: Proceedings of the Second Machine Translation Summit. Tokyo. Nyberg, E. H. and T. Mitamura. 1992. ‘The KANT system: fast, accurate, high-quality translation in practical domains’ in: Proceedings of the 14th International Conference on Computational Linguistics (COLING’92) Nantes, France. Vol. 4, pp. 1254–8. Reeder, F., B. Dorr, D. Farwell, N. Habash, S. Helmreich, E. Hovy, L. Levin, T. Mitamura, K. Miller, O. Rambow, and A. Siddharthan. 2004. ‘Interlingual annotation for MT development,’ Lecture Notes on Computer Science, 3265: 236–45. Tsakona, V. 2007. ‘Bilingualisation in practice: Terminological issues in bilingualising a specialised glossary,’ International Journal of Lexicography 20/2: 119–54. Uchida, H. 1989. ‘ATLAS-II: A machine translation system using conceptual structure as an Interlingua’ in: Proceedings of the Second Machine Translation Summit. Tokyo. Uchida, H. 2006. Universal Networking Language (UNL). Specifications version 2005. Edition 2006. Available on line at www.undl.org/unlsys/unl/ unl2005-e2006/. Vaisman, E. B. 1996. Yiddish–Hebrew–English– German–Russian–French Picture Dictionary. Available online at www.ibiblio.org/yiddish/Vort/ vort-m_files/frame.htm. Vossen, P. 1998. ‘Introduction to EuroWordNet,’ Computers and Humanities 32: 73–89. Vossen, P. 2001. ‘Condensed meaning in EuroWordNet’ in P. Bouillon, and F. Busa (eds): The Language for Word Meaning. Cambridge: Cambridge University Press, pp. 363–83. Vossen, P. 2004. ‘EuroWordNet: A multilingual database of autonomous and language-specific wordnets connected via interlingual index,’ International Journal of Lexicography 17/2: 161–73. Xarrie´, M. 2006. Glossary of Art: Conservation, Materials & Techniques, Museology. Barcelona: Balaam.
Applied Linguistics 30/1: 93–114 ß Oxford University Press 2009 doi:10.1093/applin/amn050 Advance Access published on 29 January 2009
Interpreting Inexplicit Language during Courtroom Examination JIEUN LEE Macquarie University, Australia Court interpreters are required to provide accurate renditions of witnesses’ utterances during courtroom examinations, but the accuracy of interpreting may be compromised for a number of reasons, among which is the effect on interpretation of the limited contextual information available to court interpreters. Based on the analysis of the discourse of Korean–English interpreting in Australian courtrooms, this article examines how inexplicit language used by Korean-speaking witnesses affects the accuracy of court interpreting. Such use of inexplicit language is a consequence of differences between the lexicogrammatical system of the witnesses’ language (Korean) and that of the court (English), as well as being due to the witnesses’ lack of familiarity with the courtroom discourse. This study demonstrates that the way the court interpreters cope with inexplicit language may result in inaccurate rendition of the evidence, and thus have legal implications for court proceedings.
INTRODUCTION Court interpreting may be provided at various stages of court proceedings from empanelling the jury to sentencing, but in this article the term court interpreting is limited to interpreting provided during witness examination in the courtroom of English-speaking countries. It is important that the court hears in the target language the oral evidence given in the language other than English rendered as closely as possible, in terms of content as well as the register and style, to that provided in the original utterances. It has been argued that in an adversarial court system the speech style of the witness affects the power of testimony, namely its credibility and convincingness (O’Barr 1982). As such, the court interpreter’s alterations of the style of original speech in interpreted renditions may have implications for the court proceedings, in particular by influencing the credibility of the witness and the power balance in adversarial courts (Hale 1997a, 2004; Fraser and Freedgood 1999; Berk-Seligson 2002). Nevertheless, discrepancies between the semantic and pragmatic equivalence of original speech by witnesses from culturally and linguistically diverse backgrounds (henceforth, CALD witnesses) and the interpreted rendition is not at all uncommon (see Hale 1997a, 1997b, 2002, 2004; Berk-Seligson 2002; Jacobsen 2003). An additional complexity arises where CALD witnesses with low socio-economic and educational backgrounds produce non-standard and
94
COURT INTERPRETING OF INEXPLICIT LANGUAGE
dialectically variant linguistic features, or include—as any witness can— ambiguous and confused speech in their testimony. Such incoherent1 and ambiguous speech is perceived by court interpreters as the greatest challenge in court interpreting (Hale 2004, 2007). While CALD witnesses’ pragmatic markers, such as repetition, hedging, and hesitation, that have drawn attention in the court interpreting studies generally do not affect understanding of the semantic meaning of utterances, incoherent and ambiguous speech may hinder such understanding. Despite the potential significance of incoherent and ambiguous speech in courtroom contexts, this topic has not been the subject of close analysis in the interpreting literature. In particular, there has been no research into how such speech by some CALD witnesses may be perceived as difficult to interpret, and how such difficulties can be dealt with by court interpreters. Focusing on the ambiguity contained in some CALD witnesses’ evidence, this article will examine first why such speech may be challenging for court interpreters, and second, what strategies interpreters use to cope with these situations, which may have implications for court proceedings.
Context in court interpreting Context plays a key role in determining the meaning of utterances (Linell 1997). With limitations of space, it is impossible to discuss the concept of context in detail. Based on Ochs’s definition of context (1979: 2–6), suffice it to say here, however, that the key concept of context includes the talk, and the physical setting in which a talk is situated and the extra-situational context that extends beyond the local talk. The line between text and context may sometimes be blurry. ‘Interpretation of each utterance depends on information provided by earlier utterances in the sequence and it constitutes information necessary for interpreting later utterances’ (Janney 2002: 458). Courtroom examination is illustrative of what conversational analysts label as ‘the doubly contextual’ nature of utterances (Heritage 1984: 242; Drew and Heritage 1992: 18) in that previous utterances influence the following utterances in a series of questions and answers (Ehrlich 2001: 31). This article underscores the view that talk-in-interaction during courtroom examination is both context-shaping and context-renewing (e.g. Heritage 1984; Drew and Heritage 1992; Goodwin and Duranti 1992). As participants in communication events may not share a similar level of knowledge and information, the court interpreter does not have full and equal access to a body of knowledge shared by other participants in the court proceedings. The witness has first-hand experience with the incident or the people involved, and thus may have more information about or access to the particular context. Trial judges and lawyers possess sufficient knowledge about the case through the legal procedures preceding the witness examination. In comparison, the court interpreter’s access to the context is limited to the local context unfolding at each turn in court examination.2
JIEUN LEE 95
Furthermore, the prevalent mode of interpreting during the examination of CALD witnesses, referred to as liaison interpreting, adds to the difficulty of accurate interpretation. This mode of interpreting, which can be considered as short consecutive interpreting, is characterised by constant interactions between communicative parties, namely those of the witness and the examining lawyer. Relatively short chunks of information are proffered at every turn, for example, a word, a clause or a few clauses, or a few sentences at the most. Provision of piecemeal information at every turn may cause the interpreter to draw on context heavily, since texture3 and structure4 are manifested only partially in liaison interpreting (Hatim and Mason 1997: 42). However, contextual clues available to the interpreter may be incomplete, that is restricted to local cohesion (Hatim and Mason 1997: 50). Therefore, each utterance tends to be treated as a self-contained, decontexualised unit of meaning, and the interpreter relies on the linguistic features of each utterance (e.g. Hatim and Mason 1997; Linell 1997; Wadensjo¨ 1998). This may have the effect of the court interpreter having to draw premature conclusions about the message with limited contextual knowledge. Pure inductive and deductive skills may partially compensate for the lack of contextual knowledge, but the accuracy of the interpreted evidence may nonetheless still be at risk. Pragmatic ambiguities abound in natural language use (Green 1996: 11), but communication is possible because both the speaker and the hearer are assumed to follow certain communication conventions or maxims, such as Grice’s Cooperative Principle (Grice 1975). The speaker produces utterances assuming that the hearer can understand by the same practical reasoning and contextualising operations that they apply to social conduct in general (Schiffrin 1994: 234). The speaker should also make their intended meaning comprehensible to the hearer in order to communicate. Expectation of relevance may guide the hearer to the meaning of the speaker’s utterances (Grice 1975; Sperber and Wilson 1986), but the ambiguity of the meaning should be negotiated through communicative interactions (e.g. Schiffrin 1994; Gumperz 1999). Negotiation of meaning through communicative interactions is also true of interpreter-mediated communications (Wadensjo¨ 1998; Mason 2006). Just as spoken discourse requires constant confirmation between the participants to ensure that they understand the same thing in the same way due to the inherent ambiguity of spoken language (Scollon and Scollon 2001: 78), clarification or checking may be needed in interpreting other people’s utterances in courtroom interactions. However, the constraints on the role and behaviour of the court interpreter imposed by the institution may not encourage such checking or clarification.5 The restrictive role ascribed to the court interpreter limits the interpreter’s freedom to ask the witness for clarification when it is needed. In the solemn atmosphere of the courtroom, frequent requests for clarification might be perceived as indicating the interpreter’s incompetence, and as an unnecessary disturbance of the smooth flow of the
96
COURT INTERPRETING OF INEXPLICIT LANGUAGE
proceedings, and consequently may act as a threat to the face of the interpreter who wants to maintain professional dignity. The intrusive actions of the interpreter may influence jurors’ evaluation of witnesses and lawyers; however, the impact caused by the interpreter’s interruption is far less than the one caused by the interpreter’s alterations to the pragmatic meaning of witnesses’ utterances (Berk-Seligson 2002: 195).
Inexplicit source language Courtroom discourse is a highly institutionalised discourse, constrained by evidentiary rules (Hunter and Cronin 1995; Maley 2000). Witnesses are expected to follow Grice’s (1975) conversational maxims of Quality, Quantity, Manner, and Relation (Penman 1987). Although the evidentiary rules do not always prevent witnesses from flouting them, institutional sanctions are taken against obvious violations of such rules in courtroom examinations.6 Legal discourse, which is often characterised by explicitness and precision, does not tolerate ambiguity or multiple interpretations (Luchjenbroers 1991). Witnesses are required to be explicit in giving evidence (Lakoff 1990: 100, 130). However, lay witnesses, particularly those from CALD backgrounds, may lack understanding of the discourse style expected in court proceedings, and may also lack the skills necessary to communicate adequately in such contexts. Not being aware of the level of explicitness required in the courtroom, where the alleged event and actions are reproduced in words, witnesses may inadvertently produce ambiguous and incoherent speech by not providing enough information. Perceiving a witness’s speech as coherent depends on the provision of adequate contextual information. Coherence is also dependent on the speaker’s and the hearer’s willingness to negotiate coherence in the same manner as they negotiate meaning in interactions (Bublitz and Lenk 1999: 154). To return to interpreter-mediated courtroom examination, it may be argued that a witness’s inexplicit utterances may sound incoherent or ambiguous to the interpreter whose contextual information is limited to the physical setting of the courtroom and the linguistic context created and renewed during examinations. Inexplicit language generally refers to utterances that require the hearer to rely on context to understand meaning (Cheng and Warren 1999; Warren 2006). In other words, if there is limited access to the context, the use of inexplicit language may result in pragmatic failure (Dines 1980; Cheng and Warren 2003). This article focuses on instances of pragmatic failure where the speaker misjudges the extent of the shared knowledge of the hearer, namely the interpreter and make inexplicit utterances. Thus, inexplicitness in the witness’s utterance may inevitably require clarification for accurate interpretation. I argue that, although inexplicitness in CALD witnesses’ evidence is largely due to their use of an informal discourse style, it may also be due to the linguistic differences of the language used by CALD
JIEUN LEE 97
witnesses in relation to the language of the court. The linguistic features of some languages such as Korean may create extra burdens for interpreters in maintaining appropriate levels of accuracy in interpreting. Given that research in interpreting studies has yet to examine various cross-linguistic issues in spoken language in a variety of language combinations, a study of the discourse of courtroom examination of Korean-speaking witnesses will contribute to the literature by shedding light on linguistic issues in interlingual communication involving one of the under-researched languages in courtroom settings. Some features of Korean morphology and syntax may be problematic when interpreting into English. I will briefly discuss three grammatical features of Korean that may be relevant to court interpreting. First, there is no strict marking of singularity and plurality in Korean. The plural marker attachment is not as obligatory as in English. See example (1) subsequently. (1)
Pay-ka pakwuni-ey iss-ta. Pear-NOM basket-LOC there_is-DECL pear basket-in is/exists There is a pear in the basket/There are pears in the basket. (Note: NOM = nominative case markers; LOC = locative case markers; DECL = declarative verb ending.) The noun pay (pear) can be used without any qualifier. Quantity is specified only when it is important information. When quantifiers or numerals appear in sentences, the plural marker is often not used. Secondly, Korean does not have a single form that indicates definiteness which is indicated by demonstratives in some cases (Kim 1985: 895; Sohn 1993: 278). This means that a decontexualised utterance, such as example (1), does not give the interpreter all the information necessary to meet the grammatical conventions of English, such as indicating the plurality or singularity of the pear and definiteness of the basket. Thirdly, Korean predicates do not agree in number, person or gender with their subjects, but sentence-ending forms mark varying degrees of deference and politeness (Kim 1985: 895). This means that the verb form in Korean is not an indicator of plurality/singularity, person or gender. Accordingly, in order to ensure an accurate rendition, the interpreter is required to make an appropriate decision based on the context, or to ask for clarification. Fourthly, Korean makes frequent use of ellipsis. The means of conveying inexplicitness may vary in each language, but ellipsis, substitution, deixis and reference are the main forms whereby inexplicitness is conveyed in both English and Korean. The discussion in this article is limited to ellipsis, a feature which is used for the sake of brevity, economy, and informality
98
COURT INTERPRETING OF INEXPLICIT LANGUAGE
in communication across all languages (e.g. Carter and McCarthy 1995; Swan 1995; Yang 1996; Nariyama 2004). Nonetheless, its prevalence is more common in Korean, even in written language (e.g. Kim 1989, 1998; Lim 1996; Yang 1996). Korean is a discourse/situation-oriented language, in that contextually understood elements, whether they are subject, object or any other major sentential element, are often left unexpressed (Sohn 1993: 7–8). Person pronouns may be omitted from all syntactic positions (e.g. subject, direct object, indirect object, etc.) whenever they are contextually recoverable (Sohn 1993: 282). Explicit subjects mark emphasis or contrast (Sohn 1993: 282; Yang 1996; Kim and Jung 2006: 97–100). While the subject in subordinate clauses cannot be ellipted in English, both subjects as well as indirect object can be ellipted in Korean (see example (2) subsequently). (2) Kkok kayahamyen yaykihay- cwul- key. surely/really go-have to-if tell/speak-for_one’s_benefit-I_will surely/really if Ø have to go (I)’ll tell Ø If (she/he/you/they) have to go, I’ll tell (her/him/you/them). If the hearer does not know who the speaker is going to tell, the inexplicit utterance is ambiguous to the hearer. The maxim of Quantity is violated in that case. Without contextual information, inexplicit utterances may sound ambiguous, even incoherent. Given that the case is presented through, and is dependent on, oral evidence in court proceedings, misinterpretation of inexplicit language could have serious consequences.
THE STUDY The aim of this study was to examine the circumstances in which inexplicit language used by CALD witnesses creates a problem of ambiguity of meaning, and how court interpreters handle this issue in court interpreting. The study involved Korean–English interpreting discourse, which has so far not been investigated in the body of court interpreting studies, and, it is hoped, this study may shed a new light on aspects of inexplicitness of witnesses’ speech.
The data Approximately 80 hours of audio recordings of criminal proceedings, provided by the court, form the data for this study. These hearings took place in New South Wales (NSW) local and district courts between 2003 and 2007. The basic information on the data is presented in Table 1. Cases 4 and 6 were the only trials where observation in the courtroom was at least partly possible. Five Korean interpreters and 15 Korean-speaking witnesses were engaged in the proceedings analysed in this study. Some interpreters were
JIEUN LEE 99
Table 1: Profile of the data Case No.
Jurisdiction
Matter
No. of Korean speaking witnesses in each case
Interpreter No.
1
Local court
Assault
1 and 2
2
Local court
Assault
3 4 5
District court District court District court
6
District court
Aggravated Aggravated Aggravated assault Influencing witness
Two: complainant and defendant Three: complainant, witness, and defendant One: victim Two: victim and witness Four: complainant, accused, two other witnesses Three: defendants
robbery robbery sexual the
1 and 2 1 3 and 4 2, 4 and 5 2 and 3
engaged in more than one trial. Four of the interpreters involved were interviewed by the author some time after the court proceedings to obtain their feedback about court interpreting in general. A simplified version of the Jeffersonian transcription system (see Hutchby and Wooffitt 1998: 73–92) has been used for transcribing the court recordings. (See transcription conventions in Appendix.) For the benefit of non-Korean readers, Korean Romanization according to the Yale system has been used, and an English translation has been provided in italics next to the Korean utterances in a separate column. Adaptation was needed for the presentation of Korean language utterances, however. Because of syntactical differences, such as word order, the exact positioning of pauses and overlaps was not possible in English translations. Accordingly, only the transcripts of original utterances contain such information as pauses and overlaps.7
FINDINGS The inexplicit language of the Korean-speaking witnesses created more problems at the beginning of the proceedings and during the examination-in-chief than in cross-examinations. This is not surprising considering that context is yet to be created through extended questioning and answers. Considering also the cooperative nature of examination-in-chief, inexplicitness in Korean witnesses’ utterances may not be regarded as the product of intentional evasiveness. Of the different types of ellipsis, ellipted subjects created the highest level of ambiguity of meaning and consequently posed a challenge for the interpreter who did not have sufficient contextual information to interpret accurately. Four extracts were taken from the court proceedings to highlight the issue of interpreting inexplicit language during witness examination. In the following extracts, the following abbreviations are used: C = Crown Prosecutor;
100 COURT INTERPRETING OF INEXPLICIT LANGUAGE
DC = Defence Counsel; J = Judge; I = Interpreter; W = Witness; Ø = Ellipsis. Extract (1) is taken from the Crown witness’s evidence-in-chief in Case 4. This witness’s husband had been kidnapped, assaulted, and robbed by Kim, the ex-husband of his cousin, as well as by other defendants. This excerpt took place several minutes after the examination started on Day 5 of the trial. The interpreter (Interpreter 4) had not taken part in this trial before. Consequently, this fresh interpreter lacked contextual knowledge of the specific matter before the court. In the discourse preceding this extract, it had been stated that Kim called the witness’s husband and suggested a meeting with him. Because of these previous utterances, which provided the context, three pronouns contained in the prosecutor’s question (in turn 1) were specific. Extract 1: Case 4, examination-in-chief
Original utterances
1
C:
2
I:
3
W:
4
I:
5 6
W: I:
7
C:
8
I:
9 10 11
W: I: W:
(2.4) did he say why he was going to meet him? (0.4) way kulul mannalyeko hantako malul hayssupnikka kuka? (0.4) e ihonhan ttaluy ttaluy osul cenhaycwuntako haysssupnita. (.) it was said that (0.9) uh because in order to pass, (0.2) in order to pass the:: clothings of the daughter, (1) from the divorced (.) WIFE, (1.8) uh= =in order to pass clothings (1.5) of the daughter, (1.9) who is from the divorced wife. (0.5) ok. (1) now, did did you know anything about (0.6) the: (1) daughter’s clothing? that’s the (1.4) before that phone call? (0.7) tangsinun ku cenhwaka oki ceney ku ttaluy, i: ossey kwanhayse, i: alko issnun key issessesssupnikka? (.) tulun ceki ebssupnita. [never ne-] [Central] Coast ey salmyense imi ihonul han sangthayeyse,
Author’s translation of Korean utterances
did Ø say why Ø was going to meet him, he? Ø said Ø pass divorced daughter’s, daughter’s clothing.
before the phone, did you uh know anything about the daughter’s uh clothing? Ø never heard.
JIEUN LEE
(1) amwukesto cwucil anhasssupnita. 12
I:
101
while Ø living on the Central Coast already divorced, Ø did not give anything.
(0.5) I’d never heard anything. (1) were, they were living in Central Coast, (0.2) were living in Central Coast, (0.7) and (0.8) after the situation of divorce took place, (1) u::h nothing given. (1) excuse me, prosecutor, (1.3) uh the witness has spoke with very few words meagre words. (1) in interpreting, (0.3) this is a the most difficult task how much should I ADD to make the meaning understood by the English speaker. the meaning=
The prosecutor asked if the witness’s husband told her why he was going to meet Kim. The witness’s responses contain inexplicitness (in turns 3 and 11). In turn 3, the witness was inexplicit in relating to the court the reason her husband gave for meeting Kim. With limited contextual information available to the interpreter at this stage, the ambiguity of meaning resulted from the ellipted subjects for the verb cenhaycwunta (pass/give). There was uncertainty about who was saying the statement and also who was giving or receiving the clothes because the witness did not specify the subjects. Multiple interpretations may be possible for this response: (a) ‘Kim said he wanted to meet my husband to receive the clothing divorced daughter from him’ (b) ‘Kim said he wanted to meet my husband to give him the clothing divorced daughter’ (c) ‘My husband said he wanted to meet him to receive the clothing divorced daughter from him’ (d) ‘My husband said he wanted to meet him to give him the clothing divorced daughter’.
of a of a of a of a
Prior to this extract, the witness had given evidence that Kim and his wife had divorced while Kim was in prison. This contextual information may have offered a key to interpret the utterances in question. However, the witness’s utterance ihonhan ttal (divorced daughter) posed another problem in interpreting the utterance correctly. Since Kim was in the physical setting of the courtroom, the interpreter could tell his age and could see that he was too young to have a divorced daughter, and consequently interpreted ‘divorced daughter’ as referring to Kim’s divorced wife’s daughter (see turns 4 and 6). This reveals
102 COURT INTERPRETING OF INEXPLICIT LANGUAGE
a part of a complex cognitive process of interpreting whereby the interpreter draws on context surrounding the communication settings to determine the meaning. Whether it was an ethically correct decision to convey the intended meaning of misspoken words or unidiomatic expressions is beyond the scope of this article. Brief hesitations in the interpreted renditions when the interpreter interpreted inexplicit utterances may be an indication of the interpreter’s decision-making process (see pauses marked by numbers in parentheses in turn 6). The grammatical features of interpreted renditions reveal this interpreter’s strategy to cope with inexplicit language. Interpreter 4 chose not to make the ellipted subjects explicit by using a passive voice, such as ‘it was said that’ (in turn 4), and did not indicate who was giving or receiving by using the ‘in order to’ clause, which would have sounded like an incomplete and ungrammatical sentence to the English-speaking audience. A couple of turns later, the prosecutor asked if the witness knew anything about the clothing Kim said he would give to her husband (in turn 7). The witness’s response, briefly interrupted by the interpreter, continued in turn 11, but it was inexplicit again with the subjects for three verbs, namely ‘live’, ‘divorce’, and ‘not give’, ellipted. The subject ellipses suggest that each verb would share the same subject. Otherwise, it would be very ambiguous and confusing. Plausibly, the witness may have assumed that the interpreter would have enough knowledge to follow her evidence. However, inexplicit subjects, when access to context beyond the local context was required, made it difficult for the interpreter to render her answer accurately. The interpreter first used a dummy subject ‘they’ as in ‘they were living’ and then omitted the subject when rephrasing it, as in ‘were living in Central Coast’ (in turn 12). The subject ‘they’ which was initially used by the interpreter has a reference that is not known. If Kim had been divorced from his wife, ‘they’ could not have referred to Mr and Mrs Kim, and it was not known where the former Mrs Kim lived at that time. The interpreter also avoided making ellipted subjects explicit by using the passive voice in ‘nothing given’ (in turn 12). It is noteworthy that the interpreter immediately informed the prosecutor of the interpreting issue after completing the interpretation (in line 6 of turn 12). The interpreter may have been concerned about the comprehensibility of the interpreted renditions containing ungrammatical utterances. In the discourse that follows Extract (1), which is not presented here, as the interpreter tried to explain the intended meaning of the inexplicit language, the judge interrupted, encouraging the interpreter to ‘just translate’ what was said. The judge also assured the interpreter by saying that the court followed the evidence because they had been listening to other witnesses’ evidence for the past few days. According to the judge, the courtroom audience, who had more knowledge about the extra-situational context surrounding the alleged offence, had little trouble in understanding the interpreted evidence despite the grammatical inadequacy of some translations.
JIEUN LEE
103
Most of the interpreters interviewed were reluctant to interrupt the proceedings for clarification, and their coping strategies varied. Interpreter 4 was the only interpreter who disclosed difficulties with interpreting during courtroom examination. Interviews with interpreters revealed that they were aware that inexplicit language, particularly ellipted subjects, was problematic in court interpreting, and that they often used the passive voice to avoid specifying subjects ellipted in the original utterances of witnesses. Given that grammatical choices represent a particular perspective on events (Duranti 1994), the interpreters’ dependence on the use of passive voice may unnecessarily give an impression that the witness seeks to obscure the agent or diminish their responsibility. Extract (2) is also taken from the examination-in-chief of a Crown witness, but this time from Case 3. Interpreter 1 had interpreted for this witness for the second consecutive day in this trial. The witness was a victim of the crime committed by Ben Kim and others. The prosecutor asked how his relationship with Ben Kim’s father had been strained. According to the witness’s evidence leading up to this point, Ben’s father and the witness were acquainted with each other and had introduced Ben and Mary to each other. Mary was going out with Ben, and also worked for Ben’s father. She was living at this witness’s home at the time, but she did not come home until very late on the night being discussed in the extract. She called him very late saying that she was at a karaoke bar. Ben Kim’s father was also at the bar with Mary, and he asked the witness to come and join them in the early hours of the morning.
