Download - Search

Innovation in English Language Teaching‘This volumc brings to the fore divcrsc, fundamcntal issucs about thc processes andpolitics of curriculum change and improvement, new technologies, and concepts oflanguage use, communication, and instruction vital to guiding the organization andpracticcs of teaching English internationally.’ Alister Cumming, Ontario lnstitritqfbr Studiesin Education, Ilniversiq, $TorontoTeaching English Language WorldwideA sclcction of readers’ commmts on the series:‘This thrcc-part series offkrs a map to ELT research and practice . . . it represents the bestthat EI.T, as an Anglo-Saxon institution, has dcvclopctl over thc last thirty years for the teachingof English around the \vorld . . . Readers will fnd in this scrics the Who’s Who guide to thisdynamic antl expanding community.’ Claire Kramsch, Unitwsiy !/‘Cul!/brnia, Berkele)., CulIfbrnia‘Experienced knglish language instructors sccking to tlecpen their kno\vlctlgc ant1 abilitieswill find this series forms a coherent basis to d lop their understanding of current trcntls,sociocultural diversity, and topical interests in teaching English as a second or foreign languagearound the \vorld. All thrcc \olumcs pro\ idc ample flexibility for discussion, interpretation,antl adaptation in local scttings.’ ,41ister Ctirnrning, Ontario Institute for Sttidies in Educarion,Universiy ?ffToronto‘This scrics pro\ ides a collection of essential readings \vhich will not onl! pro\ itlc theTEFI./TESOL student anti tcachcr \\ ith access to the most up-to-tlatc thinking and approachesto the subject but mill give any person interested in the suhjcct an over\ ic\z of the phenomenonof thc usc antl usage of English in the modern nurltl. Perhaps morc importantly, this series\vi11 be crucial to thosc studcnts \z ho do not haw available to them articles that providc botha \vide spectrum of information antl the neccssary analytical tools to investigate the languageturthcr.’Josepb,l. E’olej; Soiitbeu. iu Jlinirters oftducation Orpnisution, Regionul Lungiiule Centre,Singapore‘The strong rcprcscntation of the seminal Anglo- Australian tlc\clopmcnt of the Europeanfunctional tradition in the study of languagc antl language education makes this a rclrcshingl!bracing scrics, \z hich should hc \vitlcl! used in tcachcr education tor English languagcteaching.’ Liicin Reid, /nytirrite fEducution, IJniversiy of/.ondon‘In a principled antl accessible manner, thcsc thrcc 1 olumcs living together major bvritings onessential topics in the stud? of English languagc tcaching. They provide broatl coverage ofcurrent thinking and debate on major issucs, providing an in\ aluable resource for thecontcmporarq postgraduate student.’ Guy Cook, llnii crsit?, of Reading

Teaching English Language WorldwideCompanion volumesThe companion volumes in this series are:Analysing English Language ina Global Context, edited by Anne Burns and Caroline CoffinEnglish Language Teaching in its Social Contextedited by Christopher N. Candlin and NeilMercerThese three readers are part of a scheme of study jointly developed by Macquarie University,Sydney, Australia, and the open University, United I

Innovation in English Language TeachingInnovation in English Language Teaching provides both theoretical perspectives andpractical tools for analysing, developing and evaluating English language teaching curricula.It presents English language teaching in a variety of specific institutional, geographic andcultural contexts.This Reader focuses particularly on curriculum change in context. The articles - whichinclude both classic and specially commissioned pieces - have been selected and edited tohighlight the debates, discussions and current issues from different parts of the Englishspealcingand English-using world.Academics and teachers from around the world examine the role and influence not just oflanguage teachers and students, but of parents, teacher-trainers, the local community, thepress, politicians, and all who have an interest in what goes on in the language classroom.Issues are illustrated and discussed in different contexts, including: teaching migrants inEnglish speaking countries; teaching large classes in developing countries; teaching Englishfor academic purposes; using information technology in the classroom.Articles by: Michael P. Breen; I

Innovation in EnglishLanguage TeachingA ReadevEdited byDavid R. Hall and Ann Hewings2TheopenUniversityLondon and New Yorkin association with Macquarie Universityand The Open UniversityMACQUARI EUNIVERSITY - SYDNEY

FOV my parents, Ron and Anne HallSimultaneously published in the USA and Canadaby Routledge29 West 35th Street, New York, NY 10001Routledge is an imprint of the Taylor & Francis Group0 2001 Compilation, original and editorial material Macquarie University and The OpenUniversity; individual articles 0 their authorsTypeset in Perpetua and Bell Gothic by I

ContentsList of illustrationsAcknowledgementsXxiiDavid R. Hall and Ann HewingsINTRODUCTION1PART ONEDirections in curriculum changeMichael P. Breen and Christopher N. Candlin1 THE ESSENTIALS OF A COMMUNICATIVE CURRICULUMIN LANGUAGE TEACHING9David Nunan and Clarice Lamb2 MANAGING THE LEARNING PROCESS27Michael Lewis3 LEXlS IN THE SYLLABUSMichael McCarthy and Ronald Carter4 DESIGNING THE DISCOURSE SYLLABUS4655Guy Cook5 THE USES OF COMPUTERIZED LANGUAGE CORPORAAREPLY TO RONALD CARTER64Ann Hewings and Martin Hewings6 APPROACHES TO THE STUDY OF DISCIPLINARY VARIATIONIN ACADEMIC WRITING: IMPLICATIONS FOR SYLLABUSDESIGN 71

V i i iCONTENTSPART TWOPolitical and institutional constraints in curriculumdevelopmentRonald Carter7 POLITICS AND I

CONTENTSixSusan Feez17 CURRICULUM EVOLUTION IN THE AUSTRALIANADULT MIGRANT ENGLISH PROGRAMDavid R. Hall18 MATERIALS PRODUCTION: THEORY AND PRACTICESimon Sergeant19 CALL INNOVATION IN THE ELT CURRICULUM208229240PART FOUREvaluating curriculum changePauline Rea-Dickins and I

IllustrationsFigures1.12.12.22.38.110.111.111.211.312.112.214.114.214.314.415.116.117.117.217.317.418.118.219.1The curriculumThree phases or perspectives on the curriculum processThree alternative ways of grouping learnersPlanning grid for general English courseBICS and CALPAn S-shaped diffusion curveIndividualization in large classesWorlcsheet 1: radio newsWorksheet 2: self-created clozeThe initial language program frameworkThe revised language program frameworkCultural continuityProfessionally constructed image of ‘the learner’Professionally constructed image of ‘the stakeholder’‘Us’ - ‘them’ configurationThe completed syllabus gridReflective practice model of professional education/developmentCertificates in Spolcen and Written English: curriculum structureOutcomes fov Certificate I in Spolcen and Written EnglishCompetency 13: Can write a short recountMethodology to support learners working towards CSW E outcomesWorksheet 1: student A onlyWorksheet 2: student B onlyPerceived program use10374144101122131132133139147170171172175187198217218219223234235244Tables2.12.22.3Learner roles in a learner-centered curriculumLearner-centeredness in the experiential content domainLearner-centeredness in the learning process domain283030

ILLUSTRATIONSxi2.42.52.62.72.82.96.16.28.18.29.112.112.212.312.413.115.116.116.216.316.416.521.122.122.222.322.4Changing views on the nature of language and learningCurriculum decision-making in high-structure and low-structurecontextsI

Ac I< n ow I e d g e m e n t sThe editors and publishers would like to thank thc following for permission to use copyrightmaterial:Kimberley Brown and Blackwcll Pulilishers Ltd for ‘World Englishes inTESOL programs:an infusion model of curricular innovation’ in World Englishes, Vol. 12: 1, 1993.Michael P. Brccn and Christopher N. Cantllin for ‘The essentials of a communicativecurriculum in language teaching’ in Applied Linguistics, 1980. Reprintctl by permissionof Oxford University l’ress.David Carless for ‘A case study of curriculum implementation in Hong Kong’. Reprintedfrom System,Vol. 26, 1998, with pcrmission from Elsevier Science.Ronald Carter antl Michael McCarthp for matcrial from Language as Discourse: Perspectives forLanguage Teaching ( Longman Group UK Limited 1990.) Rcprintcd by pcrmission ofI’earson Education Limited.Ronald Carter and Taylor & Francis Iiooks Ltd f‘or ‘Politics and knowledge about language’in fnvestigating English Discourse, 1997.Guy Cook for ‘The uses ofrcality: a reply to Ronald Cartcr’ in ELT,/ourna/,Vol. 52, No. 1,1998. Reprintctl by permission of EI.Tlourna1 and Oxford University Prcss.Kevin Germaine and Pauline Rca-Dickins for ‘Purposes for cvaluation’ . Rcproducetl bypermission of Oxford Univcrsity Press from Erulriution by Pauline Kca-Dickins and KevinGermainc (Oxford University I’ress 1992 .)Kathleen Graves and Cambridge Univcrsity 13 . for ‘A framework of course developmentprocesses’ in K. Graves (cd.) Ecichers us Course Developers, 1996.David Hall antl SEAMEO Regional Language Centre for ‘Material production: theory andpractice’ in A.C. Hidalgo, D. Hall antl G.M. Jacohs (e&) Getting Started: Materialr Writerson Materials Writing, 1995.Adrian Holliday for ‘Achieving cultural continuity in curriculum innovation’ in C. Kennedy(cd.) Innovation and Best Practice (l’carson Education Limited 1999). Reprinted bypermission of Pcarson Education Limited.Gary M. Jones and Multilingual Matters for ‘Bilingual education and syllabus design: towardsa workable lilucprint’ in journal of’Multilingual and Multiculttiral Devclopment,Vol. 17: 24,1996.Joan Lesikin antl College ESL for ‘Dctermining social prominen a methodology foruncovering gcndcr bias in ESL textbooks’ in Colle‘qe ESL,Vol. 8, No. 1, 1998.

AC I< N 0 W L E D G E M E N T S xi i iMichael Lewis and LanguageTeaching Publications (LTP) for ‘Lcxis in the syllabus’ in TheLexical Approach:The State of E1.T and a Way Forward, 1993.Dcfeng Li and TESOL for ‘It’s always more difficult than you plan and imagine: teachers’perceived difficulties in introducing the communicative approach in South Korea’ inTESOL Quarterly,Vol. 32, No. 4, 1998.Numa Markee and Cambridge University Press for ‘The diffusion model of innovation inlanguage teaching’ in Annual Review $Applied /.inguistics, 1 3, 1993.David Nunan for ‘Action research in language education’ in J. Edge and K. Richards (eds)Teachers Develop Teacher Research Papers on Classroom Research and Teacher Derdopment, 1993.Reprinted by permission of Macmillan Press Ltd. Includes table: ‘Types of informationrequired in a lcarner-ccntered system’ from G. Krindley Needsilnalysis and Objective Settingin the Adult Mi‘qrant Program, 1984, reproduced hy permission of NSW Adult MigrantEducation Scrvice. Includes material from table: ‘Three altcrnative ways of groupinglearners’ from ESL Curriculum Guidelines, South Australian Education Department, 1990,reproduced by permission of The Department of Education, Training and Employment(SA). Includes material from table: ‘Three alternative ways of grouping learners’ from K.Willing, Learning Svles In Adult Migrant Education, 1998, reproduced by permission ofN CE LTR .David Nunan, Clarice Lamb, and Cambridge University Prrss for material from The Se!flDirected Teacher: Managing the Learning Process, 1996.William Savage and Graeme Storer for ‘An emergent language program framework: activelyinvolving learners in needs analysis’ reprinted from $stem, Vol. 20, No. 2, 1992 withpermission from Elsevier Science.Simon Sergeant for ‘CALL innovation in the ELT curriculum’ in C. Kennedy, 1’. Lloylc andC. Goh (cds) Exploring Change in English Language Teaching, 1999. Kcprinted 11: pcrmissionof Macmillan Press Ltd.Zakia Shanvar and English Teaching Forum for ‘Adapting intlividualisation techniques for’ in English Teaching Forum, April 1991 .While the publishers and editors have made every effort to contact authors and copyrightholders of lvorks reprinted in Innovation in English /,anpage Euching, this has not been possil,lcin every case. They would \celcomc correspondence from individuals or companics thcyhave bccn unable to trace.Wc \vould likc to thank thc authors who contributed thcir chapters, as kvell as colleagueswithin and outsidcThe Open University and Macquaric University \Tho gave advice on thecontents. Special thanks are due to the following people for thcir assistance in the productionof this book:Helen Boyce (course manager)Freda Barnfield, Pam Burns and Libby 13rill (course secretaries)Liz Freeman (Copublishing)Nanette Ileynolds, Frances Wilson and the staff of the Rc-source Centrc of the NationalCentre for English Language Teaching and Research, Macquarie University.Critical readersProfessor Vijay K. Rhatia (L)ept of English, City University, I long Kong)Gcoff Thompson (Applied English Language Studies Unit, 12iverpool University, UK)Professor Leo van Lier (Educational Linguistics, University of Montcrey, USA)

x i vAC I< N 0 W L E D G E M E N TSExternal assessorProfessor Ronald Carter (Dept of English Studies, Nottingham Unil ersity, UK)Developmental testersIlona C7iraky (Italy)Eladyr Maria Norhcrto da Sil\a (Brad)Chitrita Mukcrjee (Australia)Doricn GonLales (UK)Patricia Williams (Denmark)We ha\-e reproduced all original papers and chapters as faithfully as we have been alde to,givcn the inevitable restrictions of space and the nccd to produce a cohcrcnt and readablccollection for rcaders worldwide. Whrre wc have had to shorten original materialsubstantially, these chapters arc markcd as adapted. Ellipses within square brackets indicatewhere text has bcen omitted from thc original. Individual refcrcncing styles have beenretained as in the original texts.

IntroductionDavid R. Hall and Ann HewingsHEN MACQUARIE UNIVERSlTY IN SYDNEY, Australia, and TheOpen University in the UK decided to collaborate on the tlevelopment of newcurriculum materials for study at Masters level, the partnership brought together theconsiderable experience and expertise of the two universities in open antl distance learning,applicd linguistics and language cducation.The collection of essays in this book antl thetwo companion volumes is a result of that collaboration. While the edited collections havebeen designed as one part of an overall study programme, they stand alone as extensive yctfocused collections of essays which address key contemporary issues in English languageteaching antl applicd linguistics.A major concern in editing these three volumes has been thc desire to prrscnt Englishlanguage teaching (ELI‘) in a variety of specific institutional, geographical and culturalcontexts. Hence, as far as possible across the three volumes, lve have attemptcd to highlightdebate, discussion and illustration of current issues from diffcrcnt parts of the Englishspeakingand English-using world, including those where English is not learnt as a firstlanguage. In doing this we recognise that English language teaching comprises a globalcommunity of teachers antl lcarncrs in a range of social contexts.The chapters in this volume address issues relating to curriculum change in context,and all three terms in this noun phrase are important. The cssays deal with the tliffcrc-ntways in which actual classroom practices change, whether at an individual or a system 1Rccognising that languagc teaching docs not take place in a ncutral or value-freeenvironment, they look at the choices that have to be made ivithin institutional or culturalconstraints when designing curricula and evaluating their success. They examinc the roleand influence of all the stakeholders ~ h might o have an intcrcst in what goes on in thelanguage classroom: not just language teachers and students, but parents, teacher-trainers,the local community, the press, politicians, antl so on.The titles of all four parts of the volume contain the word ‘curriculum’.Thc differentusages in the literaturc of words such as ‘curriculum’, ‘syllabus’ and ‘programmc’ (or‘program’) arc cvidence that prrcise definitions of these terms are hard to pin down. Insome cases the words are differentiated Tvhilc in others they are used almost interchangcahly,and you will notice diffcrcnt meanings attached to thc tcrminology by different authors inthis volume. Here, wc intend the tcrm ‘language curriculum’ in its widest sense, coveringall the issues rclating to the planning, implemcntation and evaluation of a scrics of languagclearningevents conceived as a coherent whole nith a specified purpose.The first part, ‘Directions in curriculum change’, raises somc issues underlying

2 DAVID R. HALL AND ANN HEWINGScurriculum design by examining the theoretical underpinnings of a number of reccntapproaches. This docs not attempt, of course, to cover all the different ways of specifying acurriculum, antl is not meant to be a historical survey (but see the Feez article in part 3 fora contextualiscd account of the tlevelopmcnt of language teaching practices).’l’hc assumptionunderlying our choice of essays for part 1 is the view which few people in languagecurriculum design ~ ould now argue Lvith that content and classroom practice must beconsidered togethcr: a curriculum is not simply a list of items to be taught.Thc first chapterliy Breen and Candlin, covering work done at Lancastcr Univu-sity on communicativelanguage teaching (CIS), \vas extremely influential in bringing this \-iew into languageteaching, although the focus on the process of learning rather than just the product isconsistent with a much older tradition in gencral education.CLT is not a nionolithic packaged set of procedures, tlespitc frequent talk in theliterature of the ‘communicative method’. All of the authors of essays in part 1 ~.ould nodoubt place themselves in the communicative tradition. Where CUT has heen introducedor encouraged or atlvocatctl in specific contexts around the xvorld it has often been in thcform of a package, introduced top-hvn as a grcat antl rc+ietl ncw idea rather than arisingfrom and de\+ictl within the classroom antl the needs and problcms of particular teachersantl learners. As the name of CLT has hccn invoked as a justification for a process of changetaking place in speci tic geographical, cultural and temporal contexts, some of its underlyingprinciples and practices have lieen contested or reinterpreted, as \vi11 be seen in laterchapters in this volume. Its influence is such, howwer, that anyone putting forward analternative paradigm has to detinc it in relation to CLI’.The essays by Nunan antl Lamb and by I.c\vis both place themselves in the communicativctradition by cmphasising the ccntralitv of the learner and the learning proThe first explores task-based learning. Although the \vord ‘task’ is overused in manypublished textbooks to the extent that it is often little more than a synonjm for ‘exercise’,and although it is certainly possible to tlcvisc, context- and c.omnlunic.ation-frcc ‘tasks’ inthis sense, the proponents of‘ a task-based syllabus have something more specific in mind.Thcy take tasks to be the interactive lcarning procedures through which learners both inand out of the classroom learn to understand each other antl to make thcmselvesunderstood, so gaining conlitlencc antl cxpcricncc in using the target language.l’hc links toLewis’s proposal to replace thc traditional Prescnt~Practisc~Protlucc teaching procedurcwith Ohserve-Ilypothcsisc~Expcriment arc clear. l‘hc observing, hypothesising andexperimenting are all being undertaken by the learner in actually using the target language.The emphasis is on the learner, antl the teacher’s role changes from ‘presenter’ to somethingmore like ‘collaborator’ or ‘facilitator’.Ihcn (1984: 5 3) claims for CLT that its ‘grcatcr concern lvith capacity for communicationrather than repertoire of communication, with the acti\ ity of learning a languageitself, and with a focus upon means rather than prctlctcrminctl objectives, all indicatepriority of process ovcr content’. We cmphasisc that this docs not say ‘process to theexclusion of content’, although some have seen CLT as a methodology that rejects theteaching of formal aspects of the language altogcthcr.Thcrc are parallels here with argumcntsthat have been used about \vholc-w ord teaching of rcading as opposed to phonics-hasedmethods, antl about process writing as opposed to the teaching of scntcnce-grammar,spelling and punctuation.Thc differing roles assigned to process, product, content and formstill constitute a major sitc of struggle hvithin language curriculum tievclopment.This struggle untlerlics much of’ the argument between the Cook essay and that byMcCarthy and Carter. McCarthy antl Carter argue that the curriculum must take accountof variation and of context, and this means that learners’ interaction with texts must take