Extract 2: Case 3, examination-in-chief
1
W:
2
I:
3
W:
4 5 6
I: W: I:
7
W:
Original utterances
Author’s translation of Korean utterances
kulayse (0.4) cenun e ppalli thayksi thaywese ponyatala kuleko cenhwalul kkunhesssupnita. (0.3) so I asked him, put Mary in a cab quickly, and make he–(1) uh make her come ho:me, (0.5) and I just hung up. (3) kulikon tto cenhwaka wasssupnita, (0.2) but I had a call again, tto naolako haysssupni8ta8. (0.6) uh (2) requesting me to come out and meet him= =cenun mosnakantako hako,=
and Ø asked Ø to send Ø home by taxi, and Ø hung up phone.
and there was phone call again
Ø said again come out
I said Ø couldn’t go out,
104 COURT INTERPRETING OF INEXPLICIT LANGUAGE
8 9
J: I:
10 11
J: I:
12
DC
13
I:
14
W:
15 16
I: W:
17 18
I: W:
19
I:
20 21
W: I:
22 23 24
C: I: W:
=can you just repeat that? uh I had a call from HIM again, yeah= =req-requesting me, to come and meet, (0.3) with him. (2.7) can I ask him who he is your honour. kuka nwuka nwukwunyakuyo= =Be-Ben Kim 8apeci8 Ben Kim’s father. (2) kulayse e mosnakanta kulehko, so I told him I can’t. elma hwuey iltan icey thayksi thako cipulo ONtako, (0.4) cenhwaka tasi wasssupnita. (0.2) I had a call uh little bit later that, (0.6) she was coming home. (0.8) so, nwuka ontakoyo? (0.8) Ben Kimhako Maryhako it was Ben Kim’s father and Mary were coming home. (1.5) and to your place (1) ponin cipuloyo? yey
who-who is he? Be-Ben Kim’s father
so Ø said Ø can’t go out
a while later there was phone call again saying that Ø coming home by taxi
who’s coming? Ben Kim and Mary
to your home? yes
This witness’s utterances at every turn contained ellipted subjects. Nevertheless, when the preceding utterances provided the context for the interpreter to understand the ellipted subjects and objects, it did not hinder interpreting the witness’s utterances in turns 1, 3, 5 and 16. The interpreted renditions indicate that the interpreter made all the ellipted subjects explicit to produce grammatically adequate renditions in English (see turns 2, 4, 6 and 17). This extract shows that judges as well as lawyers, both examining lawyers and other lawyers, do clarify the meaning during examination. When the witness resumed giving evidence, the judge asked for repetition of the previous interpreted evidence (in turn 8). Immediately after the interpreter finished rendering the witness’s utterances, one of the defence lawyers asked for clarification of the ‘him’ which was contained in the interpreted rendition
JIEUN LEE
105
(in turn 12). The lawyers representing each defendant were interested in finding out exactly who did what and how it was done because such information was crucial for determining the evidence and consequently for the outcome of the case. It is noteworthy that a few turns later, in turn 18, the witness’s inexplicit language caused a problem in interpreting since there was more than one interpretation as to who was coming home. There was a possibility that Ben Kim’s father, who was allegedly drunk and had insisted that the witness come out at that hour, might have invited himself to the witness’s home. Since there is no subject–verb agreement in Korean, it was impossible to determine whether Mary was coming by herself or whether she had company. The interpreter asked the witness directly who was coming home (in turn 19) without asking for the court’s permission to clarify this. The witness responded that it was Ben Kim and Mary that were coming home at the time. It is not known whether it was the witness’s slip of the tongue and whether the interpreter intentionally altered the interpreted rendition. The interpreted rendition (in turn 21) shows that the interpreter translated ‘Ben Kim’ into ‘Ben Kim’s father’. In fact, Ben Kim had not been mentioned earlier in his testimony or later.8 Whenever clarification was needed due to ellipted subjects creating a difficulty in interpreting, Interpreter 1 did not ask for the court’s permission to seek clarification, and did not disclose to the court what the minor conversation was about or why such a clarification was needed. As a consequence, the court had no knowledge of the difficulty faced by the interpreter, and the court might have formed a suspicion that something underhand had transpired (Popovic 1991: 46). Extract (3) is from another case (Case 4) in which Interpreter 3 had to interpret inexplicit utterances. The Crown witness in this case gave evidence at two separate trials on the same matter involving different co-defendants, and is thus the same witness as the one appearing in Extract (2). Up to this point in the examination, the court had heard from the Crown witness that he was assaulted, robbed, and detained in a residential unit. However, it was not revealed at this stage of the examination that the victim had travelled with his kidnappers until the police found him. Therefore, it can be said that a new context was being created when new information was sought by the prosecutor, who asked whether the defendants mentioned where he was going (in turn 1). Extract 3: Case 4, examination in-chief
Original utterances
1
C:
and did anybody say where you were going?
Author’s translations of Korean utterances
106 COURT INTERPRETING OF INEXPLICIT LANGUAGE
2
I:
3
W:
4
I:
5 6
C: I:
7
W:
8
I:
9 10 11
C: I: C:
(0.7) nwukwulato mwe tangsinul (0.2) etilo teyliko kantatenci mwe eti kantanunci yaykihaysseyo? (0.7) kulen yaykinun ah (1.3) 8uh8 (1.9) Ben Kimi Central Coastlul ku kaya toentako kulen yaykilhaysseyo. (0.5) Ben Kim said (1.9) someone is going to Central Coast. (1.2) to the Central Coast? (0.2) Central Coastlo nwuka kantayyo? icey ku Central Coast ccokulo (1) icey kal kelako incey kulehkey yaykilul, (0.3) Ben Kim said (1) they were going towards Central coast. (1.2) did he say why= =way= =you be going to the Central Coast?
did anybody say where to take you or going anywhere? such talk uh Ben Kim said Ø had to go to Central Coast.
who’s going to Central Coast? Ø said Ø (be) going toward Central Coast Ø said so
why
The witness simply responded that Ben Kim talked about going to the Central Coast (in turn 3) without specifying who was going to the Central Coast. Although the prosecutor’s question implied that the witness was going somewhere, the ellipted subject in the witness’s utterance may have been perceived as ambiguous by the interpreter. It may be due to the interpreter’s tendency to treat each utterance as a decontextualised utterance. The interpreter used a dummy subject, ‘someone’, in the interpreted rendition (in turn 4) in place of the ellipted subject. A brief pause before the interpreter uttered ‘someone’ may be an indication of extended processing or momentary hesitation by the interpreter. Hesitations were common when interpreting inexplicit language of Korean witnesses, which were ambiguous or polysemious. The prosecutor did not seek to clarify who was referred to as ‘someone’ in the interpreted rendition (see turn 5). To the prosecutor, this information was already known and contained in her question. The prosecutor simply checked whether it was the Central Coast. However, the interpreted rendition of the prosecutor’s checking question points to the interpreter’s curiosity. While the prosecutor asked ‘to the Central Coast’ to confirm the place of destination, the interpreter asked who was going to the Central Coast, which was an attempt to clarify the inexplicit subject (see turns 5 and 6). Despite such an attempt, the witness, perhaps not aware of the interpreter’s problem, merely responded that Ben Kim talked about going to the Central Coast, thus not answering the interpreter’s question. As a consequence, the interpreter used
JIEUN LEE
107
another dummy subject, ‘they’, in the interpreted rendition in place of the ellipted subject (in turn 8). However, the prosecutor ignored these inexplicit references and asked only about the reason behind the trip to the Central Coast (in turns 9 and 11). It may be because the prosecutor thought the subject was already explicit in her questions (see turns 1 and 11). Other participants, such as the defence lawyer and the judge, did not seek clarification either, perhaps because this information was obvious to them or insignificant. Interpreter 3 covertly sought clarification from the witness without informing the court of the recurring interpreting problem deriving from inexplicit utterances. Extract 3 shows that Interpreter 3 produced grammatically adequate renditions adding subjects arbitrarily. Such restraint from disclosing the perceived difficulty in interpreting resulted in somewhat inconsistent evidence (e.g. the use of ‘someone’ and ‘they’ in turns 4 and 8). After many turns, the following discourse continued (Extract 4). The witness was asked to explain briefly why Ben Kim wanted to go to the Central Coast. The witness’s speech was somewhat incoherent because of frequent ellipses, fillers, and hesitation markers, as well as multiple embedded clauses within fragmented sentences (in turn 41). My translation below does not fully reproduce the grammatical features of the original utterances due to the linguistic differences, but the original utterances were mostly cut off and followed by new embedded clauses, which may be due to the witness’s attempt to elaborate and provide more background information. Extract 4: Case 4, examination in-chief Original utterances
37
C:
38 39
I: C:
40
I:
41
W:
well can you just exp- tell us briefly= [kantanhi] =[what], why he, spoke about YOU and HE going to the Central Coast, for what purpose? (0.4) com kantanhi malhayposeyyo. mwusun mokcekulo tangsini ku salamhako Central Coastlo kaya twayssnunci, (0.5) u::h Central Coastey (0.3) e Ben Kimi kekise incey ku (1.2) kekise incey, kal kaciko wuli cousin ce cwukillako kulenikka, (1) ku nayka anun ku yenge calhanun yecahanthey, ku nayka ceney alten
Author’s translations of Korean utterances
briefly
please say briefly. for what purpose did you have to go to Central Coast with the person uh . . . in Central Coast uh Ben Kim there uh there with knife tried to kill my cousin, (so) uh woman speaks English well, uh woman liv(es/ed) there who I knew before speaks English well. called Bae Sohee. I asked Mrs. Bae Sohee to
108 COURT INTERPRETING OF INEXPLICIT LANGUAGE
42
I:
43
W:
44 45 46
I: W: I:
47
C:
yecaka keki sanuntey yengelul calhayyo. Bae Soheelako. ku Bae Sohee acwumenihanthey kyengchaley sinkohaykaciko, cepkun kumcilul sikheytalako nayka yaykihassteni, icey kekiey kase, (1.5) Ben Kim was going to kill, (0.7) people with a KNIFE, (0.3) and then there was a Korean lady who speaks English well, (0.2) her name is Sohee Bae, who, she lives in Queensland, Central Coast, and, (1) sorry, I have to stop. I have to [stop].((shrugging her shoulders) [u::mm] kuleko incey ku kati kakaciko, (0.3) ku: ku tangsiey issessten yaykitulul, selo a (0.8) kekise incey (0.6) yaykilul kati hacako, ku yaykika manyakey ni mali thuliko nay mali thulimyenun e kekise cwukintatunka mwe incey kulen sikulo= he said =kepul cwunkeci. he was scaring me off (0.3) basically, (.) and he said, let’s go and find out, uh who’s killing who, and uh what, find out. (1) he was just scaring me off. ((rolling her eyes and shrugging her shoulders) (1) ok. (0.7) well, (1.1) when you le:ft (.) the unit, (0.2) did you get into a car,
report to the police and apply for the restraining order, so by going there,
u::mm and then, Ø go there and, uh Ø talk about what happened at that time, uh Ø talk about it together and, if the story, what you say is wrong, if what I say is wrong, Ø will kill Ø there, in that way,
Ø threatened.
Without prior knowledge about the extra-situational context the witness was referring to, the interpreter could not easily understand his inexplicit utterances. However, without clarification attempts, Interpreter 3 interpreted the witness’s utterances into coherent and explicit utterances, thus making them comprehensible for the courtroom audience (in turns 42 and 46). This interpreter also omitted part of the information she probably did not understand, including reference to the witness’ cousin. She translated ‘cousin’ as ‘people’ (in turn 42). After a short pause, the interpreter gave up interpreting,
JIEUN LEE
109
saying ‘I have to stop.’ The interpreter shrugged her shoulders with a bewildered look when she said this. Since the interpreter did not specify that it was her own utterance, rather than the witness’s utterance, it could have been perceived as the witness’s utterance. To the English-speaking courtroom audience, it might have been perceived as the witness’s remark that he was sorry for speaking too long without giving the interpreter a chance to interpret, but he could not help it. The witness continued giving evidence (in turns 43 and 45), but because of ellipted subjects and ambiguous utterances and the lack of contextual information, it might have been difficult to render his evidence accurately. The data reveals that Interpreter 3 summarised the evidence (in turn 46), but here misinterpretation occurred. The witness meant that Ben Kim threatened that he would kill someone, possibly Mrs Bae or the witness, but this was interpreted as ‘who’s killing who’ (in turn 46). This is another example of a distortion of the semantic meaning arising out of inexplicit utterances. In an interview, Interpreter 3 stated that subjectless sentences were one of the challenges of court interpreting, but she said that she tended to draw out the intended meaning from the context if possible, and refrain from interrupting the proceedings unless it was a matter of significance. The results of this study indicate that ambiguity deriving from inexplicit language, particularly ellipted subjects, poses a challenge for the accurate interpretation of evidence when ellipted subjects are not recoverable from the context, particularly when there is a possibility that more than one person is involved. In addition, when references are related to the extra-situational context about which the interpreter lacks knowledge, there is an increased risk of misinterpretation. The findings reveal that court interpreters generally do not reproduce the ambiguity of the meaning deriving from inexplicit utterances. They omit or modify the witnesses’ utterances to make them comprehensible and grammatically adequate, based on their judgement of the intended meaning, which may not only be unethical but may also result in misinterpretation of the original meaning. Such modifications corroborate other studies into Spanish–English court interpreting which indicate that interpreters tend to raise the register by rendering more coherent and comprehensible interpretations than the original (e.g. Hale 1997a: 204–5, 1997b: 52, 2004: 156–7; Berk-Seligson 2002: 142–5). Interpreters’ non-disclosure in addition to unwillingness to self-correct or admit errors may be considered as unprofessional conduct and may even be thought to be a matter of contaminating the evidence. When the interpreter misjudges the gravity of misinterpretation of inexplicit language, it may have significant legal implications. This study indicates that the court cannot be aware of the challenges faced by court interpreters in inter-lingual communication unless interpreters inform the court of inexplicit utterances and their need for clarification. If this clarification does not occur, the court will not be alerted to any potential inaccuracy in the interpreted renditions caused by the inexplicit utterances of CALD witnesses.
110 COURT INTERPRETING OF INEXPLICIT LANGUAGE
Notwithstanding this, interpreters’ attempts to draw the attention of the court to the ambiguity of the precise meaning of witnesses’ utterances and to request clarification have not always been welcome in the court (Morris 1995).9 Considering that there has been little discussion to date on when court interpreters’ intervention or mediation is justified and when it is not—because it is often perceived as something other than interpreting—it is understandable why most of the court interpreters tended to provide comprehensible renditions rather than interrupting the proceedings and reproducing inexplicitness in interpreted renditions. As long as the court assumes that interpreting is a mechanical process, interpreters may have to make hard decisions as to how to deal with inexplicit utterances. This may result in their overstepping the boundaries of their prescribed role of faithfully rendering original utterances into the language of the court.
CONCLUSION Drawing data from the discourse of Korean–English interpreting in Australian court examinations, this article has examined ambiguities contained in the inexplicit language used by Korean-speaking witnesses and the ways interpreters deal with this issue in the court. This article has demonstrated that cross-linguistic differences tied with grammatical conventions as well as context-specific discursive practice conventions add to the challenge of interpreting Korean witnesses’ utterances into English. Ambiguity of witnesses’ inexplicit utterances was more evident in examinationin-chief than in cross-examination in this study. Given that key information related to the offence is presented before the court during examination-in-chief, and evidence given in examination-in-chief through the interpreter remains in the official court records, misinterpreted evidence, which may seem trivial, may take on greater significance in adversarial court proceedings. The data analysed in this article was limited to criminal court proceedings, but the findings may equally apply to civil proceedings. These findings, based on Korean interpreting in Australian courts, may not be generalised to all CALD witnesses and all community languages used in the courtroom. However, given that Asian languages, such as Chinese and Japanese, display similar grammatical features of subject ellipsis to Korean (e.g. Oh 2006, 2007; Lee and Yonezawa 2008), the cross-linguistic issues examined in this article may also cause a problem related to the accuracy of interpreted evidence in court cases involving such languages that may not require explicit subjects in as strict manner as English. This theme needs further research, but clearly has potentially significant implications for the courts, whose duty it is to seek to obtain evidence as accurately as possible. This article infers from the study that court interpreters should not be held responsible for making sense of ambiguous utterances. Instead, it is
JIEUN LEE
111
recommended that they disclose any interpreting issues which derive from cross-linguistic differences, namely the use of inexplicit language, to the court or, if this proves to be too difficult or inappropriate, they should reproduce the inexplicitness in interpreted renditions so that the court may direct the witness to be more explicit. Interpreters should make every effort to avoid altering oral evidence. Considering the significance of potential implications of alterations and modification of original utterances in adversarial courtrooms, legal professionals as well as court interpreters need to appreciate that clarification may be necessary for the sake of achieving interpreting accuracy. The court should be informed that seeking or providing clarification is not necessarily a matter of displaying some deficiency in the interpreter’s language ability or interpreting skills. The objective should be to create an environment where the interpreters feel encouraged to alert the court to interpreting issues without the fear of putting their competence in doubt. Only when interpreters feel free to disclose such issues related to the integrity of evidence, unafraid of losing face, will the court be able to hear the evidence accurately, as it wishes and as it is required.
ACKNOWLEDGEMENTS I would like to thank Senior Research Professor Christopher N. Candlin and two anonymous reviewers for their comments on earlier version of this article. I am grateful for research support from Department of Linguistics of Macquarie University.
APPENDIX Transcription conventions [] = : CAPITALS , . ? () () (( )) 88
Overlapping talk Latching utterances Elongated vowel sounds. The more, the slower the enunciation is (e.g. :::) Emphasis Continuing intonation A stopping fall in tone. It does not necessarily indicate the end of a sentence Rising intonation The number in brackets indicates a time gap in seconds Empty parentheses indicate the presence of an unclear fragment Non-verbal activity Soft voice
112 COURT INTERPRETING OF INEXPLICIT LANGUAGE
NOTES 1
2
3
4 5
6
Incoherent speech in this article is distinct from incoherence caused by impaired speech ability or complete lack of coherence. The interpreter’s prior knowledge of the witness’s evidence is limited. The interpreter is not allowed to engage in conversation with the witness except for the purpose of checking whether they can understand each other. It is also the common practice of the New South Wales (NSW) courts to prefer a fresh interpreter who has not been involved in pre-trial conferences and police interviews, to enhance neutrality in interpreting. Texts possess the property of texture, which means that the lexicogrammatical units representing text hang together and linguistic cohesion exists within the text (Hasan 1978). See Hasan 1978: 229. See Morris (1993, 1995), Mikkelson (2008) and others for the controversy over the role of the interpreter. Giving false evidence is subject to a perjury charge. Irrelevant evidence is struck out by the judge. Lay witnesses
7
8
9
are often cut short when they speak too long. All the names that might be indicative of identities have been changed to protect the privacy of the people involved. I tried to provide a literal translation, even if it looks ungrammatical, to indicate the absence of grammatical equivalents. Alternative translation is also provided when there is ambiguity of meaning. An English place name, such as Central Coast uttered by Korean witnesses has not been translated in the Korean Romanization transcription. A speaker’s slip of the tongue is one of the sources of dilemmas for court interpreters (Morris 1995: 34–5), but the literature indicates clearly that court interpreters should refrain from correcting even obvious errors and do their utmost to faithfully reproduce original utterances (e.g. Gonzalez et al. 1991; Edwards 1995; Hale 1997b, 2002, 2004). The trial of Demjanjuk (The State of Israel v. Ivan John Demjanjuk, Criminal Case 373/86).
REFERENCES Berk-Seligson, S. 2002. The Bilingual Courtroom: Court Interpreters in the Judicial Process. 2nd edn. Chicago: University of Chicago Press. Bublitz, W. and U. Lenk. 1999. ‘Disturbed coherence: ‘‘Fill me in’’ ’ in W. Bublitz, U. Lenk, and E. Ventola (eds): Coherence in Spoken and Written Discourse. Amsterdam/ Philadelphia: John Benjamins. Carter, R. and M. McCarthy. 1995. ‘Grammar and the spoken language,’ Applied Linguistics 16/2: 141–58. Cheng, W. and M. Warren. 1999. ‘Inexplicitness: What is it and should we be teaching it?’ Applied Linguistics 20/3: 293–315.
Cheng, W. and M. Warren. 2003. ‘Indirectness, inexplicitness and vagueness made clearer,’ Pragmatics 13/3: 381–400. Dines, R. R. 1980. ‘Variation in discourse— ‘‘and stuff like that’’,’ Language in Society 9: 13–31. Drew, P. and J. Heritage. 1992. ‘Analysing talk at work: An introduction’ in P. Drew and J. Heritage (eds): Talk at Work: Interaction in Institutional Settings. Cambridge/New York: Cambridge University Press. Duranti, A. 1994. From Grammar to Politics: Linguistic Anthropology in a Western Samoan Village. Berkeley/LA: University of California Press.
JIEUN LEE
Edwards, A. B. 1995. The Practice of Court Interpreting. Amsterdam/Philadelphia: John Benjamins. Ehrlich, S. 2001. Representing Rape: Language and Sexual Consent. London/New York: Routledge. Fraser, B. and L. Freedgood. 1999. ‘Interpreter alterations to pragmatic features in trial testimony’. Paper presented at the annual meeting of American Association for Applied Linguistics, March 6–9. Goodwin, C. and A. Duranti. 1992. ‘Rethinking context: An introduction’ in A. Duranti and C. Goodwin (eds): Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Gonzalez, R. D., D. F. Vasquez, and H. Mikkelson. 1991. Fundamentals of Court Interpretation: Theory, Policy, and Practice. Durham, NC: Carolina Academic Press. Green, G. M. 1996. Pragmatics and Natural Language Understanding. 2nd edn. Mahwah, NJ: Lawrence Erlbaum Associates. Grice, H. P. 1975. ‘Logic and conversation’ in P. Cole and J. Morgan (eds): Syntax and Semantics, vol. 3: Speech Acts. New York: Academic Press. Gumperz, J. J. 1999. ‘On interactional sociolinguistic method’ in S. Sarangi and C. Roberts (eds): Talk, Work and Institutional Order: Discourse in Medical, Mediation and Management Settings. Berlin: De Gruyter. Hale, S. 1997a. ‘Clash of world perspectives: The discursive practices of the law, the witness and the interpreter,’ Forensic Linguistics 4/2: 197–209. Hale, S. 1997b. ‘The treatment of register variation in court interpreting,’ The Translator 3/1: 39–54. Hale, S. 2002. ‘How faithfully do court interpreters render the style of non-English speaking witnesses’ testimonies? A data-based study of Spanish-English bilingual proceedings,’ Discourse Studies 4/1: 25–47. Hale, S. 2004. The Discourse of Court Interpreting. Amsterdam/Philadelphia: John Benjamins. Hale, S. B. 2007. Community Interpreting. Basingstoke: Palgrave Macmillan. Hasan, R. 1978. ‘Text in the systemic functional model’ in W. Dressler (ed.): Current Trends in Textlinguistics. Berlin/New York: Walter de Gruyter. Hatim, B. and I. Mason. 1997. The Translator as Communicator. London/New York: Routledge, pp. 36–60.
113
Heritage, J. 1984. Garfinkel and Ethomethodology. Oxford: Basil Blackwell. Hutchby, I. and R. Wooffitt. 1998. Conversation Analysis. Cambridge: Polity, pp. 73–92. Hunter, J. and K. Cronin. 1995. Evidence, Advocacy and Ethical Practice: A Criminal Trial Commentary. Sydney/Adelaide/Brisbane/ Canberra/Melbourne/Perth: Butterworths. Jacobsen, B. 2003. ‘Pragmatics in court interpreting: Additions’ in L. Brunette, G. Bastin, I. Hemlin, and H. Clarke (eds): Critical Link 3: Interpreters in the Community. New York/ Amsterdam: John Benjamins. Janney, R. W. 2002. ‘Context as context: vague answers in court,’ Language & Communication 22: 457–75. Kim, J. N. 1998. ‘A retrospection on the phenomena of ellipsis in Korean: Ellipted subjects in embedded clauses,’ Kukeohak 32: 201–15. Kim, N. I. 1985. ‘Korean’ in B. Comrie (ed.): The World’s Major Languages. London/Sydney: Croom Helm. Kim, S. R. 1989. ‘The phenomena of ellipsis in Korean and its types,’ Keonkuk University Nonmunjib 29/2: 53–70. Kim, E. I. and Y. C. Jung. 2006. ‘Subject ellipsis and ambiguity condition,’ Eoneokwahakyeonku 37: 93–112. Lakoff, R. T. 1990. Talking Power: The Politics of Language in our Lives. New York: Basic Books. Lee, D. and Y. Yonezawa. 2008. ‘The role of the overt expression of first and second person subject in Japanese,’ Journal of Pragmatics 40: 733–67. Lim, G. H. 1996. ‘A study of Korean ellipsis phenomenon,’ Eomunhak 57: 281–319. Linell, P. 1997. ‘Interpreting as miscommunication’ in Y. Gambier, D. Gile, and C. Taylor (eds): Conference Interpreting: Current Trends in Research. Amsterdam/Philadelphia: John Benjamins. Luchjenbroers, J. 1991. ‘Discourse dynamics in the courtroom,’ La Trobe Papers in Linguistics 4, available at http://www.latrobe.edu. au/linguistics/LaTrobePapersinLinguistics/ Vol%2004/06Luchjenbroers.pdf. Accessed 15 November 2007. Maley, Y. 2000. ‘The case of the long-nosed potoroo: The framing and construction of expert witness testimony’ in S. Sarangi and M. Coulthard (eds): Discourse and Social Life. Harlow/New York: Longman.
114 COURT INTERPRETING OF INEXPLICIT LANGUAGE
Mason, I. 2006. ‘On mutual accessibility of contextual assumptions in dialogue interpreting,’ Journal of Pragmatics 38/3: 359–73. Mikkelson, H. 2008. ‘Evolving views of the court interpreter’s role: Between Scylla and Charybdis’ in C. Valero-Garces and A. Martin (eds): Crossing Borders in Community Interpreting. Amsterdam/Philadelphia: John Benjamins. Morris, R. 1993. ‘The interlingual interpretercypher or intelligent participant?,’ International Journal for the Semiotics of Law 6/18: 271–91. Morris, R. 1995. ‘The moral dilemmas of court interpreting,’ The Translator 1/1: 25–46. Nariyama, S. 2004. ‘Subject ellipsis in English,’ Journal of Pragmatics 36/2: 237–64. O’Barr, W. M. 1982. Linguistic Evidence: Language, Power, and Strategy in the Courtroom. New York: Academic Press. Ochs, E. 1979. ‘Introduction: What child language can contribute to pragmatics’ in E. Ochs and B. Bambi (eds): Developmental Pragmatics. Academic Press: New York. Oh, S. Y. 2006. ‘English zero anaphora as an interactional resource,’ Research on Language an Social Interaction 38/3: 267–302. Oh, S. Y. 2007. ‘Overt reference to speaker and recipient in Korean,’ Discourse Studies 9/4: 462–92.
Penman, R. 1987. ‘Discourse in courts: Cooperation, coercion and coherence,’ Discourse Processes 10: 201–18. Popovic, J. 1991. ‘The magistrates’ court’ in G. Bird (ed.): Law in a Multicultural Australia. Melbourne: National Centre for Cross-Cultural Studies in Law. Schiffrin, D. 1994. Approaches to Discourse. Singapore: Blackwell. Scollon, R. and S. W. Scollon. 2001. Intercultural Communication. 2nd edn. Oxford: Blackwell. Sohn, H. M. 1993. Korean. London/New York: Longman. Sperber, D. and D. Wilson. 1986. Relevance: Communication and Cognition. Oxford: Basil Blackwell. Swan, M. 1995. Practical English Usage. 2nd edn. Oxford: Oxford University Press. Wadensjo¨, C. 1998. Interpreting as Interaction. London/New York: Longman. Warren, M. 2006. Features of Naturalness in Conversation. Amsterdam/Philadelphia: John Benjamins. Yang, M. H. 1996. ‘Ellipsis in Korean,’ Kukeokukmunhak 117: 125–57.
Applied Linguistics 30/1: 115–137 ß Oxford University Press 2009 doi:10.1093/applin/amn052 Advance Access published on 28 January 2009
MEMORIAL ARTICLE: JOHN SINCLAIR (1933–2007)
The Search for Units of Meaning: Sinclair on Empirical Semantics MICHAEL STUBBS Universita¨t Trier, Germany John McHardy Sinclair has made major contributions to applied linguistics in three related areas: language in education, discourse analysis, and corpusassisted lexicography. This article discusses the far-reaching implications for language description of this third area. The corpus-assisted search methodology provides empirical evidence for an original and innovative model of phraseological units of meaning. This, in turn, provides new findings about the relation between word-forms, lemmas, grammar, and phraseology. The article gives examples of these points, places Sinclair’s work briefly within a tradition of empirical text analysis, and identifies questions which are currently unanswered, but where productive lines of investigation are not difficult to see: (1) linguistic-descriptive (can we provide a comprehensive description of extended phrasal units for a given language?) and explanatory (what explains the high degree of syntagmatic organization in language in use?), and (2) sociopsychological (how can the description of phrasal units of meaning contribute to a theory of social action and to a theory of the ways in which we construe the social world?).
John McHardy Sinclair (14 June 1933–13 March 2007) contributed significantly to three central areas of applied linguistics: language in education, discourse analysis, and corpus-assisted lexicography. Throughout all this work, his method of linguistic analysis was to search for patterning in long authentic texts, and he argued consistently against the neglect and devaluation of textual study in much recent linguistics. In the 1960s, his early corpus work followed the principle that conversation is ‘the key to a better understanding of what language really is and how it works’ (Firth 1935: 71), and argued that spoken English would provide evidence of ‘the common, frequently occurring patterns of language’ (Sinclair et al. 1970/2004: 19). In the 1970s, his work on audio-recorded spoken language in school classrooms described characteristic units of teacher–pupil dialogue, and developed structural categories for analysing long texts, as opposed to the short invented sentences which were in vogue at the time (Sinclair and Coulthard 1975). In the 1980s and 1990s, he studied patterning which is visible only across machine-readable corpora of hundreds of millions of running words (Sinclair 1991, 2004a). This led to his theory of phraseology
116 SINCLAIR ON EMPIRICAL SEMANTICS
and to the Cobuild series of language reference materials for advanced learners of English. The first Cobuild dictionary appeared in 1987 and was followed by further dictionaries, grammars and teaching materials, as well as detailed discussions of their theoretical principles (Sinclair (ed.) 1987; Barnbrook 1996). Accounts of Sinclair’s life and work have appeared in several places,1 and his publications are comprehensively listed in the Supplementary Data, which accompanies the on-line version of this article. I will not repeat this material here, but will concentrate on his theory of phraseology. It is this which probably has the most far-reaching implications for language description and language teaching.
EMPIRICAL SEMANTICS: TECHNOLOGY AND OBSERVATION Sinclair’s work draws on the British tradition of empirical text analysis developed by Firth, Halliday, and others, but it is Sinclair who develops the most radical implications, namely that multi-word units of meaning can be discovered by observing recurrent patterns across large text collections. His best known work is based on ‘a new view of language and the technology associated with it’ (Sinclair 1991: 1), and on the observation that ‘the language looks rather different when you look at a lot of it at once’ (Sinclair 1991: 100). Technology leads to a radically revised perception of the object of study for linguistics, because it becomes possible to observe patterns of language use which are otherwise invisible. Sinclair is one of the very few linguists who has discovered many things which people had simply not noticed, despite thousands of years of textual study—because they are observable only with the help of computer techniques which he helped to invent. His discoveries are due to a number of related principles and insights. First is his plain text policy (Sinclair 2004a: 190–1). In contrast to much other corpus work, Sinclair’s principle is to rely as little as possible on annotating the data, since grouping word-forms into lemmas and part-of-speech categories can hide the patterns. In the worst case, a linguist tags data with categories from pre-corpus studies, searches the data, and finds the categories, in a vicious methodological circle. A widespread belief, that raw text is not amenable to systematic analysis, led to Saussure’s despair of studying parole and to Chomsky’s withdrawal into introspection. Sinclair has shown that we now have the technology to discover patterns in raw textual data—and that these patterns are rather different from what we previously thought. His slogan for the approach is ‘trust the text’ (Sinclair 2004a). Second, as Sinclair (2008: 31, 38) points out, although linearity, directionality, and redundancy are all obvious features of text, they are largely ignored in linguistic theory. The role of paradigmatic choice has often been overestimated, and syntagmatic constraints on linear sequences correspondingly underestimated. A major motivation for his approach is therefore to remedy this neglect of syntagmatic organization (Sinclair and Mauranen 2006: xviii).
MICHAEL STUBBS 117
In an individual text, we can observe neither repeated syntagmatic relations nor any paradigmatic relations at all, but it is precisely these two things which concordances make visible (Tognini-Bonelli 2004: 18). An individual text is designed to be read as a whole, linearly, from left to right. A concordance is designed to be read as a series of fragments, vertically, from top to bottom. A concordance brings together utterances which have been produced at different times by different speakers, makes visible recurrent patterns, and allows us to count them. It freezes instances of language behaviour, and provides public data which can be studied at leisure (Mauranen 2004: 103). This is analogous to the effect of a camera, which can freeze a series of events in time, thereby turning a process into a product, which is publicly accessible to different observers. The development of the natural sciences was made possible by the powers of observation which were, in turn, made possible by the invention of the lens, and hence the telescope, the microscope, and the camera (Macfarlane and Martin 2002). Presumably no-one these days claims that observations can entirely avoid subjectivity. We select what to study because we have ideas about what is interesting, and different methods of observation give access to different phenomena. Nevertheless, corpus data allow us to study language ‘with a degree of objectivity . . . where before we could only speculate’ (Kilgarriff 1997: 137). It is sometimes objected that corpus methods over-emphasize quantitative aspects of language use. However, no language teacher could conceivably ignore frequency in presenting vocabulary, since the distribution of units in language use is so uneven. A few words occur once every few seconds or so, but most words occur only once every few weeks or so (as Sinclair once summarized Zipf’s law to me). For example, ten of the high frequency words in English (the, of, and, to, a, in, that, is, was, it) make up over 20 per cent of most running text. Sinclair (e.g. 1999a) often points out that linguistic theory underestimates the importance of high frequency words. In any case, observational tools are designed to reveal something which we could not otherwise see, such as recurrent collocations. The notion of what counts as the same in a language system is a foundational question for linguistics, and a central task is therefore to discover the units (morphemes, phonemes, distinctive features, etc.) which regularly recur, and which are the ‘very essence of language’ (Harris 1988: 19). In summary: The computer-assisted methods which can help in discovering patterns of language use involve new observational techniques, they are quantitative, and they provide a way of studying the relation between paradigmatic oppositions and syntagmatic constraints. Major theoretical advances have often come when linguists have realized the significance of different forms of data. Corpora are just data and quantitative methods are just methods, but their combination has led to a major shift in theory, and it is this theory which has to be evaluated.
118 SINCLAIR ON EMPIRICAL SEMANTICS
A concordance presents parallel cases in tabular form—typically dozens or hundreds of instances—and this visual display is part of the argument (Fahnestock 2003: 142). The methods based on the use of concordance data are often thought of as ‘induction’: many observations (down the vertical axis of a concordance) lead to a generalization. However, we are always working with a sample drawn from an indefinitely large population whose parameters of variation are unknown. If we increase the size of the sample, we increase its diversity, but no matter how large the sample becomes, we will never know whether it includes all the patterns of the language in their typical proportions. Indeed, because of differences between text collections due to genre and topic, it is most unlikely that this will be the case (Sinclair 2008: 30). It is therefore useful to distinguish between ‘induction’ and what is sometimes called ‘eduction’. Induction reasons from particulars to the general, and is notoriously susceptible to refutation by counter-example. Eduction reasons from particulars to more particulars, and the concept of a single counterexample does not apply. Many observations of a pattern lead to a prediction that similar patterns will be observed in the future. But absolutely fixed patterns are extremely rare, and a frequent conclusion is that a given pattern is ‘typical’ or ‘canonical’, but that it has variants. Fahnestock discusses these points in detail, and shows how parallelism across both textual and visual arrays constitute ‘schematic presentations of evidence in an argument’ (2003: 139).
‘THE SEARCH FOR UNITS OF MEANING’: OSTI TO COBUILD It is simplistic to pick out just one theme in all this work, but ‘the search for units of meaning’ (the title of Sinclair 1996) is central, and signals a defining problem in linguistics. In the 1960s, Sinclair et al. (1970/2004: 3) had asked how objective data and subjective meaning are related: ‘(a) How can collocation be objectively described?’ and ‘(b) What is the relationship between the physical evidence of collocation and the psychological sensation of meaning?’ Sinclair’s early corpus work, informally published as the OSTI Report (UK Government Office for Scientific and Technical Information, Sinclair et al. 1970/2004), describes quantitative research on computerreadable data, carried out between 1963 and 1969, but not formally published until 2004. It is difficult to project oneself back to a period in which there were no PCs, and in which the university mainframe machine could only handle with difficulty the OSTI spoken corpus of 135,000 running words. The project was in touch with work by Francis and Kucˇera (1967) on their one-million-word corpus of written American English, but little other comparable work was available. Yet the OSTI Report explicitly formulated many ideas which are still central to corpus linguistics. The unit of lexis is unlikely to be the word in all cases. Homonyms can be automatically distinguished by their collocations.
MICHAEL STUBBS 119
Collocations differ in different text-types. Many words are frequent because they are used in frequent phrases. One form of a lemma is regularly much more frequent than the others (which throws doubt on the lemma as a linguistic unit). There is a relation ‘between statistically defined units of lexis and postulated units of meaning’ (Sinclair et al. 1970/2004: 6). As Sinclair puts it in the 2004 preface to the OSTI Report, we have a ‘very strong hypothesis [that] for every distinct unit of meaning there is a full phrasal expression . . . which we call the canonical form’. This tradition of computer-assisted language analysis was concerned, from the beginning, with a theory of meaning, and it provides the context for Sinclair’s ambitious aim: ‘the ultimate dictionary’ would list all the lexical items in the language with their possible variants (Sinclair et al. 1970/2004: xxiv). Because machines in the 1970s were not powerful enough to handle large quantities of data, the work was shelved, and started again in the 1980s as the COBUILD project in corpus-assisted lexicography (Sinclair (ed.) 1987; Moon 2007). Sinclair’s long-term vision of linguistics was formulated in the 1960s, and he then waited till the technology—and everyone else’s ideas—had caught up with him.
CO-SELECTION: LEMMAS AND WORD-FORMS Sinclair proposes co-selection as a central descriptive mechanism of language in use. Examples of lexical co-selection which tend to spring to mind are cases where a relatively infrequent word has a strong tendency to co-occur with a restricted set of collocates. BOGGLE usually co-occurs with mind, or QUAFF usually co-occurs with beer or wine. More generally, collocation is often thought of as a relation between two lemmas. For example, different forms of the lemmas HARD–WORK, PLAY–ROLE, STRONG–ARGUE and HEAVY–RAIN can co-occur as follows:
hard work, hard-working, works very hard, to work harder, a hard worker play a role, play a key role, the role played by, role-play, a new role to play a strong argument, strongly argued, the argument will be strengthened heavy overnight rain, heavy autumn rains, heavy rainfall, raining heavily
However, there are frequently restrictions on the forms of the lemma. For example, the lemmas HEAVY–DRINK co-occur as heavy drinker and drink heavily, but not as heavily drunk. In addition, as Greaves (2007) and Warren (2007) point out, in the collocation PLAY–ROLE the noun is often preceded by roughly synonymous adjectives. This provides evidence of a longer pattern, in which the adjective signals the speaker’s evaluation:
a(n)
role
PLAY
120 SINCLAIR ON EMPIRICAL SEMANTICS
Greaves, Warren, and Cheng (Cheng et al. 2006, 2009) have developed software (illustrated below) to investigate empirically the extent to which collocational units can vary on three dimensions: the forms of the lemmas, their relative position to left and right of each other, and their distance from each other. One of the clearest findings of corpus analysis is that different forms of a lemma often have quite different frequencies and collocates, and therefore different meanings. Sinclair (2004a: 31) notes that plural eyes and singular eye have little overlap in their top 20 collocates: blue and brown collocate only with plural eyes, and singular eye occurs in expressions to do with visualizing and evaluating.
an eye on; TURN a blind eye to; in the public eye; with the naked eye; as far as the eye can see, in his mind’s eye; more than meets the eye; KEEP an eye out for
KEEP
Similarly, the collocation SEEK–asylum occurs in various forms (asylum seekers, seeking asylum, etc.). However, different forms of SEEK co-occur with different collocates. In a 200-million-word corpus, I studied the 20 most frequent collocates of the different word-forms (Stubbs 2001: 27–8). The forms seek, seeking and sought all shared the collocates asylum, court, government, help, political, support. The forms seeks and seek shared only one collocate: professional. And the pairs seeks/sought and seeks/seeking had no shared collocates. The word-form seeks is frequent in lonely hearts ads, where its frequent collocates include attractive, black, caring, female, guy, lady, male, man, professional, similar: female 31, single, seeks well educated gentleman The overlap in the collocates provides a measure of the semantic distance between the word-forms. Three word-forms (seek, seeking, sought) often occur in texts on political and legal topics, with collocates from the semantic field of ‘‘help and support’’, but the word-form seeks is only distantly attached to this cluster. How lemmas and their forms should be divided between headwords in dictionaries is an empirical question.
CO-SELECTION: LEXIS AND GRAMMAR Sinclair (1992: 14) illustrates co-selection of lexis and grammar with the example lap. In its body-part sense, the word does not occur as the subject or object of a verb, but only in prepositional phrases: her knitting lying in her lap, he lifted the cat off his lap. In other phraseology, it has different meanings (e.g. we’re on the last lap now). It therefore makes little sense to ask what the individual word means, since this depends on how it is used in different phrases and grammatical constructions. The word ebb provides a similar example. It occurs around 250 times in the British National Corpus (BNC). Only a few uses are literal (rocky reefs exposed by
MICHAEL STUBBS 121
the ebb tide). Most occurrences are of the noun, most frequently in phrases such as a low ebb and ebb and flow. The descriptive problem lies in the variants:
at (such) a low ebb; at this low ebb at an/a low ebb at lowest ebb
The most frequent form is at a low ebb, but the unit is more abstract and has indeterminate boundaries. The collocation at-LOW-ebb is typically used to talk about people’s morale or spirits, which are at a lower ebb than for some time in the past. The most frequent verb is BE, but a few other verbs occur:
her spirits were at their lowest ebb with teachers’ morale at its lowest ebb in living memory at their lowest ebb for 20 years at its lowest ebb in history staff levels have reached a low ebb credit had sunk to its lowest ebb
This is a clear example of a word occurring in restricted patterns, typically in a unit which is ‘a single lexical choice whose realization is six or seven words long, and within which there is some variation’ (Sinclair 2004b: 290). There is no possible paradigmatic contrast between definite and indefinite articles, or between HIGH and LOW. In practical terms, there is little point in knowing the word ebb, without knowing its phraseology. In theoretical terms, the problem is to establish the internal variability and external boundaries of the phrasal unit. The membership of the category (as with most linguistic categories) is a matter of degree. Taylor (2002: 102) discusses the grammatical pattern N1 by N1. Examples in the BNC which occur 20 times or more are:
step by step (165), day by day (117), year by year (105), bit by bit (47), week by week (32), line by line (31), case by case (30), month by month (24), stage by stage (21), inch by inch (20)
The lexis usually denotes small units of time or space. The phrase day by day is frequent, decade by decade much less frequent (3), and century by century does not occur in the BNC, although a search of the worldwide web provides examples such as
it was possible, century by century, to follow the town’s urban development
The construction has a few conventionalized exponents, but it is the idiomatic pattern itself which carries the meaning of a gradual, steady, often deliberate and methodical process. It is not possible to give a definitive list of its lexis, since it is partly productive:
she washed a Cos lettuce, leaf by leaf they worked their way up floor by floor
122 SINCLAIR ON EMPIRICAL SEMANTICS
The observation that different words occur in different grammatical patterns leads to the idea of contrasting a global with a local grammar (Gross 1993). Traditional grammatical models have never been able to cope with everything which occurs in all varieties of text. So, what is required is descriptions of specific areas of language, where specific meanings are expressed. Hunston and Sinclair (2000) discuss the language of evaluation and show that evaluative adjectives often occur in the pattern: there + link verb + something + adjective + about The BNC has over 190 occurrences of the pattern there BE something ADJ about. The adjectives are evaluative, and usually unfavourable. The most frequent express the concept of ‘‘strange’’ (e.g. abnormal, curious, eerie, fishy, funny, paradoxical, peculiar, strange, odd, unusual) or of ‘‘familiarity versus difference’’ (e.g. different, familiar, new, special, unique). But it is not possible to provide a definitive list: there are also single occurrences of absurd, crude, delicious, evil, God-like, horrible, insulting, irritating, naive, primitive, right, risible, spectacular, suspicious, tragic, unsatisfactory, vital, wonderful, and others. Again, we have a strong central pattern with a long tail of variants.
FREQUENT WORDS IN FREQUENT PHRASES Sinclair argues that very frequent words need to be described in their own terms: ‘their frequency makes them dominate all text’, but few of them ‘have a clear meaning independent of the cotext’. For example, the word way ‘appears frequently in fixed sequences’, where different patterns characterize different meanings, and where the resultant phrases ‘are frequently used metaphorically rather than literally’ (Sinclair et al. 1970/2004: 157–59, 163, 110–11). It looks like a high frequency noun, but in terms of its usage is in a class of its own (Sinclair 1999a: 166–72). The word is frequent because it occurs in frequent phrases, and its meaning depends on the phraseology. On its own it seems to have many meanings, but the phrases are unambiguous. all the way to school, half way through, the other way round, a possible way of checking, by the way, etc. The word frequently occurs in longer conventionalized phrases which express topic-independent pragmatic meanings, as in these quasi-proverbs and cliche´s, speech acts, and discourse markers: see/know which way the wind is blowing; has become a way of life; if that’s the way you want it; laughing all the way to the bank; let me put it this way; only one way to find out; that’s one way of putting it; that’s the way I look at it; there is no way of knowing/telling; well on the way to recovery The word point is also very frequent: at about rank 30 in a frequency list of nouns in the BNC. Under point as a head-word, Cobuild (1995a) gives about 30 senses, which seems to imply that the word is highly ambiguous, but the
MICHAEL STUBBS 123
Concordance 1: Illustrative examples of strong point preceded by a negative. 01It was soon clear that rowing was not my strong point. At hockey there was a v 03ting as usual for arithmetic was not her strong point especially mental arithm 05nating between men is evidently not your strong point. Perhaps a few lessons m 06rope. Zoological accuracy was not Tulp’s strong point. The animal was a chimpa 07knowledge that, cooking was not Stella’s strong point, for it had turned out 08 n turmoil consultation is not the BMC’s strong point when it comes to mountai 09sion or argument anyway. that wasn’t her strong point. Her eyesight was her st 10 organisation of business er isn’t their strong point at the moment, whether i 11re that thinking doesn’t seem to be your strong point. So why don’t you try li 12d. The original XR’s gearbox was never a strong point. Clean changes were poss 13need improving? Electronics was never my strong point. They hadn’t invented el 14 f things I shouldn’t. Tact never was my strong point, as Maxim will tell you. 15onfesses that finance has never been his strong point, broadens his horizons o 16at the young characters, never O’Casey’s strong point, were played with great 17 Economic analysis was never Trevelyan’s strong point and the England of the 18nomic management has never been Labour’s strong point. The opinion polls conti 19isation has never been the IT industry’s strong point, and the answer is “prob 20r. Contemporary art was anything but the strong point of the Salon, with Paris
phrases in which it occurs are not ambiguous. Some of the most frequent in the BNC are
from the point of view of; it is at this point that; point you in the right direction; but that’s not the point; she was on the point of; at pains to point out that; I don’t see the point
The two-word string strong point is still theoretically ambiguous, but in practice it is almost always used in an abstract sense. It can be used positively (My hon. friend makes a strong point). But, as in the illustrative lines in Concordance 1, it is most often the core of a speech act which has the form:
x
BE
NEG y’s strong point
Variable y is always a possessive pronoun or a proper name. Variable x is often something technical and/or numerical, and relates to the topic of the co-text. The whole unit is a conventional and ironic way of criticising someone by understatement. If you say that cooking is not her strong point, you mean ‘‘her cooking is terrible’’. We have a canonical form, with minor variants, and a clear pragmatic force.