INTRODUCTION 3account of the ‘dynamism inherent in linguistic contexts’. The understanding of thedynamism of language and contextual variation is reinforced in McCarthy and Carter’s ownwork on language corpora, large collections of actual texts stored on computer andaccessible for quantifiable linguistic analysis. To an extent, then, thcy may he seen asadvocating teaching programmes based on quantitative data, an argument that dates hackmany ycars (see, for example, West’s General Servicc List of English Words, liascd on kvordcounts, and first published in 1936). Cook argues that language teaching should not br aslave to language description, that thc processes of learning antl traching are not coincidentwith the records of language behaviour evidenced through corpora. The argumcnt comcsback, as many times before in the history of language pedagogy, to the role played liy formallanguage description in helping teachers to teach and learners to learn.No linguist \voulcl dispute the ubiquity and importance of language variation over time,space, context, purpose, etc. The essay by Hewings and Hc\vings, which ends part 1 antlprovides a link with the first chapter of part 2, dcals with the variation in language usebetwccn academic disciplines. Particularly when applied to the teaching of English forSpecific Purposes, this is an area which is attracting growing attention, \vith increasing Ic\-clsof research being undertaken around thc world (see, lor example, the essays in Candlin andFIvlantl, 19YY).The capacity of computers to store large and spccialised corpora of languagt,as it is actually used is, whatcvcr we may think of the Cook argumcnt, beginning torcvolutionisc the way we see language. Where corpora are sufficiently spccialised, theyunderline differences in lexical patterns, word meanings, grammatical pattcrns and tcxtualcohesion in a way that ivas possible only in a limited and largely intuition-based way beforcthe corpus databases ivere created. This development has already hatl an influence on tieltlssuch as lexicography, forensic linguistics, English for academic purposes and curriculumantl syllabus design. The approaches takcn by Hewings and Hclvings and by McCarthy andCarter are illustrative of these last two. Further dcvelopmcnts in thcsc areas may scrve toinform and/or challcnge both learners and teachers in the futurc.The incorporation of variation into language curricula is not, at present, a midesprcatlpractice, and the essay hy Cartcr which opens part 2 gives some indication of why it is not.In fact, all of the essays in part 2 deal in one way or another with the political and institutionalconstraints in curriculum development. Language, as an intimate component of individualand national identity, is a particularly emotional issur, and cui-riculum developers tamperwith ‘the hvay things have always been donc around herc’ at their o\vn risk. Cartcr \vas brantlcdin some sections of the press as a revolutionary traitor to the nation for his emphasis onknowletlge about language and how it functions. Even at a local bel, teachers working bythcmsclves in individual classrooms and taking ivhat seem like innocent decisions ahout theday’s activities can easily tind themselves the subject of parents’ or the school principal’s ire.All but one of thc seven chapters in part 2 deal \vith attcmpts to change curricula inspccific contcxts.The cxccption is the cssay by Markce, which prcsents a framchvork for thcintroduction of innovations. As might hc expected from an examination of Markee’scategorics and criteria, none of the projects reported could be judged uncquivocallq. as afailure or a success. He dcals with issues such as: At \vhat point can an innovation be judgctlas satisfactor?? What is the critical mass which determines whether an innovation has beenadopted by the population for which it was intcndrd? Each of the other essays in this sectioncould lie examined in thc light of Markee’s criteria and follo\vetl up to see what has happenedin the period since the first publication of the essay. Have the innovations discusscd by Carter,Jones, Brown, Sarwar, Savage and Storer, and Li actually hatl any lasting effect? Iiave, forexample, Carter’s ideas set out in the IJNC project influenced British tcachers’ views onlanguage? Has the mix of Malay and English in thc Brunei school system actually liccn

4 DAVID R. HALL AND ANN HEWINGSmodified following the publication of Jones’s essay? Has Kimberley Brown managed topersuade American libraries to stock more world-English titles? Has Zakia Sarwar convincedher colleagues to introduce similar methods in their classes, or transferred what she wasdoing to the mainstream, official curriculum? Have the needs analysis procedures of Savagcand Storer been extended to other contexts? And have Defeng Li’s Korean teachers adapted,adopted or rejectcd a communicative approach? These would be interesting questionsto follow up, and sometimes they can lie pursued through publications in journals (see, forexample, Storer and Savage, 1999, for furthrr extensions of their own work) andincrcasingly through internet discussion lists.Part 3 examines both the planning and the implementation of curriculum change. Manycurriculum dcvelopment tcxtbooks treat these two aspects separately, but it is clear thatthe processes involved are circular rather than linear, and that both need to involve or atleast take into consideration all of the stakeholders. This part, then, continucs the theme ofcurriculum development in its social and institutional context.All the essays in this part examine the ways in which change is brought about.They alsoraise the question of whether change is always desirable, particularly in cases wheresuccessful models from one context arc imported into new contexts. This is currently amuch-disputed point in language teaching in rclation to the recent insistence in some circleson learner autonomy as a necessary condition for successful language learning.Thc chaptersIiy Sarwar and Savagc antl Storer in part 2 hoth deal with the introduction of autonomousapproaches. In the first essay in part 3, however, Holliday presents a rather differcnt andprovocative view of learner~centrcdness. For Holliday, ‘learner-centredncss’ has becomea short-hand way of referring not to individuals but to thc skills antl compctencics we canequip thcm with and thc evaluative mechanisms that can lie used to test how effective wcas teachers have bcxn.This teachcr-centred interprctation of learner-centredness is a highlycontentious position, with those advocating learner-ccntredncss strongly disputing thisunderstanding oftheir approach (see, for cxamplc, Savagc, 1997). Clcarly, thcrc arr complexantl sometimes contradictory arguments involved here, with both sides claiming to hold themoral high ground.Thc relationships lxtwccn different stakeholtlers ~ between donors andrecipients, policy-makers antl practitioners, native and non-native speakers, teachers andlearners, insiders antl outsiders, cxpcrts antl novicrs arc dclicatc antl involve many moreissues than how to tcach language, as \vc see again antl again in these essays. While Hollidayassumes a top-clown model (Lvhich he attacks), the chapters by Graves and liy Nunan 110thlook at ways in which innovation can be instigatctl by the teachcr or by teachers and othersworking togethcr. Feez givcs cxamplrs of this collalxxative approach when she describesin some detail the \vays in irhich curriculum change has taken place within a large systcm.Although systcmic innovation necessarily in\ olvcs some form of imposition, it will be seenthat through consultation and Lrorkshops, teachers at all levcls have been involvctl in thcvarious reformulations of the curriculum. Hall tlc ribes four tliffercnt curriculumdevelopmentprojects antl examines thrm in thc light of their capacity for helping thelearners to learn. Hall’s introduction of the irnportancc of defining what you are trying todo in tlcvcloping curricular innovations leads on to the final chapter in this part, in whichSergeant analyses the various uses made of computcrs in the language classroom and themotivations attachetl to those uses. He makes a va1ual)le distinction between ‘change’brought about by computer-assisted language learning (CALL), which he sees as a superficialaddition to the curriculum, antl ‘innovation’ using CALL, which is embedded in thecurriculum and encourages new ways of teaching and lrarning. This separation of superficialversus embeddcd, change versus innovation is applicablc to all areas of the curriculum, notjust CALL.

INTRODUCTION 5Part 4 focuscs on evaluating curriculum change, placing this as an integral componentof the planning and implementation cycle. While evaluation is an essential elementunderpinning all the essays in the volumc, this section looks in some detail at different kindsof evaluative processes. Rea-Dickins and Germainc, in answer to the question of why wemight need to evaluate at all, provide an overview of the different purposes for whichevaluation is conducted. The kind of evaluation exemplified in the Carlcss essay is one of agrowing number of research projects using qualitative methods. His case-study approach,involving what is sometimes called a ‘thick’ description, turns a microscope on onc specificcvent in one specific context. As a rcsearch methodology, it may not yet havc the academiccachet of quantitative and controllcd research methods, and may not he able to upholdclaims to generalisability in the same way as quantitative methods do, but it can rcveal muchmore about why ideas that seem fine in theory work rather differently, if at all, in real life.In the longer term, as more and more small-scale context-cmtieddcd research projects arereported, it may tie that these stories will constitute the most \-aluablc resource for theteacher and curriculum dewloper.The final chapter, by Lcsikin, provides a framework for evaluating and analysinglanguage-teaching materials. Using the tools of systemic-functional grammatical analysisshc uncovers systematic gender bias in a published textbook. Lesikin’s procedures are \-crymuch in thc tradition of critical discourse analysis, and it is clear that they could be uspd incontexts other than that outlined here.The essays in this collection are designed as a source of thought-provoking ideas for allstudents of language teaching, language teachers, teacher-trainers, curriculum dedopersand educational administrators. They exemplify a range of work by academics with wideexperience in different parts of the world and by teachers who are still very close to thechalk-face. They make the link between theory and the actual circumstances in whichlanguage learning takes place or fails to take place. This volumc is not a handbook, and hasno predetermined answers to the problem of finding appropriate ways of putting togetherlanguage curricula so that lcarners can learn. Instcad, it provides a series of conccptualframeworks within which such a quest can be undertaken.ReferencesBreen, M. P. 1984. ‘Proccss syllabuses for the language classroom’, in C. Brumfit (ed.) GeneralEnglish Syllabus Design: Crirriculum and Syllabus Design-for the General English Classroom, ELTDocuments I 18. Oxford: British Council and Pcrgamon Press.Candlin, C. N. and Hyland, K. (eds) 1999. Writing: Em, Procecses, and Practices. Harlow:1,ongman.Savage, W. 1997. ‘Language and development’, In B. Kenny and W. Savagc (cds) Language andDevelopment. Harlow: Longman, 283-325.Storer, G. and Savage, W. 1999. ‘Extending an emergent frarnc\vork to other contexts’, System,27. 3: 421-5.

PART ONEDirections in curriculumchange

Chapter 1Michael P. Breen andChristopher N. CandlinTHE ESSENTIALS OF A COMMUNICATIVECURRICULUM IN LANGUAGE TEACHINGIntroductionT A TIME WHEN THERE IS A RECOGNISED NEED inlanguagcteachingA to give adequate attention to language use as well as language form, various ‘notionalfunctional’or so-called ‘communicative approaches’ to language teaching are lxingadvocatcd. In this context, the present paper is offered as a sct of proposals in an effort todcfine the nature of communicative language tcaching.Any teaching curriculum is designed in answer to three interrelated questions: What isto be Icarned? How is the learning to be undertaken antl achieved? To what cxtent is theformer appropriate and the latter effective? A communicative curriculum will place languageteaching within the framework of this relationship between some specified purposcs, themethodology which will be the means towards the achievcment of those purposcs, antlthc evaluation procedurcs \vhich will assess the appropriatencss of the initial purpoand the ell’cctiveness of the methodology.This chapter presents the potential charactcristics of communicative languagc teachingin terms of such a curriculum framework. It also proposes a set of principles on \vhichparticular curriculum designs can be based for implemcntation in particular situationsand circumstances. Figure 1.1 summarises the main areas with w.hich this chapter \vi11 dcal.In discussing the purposes of language teaching, we will consider (1) communication as ageneral purpow, (2) thc underlying demands on the learner that such a purposc may imply,and (3) thc initial contributions which learners may bring to the curriculum. In discussingthe potential methodology of a communicative curriculum, \ye \vi11 consider (4) thc proccssof teaching and learning, (5) thc roles of teacher and learncrs, antl (6) thc role of contentwithin the teaching antl karning. Finally (7) we \rill discuss the placc of evaluation of learnerprogress and cvaluation of the curriculum itself from a communicativc point of view. ’Inevitably, any statement almut the components of the curriculum runs thc risk ofpi-csenting in linear form a framettsork which is, in fact, char-actcrised hy intcrtlependcnceand overlap among the components. In taking purposes, methodology, antl evaluation inturn, therefore, wc ask readcrs to bcar in mind the actual interdependence between them.What follows is a consitleration of those minimal requirements on communicativelanguage learning and teaching which, in our view, must now he taken into account incurriculum design and implcmcntation.

10 MICHAEL P. BREEN AND CHRISTOPHERTHE CURRICULUM1 Communication\ /The classroomprocessTeacher / learnei ' roles\ 6Role of contentI \7 Of learner Of curriculumFigtire I. 1 Thc c.urriculum1 What is the purpose of the curriculum?The communicative curriculum tlcfincs languagc lcarning as lcarning how to communicateas a membcr of a particular socio-cultural group.Thc social convcntions governing languagcform and hehaviour within the group arc, therefore, central to the process of languagclearning. In any communicative evcnt, intlivitlual participants bring with them priorknowledge of meaning and prior knowledge of how such meaning can be realiscd throughthe conventions of language form antl Iichaviour. Since communication is primarily intcrpersonal,these conventions arc subject to variation \vhilc they are being uscd. In cxploringshared knowledge, participants \fill he modifying that knowlctlgr. They typically exploit atcnsion between the conventions that are cstahlishcd and the opportunity to modify theseconventions for their particular communicative purposes. Communicating is not merely amattcr of folloning conventions hut also of negotiating through and ahout thc convcntionsthcmselvcs. It is a convention-creating as well as a convention-follo\ving activity.In communication, speakers and hcarcrs (and writers antl readers) are most oftenengaged in the work of sharing meanings \vhich arc 110th dcpendcnt on the conventions ofinterpersonal hehaviour antl created by such Iiehaviour. Similarly, thc itleas or conceptswhich are communicated about contain differcnt potential meanings, and such potentialmeanings are expressed through antl tlerivctl from the formal system of text during theprocess of communication .To understand the conventions which underlie communication,therefore, we not only have to understand a system of ideas or concepts and a system ofinterpersonal bchaviour, we have to understand how thcse itlcas and this interpersonalbchaviour can be realiscd in languagc ~ in connectcd texts. Mastering this unity ofideational,interpersonal and textual knowledge allows us to participate in a crcative mcaning-makingprocess and to express or interprct the potential meanings within spoken or written text(Hallitlay, 1973).There is an additional characteristic of this unificd system of knowletlgc. The social orinterpersonal nature of communication guarantees that it is permeated liy personal andsocio-cultural attitudes, values and emotions.These different ufects will determine what wechoose to communicate about and how we communicate. The convcntions governing ideasor concepts, interpersonal hehaviour, and their realisation in texts all scrve and createattitudes, judgements and feelings. Just as communication cannot be affectively neutral,

A COMMUNICATIVE CURRICULUM 11learning to communicate implies that the learner will come to terms with the ne\v learningto the extent that his own affects will be engaged. At that point, thc learner’s affccts becomefurther involved in a process of negotiation with those affects which are embodied withinthe communicative performance of the target community. So, affective involvement is boththe driving-force for learning, and also the motivation behind much everyday communicationand the inspiration for the recreation of the conventions which govern suchcommunication.Communication in everyday life synthcsises ideational, interpersonal, and textualknowledge - and thc affects which are part of such knowledge. I3ut it is also related to andintcgrated with other forms of human lxhaviour. In learning how to communicate in a newlanguage, the learner is not confronted by a task which is easily separable from his otherpsychological and social experiences. The sharing and negotiating of potential meanings ina new language implies the use and refinement of perceptions, concepts and affects.Furthermore, learning the conventions governing communication within a new social groupinvolves the refincrnent and use of the social roles and the social identity cxpectcd bythat group of its members. Thus, learning to communicate is a socialisation process. [. . .]Therefore, it makes sense for the teacher to see the overall purpose of language teaching asthe development of the learner’s communicative knowlcdge in the context of personal andsocial de\ 7c ~1 o p ment.2 What underlies the ultimate demands on the learner?A language teaching curriculum, from a communicative point of view, will specify itspurposes in terms of a particular target repertoire (Gumperz, 1964). Different curriculawill hopefully select their own particular repertoires from a pool of communicativepcrformance on the basis of a sociolinguistic analysis of thc target situation. This does notimply that any one curriculum will 11c necessarily entirely distinctive in the target reprrtoireto which it is devotcd. At the surface there will be inevitable overlap among differentrepertoires. However, underlying any selected target repertoire there will be an implicittarget competcnce. It is this target competence which we may define as the capacity foractual use of the language in thc target situation. So, in specifying the purposes of thecurriculum, a rcquiremcnt for thc communicative approach would be to make an initialdistinction between the target repertoire ultimately demanded of the learner and the targetcompetence which will undcrlie and generate such a repertoire.How can tve characterisc this target competence? We have already proposed thatlearning to communicate involves acquiring a knowledge of the conventions which governcommunicative pcrformance. In addition, we have proposed that such communicativeknowledge can lie seen as a unified system ofideational, interpersonal, and textual knowledge,which incorporates a range of affccts.We have also suggested that communication and learning how to communicate involvethe participants in the sharing and negotiating of meanings and conventions. Such sharingand negotiating implies the existence of particular communicative abilities as an esscntialpart of competence. Therefore, we may identify within competence both the knowledgesystems and the abilities which call upon and act upon that knowledge. These abilities canbe distinguished within competence more precisely. In order to sharc meaning, theindividual participant needs to be able to interpret the meanings of others and to expresshis own meanings. However, such interpretation and expression will most often take placein the context of interpersonal and personal negotiation. The ability to negotiate operatesbetween participants in communication and within the mind of the individual participant

as12 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLIN- the latter negotiation is perhaps more conscious during new learning. More obviously,participants in communication negotiatc hvith one another. Rut, in endeavouring to interpretand express with a new language, thc learner will himself negotiate lictwecn the communicativecompetence he already possesses and that which underlies the new learning.’Wc suggest, thcreforc, that the communicative ahilitics of interpretation, expression,antl ncgotiation arc the essential or ‘primary’ ahilitics within any target competence, It is alsolikely that these three ahilitics continually interrclatc with one another during communicativeperformance and that they are complex in naturc.l’hey will involve psychological processesfor the handling of rich and variablc (lata ~~ thc attcmtion and memory processes, for cxample- antl they may contain within them a rangc of sccontlary abilities such as ‘coding’,‘codesubstituting’ and ‘style-shifting’(Ret-nstcin, 1971, I Iymes, 1971, LAov, 1970).The use of these communicative abilitics is manifested in communicative performancethrough a set of skills. Speaking, listening, reading antl lvriting skills can lie sccn to scrveand depend upon the untlerlying abilities of interpretation, expression antl negotiation. Inthis \vay \ve arc suggesting that the skills i-cprcscnt or realise underlying communicativeabilities. The skills arc the meeting point lict\vccn underlying communicative competenceand ot)serval)le communicati1.e pcrformancc; they arc the means through which knodedgcand ahilitics are translated into pcrfoi-mancc, and vice vcrsa.In sclecting any target repertoire, thcrefoi-c, a communicative curriculum alsodistinguishes antl specifics the target compctcncc on \vhich the performance of such arcpertoire depends and through which it is achic\.ctl. This specification would indicate theideational, interpersonal and tcxtual conventions ~ and the affective aspccts of suchconventions ~ a related antl underlying tcm of knoLvlcdgc which is shared anddeveloped within the target community. The specification ~vould also indicate the demandsupon the lcarncr’s communicative aliilitics of intcrprctation, cxprcssion, and negotiationsimilarly underlying communicative performance in the target community and the rangeof skills which manifest these abilities. Such a specification \zould account for what thclearner nerds to know, antl how thc lcarncr needs to bc alilc to use such knowledge. Thcultimatc demands on the learner in terms of some specific target repertoire will, in ourvie\v, derive from antl depend upon this underlying competence of communicativekno\vledgc antl communicative abilities.3 What are the learner’s initial contributions?[. . .] A communicativc specification of purposes supports the principle that the roots ofour ohjecti\-cs can already be discovered in our learners ~ however liencath the surfaceof the actual targct repcrtoii-c these roots may lie. We need to try to rccognisc what thelearner knoivs and can do in communicative performance with the first language and notassumc that thc Icarncr’s ignorance of the target rcpcrtoii-c implies that the learner is a naivecommunicator or someone who evaluates communication in only a superficial \lay.This principle, which scems to rcquirc us to credit thc kat-ner with a highly relevantinitial competcncc, of communicative kno\vletlgc antl aliilitics, has often been overlookedor only partially applied in language teaching. In the past, it has seemed easier to somchon.separate the learner from the knonletlgc to I x lcarnctl to ‘oljcctily’ thc targct languageas something completely unfamiliar to the learner. This olijrctification of the language inrelation to the learner has perhaps been encouraged hy a narrow definition of what theobject of learning actually is, antl by an incomplete view of what the learner has to offer.Wc have tended to see the target only in terms of ‘linguistic competence’ or tcxtualknowledge, and we have limited such knowledge to the level of syntax without reference