A MODEL OF EXTENDED LEXICAL UNITS (ELUs) In a series of articles, Sinclair (1996, 1998, 2005) draws together his observations about co-selection and identifies semantic units of a kind which had not
124 SINCLAIR ON EMPIRICAL SEMANTICS
previously been described. He argues consistently that ‘the normal carrier of meaning is the phrase’ (Sinclair 2005), and that the lack of a convincing theory of phraseology is due to the faulty assumption that the word is the primary unit of meaning, and to the misleading separation of lexis and grammar. The model is extremely productive, and many further examples have been discovered by other linguists (e.g. Channell 2000; Francis 1993; Hunston 2007; Partington 1998, 2004; Stubbs 1995, 2001). In one of his best known case studies, Sinclair (1998) discusses the verb BUDGE. were stuck fast. We couldn’t the handle didn’t the rope would not neither bribe nor threat will Mrs Thatcher refused to
budge them. budge. She tried again and again budge. A tug-o’-war ensued budge him from the truth budge in her hostility to the EC
The meaning is not conveyed by the individual word alone. The verb regularly occurs in a longer construction, with either a grammatical or implied negative to the left (didn’t budge; refused to budge). A larger set of examples would show the things that typically don’t budge (e.g. doors, lids of jars, obstinate people): this is the ‘semantic preference’ of the verb, which relates to the topic of the text. The whole construction is used to tell a little narrative whose typicality we all recognize: the speaker has tried repeatedly to do something, has failed, and is now annoyed. This overall evaluative ‘semantic prosody’ is the communicative function of the whole unit. One of Sinclair’s main contributions to linguistic theory is his model of an ELU which is held together by relations of lexis, syntax, semantics, and pragmatics. These units (Sinclair 1996, 1998, 2005) have an obligatory core and an obligatory semantic prosody. Their internal structure has four parameters, which take different values and go from concrete to abstract: from observed word-forms (1) to hypothesized communicative functions (4). (1) COLLOCATION is the relation of co-occurrence between an obligatory core word or phrase (the node) and individual COLLOCATES: wordtokens which are directly observable and countable in texts. (2) COLLIGATION is the relation of co-occurrence between the node and abstract grammatical categories (e.g. past participles or quantifiers). A traditional category such as ‘‘negative’’ may be realized grammatically (would not budge) or semantically (would hardly budge, refused to budge).
MICHAEL STUBBS 125
(3) SEMANTIC PREFERENCE is the relation of co-occurrence between the phrasal unit and words from characteristic lexical fields. Recurrent collocates provide observable evidence of the characteristic topic of the surrounding text (e.g. typical subjects or objects of a verb). (4) SEMANTIC PROSODY is the function of the whole extended unit. It is a generalization about the communicative purpose of the unit: the reason for choosing it (and is therefore related to the concept of illocutionary force). Two further criteria in defining units are the strength of attraction between node and collocates (though there is considerable debate about which statistics are appropriate: see Evert 2005); and their distribution in text-types: whether they occur widely in general English or are restricted to broad varieties (e.g. journalism or technical and scientific English) or specialized text-types, such as recipes (add to taste), weather forecasts (warm front), or horse racing commentaries (entering the final furlong). In summary, the model has these components:
COLLOCATION COLLIGATION SEMANTIC PREFERENCE SEMANTIC PROSODY
tokens co-occurring word-forms classes co-occurring grammatical classes topics textual coherence motivation communicative purpose
Relations (1) and (2) deal with how linguistic signs relate syntagmatically to one another; (3) deals with how linguistic signs relate semantically to the topic of the text; (4) deals with how linguistic signs relate pragmatically to their users. This model of meaningful human action is therefore based on two ways of grasping linguistic units: (1) and (2) are evidence of (3) and (4). Relation (1) collocation and (2) colligation provide an objective view from the outside. They describe the observable behaviour of a social group. Collocation operationalizes the search for phrasal units: it makes only minimal assumptions about the division of text into orthographic words. Colligation assumes an analysis into grammatical classes: it is not directly observable. Collocation and colligation are part of our unconscious linguistic behaviour. Relation (3) semantic preference and (4) semantic prosody provide a subjective view from the inside. Textual collocates can be identified automatically by the software, and provide observable evidence, but discourse topics can be identified only intuitively, and semantic prosody is a hypothesis about the conscious and intentional social action. The pragmatic function is often difficult to formulate, partly because there may be no word in the language which can serve as a descriptive label (Sinclair 1998: 20). The hypothesis is based on observable data, but improved formulations—and counter-examples—are always possible.
126 SINCLAIR ON EMPIRICAL SEMANTICS
The model, especially the concepts of semantic preference and semantic prosody, has received substantial commentary (Partington 2004; Hunston 2007), but readers seem to have had problems in interpreting the model in three areas. (a) The model has largely been illustrated by individual case studies, and it remains unclear whether all phrases, or only some, have semantic prosodies. (b) The term ‘semantic prosody’ is often regarded as a rather vague indicator of the attitudinal connotation of a word: something good/desirable or bad/undesirable. This was perhaps the case with early examples (e.g. Sinclair 1991: 70–5; Louw 1993), but more recent analyses propose much more specific communicative functions, as in the BUDGE example (‘‘failed attempt and frustration’’). (c) There is confusion between semantic preference and semantic prosody, and it might be helpful to use different terms here, in order to distinguish between semantic relations (to the topic of the surrounding text), and the pragmatic function (of the whole phrasal unit). These problems can largely be solved if ‘semantic prosody’ is used in Sinclair’s own sense, to refer to communicative purpose. This is also Hunston’s (2007) interpretation: she distinguishes clearly between the attitudinal connotation of an individual word and the discourse function of an extended unit of meaning.
SEARCH METHODS: AN EXAMPLE In relatively early papers, Sinclair (1991: 110–15) distinguished between two principles: the idiom principle (that a speaker has available a large number of semi-preconstructed phrases) and the open-choice principle (to which speakers have recourse when no conventional phraseology is available). He later argued (Sinclair 1996) that this distinction is too sharp, and talked of two tendencies: the phraseological tendency (where words ‘go together and make meanings by their combinations’) and the terminological tendency (where words ‘have a fixed meaning in reference to the world’). The final move (Sinclair 1999b: 2–3) was to argue that ‘most of the so-called fixed expressions are not fixed at all’: there are just ‘variable expressions’. The ebb example above is typical: a phrasal unit has one frequent canonical form and a long tail of variants (cf. Moon 1998). This finding has consequences for search methodology (Tognini-Bonelli 2007) and for the level of abstraction at which the units must be described. In a project on which Sinclair was a consultant, Cheng et al. (2006, 2009) have developed a new generation of concordance software (ConcGram), which can search for variants of two or more lemmas, which co-occur in different sequences, either adjacent or not. For example, the lemmas CAUSE– PROBLEM frequently co-occur, but the lemmas vary in form (e.g. cause and causing) and in their relative position (e.g. causes a problem or a problem caused by),
MICHAEL STUBBS 127
Concordance 2. Illustrative examples of CAUSE–PROBLEM in a span of 6:6 in a 6-million-word corpus. Every third example. 01 dvantage of flexibility which can cause problems during installation. Fibregla 02 our behaviour. this can sometimes cause problems, for an over-night change in 03 go But money brings yu down Money causes problems anywea money is found, Food 04 t in present day Kazakhstan, were causing problems ( see below for more detail 05 under the rating system. This is causing problems for the local authority. It 06 ing the raft of legislation could cause real problems. On the single market, t 07 nsion on It’s not actually cos it causes real problems. In other words it can 08 ad confessed to the breaches that caused these problems. With this personal co 09 erratic nature of Patrice Lumumba caused constant problems- he frequently anno 10 ol it is only the fish leech that causes any serious problems. This fastens it 11 longer wanted as first homes they cause relatively few problems; but when they 12 cleared away and areas likely to cause fire or overheating problems—electri 13 at at once trivializes systematic causes of poverty and magnifies the problems 14 ty Council to act on the problems caused by the gypsies at the end of the Abin 15 shington to confront the problems caused by its continuing non-military suppor 16 at percentage of our problems are caused by hail, rain, wind, snow, blow, fire 17 variety of other problems can be caused by echo, the two most important of wh 18 owed when the problem is directly caused by an error in the sub-system being t 19 s and problems on the borderlands caused the country’s chieftain, Khama III, t 20 cant problem on college campuses, causing faculty, staff, students, and parent
and the collocates may be adjacent or not (e.g. causes problems and causes severe problems). To generate Concordance 2, the ConcGram software was set to find all examples of CAUSE–PROBLEM within a span of 6:6 in a corpus of 6 million running words. In this small illustrative sample, CAUSE is much more frequent as a verb, and PROBLEM is much more frequent in the plural. This could be checked in a larger corpus. Other software reveals further semantic variation. In the BNC, the top ten occurrences of CAUSE + (ADJ) + NOUN are: cause problems causing death cause trouble causes problems causing damage causing problems cause damage cause cancer cause difficulties cause injury
180 99 60 45 41 41 40 37 36 36
causing unnecessary suffering causing serious injury cause serious injury cause mental handicap cause serious damage cause severe damage caused extensive damage causing criminal damage caused considerable damage caused great concern
14 11 10 9 8 8 7 7 6 6
128 SINCLAIR ON EMPIRICAL SEMANTICS
The top ten occurrences of nouns before and after caused by are: damage caused by problems caused by injuries caused by pollution caused by difficulties caused by losses caused by disruption caused by loss caused by pain caused by delays caused by
165 103 36 29 21 19 18 18 18 17
caused caused caused caused caused caused caused caused caused caused
by by by by by by by by by by
negligence changes fire lack people differences smoking alcohol exposure bacteria
20 16 14 13 13 11 10 9 8 7
As Barlow (2004: 206–8) shows very clearly, these methods transform texts, by ripping them apart into concordance lines and lists of words and phrases. When these textual fragments are further rearranged into tables, alphabetically or by frequency, the arrangement becomes part of the argument. The methods are a necessary estrangement device, which force necessary distance between the observer and the way in which we normally experience running text. Without them, we cannot see the patterns, but because different methods of observation provide multiple views of the text, which highlight some patterns and obscure others, such methods are exploratory and corpus data are ‘essentially indirect’ (Sinclair 1999b: 1; 2004c: 7). Table 1 shows a lexical profile generated from all 3,250 occurrences of causing in the BNC. It shows, with their frequencies, the top 15 noun and verb collocates within a span of 2:2. Some collocates, especially to the right, are hyponyms of problems (e.g. alarm, confusion, delays, explosion, havoc). Several collocates are medical (e.g. cancer, cells, disease, gene, injury). Others are due to legal phrases such as charged with causing grievous bodily harm denies causing actual bodily harm So far, this analysis of CAUSE above generalizes across mixed reference corpora, looks only at the surrounding lexis, and ignores the detailed interaction between semantic field, grammar, and text-type. First, with an object such as damage or trouble, CAUSE can take a double object construction, but this is not possible with words for illnesses and diseases (Klotz 1997: 102–4). Semantic preference interacts with grammar: it caused a lot of trouble it caused him a lot of trouble smoking causes cancer *smoking caused him cancer Second, CAUSE has no necessarily undesirable connotations in scientific and technical English. Semantic prosody interacts with text-type: their effectiveness in causing the plastic flow of materials
MICHAEL STUBBS 129
Table 1: Lexical profile of causing in the BNC. The top 15 noun and verb collocates within a span of 2:2 N 2
N 1
NODE
N+1
N+2
charged (38) capable (26) risk (26) guilty (20) offence (20) responsible (13) accused (12) charges (12) convicted (11) blamed (9) fled (9) murder (9) person (7) body (6) court (6) charge (6) manslaughter (6) offences (6) suspected (6) admitted (5)
admitted (31) assault (28) avoid (12) face (9) factors (9) denies (7) risk (7) time (7) denied (7) deny (6) disease (6) factor (6) example (5) gene (5) people (5) person (5) system (5) virus (5) cells (4) cancer (4)
causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing causing
death (99) grievous (67) actual (47) serious (43) damage (41) problems (41) trouble (35) concern (31) severe (29) considerable (22) great (22) injury (22) unneccesary (21) pain (18) harm (17) havoc (15) widespread (15) further (14) offence (14) extensive (11)
bodily (115) damage (87) problems (51) problem (38) death (37) concern (32) injury (29) deaths (21) pain (19) trouble (18) suffering (16) harm (15) explosion (14) alarm (13) loss (13) delays (12) permitting 12) confusion (10) injuries (10) disease (10)
But third, Hunston (2007: 263) points out that this formulation is not quite right. It is when a human being is involved that CAUSE implies something undesirable. If cause and effect are inanimate (which is often the case in scientific and technical texts), then there is no necessary attitudinal connotation. In a word, there is co-selection of lexis, syntax, semantics, pragmatics, and text-type. We now have criteria for the prototype structure of the ELU. Canonical phrases include CAUSE problems and CAUSE serious damage. The most frequent nouns (problems and damage) have general superordinate meanings, and can therefore occur in a wide range of texts. The most frequent adjectives (severe and serious) provide evidence of the evaluative semantic prosody of the unit. Hyponyms of problems and damage (e.g. disruption, injury) are topic-dependent and make cohesive links within specific texts. The paradigmatic variation in noun collocates to left and right makes the unit fit into its co-text, and are evidence of the semantic preference of the unit. The topic might be pollution and the cause might be negligence. More specialized again are technical
130 SINCLAIR ON EMPIRICAL SEMANTICS
expressions, such as CAUSE grievous bodily harm, which typically occur in legal and journalistic texts. In scientific and technical texts the semantic preference and the semantic prosody are likely to be cancelled. The vertical and horizontal axes of concordance data and lexical profiles provide evidence of form and meaning respectively. The vertical axis provides evidence of recurrent forms across many instances of language use: this is often very obvious even to superficial inspection. The horizontal axis provides evidence of meaning, both in individual instances (individual concordance lines) and also in the language system. This is often less obvious, because different collocates typically occur in different instances, and it needs the experience and intuition of the analyst to group these collocates into semantic sets. In addition, we typically have a few prominent items, with a long tail of less frequent or unique exponents (Sinclair 2004b: 286), and the appropriate level of delicacy in description remains a decision for language teachers or lexicographers. The software provides objective and quantifiable evidence, but the best way to present such findings requires judgement. As Labov (1972: 258) puts it: ‘By the time the analyst knows what to count, the problem is practically solved’.
SUMMARY AND IMPLICATIONS John Sinclair’s work shows how to use empirical evidence to tackle one of the deepest questions in the philosophy of language: the nature of units of meaning. As Kant put it (Scruton 2001/2004: 31, 41): pure empiricism is content without form, but pure rationalism is form without content. Sinclair’s model has both: corpus observations provide high empirical content, and interpretation provides the elegant phraseological model. Sinclair questions traditional distinctions between logic, rhetoric, and grammar, which have been familiar since the medieval Trivium, and which are still often taken for granted. He is sceptical of logical approaches to language (e.g. natural language processing), he reinterprets rhetoric as discourse and builds concepts of communicative function into ELUs, and he severely questions the invented introspective data which were prominent in grammatical theory from the 1960s onwards. As he says, alluding to this tradition, ‘one does not study all of botany by making artificial flowers’ (Sinclair 1991: 6). His work supports a long and very different tradition of text and corpus analysis, derived from his own British teachers and colleagues (especially Firth and Halliday), but represented in a broader European tradition (for example, Jespersen) and in a much more restricted American tradition (for example, Fries). This tradition has always emphasized the description of meaning, but it is Sinclair who makes the most sustained attempt to develop an empirical semantics: what Hanks (1997) has called his ‘ferocious empiricism’. However, his research programme remains to be developed in several directions. Implications and applications for teaching materials are discussed by
MICHAEL STUBBS 131
Willis (1990), for pattern grammar by Hunston and Francis (2000), and for lexicography by Moon (2007) and by the articles in the special issue of the International Journal of Lexicography (21/3, 2008) which is devoted to Sinclair’s work. I will mention here some implications of the model of phraseology.
1. ‘The ultimate dictionary’ (Sinclair et al. 1970/2004: xxiv) A major descriptive task, for both theoretical and applied linguists, is a comprehensive list of the units of meaning in a given language. The general method is clear. If we discover phrases which are both frequent and widely distributed across a corpus, then they are not text-dependent, but part of the patterns of the language (Sinclair et al. 1970/2004: 79). The major descriptive problem is that the units are internally variable and have indeterminate boundaries, and that similar units are often related to each other in taxonomic hierarchies (Croft 2001: 25; Stubbs 2007). The general solution is to describe their canonical, prototypical forms, but deciding the appropriate level of delicacy for different purposes is a matter of interpretation. We have many convincing case studies and the beginnings of such descriptions in corpus-based dictionaries, but we still need thorough-going phraseological dictionaries.
2. A theory of textual cohesion The question then arises as to why certain phrasal units occur in many different texts. One answer is that they do not depend on the topic of the text, but serve text-management functions, such as signalling narrative structure, topicalization, point of view, and the like. This provides the link between text and corpus analysis. Here are just two examples. First, the double verb construction went and VERBed is a conventional way of marking the end of a segment of narrative.
he put the phone down and went and got himself a malt whisky so I went and toddled off to find somebody and then, would you believe it, she went and married him then you went and got married again, you can’t help yourself then he went and jumped out of a plane then she went and spoiled everything by behaving as if pissed Paul Bodin then went and missed a penalty
In (a) and (b) the verb went seems redundant: it is assumed by the action which follows. In (c) to (g), went cannot be interpreted literally at all: it is a pragmatic signal of something the speaker did not expect or did not approve of. Along with the co-occurring (and) then or so, it also signals the conclusion of part of the action.2 Second, there are several words (e.g. blatant, downright, mere, outright, sheer, utter) which have purely pragmatic meanings. They do not denote anything
132 SINCLAIR ON EMPIRICAL SEMANTICS
in the world, but signal the attitude of the speaker and a textual contrast. For example, the x and/or outright y construction contributes to textual coherence by contrasting two points on a scale: a middle point and an extreme point of which the speaker disapproves. fraught with inaccuracies and through ruthless speculation and deeply ingrained suspicion and signs of strain or misguided speculation or imbued with deep suspicion or
outright outright outright outright outright outright
fallacies fraud hostility contradiction corruption opposition
An increasing number of case studies (e.g. Hunston and Francis 2000: 185–8; Partington and Morley 2004; Hoey 2005; Mahlberg 2005) show how lexis and phraseology contribute to textual structure and organization. If they could be integrated, such case studies would contribute to a functional theory of lexis.
3. A theory of collocation We know from many descriptive studies that the syntagmatic attraction between linguistic units is much stronger than is often realized, and this attraction can be measured in different ways. For example, the Cobuild (1995b) database gives the 10,000 most frequent word-forms along with their 20 most frequent collocates in a 200-million word corpus. In the following illustrations (from Stubbs 2006), a statement such as node means that a node word co-occurs in 10 per cent of cases with its top collocate within a span of four words to left or right:
drive ; origin ; basis room ; hate ; ending chopped ; answers ; pose solving <problem 37%>; conditioning ; eighteenth
Indeed, the strength of attraction between collocates is much stronger than such figures suggest, since such calculations, between two individual wordforms, would count separately inflicts damage, inflicted a wound, and inflicting an injury, although all three phrases seem part of a single semantic pattern. Kilgarriff (2006) goes a step further and identifies those words which usually co-occur with a small set of particular collocates, rather than appearing freely with large numbers of collocates. Using data from the BNC, he identifies the 100 nouns and verbs which show the ‘most collocational’ behaviour (for nouns, with respect to the verbs they are objects of, and for verbs, with respect
MICHAEL STUBBS 133
to their object-nouns). Many high frequency nouns frequently occur with light support-verbs, and low frequency nouns often occur largely in idioms:
take place, take care, take advantage, pay attention take the mickey, come a cropper, enter/join the fray
Corpus studies have now documented in detail that people’s language use is typically very repetitive, but it is far from trivial to explain what level of repetitiveness is optimum in a functioning system. Semantic and functional relations between collocates are studied by Mel’cˇuk (1998) and Jones (2002), and psycholinguistic factors are studied by Wray (2002, 2008) and Ellis et al. (in press). We need to move from description to explanation by using both semantic and psycholinguistic data.
4. A theory of social action In his inaugural lecture, given only a few years after the publication of How to Do Things with Words (Austin 1962), Sinclair (1966) argued that ‘utterances do things rather than just mean things’ (emphasis in original). Sinclair (2008: 24) returns to this point that, before Austin, linguists were ‘embarassingly short’ of ideas about language interaction. Both speech acts and ELUs build the speaking agent into units of language structure, and the concept of illocutionary force seems close to the concept of semantic prosody. Yet, despite his early work on classroom discourse, and despite the social orientation of work by Firth and Halliday, Sinclair remained notoriously unwilling to draw out the social-psychological implications of his work on phraseology. It is only corpora which can provide data for studying prosodies from the bottom up, and therefore show how we could do real ‘ordinary language philosophy’. Austin and Searle work deductively. They use only invented introspective data. They propose a powerful theory of speech acts, and Searle (1995) is strong on social theory. Conversely, Sinclair works inductively. He uses authentic empirical data. He also proposes a powerful theory of functional language units, but his work lacks a general social theory. We need a clearer understanding of the relation between these two traditions.
5. A theory of how people construe the world Despite his resistance to socio-psychological approaches, Sinclair’s phraseological model does use unashamedly psychological concepts such as motivation and evaluation. ELUs, such as the wouldn’t budge construction (Sinclair 1998), are little schemas of cultural knowledge. The par for the course construction signals that things have gone wrong, yet again, in just the way that we would have predicted (Channell 2000: 47–50). The not the end of the world construction signals sympathy for someone who has suffered a disappointment (Stubbs 2007). ELUs assume typical scenarios in the everyday world, and presumably we learn how to construe the social world through the phraseology. Speakers are conscious social agents whose constant comments
134 SINCLAIR ON EMPIRICAL SEMANTICS
on the actions of themselves and others point to phenomena which really exist (Harre´ and Secord 1972). To complain is part of being discontented, to express frustration is part of being frustrated. ELUs are part of the way in which actions and states of mind are mediated by language. Meaning depends on intertext: it derives from what has been said frequently and successfully by innummerable speakers in the past. But often the semantic prosodies have no conventional name (Sinclair 1998: 20) and can be discovered only by studying how the lexico-grammar expresses taken-for-granted cultural meanings. We need to compile ‘a grammar of the typical meanings that human communication encodes’ (Francis 1993: 155) and relate the linguistic and cognitive schemata which underlie how we understand social behaviour (Teubert 2008). These five points indicate how Sinclair’s empirical work could be developed in linguistic, cognitive and social directions. Although he seems generally unwilling to speculate on such topics, Sinclair (2007: 1) points out that, whereas corpus linguists like nothing better than empirical findings supported by levels of statistical significance, people outside this narrow circle want to know how it all hangs together, and how all the empirical information contributes to solving the great intellectual puzzles of language in society. He asks: How should all this work be evaluated? How does empirical linguistics contribute to ‘wider issues’, and how can it be used ‘as a foundation for a broad range of intellectual exploration’? His approach to language may be ‘ferociously empiricist’, but it is the antithesis of naive positivism. By changing the frame of reference for our observations, Sinclair has shown ways of describing the order which underlies the apparent chaos of parole. His approach to phraseology is a major advance in describing the facts. We now have to ask how far this approach can help in explaining the facts.
SUPPLEMENTARY DATA There is online Supplementary Data available at Applied Linguistics online.
CONVENTIONS AND CORPORA LEMMAS are represented as small caps. Word-forms are represented in lower case italic. ‘‘Meanings’’ are in double quotes. ‘Quotes’ from other authors and technical terms are in single quotes. enclose typical collocates of a node-word. Many of the examples are from the British National Corpus (BNC): 100 million running words of contemporary English, 90 million written and 10 million spoken, sampled from over 4,000 texts from a wide range of text-types. Many of the examples of n-grams (recurrent multi-word strings) were extracted from the BNC by using William Fletcher’s BNC-interface at http://www.phrasesinenglish.org. This database allows us to start from a node word, and extract all recurrent n-grams with the node at each position, down to a specified cut-off frequency. References to a six-million word corpus are to a collated version of the Brown, LOB, Frown, FLOB and BNC-baby
MICHAEL STUBBS 135
corpora: this comprises five million written and one million spoken running words.
ACKNOWLEDGEMENTS For comments on a previous version of this paper, I am grateful to Bettina Fischer-Starcke and to three anonymous reviewers.
NOTES 1
Obituaries appeared in The Guardian (3 May 2007), The Scotsman (10 May 2007) and Functions of Language (14/2: 2007). Special issues of two journals are devoted to papers on Sinclair’s work: International Journal of Corpus Linguistics (12/2: 2007) and International Journal of Lexicography (21/3: 2008).
2
Airola (2007) studies the function of similar double verb constructions in Finnish, and shows that they seem semantically redundant if regarded out of context, but that they can be explained by looking at their pragmatic function.
REFERENCES Airola, A. 2007. Verb Patterns in Texts. Publication 41. Dept of General Linguistics, University of Helsinki. Austin, J. L. 1962. How to Do Things with Words. Oxford: Clarendon. Barlow, M. 2004. ‘Software for corpus access and analysis’ in J. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam: Benjamins, pp. 205–21. Barnbrook, G. 1996. Defining Sentences. Amsterdam: Benjamins. Channell, J. 2000. ‘Corpus-based analysis of evaluative lexis’ in S. Hunston and G. Thompson (eds): Evaluation in Text. Oxford: Oxford University Press, pp. 38–55. Cheng, W., C. Greaves, and M. Warren. 2006. ‘From n-gram to skip-gram to concgram,’ International Journal of Corpus Linguistics 11/4: 411–33. Cheng, W., C. Greaves, J. Sinclair, and M. Warren. 2009. ‘Uncovering the extent of the phraseological tendency: towards a systematic analysis of concgrams,’ Applied Linguistics 30/2: (in press). Cobuild. 1995a. Collins Cobuild English Dictionary. London: HarperCollins. Cobuild. 1995b. Collins Cobuild English Collocations on CD-ROM. London: HarperCollins.