evenontheforA COMMUNICATIVE CURRICULUM 13to structure above the sentence. Thus, ideational and interpersonal knowledge, whichcontinuallv interact with textual knowledge and from which textual knowledge evolves,have tended to be overlooked or neutralised. We have often seen the learner primarily interms of the first language, and we have often assigned to it ‘interference’ value alone ~again taking a narrow textual knowledge as our criterion. More recently, due totlevclopmcnts within sociolinguistics, we haw recognised the significance of ‘ sociolinguisticcompetence’ and also of the ‘functional’ aspect of language. However, a partial antlknowledge-based view of learner competence seems to remain with us and the lcarncr’scommunicative abilities underlying the initial repertoire still need to bc more thoroughlyexploited. Rather than just allowing the use of the first language in the classroom, we shouldperhaps be more conccrned with activating that which underlies thc initial repertoire ofthe learner, and to evoke and engage what we mav describe as the lcarncr’s ongoing or processcompctencc.j Once we define the object of learning as communication, then \ve arc enahledto perceive the learncr in a new light. His initial textual knowlcdge is placed in its properperspective - it is merely the tip of the iceberg. Language teaching need no longer beprimarily conccrned with ‘linguistic competence’ . We can begin mith the assumption thattext is the surface realisation of communicative knowledge antl abilities antl that text is usedand created and learned ~ the basis of thcm.7’he communicative curriculum seeks tofacilitate ~ guarantee ~ involvement of the learner’s communicative knowledgeand abilities from the outset rather than overlook them for the sake of some apparent‘fluency’ \vith text. I. . .] Ho er, lcarners not only contribute prior kno\vledge antlabilities, they also have expectations about the le ing of a language. What the curriculumseeks to achicve in terms of any specified purp must be balanced by what the learnerpersonally expects of the curriculum. Perhaps the current interest in teaching language for‘special purposes’ may eventually reveal the challenge to curriculum designers: that alllearners regard themselves as learning a language for some special purpose.We can identify several types of learner expectations and these may, of course, influenccone another. We can ask: What is the learner’s own vicw of the nature of language? What isthe learner’s view of learning a languagr? (Thc answers to these questions ma)- lie in thelearner’s previous formal education, and how he reacted to that experience.) We can alsodistinguish between, first, ho\v the learner defines his o\.cn language learning ncrds;secondly, what is likely to interest the learner both within the target reprrtoire antl thclearning process; and, third, \\,hat the learner’s moti\rations are for learning the targetrcpertoire. All these initial expectations arc disti ti need to be discovered in somc \vayso that areas ofpotential match antl mismatch be learncr expectations antl the selectedtarget repertoire antl its underlying competence can be best anticipated.Two important problems nrctl to be identified hcrr in accounting for learncrexpectations. Thesc expectations are inwitably various antl ~ more significant1)- - thc)- aresubject to change over time. So, the curriculum will need to accommotlatc and allow for aheterogeneity of lcarner expectations. It will also ncetl to allow for changes in differentlearncrs’ perceptions of their needs, in ivhat interests different learncrs, and in themotivations of differrnt learners. In this way, curriculum purposes should account for initialcxpectations of lcarners and anticipate changes in expectations during the learning teachingprocess. Such an account and such anticipation may appear to be an impracticable dreamwhen confronted with the variety and fluctuation in the real expectations of learners. Thatwe should try to account for and anticipate these is a further motivation for a communicativecurriculum, and more particularly ~ a communicatjvr methodology (scc sections 4ff).However, there is a second important aspect of learner cxpectations: expectations can tieeducated. For this to happen, learners need to be enabled to express their own expectations;

~ spoken,~ whatwhichincludingand14 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINto explore them and the sourccs from which they derive. They also need to be enabled tointerpret the expectations which the specific purposes of the curriculum make upon themas learners. They need to interpret ~ at the start of the learning- teaching process andthroughout this process ~ the target repertoire and its underlying competence demandsof them. Howcver vague a learner’s initial interpretation may be, he is not going to learnanything unless he has an idea of what he is trying to achievc. Therefore, a process ofnegotiation between the learner’s contributions ~ expectations ~~ the targetrepertoirc, and the means b? which thesc two are brought together, is likely to bccharacteristic of a communicative methodology. Curriculum purposes inform and guidemethodology, and an account of learner cxpcctations within purposes can enablemethodology to involve these subjective contributions of the learner and, thereby, call uponthc genuine intersubjective rcsponsibility of that learner.4 How are the curriculum purposes to be achieved?4. I Methodology as a communicative processLanguage learning within a communicative curriculum is most appropriately seen ascommunicative interaction involving all the participants in the learning and including thevarious material resourccs on which the learning is cxercised.Therefore, language learningmay be seen as a process which grows out of the interaction lietween learners, teachers,texts and activities.This communicative interaction is likely to engage the abilities within the learner’sdeveloping competence in an arena of cooperative negotiation, joint interpretation, and thesharing of rxprcssion. The communicative classroom can serve as a forum characterised liythe activation of these abilities upon the learners’ new and developing knowledge. Thisactivation will depend on the provision of a rangc of differcnt text-types in different mediawritten, visual and audio-visual ~ the participants can make use of todevelop their competencc through a variety of activities and tasks. The presence of a rangeof text-types acknowledgcs that the use of communicative abilities is not restricted to anyone medium of communication.The earlier distinction we saw between underlying abilitiesand the set of skills which serve and depend on such abilities enables us to perceive that thelearner may exploit any selected skill or combination of skills to develop and refine hisinterpretation, expression and negotiation. Thr learner need not be restricted to thcparticular skills performance laid down by the target repertoire. Because communicativeabilities permeate each of the skills, they can be scen to underlie speaking, hearing, readingand writing and to be independent of any prcscribcd selcction or combination of theseskills. Similarly, just as no single communicative ability can really develop independentlyof thc other abilities, so the developmcnt of any single skill may well drpend on the appropriatedevelopment of the other skills. In other words, a refinement of intcrprctation willcontribute to the refinement of expression, and vice-versa; just as a refinement of the skillof reading, for example, will contribute to the refinement of the skill of speaking and viceversa.Classroom procedures and activities can involve participants in 130th communicatingand metacommunicating. We have rcferrcd to the characteristics of communicating insection 1 of this paper. By metacommunicating we imply the learner’s activity in analysing,monitoring and evaluating those knowledge systems implicit within the various text-typesconfronting him during learning. Such metacommunication occurs within the communicativeperformance of thc classroom as a sociolinguistic activity in its own right.

A COMMUNICATIVE CURRICULUM 15Through this ongoing communication and metacommunication, learners not only becomeparticipants in the procedures and activities, they may also become critically sensitised tothe potential and richness of the unified system of knowledge, affects and abilities uponwhich their communication depends. [. . .] In particular, the involvement of all theparticipants in a process of communicating through texts and activities, and metacommunicatingabout texts, is likely to exploit the productive rclationship hetwcen usingthe language and learning the language.4.2 Methodology as a diflerentiated processThe emphasis given in the previous section to the interactive nature of the communicativecurriculum suggests, in turn, the need for a communicative curriculum to be differentiated.A communicative curriculum begins with the principle that we should differentiate withinpurposes between the target repertoire and the communicative knowledge and abiliticswhich underlie it. A second principle is that the learner’s process competence needs tobe differentiated from the target competence, and that different learners may exploitdifferent process competenccs as the means towards some particular target. These kinds ofdistinctions involve differentiation at the curriculum level bctwecn purposes and themethodology adopted to achieve such purposes.Within mcthodology, differentiation is a principle which can be applied to theparticipants in the learning, the activities they attempt, the text-types with which theychoose to work, and the ways they use their abilities. It is worth considering differentiationwithin these areas in more detail:(a) Learners’ contributionsIndividual learners bring individual contributions to the language learning process in termsof their initial competence, their various expectations about language learning, and theirchanging needs, interests and motivations prior to and throughout the language learningprocess. Wc can recognisc that, even in the achievement of some common target conipctcnre,diffcrent lcarners through their changing process competcnce may well adoptdifferent mcans in attempting to achieve such competence.(b) RoutesThe emphasis within a communicative curriculum on the communicative process oflanguage learning, with the consequent emphasis on cooperative learner activities, offers anatural means for differentiation, Different learners need the opportunity of followingdifferent routcs to the accomplishment of some individual or common group objective .Such variation in choice of route typically involves selection among alternative skills orcombinations of skills, and hence the choice of alternative media. The variation may bemotivated by the need to work at a different pace from other learners, or by the desire topursue alternative content. This sclection among routcs can itself be open to jointinterpretation, the sharing of expression and cooperative negotiation.(c) MediaIn order to allow for differences in personal interest and ease of access, or to permit thesearch for alternative perspectives on the content, learners should be offered the possibilityof working with one or more of a range of media. We mean by this that learners would be

16 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINexpected to act upon text-types in the appropriatr medium: written texts would be read,spoken ones listened to, visual ones seen. Just as communication is governed I>y conventions,so we can see that the different media represent and obey conventions specific to themselves.Learning dialogue by reatling, for cxample, may neutralise the authentic conventions ofspoken discourse, and bve may tie asking thc learner to hccomc involved in using andapplying knowledge in a distorted \vay.Whatever the route chosen or the media antl tcxt-types selected for communicativelearning, different learners will have differentiated ways of making use of the abilities xvithintheir communicative compctence, antl \vi11 therefore adopt different learning strategies.Such heterogeneity is often secn as problematic for the teacher, but a communicativemethodology would take advantage of this differentiation among learning strategies, ratherthan insisting that all learners exploit the samc kinds of strategy.Thcse four illustrations of the principle of tliffercntiation within a communicativemethodology imply morc than merely offering to individual lcarncrs opportunities Fordifferential communication and learning, or acknowledging differences between pcrformanccrepertoires antl the developing compctcnces underlying them. Differentiationdemands and authcnticates communication in the classroom. The various perspectivesoffered by alternative media, the accomplishment of shared objectives through a varietyof routcs, and the opportunities for exploiting different learning strategics, all facilitatethe conditions for authentic communication among the participants in the learning.Differentiation also enahlcs the lcarner to authenticate his own learning and thcreby becomeinvolved in genuine communication as a means to\vartls it. Further, if ~ve confront learnerskvith texts and text-types lvhich are also authentic, this obliges us to allow for differentinterpretations antl differences in holv learners will themselves negotiate with texts.4.3 Methodology exploits the communicative potential of thelearning-teaching contextWe are easily tcmptcd to cxcuse the classroom as an artificial or synthetic language learningcontext ~ as distinct lrom somc natural or authentic environment, The communicativecurriculum sceks to exploit the classroom in terms of \vhat it can realistically offer as aresource for learning. This would not necessarily mean changing or disguising the classroomin the hope that it will momentarily servc as sonic kind of ‘communicativc situation’resembling situations in the outside \vorltl. The classroom itself is a unique socialenvironmcnt with its own human activities and its own conventions governing thescacti\itics. It is an environment where a particular social-p hological and cultural realityis constructcd. This uniqueness antl this rcality implies a communicativc potential to beexploited, rather than constraints lvhich haw to be overcome or compensated for.Experimcntation Lvithin the prior constraints of any communicativc situation is, as wc haveseen, typical of the nature of communication itself, and thc prior constraints of classroomcommunication need he no exception.We can make a distinction betwren the different contributions offered to learning by,on the one hand, the ‘formal’ language learning contexts ofthc classroom and, on the other,the ‘informal’ learning which takes place at any time, anywhere. The classroom can becharactcrised by the kinds of learning which are best generated in a group context, while‘informal’ lcarning undrrtaken beyond the classroom is often an individual commitment,

andA COMMUNICATIVE CURRICULUM 17espccially in the contcxt of foreign language learning. Thus the ‘formal’ context is one wherethe interpersonal relationships of the classroom group have their own potential contributionto make to the overall task. Within the communicativc curriculum, thc classroom i and theproccdurcs and activities it allows ~ can serve as the focal point of the learningteachingprocess. In adopting a methodology characterised by learning and teaching as a communicativeand differentiated process, the classroom no longer needs to bc seen as a palerepresentation of somc outside communicative reality. It can become the meeting-place forrealistically motivated communication-as-learning, communication about learning, andmetacommunication. It can he a forum where knowledge may be jointly offered and sought,rcflectcd upon, and acted upon. The classroom can also crucially serve as the source offeedback on, and refinement of, the individual learner’s own process competcnce. And itcan serve as a springboard for the learner’s ‘pcrsonal curriculum’ lvhich may be undertakenand developed ‘informally’ outsidc the classroom. As a coparticipant in the classroom group,the learner’s own progress can lie both monitored and potentially sustained by himself onthe basis of others’ feedback and by others within some shared undertaking.To cnsurc that the special and differing contributions offered by both ‘formal’ and‘informal’ contexts of learning can be fully exploited, a communicative mcthodologyhas to try to relate the two.The classroom can deal with and explore phenomena which aresignificant in the cxpericnced ‘outsitlc world’ of the lcarner, and it can hccomc anobservatory of communication as everyday human behaviour. As well as looking outwards,the classroom has a rcflcxivc role as a laboratory where observations can ticcome thc mcansfor the discovery of new knowledgc and the tlevclopmcnt of abilities.A communicative methodology \vi11 therefore exploit the classroom as a resource lvithits own communicative potential. The classroom is only one resource in languagc teaching,but it is also the meeting-place of all other resources ~ learners, teachers, and texts. Eachof these has sufficiently hctcrogcncous characteristics to makc classroom -based negotiationa necessary undertaking. 1. . .] The authenticity of the classroom lies in its dual rolc ofohscrvatory and laboratory during a communicative learning-teaching process.5 What are the roles of the teacher and the learners within acommunicative methodology?5.1 The teacherWithin a communicative methotlology thc teacher has two main roles. The tirst I-ole is tofacilitate the communicative process between all participants in the classroom, and lictuwnthese participants and the various activities and texts. The second role is to act as aninterdependent participant within the learning-teaching groups. This latter role is closelyrelated to the objective of thc first role and it arises from it. Thcsc roles imply a set ofsecondary roles for the teacher: first, as an organiser ofresources and as a resource himself.Second, as a guide within the classroom procedures and activities. In this role the teacherendeavours to make clear to the Icarnei-s what they need to do in order to achirvc somcspecific activity or task, if they indicate that such guidance is necessary. This guidance rolcis ongoing and largely unpredictable, so the tcacher needs to share it with other learners.Related to this, the teacher ~ other learners can offcr and seck feedback at appropriatemoments in learning-teaching activities. In guiding and monitoring the teacher needs to bea ‘seer of potential’ with the aim of facilitating and shaping individual anti group knowledgeand exploitation of abilities during learning. In this way the teacher will he concentratingon the process compctcnces of the learners.

in18 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINA third rolc for the teacher is that of researcher and lcarner - with much to contributein terms of appropriate knowledge and abilities, actual and observed experience of thenature of learning, and organisational capahilitics. As a participant-observer, thc teacherhas the opportunity to ‘stcp back’ and monitor the communicative protcaching.As an interdrpcndent participant in thc process, the teacher needs to actively share theresponsibility for learning and teaching with the 1carncrs.This sharing can provide the basisfor joint negotiation which itself releases thc teacher to become a co-participant. Perceivingthc learners as having important contributions to make in terms of initial comprtenceand a rangc of various and changing expcctations can enable thc teacher to continuallyseek potential antl exploit it. A requirement on the teacher must he that he distinguishbetween learning and thc performancc of Irhat is being learned. The tcachcr must assumethat the performance within any target rcpertoire is separablc from thc means to theachievement of that rcpertoirc. Also, he must assume that learners are capablc of arrivingat a particular objective through tlivcrsc routes. The teacher nccds to recognise learning asan interpersonal undertaking over which no single person can have full control, and thatthere will be differences between ongoing learning processes.The teacher has to accept thatdifferent learners learn different things in different bvays at different times, and he needs tobc patiently awarc that somc learners, for cxamplc, \vi11 enter periods when it swms thatlittle or no progress is being made and that, sometimes, learning is typified by silentrefkction.5.2 The learnerRegardless ofthe curriculum in lvhich they \vork and rcgartllcss of whether or not they arebeing taught, all learners ofa language arc confronted by the task of discovcring how to learnthe language. All lcarncrs \vi11 start with differing cxpcctations about the actual learning,but each individual learner will be required to adapt and continually readapt in the processof relating himself to what is being learned. The knowledge will be redefined as the learneruncovcrs it, and, in constructing and reconstructing his own curriculum, the learner maydiscover that earlier strategies in the use of his abilities need to be replaced hy otherstrategies. Thus, all learners ~ thcir own wys have to adopt the role of negotiationbet>vecn thcmselvcs, thcir learning process, and the gradually revealed object of learning.A communicative methodology is charactcrised by making this ncgotiative role - thislearning how to learn ~ a public as well as a private undertaking. Within the context of theclassroom group, this role is shared and, thereby, made interpersonal. If we recognise thatany knowledge which we ourselws haw mastered is always shared knowledge and that wealways seck confirmation that \vc ‘know’ something by communicating with other people,\re have to conclude that knowledge of anything and thc learning of anything is an interpersonalmatter. Also, if we recognisc that real knowledge is always set in a context and thiscontext is both psychological and social ~ what is known will always be contcxtualised withother knowlcdge in our minds antl will always carry with it elements of the social contextin which it was experienced ~ then we also have to conclude that a significant part of ourlearning is, in fact, socially constructed. These justifications for a genuinely interpersonalmethodology are quite independent of the nature of what is to bc learned. If the object oflearning is itself communication, then the motivation to enable the learner to adopt aninterpersonal means to that learning is doubly justified. Quite simply, in order to learn tocommunicate within a selected target repertoire, the learner must be encouraged tocommunicate ~ to communicate about the learning process, and to communicate about the

evenbecauseA COMMUNICATIVE CURRICULUM 19changing object of learning on the basis of accepting that ‘learning ho\v to learn’ is a problemshared, and solved, h); othcr learners.Within a cornmunicativc methodology, the rolc of learner as negotiator ~~ between theself, the learning proccss, and the object oflcarning ~ emerges from and interacts with therole of joint negotiator mithin the group and within the classroom procedures antl activitieswhich the group undcrtakes.l’hc implication for the learner is that he should contribute asmuch as he gains, and thereby learn in an interdependent way. The learner can achieveinterdependence by recognising responsibility for his own learning and by sharing thatresponsibility with other learners and the teacher. This commitment can be initiated antisupportcd by a milieu in which the learncr’s own contributions - interprctations, exprcssions,and cfforts to negotiate ~ are recognisrd as valid and valuable. Such a context \vouldhe typified by the acceptancc of ongoing success andfuilure as necessary prcrcquisites to\vardssome ultimate achievement, where it is assumed that lcarners inevitably bring with thcm‘mixed abilities’ antl that such a ‘mixture’ is, in fact, positively useful to the group as a whole.Commitment to communication on the learner’s part need not he regarded as somethingunattainable or threatening ~ for the ‘beginning’ learner ~ he is expected torely on and develop that which is familiar: his own proccss competence and experience ofcommunication.As an intcrdcpendent participant in a cooperative milieu where the lcarner’scontributions arc valued and used, the individual learner is potentially reLvardct1 by havinghis own subjectivc expectations antl decisions informed and guided by others. In a contextwhere different contributions and differential learning are positively encouraged, the learneris allowed to dcpend on othcr learners and on the tcacher kvhen the need ariscs, and alsoenablctl to be independent at appropriate moments of the learning. He can feel free toexploit independent strategies in order to learn, to maintain and dcvclop personal affectivemotivations for learning, and to decide on different routes and mcans which liecomeavailable during learning. The paradox here, of course, is that genuine independence arisesonly to the extent that it is intcrdcpendently granted anti interdependently aLearning seen as totally a personal and subjective matter is seeing learning in a vacuum;indeed w-c may wonder whether such learning is ever possible.Lcarners also have an important monitoring role in addition to the degrec of monitoringwhich they may apply subjectivcly to their own learning. The learner can be a provider offeedback to others concerning his own interpretation of the specific purposcs of‘ thecurriculum, and the appropriateness of methodology to his own learning experiences andachievemcnts. In expression antl negotiation, the learner adopts the dual role of being, first,a potential teacher for other learners and, second, an informant to the teacher concerninghis own learning progress. In this latter role, the learner can offer the teacher and otherlearners a source for new directions in the learning-teaching process of thc group.Essentiall?, a communicative methodology would allow both the teacher and thc learner tobc interdependent participants in a communicative proccss of learning and teaching.6 What is the role of content within a communicativemethodology?Language teaching curricula have often been traditionally defined by their content. Suchcontent has itself been dcrivetl from a target repertoire in tcrms of somr sclccted invcntoriesof items analysed prior to thc commenccment of the teaching-learning proccss and oftenacting as predeterminants of it. Similarly, sets of formal items takcn from an analyticgrammar of the language, or sets of ‘functions’ taken from some list of semantic categories,

20 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINhave been linked to themes antl topics tlccmctl in advance to lie appropriate to theexpectations of the particular learners.Communicative curricula, on the other hand, do not look exclusively to a selectcdtarget repertoire as a specifier of curriculum content, for a number of reasons. First, theemphasis on the process of Iiringing ccrtain basic abilitics to bear on the dynamicconventions of communication prccludcs any specification of content in terms of a staticinvcntory of language items grammatical or ‘functional’ ~ to be learned in some prescribedway. Sccond, the central concern for the development and refincmcnt of underlyingcompctcncc as a basis for a sclcctcd target repertoire requires a distinction bctwccn thattarget and an)’ content which could be used as a potential means towards it. Third, theimportance of the curriculum as a means tor the activation and refincmcnt of the processcompetcnccs of different learners IiresupIioses differentiation, ongoing change, antl onlyshort-term prcdictability in \\.hat may lie appropriatc contcnt.The communicative curriculum \vould place contcnt within methodology and provideit with the role of servant to the Icarning-tcaching process. Thus, content would notnecessarily be prescrihcd by purposes but selected and organisctl within the communicativeand differentiated prucess by learners and teachers as participants in that process.Therefore,the learner would usc the content 01’ the curriculum as the ‘carrier’ of his processcompetence and as the provider of opportunities lor communicativc cxpcricnces through\vhich personal routes may lie sclcctcd antl explored as a means to the ultimate targetcompetence.From this concern with mcans rather than ends mith the process of learning-teachingrather than with the product ~ the communicative curriculum will adopt critcria for theselection and organisation 01’ content which will be suliject to, and defined by,communicative learning and teaching. ’l’hc content of any curriculum can lie selected andorganised on the basis of some adolited criteria, and th criteria will influence five basicaspects of thc content: its focus, its scqucncc, its subdivision (or hrcakdo\vn), its continuity,and its direction (or routing). What arc the critcria for the selection and organisation ofcontent within the communicative curriculum?(a) FocusContent within communicative mcthotlology is likely to focus upon knowledge ~ bothcognitive and affective ~ \vhich is personally significant to the Icarncr. Such knowledgen-ould 1)c placcd in an interpersonal context hvhich can motivate personal and jointnegotiation through the provision of authcntic and ~irolilcm-posing tcxts. If content is to bcsensitive to thc process of learning antl to the interpersonal concerns of the group, it needsto reflect and support the integration of language hvith other forms of human experienceand hehaviour.(/I,Scy1renceIf \vc accept that the communicative process requires that \vc deal with dynamic and creativcconvcntions, \vc cannot assume that any stcii-by-stcp or cumulativc sequcnce of content\vi11 ncccssarily be appropriatc. In learning, the various antl changing routes of the lcarncrscrucially affect any orticring of content, so that scqucncing derives from the state frhe learnersrather than from the implicit ‘logic’ of the content itsclf. It may be naive to assume thatwhat may he ‘simple’ for any one learner is likely to be ‘simplc’ for all learners. Sequencingin communicative content is therefore likely to be a cyclic process where learners arecontinuall? developing related lrameworks or aggregations of knowledge and ability use,

ideational,andA COMMUNICATIVE CURRICULUM 21rather than accumulating separalde blocks of ‘static’ knowledge or a sequence of ordcrctlskills.Thus, content hccomcs something \vhich lcarncrs move into antl out from, antl towhich they return in a process of finer analysis and rcfincd synthesis. Curriculum dcsigntmcannot, thcrcforc, predict \z ith any certainty thr ‘levels’ of contcnt on which learner5 \vi11tlecitlc to cvolve their own sequencing in learning. [. . .](c) SuhtlrvisionTraditionally content has been subdividetl into scrialisrd categories of structurrs or‘functions’. A communicative view of content precludes this fragmentation and argucs forsubdivision in terms of whole frameworks hvhercin there is interaction Iictwern all thevarious componcnts of the knowlcdge system ~ interpersonal antl textual ~all the abilitics involved in using such kno\vlcdge. Content would he subdivided or brokendown in terms of activities and tasks to he undertaken, whcrcin both knowledge and abilitieswould lir engaged in the learners’ communication and mctacommunication. ‘l‘he variousactivities antl tasks would be related liy sharing a holistic ‘core’ of knowledge and abilities.So, we \vould not be concerned with ‘units’ of contcnt, hut with ‘units’ of activity \vhichgencratc communication and metacommunication.((1) Con t in u IThc need to providc continuity for the learner has, in the past, bccn liased upon contcnt.Within a communicative methodology, continuity can he identified within at least four areas.First, continuity can reside in the activities and the tasks within cach activity; antl from oneactivity to another and from one task to another. Second, continuity potentially resideswithin communicativc acts during the learning and teaching: either at the ‘macro’ level interms of the \\.hole lesson and its ‘micro’ sequenccs of negotiation, or lvithin the structureof discourse in terms ofthc ‘macro’ communicati\-c act with its ohvn coherent scqucncv ofuttcrances.Thirt1, continuity is provided through the ideational system. At thc ‘macro’ levelthe learner may have access to continuity of theme, VI hilc at the ‘micro’ level the learnercan have access to conceptual or notional continuity Ideational continuity is rcalised througha rcfincment of tcxtual kno\vlcdgc ~ the rcfincment of a concept, for example, can implya rcfinernent of its linguistic cxprcssion, and vice-versa. Fourth, antl finally, continuity canreside bvithin a skills repertoirc or a cycle of skill-use during an activity. I;or examplc, thcrecould he a progression from reading to notr-taking to speaking for the achic\cment of aparticular activity. A communicativc mcthotlology would exploit cach of these areas ofcontinuity as clusters of potential continuitics, rather than cxploit any one alone. All can l xinherent in a single activity.‘I‘hcsr kinds of continuity offcr two important advantagcs.‘I‘hcycan servc the full proccss competcnccs of Iearncrs kno~vlt~lgc systems antl abilities ~ antlthey can allo\v for differentiation. Learners need to bc cnaliletl to scck and achic\-c theirown continuity and, thcrcfore, the criteria for their onm progress. In thc process ofaccomplishing some immediate activity, lcarners will impose their o\vn personal andinterpersonal order and continuity upon that acti\ it?, the communication lzhich the activitygenerates, the interpersonal, ideational and textual data which they act upon, antl on thcskills they nccd to use in thc activity’s achievemcnt. As a result, the progrcssivc refinemcntof the learner’s om n pro compctcnce can provide an overall lcurning continuity. Oncethe tcacher can acccpt that cach of thew areas provides potential continuity for differentto he a problem if different learners pursue scvcral routcs or progress atdifferent rates.

incould1 or22 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLIN(e) DirectionTraditionally, learners have been expected to follow the direction implicit in someprescribed content. Typically an emphasis on content led the learner from the beginning,through the middle, to the end. From \\.hat has been indicated so far, a communicativemethodology would not cxxploit contrnt as somc’ lire-drtcrminctl route with specific entryand exit points. In a communicative methodology, content ceases to become some externalcontrol over learning-teaching proerdurw. Choosing directions becomes a part of thecurriculum itself, and involvcs negotiation bct\vccn lcarncrs and learners, learners andteachers, and learners and text. Who or what directs content becomes a justification forcommunication about the selection and organisation of content with methodology, andabout the various routes to he adoptctll>y thc learners through any agreed content. Contentcan be predicted within methodology only to the extent that it serves the communicativelearning process of thc participants in the group. It might \vcll be that the teacher, innegotiation with learners, will ~iroposc the adoption of aspects of the target rcpcrtoirc asappropriate content. Ilo\vcvcr, the teacher \vould recognisc that thc ccntral objective ofdeveloping underlying communicative knowledge antl atiilities can lic achirange of alternative content, not nccesrarib. including aspects of the target repertoire. Such*carrier’ content can tic as tli\crsc as the different routes learners may take towards acommon target: perhaps contcnt can bc more various antl morc varialile. Also, the teacher\voultl remain frer to build upon the contriliutions of learners their initial competencesand expectations and exploit thc inevitably different \vays in which learners may attainthe ultimate target. [. . .]7 How is the curriculum process to be evaluated?Thc communicativc curriculum insists that c,valuation is a highly significant part ofcommunicativc intcraction itscll. Wc judge ‘grammaticalit?’, ‘appropriatcncss’, ‘intelligibility’, and ‘cohcrrncc’ in communicative performance on the basis of shared, negotiated,antl changing convcntions. Evaluation ivithin the curriculum can exploit this ‘judging’clement of everyclay communicative hchaviour in the asscssmcnt of learners’ communicationantl mctacommunication.‘I‘hc highlj evaluative aspect of communication can be atloptcd asthc evaluation proccdurc of thc curriculum. If so, the csscntially intcrsuhjectivc nature ofevaluation can be seen as a strong point rather than, possilily, a Lveakncss.How might lve evaluate lcarncr progrc Evaluation of oncsclf, cvaluation of others,and evaluation of self‘ by others is intersuhjcctivc. In this xvay, evaluation need not beregarded as external to the purposcs of the curriculum or external to the actual process oflearning and teaching. In rccognising that relative su or failure in the sharing of meaning,or in the achievement of somc particular task, is most often an intcrsubjccti\-c matter, thecommunicative curriculum ~vould rely on shared and negotiated evaluation. Criteria foreventual succcss ~ some particular task ~ lie initially ncg ctl, achirvemcnt ofthe task could be rclatctl to thcsc agreed criteria, antl degrees of su failure could hethemselves further negotiated on thc tiasis of the original criteria. Evaluatiyr criteria,therefore, \vould be rstahlishcd antl applied in a thrcc-stage process: (i) What might ‘success’mean? (ii) Is the learner’s perhrmancc of the task succcssful! (iii) If so, hom succcssful isit? Each stage \vould lit, a matter for communication. Instcad of the teacher hcing obligedto teach toxvards somc cxtcrnally imposed criteria ~ manifested most often by some externalexamination or stantlartlisetl test ~ he can exploit the interpretation of these external orstandardiscd criteria as part of the joint negotiation within the classroom. The group’s

~ cannotA COMMUNICATIVE CURRICULUM 23discovery of the critcria inherent in such cnd-of-course or summativc assessment \loultl lieone means for the establishment of the group’s own negotiatcd criteria and, crucially, forthe sharing of responsibilities during the learning-teaching process.In a communicative curriculum we are dealing with an intcrdcpcntlencc of thecurriculum components of purposes, methodology, antl evaluation. It follolvs that anyevaluation within the curriculum also involves an evaluation of the curriculum itself. Anyjoint negotiation among the various participants within the curriculum may obviously dealwith the initial purposes and ongoing methodology which have been adopted. Indeed,communicative evaluation may wcll lcad to adaptation of initial purposes, of methodology,antl of the agreed criteria of c\-aluation themselves. Evaluation within and of the curriculumcan lie a pokverful and guiding force. Judgements are a crucial part of kno\vlcdgc, Icarning,and any educational process. 1 3 applying ~ judgements to the curriculum itself, evaluation bythe uscrs ofthat curriculum can be brought into the classroom in an immediate and practicalsense. Once within the classroom, evaluation can be made to scrvc as a basis for new. of teaching and learning.nicative use of evaluation will lcad towards an emphasis on firmuriveor ongoing evaluation, rather than summativc or cntl-of-course e\duation Ivhich maybe based on some prcscribetl criteria. That is, it can shape antl guide learning and guidedecisions within thr curriculum process. Any shared antl negotiatcd cvaluation within theclassroom will generate potentially formative feedback for and Iwtween learners antlbetween lcarncrs and the teacher. Formative evaluation may not only indicate the relativesuccesses and failures of both learner and curriculum, it can also indicate new and differentdirections in which both can mo\e and dcvclop. [. . .]This placing of evaluation within the communicative process as a formative activity initself docs not necessarily invalitlatc thc place of summativc cvaluation. Summativcevaluation becomes valuable if it can reveal the learners’ relative achi ment ola particulartarget repcrtoirc. I Iowever, we have already proposed that any target repertoire needs tobe seen as the tip of an iceberg. Therefore, an essential requircmcnt on any summativcevaluation \rould be that it can adcquately account for the learner’s progress in therefinement of a particular underlying competence ~ the communicative knowledge antlaliilities which provide the capociy for the use of a target repertoire. Summativc evaluation,in other words, needs to be sensitive to differential competences which may undcrlie somecommon target. As such, summative evaluation within a communicative curriculum ncctlsto focus on the assessment of the learner’s developing communicative knowlcdgc andaliilities as well as on his actual pcrformance \\ ithin the target rcpertoire. [. . . ITherefore, the essential characteristics of evaluation within a communicative curriculumLvould be that such evaluation is itself incorporated within the communicatily process ofteaching and learning, that it serves the dual role of evaluating lcarncr progress and theongoing curriculum, and that it is likely to hc formative in the achievement ofthis dual role.8 Achieving communicative language teachingWe cmphasised at the outset of this paper that any curriculum framework for languageteaching and learning nccrssarily in\ olvcs designers, materials writers, teachers and learnci-sin a process of relating the three components of purpose’, mcthodology and evaluation. Evenso, we need to ackno\vlcdgc that any curriculum ~~ including a communicati\c curriculumstrictly be designed as a whole from the start. We can only deduce and proposethc principles on which a varicty of communicative curricula may bc based. Any curriculumis a personal and social arcna. A communicative curriculum in particular, with its emphasis

24 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINon the learning and teaching of communication, highlights a communicative processwhereby the intcrrclating curriculum components arc themselves open to negotiation andchange.From this it follows that the communicative curriculum no more than any other -can ncvcr bc one uniquelp itlcntifialde language teaching curriculum. In a real sense thcrccan lie no such thing as an ideal antl uniquclp applicablc language teaching curriculum sinceany realisation of the curriculum must rctlcct a realistic analysis ofthe actual situation withinwhich the language teaching will take place. 'lh cope cvith this requirement of appropriatenessto situation, thc communicative curriculum has to be proposcd as a flcxihlc andpractical set of hasic principles 1% hich underlie a \vholc range of potential communicativecurricula. It is this set of principles which \vc have tried to present in this paper, in theknowledge that such proposals need to Iic translated into action in thc classroom in ortlcrto test their obvn valitlity.This is, after all, the only means by which curriculum theory andpractice can develop. Even though thc curriculum dcsigncr may haw takcn account of theactual language teaching situation, hc has to rccognise that from tlcsign to implementationis itsclf a communicative proc . J. M. Stephens (1967) idcntificd this process when hesaid:The curricular rcforms emanating from the conference room \vi11 be cffcctive onlyinsofar as they become incorporated into the concerns that the teacher is led toexpress. Any statcmcnts or decisions coming from thc curriculum committee will notlie transportctl intact into the li of pupils. Such statements must work through acomplex chain of interactions. 'l'he original statements of the committee will act asstimuli for one set of pcoplc such as sulijcct-matter super\-isors.Thcse pcoplc, in turn,\vi11 react to the stimuli, possililp mcrclp mirroring \vhat they rcceive, more likely,incorporating much of themselves into thc reaction. Their reactions will then act asstimuli for a second sct ofp~oplc \z ho cvill also rcact in their mz-n way. After a numberof such intermctliary transactions somconc, thc tcachcr, will apply some stimuli tothe pupil himself.(pp. 12-13)While Stephens, in talking about stimuli, docs not cmphasisc transactions as a t\vo-\vayprocess, he clcarlp implies that the translation from principlcs through design toimplementation is most oftcn a proc ' ofi-cintcrprctation of the curriculum, and a processof negotiation lietcvccn the curriculum antl its users. If adopted lcithin the design andimplementation procedure, the conditions or minimal rcquircmcnts on any communicativecurriculum must take account of those situational constraints which arc unchangeable.However, such minimal requirements should also serve as the gcncral criteria against whichany situational constraints \vi11 lie tcstctl in order to assess lvhcthcr or not the constraint isgenuincly immutal)lc or lvhcthcr it may I>c ovt~rconit~.If a curriculum Inscd upon the principles \vhich \vc havc examined here is notimplementalilc \Tithin a particular situation, then it tnay lie that a gcnuincly communicativecurriculum is simply not ialdc. It may Iic the caw that curriculum dc-signers antl teachersin such a situation nccd to consider lvhcthcr thc achicv,mcnt of language learning ascommunicution is appropi-iate.Communicative curricula ncctl through timc and according to situation - to lie openand suhjcct to ongoing developments in theory, research, antl practical classroomexperience. Communicative curricula arc essentially the means of capturing variability.Variability \vi11 exist in selected purposes, methods, and evaluation procedures, hut

A COMMUNICATIVE CURRICULUM 25variability must also be seen as inhrrrnt in human communication and in the ways it isvariously achieved 11y different learners antl teachers. Thc classroom ~ its socialhologicalreality, its proccdurcs and activities is potentially a communicative cnvironmcntwhcrc the effort to pull together such variability is undcrtakcn.Thc learning-teachingprocess in the classroom is the meeting-point of all curriculum components anti it is thrplace \z here their coherence is continually tested. The learning-teaching process in theclassroom is also thc catalyst for thc dcvclopmcnt and rcfincmcnt of thosc minimalrcquircmcnts which will underlie future curricula.Notes‘Curriculum’ can he distinguished from ‘ayllabus’ in that a syllabus is typically aspecification of the content of’ the teaching anti learning antl the organisation antlsequencing of the content. Content and its organisation is subsumed n ithin a curriculumas part of methotlolog! (Section 6 of this paper). A syllabus is thcrcforc only part of thcoverall curriculum M ithin which it operates. For interesting discussions of curriculumtheory antl ticsign scc, inter ulio, Iavton, 1973; Stenhouse, 1975; Colby et ul., 1975.This ncgotiativc interaction u itliin the learner hctwccn prior kno\vlctigc and the ncnlearninghas Iwcn a conccrn \vi hology for many years. See, [or rxamplc, Piagct(1953), Bruner (197 3), and Nc’l‘his ‘process competcncc’ ischanging antl de\ eloping communicatil ckno\vletigc and abilities as learner moves from initial compctcncc ton.ards the targetcompetence. It is partly I- alcd through a series of ‘Interlanguagcs’ (Sclinkcr, 1972,Taronc, 1977, Corder, 1978).ReferencesBernstein, B., 1971. Clciss, Codes unci Control, Volume 1 : Theoretical Stutlies tori-urds u Sociolo

26 MICHAEL P. BREEN AND CHRISTOPHER N. CANDLINStephens, J. M., 1967. The Process ?f Schooling: A P.y.chological Examination. New York: Holt,Rinehart &Winston.Tarone, E., 1977. ‘Conscious communication stratcgics in inter-language: a progress rcport’ .Paper prescntcd at the 1 1 thTtSOL Convention, Miami, F1. 1977.