Croft, W. 2001. Radical Construction Grammar. Oxford: Oxford University Press. Ellis, N., E. Frey, and I. Jalkanen. in press. ‘The psycholinguistic reality of collocation and semantic prosody (1): lexical access’ in U. Ro¨mer and R. Schulze (eds): Exploring the Lexis–Grammar Interface. Amsterdam: Benjamins. Evert, S. 2005. ‘The Statistics of Word Co-occurrences: Word Pairs and Collocations’, Ph.D. thesis, University of Stuttgart. Fahnestock, J. 2003. ‘Verbal and visual parallelism,’ Written Communication 20/2: 123–52. Firth, J. R. 1935. ‘The technique of semantics,’ Transactions of the Philological Society 36–72. Firth, J. R. 1957. ‘A synopsis of linguistic theory 1930–1955,’ Transactions of the Philological Society. Special Volume. Oxford: Blackwell, pp. 1–32. Francis, G. 1993. ‘A corpus-driven approach to grammar’ in M. Baker, G. Francis, and E. Tognini-Bonelli (eds): Text and Technology. Amsterdam: Benjamins, pp. 137–56. Francis, W. N. and H. Kucˇera. 1967. Computational Analysis of Present Day American English. Providence: Brown University Press. Gerbig, A. and O. Mason. (eds) 2008. Language, People, Numbers. Amsterdam: Rodopi.
136 SINCLAIR ON EMPIRICAL SEMANTICS
Greaves, C. 2007. ‘Software demonstration,’ Keyness in Text, Certosa di Pontignano, University of Siena, Italy, 26–30 June. Gross, M. 1993. ‘Local grammars and their representation by finite automata’ in M. Hoey (ed.): Data, Description, Discourse. London: Harper Collins, pp. 26–38. Hanks, P. 1997. ‘Ferocious empiricism: Review of Sinclair’s On Lexis and Lexicography,’ International Journal of Corpus Linguistics 2/2: 289–95. Harre´, R. and P. F. Secord. 1972. The Explanation of Social Behaviour. Oxford: Blackwell. Harris, R. 1988. Language, Saussure and Wittgenstein. London: Routledge. Hoey, M. 2005. Lexical Priming. London: Routledge. Hoey, M., M. Mahlberg, M. Stubbs, and W. Teubert. 2007. Text, Discourse and Corpora: Theory and Analysis. London: Continuum. Hunston, S. 2007. ‘Semantic prosody revisited,’ International Journal of Corpus Linguistics 12/2: 249–68. Hunston, S. and G. Francis. 2000. Pattern Grammar. Amsterdam: Benjamins. Hunston, S. and J. Sinclair. 2000. ‘A local grammar of evaluation’ in S. Hunston and G. Thompson (eds): Evaluation in Text. Oxford: Oxford University Press, pp. 74–101. Jones, S. 2002. Antonymy. London: Routledge. Kilgarriff, A. 1997. ‘Putting frequencies in the dictionary,’ International Journal of Lexicography 10/2: 135–55. Kilgarriff, A. 2006. ‘Collocationality (and how to measure it)’ in E. Corino, C. Marello Carla, and C. Onesti (eds): Proceedings XII Euralex International Congress. Turin, Italy, Alessandria: Edizioni dell’Orso. Klotz, M. 1997. ‘Ein Valenzwo¨rterbuch englischer Verben, Adjektive und Substantive,’ Zeitschrift fu¨r Angewandte Linguistik 27: 93–111. Labov, W. 1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press. Louw, B. 1993. ‘Irony in the text or insincerity in the writer? The diagnostic potential of semantic prosodies’ in M. Baker, G. Francis, and E. Tognini-Bonelli (eds): Text and Technology. Amsterdam: Benjamins, pp. 157–76. Macfarlane, A. and G. Martin. 2002. The Glass Bathyscaphe: How Glass Changed the World. London: Profile.
Mahlberg, M. 2005. English General Nouns. Amsterdam: Benjamins. Mauranen, A. 2004. ‘Spoken corpus for an ordinary learner’ in J. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam: Benjamins, pp. 89–105. Mel’cˇuk, I. 1998. ‘Collocations and lexical functions’ in A. P. Cowie (ed.): Phraseology. Oxford: Oxford University Press, pp. 23–53. Moon, R. 1998. Fixed Expressions and Idioms in English. Oxford: Clarendon. Moon, R. 2007. ‘Sinclair, lexicography and the COBUILD project,’ International Journal of Corpus Linguistics 12/2: 159–81. Partington, A. 1998. Patterns and Meanings. Amsterdam: Benjamins. Partington, A. 2004. ‘Utterly content in each other’s company: Semantic prosody and semantic preference,’ International Journal of Corpus Linguistics 9/1: 131–56. Partington, A. and J. Morley. 2004. ‘At the heart of ideology: Word and cluster/ bundle frequency in political debate’ in B. Lewandowska-Tomaszczyk (ed.): Practical Applications in Language and Computers. Frankfurt am Main: Lang. Scruton, R. 2001. Kant. New edition. Oxford: Oxford University Press. (Page refs to German translation, Freiburg: Herder, 2004.) Searle, J. 1995. The Construction of Social Reality. Harmondsworth: Allen Lane. Sinclair, J. 1966. ‘Indescribable English’ Inaugural lecture, University of Birmingham. (Abstract in Sinclair, J. and R. M. Coulthard. 1975. Towards an Analysis of Discourse. Oxford: Oxford University Press, p. 151.) Sinclair, J. (ed.) 1987. Looking Up: An Account of the Cobuild Project in Lexical Computing. London: Collins. Sinclair, J. 1991. Corpus Concordance Collocation. Oxford: Oxford University Press. Sinclair, J. 1992. ‘Trust the text’ in M. Davies and L. Ravelli (eds): Advances in Systemic Linguistics. London: Pinter, pp. 5–19. (Reprinted in Sinclair 2004a: 9–23.) Sinclair, J. 1996. ‘The search for units of meaning,’ Textus 9/1: 75–106. (Reprinted in Sinclair 2004a: 24–48.) Sinclair, J. 1998. ‘The lexical item’ in E. Weigand (ed.): Contrastive Lexical Semantics. Amsterdam: Benjamins, pp. 1–24. (Reprinted in Sinclair 2004a: 131–48.)
MICHAEL STUBBS 137
Sinclair, J. 1999a. ‘A way with common words’ in H. Hasselga˚rd and S. Oksefjell (eds): Out of Corpora. Amsterdam: Rodopi, pp. 157–79. Sinclair, J. 1999b. ‘The computer, the corpus and the theory of language’ in G. Azzaro and M. Ulrych (eds): Transiti Linguistici e Culturali. Trieste: EUT, pp. 1–15. Sinclair, J. 2004a. Trust the Text. London: Routledge. Sinclair, J. 2004b. ‘New evidence, new priorities, new attitudes’ in J. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam: Benjamins, pp. 271–99. Sinclair, J. (ed.) 2004c. How to Use Corpora in Language Teaching. Amsterdam: Benjamins. Sinclair, J. 2005. ‘The phrase, the whole phrase and nothing but the phrase’ Plenary lecture, Phraseology 2005, Louvain-la-Neuve, October. Sinclair, J. 2007. ‘Introduction’ in M. Hoey, M. Mahlberg, M. Stubbs, and W. Teubert. 2007. Text, Discourse and Corpora: Theory and Analysis. London: Continuum, pp. 1–5. Sinclair, J. 2008. ‘Borrowed ideas’ in A. Gerbig and O. Mason (eds): Language, People, Numbers. Amsterdam: Rodopi, pp. 21–42. Sinclair, J. and R. M. Coulthard. 1975. Towards an Analysis of Discourse. Oxford: Oxford University Press. Sinclair, J. and A. Mauranen. 2006. Linear Unit Grammar. Amsterdam: Benjamins. Sinclair, J., S. Jones, and R. Daley. 1970/2004. English Collocation Studies: The OSTI Report, ed. R. Krishnamurthy. London: Continuum. (Originally mimeo 1970.) Stubbs, M. 1995. ‘Collocations and semantic profiles,’ Functions of Language 2/1: 23–55.
Stubbs, M. 2001. Words and Phrases. Oxford: Blackwell. Stubbs, M. 2006. ‘Corpus analysis: The state of the art and three types of unanswered questions’ in G. Thompson and S. Hunston (eds): System and Corpus. London: Equinox, pp. 15–36. Stubbs, M. 2007. ‘Quantitative data on multiword sequences in English: the case of the word world’ in M. Hoey, M. Mahlberg, M. Stubbs, and W. Teubert. 2007. Text, Discourse and Corpora: Theory and Analysis. London: Continuum, pp. 163–89. Taylor, J. R. 2002. Cognitive Grammar. Oxford: Oxford University Press. Teubert, W. 2008. ‘Some notes on the concept of cognitive linguistics’ in A. Gerbig and O. Mason (eds): Language, People, Numbers. Amsterdam: Rodopi, pp. 61–84. Tognini-Bonelli, E. 2004. ‘Working with corpora’ in C. Coffin, A. Hewings, and K. O’Halloran (eds): Applying English Grammar. London: Arnold, pp. 11–24. Tognini-Bonelli, E. 2007. ‘Bits and pieces’ Plenary lecture, Conference on Corpus and Cognitive Linguistics, University of Erlangen, Germany, 25–27 October. Warren, M. 2007. ‘Making sense of phraseological variation’ Plenary lecture, Keyness in Text, Certosa di Pontignano, University of Siena, Italy, 26–30 June. Willis, D. 1990. The Lexical Syllabus. London: Collins ELT. Wray, A. 2002. Formulaic Language and the Lexicon. Cambridge: Cambridge University Press. Wray, A. 2008. Formulaic Language: Pushing the Boundaries. Oxford: Oxford University Press.
John McHardy Sinclair: Publications This is an attempt to compile a comprehensive bibliography of John Sinclair's publications: not an easy task, since some articles appeared in obscure journals, festschriften and other collections, and some were then reprinted, sometimes revised and/or with different titles, sometimes in equally obscure places. (An extreme example of an obscure place is 2006a, an on-line community newspaper from the west of Scotland.) For some publications which I have not been able to see, the page references are missing. A couple of items are preconference abstracts, but contain careful statements of basic principles. (An example is 2005a, a quite substantial statement about phraseological theory.) This version of the bibliography was last updated on 3 October 2008.
Michael Stubbs, FB2 Anglistik, University of Trier, Germany.
1.
BOOKS (single and joint authored and edited) Sinclair, J. & Mauranen, A. (2006) Linear Unit Grammar. Integrating Speech and Writing. Amsterdam: Benjamins. Sinclair, J. (2004a) Trust the Text. London: Routledge. Sinclair, J. (ed.) (2004b) How to Use Corpora in Language Teaching. Amsterdam: Benjamins. Sinclair, J., Jones, S. & Daley, R. (2004) English Collocation Studies. The OSTI Report. Edited by R. Krishnamurthy. London: Continuum. [Originally circulated as a mimeoed report, Sinclair, Jones, & Daley (1970), see below.] Sinclair, J. (2003a) Reading Concordances: an Introduction. London: Pearson Longman. Sinclair, J., Payne, J. & Hernández, C.P. (eds) (1996) Special issue. International Journal of Lexicography, 9, 3. Sinclair, J., Hoelter, M. & Peters C. (eds) (1994) The Language of Definition: The Formalization of Dictionary Definitions for Natural Language Processing. Vol. 7, Studies in Machine Translation and Natural Language Processing. Brussels: The European Commission. Sinclair, J., Hoey, M. & Fox, G. (eds) (1993) Techniques of Description: Spoken and Written Discourse; a Festschrift for Malcolm Coulthard. London: Routledge. Sinclair, J. & Coulthard, R. M. (eds) (1992) Advances in Spoken Discourse Analysis. London: Routledge.
2
Sinclair, J. (1991a) Corpus, Concordance, Collocation. Oxford: Oxford University Press. Sinclair, J. (ed) (1987) Looking Up: an Account of the Cobuild Project in Lexical Computing. London: Collins. Sinclair, J. & Brazil, D. C. (1982) Teacher Talk. Oxford: Oxford University Press. Sinclair, J. (ed) (1980) Skills of Learning. 4 vols. Nelson: University of Malaya Press. Sinclair, J., Coulthard, M. & Krumm, J. (1977) Analyse der Unterrichtssprache: Ansätze zu einer Diskursanalyse dargestellt am Sprachverhalten englischer Lehrer und Schüler. Heidelberg: Quelle & Meyer. Sinclair, J. & Coulthard, R. M. (1975) Towards an Analysis of Discourse: the English Used by Teachers and Pupils. Oxford: Oxford University Press. Sinclair, J. (1972a) A Course in Spoken English: Grammar. Oxford: Oxford University Press. Sinclair, J., Coulthard, R. M., Forsyth I. J. & Ashby M. C. (1972) The English used by Teachers and Pupils. Final Report to the Social Science Research Council for the period September 1970 to August 1972. Birmingham: University of Birmingham Press. [Revised as Sinclair & Coulthard 1975.] Sinclair, J., Jones, S. & Daley, R. (1970) English Lexical Studies: The OSTI Report. Birmingham: University of Birmingham Press. [Republished as Sinclair, Jones & Daley 2004.] 2.
COBUILD LANGUAGE REFERENCE MATERIALS Sinclair was 'Editor in Chief' and then 'Founding Editor in Chief' of a large series of dictionaries, grammars and other language reference materials. The first dictionary, based on a 20 million word corpus, was (1987) Collins Cobuild English Language Dictionary. London: HarperCollins. The first grammar was (1990) Collins Cobuild English Grammar. London: HarperCollins. These books went through many different editions, for example a Collins Cobuild Essential English Dictionary (1988) for intermediate learners, and then revisions based on larger corpora, plus dictionaries for other languages some of which are listed in Sinclair (1999d). These materials are not further listed here.
3
3.
ARTICLES Sinclair, J. (2008) Borrowed ideas. In A. Gerbig & O. Mason (eds) Language, People, Numbers: Corpus Linguistics and Society. Amsterdam/New York: Rodopi. 21-41. Sinclair, J. (2007a) Introduction. In M. Hoey, M. Stubbs, W. Teubert & M. Mahlberg (eds): Text, Discourse and Corpora: Theory and Analysis. London: Continuum. 1-5. Sinclair, J. (2007b) Preface. International Journal of Corpus Linguistics. (Special Issue. Words, Grammar, Text: Revisiting the Work of John Sinclair.) 12, 2: 155-57. Sinclair, J. (2007c) The exploitation of meaning: literary text and local grammars. In I. Bas & D. Freeman (eds): PALA Papers 2: Challenging the Boundaries. Amsterdam: Rodopi. 1-36. Sinclair, J. (2006a) A language landscape. West Word. January 2006 Issue. http://road-tothe-isles.org.uk/westword/jan2006/.html (Accessed 3 Nov 2007). Sinclair, J. (2006b) Small words make big meanings. Plenary, AAAL (American Association for Applied Linguistics). [http://www.aaal.org/aaal2006/program.pdf, Abstract p.92]. Tognini-Bonelli, E. & Sinclair, J. (2006) A short history of the corpus. [Unpublished?] Sinclair, J. (2005a) The phrase, the whole phrase and nothing but the phrase. [Abstract in C. Cosme, C. Gouverneur, F. Meunier & M. Paquot (eds). Phraseology 2005. The Many Faces of Phraseology. An Interdisciplinary Conference. Louvain-la-Neuve. 1922.] Sinclair, J. (2005b) Corpus and text: basic principles. In M. Wynne (ed). Developing Linguistic Corpora: a Guide to Good Practice. Oxford: Oxbow Books. 1-16. Available (accessed 21 February 2008) at http://www.ahds.ac.uk/creating/guides/linguistic-corpora/chapter1.htm. Sinclair, J. (2005c) Appendix to chapter one: how to make a corpus. In M. Wynne (ed). Developing Linguistic Corpora: a Guide to Good Practice. Oxford: Oxbow Books. 79-83. Available (accessed 21 February 2008) at http://www.ahds.ac.uk/creating/guides/linguistic-corpora/appendix.htm. Sinclair, J. (2005d) Language as a string of beads: discourse and the M-word. In E. Tognini-Bonelli & G. Del Lungo Camiciotti (eds). Strategies in Academic Discourse. Amsterdam: Benjamins. 163-68. Sinclair, J. (2004c) Introduction. In Sinclair 2004b: 1-10. Sinclair, J. (2004d) New evidence, new priorities, new attitudes. In Sinclair 2004b: 271-99.
4
Sinclair, J. (2004e) Meaning in the framework of corpus linguistics. Lexicographica, 20: 20-32. [Reprinted in W. Teubert (ed) (2005) Lexicographica. Tübingen: Niemeyer. 20-32.] Sinclair, J. (2004f) Intuition and annotation: the discussion continues. In. B. Altenberg & K. Aijmer (eds). Advances in Corpus Linguistics. [Papers from ICAME 23.] Amsterdam: Rodopi. 39-59. Sinclair, J. (2004g) Corpus creation. In G. Sampson & D. McCarthy (eds). Corpus Linguistics: Readings in a Widening Discipline. London: Continuum. 78-84. [Excerpted from Sinclair 1991a.] Sinclair, J. (2004h) In praise of the dictionary. In G. Williams & S. Vessier (eds) Proceedings of the Eleventh EURALEX Congress, EURALEX 2004. Lorient: Université de Bretagne Sud. 1-11. Sinclair, J. (2003b) Corpora for lexicography. In P. van Sterkenberg (ed). A Practical Guide to Lexicography. Amsterdam: Benjamins. 167-78. Sinclair, J. (2003c) Corpus processing. In P. van Sterkenberg (ed). A Practical Guide to Lexicography. Amsterdam: Benjamins. 179-93. Sinclair, J. (2002) Phraseognomy. In S. Nuccorini (ed). Phrases and Phraseology, Data and Descriptions. Bern: Lang. 17-26. [Reprinted in 2004a: 177-84.] Sinclair, J. (2001a) A tool for text explication. In K. Aijmer (ed). A Wealth of English. Göteborg: Acta Universitatis Gothoburgensis. 163-76. [Reprinted in 2004a: 115-27.] Sinclair, J. (2001b) The deification of information. In M. Scott & G. Thompson (eds). Patterns of Text: in Honour of Michael Hoey. Amsterdam: Benjamins. 287-313. Sinclair, J. (2001c) Lexical grammar. In M. Gellerstam, K. Jóhanesson, B. Ralph & L. Rogström (eds). Nordiska Studier I Lexicografi 5. Proceedings of the Fifth Conference in Nordic Lexicography. Göteborg: Göteborgs Universitet. 323-43. [Reprinted in B. Hary (ed) (2003) Corpus Linguistics and Modern Hebrew. Tel-Aviv University: Chaim Rosenberg School of Jewish Studies. And in Sinclair 2004a: 16476. Sinclair, J. (2001d) Preface. In M. Ghadessy, A. Henry & R.L. Roseberry (eds). Small Corpus Studies and ELT. Amsterdam: Benjamins. vii-xvi. Sinclair, J. (2001e) Passion speechlesse lies. In C. K. Tong, A. Pakir, B. K. Choon & R. Goh (eds). Ariels: Departures and Returns. Essays for Edwin Thumboo. Oxford: Oxford University Press. 211-30. Sinclair, J. (2001f) Review of The Longman Grammar of Spoken and Written English. International Journal of Corpus Linguistics. 6, 2: 339-59.
5
Barnbrook, G. & Sinclair, J. (2001) Specialised corpus, local and functional grammars. In M. Ghadessy, A. Henry & R. L. Roseberry (eds). Small Corpus Studies and ELT. Amsterdam: Benjamins. 237-78. Sinclair, J. (2000) Current issues in corpus linguistics. In R. R. Favretti (ed) Linguistica e Informatica. Rome: Bulzoni. 2-38. [Reprinted in B. Hary (ed). Corpus Linguistics and Modern Hebrew. Tel-Aviv University: Chaim Rosenberg School of Jewish Studies. Pages ?. And in 2004a: 185-93.] Sinclair, J. & Hunston, S. (2000) A local grammar of evaluation. In S. Hunston & G. Thompson (eds). Evaluation in Text: Authorial Stance and the Construction of Discourse. Oxford: Oxford University Press. 74-101. Sinclair, J. (1999a) The internalisation of dialogue. In R. R. Favretti, G. Sandri & R. Scazzieri (eds). Incommensurability and Translation, Essays in Honour of Thomas S. Khun. Cheltenham: Elgar. 391-406. [Reprinted in Sinclair 2004a: 102-14.] Sinclair, J. (1999b) A way with common words. In H. Hasselgård & S. Oksefjell (eds). Out of Corpora, Studies in Honour of Stig Johansson. Amsterdam: Rodopi. 157-75. Sinclair, J. (1999c) The lexical item. In E. Weigand (ed). Contrastive Lexical Semantics. Amsterdam: Benjamins. 1-24. [Reprinted in Sinclair 2004a: 131-48.] Sinclair, J. (1999d) Data-derived multilingual lexicons. In E. Arcaini (ed). La Traduzione (IV), Quaderni di Libre e Riviste d'Italia, 43, Ministero per i Beni e le Attivita Culturali. 33-46. [Also in International Journal of Corpus Linguistics, 2001, 6: 7994. Reprinted in W. Teubert (ed) (2007) Text Corpora and Multilingual Lexicography. Amsterdam: Benjamins. 69-81.] Sinclair, J. (1999e) New roles for language centres: the mayonnaise problem. In D. Bickerton & M. Gotti (eds). Language Centres: Integration through Innovation. CercleS. (Confédération Européenne des Centres de Langues de l'Enseignement Supérieur). University of Plymouth: Department of Modern Languages. 31-50. Sinclair, J. (1999f) The computer, the corpus and the theory of language. In G. Azzaro & M. Ulrych (eds). Transiti Linguistici e Culturali. Trieste: EUT. 1-15. [Reprinted in LMS Lingua, 1999, 1, 9: 24-32.] Sinclair, J. (1998a) Korpustypologie: ein Klassifikationsrahmen. In W. Teubert (ed). Neologie und Korpus. Tübingen: Narr. 111-28. Sinclair, J., Mason, O., Ball, J. & Barnbrook, G. (1998b) Language independent statistical software for corpus exploration. Computers and the Humanities, 31, 3: 229-55. Sinclair, J. (1998c) Large corpus research and foreign language teaching. In R. de Beaugrande, M. Grosman & B. Seidlhofer (eds). Language Policy and Language Education in Emerging Nations. Stamford. Conn.: Ablex. 79-86.
6
Sinclair, J. (1997) Corpus linguistics at the millennium. In D. Wolff, J. Kohn & B. Ruschoff (eds). New Horizons in CALL. Szombathely: Bersenyi Dániel College. Pages ? Sinclair, J. (1996a) The search for units of meaning. Textus, 9, 1: 75-106. [Reprinted in G.Corpas Pastor (ed) (2000) Las Lenguas de Europa. Granada: Comares. 7-38. And in Sinclair 2004a: 24-48.] Sinclair, J. (1996b) The empty lexicon. International Journal of Corpus Linguistics, 1, 1: 99-119. [Reprinted in Sinclair 2004a: 149-63.] Sinclair, J. (1996c) Prospects for automatic lexicography. In A. Zettersten & V. Pedersen (eds). Symposium on Lexicography VII. [Proceedings of Seventh Symposium on Lexicography, May 1994, University of Copenhagen.] Tübingen: Niemeyer. 1-10. Sinclair, J. (1996d) Whose English? In J. Thomas (ed). The Birmingham Magazine. Birmingham: University of Birmingham Press. Pages ? Sinclair, J., Payne, J. & Hernández, C. P. (1996) Corpus to corpus: a study of translation equivalence. International Journal of Lexicography, 9, 3: 171-78. (= Sinclair, Payne & Hernández (eds) 1996.) Sinclair, J. (1996f) Multilingual databases: an international project in multilingual lexicography. International Journal of Lexicography, 9, 3: 179-96. (= Sinclair, Payne & Hernández (eds) 1996.) Sinclair, J. (1995a) From theory to practice. In G. Leech, G. Myers & J. Thomas (eds). Spoken English on Computer: Transcription, Mark-up, and Application. London: Longman. 99-109. Sinclair, J. (1995b) The world of woman in the bank of English: internal criteria for the classification of corpora. Journal of Literary and Linguistic Computing, 10, 2: 99110. Sinclair, J. (1995c) Computers and language teaching. In I. Lee et al (eds). Linguistics in the Morning Calm 3: Selected Papers from SICOL-1992. Seoul: Hanshin. 287-97. [Reprinted as 'Corpus evidence in language description', in A. Wichmann et al (eds) (1997) Teaching and Language Corpora. London: Longman. 27-39.] Sinclair, J. (1995d) Fictional worlds revisited. In E. Siciliani (ed). Le Trasformazioni del Narrare. Atti del XVI Convegno Nazionale, Ostuni 14-16 Ottobre 1993. Fasano di Brindisi: Schena Editore. 459-82. Sinclair, J. (1994) From scarcity to superfluity. In J. D'Arcy (ed). Proceedings of the Third Nordic Symposium for English Studies. Publisher?. 1-17. Sinclair, J. & Barnbrook G. (1994) Parsing Cobuild entries. In Sinclair, Hoelter & Peters (eds) 1994: 13-58.