Chapter 2David Nunan and Clarice LambMANAGING THE LEARNING PROCESS“I dunno,” Jimmy said, “I forget what I was taught. I onl) remember what I’ie learnt.”(Patrick W hitc)You are glen the experiences 4ou need to understand thc norld.(Paulo Coelho)IntroductionHE DECISIONS THAT TEACHERS ARE REQUIRED tomakc duringthcT instructional proccss are all driven by the nature of the program, the goals ofinstruction, and the necds of the individual learners. It is therefore critical for us to considcrthcsc issues before turning to thc management of the learning proccss in the classroom.[. ‘ .IIn this chapter we cover the following issucs and concepts:Setting the conreit and cle_f;ning term.s key terms defined arc “learnci--ccntci-c-dnC‘ss,”“cxperiential learning,” “humanism,” “learning-centeredness,” “communicativelanguage tcaching,” “high-structure and low-structure teaching”Curriculum processes the scope of curriculum development and the importancc ofcurriculum development for the managcmcnt of learningNeeds analysis definition and examples of needs analysisSetting goals and objectives from learner needs to learning goals, illustration of goals antlobjectives, how clcarly stated goals and objectives provide a sound basis for managingthc lcarning processSetting the context and defining termsI. . .ILearner-centerednessThe concept of learner-crnteredness has been invoked with increasing frequency in rcccntyears. What docs thc tcrm mean? Likc many widely used terms, it probably means ratherdifferent things to different people (Nunan antl Brindley 1986). For us, lcarner-centcrcdclassrooms are those in which learners are actively involved in their o\vn learning processcs.

28 DAVID NUNAN AND CLARICE LAMBThe extent to mhich it 15 possil~lc or tlcsiratilc tor learners to lie in\ol\cd in their omnlcarning nil1 olniouslq \ar) trom context to context (and, intlccd, from Icarncr to learner).If learners are to lcarn anJthing at all, ho\zc\cr, ultimatel! thcy ha\e to do the learning forthemsel\ es Thus it 15 a truism to sa! that the> \hould I,c in\ ol\ et1 in their OM n learning. Inan ideal learning-centcrcd contcxt, not on14 \\ill deciwms about M hat to lcarn antl hov tolearn be made nith reference to the lcarnci s, Init the learner5 themsel\cs will be iniohedin the decision-making pi-oc tach clcmcnt in the curriculum proce5s will inbolkc thelearner, aslalilc 2 1 .;ho\zsTuhlc 2. l Learner roles in a learner-ccntcrctl curriculumCurriculum ctagcRole qflecirneiPlanningImplementationAsscs\mcnt antl c\aluationI earners arc consultcd on \\.hat thcy Lvaiit tolearn antl ho\v thcy want to go ahout learning.this process. Lcarncrs arc in\olvctl in setting,monitoring, antl modifying the goals andobjccti\ cs of the programs being dcsignctl forthcm.I cai-ncrs' language skills tlof ncctls analysis facilitatesIcarncrs actively using ant1 reflecting on thelanguage insitlc and outsitlc the classroom. Theyare also involved in modifying antl crcating theiro\vn Icarning tasks and language data.Lcarncrs monitor ant1 a.The! arc also activclv involvrtl in the evaluationantl modification of tcaching and Icarning duringthc course and aftcr it has hccn completed.Thc philosophy of Icarner-ccntcreclncss has strong links \vith experiential Icarning,humanistic psychology and task-liasrtl language tcaching. Thcsc links arc evident in thefollowing quotes:[A Icarnci--centered] curriculum \vi11 contain similar clcmcnts to thosc contained intraditional curriculum tlcvclopment, that is, planning (including needs analysis, goaland objective setting), implementation (including methodology antl materialsdcvclopment) antl evaluation (see for cxamplc Hunkins 1980). However, the keydifferrncc lietween learner-ccntrctl antl traditional curriculum tlcvclopment is that,in the formcr, the curriculum is a collaborativc cf'fort between tcachcrs and learncrs,since learners arc closcly involved in the decision-making process regarding thecontent of the curriculum antl how it is taught. This change in oricntation has majorpractical implications for the entire curriculum pro ', since a negotiated curriculumcannot be introduced antl managed in the same \ray as one which is prescribed hy theteacher or tcaching institutions. In particular, it places the liurden for all aspects ofcurriculum de\ elopmcnt on thc teacher.(Nunan 1988: 2)

MANAGING THE LEARNING PROCESS 29The proponents of humanistic education have broadened our concept of lcarning byemphasizing that meaningful learning has to he self-initiated. Even if the stimuluscomcs from outsidc, the sense of discovery, ho\vever, and the motivation which that\,rings has to come from inside drivrn hy the basic human dcsirc for self-realization,well-bcing and growth. . . . [I]. terms of personal and interpersonal compctcncc the-oriented classroom revolves around issues of risk and urity, cooperationand competition, self-dircct~dness and other-directedness; antl meaningful andmeaningless activities. We ha\,e also tried to make clear that “teachers \vho claim it isnot their job to takc these phenomena into account may miss out on some ofthc mostessential ingredients in the management of successful learning” (Undcrhill 1989,12. 252).(Lcgutkc andThomas 1991 : 269)Wt. can scc from these extracts that learnei--centeredncss is stronglj rooted in traditionsdcri\cd from general education Our \ iem is that language pc(lagog> nerd\ to drav on itsgeneral educational roots for sustenancc, 1% hich it has not not al\z a) s done. In fact, somelanguage programs seem to ha\ e suffcrcd an “educational 11) pass ”Learning-centerednessTahlc 2 1 , M hich scts out the role of the learner in relation to curriculum planning,implementation and elaluation, represents the ideal As tcac hers and course dcsigncrs, mehac heen in relatncly fc\z situations in \\ hich leal ncrs trom an cad! stage in the learningprocess hac been able to make criticall? infoi-mcd decision\ about \z hat to lcai n and ho\\to lcarn In our experience, learners nccd to be s\ sternaticall) taught the skills ncetletl toimplement a learner (cntcrcd approach to petlagog) In other words, language programsshould ha\e t\\ in goals language content goals antl learning piocc\s goals \uch a program\\e \zould charactcri7c as being “learning centered ” UJ tcmaticall) educating learnersabout \\hat it means to lie a learner, Icarncrs reach a point \\here the) are ahlc to inakcinformed ticcision, aliout 1% hat thri nant to lcarn and ho\z thc! mant to lcarn It is at thi,point that a truly learner-ccmtcrcd curriculum can lie implemented Lcai ning ccntci chessI\ thus designed to lead to learner-centerednrssThe pre\ ious discussion undci lines the fact that learner ccntci cdncss is not an all-ornothingprocess Rather it is a continuum trom rclati\cl) lcss to relatnel! more learnercentered Nunan (1 99511) has captured this continuum in the tollow ing tables, which shobzthat learner ccntcrctlness can be implementcd at a numlicr of different lc\cls The tablesalso illustrate some ot the practical steps that can he taken in implementing a lcai ncioriented approach to in\tructionTable 2.2 relates to the experiential contcnt domain It demonstrates that, all otherthings licing equal, a classroom in \z hich learners are made am arc of thc pedagogical goalsand content of instruction is morc lcarncr-centered than one in \z hich goals antl contentarc left implicit We \\auld argue that all learners should, in thc hrst instancc, IIC alerted togoalr and content In collecting data for this book me nere surprised at hom tntrequcntl)this step happened Ho\ze\er, wc \zould go furthcr, and arguc that it is just a first step alonga path that, gi\en the appropriate context and types of lcarncrs, could takc the lcarnei 5through a gradual learning process in mhich the) made selections trom a range ofaltcrnatii es, modihed and adapted goals and content, created their 0x2 n goals and selectedtheir own cxpericntial content area, antl finall! mo\ cd be> ond the classroom itself (korpractical descriptions antl illustrations of thew processes, see Nunan 19951) ) Ho\z far one

30 DAVID NUNAN AND CLARICE LAMBchooscs to movc along the continuurn depends on onc’s learners and the context andenvironment of the instructional process.Table 2.3 shoxvs how the continuum can apply to the learning process domain. Onceagain, we see that learner-ccntcrcdnws is not an all-or-nothing process, but can licimplemented in a series of gradual steps.Tuble 2.2 Learner-ccntcredncss in the experiential content domainLevel Leurner act ion Glo,r1 Abvarcncss 1.carnc.n arc matlc aware of the pdagogical goals antlcontent of the course.2 In\ olvcmcnt Lcarncrs arc involved in sclccting thcir o\vn goals andol>jcctivcs from a range of altcrnati\-cs on offcr.3 Inter\ cntionLcarncrs arc in\ olvctl in modifying and adapting theqoals and content of the learning program.4 Crcation Lcarncrs crcatc thcir o\vn goals ancl ohjcctivcs.5 Transccntlcncc I.carncrs go Ixyond thc classroom and make linkst)ct\\ccn the contcnt of the classroom antl thc \vorldbcvond thc classroom,Tuble 2.3 Lcarncr~ccntcrcdncss in the lcarninc process domainI.cve1 Lecirner action Gl0s.s1 A\varcncss L,carncrs itlcntify stratcg! implications 01’ pcdagogicaltasks antl idcntil) thcir o\vn prctcrrctl learningstj Ics/stratcgics.2 Involvcnicnt Lcarncrs makc choices among a range of options.3 Intervention Lcarncrs niodif) /adapt tasks4 Crcation Lcarncrs crcatc thcir own tasks5 Transccndcncc Lcarncrs \>cconie teachers antl rescarchcrsCommunicative language teachingCommunicatiw language teaching cmcrgcd from a numlicr of disparatc sources. Duringthc 1970s and 1980s applied linguists antl languagc educators began to re-evaluatepedagogical practice in the light of changrtl views on the nature of language and learning,and the role of teachers and lcarners in thc light of these changing vicws. The contrastlietween what for want ofhettcr terms we have called “traditionalism,” and communicativelanguage teaching (CLT), is shown in Tablc 2.4 in relation to a numlicr of key variableswithin the curriculum. The table prcscnts contrasts in relation to theories of language andlearning, and in relation to objectives, syllabus, classroom activities and the roles of learners,teachers antl materials. The vicws illustrated represent points on a continuum, rather than

interaction.MANAGING THE LEARNING PROCESS 31cxclusive categories, and most teachcrs \vi11 mow back and forth along the continuum inresponse to the nerds of the students and thc overall contcxt in which they arc teaching.The truth is that language is, at one and the same time, hoth a system of rule-govcrnetlstructurcs and a system for the expression of meaning. Learning is a matter of habitformation as \vel1 as a proccss of activation through the deployment of' communicative tasks.The challenge for the teacher, the textbook writer and the curriculum developer is to shobvhow the rule-governed structures enable the language user to make meanings.Table 2.3 Changing vic\vs on thc nature of language antl learning: Traditionalism and CLTTeac h I ng Traditionalism Commun ica ti re lung uagcTheory of languageThcor! of learningOhjectii e\Language is a svstcm of rule- Language is a s! stem for thegovcrnctl structures hierarchically expression of meaning: primar)arrangctl. function ~Habit formation; skills arc leal-nctl Acti\ itics inwlving realmore effectively if oral precedes communication; carrying out\vrittcn; analogy not anal! meaningful tasks antl usinglanguage that is meaningful tothe Icarncr promote learning.Control of' thc structures of sound, Olijcctiws \vi11 reflect thc nccdsform and order, master) oicr of the learner; they \z ill includemhol\ ot the language; goal lunctional skills as \vel1 asnatil c spcakcr master!.linguistics objectives.SvllahusActi] iticsRolc ol lcarncrRolc of tcac hcrRolc of materialsGradctl syllabus of phonology,morphology, antl syntax.Contrastiw analysis.Dialogues and drills; repetitionand memorization; patternpi-acticc.Organisms that can he tlircctctl bqskilled training tcchniqucs toproduce correct responses.Central antl active; tcachcrdominatctlmethod. I'ro\ itlcsmo(lcl; controls tlircction antlpacc.Primarily tcachcr oricntctl. Tapesand visuals; language lab oftcnused.Will includc some or all of thefollo\ving: structures, functions,notions, thcmcs antl tayks.Ordering \vi11 Iic guided bylcar tier ne et l s .Engage learners incommunication; in\ol\ cproccsscs such as informationsharing, negotiation of meaningand interaction.Learner as ncgotiator,interactor, gi\ ing as nell astaking.Facilitator of thecommunication process, ncctl\analyst, counselor, processmanager.Primary role of promotingcommunicative language use;task based, authcntic matcrials

32 DAVID NUNAN AND CLARICE LAMBWe do not Ilclievc that many classrooms can he defined exclusively in terms of aparticular methodology. Whether a classroom is characterized as “traditional” or“communicativc” is therefore determined by the relative emphasis and degrcc to which theviews listed in the table underpin \\.hat happens in the classroom rather than on the cxclusiwadherencc to one set of views to thc exclusion of an); other. Thc difference lies, not in therigid adherence to onc particular ap[iroach rather than another, but in the basic orientation.Somc teachers operatc out of a traditional paradigm, making occasional forays into CL2T,and for others it is the other \.ray around. In the ESI. and EFIJ classrooms vvc have workedin and studied in rcccnt !cars, the Ix-cvailing t rcd has hcrn to\vard CLT, although by nomeans exclusively so.High- and low-structure teachingThe insight that communication \vas an intcgratetl process rather than a set of discretelearning outcomes created a dilcmma for language cducation. It meant that the destination(functioning in another language) and thc routc (attcmpting to learn thc target language)moved much closer togethcr, antl, in some instances (for examplr, in role plays andsimulations), became indistinguisha1,lc.. The challenge for curriculum devclopcrs, syllahusdesigners, materials writers and classroom teachers revolved around decisions associatedwith thc movements 1,ctween points on the continua set out in the tables in the precedingsection. Questions such as the follo\ving therefore appeared Ivith increasing frequencyin teacher-training kvorkshops: Ho\v do I integrate “traditional” excrcises, such as drills,controlled conversations antl the like, Lvith communicative tasks such as discussions, tlebatcs,role plays, etc.? Ho\v do I manage decision making and the learning proclassroom sessions devoted to communicative tasks Lvhich, by definition, require mc to handover substantial amounts of tlccision-making pvcr antl control to the Icarners? How canI equip learners thcmselvcs \vith the skills thcy \vi11 nccd to makc tlccisions \viscly and toembrace po\vcr cffccti\-ely?For some individuals the solution la? in wjccting the changing vie\vs along with theirinconvenient pedagogical implications. Others lvcnt to the oppositc extreme, eschewing“traditional” solutions to their materials clcvelopmcnt antl language-teaching challenges. Inmost contcxts, hou cr, a more Iialanccd \icw prcvailctl.For some time after thc rise of CL‘I; thc status of grammar in the curriculum \vasrather uncertain. Somc linguists maintained that it \vas not negrammar, that the abilit? to LISC a second language (“knowing how”) \vould developautomatically if the learner \vcrc required to focus on meaning in the proccss of usingthe language to communicatc. In rcccnt ?cars, this vicw has come under seriouschallenge, and it now seems to 1~ widely accepted that there is value in classroomtasks kvhich require lcarncrs to focus on form. It is also accepted that grammar is anessential resource in using languagc communicatively.(Nunan 1989: 13)In educational terms, a useful \.ray of viewing this emerging dilemma in language cducationis in terms of high- antl low-structure tcaching. Iligh-structure tasks arc those in whichteachers have all the pvcr and control. Low-structure tasks are those in which power andcontrol are devolved to the students. We have borrowd the terms “high-structure” and “lowstructure”from Biggs and Telfcr (1 987). They suggest that the successful management ofthc, learning process depcntls on teachers knowing kvhere to locate themselves on the high-

MANAGING THE LEARNING PROCESS 33to Ion-structure continuum in relation to a given task. In a high-structure task, studentsare placed in reactive roles and accorded relatively little choice. In a low-structure context,students haw many options antl maximum autonomy. EIomww-, ~ vc do not equate highstructurewith non-communicative and Ion-structure with communicativc tasks. In ccrtaincommunicative tasks, learners have relatively little freedom of maneuver. Howcvcr, ~ vc dobelieve an association exists bctwecn lolv-structure antl CLT and that the incorporation ofcommunicativc tasks Lvith Ion-structure implications into the classroom increases thtcomplexityof the decision-making process for the teacher.We \vould argue that the kinds of managerial issues that arise and the sorts of decisionsthat teachers arc required to make will be largely driven by the degrcc of structure implied.This concept is illustrated inTablc 2.5, which provides exemplary questions relating to highandloM--structure contexts as these apply to key elemcnts at the levels of curriculumplanning, implemcntation, and evaluation.This schema will be referred to constantly in thepages that follow, as it is one of the key organizational framcworks underpinning the workas a whole. It allows us to deal coherently with the following key managerial questions antlto demonstrate that the answers \.rill vary according to the tlcgrec of structuring called forby the instructional goals guiding the intcraction at that particular time.What aspects of tcachcr talk (direct instruction, feedback, instructions, antl questioningstrategies) facilitate or impair cffcctive learning?What issues nccd to bc taken into consideration in lesson planning and preparation?How can the tcacher most effectively cxploit resources in the classroom?What stratcgics cxist for setting up diffcrcnt modes of classroom interaction, fromteacher-fronted through small group, pair and individual Lvork?What arc the implications of affective attitudes (e.g., motivation, attitude and aptitude)for the effective managcmcnt of learning?What tools, tcchniques, and strategies cxist for the ongoing monitoring and evaluationof classroom interaction and acquisition?(All of these questions can be explored through thc investigative procedures suggestedin Nunan 1990, 1992).The curriculum in outlineImplicit in the foregoing discussion is thc fact that classroom dccision-making antl the effectivemanagement of thc learning process cannot be made without rcfcrcncc to the larger contextwithin which instruction takcs place. The context and cnvironmcnt of the learning proincluding the curriculum plans that should tlri\ e the pedagogical action, arc criticall!important hcrc. In other \vortls, classroom dccisions cannot be made \vithout rcfcrcncc tostructures operating outsidc of the classroom, at the level of the curriculum.Language curriculum development has been greatly influenced by changing vithe naturc of teaching and learning.’l’hesc changing viclvs are reflected in the objectives andcontent of language programs, as well as acti\ities, materials, and tcachcr/lcarner roles.The influence of these diffcrcnt viovs was made clear in the Table 2.4, which contrastedtraditionalism nith CLT.As hve can sec from Table 2.5, communicative language teaching has had a majorinfluence on languagc curriculum dcvclopment. First, curriculum tlcvclopmcnt has hccomcmuch morc complcx. Whereas txvcnty or thirty years ago, thc point of tlcparturc forcurriculum development trntlrd to be restricted to the identification of the Icarncr’s currrntlcvcl of proficiency, \z ith thc dcvelopmcnt of communicativc language tcaching and thcinsight that curricula should reflect learners’ communicative ncetls antl learning prcfcrcnccs,

Table 2. J Curriculum decision-making in high-structure and lo\\ -structure contextsCurricular elements High-structure contexts Lowstructiire contexts.It the planning ctageCourse dc5ignWhat docs the institution tell nic to teach?What arc the managerial decisions entailed in thcteacher’s manual?Ho\v do I tlesign/adapt my o\vn content/goalsltasks?Nectls analysisHo\v can I identify the learning prcfci-cnccs of mystudents?Ho\\ can I in\ol\c my learners in identifying andarticulating thcir o\vn needs?CollegialHo~v can I coopcratc \\-ith collcagucs in Courseplanning?Ho\\ can 1 get the most out of staff meetings?Holy can staff meetings contributc to cffccti\ cplanning?What opportunitics exist hi- team teaching?ResourcesHOM do I manage ujc of act text?Hmv do I niotlifq/adapt the text?Ho\v do I crcatc my olvn resources?Holy do I design split information tasks that \vi11 becffcctirc in mv context?At the implementation rtagcTalk/interactionWhat arc cffectiw strategies for direct instruction?Hoiv do I give feedback on high-structure tasks?What questioning strategies facilitate learnercontributions to low-structure tasks?How do I give feedback in lo\\--structurc tasks?What types of teacher questions maximize studentoutput?