7
Sinclair, J. & Francis, G. (1994) “I bet he drinks Carling Black Label”: a riposte to Owen on corpus grammar. Applied Linguistics, 15, 2: 190-200. Sinclair, J. (1993) Written discourse structure. In Sinclair, Hoey & Fox (eds) 1993: 6-31. [Reprinted in Sinclair 2004a: 82-101.] Sinclair, J. (1992a) The automatic analysis of corpora. In J. Svartvik (ed). Directions in Corpus Linguistics, Proceedings of Nobel Symposium 82, Stockholm 4-8 August 1991. Berlin: Mouton de Gruyter. 379-97. Sinclair, J. (1992b) Trust the text. In M. Davies & L. Ravelli (eds). Advances in Systemic Linguistics: Recent Theory and Practice. London: Pinter. 5-19. [Reprinted in R. M. Coulthard (ed). (1994) Advances in Written Text Analysis. London: Routledge. 12-26. And in Sinclair 2004a: 9-23.] Sinclair, J. (1992c) Priorities in discourse analysis. In R. M. Coulthard (ed). Advances in Spoken Discourse Analysis. London: Routledge. 79-88. Sinclair, J. (1992d) The corpora: lexicographers' needs. Zeitschrift für Anglistik und Amerikanistik, 41, 1: 5-14. Sinclair, J. (1992e) Corpus typology: a framework for classification. In G. Melchers & B. Warren (eds). Studies in Anglistics. Stockholm: Almqvist & Wiksell. 17-33. Sinclair, J. (1991b) Shared knowledge. In J. E. Alatis (ed). Paper presented at the Georgetown University Round Table on Languages and Linguistics 1991. Washington D.C.: Georgetown University Press. 489-500. Sinclair, J. & Renouf, A. (1991) Collocational frameworks in English. In K. Aijmer & B. Altenberg (eds). English Corpus Linguistics: Studies in Honour of Jan Svartvik. London: Longman. 128-43. Sinclair, J. (1990a) The nature of lexical statement. In A. Yoshimura et al (eds). Linguistic Fiesta: a Festschrift for Professor Hisao Kakehi’s 60th Birthday. Tokyo: Kurashio. 183-97. [Revised version as 'Words about words' in Sinclair 1991a: 123-37.] Sinclair, J. (1990b) Methods and madness. In V. Bickley (ed). Language Use, Language Teaching and the Curriculum. Hong Kong: Institute of Language in Education. 23140. Sinclair, J. (1990c) Teaching English: the decade ahead. In M. Abousenna (ed). Proceedings of the Tenth National Symposium on English Language Teaching in Egypt, CDELT. Cairo: Publisher? 18-24. Sinclair, J. (1990d) The Structure of Teacher Talk. Discourse Analysis Monographs 15. Birmingham: University of Birmingham Press. [Reprint of first part of Sinclair & Brazil 1982.]
8
Sinclair, J. & Kirby, D. (1990) Progress in computational lexicography. In L. Cignon & C. Peters (eds). Computational Lexicology and Lexicography. Vol. 7, special issue dedicated to B. Quemada. Pisa: Giardini. 233-57. [Reprinted in World Englishes, 1990, 9, 1: 21-36.] Sinclair, J. (1989a) Corpus creation. In C. Candlin & T. McNamara (eds) Language, Learning and Community. NCELTR. Macquarie University. 25-33. [Originally published as ‘Corpus creation’, paper presented to Council of Europe , February 1987. Revised version in Sinclair 1991a: 13-26.] Sinclair, J. (1989b) Uncommonly common words. In M. L. Tickoo (ed). Learners’ Dictionaries: State of the Art. Anthology Series 23. Singapore: RELC. 135-52. [Revised version as 'The meeting of lexis and grammar' in Sinclair 1991a: 81-98.] Sinclair, J. (1988a) Mirror for a text. Journal of English and Foreign Languages (Hyderabad), 1: 15-44. Sinclair, J. (1988b) Sense and structure in lexis. In J. D. Benson, M. Cummings & W. S. Greaves (eds). Linguistics in a Systemic Perspective. Amsterdam: Benjamins. 73-97. [Revised version in Sinclair 1991a: 53-65.] Sinclair, J. (1987a) Collocation: a progress report. In R. Steele & T. Threadgold (eds). Language Topics: Essays in Honour of Michael Halliday. Amsterdam: Benjamins. 319-31. [Revised version as 'Collocation' in Sinclair 1991: 109-121.] Sinclair, J. (1987b) Grammar in the dictionary. In Sinclair (ed) 1987: 104-15. Sinclair, J. (1987c) The nature of the evidence. In Sinclair (ed) 1987: 150-59. [Revised version as 'Words and phrases' in Sinclair 1991a: 67-79.] Sinclair, J. (1987d) The dictionary of the future. In J. A. Foley (ed). J. M. Sinclair on Lexis and Lexicography. Singapore: UniPress. 121-36. [Collins English Dictionary Annual Lecture, University of Strathclyde. Reprinted (1987) Library Review. Volume? And Focus on English. Volume?.] Sinclair, J. (1987e) Classroom discourse. Regional Language Centre, 18, 2: 1-14. Sinclair, J. (1987f) Language models and monuments. Britain Abroad, 5: 8-9. [Reprinted in English Today, 1988, 4, 15: 3-6.] Sinclair, J. (1987g) Tools of the trade. TESOL France Newsletter. Issue? Pages? Sinclair, J. & Renouf, A. (1987h) A lexical syllabus for language learning. In R. Carter & M. McCarthy (eds). Vocabulary and Language Teaching, London: Longman. 140-58 Sinclair, J. (1987i) Compressed English. In M. Ghadessy (ed). Varieties of Written English. London: Pinter. 130-36
9
Sinclair, J. (1986a) Fictional worlds. In R. M. Coulthard (ed). Talking about Text. Discourse Analysis Monograph No 13. English Language Research: University of Birmingham. 43-60. Sinclair, J. (1986b) First throw away your evidence. In G. Leitner (ed). The English Reference Grammar. Tübingen: Max Niemeyer Verlag. 56-65. [Revised version as 'Evaluating instances' in Sinclair 1991a: 99-108.] Sinclair, J. (1986c) Basic text processing. In G. Leech & C. Candlin (eds) Computers in English Language Teaching and Research. London: Longman. Pages? [Revised version in Sinclair 1991a: 27-36.] Sinclair, J. (1985a) On the integration of linguistic description. In T. A. Van Dijk (ed). Handbook of Discourse Analysis. Vol. 2. London: Academic Press. 13-28. [Reprinted in Sinclair 2004a: 67-81.] Sinclair, J. (1985b) Selected issues. In R. Quirk & H. G. Widdowson (eds). Language in the World. Cambridge: Cambridge University Press. 248-54. Sinclair, J. (1985c) Basic computer processing of long texts. In C. Candlin & G. Leech (eds). Computers and the English Language. London: Longman. 185-203. Sinclair, J. (1985d) Lexicographic evidence. In R. Ilson (ed). Dictionaries, Lexicography and Language Learning. ELT Documents 120, Pergamon. 81-94. [Revised version as 'The evidence of usage' in Sinclair 1991a: 37-51.] Sinclair, J. (1984a) Naturalness in language. In J. Aarts & W. Meijs (eds). Corpus Linguistics. Recent Developments in the Use of Computer Corpora in English Language Research. Amsterdam: Rodopi. 203-10. [Reprinted in English Language Research Journal, 1985, 2: 11-20. And in A Bilingual Journal of Language and Literature, 5, 11: 45-55.] Sinclair, J. (1984b) Lexicography as an academic subject. In R. K. K. Hartmann (ed). LEXeter 83 Proceedings, Lexicographica Series Maior No 2. Tübingen: Max Niemeyer Verlag. 3-12. Sinclair, J. (1984c) The teaching of oral communication. Nagoya Gakuin Daigaku Gaikokugo Kyoikukiyo, 10: 1-12. Sinclair, J. (1984d) Language awareness in six easy lessons. In B. G. Donmall (ed). Language Awareness, NCLE Papers and Reports 6, National Congress on Languages in Education, 4th Assembly. 33-6. Sinclair, J. (1984e) Poetic discourse: a sample exercise in style, structure and criticism. Indian Journal of Applied Linguistics. 10, 1-2: 9-27. [Reprinted in W. van Peer (ed) (1989) The Taming of the Text. London: Routledge. Pages?]
10
Sinclair, J., Donmall, B. G. & Tinkel, A. J. (1984) Evaluation and assessment. NCLE Papers and Reports 6, National Congress on Languages in Education, 4th Assembly. 71-95. Sinclair, J. (1983) Teacher training and project design. In B. Coffey (ed). Teacher Training and the Curriculum: An Investigation. The British Council, papers related to the Dunford House Seminar, 19-29 July 1982. Sinclair, J. (1982a) Linguistics and the teacher. In R. Carter (ed). Linguistics and the Teacher. London: Routledge. 16-30. Sinclair, J. (1982b) Reflections on computer corpora in English language research. In S. Johansson (ed). Computer Corpora in English Language Research. Bergen: Norwegian Computing Centre for the Humanities. 1-6. Sinclair, J. (1982c) Planes of discourse. In S. N. A. Rizvi (ed). The Two-Fold Voice: Essays in Honour of Ramesh Mohan. Salzburg: Institut für Anglistik und Amerikanistik, Universität Salzburg & New Delhi: Pitambar Publishing Co. 70-91. [Reprinted in Sinclair 2004a: 51-66).] Sinclair, J. (1981) The development of skills for learning. English for Specific Purposes, 47: 1-5. Sinclair, J. (1980a) Some implications of discourse analysis for ESP methodology. Applied Linguistics, 1, 3: 253-61. Sinclair, J. (1980b) Computational text analysis at the University of Birmingham. In S. Johansson (ed). Newsletter of the International Computer Archive of Modern English. Bergen: The Norwegian Computing Centre for the Humanities. 13-16. Sinclair, J. (1980c) Discourse in relation to language structure and semiotics. In S. Greenbaum, G. Leech & J. Svartvik (eds). Studies in English Linguistics for Randolph Quirk. London: Longman. 110-24. Sinclair, J. (1980d) Language for specific purposes. English Language Research Journal, 1: 3-13. Sinclair, J. (1979a) Issues in current ESP design and management. In S. Ziahosseiny & A. Mountford (eds). English for Special Purposes, papers from the 2nd Regional ESP Conference, Isfahan. 1-22. [Another version (1978) MALS Journal, 104-25.] Sinclair, J. (1979b) What purpose to English for special purposes? In R. Payne (ed). Papers on English for Special Purposes. Bourguiba Institute of Modern Languages Tunisia: Language and Linguistics Series. 39-58. Sinclair, J. & Coulthard, R. M. (1979) The system of analysis. In Supplementary Readings, Block 5, PE 232 Language Development, Open University. [Reprinted from Sinclair & Coulthard 1975.]
11
Sinclair, J. (1978) Language for specific purposes. In M. Linnarud & J. Svartvik (eds). Kommunikativ Kompetens och Fackspråk. Uppsala : Svenska föreningen för tillämpad språkvetenskap. 259-68. [Reprinted in D. Harper (ed) (1979) English for Specific Purposes: Papers from the Second Latin-American Conference, 11-17.] Sinclair, J. (1975a) Discourse in relation to language structure and semiotics. Paper read to Burg Wartenstein Symposium. [Reprinted in S. Greenbaum, G. Leech & J. Svartvik (eds) (1980) Studies in English Linguistics for Randolph Quirk. London: Longman. 110-24.] Sinclair, J. (1975b) The linguistic bases of style. In H. Ringbom (eds). Style and Text. Studies Presented to Nils Erik Enkvist. Stockholm: Skriptor, 1975. 75-89. Sinclair, J. (1974) English lexical collocations. Cahiers de Lexicologie. Paris: Institut des Professeurs de Français à l'Étranger. 15-61. Sinclair, J. (1973a) English for effect. Commonwealth Education Liaison Committee Newsletter, 3, 11: 5-7. [Reprinted in M. Stubbs & H. Hillier (eds) (1983) Readings on Language, Schools and Classrooms. London: Methuen. 238-45.] Sinclair, J. (1973b) Linguistics in colleges of education. Dudley Journal of Education, 3: 17-25. Sinclair, J. (1972b) Linguistic stylistics by candlelight. Nottingham Linguistic Circular, 11, 1: pages? Sinclair, J. (1972c) Lines about lines. In B. Kachru, B. Stahlke, F. W. Herbert (eds). Current Trends in Stylistics. Edmonton: Linguistic Research Inc. 251-61. [Reprinted in R. Carter (ed) (1982) Language and Literature. London: Allen & Unwin. 163-76.] Sinclair, J. (1971) The integration of language and literature in the English curriculum. Educational Review, 23, 3: 220-34. [Reprinted (1982) in R. Carter & D. Burton (eds). Literary Text and Language Study. London: Arnold. 9-27.] Sinclair, J. (1968a) A technique of stylistic description. Language and Style, 1: 215-42. Sinclair, J. (1968b) English language in English studies. Educational Review, 20: 82-94. Sinclair, J. (1968c) Linguistics and the teaching of English. In A. H. Marckwardt (ed). Language and Language Learning. Illinois (Champaign): NCTE, 31. 31-41. Sinclair, J. (1966a) Beginning the study of lexis. In C. E. Bazell, J. C. Catford, M. A. K. Halliday & R. H. Robins (eds). In Memory of J. R. Firth. London: Longman. 410-30. Sinclair, J. (1966b) Indescribable English. Unpublished inaugural lecture,. University of Birmingham. [Abstract in Sinclair & Coulthard 1975: 151.]
12
Sinclair, J. (1966c) Taking a poem to pieces. In R. Fowler (ed.). Essays on Style and Language. London: Routledge. 68-81. [Reprinted in D. M. Freeman (ed.) (1970) Linguistics and Literary Style. New York: Holt, Rinehart and Winston. 129-42.] Sinclair, J. (1965) When is a poem like a sunset? A Review of English Literature, 6, 2: 7691. [Reprinted in E.B. Lyle (ed.) (1976) Ballad Studies. Brewer/Rowman and Littlefield. 153-169.]
ACKNOWLEDGEMENTS I am grateful to two of my students, Simone Dausner and Julia Schmitt, and also to Susan Hunston and Michael Toolan, for help in tracking down some references. The website at http://www.worldcat.org/advancedsearch was a great help for details of books. If any readers can supply corrections or additions, this will be very welcome.
Applied Linguistics 30/1: 138–143 ß Oxford University Press 2009 doi:10.1093/applin/amp005 Advance Access published on 26 February 2009
FORUM
Ideology in Applied Linguistics for Language Teaching ALAN WATERS Lancaster University, UK It is contended that much of present-day applied linguistics for language teaching (ALLT) fails to mediate effectively, primarily because an ideological construction, emanating from a critical theory perspective, is too often imposed on everyday pedagogical practices. This has resulted in an exaggerated level of concern about the power imbalances that language teaching involves, leading to the promotion of approaches which attempt to subvert teaching and the teacher role, irrespective of the pedagogical consequences, while simultaneously proscribing others, which, from a practitioner perspective, are widely regarded as axiomatic. Such a stance has taken root, it is also argued, because of insufficient critical questioning of its moral and intellectual underpinnings. As a remedy, more should be done by ALLT to appreciate the rationale for and build on existing pedagogical traditions.
What can be termed ‘applied linguistics for language teaching’ (ALLT) has been characterised as ‘a mediating process which explores ways in which the concerns of linguistics as a discipline can be relevantly related to those of the language subject’ (Widdowson 2003: 13—my emphasis). When ALLT operates as such, it can be highly beneficial. However, as Cook and Seidlhofer (1995: 8) indicate, ‘[t]his is the ideal’, and it is argued in what follows that, unfortunately, much of present-day ALLT lacks the relevance necessary for carrying out its mediating role in an effective manner. This is seen to occur because a good deal of its discourse promotes or proscribes language teaching ideas on the basis of ideological belief rather than pedagogical value. The debate about ‘authenticity’ vs. ‘artificiality’ in language teaching is a representative example of this tendency. From the 1970s onwards, the view gained ground in ALLT that learners should experience not only artificially constructed texts, but also naturally-occurring ones. It was argued that this would increase their motivation, because, e.g. they would see the immediate relevance of what they were studying, be more confident in coping with real-life language use by developing strategies for dealing with its complexity, learn language as it is actually spoken and written, and so on (see, e.g. Wilkins 1976: 79). All of these reasons can be seen as having had pedagogical plausibility, and, importantly, were regarded by some
FORUM
139
(see, e.g. Widdowson 1979: 165) as grounds for supplementing rather than abandoning ‘artificial’ texts, in order to extend the range of learning opportunities available to learners. However, from the 1980s onwards, the status of ‘authenticity’ was elevated to that of a ‘moral imperative’ (Clarke 1989: 73). A typical expression of this attitude is the view that the findings of corpus-based descriptions of spoken English, such as those provided by the CANCODE project, should, of necessity, be incorporated into EFL teaching materials. Thus, as Carter and McCarthy (1996) put it: We know from our knowledge of our first language that in most textbook discourse we are getting something which is concocted for us, and may therefore rightly resent being disempowered by teachers or materials writers who, on apparently laudable ideological grounds, appear to know better. Information or knowledge about language should never be held back; the task is to make it available, without artificial restrictions, in ways which answer most learners’ needs. (p. 369) From this perspective, the use or otherwise of ‘real’ language in the classroom is seen in politicised terms, as a struggle between ideologically driven language-teaching practitioners and ‘disempowered’ learners. But in actual fact, of course, there might be very good pedagogical grounds why language knowledge should be ‘held back’, e.g. when it is too confusing or daunting for the learner to cope with, and such a policy may also be viewed as empowering the learner, by increasing the potential for learning. It is also not unreasonable, from a pedagogic perspective, for a teacher to be regarded by learners as indeed to ‘know better’ in such matters. Ironically enough, it is therefore actually Carter and McCarthy’s own point of view which is ideological. From a pedagogical angle, choice of teaching methodology is at root not so much (or should not be) a matter of political rights or wrongs, but, rather, an attempt to determine, at any given point in the learning process, what is likely to best enhance the learners’ opportunities for learning (cf. Hutchinson and Waters 1987: 158–160). This may involve, inter alia, the use of ‘authentic’ language, but, equally, it may well not. Thus, as Richards (2006) puts it: what is important in writing materials for EFL learners is not necessarily native speaker usage, but rather what will provide the means of successful communication both within and outside the classroom. This means providing learners with a repertoire of well selected vocabulary, sentence patterns and grammar, as well as a stock of communication strategies. . . . how native-speakers ask for and give directions is largely irrelevant. . . . my goal is to give them the resources to have successful experiences using English for simple classroom activities. Whether or not they employ native speaker-like language to do so is irrelevant. (p. 22)
140 FORUM
In other words, rather than because of ideological bias on the part of ‘materials writers’, lack of ‘real English’ in EFL textbooks can be seen to relate to the pedagogic advantages which can accrue, in terms of many aspects of classroom language learning, from the use of non-authentic material (cf. Waters 2009 (in press); Widdowson 2003: Ch. 9). This is not to deny, of course, that there are also occasions when the use of ‘authentic’ language can have an important pedagogic function. Nor is it to dispute the value of the role which ALLT played in earlier days in helping to make the option of exploiting not only artificial but also authentic texts an established pedagogical principle. The issue, rather, is the way in which, in much of the current ALLT discourse, inappropriate or impracticable teaching ideas, such as the wholesale use of ‘real language’, are being advocated for ideological reasons, and alternative perspectives, rooted in an awareness of everyday pedagogical practice, are viewed with suspicion. I have tried to show elsewhere (see Waters 2007a, b, 2008) that the primary driving force behind this trend has been the adoption, either directly or indirectly, in nearly all the main critiques by ALLT of language teaching pedagogy in recent years, of a ‘critical theory’ (CT) perspective (see, e.g. Phillipson 1992; Pennycook 1994; Canagarajah 1999; Holliday 2005; Edge 2006). From this point of view, asymmetry in language teaching structures is seen as resulting from the oppressive exercise of power. As a corollary, it is regarded as necessary to put in place alternative policies and procedures which will result in a more balanced distribution of power. Opposition to these innovations by language-teaching practitioners is viewed as political naivety, a form of ‘false consciousness’ (Holliday 2007). Thus, in terms of the issue just examined, the overweening concern in much of ALLT for maximizing ‘authenticity’ in language teaching can be seen as a CT-motivated attempt to restrict what is regarded as the unhealthy exercise of power by the textbook writer, by minimizing the possibility of prejudgement about how the learner might wish to view and use language data. Mutatis mutandis, the same underlying attitude can be seen to manifest itself in relation to the overselling in modern-day ALLT of many other academic ideas, for example, the anti-textbook stance (Kumaravadivelu 2006), the learnercentred approach and task-based learning (Nunan 1999), the proscription of cultural generalizations (Kubota 1999), the use of non-metropolitan language models as a basis for syllabus design (Matsuda 2006), and so on. However, in most language teaching situations, where practice is configured primarily in terms of pedagogical priorities, ideas of this kind tend to be viewed as having only limited relevance (Waters 2007a, b). This occurs because, in contradistinction to the typical practitioner perspective, CT-based ALLT, rather than properly acknowledging the centrality to language teaching pedagogy of language learning and the language learner role—since they are seen as liable to the exercise of hegemony via teaching and teachers—attempts instead to construct teaching and the teacher role as largely redundant, by, for example, giving pride of place to language use and the language user role, as in the
FORUM
141
excessive advocacy of authenticity (cf. Hutchinson and Waters 1987: 14; Ellis 2003: 251–4). In other words, in overall terms, the CT perspective can be viewed as aimed at suborning much or even all of what is at the heart of the pedagogical enterprise, since its everyday modus operandi are seen to conflict with the CT ‘Weltanschauung’. As a result, on the one hand, aspects of pedagogy such as fostering learner autonomy, which, though of importance, are, as Spratt et al. (2002) have shown, secondary to developing the motivation that stems from success in more basic aspects of learning, tend to be over-represented in terms of ALLT research and theorising, because they conform to the priority, from the CT perspective, of ‘liberating’ the learner from the teacher; and on the other, areas of pedagogy which are arguably more primary, such as the need to provide the level of classroom ‘structure’ that will imbue learners with the confidence to ‘stick their necks out’ (Stevick 1982: 7), are given a good deal less attention, because they conflict with the CT view of how interpersonal relations should be ordered. Why, however, despite these deficiencies, has the CT perspective nevertheless come to exercise such a strong influence on the discourse of ALLT? The answer would seem to be because, as Pinker (1998) argues with respect to a similar phenomenon within the social sciences in general, CT ‘not only has become an intellectual orthodoxy but has acquired a moral authority’ (p. 45). To take the second of these aspects first. As already noted, CT is concerned with championing what it sees as issues of social injustice. To challenge its views is therefore to be seen to oppose the moral values it espouses (cf. Widdowson 1998). Thus, as has been shown, restricting the use of authentic language is to be seen to ‘disempower’ learners; similarly, to criticise the concept of ‘learner-centredness’ is to run the risk of being stigmatised as authoritarian; to formulate cultural generalisations courts the danger of being accused of racism; to advocate the use of ‘standard English’ as a pedagogic model opens the door to the charge of native-speakerism; and so on. As a result, because of the discomfort associated with such ad hominem labelling, there has been a considerable lack of willingness to challenge the CT-based ‘line’, regardless of the consequences for pedagogy. Such a situation also contributes to the former aspect, the attainment of ‘intellectual orthodoxy’. Because of its perceived moral stature, CT, as it itself is wont to argue with respect to the habits of thought and practices it criticises (see, e.g. Fairclough 2001: 27), can be seen to have attained ideological status, i.e. to be regarded by many in academe as ‘common sense’, so customary and deeply ingrained that it is no longer noticed, but simply unconsciously taken for granted. As a result, it has become, ironically enough, a (reverse) form of ‘false consciousness’, i.e. an oppressive but largely unquestioned orthodoxy, a form of intellectual hegemony in its own right. In short, for the reasons I have tried to explain, the overall tenor of a good deal of the ALLT discourse in recent years can be seen as boiling down to not so much an attempt to mediate between linguistics and language teaching in
142 FORUM
an even-handed way, but, rather, the imposition of an ideologically biased view of how language teaching ought to be constituted, regardless of pedagogical relevance. How, then, might the problem be remedied? First, it is crucial for much more effort to be invested by ALLT in attempting, in the first instance, to understand and appreciate the pedagogical rationale for existing language teaching practices on their own terms, in the manner illustrated in some parts of the foregoing (cf. Brumfit 2001: Part Six), rather than reductively essentialising the motives of language-teaching practitioners so that they are automatically viewed in a negative light. Second, such a stance would, in turn, enable ALLT to be a good deal more heedful of the lessons of innovation theory, particularly the way in which, characteristically, ‘tradition guides the transition’ (Mensch 1979: Ch. 8; cf. Widdowson 1992). In other words, having identified the strengths and weaknesses of current trends in pedagogical practice, the strategy should not be to attempt wholesale replacement of them, on ideological (or any other) grounds, but, rather, in a manner similar to the approach adopted by ALLT in the earlier days of the authenticity debate, to identify how, by building creatively on their foundations, their potential for further development might occur. Only in such a way are we likely to increase the capability of ALLT, in the ideal manner, to mediate relevantly.