Learner languageH ~ do M I correct learner errors?Hon- can I provide language models in small grouprole plays in xvhich the principal focus is on theexchange of meanings?Learner attitudeHoxv do I deal \r-ith group conflicts?Hmv do I deal with student resistance to learnerinitiated tasks?Group configurationHoxv do I organize controlled practice?Ho\v do I managc teacher-fronted instructioneffectively ?Ho\v do I set up small group learning?What strategies exist for setting communicatix-etasks in Lvhich students Lvork independently?Ar the eraluation stageLearner assessmentSclf-cvaluation of the learning processWhat techniques \vi11 help me to asscss theachic\cment of my learners?How can I help my learners de\-clop effectivetcchniqucs for sclf-assessmcnt?Formal e\ aluationHo\\ can learnen be in\olxcd in proxidlng input tothe e\ aluation procc$s?

36 DAVID NUNAN AND CLARICE LAMBmuch more information about and by learncrs came to be incorporated into the curriculumprocess. Thc other major modification occurred with the emergence of the communicativetask as a central building block within the curriculum. Instead of being designed to teach aparticular lexical, phonological or morphosyntactic point, tasks were designed to reflectlearners’ communicative necds. I ,anguage focus cxerciscs \vere developed as a second-orderactivity.In summary, we can say that curriculum dcvclopmcnt represents a delicate juggling actinvolving the incorporation of information about the learner, about the language, and aboutthe learning process. Language content questions include what are we tcaching, why arewe teaching it, antl when ~ ’ arc e teaching it. Learning process qucstions, which aremethodological in character, include how are \vc arranging the learning environment.Among other things, when we focus on the lcarner, we must ask how well the learner hasdone and how well the curriculum has done in serving the necds of the learner.Wc can relate thcsc key qucstions to each other in terms of the central curricularelements of syllabus design, which has to do Lvith thc selection, sequencing and grading ofcontent; methodology, which is concerned with task selection and sequencing; andassessment and evaluation, which are concerned with determining how well students havedone, as well as evaluating how well the instructional process has met curricular goals.Theserelationships are set out schematically inTahlc 2.6.Table 2. h Kc! curriculum questions, prowdurcs, and arcasQuestion., ProceJirrcs , lrcusContcntWhat?Why?When?I’roccsscsEIO\V?When?OutconicsHo\v \vcll?How cffccti\c?SclcctingJustifyingGradingtnactingSequencingAsscssingE\ aluating11 Methotlolog!Asscssmcntk\ aluationOne vicw of “curriculum” has it that curriculum processes have to do with thcdc\-clopment of tactical plans for action. In this vicw, “curriculum” is taken to refcr tostatements about what should happen in the teaching and lcarning situation. According tothis vie\v, the curriculum specialist’s task ends when the ink is dry on the various documentsthat have been produced to guide teaching and learning. Wc believe that this vicw is simplisticand nai‘vc, that while “curriculum” includes the planning process, it also includes theprocesses of implementation antl evaluation .These three phasc-s are captured in Figure 2.1 .The final point we wish to makc is that the language curriculum should concern itself,not only bvith language content goals, but also with learning pro goals. Learners shouldbe focused on the processes through which lcarning takes place as well as on thc targetlanguage they arc learning. It is our contention that learners cvho haw developed skills inidentifying their own preferred learning skills and strategies \vi11 be more effective languagelearners.

~~~~MANAGING THE LEARNING PROCESS 37Phase I:Phase 11:Phase 111:Planning (initial nccds analysis, goals antl (hjcctives, contcmt, antl process)Implcmrntation (ongoing ncrtls analysis, monitoring, action I-cscarch)E\ aluation (assessmcnt, self-asscssmrnt, program cvaluation)hprc 2. IThrcc phascc or pcrspccti\cs on the curriculum processCurriculum goals I anguage content For example, to tle\rlop theabilit! to olitain goods andscr\ ices in the targetlanguageLcarning processFor cxamplc, to del elopskills in learning hov tolcarnTaskAim To familiarize you \z ith some ofthc key tasks conccrncd nith cui-riculum dcwlopmentand to provide an opportunity for you to relatc thcsc to your on-n tcaching situation.Procctlurc1 . The following list contains somc of the tasks that need to be carried out in the courseof dcsigning and implcmcnting a curriculum. Study the activitics and decide kvhichof them, in relation to a context Ivith which you arc familiar, should be carried outby a teacher, a curriculum specialist, a counsclor, a director of studics, ctc. Write thesedown in thc spaces provided.2. Sclcct those areas for xvhich thc teacher has primary responsibility. What arc somc ofthe decisions that need to be made? Exprcss these as questions.DataIntel-\ iem students ~~Conduct needs anal!iis ~Assign studcnts to class group ~Carry out diagnostic test ~Arress studcnts’ current le\el of EnglishIhagnosc indn idual learning difhculticsIdentify indn itlual lcarning \t!leiSclcct and grade linguistic contcnt (grammar, 1 ocabular!, functions, notion\)-~Select cxpcriential content (topics, thcmcs, situations, settings, etc.) _ _Sct out course goalsWrite performance objecti\csSelect, adapt or develop lcarning tasks and material\Monitor student progrcss ~__

~Student~~~ ~ _________~ ___________~ ~ ~ ~38 DAVID NUNAN AND CLARICE LAMBAsess learning outcomes ~ELaluatc language program(s) ~~In some teaching contexts, teachers will lie rcsponsiblc for all these tasks. In others, theywill have little control. Some of the questions r aid by teachers in rclation to interviews,needs analysis, and assigning students to groups include the follo\ying:intervicu,s Should these he carried out Idorc, during, or after the course hasbegun? Should the learners he forced to respond in the target language? HOW do I getinformation from lo\v-proticiency learners \vhen I don’t spcak their language?A‘eeds ana@s What techniqucs exist for doing nccds analysis? How can the resultinginformation be used for writing course goals and ohjcctivrs? What if my learners haveconflicting needs?dssigning students to groups What criteria, other than proficiency level, can be used toassign students to groups? Is it possible to have diffcrent configurations at differenttimes during thc- teaching day?Needs analysisIn the course of designing a teaching program from scratch or modifying an existing one,it is generally tlesirahlc to collect and interpret (lata about the learners and the institutionalcontext in which they learn.l’his information may Ile collected formally or informally beforethe course and once the course has begun. A variety of different tycs of information canbe collectctl. Such information might include biographical information about the learners,data on the types of communicative tasks that learners might want or nccd to carry out inthe target language, information on the ways in tvhich the learners prcfcr to learn, and soon. A \vide range of information can hc collected through nccds analysis procedures ofvarious kinds, as will be seen in the sample instruments provided in this section. In the initialplanning stages, the extent to which Icarners’ subjcctive nccds can be canvassed dcpendson the range and extent of lcarncrs’ previous cxpcricnccs. (It \vould l ~c unrealistic, forexample, to ask learners lvhcthcr they like to learn through rolc play and simulations ifthcyhave never expericnccd such activities.)In attempting to obtain information from learners, as well as allout learners, additionallimitations and constraints will apply \vith young Icarncrs, or with lowproficiency learnersif the teacher docs not speak the learners’ first language and docs not have the benefit ofbilingual assistants or other first language resources.Rrintllcy (1989) suggests that there arc basically thrcc different approaches to nccdsanalysis. He calls thrse the language proficiency orientation, the psychological/humanisticorientation and the specific purposc orientation. The thrcc approaches arc differentiatedaccording to their educational rationale, the type of information collected, the method ofdata collection and the l~ur~ioses for \vhich the data arc collcctetl.The salient characteristicsofthc three approaches are set out inTable 2.7.In learner-oriented contexts, the types of information requircd and the purposes towhich thc information \vi11 be put will vary somewhat from programs tlevelopctl withoutreference to the learners themselves, and those for which any preliminary analysis will belargely restricted to thc needs of the institution or thc educational system that thecurriculum is intended to serve. Within a second, rather than foreign, language context,Rrindlcy suggests types of information and purposes that are important (scc*Ial~le 2.8).

MANAGING THE LEARNING PROCESS 39Table 2.7 Approaches to needs analysisLungnage proficicnc). orientation P.y chological /humanistic orientation Spec$c purpose orientationEducational rationaleLearners Icarn morc txffectivclyif gi-oupcd according toproficiency.Lcarners learn more effectivelyifini-ol\cci in the learningprocchs.Learners Icarn morc cffecti\ el\ ifcontent is relevant to theirspecific arcas of nrrd/intcrcst.7jy of informationLanguage proficitmcy/languagctlifficul ticsAttitude$, moti\ ation, lrarningTtratcgy pretcrcnccuInformation on nativc speakcruse of language in learners’ targetcommunication situationMethod ?f collectionStandardized forms/tcstsObscr\ ationStantlartlizetl formsObservation, inter\ icws andsurvcysI anguagc analysisSurveys of learners’ pattcrns oflanguage usePurposeSo lcarncrs can be placed ingroups of homogeneouslanguage proficiencySo tcaclicrs can plan languagecontent relevant to lcarncrs’proficirncy Ic~clSo Icarncrs’ individualcharacteristics as learncrs can bc@\,en due considerationSo Icarncrs can bc hclpd to1,ccomr sclf-dirccting by bringinvolved in decision makingabout their learningSo that lcarncrs \vi11 be prescntctlivith language data rclcyant tothrir communication goalsSo motivation will I)c cnhanccdI)? relativeness of languagecontentSource: After Rrindlry 1989: 67 69. Used by pc.rmissionA major purpose for- conducting needs analyses is to categorize and group lcarners.This grouping process facilitates the specification of content and learning procedures thatarc consonant with some aspect of the learner data that has been gathered. Figure 2.2excmplifics some ways in which data can be used for grouping purposes.Setting goals and objectivesIn the contcnt domain, needs analysis provides a basis for setting goals and objectives. Goaland objective setting arc important tasks in most educational contexts, because they pro\ idea rationale for selecting and integrating pedagogical tasks, as well providing a point ofreference for thc decision-making process. Goals arc broad statements that provide gcncralsignposts for course development. Thc following sample goals have hcen extracted from avariety of second and foreign language programs. They are expressed in the 1)roadest possibleterms.To dcvelop sufficient oral and written skills to obtain promotion from unskilled \vorkerto site superyisorTo establish and maintain social relationships through exchanging information, itleas,opinions, attitudcs, feelings, and plans

~To~40 DAVID NUNAN AND CLARICE LAMBTuhle 2.8 T! pes of information required in a Icarncr-centered yystcmLcarncrs’ lift, goalsLanguage goals, communi( ativc net\\ orksantl social rolcsOlljcctivc nccds, patterns of language usc,personal rcsourccs (including time)Imguagc. proticicncy antl language clifficulticsSul>jcctivc ncctls including learning stratcg!prcl‘crcnccs, atlccti\ c nccds, learning ad\it!prcfci-cncrs, pac~of learning, attitudc to\vx(lcorrectionInformation about Irarncrs’ attainnicnt ofobjccti\ csInloi-mation almut tlcvclopnicntal prou~sscsin second language learning, includinglearners’ communicative strategic.;So that tcachcrs have a Iiasis on \vhich to tlctcrmincor prcdict Icai-ncrs’ language goals, communicativenct\\.orks antl social rolesSo lcai-iicrs ilia) Ilc placrd in a group basctl on(‘oninion social roles, antl tcachcrs may makcpi-c>lirninary tlccisions alwut coui-cc contentappropriate to learners’ social rolesSo Icai-ncrs can 1)c grouped according to theiitietd\ and /or intcrc-stsSo Icarnci-s can Iw grouped accoi-ding to thcirlanguage proficicne\So that tcachcrs ma! adapt learning activities toIcarning Ytratcgr prclci-cnccs, indi\ itlual needsSo that thr tcachcr can monitor Iicrformancc andinotlitj progr,im accordingl!So that tcxhci-s can gear Ianguagc content antltiiatci-ids to learners’ stage of tlcvclopmcntSourre: Adapted from Rrintllcy 1984. Used liy pc~riiiission.tlevclop communicati\ kills in ortlcr to acquirc, rccnrd and use intwmation froma variety of aural sourccs‘lh tlcwlop acatlcmic Iistcning skills in order- to cxtract key information fromuniversity lccturcsTo dcvclop Iiasic communicativc skills in ortlcr to olitain basic goods and services asa touristMorc limited goals, couched in functional terms, can Ilc found in tcaching matcrials ofvarious sorts. The follo\ving haw been taken from an intcrmcdiate~lcvcl textbook.In this book you will:Make comparisonsAsk for and give adviccExpress obligationTalk ahout past cxpc,ricnccsExprcss opinions aliout cntcrtainmcnt.(Nunan 1995a)These goal statements arc very general in nature and can encompass numcrous subsidiary. Most curriculum tlocumcnts Iiasctl on a goal and olijcctivcs approach contain a

~Working~~MANAGING THE LEARNING PROCESS 41[ Languagc proticicnc,y protilc12345Students \\ ith oral skills, but \I ith re\\ or no literacy skills in 1-1Students \vho belong in a nc\v arrivals programStudrnts \\ ho rcquirc gcncral support in the mainstrcamStutlcnts with spccific affrcti\ e, language antl communication ncctlaStudents \vho arc approximating nativclikc proficirnc!(Atlaptctl lrom S.A. ESI Guitlrlincs)11 I earning strategy protile1 "Concrete" learners 'l'hcsc learners tend to like gamcs, picturcs, films, vitlro, using cassettes,2talking in pairs antl practicing English outside class.". tna~trcii/"learner.s 'l'hcsc Icarnrrs like to study grammar, stucly English hooks, and rratlne\\ spapus; they also likr to study alonc, find their own mistakes, anti \vork on problemsset by the tcachcr.3 "Cornrnunicarivc" learners Th tudcnts like to learn b! bvatching, listening to nativespeakers, talking to fricnds in English and Lvatching t sion in hglish, using tnglish outot'class in shops, trains, ctc., learning nc\v \vortls hy hcaring them and learning I>)con\ crsations.4 ";I~ithori~-or~~nted" /eurner.s 'I'h learners prclcr the trachcr to explainlikr to ha\c their o\vn tcxtbocrything in a notebook, to study grammar, learn1 ) ~ rratling, and lcarn ne\? \vortls(Atlaptctl from Willing 1988)111 Ilcarning purpose1 Ne\\- arrivals2 English in thr \vorkplacc3 English for further study45English lbr professional cmplo! nicntEnglish for access to \ ocational tl-aining and cmploymcntFigtire 2.2 Three altcrnati\ e wavs of grouping learnerslimited number of goals (perhaps five or six) that prodc a basis for the development ofobjectives. Formal pcrformancc objectives specify kvhat learncrs should lie able to do as aresult of instruction. Formal objcctivcs should contain a perfbrmancc (which sets out whatlearncrs arc to do), conditions (specifying the conditions and circumstances under whichthe learners should perform) antl standards (setting out how \vel1 they should pcrloi-m).The three objcctivcs that follow illustrate the thrcc components of performance, conditions,and standards.in pairs, learnci s \T ill pro\icle cnough information for their pai-tnrr to drautheir famil! trcc The) \z ill pro1 itlc cnough information for a thrcc gcncration famil!trcc to lie dra~n.Students m ill extract and record estimated minimum antl maximum tcmperatui e4from a tapcd radio neather forcca5t Thc) must accuratclj record tour of the ~ I Xregions Cwercd by the forecast.While matching a I idcotaped conxerqation bctneen t\zo natixc speaker5,5tutlcnts v ill

42 DAVID NUNAN AND CLARICE LAMBidentify the various topics discussed and points at which they are changed. All topicsand change points arc to lxx idcntificd.The use of an objectives approach has ken criticized in general education on the groundsthat precise statements of what thr learner should lie able to do at the end of a course issomehow undemocratic and neetllr~ssly restricting on both the student and the teacher.Others argue that such precise specification greatly facilitates other steps in the designprocess. It forces the designcr to he realistic about what learners can achieve and helps guidethe selection of appropriate materials and classroom activitics. It is also an essentialprerequisite for devising appropriate forms of learner assessment.Some years ago, an interesting set of specifications was developed in Australia. Calledthe Australian Language Levels (ALL) guidclines, these specifications were intended to begeneral enough to help materials writers and teachers nm-king in a range of second andforeign languages. The ALL guidclines take as their point of departure a number of broadgoals that are refined into specific goals, as shown inliblc 2.9.You can get some idea from this furthcr example of thc breadth of the goal-settingexercise. You can also see how numerous subsidiary objectivcs could be formulated fromeach of the goal statements. Interestingly, the designers of the ALL guidelines chose to movedirectly from goals to the specification of task or activity types without elaborating detailedsets of objectives. We also have employed this procedure in some of our work. Although wedo not feel it necessary to dcvrlop formal three-part objectives for everything we wish toteach our learners, we do believc that a sample set ofohiectivcs can greatly assist in managingthe learning process. They can be particularly useful in the ongoing monitoring andassessment of the learning process.The latest manifestations of the goals and objective approach to curriculum devrlopmcnthave appcareti in competency statements that attempt to specify what learners shouldbe able to do at different levels. The following arc extracts of core competencies designedfor an adult immigrant program. Once again, you can see they arc formulated in terms ofwhat the learners should be able to do as a result of instruction.Englishfor Stti(+1 .2.Can understand the context of further cducation/training in AustraliaCan utilise a range of learning strategies relevant to further cducationltrainingcontext\3. Can Understand an oral prcscntation relevant to further education/trainingcontexts4. Can negotiate complex/problematic spoken cxchangcs rclated to furthereducational/ training contexts5. Can participate in group discussions relevant to further cducational/trainingcontexts6. Can deliver short oral presentations relevant to further educational /trainingcontcxts[. . .IVocational English1. Can understand the context of work in Australia

~ establish~~ obtain~ obtain~ cognitive~ communicationMANAGING THE LEARNING PROCESS 43Table 2 9 Communi( ation and learning ho\\ -to learn coalsBroad goalSpectf;c godsCommunicationBy participating in acti) itics organized arounduse of the targct language, learners will acquirecommunication skills in the targct language, inorder that they may widen their nct\vorku ofinterpersonal relations, havc direct access toinlorniation and use their language skills forstudy, vocational anti lrisurc-based purposesTo be able to use the targrt language to:and maintain relationships and discusstopics of intcrcst (e.g., through exchange ofinformation, ideas, opinions, attitudes, fcclings,experiences, plans)participate in social interaction rclated tosolving a problem, making arrangements,making decisions with others, and transactingto obtain goods, srrviccs, and publicinformationinformation by scarching for spccihcdctails in a spoken or written trxt and thcnprocess and use the information obtainedinformation by listening to or reading aspoken or Lvrittcn tcxt as a wholc, and thcngive information in spoken or written form(c.g., givr a talk, write an essay or a set 01‘instructions)and use the information obtaincdlisten to, read or vie\\, and respond personallyto a stimulus (c.g., a story, play film, song,poem, picture, play)1.enmin~~hon-to-learnLeamcrs \vi11 be able to take a grobvingresponsibility for the management of their ownlearning so that they learn how to Irarn, anti houto lcarn a languageTo develop:processing skills (to enahlc them tounderstand values, attitudes and feelings toIirocc’ss information, and to think and respondcrcativcly)Icarning-ho\\ -to-lcarn skillsstratcgirs (to enablc them tosustain communication in thc target languagr)Source: Atlaptcd from Scarino ct al. 1988.2. Can utilise a rangc of learning strategies relevant to employment contexts3.4.5.6.Can understand an oral presentation relevant to workplace contextsCan negotiate complex /problematic spoken exchanges relevant to employmentcontextsCan participate in group discussions/mcetingsCan participate in casual conversations1. ’ .IEnglish for community access1. Can understand the context of wclfarc/community services in Australia