REFERENCES Brumfit, C. 2001. Individual Freedom in Language Teaching: Helping Learners to Develop a Dialect of Their Own. Oxford University Press. Canagarajah, A. S. 1999. Resisting Linguistic Imperialism in English Teaching. Oxford University Press. Carter, R. and M. McCarthy. 1996. ‘Correspondence,’ ELT Journal 50/4: 369–71. Clarke, D. 1989. ‘Communicative theory and its influence on materials production: state-ofthe-art article,’ Language Teaching 22/2: 73–86. Cook, G. and B. Seidlhofer. 1995. ‘An applied linguist in principle and practice’ in G. Cook and B. Seidlhofer (eds): Principle and Practice in Applied Linguistics: Studies in Honour of H.G. Widdowson. Oxford University Press, pp. 1–25. Edge, J. (ed.). 2006. (Re-)Locating TESOL in an Age of Empire. Palgrave Macmillan. Ellis, R. 2003. Task-based Language Learning and Teaching. Oxford University Press. Fairclough, N. 2001. Language and Power, 2nd edn. Longman. Holliday, A. 2005. The Struggle to Teach English as an International Language. Oxford University Press.
Holliday, A. 2007. ‘Response to ‘ELT and ‘‘the spirit of the times’’ ’,’ ELT Journal 61/4: 360–6. Hutchinson, T. and A. Waters. 1987. English for Specific Purposes: A Learning-centred Approach. Cambridge University Press. Kubota, R. 1999. ‘Japanese culture constructed by discourses: implications for applied linguistics research and ELT,’ TESOL Quarterly 33/1: 9–35. Kumaravadivelu, K. 2006. ‘Dangerous liaison: globalization, empire and TESOL’ in J. Edge (ed.): (Re-)Locating TESOL in an Age of Empire. Palgrave Macmillan, pp. 1–26. Matsuda, A. 2006. ‘Negotiating ELT assumptions in EIL classrooms’ in J. Edge (ed.): (Re-)Locating TESOL in an Age of Empire. Palgrave Macmillan, pp. 158–70. Mensch, G. 1979. Stalemate in Technology. Ballinger Publishing Company. Nunan, D. 1999. Second Language Teaching and Learning. Heinle and Heinle. Pennycook, A. 1994. The Cultural Politics of English as an International Language. Longman.
FORUM
Phillipson, R. 1992. Linguistic Imperialism. Oxford University Press. Pinker, S. 1998. How the Mind Works. Penguin. Richards, J. C. 2006. ‘Materials development and research-making the connection,’ RELC Journal 37/1: 5–26. Spratt, M., G. Humphreys, and V. Chan. 2002. ‘Autonomy and motivation: which comes first?,’ Language Teaching Research 6/3: 245–66. Stevick, E. W. 1982. Teaching and Learning Languages. Cambridge University Press. Waters, A. 2007a. ‘ELT and the spirit of the times,’ ELT Journal 61/4: 353–9. Waters, A. 2007b. ‘Native-speakerism in ELT: plus ca change . . . ?,’ System 35/3: 28–292. Waters, A. 2008. ‘Review of ‘‘(Re)-Locating TESOL in an Age of Empire’’,’ ELT Journal 62/1: 89–92.
143
Waters, A. 2009. ‘Advances in materials design’ in M. H. Long and C. J. Doughty (eds): Handbook of Language Teaching. Blackwell, in press. Widdowson, H. G. 1979. Explorations in Applied Linguistics. Oxford University Press. Widdowson, H. G. 1992. ‘Innovation in teacher development,’ Annual Review of Applied Linguistics 13: 260–75. Widdowson, H. G. 1998. ‘Review article: the theory and practice of critical discourse analysis,’ Applied Linguistics 19/1: 136–51. Widdowson, H. G. 2003. Defining Issues in English Language Teaching. Oxford University Press. Wilkins, D. A. 1976. Notional Syllabuses: A Taxonomy and its Relevance to Foreign Language Curriculum Development. Oxford University Press.
Applied Linguistics 30/1: 144–157
ß Oxford University Press 2009
REVIEWS Nikolas Coupland: STYLE: LANGUAGE VARIATION AND IDENTITY. Cambridge University Press, 2007. Coupland’s sociolinguistic perspective on style is both the source of its very stimulating insights as well as some of the reservations I have. Coupland, as a leading British sociolinguist, is eminently well-qualified to write this title on a concept central to his own research. Indeed, one fascination of the book is the way ontogeny recapitulates phylogeny as the cliche´ goes. The account of sociolinguistics’ origins in Labovian studies in variation segueing to more late modern studies of styling-in-performance can be seen to mirror Coupland’s own research itinerary, as his earlier uneasy post-Labovian studies of Cardiff English transform into more recent evaluations of media performances of a more reflexive ‘mythology of working-class Welshness’ (p. 162). The argument is for style as a key to studying people, rather than for variation as a branch of linguistics. However, ‘variation’ is still there in the title. Those from outside the sociolinguistic charmed circle will be a little surprised to learn, then, from the very subtitle of the book, that style is to be reduced to ‘variation’, and with particular reference to ‘identity’. Is style reducible to variation and identity? Variation is an increasingly problematic concept for many, implying as it does the contrastive notion of a ‘standard’ which Coupland himself has done so much to prove to us is an ideology rather than a set of empirical features that can be observed in everyday interaction. If, rather, variation is all we have in language use, the analytic potential of the concept seems seriously compromised. Similarly, for many, style in specific instantiations may or may not involve notions of identity, unless ‘identity’ is to become a rather meaningless catch all identifier, to the detriment of other motivations in human interaction. The traditional sociological identity categories of class, gender, or ethnicity, here also age, are both recognised to be inadequate, and yet insisted on as centrally relevant. The book frustrates a little by showing the limitations of such key sociolinguistic terms, even as it shows their value in new more sophisticated elaborations. The opening Preface carefully emphasises Labov’s own point that variation is observable within the individual speaker using language in different contexts as well as between individuals and groups, but the founding father and his epigones are subsequently criticised for losing the first insight. The effort of this book, then, is to rescue stylistics from the linguists, to show how ‘social reality is creatively styled’ (p. xii) by individual social agents. The stress is to be on style as discourse, a contextualised activity, rather than on forms, structures, and systems. Next, the Introduction goes on to argue the particular relevance of styling to late modernity with its lifestyle cultures and pervasive
REVIEWS 145
mediatisation. ‘Late modernity’ is preferred over ‘post-modernity’ because Coupland wants to hold on to a narrative of continuities and growing enlightenment, as well as the persisting relevance of more traditional structures and systems however radically they are apparently being interrogated in contemporary social practice. From Peirce, the notion of style as indexicality is invoked, and from Bakhtin, the terminology of ‘stylisation’. Both advance the idea of a necessarily interactional and contextualised dimension to actual acts of styling. Halliday, or later Bourdieu and others, are taken to task for being too deterministic for the more Giddens-eye-view of late modernity taken here. In the same cultural perspective, fixed or reified ‘authentic’ identities will be taken to task in the ‘anti-essentialist’ argument of later chapters. Chapter 2 is an intelligent, well-informed but critical account of Labov and related stratificational or structural ideas. Labov is particularly taken to task for his emphasis on variable ‘attention to language’. Coupland’s position is that speakers always attend to their language, and he rejects any straightforward calibration of ‘attention’ with a scale of standard to ‘natural’ vernacular features of spoken language. A related difficulty is that Labov still understands language and society as two discrete entities. Ideas of, for example, ‘covert prestige’ are of interest but still too context independent in Coupland’s estimation. More reciprocal models of ‘communities of practice’ are preferred to a society understood as structures: ‘Variation does not simply reflect a readymade social meaning; it is part of the means by which that meaning emerges’ (Eckert 2000: 43; quoted on p. 50). The detail of the chapter repays close consideration, but we hurry on in our grand narrative: ‘what was needed as a next stage’ (p. 53, sic) follows in Chapter 3, ‘Style for Audiences’. Here, Bell and Giles in particular are singled out for their more nuanced account of the meanings of stylistic choices as relative to contexts and interlocutor perception, even if ‘audience’ is still too limited a notion. Importantly, however, style is coming to be seen as a resource for interactional activity by creative agents. In this work, style demonstrably exceeds or is not reducible to traditional social identities, though they are still relevant. ‘The great achievement of audience design and accommodation research is to show the malleability of sociolinguistic identity’ (p. 76). Critically, though, Coupland wishes to continue to endorse the idea of variation as a primary given, even if now seen as a resource for, rather than determiner of, speech. Stylisation is then a secondary use or strategic deployment of variation with which speakers mean. Chapter 4, a kind of bridge between traditional sociolinguistics and the newer understandings proposed of ‘styling as an interactional practice’ (p. 105), is the weakest in my judgement. Nevertheless, the narrative is purposefully developed through characters like Bourdieu, Butler, and Bakhtin, to argue the need to recognise fully the significance of individual linguistic creativity. In the posited late modern setting: ‘social meanings for variation are clearly multi-dimensional,
146 REVIEWS
inherently variable and potentially unstable’ (p. 99), but not completely open choices. Chapters 5 and 6 are the most original and least conventional sociolinguistically. ‘Stylising Social Identities’, Chapter 5, argues from data for ‘identity as an active discursive process’ (p. 106). Here, Le Page and Tabouret-Keller’s ‘Acts of Identity’ (1985) are seen as prescient harbingers. Cardiff travel agent data are re-examined from this angle, and Rampton’s (1995) ideas of crossing also instantiated—‘social class and ethnic meanings are a resource available to be invoked and inflected in many different ways in relation to local interactional concerns’ (p. 144). Such work is valued for its potential to take back empirically originating issues of social theory to sociology as well as to undertheorised sociolinguistics. Chapter 6 then charts the outer reaches of late modernity with the acknowledgement that parody, meta-parody, rampant intertextuality, metalinguistic, and metacultural performances are the distinctive salient features of our ‘knowing’ mediatised age. Identities are pervasively ascribed, appropriated, invoked, contested, and resisted. The claim for style in such a cultural environment is that it holds up for our inspection and evaluation exaggerated images of the ‘way we are now’ as seen in examples from BBC Radio Wales or Nye Bevan speeches. Coupland attempts to describe rather than to celebrate or deplore, as a cultural studies orientation might, but it has to be said that some of his interpretations and arguments depend on insider perspectives, a variety of ‘linguistic ethnography’. ‘Coda’, Chapter 7, closes the book by arguing the value of ‘style’ as a central device for the study of social reality in all its late modern complexity, contingency, and pervasive heteroglossia (as Bakhtin would have had it). Questions of authenticity and of the relative decline of ideologies of standards and strategic privileging of vernaculars are suggested as examples of central areas of contemporary social enquiry benefiting from such studies. This is an enormously stimulating, succinct, and subtle yet fully readable state of the art study written by a leading sociolinguist at the top of his game. Every line witnesses an unusual level of understanding, relevant reading, and experience. The book may be over-ambitious in its claims for style, for identity, and for late modern sociolinguistics. It is also paradoxically too modest too. Style as contextually relevant micro-phonological variation and identity work is not the whole story. The narrative is suspiciously progressivist in any case. And the emphasis on the need for local and ethnographic understandings may risk a fall into the provincial and the inconsequential without ongoing reference back to larger theories. There are ‘no last words’, as Bakhtin taught us to know. This is a great first instalment, but there is more to be said. In fact, my final comment is that in what is ultimately a textbook, however advanced, more acknowledgement should have been made to the Russian pioneer of so much that is here. The index, in this respect as in others, is not of much help. But anyone interested in language use in the real world—the brief of applied
REVIEWS 147
linguists—should read this subtle, closely argued, and richly informed book, if only to dream about a more complete account. Reviewed by Geoff Hall Swansea University, UK doi:10.1093/applin/amp002
Advance Access published on 24 February 2009
REFERENCES Eckert, P. 2000. Linguistic Variation as Social Practice. Oxford: Blackwell. Le Page, R. B. and A. Tabouret-Keller. 1985. Acts of Identity: Creole-based Approaches to
Language and Ethnicity. Cambridge: Cambridge University Press. Rampton, B. 1995. Crossing: Language and Ethnicity among Adolescents. London: Longman.
Malcolm Coulthard and Alison Johnson: AN INTRODUCTION TO FORENSIC LINGUISTICS: LANGUAGE IN EVIDENCE. Routledge, 2007. In reviewing Coulthard’s An Introduction to Discourse Analysis for this journal back in 1980, Sankoff begins ‘I wouldn’t have thought it possible, in a slim little volume like this, to review most of the current major lines of work [in the field]’ (p. 179). She concludes ‘Coulthard succeeds in covering a remarkable amount of material. Even specialist readers will probably find something that is new to them. The book is coherent, well organised and successfully unites practical and theoretical concerns’ (p. 180). It is uncanny how well every detail of Sankoff’s comments apply too to Coulthard’s most recent publication, with Alison Johnson as co-author. An Introduction to Forensic Linguistics is likely to remain influential in a variety of ways long after its publication. It exposes an emergent area of academic endeavour exactly as its title suggests by considering language and legal processes (Part 1) and language as evidence (Part 2). Coulthard and Johnson’s backgrounds as academics, expert witnesses and, for Johnson, as a police officer, leave them well placed to write this insightful text. Following chronologically, and in some cases conceptually, from other recent books on part or all of the subject matter (e.g. Gibbons 2003; Shuy 2005, 2006; Solan and Tiersma 2005), An Introduction to Forensic Linguistics nonetheless fills a gap in the literature through its breadth. From its opening, with presentation of case summaries departing from media reports, to its close, with consideration of the effective practitioner, the book tightly focuses on issues which will stimulate and engage. This is balanced throughout by avoidance of sensationalism and a stated respect for the human lives discussed (e.g. p. 9). This book is rich in worked textual examples, leaving readers clear about how Coulthard and Johnson do their work and showcasing a range of other approaches. This is first evidenced in Chapter 1 which is based around
REVIEWS 147
linguists—should read this subtle, closely argued, and richly informed book, if only to dream about a more complete account. Reviewed by Geoff Hall Swansea University, UK doi:10.1093/applin/amp002
Advance Access published on 24 February 2009
REFERENCES Eckert, P. 2000. Linguistic Variation as Social Practice. Oxford: Blackwell. Le Page, R. B. and A. Tabouret-Keller. 1985. Acts of Identity: Creole-based Approaches to
Language and Ethnicity. Cambridge: Cambridge University Press. Rampton, B. 1995. Crossing: Language and Ethnicity among Adolescents. London: Longman.
Malcolm Coulthard and Alison Johnson: AN INTRODUCTION TO FORENSIC LINGUISTICS: LANGUAGE IN EVIDENCE. Routledge, 2007. In reviewing Coulthard’s An Introduction to Discourse Analysis for this journal back in 1980, Sankoff begins ‘I wouldn’t have thought it possible, in a slim little volume like this, to review most of the current major lines of work [in the field]’ (p. 179). She concludes ‘Coulthard succeeds in covering a remarkable amount of material. Even specialist readers will probably find something that is new to them. The book is coherent, well organised and successfully unites practical and theoretical concerns’ (p. 180). It is uncanny how well every detail of Sankoff’s comments apply too to Coulthard’s most recent publication, with Alison Johnson as co-author. An Introduction to Forensic Linguistics is likely to remain influential in a variety of ways long after its publication. It exposes an emergent area of academic endeavour exactly as its title suggests by considering language and legal processes (Part 1) and language as evidence (Part 2). Coulthard and Johnson’s backgrounds as academics, expert witnesses and, for Johnson, as a police officer, leave them well placed to write this insightful text. Following chronologically, and in some cases conceptually, from other recent books on part or all of the subject matter (e.g. Gibbons 2003; Shuy 2005, 2006; Solan and Tiersma 2005), An Introduction to Forensic Linguistics nonetheless fills a gap in the literature through its breadth. From its opening, with presentation of case summaries departing from media reports, to its close, with consideration of the effective practitioner, the book tightly focuses on issues which will stimulate and engage. This is balanced throughout by avoidance of sensationalism and a stated respect for the human lives discussed (e.g. p. 9). This book is rich in worked textual examples, leaving readers clear about how Coulthard and Johnson do their work and showcasing a range of other approaches. This is first evidenced in Chapter 1 which is based around
148 REVIEWS
a vignette contrasting the language of the former British Prime Minister, Tony Blair, with that of the Little Britain comedy caricature, Vicky Pollard, both as courtroom witnesses. Not an obvious comparison, it accessibly illustrates how quantitative and qualitative techniques provide ways into analysing language in legal settings. The vignette’s early inclusion takes readers straight to the action of both the courtroom and applied analysis. These particular texts illustrate how readers might successfully source data from the public domain. Further involvement is promoted through expansive and enticing recommended readings and research tasks which close each chapter and have obviously been selected with, respectively, accessibility and practicality in mind. The tasks are easily adapted to various teaching circumstances and indeed undergraduates and postgraduates should be able to develop them independently. Much student work will set sail from these pages. The strategy of involving readers appears to be paramount to the authors and is, for me, one of the book’s major strengths. Readers could hardly be more directly involved than when the book itself is charmingly transformed into a forensic task as Coulthard and Johnson invite them to establish who authored which chapter. Answers are provided on an associated web page (not live at the time of writing). The courteous, attentive reviewer should perhaps have responded by writing anonymously. The first chapter sets out the book’s stall and provides a succinct history of forensic linguistics extending as far as 1949, further than some might expect. Chapter 2 examines the language of law, considering style, register, and meaning, through reference to comprehension, comprehensibility, interpretation, and applicability. Like its predecessor, this chapter embraces exemplification and application; here, comparative corpus data replace transcripts. In Chapter 3, generic relationships, both conceptual and actual, are explored through usefully referenced commentary. The authors are not afraid to leave the forensic linguistic canon, invoking relevant literature on medical interactions. Where Chapter 3 uses police interviews to illustrate a linguistic phenomenon (genre), Chapter 4’s interview data exemplify a stage in the legal process (the early stage). Combining these with work on emergency calls and police–bomber negotiations presents established literature in new configurations. The view of the courtroom in Chapter 5 is contemporary and develops key themes, such as the influence of both narrative and question– answer forms. Yet the chapter, like its predecessors, is adventurous. In this case, wide-ranging data enable the authors to extend familiar points about friendly and opposing counsel by considering the influence of dialogue, collaboration, disruption, and fragmentation of narratives from the privileged position of the interview room. In Part 2, Chapter 6 describes cases which, whilst familiar to those already in the field (e.g. Coulthard 2005), will fascinate newcomers. Advanced readers might hunger for some theorising around application but the authors rightly show that this book is not its home. Chapter 7 covers the contrasting endeavours of forensic phoneticians and document examiners. Chapter 8
REVIEWS 149
outlines the notion of idiolect before justifying authorship investigations centred on that notion. Interesting history, stretching back 2000 years, accompanies recent scholarship on authorship attribution methods and possible markers. Some might say that plagiarism, not a criminal offence in most countries, is not ‘forensic linguistics’. However, Coulthard and Johnson’s ninth chapter takes such unacknowledged ‘borrowing’, deliberate or accidental, as its focus and convincingly justifies its presence through reference to copyright law, publishing, and ownership. The authors oriented to plagiarism as civil offence in the Introduction (p. 2) so this does not feel contrived. As in the previous chapter, a historical perspective precedes examples of recent scholarship including Coulthard’s own casework. The final chapter, on linguists as expert witnesses, carefully considers challenges of expressing expert opinions semantically and statistically. It is perhaps ironic, perhaps fortuitous that linguistics, a scholarly field so interested in the expression of professional concepts (e.g. Sarangi and Candlin 2003), should find itself reflecting critically about how it expresses its own expertise. Chapter 10 acknowledges that matters of admissibility can keep linguists out of court altogether but appropriately closes, undeterred, by illustrating how they might contribute to legal decisions nonetheless. The aim and construction of almost every chapter here is distinctive. This is illustrated by contrasting Chapters 1 and 2, for example, with their different data and methods, and by contrasting these data-driven sections with Chapters such as 4, 7, and 8, which orient to existing literature and historical contextualisation much more. Whilst such differences might be addling elsewhere, here they flag the diversity of forensic linguistic scholarship and provide complementary points of engagement. Coulthard and Johnson assume some background knowledge of legal systems or a willingness to read without. For example, standards of proof are not introduced until page 72 and then somewhat nonchalantly. Similarly, novice readers might overlook the selectivity of the predominant focus on adversarial legal systems which is rarely highlighted (e.g. p. 95; pp. 207–210; cf. Gibbons 2003: 5). Nonetheless, this lack of explicit pandering to readers who might be expected to look up somewhat tangential points, or notice rather obvious ones, is not problematic. The book is implicitly positioned as being about forensic discourse analysis (pp. 7–8). Certainly, discourse analytic techniques dominate its pages, yet it also considers other approaches (e.g. pp. 56–60) and matters which are not amenable to discourse analysis (e.g. pp. 144–156). Clearly Coulthard and Johnson faced tricky decisions about whether to confine themselves to reviewing and indeed using only discourse analytic methods. They have opted instead to present forensic linguistics as a broad church, entirely sensibly given their title. Nonetheless it would have been intriguing to see them tackling this mismatch directly. Similarly the notions of ‘tools’ and the linguistic ‘toolkit’ feature throughout (e.g. Chapter 1, p. 121) and more attention to the minutiae of selecting and even identifying tools would have helped
150 REVIEWS
newcomers. This is especially important as novice discourse analysts are encouraged to ‘try your hand at some forensic linguist problems’ (p. 159). This encouragement is confusing in the context of Chapter 7 where readers are warned off phonetic analysis due to insufficient equipment or knowledge. It would have been fascinating to see the authors explore the reasons for and implications of this contrasting advice although realistically beyond the scope of the book. As scholarly work in language, law, and evidence proliferates, we can expect that this will not be one of relatively few introductory books on ‘forensic linguistics’ for long. Nonetheless this thorough, engaging and imaginative work sets the bar high. This is a beautifully crafted textbook which appears at exactly the right time to make its discussion useful to linguists in classrooms as they present forensic linguistics to students, in seminar rooms as they discuss the sub-discipline with colleagues and in police stations and courts as they explain themselves to legal professionals. Reviewed by Frances Rock Cardiff University, UK doi:10.1093/applin/amp003
Advance Access published on 26 February 2009
REFERENCES Coulthard, M. 2005. ‘The linguist as expert witness,’ Linguistics and the Human Sciences 1/1: 39–58. Gibbons, J. 2003. Forensic Linguistics: An Introduction to Language in the Justice System. Oxford: Blackwell. Sankoff, G. 1980. ‘Malcolm Coulthard, An Introduction to Discourse Analysis, London: Longman, 1977,’ Applied Linguistics 1/2: 179–80. Sarangi, S. and C. Candlin. 2003. ‘Categorization and explanation of risk: a
discourse analytical perspective,’ Health, Risk and Society 5/2: 115–24. Shuy, R. 2005. Creating Language Crimes: How Law Enforcement Uses (and Misuses) Language. Oxford: Oxford University Press. Shuy, R. 2006. Linguistics in the Courtroom: A Practical Guide. Oxford: Oxford University Press. Solan, L. and P. Tiersma. 2005. Speaking of Crime: The Language of Criminal Justice. Chicago: University of Chicago Press.
D. Block: SECOND LANGUAGE IDENTITIES. Continuum, 2007. Block starts off by stating that the ‘book is about how identity is a key construct in different ways in different second language learning contexts’ (p. 1). The three chapters of Part I provide a background to studies on identity in the social sciences, and identity in early second language (L2) research, whereas in the three chapters of Part II, the author touches on three common contexts in which second language learning (SLL) takes place: adult migration, foreign language learning, and study abroad. In the last chapter, Block proposes directions for future research.
150 REVIEWS
newcomers. This is especially important as novice discourse analysts are encouraged to ‘try your hand at some forensic linguist problems’ (p. 159). This encouragement is confusing in the context of Chapter 7 where readers are warned off phonetic analysis due to insufficient equipment or knowledge. It would have been fascinating to see the authors explore the reasons for and implications of this contrasting advice although realistically beyond the scope of the book. As scholarly work in language, law, and evidence proliferates, we can expect that this will not be one of relatively few introductory books on ‘forensic linguistics’ for long. Nonetheless this thorough, engaging and imaginative work sets the bar high. This is a beautifully crafted textbook which appears at exactly the right time to make its discussion useful to linguists in classrooms as they present forensic linguistics to students, in seminar rooms as they discuss the sub-discipline with colleagues and in police stations and courts as they explain themselves to legal professionals. Reviewed by Frances Rock Cardiff University, UK doi:10.1093/applin/amp003
Advance Access published on 26 February 2009
REFERENCES Coulthard, M. 2005. ‘The linguist as expert witness,’ Linguistics and the Human Sciences 1/1: 39–58. Gibbons, J. 2003. Forensic Linguistics: An Introduction to Language in the Justice System. Oxford: Blackwell. Sankoff, G. 1980. ‘Malcolm Coulthard, An Introduction to Discourse Analysis, London: Longman, 1977,’ Applied Linguistics 1/2: 179–80. Sarangi, S. and C. Candlin. 2003. ‘Categorization and explanation of risk: a
discourse analytical perspective,’ Health, Risk and Society 5/2: 115–24. Shuy, R. 2005. Creating Language Crimes: How Law Enforcement Uses (and Misuses) Language. Oxford: Oxford University Press. Shuy, R. 2006. Linguistics in the Courtroom: A Practical Guide. Oxford: Oxford University Press. Solan, L. and P. Tiersma. 2005. Speaking of Crime: The Language of Criminal Justice. Chicago: University of Chicago Press.