44 DAVID NUNAN AND CLARICE LAMB2. Can utiliw a I-angc of learning strategies relevant to the local community context3.4.5.Can understand an oral report relevant to the local community contextCan negotiate com~)lcx/l~rol~lcmatic spoken exchanges for personal business andcommunity purposcsCan participatc in casual con\w-sation[. ’ .I(NSW Adult Migrant kducation Scri ice Ilraft Competencies)Anothcr useful tool is the curriculum-planning qritl. Planning grids such as Figure 2.3can be used to rclatr goal and olijcctivc statements kvith other curricular elements (such asgrammar, functions, or topics). In Figure 2.3 the task or pcrformancc elements from a setof olijectives are cross-refercncctl ivith scttings.Thc gritl \vas dcvclopcd for a gcnci-al Englishspeaking course.Key to settings1 At work 3 Using public transport 5 On holiday 7 At the market 9 At a dinner party2 At home 4 In barkoffee shop 6 In a store 8 At school 10 In a government officehprc 2 3 Planning gritl tor gcncral Lnglish coui vTaskAim To appl> the planning gritl dc\crilictl in this wction to >our own tcaching situation.Develop a planning grid, similar to thc one in 1;igurc 2.3, to a course of your choosing.

MANAGING THE LEARNING PROCESS 45In th15 section \\e haie tried to illustrate a range of \Ea)\ in \\ hich goals and olijcctiicscan be exprcssctl Dcspite tht-ir tlitlercncc\, all of these goal5 antl oblccti\es share somcthlngin common; the! all describe what learners should lit. able to do as a result 01 instructionWe l>cIie\e that all language programs should take as their point ot tlrparturc goals antlobjectii CY, ho\z ei (’1- couchcd, that ha\ e been dcrii et1 from an analSummary and conclusionsThe hasic themc of this chaptcr is that a firm his for d‘frctivc classroom decision makingand managemrnt must hr laid cwll bcfore the teacher sets foot in the classroom. It isdifficult, if not impossible, to sa); whcther many managerial decisions are eithcr good or hadLvithout refercncc to the ncctls of thr learners or thc goals and objectives ofthr curriculum.ions that teachcrs are required to make during the insti-ucnature of the program, the goals of instruction antl thindividual learners, i~-c inclutlc a detailed description and discussion of tprocedures in the chaptcr. I. . .]ReferencesRiggs, 1. and R.Tr1ft.r 1987. The Process oflearning. 2nd ctln. Sydney: Prenticc-1 Iall.Brindle!; G. 1984. A7eeds .hub,.sis und 0bjcctic.c Setting in [he :tdiilt Illigrunt Llocution frogrum.Svdncy : NSW Adult Migrant Education Service.Rrintlley, G. 1989. Ac.se.s.sing .4chie~eincnt in u Leurner-Centred Curriculum. Sydney: NCELTR.Flunkins, F. 1980. Crirriculoin Deielopment: frogrum Improrcmcnt. Columl>us, Ohio: CharlcsMrrrill l’ublishing Co.Iqutke, M. and H. l’honias 1991. Process und Experience in the hnguuge Classroom. London:I nngman ,Nunan, 1). 1988. The Lcurner Centred Curriculum. Cambridge: Cambridge. Univcr-sity Pres.Nunan, D. 1989. Designing 7b.sk.s fir the Communicative Clussroom. Cainbritlgc: Cambi-itlgrUniversity Press.Nunan, 11. 1990. “Action rcscarch in thc language clas.;room.” In J. Richards and I). Nunan(ctfs.), Second Lungtiup Teucher Educarion. Ne\vYork: Cani\)ritlgc Univcrsit) Prcss.Nunan, 1). 1992. Reseilrch .l)fcthoJs in I angtiqe Leurnlng. NcivYork: Cambridge Univcrsity Press.Nunan, D. 1995a. ;ITL:lS: ~.ctrrn1n,4~C‘entcreti Communication. Boston: Hcinlc 8( Heinlc.Nunan, I). 199511. Closing the gap I)eti\ccn lcarning antl in\truction. TE5OL Qiurtcrb , Spring1995.Nunan, D. antl G. Brindle!.1986. “The leal-ncr-ccntretl curricdum in theory antl pi-acticr,”paper prcscntctl at the AnnualTESOI. Convrntion, Anahrim, April 1986.Scarino, A,, D. Vale, 1’. McKay antl J. Clark. 1988. .4tistruliun Lmgiiuge Lcids Guiclelincs.Canberra: Curriculum Devclopmcnt Centre.Untlcrhill, A. 1989. “Pro in humanistic education.” Lnglish Lungiiugc Teuching ]ournul, 43,250 256.Whitc, P, 1961. The Tree ?[,bfun. London: Penguin.Willing, K. 1988. Learning Sr,iiles in .Itloll ,1Iigrunt Etlncution. .4tIclaitle: National CurriculumRrsourcc Ccntrc.

Chapter 3Michael LewisLEXIS IN THE SYLLABUSY L L A B U S IN T H I S C H APT E R is interpreted in what Nunan calls the ‘narrolv’S sense ~ thc content of the teaching programme. Willis, in The Lexical Syll~h, observesthat an approach involves both syllabus specitication and methodology, and that syllabus andmcthotlology arc not discrete options: indeed, syllabus may be specified in terms of goals,performancc ohjcctiws, or other critcria such as l’rabhu’s procedural syllabus. Here, I amconcerned with the contribution lcxis may make to the specification of content. Historically,syllabuses wcre structural; the Communicative Approach introduced functions, antl certainre-ordcrings. The question naturally arises as to kzhat similar changes are called for by theLexical Approach. The search for a strictly lcxical syllabus is likelv to be frustrating fortheorist, tcacher and studcnt. Witldo\vson has olxcrved that a strictly lrxical syllabus wouldbegin Ivith one word texts each complete in itself, proceed to two word tcxts, and so onto ever more complex tcxts but where, at all timcs, any grammatical complexity wasobligatory as thc language uscr’s meaning became incrcasingly complex, and demandedadditional grammaticalisation. Even if such a syllalius were I)ossihlc to devise, it is difficultto imagine it being pedagogically acccptablc.Similarly, thc attempt by Cobuiltl to ticfine a Icxical syllalx~s around the most frequentwords ofthc languagc has not, despite its fascinating theoretical base, met with widespreadacceptancc. Some of the reasons I perceive for this arc discu. ti 11rlow. I cmphasisc that myown conccrn is to look at thc contribution ivhich lcxical items of different kinds canmake in determining content.Educational syllabusLanguage teaching is part of a \vidcr M hole, the education of individuals. Every learningcxyeriencc should contributc to thc dcwlopment of maturc individuals. Althoughcclucational experiences will differ in the \vay they contributc for every participatingin&\ itlual, effective educational experience should increase curiosity, wonder and awe,confidence and self-worth. In addition it should increase the individual’s abilityto concentrate, appreciate, argue a case, tolerate, take responsibility antlco-operate.There is in all education a hidtlcn agenda which secks to tlc lop particular intellectualskills, the most important of lvhich arc inwlvcd in:1 Itlcntifying problem\.2 Collecting information, data antl cvidence.

~AllLEXIS IN THE SYLLABUS 47345Classifying data, bv recognising similarity antl difference.Ranking, making hierarchies, separating more from less important.Evaluating evidence and argument.6 Estimating, so that the plausiliility of an answw may bc ci-aluatctl.7 Taking decisions ~ hasetl on complete or partial data.8 Communicating results effectively.It will he noted that much traditional language teaching is in direct conflict with some ofthese otijectivcs.The I’-P-P (prcscnt, practise, produce) paradigm, repetition, and controlledpattern practice arc elements of this kind. A task-hascd methodology, and an O-H-E(obscrl e, hypothesisc, experiment) paradigm arc in sympathy with the \vitler educationalsyllabus.This is important, for nothing which happens in the classroom should conflict \viththe educational ideals \vhich the ahove summary expresses.The single most distinctive feature of the Lexical Approach is that it proposes afundamentally differcnt attitude to the treatment of text. Firstly, it is suspicious of dccontcxtualiscdlanguagc, recognising the importance of co-text, antl thcreforc preferringextended text or discourse. Secondly, it proposes a range of aivarencss-raising actiliticsdirecting students’ attention to the chunks of which text is compowd. Texts play a rolc inintroducing interesting content, but also act as a major linguistic resource from \vhichstudents can extract lexical itcms for study, expansion, and recording in appropriateformats. A basic classroom strategy will he helping students to avoid becoming preoccupiedhv grammar or vocabulary, concentrating instead on different kinds of lcxical item.Syllahuscs are normally thought of as listing, and perhaps sequencing, course content.In tact, thrcc factors arc important: inclusions, exclusions and sequencing,Inclusions, exclusions and sequencingAs all teachers kno\v, courses are invariably too short. Although a case can lie macle forincluding any language which is ne\v for the student, a principal role for the syllabus is toprovidc- principled ways of including only maximally useful itcms. What is maximallyuseful is not intrinsic to the languagc, but relatcs to particular courses, antl ei.en particulaistudcnts.A primarv distinction is lwt\vccn long courscs ~ perhaps over sevcral years inschool and short intensive coui-scs intended to hale a high surrender valuv. Too manycoursvs are constructed on the implicit assumption that thcy arc intermcdiatc stages on theway to full language comprtcncc. Only rarely is this the ca most students \vi11 remainintermediate and this should influence the language selected for inclusion.Within the I .cxical Approach:lo\\ lexel tourws \\ill gi\c students a large \ocal)ular), eicn if the! are initiall)unable to grammaticalise itPragmaticallj useful lcxlc al item\, partitularl) in\titutionalited uttcrancct, Iorm asignificant component of all coursc5A halance v 111 he maintaincd bet\\ n (relati\ el! rai e) I\ ord5 cari j mg considcralilcincaning, and (rclati\el\ v idc antl frrqucnt) pattern5 1% ith Ion mcaning contentThree principal reasons may he itlcntifietl for excluding material: it is not identified,not valued, or not prioritised. In the days of structural syllabuses, mastery of structurewas rcgarticd as synonymous with language learning; the consequent emphasis of structurewithin syllabuses \vas wholly to lie expected. When the influence of pragmatics \vas felt

48 MICHAEL LEWISin language teaching, functions became a familiar term to teachers. As a result CZbtildyoulike. . .? was re-identified as Ofleering; its re-identitication allowctl it to bc re-valued, and replaced,much carlier in courses. Within the Lexical Approach diffcrcnt kinds of lexical itemmay lie identified, or in relation to traditional language teaching, re-identified. Examplesare trcating would as a single \vord lexical item, rather than part of ‘the conditional’ (seehclow), or the recognition of fully institutionalised utterances which may be introducedand treated as unanalyscd wlmles contributing to, rather than tlcrived from grammaticalcompetence.The tension of syllabus v language and learningMost language syllabuscs still list discrete items; this listing naturally, but misleadingly,suggests that languagc: may be learned in a similar kvay, by ‘accumulating entities’. Nunan(1988: 34), in his comprehensive survey 5jllubu.s Design, remarks that ‘there arc generalarguments against grammatical grading of content, whethcr this grading be based ontraditional critcria or on more reccnt critcria stemming from SLA research’. And he quotesWidtlowson as observing, as early as 1979:Inventories of functions antl notions do not necessarily reflect the way languages arclearncd any more than the inventories of grammatical points and lexical items. Thiscomment reflects WitldoLvson’s claim that ‘Dividing language into discrete units ofwhatever type misrepresents thc naturc. of language as communication’.The tension bctwecn language as communication antl the supposcd neccssity for tliscrctcitem listing for language sTllahuses is reflected in Willis’ comment (1 990: viii):An approach \vhich itcrniscs language seems to imply that items can be learneddiscretely, and that the language can he built up by an accretion of thew items.communicative methotlologj is holistic in that it relics on the ability of learners toabstract from the language to which thcy arc exposed, in order to rccrcate a pictureof the target languagc.7’hc lexical syllabus attempts to rcconcile thrse contradictions.It docs itemisc language. It itcmi language minutely, resting on a large body ofresearch into natural language. On the basis of this research it makes realistic andeconomical statements about lvhat is to be Icarned. But the methodology associatedwith the lexical syllabus does not depend on itcmisation.Wilkins distinguishes lict\vecn synthetic antl analytical syllahuscs, the former being ‘a processof gradual accumulation of parts until the whole structure of language has hccm built up’,\vhile in the latter, Nunan (1 988: 28) suggests:Learners arc prcscntrtl ivith chunks of language which may include structures ofvaryin J dc recs oftlifficulty. A starting point for syllabus design is not the grammaticalb .gsystem ot the language, liut the communicative purpose for ichich the language isused.Prahhu (1 987: 1 ), tlcscrihing his \\ell documented Bangalore Project, tlcscrilio its origins:A strongly-felt pedagogic intuition that the development of compctcncc in a secondlanguage requires not systcmatisation of language inputs or maximisation of plannedpractice, liut rather thc creation of conditions in which learners engage in an effortto cope with communication.

LEXIS IN THE SYLLABUS 49Pratihu, Widdowson, Nunan, Willis and indced many others would concur with Nunan’sjudgement that:Evidence from second language acquisition research suggests that learning docs notoccur in a simple additive fashion.(1988: 30)Syllabuses tend to isolate, divide and sub-dividc. The tacit assumption is that macro-skillsare a synthetic assembly of micro-skills; that larger units of discourse are assembled fromwords and structurcx. Thesc assumptions arc almost certainly untrue but this raisespedagogical difficulties. Thc implications are that we should adopt a morc holistic view oflanguage, and a task-based approach to learning, but, as Willis (1 990: 1 29) observes:A shortcoming oftask-tiascd approaches is that they make it difficult to specify syllabuscontent, and as teachers \IT cannot he sure what has lm-n learned in the course of agiven language activity or a given unit.‘I’here is a fundamental conflict between the teacher’s natural desire to give clearly focuscdand effective lessons, and the non-linear nature of language and learning. Although therc issubstantial theoretical support for task-based goal-orientated syllabus specification, mostteachers continue to demand much more specific linguistic objectives for cach lesson. Whileendorsing and encouraging a mcthodology based on tasks and skills, rathcr than spccificallylinguistic criteria, we can identify explicitly linguistic changw which arc consistent with theLexical Approach.Content specifying listsOne of the most intlucntial attempts to specify contcnt \vas the Threshold Level, whichattempted to dc-vclop in detail the \vork summarised in A’otional Syllabuses. Somewhatsurprisingly, whilst claiming a primary focus on meaning, Wilkins (1 976: 2 1) had a rathcrcavalicr attitude to vocabulary:Rut it is tlicreforc with the general aspects of mcaning and use that thc categoricspresented here are concerned, though they arc not less significant for being gencralin character.lhis also explains why no attempt is made in this framework to accountfor a lexical contcnt of 1earning.This is probably hettcr approached in terms of subjectmatterand situation. At the same time, lcxical aspects cannot bc entirely excludedsince grammatical and lexical tleviccs often interact significantly.To a certain, though limited, cxtcnt the scmantico-grammatical categories themsclvcshave applications for the lexical content . . . The lexical content of Icarning, therefore,can be largely derived from an analysis of the typical topics which occur in the languageuse of a given group.(1 976: 76)Wilkins’s vicw is, thus, that ho\vrver important vocabulary may be, it has no defining roleto play within syllahus design.In contrast, Willis (1 990: v), dacloping Sinclair’s ideas, regards vocatiulary, and quitespecifically words, as the key to syllabus specification:

50 MICHAEL LEWISSinclair advanced a number of arguments in favour of the lcxical syllabus, liut theunderlying argument \vas to do with utility and with the po\vcr of thc most frcqucntlvords of English. . . . We tlccitlctl that \vord frequency \vould determine the contentof our courseInstead of specifying an inventory of grammatical structures or a set of functions, eachstagc of thc course \vould bc h ilt round a lexical syllabus.This lvould specify \vords,then meanings antl thc common phrases in which thcy were used.(1 990: 15)It \vi11 bc notcd that, despite the retercncc to ‘phrases in which they occur’ Sinclair andWillis largely equate the lexical syllalius with a word-based syllabus. Inherent in thisintcrpretation are three problems which manifcst themselves in thc coursc tlescrilied inWillis’s The Lexical Syllabus:1. The most frequent ‘\vurds’ are lrcquently items previously regarded as structural and,ironically, \vords of low semantic contcnt.Thcsc largely delexicalised words are highlyfrequent precisely becausc they often have scvcral meanings, antl their pattern profilesare cxtremcly complex. Mastcry of wortls like to, with, have is considerably moredifficult than mastcring a voca1)ulary itrm with highcr meaning content: accident, soot,slump.2. The word-based syllabus introduced words with Imth their highly frequcnt and muchrarer meanings together. A preoccupation with the word as a unit meant infrequentmeanings of high]! frequent lvords \vcre givcn preferc:nce over highly frequentmeanings of rather lcss frcqucnt words within thc corpus. Some of these rarermeanings of high frequency words appear as of relatively low utility, and a relativelyhigh confusion-factor for elcmcntary studcnts.3. Multi-word lcxical itcms arc untlcr-valucd antl under-cxploitcd.The Lexical Approach I propose avoids these dangers. It is specifically not a lexical syllabus,and explicitly recognises word patterns for (relatively) delexical words, collocational powerfor (relatively) semantically powerful words, and longer multi-\vord items, particularlyinstitutionaliscd sentences, as requiring tlifferrnt, and parallel, pedagogical treatment.The old structural syllabuses specifically restricted vocabulary to the level necessary toexemplify structural patterns. Ironically, Willis ( 1990: 74) in his word-based approachexplicitly espouses the samc principle: We .set out to achieve the best coverage we could with aslittle extraneous lexis as possible (i.e. extraneous to the most frequent 700, 1,500 and 2,500‘words’ which they selected as the basis for Parts 1 , 2, 3 of their course). In contrast to theirurge to restrict vocahulary at low levels, I advocate encouraging the learning of acomparatively large repertoire of high-meaning content nouns, adjectives and vcrbs.Although the \vords learned will inevitably he in corpus terms comparatively low-frequency,by definition they carry meaning. Rut words carry more meaning than grammar, and if it iscommunicative power which is thc primary objective, increased vocabulary will play a largercontribution than additional mastery of even the most highly frequent patterns of highfrcquency words.There is an additional, pedagogical advantage. Willis observes that ‘profileshccome lcss complex as one moves down the frequency scale’.This means that from a naive,student point of vicw the words are easier to learn, and any L2 L1 equivalence, whichstudents almost incvitably make, is more likely to be accurate. ‘Learnability’ andcommunicative power arc at least as important in selecting words for inclusion as frequency.