D. Block: SECOND LANGUAGE IDENTITIES. Continuum, 2007. Block starts off by stating that the ‘book is about how identity is a key construct in different ways in different second language learning contexts’ (p. 1). The three chapters of Part I provide a background to studies on identity in the social sciences, and identity in early second language (L2) research, whereas in the three chapters of Part II, the author touches on three common contexts in which second language learning (SLL) takes place: adult migration, foreign language learning, and study abroad. In the last chapter, Block proposes directions for future research.
REVIEWS 151
The book begins with the various definitions of identity: (i) how individuals are defined, contained, and enabled by documents as opposed to the actions they take; (ii) identity as open-ended and unstable, which captures common themes in current discussions of identity, such as authenticity and the extent to which it is a self-conscious and ongoing project; and (iii) since the 1990s, identity as an academic concept. On page 4, the author declares that his analytical approach to the book is borrowed in part from Selinker (1992), and that he is deliberately ‘misreading’ the studies, without quite explaining though what that means. Block entitles Chapter 2 ‘Identity in the social sciences today’ and discusses identity as a construct for social scientists in the post-structuralist world. He notes that traditional determinants of identity such as biology gave way to a social structuralist approach in which individuals are determined by their membership of social categories based on social class, religion, education, family, peer groups, and so forth. The current, post-structuralist social science literature, however, includes ‘more nuanced, multileveled and complicated framings of the world around us’ instead of an attempt to locate ‘universal and invariant laws of humanity’ (p. 13). Terms such as performativity (e.g. Butler 1999), positioning/footing (e.g. Davies and Harre´ 1999; Goffman 1981), ambivalence and hybridity (e.g. Bhabha 1994; Mercer 1990), communities of practice (e.g. Lave and Wenger 1991), and power and recognition (e.g. Foucault 1981) have also begun to appear. Identity in early SLL research—the third chapter—begins with Lambert’s (1972) study of motivation and French–English bilingualism. Other prominent studies include Schumann’s Acculturation Model (1974), Krashen’s Input theory (1981), Long’s (1985), and Gass’s (1988) theories on interaction. All these point to two trends in second language learning research: that of identity becoming an object of interest in its own right, and identity being framed in less fixed and measurable terms, quite independently of success in second language learning. Chapter 4 begins the second part of Block’s book where he moves on to the different contexts in which language learning takes place. This chapter focuses on the adult migrant, and starts off with Broeder et al. (1996) on gate-keeping encounters in Western Europe. Block also touches upon several other narratives in the North American context, as well as his own interviews of Spanish-speaking Latinos in London. In the next chapter, Block turns to the foreign language context, and again presents narrative accounts of various studies, such as how pragmalinguistics, sociopragmatics, and identity work come together. He elaborates on intercultural language learning and the development of intercultural speakers in foreign language contexts; textual identity, language play, and identity of the learner; as well as interpersonal learners such as Silvia in his own research. Block also reviews articles on feminism, the English as a Foreign Language classroom in Japan, and internet-mediated foreign language learning and identity. At the end of the chapter, he concludes that the foreign language context differs markedly from the adult migrant context in terms of target language-mediated identity work.
152 REVIEWS
The last chapter in this section deals with the context of study abroad, and Block provides an overview of it through studies of sexual harassment in Russia, Costa Rica, and Spain. He states on page 157 that ‘sexual harassment can be a defining factor of the study abroad experience and one that in effect, places strong constraints on the prospects of developing new target languagemediated subject positions’. Other related narratives include gendered subject positions, teacher–student subject positions, and enhanced national identity. To Block, the context of study abroad is as varied as the different nationalities enrolled in such programs and the various countries which receive them, and therefore, there is scope for opening up related identity research, especially in contexts which are under-represented in the available literature. In the concluding chapter, Block appeals to five main considerations for future research: (i) a greater emphasis on social class as symbolic capital; (ii) a move away from the monolingual bias and the L1–L2 dichotomy to expanding the ambit of L1 to include people with two or more first languages and people who speak more than one dialect of the same L1; (iii) investigations into the emergence of local lingua francas, which are languages that are mutually accessible to individuals from other language backgrounds but are not the predominant language of the host community; (iv) electronically mediated second language learning experiences; and (v) the psychoanalytic perspective. Block offers a very readable overview of the identity issues surrounding second language learning, and a comprehensive review of the available literature. He refers to numerous narrative-based studies and quotes liberally from them. However, the general, descriptive tenor of the book may inadvertently disfavor a more critical analysis. For example, while Block describes identity as open-ended and unstable, suggesting that it is an ongoing process that is co-constructed and performed by the interactants, he does not seem to take that up very often in the analyses of his own data and those of other writers whose work he reviews. As Silverman (2006: 287) warns, interview studies should ‘satisfy the criterion of using low-inference descriptors’, such as ‘carefully transcribing [the tape-recordings of all face-to-face interviews] according to the needs of reliable analysis,’ and ‘presenting long extracts of data . . . including, at the very least, the question that provoked any answer’. Subtleties are lost when the details of talk-in-interaction are deleted and the interaction is reduced to a script, or even prose, and the reader is left to rely entirely on the accuracy of the author’s translations and account, rather than interpret the data for herself. Moreover, there appears to be an over-emphasis on narratives and interviews whereas other data types exist, such as the identity/identification documents alluded to by Block in his definition of identity earlier on. At the same time, greater care could be paid when methodologies are selected from an available repertoire. For instance, Atkinson and Delamont (2006: 167) note that ‘. . . narratives, and the accounting devices they enshrine, create the realities they purport to describe’. Therefore, such accounts should
REVIEWS 153
not be taken at face value, but ‘the research interview should be examined analytically as a performative act, through which identities are enacted, actions are justified and recounted events are retrospectively constructed’. Third, while Block explains why he chose these three specific contexts (pp. 4–5), and makes a final call to expand the L1 ambit, I believe that call should be a clarion one that refers specifically to the common context where someone from a multilingual community is learning an additional language, whether at home or abroad. This is a ubiquitous yet complex scenario throughout the world, and certainly deserves more attention than merely a summary in the concluding chapter. On the whole though, Block’s book offers a general introduction to the study of second language identities that is highly comprehensible, and provides even novices easy access to the subject. The book would, however, benefit from a sharper focus on the analytical framework, especially when critically examining and evaluating the studies which are reviewed, and thus avoid the pitfalls suggested by some of the authors cited here. Reviewed by Samantha Ng University of Hawai‘i at Ma¯noa doi:10.1093/applin/amp004 Advance Access published on 24 February 2009
REFERENCES Atkinson, P. and S. Delamont. 2006. ‘Rescuing narrative from qualitative research,’ Narrative Inquiry 16/1: 164–72. Bhabha, H. 1994. The Location of Culture. Routledge. Broeder, P., K. Bremer, C. Roberts, M. T. Vasseur, and M. Simonot. 1996. Achieving Understanding: Discourse in Intercultural Encounters. Longman. Butler, J. 1999. Gender Trouble: Feminism and the Subversion of Identity (10th Anniversary Edn). Routledge. Davies, B. and R. Harre´. 1999. ‘Positioning and personhood’ in R. Harre´ and L. van Langenhove (eds): Positioning Theory. Sage, pp. 32–52. Foucault, M. 1981. The History of Sexuality. Vol. 1: An Introduction. Pelican. Gass, S. 1988. ‘Integrating research areas: a framework for second language studies,’ Applied Linguistics 9/2: 198–217. Goffman, E. 1981. Forms of Talk. Blackwell. Krashen, S. 1981. Second Language Acquisition and Second Language Learning. Pergamon. Lambert, W. 1972. ‘A social psychology of bilingualism’ in A. S. Dil (ed.): Language, Psychology,
and Culture. Essays by Wallace E. Lambert. Stanford University Press, pp. 212–35. Originally published in 1967 in The Journal of Social Issues, 23/2: 91–109. Lave, J. and E. Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge University Press. Long, M. 1985. ‘Input and second-language acquisition theory’ in S. Gass and C. Madden (eds): Input in Second Language Acquisition. Newbury House, pp. 377–93. Mercer, K. 1990. ‘Welcome to the jungle: identity and diversity in postmodern politics’ in J. Rutherford (ed.): Identity: Culture, Community, Difference. Lawrence and Wishart, pp. 43–71. Schumann, J. 1974. ‘The implications of interlanguage, pidginization and creolization for the study of adult second language acquisition,’ TESOL Quarterly 8/1: 145–52. Selinker, L. 1992. Rediscovering Interlanguage. Longman. Silverman, D. 2006. Interpreting Qualitative Data. Sage.
154 REVIEWS
Alastair McLauchlan: THE NEGATIVE L2 CLIMATE: UNDERSTANDING ATTRITION AMONG SECOND LANGUAGE STUDENTS. Sasakawa Fellowship Fund for Japanese Language Education, 2007. The United Kingdom is a multilingual country. Indigenous languages and generations of migration have built up a complex pattern of language use. Yet the public perception, shaped and reflected by the media, is of a monolingual, monocultural society whose standardised English, universally spoken across the United Kingdom, has proved so attractive to outsiders across the world that Britons have no need to explore other languages. In such a context, it is an uphill task to motivate secondary school students to continue studying a foreign language, once it is no longer compulsory. A generation of quantitative and qualitative studies (reviewed in Coleman et al. 2007, cf. Macaro 2008) have shown that secondary school pupils generally regard the learning of foreign languages as difficult, boring, and—perhaps surprisingly for a nation which travels so extensively—of little practical use. New Zealand faces a very similar scenario. And despite its generic title and subtitle, it is actually in the context of New Zealand that this book explores language learner motivation in secondary education. Christchurch is New Zealand’s second city and in most respects representative of the country’s entire school population. McLauchlan’s study traces a cohort of pupils in all Greater Christchurch’s schools across the three final years of secondary schooling. The study begins in year 11. At this point, students may choose to take an examination in one or more foreign languages. This opt-in may be followed by two further years of school study, or by an opt-out at the end of year 11 or of year 12. The survey is set in its national socio-political context, and the author makes comparisons with the situation in the United Kingdom and Australia. The study focuses on six languages: French, German, and Japanese are traditional target languages now in decline; Spanish and Chinese are growing from a low base; Russian has gone, Samoan is too recent to figure, and McLauchlan also includes Latin, although the profile of learners is distinctive and some overall results are skewed as a consequence. Recent trends have seen not only a fall in the numbers initially opting for language study, but also drop-out rates higher than for other disciplines: only one in three students completes the 3 years of language study which would prepare them for university language study. The proportion of girls to boys is about 2:1 in upper classes, where groups are often small enough to attract adverse comment from other teachers. ‘Dedicated but despairing’ language teachers suggest that high flyers are opting for vocationally relevant subjects, while other youngsters see languages as difficult and boring, with little relevance, and are easily persuaded to drop languages by teachers of
REVIEWS 155
other disciplines, by timetabling, or by the attraction of novel and apparently less demanding subjects. McLauchlan’s longitudinal questionnaire study explored, in 2004, what motivated the year 11 students (N = 765) to choose to take an NCEA level 1 examination in a foreign language. In 2005 and in 2006, a follow-up questionnaire investigated the reasons of all those who, while not opting to leave school, had dropped out of language study. Motivation is thus operationalised as persistence, or rather a lack of it, and the ‘attrition’ of the title is to be understood as discontinuation of formal study in a school context. The study does not build on the extensive literature on language-learning motivation as a construct, nor does it draw directly on existing research instruments. McLauchlan’s very practical concern is to understand the reasons which lead students into the language classroom in year 11, and why these reasons are insufficient to keep them there. The first questionnaire established that the initial choice was the student’s own, unconstrained, though perhaps influenced, by school or parental pressure. Among the four top reasons for LOTEM study (Language Other Than English or Maori) given by the cohort, ‘language study is important’ and ‘interest in the country’ outweighed career and travel, though less markedly so for Chinese and Japanese, and among girls. At this point, one in four had already decided to go no further with languages, but most intended to persist at least to year 13, and 26% of girls and 32% of boys planned to study a language at university. McLauchlan shares with his New Zealand colleagues a feeling that languages are ‘academically more difficult’ than other subjects, so he asks students whether they find languages ‘difficult’. The result—a normal distribution either side of the fence—is unsurprisingly unenlightening, given the lack of comparison with other subjects. An open question asking for students’ ‘biggest problem’ brought equally predictable results—grammar, vocabulary, time, speaking, and (for Japanese and Chinese) kanji. What students liked most was talking to visitors, the culture, and the process itself: few mentioned ever travelling to the country in question. At this early stage, two out of three students are pleased to be doing a language, pleased with their progress, and determined to work hard. McLauchlan notes correlations among enjoyment, intentions, gender, and school type. By 2005, 41 students had left and 306 dropped languages; of the latter group, which had the same gender distribution as the initial sample, 264 completed the follow-up questionnaire. At around 40%, Christchurch had a lower drop-out rate than the rest of New Zealand. There was considerable variation across schools, though no link with target language or gender. The principal reasons given for discontinuation were difficulty, loss of interest and timetable difficulties—this last apparently fictitious since there were no actual clashes. Few claimed not to have enjoyed the classes. A year
156 REVIEWS
later, 62 more had left, and 128 discontinued (91 returned a questionnaire). Loss of interest now supersedes perceived difficulty as a reason for dropping out, but otherwise responses followed a similar pattern. Even those who abandoned their languages remained pleased they had once studied them. Perhaps, the principal finding is how little reliance can be placed on students’ intentions: two-thirds planned to stick it out at least to the end of school; two-thirds failed to do so. The study is methodical, meticulous, and highly informative, but the book is open to a number of criticisms. There is a wealth of tables giving percentages, but very few more sophisticated statistical analyses, and the study eschews many features of a standard research report, such as the separation of results and discussion. At times the presentation is too detailed, and it is hard to see the wood for the trees. At other times, the passionate commitment to language learning shades analysis into partial commentary, coloured by a tone of exasperation and disenchantment. I feel the book would have benefited from editing to firm up the structure, maintain strict objectivity, and provide executive summaries. McLauchlan can no more find answers to the ‘negative L2 climate’ than his counterparts in other Anglophone countries. The suggestions he voices include: varying learning outcomes by ability, a different approach to vocabulary learning, as well as seeking to enrol allies such as universities to influence government policy and change the national attitude to foreign languages. McLauchlan speculates that, as in the United Kingdom, languages may be turning back into an elite subject, a middle class accomplishment. Compulsory study, abandoned in the public sector because of student resentment or ‘negative backlash’, is now limited to private schools. In criticising the pervasive climate, McLauchlan does not single out the media for criticism, although in the United Kingdom it is now well established (e.g. Ensslin and Johnson 2006, Gieve and Norton 2007, Jaworski 2007) that the erasure or problematising of multilingualism and of inter-language issues by the British print and broadcast media is systematic, and has the effect of depressing take-up of language study in schools. The increasing focus on involving parents in new language-teaching initiatives at primary and secondary level reflects British teachers’ recognition of the need to address negative perceptions, so that individual pupils are not disadvantaged, even if wider society cannot be influenced out of its ethnocentrism. Whether a change of attitude can ever be achieved or not, most linguists will share the sentiments expressed in the phrase which McLauchlan has chosen as epigram: He taonga nga reo katoa . . . all languages are to be treasured. Reviewed by Jim Coleman The Open University, UK doi:10.1093/applin/amp006
Advance Access published on 24 February 2009
REVIEWS 157
REFERENCES Coleman, J. A., A. Galaczi, and L. Astruc. 2007. ‘Motivation of UK school pupils towards foreign languages: a large-scale survey at Key Stage 3,’ Language Learning Journal 35/2: 245–80. Ensslin, A. and S. Johnson. 2006. ‘Language in the news: investigating representations of ‘Englishness’ using WordSmith Tools,’ Corpora 1/2: 153–85. Gieve, S. and J. Norton. 2007. ‘Dealing with linguistic difference in encounters with others
on British television’ in S. Johnson and A. Ensslin (eds): Language and the Media. Continuum, pp. 188–210. Jaworski, A. 2007. ‘Language in the media: authenticity and othering’ in S. Johnson and A. Ensslin (eds): Language and the Media. Continuum, pp. 271–80. Macaro, E. 2008. ‘The decline in language learning in England: getting the facts right and getting real,’ Language Learning Journal 36/1: 101–8.
Applied Linguistics 30/1: 158 doi:10.1093/applin/amp001
ß Oxford University Press 2009
ERRATUM In the Notes on Contributors section of Issue 29/4, biographical information on Jasone Cenoz was included when the work of this author did not appear in this issue. Information on Carmen Mun˜oz, whose paper did appear in this issue, was not included. Information on this author is therefore given below: Carmen Mun˜oz received her MA in Applied Linguistics from the University of Reading, UK and PhD in English Linguistics from the University of Barcelona, Spain, where she is now a Professor of English Linguistics and Applied Linguistics at the University of Barcelona. Her research interests include second and foreign language acquisition, as well as bilingual acquisition. She is the coordinator of the Barcelona Age Factor (BAF) Project. Among her more recent publications are the edited volume Age and the Rate of Foreign Language Learning, Multilingual Matters (2006), and the chapter ‘Age-related differences and second language learning practice’ in the volume edited by R. DeKeyser. 2007. Practice in a Second Language. Perspectives from Applied Linguistics and Cognitive Psychology. Cambridge University Press. <[email protected]>
Applied Linguistics 30/1: 159–161 doi:10.1093/applin/amp013
ß Oxford University Press 2009
NOTES ON CONTRIBUTORS Igor Boguslavsky is an associate professor at Universidad Politecnica de Madrid (Spain), Director of the Russian Language Centre of the UNL Program which is devoted to the multilingualism support on the internet, and one of the main authors of the ETAP machine translation system developed in the Russian Academy of Sciences. His major research interests include syntax and semantics of natural languages, lexicography, computational linguistics, interlingua design, and machine translation. He has more than 100 publications on various aspects of linguistics and computational linguistics. Address for correspondence: Validation and Business Applications Research Group (VBA), Facultad de Informa´tica, Universidad Polite´cnica de Madrid Campus de Montegancedo, 28660 Madrid, Spain. Cade Bushnell is a PhD candidate in Japanese Linguistics at the Department of East Asian Languages and Literatures, University of Hawai‘i at Ma¯noa. He earned his MA in Japanese Pedagogy and Language Studies from the University of Tsukuba, Japan. His research interests include conversation and membership categorization analysis of second language data, second language acquisition (especially from situated learning and sociocultural theoretical perspectives), foreign/second language pedagogy, and Japanese linguistics. Cade is a recipient of several awards, including the Monbukagakusho Scholarship, and the Japan Foundation Doctoral Fellowship for 2008–9. Address for correspondence: Department of East Asian Languages and Literatures, University of Hawai‘i at Ma¯noa, 1890 East-West Road, Moore Hall Room 382, Honolulu, Hawaii 96822. Jesu´s Carden˜osa is an associate professor at Universidad Polite´cnica de Madrid (Spain) and currently head of the Validation and Business Applications research Group, and since 1997 head of the Spanish Language Centre of the UNL Programme for the support of multilingualism in Internet. He has been the project leader in more than 30 national and international research projects from the European Commission and the United Nations. He has established international agreements with more than 25 countries. His main research lines include validation of intelligent systems, intelligent information retrieval, information extraction, text mining, interlinguas, ontology learning, and multilingual lexicons and dictionaries. He has published more than 100 papers in international journals and conferences. Address for correspondence: Validation and Business Applications Research Group (VBA), Facultad de Informa´tica, Universidad Polite´cnica de Madrid, Campus de Montegancedo, 28660 Madrid, Spain. Asta Cekaite is a postdoctoral fellow at the Department of Child Studies, Linko¨ping University, Sweden. Her main research focuses on second language
160 NOTES ON CONTRIBUTORS
socialization, notably early L2 conversations, among beginner learners in an immersion classroom. A wider field of interests includes situated microapproaches to discourse in educational and informal family settings. Address for correspondence: Department of Child Studies, Linko¨ping University, Sweden. Jim Coleman is a professor of language learning and teaching at the Open University, UK. His research encompasses many aspects of language learning, especially in higher education, including policy, pedagogy, new technologies, affect, and study abroad. Address for correspondence: Professor of Language Learning and Teaching Department of Languages, The Open University, Stuart Hall Building, Walton Hall, Milton Keynes MK7 6AA, UK. <[email protected]> Carolina Gallardo is an associate professor at Universidad Polite´cnica de Madrid and member of the Validation and Business Applications Research Group. She has participated as a research member in several international and national projects, including the UNL Programme of the United Nations where she has led the coordination of the linguistic works. Her main research lines include machine translation based on interlinguas, multilingual dictionaries and information extraction. She has participated in top conferences in computational linguistics and artificial intelligences like CICLING, MICAI, or LREC. Address for correspondence: Validation and Business Applications Research Group (VBA), Escuela Universitaria de Informa´tica, Universidad Polite´cnica de Madrid, Carretera de Valencia, km 7, 28031 Madrid, Spain. Geoff Hall is a senior lecturer and head of Applied Linguistics at Swansea University, and an Editor of Language & Literature (Sage). Research interests in literary stylistics, discourse analysis and TESOL. Publications include Literature in Second Language Education (Palgrave, Macmillan, 2005). Address for correspondence: Applied Linguistics, School of Arts, Swansea University, Singleton Park, Swansea SA2 8PP, UK. Jieun Lee earned her master’s degree at the Graduate School of Interpretation and Translation of Hankuk University in Seoul, Korea in 1995 and received her second master’s degree in Applied Linguistics at Macquarie University in 2005. She is currently a PhD candidate in Linguistics at Macquarie University. She has extensive experience in interpreting and translation and teaches in the Program of Master of Translation and Interpreting of Macquarie University. Address for correspondence: Department of Linguistics, Macquarie University, North Ryde NSW 2109, Australia. <[email protected]> Samantha Ng is an MA student in the Department of Second Language Studies at the University of Hawai‘i at Ma¯noa. Her research interests include second language acquisition and sociolinguistics, especially the use(s) and users of
NOTES ON CONTRIBUTORS
161
English as an international language. She is currently working on her capstone paper provisionally entitled ‘Providing equal language access to the LEP person: At what costs?’. Address for correspondence: 1711 East-West Road #927, Honolulu, HI 96848-1711, USA. <[email protected]> Frances Rock is a lecturer in Language and Communication at Cardiff University. Her research interests are in discourse analysis, interactional sociolinguistics, and new literacy studies and she is currently working on applications of language study to policing. Address for correspondence: Centre for Language and Communication, Cardiff University, ENCAP, Humanities Building, Colum Drive, Cardiff, CF10 3EU, UK. Michael Stubbs has been professor of English Linguistics, University of Trier, Germany, since 1990. He was previously professor of English in Education, Institute of Education, University of London (1985–90) and Lecturer in Linguistics, University of Nottingham, UK (1974–85). He was Chair of BAAL (1988–91). His most recent books are: Text, Discourse and Corpora (with M. Hoey, M. Mahlberg and W. Teubert, 2007 Continuum), Words and Phrases (2001, Blackwell), and Text and Corpus Analysis (1996, Blackwell). His main areas of teaching and research are applied linguistics, corpus linguistics, semantics and pragmatics, and stylistics. Address for correspondence: FB2 Anglistik, Universita¨t Trier, 54286 Trier, Germany. <[email protected]> Masahiro Takimoto is an associate professor in the Department of English Language and Cultural Studies at the Faculty of Humanities at Tezukayama University, Nara, Japan. His research interests are in the roles of instruction in learning second language (L2) pragmatics and the relationship between the development of L2 grammatical competence and L2 pragmatic competence. Address for correspondence: Tezukayama University, 7–1–1 Tezukayama, Nara 631–8501, Japan. Alan Waters is a senior lecturer in the Department of Linguistics and English Language at Lancaster University. He has taught EFL and has trained teachers in the UK and several other parts of the world, as well as published a number of books and articles on a range of ELT topics. His main current research interests are in the areas of teacher learning and the management of innovation in language education. Address for correspondence: Department of Linguistics and English Language, Lancaster University, Lancaster LA1 4YT, UK.
APPLIN-30(1)Cover.qxd
3/10/09
11:47 AM
Page 2
APPLIN-30(1)Cover.qxd
3/10/09
11:47 AM
Page 1
APPLI ED LI NGUISTICS
Applied Linguistics
Volume 30 Number 1 March 2009
Volume 30 Number 1 March 2009
ISSN 0142-6001
Published in cooperation with AAAL American Association for Applied Linguistics AILA International Association of Applied Linguistics BAAL British Association for Applied Linguistics
OXFORD
www.applij.oxfordjournals.org