huve,LEXIS IN THE SYLLABUS 51Lexis contributes as a syllabus component in the folloicing lvays:1. Certain words deserve lexical rather than grammatical treatmentare typically high frequency, de-lexicalised items. Those items \vhich enter into thewidest range of patterns, and are thus usefully if not maximally generative, arc words whichthemselves carry least meaning. De-lcxicalised verbs ~ get, put, take, make, do ~represent an important subcategory. Function words, often thought of as prepositions ~of, with,j;)r, by, are anothcr. The modal auxiliaries, including would, are a third. Mostimportantly, would should be dealt with early in a course from a lexical point of view.I!I/ould was trcatcd in structural courses as ‘the conditional’; functions moved it to an earlier,but comparatively marginal, non-generative position. It dcscrvcs high priority as a onewordlcxical item. Interestingly, it is one ofthe items which mcrits fullest discussion in TheLexical .~llahus.2. Increased attention to the base form of lexical verbsA preoccupation with grammar and structure has obscured the importance ofthe base formof the verb in English. Willis comments on courses which ‘spend an inordinate amount oftimc on the verb phrase’, that is, on the structure of the verb, and so-called tenseformation. In fact, the simple present is about eight times as common as the presentcontinuous in naturally occurring English and is, with thc marginal inconvenicncc of thethird person -s, identical \vith thc base form. The Lexical Approach advocates the nred fora large rcpcrtoire of verbs in their lmsr or lexical form with increased attention to the highlyfrcquent present simplc.3. De-contextualised teaching of semantically dense itemsCommunicative power is most rapidly increased by expanding studcnts’ vocalmlarics,meaning their repertoire of lcxical items, but particularly simple high-contrnt words.Thcrc is no need for over-elaborate contextualisation in the early stagcs of learning: simplcidentification of signification, although in no sense mastery of the word, is an appropriateand valuable basis for increased communicative power.4. CollocationsAs soon as the inadequacy of the grammar/vocabulary dichotomy is rccognised, it heconicsnatural for collocation to assume an important syllabus generating role. This applirsparticularly to relatively high content nouns. When these arc introduced, it should he naturalto introduce with them verbs and adjectives which form pow.l-erfu1 or relatively fixedcollocations. The statistical evidence of corpus lexicography hcrc clearly reveals the nccessityof acknowledging both literal and metaphorical meaning. Often it is the lattr-r which is morefrequent.5. Institutionalised utterancesTraditional grammar exercises usually include a sample sentence which providcs the modelfor students to produce ‘similar’ sentences. Modern research into both grammar andlearning suggests that students could usefully be offered a group of sentences forcomprehension and reflection. These would not exemplify ‘the grammar’, hut bepragmatically identifiablc institutionaliscd utterances which students could both useimmediately to increase communicative power, and as a resource the analysis of whichwould provide a basis for the gradual perception of pattern.6. Sentence headsThese are very similar to institutionalised utterances. Scntcnce heads can frcqucntly heidentified and provide both an immediate increase in communicative power, and a resource

52 MICHAEL LEWISto aid acquisition. These scntcncc heads frequently lie somc\vhcrc Iwtwccn grammar andfunction on a conventional syllabus. ‘Grammar’ in grammar practices frequently tried tocover all elements of the paradigm, consciously introducing first, second and third personsubjects, singulars antl plurals; in functional practice a single sentence head Cfbuldpi liketo . . . requires students to complete thc scntcncc in different \.rays. Introspection orstatistical data, ho\ve\ cr, 110th re\d that some combinations of, for example, a particularmodal antl a particular person are much morc frequent than others; compare Could~ou. . . and Coiild she . . .; contrast 1 might . . . and Might 1 . . .? Doyon thinkyou might . . .? andDojoti think 1 might . . .? Paradigms cxcmplify the possible sentences of English; \vcllchosengroups of scntcncc heads exemplify the frequent or probable patterns of English.Functions arc all too often ungcneralisablc, \zhilc scntcncc head groups arc gcncralisablc.It is noticeable that the institutionalisctl uttcranccs antl scntcncc heads of spoken Englisharc vclry different from those of the \\ rittcn language. McCarthy is only one of many tosuggcst that ‘vocabulary \vork in spoken language requires separate and additionalprocedures from vocaliulary teaching using written texts’.7. Supra-sentential linkingTraditionally this has liccn practised only on a grammatical level, concerning tags, interestedresponses etc. In fact, supra-sentcntial lexical linking is an important cohesive device inspontaneous conversation, suggesting lexically, rather than structurally, based cxcrciscswould be morc natural and morc pragmatically c4l’cctivc. McCarthy (1 99 1 : 7 1 ) quotcs datain which:People did not typically agree or disagree with phrases such as ‘I agree’ or ‘I disagree’(beloved of English course book writers); rather, there seems to lie a preference forsimply using some sort of lexical relation between turns.This suggestion is borne out in Willis’s work, antl hc goes further, suggesting that muchspontaneous conversation is based on joint production, in which participants contributematching, complementary or contradictory lexical items in the devclopmcnt of a single unitof meaning.Supra-sentential linking of this kind is central to spoken discourse, but quite different,and equally important features apply to the crration of cohrrent and cohcsivc written text.A central requirement of the Lexical Approach is that language material should be text andtliscoursc, rather than scntcncc hascd. Again Willis agrees, constantly reasserting that ‘onlyby drawing attention to occurrences in text’ can learners begin to h ild up an adcquatcpicture of language in use.8. Synonyms within the existential paradigmThis is a particular cxamplc of supra-sentential linking. Observations of real data show thatin spontaneous speech the ability to usc altcrnativc language items as value-synonyms,although they have different signification, is a key fcaturr of fluency. Thcsc value-synonymsmay be both individual words (daffbdils/flower5) or fully grammaticalisctl utterances realisingthe same pragmatic function (That has my full support. /ilbsolut+, I’dgo dong with that).9. ‘Synopsising’ wordsTraditional grammar taught so-called reported speech. As discussed elsewhere, this categoryis wholly untypical of naturally occurring data. Most often, thc speaker reports a wholeevent, rather than manipulating the words that were spoken. The ‘reporter’ summarisesor synopsises the whole nt lexically and so rcquircs an adcquatc repertoire of synopsisingverbs.

LEXIS IN THE SYLLABUS 5310. Metaphorical patterningMetaphor is often perceived as an essentially literary device. Modern philosophical andlinguistic research rcveals that far from king rcstrictcd to literary language, it is intrinsicto thc nature of language itself. Lakoff and Johnson (1 980: 7ff), in a seminal 1vol-k belongingntiallv to the field of philosophy, have demonstratctl convincingly that there arc manyconcepts which cannot be discussed except in metaphorical language. ‘l’hcy give manyexamples but here a single example must suffice: TIME IS MONEY.Thcy point out that this is an English proverb, but more than that, it is impossible totalk about time without basing the conceptualisation on the metaphor ‘I‘IMI. IS MONEY .Theydeveloped the idea as fol1on.s (in slightly abbreviated form):Time is moncy is a metaphorical concept. It is metaphorical since \vc are using oureveryday experiences of money, limited resources and the valuable commodities toconccptualise timc.This isn’t a necessary way for human beings to conceptualise time;it is tied to our culture. Thcre are cultures where time is none of these things.Wc are adopting thc practice of using the most specific mctaphorical conccpts, in thiscasc time is money, to charactcrise the entire sjstcm.This is an example of the \Yay in lvhich metaphorical entailments can catcgorise acohcrent system of metaphorical concepts and a con-csponding coherent system ofmetaphorical expressions for those concepts.They point out that in English many of the lvords used to descrilie time can also he used todescribe moncy: spend, invest, hudger, profitah!,y. Here are somc of their examples:How do you spendyour time these duys.;1 haven’t enough time to sppnre-fir thut.Is it worthjrour while?You don’t usegour time proptuhb.You ore wasting my time.This p dpt will smej’ou hours.Clearly, there is a pattern here which it is \vorth\z-hilc to draw to thc attention ofstudents.Many ofthe \vords which arc uscd to talk about moncj can also be uscd to talk about timcy.This is not fully gencralisahle, but it still constitutes a powerfully generati\ e pattern s~ stem.The importance of I.akoff and Johnson’s \vork is difficult to over-emphasisc. It is essentialreading for anyone inttwstetl in how language works.When the Berlin ndl \vas breached, at first a trickle of peoplc camc through. Latcr, asthe gap was widened, pcop1e.Jloodcd through. There \vas a constant streum of people anxiousto visit friends, or rcstorc family contacts. Oncc the initial excitement wort off, thc,porr. ofpeople dried up.The above passage rcprcsents my own observations of the language used hy the K.B. C.Nexvs to rcport the destruction of thc Berlin wall. An important metaphor is involvctl:cro\vtls of people movc like VI atcr. It is almost impossil)lc to descrilir those events withoutresorting to ‘water-\vords’. Rut notice, as Lakoff and Johnson constantly emphasise,metaphor highlights only at the expcnse of supprcssing. Peoplc in movement may movc likewater, but they are not water, water docs not rc-cstablish family contacts.l’hcrc is a usefullinguistic pattern, but not an identification.Editors on the Cobuild project were initially surprised at the prcpondcrancc of

54 MICHAEL LEWISmetaphorical usage ~ torrents arc morc likely to be of abuse or French than water.Lexicographic difficulties arise ~ if metaphorical use is morc frequent than thc litcral, andtherefore supposedly core, use should it be placed first in the dictionary?Their editors haveobserved, for example, the importance of plant-based metaphor in discussing abstractionssuch as government policy: The problem has its roots . . ,;Since the beginning oftheyear, we haveseen ajowering. . . .For language tcaching, thc importancc lics in rccognising:abThat metaphor is a part of everyday language.That such metaphorical usage is patterned, often in accessible, gencralisable ways.Functions and skillsThe development of communicative poivcr will be aided by incorporating a well-balancedrange of lexically derived activities in the classroom. These must reflect the different kindsof lexical item. The change, however, is a mattcr of emphasis not revolution. Grammarretains a place, but a reduced onc; lcxis plays an incrcascd role. Language content can,however, never be wholly scparatcd from othcr elements of syllabus specification. Mostfunctional syllabuses concentrate on micro- rather than macro-functions and ‘nice’ ratherthan ‘nasty’ events. For many students such functions as cxprcrsing irritation, expressing dishelie_f;distancing the speaker.from the content ~ fwhat is said, expressing condolence, telling and respondingto jokes may be at least as important as accepting and reJising invitations politely. In a similarway, a lexical approach suggests that thc skills syllabus needs to be broadened. Two skillscentral to the Lexical Approach arc developing the students’ ability to use the dictionary asa learning resource, rather than reference work, and, most importantly of all, helpingstudents to identify lexical phrases in tcxt.This rcturns us to the single most powerfulmethodological implication, namcly a tliffcrcnt attitude to, and use of, texts.BibliographyLakoff, G. and Johnson, M. MetaphorslVe Lire By. Univ. of Chicago Press 1980McCarthy, M. Discourse Analy~is~fbr Language Teachers. CUP 1991Nunan, D. Syllabus Design. OUP 1988Prabhu, N. S. Second Language Pedagogy. OUP 1987Widdowson, H. Proper Words in Proper Places. ELT‘ News No. 8. British CouncilVienna July1989Wilkins, D. Notional S,vllahuses. OUP 1976Willis, D. The Lexical Syllabus. Collins Cobuild 1990

Chapter 4Michael McCarthy and Ronald CarterDESIGNING THE DISCOURSE SYLLABUS*1 IntroductionHE ADEQUATE DESCRIPTION OF LANGUAGE is vital as a precursor ofT languagc tcaching syllabuses. At the macro- and micro-level, from issues of genre downto individual grammatical and lexical choices, our findings (McCarthy and Carter 1994)have implications for how we look at the syllabus and, consequently, its content and thekinds of activities that it generates in the class. This chaptcr concentrates on thoseimplications in discussing the design of the discourse syllabus.2 The notion of ‘discourse competence’Ever since Chomsky (1 965) made the distinction between competence and performance,that is what a person knows about his or her language as opposed to what can beobserved from manifestations of actual use, linguists have debated just what ‘competence’might mean. Chomsky was concerned with the fact that native-speakers have an underlyingknowledge of what constitutes a well-formed scntcncc in their own language, and hc sctabout trying to account for such knowlrdgc. But it was not long before thc notion ofcompetence was expanded to embrace what a speaker needs to know about how a languageis used in particular situutions for effective and appropriate communication, in other wordscommunicative competence (see Hymes 1971).The notion of communicative compctcncc has had a very powcrful influence onlanguage teaching, both in terms of methodology and the goals set by syllabus plannerswhich learners are supposcd to achieve. Thus the term communicative syllabus is afamiliar one to most language teachers. Typically, a communicative syllabus will set out avariety of communicative abilities that thc learner should be able to dcmonstrate at the endof a prescribed course or period of learning. One such English Language syllabus, a pioneerin its day, recommended that learners should be able to (among other things) makc andreceive telephone calls, handle friendly and social correspondence, makc short notes torecord salient information, ask questions and makc comments for gleaning furtherinformation, and so on.This was the Malaysian (1 976) Communicational Syllabus for Forms4 and 5 of secondary school (see British Council 1983; 1986). The syllabus was a classic* Thi\ text ha\ heen atlaptcti

56 MICHAEL McCARTHY AND RONALD CARTERcommunicative one, \vith no rcal emphasis on rorrcctncss in grammar and vocabulary, andevery cmphasis on the ability to communicate antl achi goals, a balance ofprioritics forwhich it came into much criticism in its o\vn country (See Mohidccn 1991). It was criticismofthis swing ofthc pcmclulum an.ay fi-om linguistic. (i.c. grammatical and lexical) competenceto a prcoccupation bvith communicativc. compctcncc alone, not just in Malaysia, hvhich ledapplied linguists to question whrthcr compctcncc could ever lie seen as a monolithicconcept. Might it make morc sense to think of the lcarnrr developing a set of competences,each one csscntial to using language cffcctivcly, but each one separable in termsof what could bc dcscribed and prescribed tor the syllalius and lcarning programmc?Thusgrammatical and lexical kno\vlctlgc as one of thc scvcral compctences came to the fore againas an issue in language teaching. Applied linguists argued that communicativc ability was ahollow notion without kno\vlctlgc of thc grammatical system that cnablctl actual realizationsof communicative acts (but also vice wrsa; scc Canalc antl S\vain 1980). Equally, there \vasa return of interest in the prol)lcm of i-ocahulary building, lvithout Lvhich little rcalcommunication hias possible (McCarthy 1984; s ~ also c Carter antl McCarthy 1988: ch. 3for a survey of these arguments). Linguistic compctencc, it mas argued, was a nethough not sufficient, condition for communicative ability. From such pressures h\\.hat most \vould agrcc is a healthier balancc hctm n the tlcvclopment of competence intem antl compctcncc in its use, as exemplified in so-called eclectic.an and Walter 1984 Cambridge English Course is a good example), and inwhat Yaltlcn (1983) calls the proportional syllalius, \vherc the proportions of systemorientedknowledge and communication-oi-icntctl skills arc increasingly altcrcd in favourof the latter as the learner progresses from beginner level. The lcxical syllabus (Sinclair andRcnouf 1988;Willis 1990), based on a faithlul description of how words arc used, representsanothcr move in the direction of integrating knowlcdgc of the system antl knou ledgeof use.Rut othcr questions remain for thc guagc tcachcr. If the description of language isincomplete without a description of the 1 of discoursc, and if discourse-level constraintsoperate simultaneously with Icxico-grammatical ones, then is thcrc something akin toa discourse competence that can he tlescrihetl antl articulatcd as a sct of‘ goals for thesyllabus to aspire to? Rcccnt tlcliatcs in syllabus design have tended to assume that there is.Those linguists antl applied linguists who have moved amy from the idea of competence asa monolithic concept have already addcd to thc basic notion ol’communicativc competencesubdivisions such as socio-linguistic competence and strategic competence. As Canalc (1983)uses these terms, they ma); I>

DESIGNING THE DISCOURSE SYLLABUS 57taught and goals to lie achieved. The first problem, the separation of socio-culturalfeatures from discourse ones, is cspccially problematic givcn, as ~ vc have argued clsewhcrc(McCarth? and Carter 1994), that such things as register antl mode are integral to thecreation ot’tliscourse, not in some \vay ‘parallel’ or complementary to it. Wc haw also soughtto demonstrate that isolated lists of spccch acts are insufficient to tlrscribe \\.hat sIieakcrs/writers do and how they manage interaction over extended language cvcnts. In other Xvords,\vc see the chaining together of’functions or speech acts as inseparalile from the creating oflargcr pattcrns and gcmres in discourse. By the same token, ~ve see the realization ofrcgistcrs, attitudinal features and topics as inseparable from cohcrcncc and its manifestationsin surface cohesion. Even more to the point, grammar antl \ ocabulary kno\vledge shouldinvol\e ho\\. these aspects of linguisticjbrrn create discourse; in other \vords, linguisticcompetence cannot be separated from discourse competence.These views have a direct bearing on the second concern, xvhcther things can heitemized for teaching and given socio-cultural, strategic or discourse laliels antl thereti?allotted their rightful place in the syllabus inventory or check-list. How we analyse andclassify language for our syllabus necessarily affects our methodology and Lvhat \ve do inthc classroom.3 Analysis and classificationSome notable writers on syllabus design follow the view that the analysis of language intoits various levels antl the classification of fcaturcs within those 1 is a feasihle hasis forsyllabus specifications.Yalden’s (1983) description of syllabus components sccms implicitlyto accept this with a section cntitlcd ‘A further component: discourse structure’ (1983: 78),and her syllahus chcck-list (1983: 169-72) includes the following discourse components:A13C1)Cohesion and rcfcrcncc (basrd large17 on Halliday antl Hasan 1976)Operations on text (for cxamplc extracting salient information, expanding a text)Rhetorical organization (textual functions such as generalization, classification, ctc)Overt transactional skills in spoken discourse (for example initiating, introducingtopics, closing, turn-taking) .Thcse categories certainly represcnt innovativc clcmcnts in syllabus specifications and arefaithful to \vhat discourse analysts have described as above-sentence features. Wc shouldnote, though, that categories A and C seem to be languagc fcaturcs, while I3 and D w ~ ~ ~ l dseem to fit bcttcr under the heading of‘ skills or stratcgics. This is no mere hair-splitting,antl is at the heart of thc process of analysis and classification that precedes specification antlitcmization. For instance, it could I)c argued that a feature such as lexical cohesion is anaspect of the language system antl can thus be taught as languagc knowledge, just liketeaching the grammatical facts about tenses or dctcrmincrs. This would mean not only tellinglearners what the synonyms and hyponyms ofa particular word or set of words arc, but alsodemonstrating that synonymy and hyponymy in tise are often involved in the creation of wellformedtext and interacti .pccch (see McCarthy 1984; 1988). Howcvcr, another vie\vmight he that lexical cohesion is a language universal; as such, it liecomes more a matter ofskill-training, practice and training in an intuitive skill in order to improve one’s proficiencyin its use, without any need to ‘present it’ as knowlcdgc or fact.This is a crucial decision inthe categorizing of syllabus components: Yaldcn (1983), for example, has clearly flaggcdfeatures such as turn-taking and closing as ‘skills’, suggesting a different emphasis from thatattached to cohcsion antl reference, while ‘opcrations on a text’ are unambiguously things

58 MICHAEL McCARTHY AND RONALD CARTERwe ‘do’ with language, rather than fcaturcs Lvhich ‘exist’ in the language system. Butseparating the ‘ivhat’ of thc language . tcm from the ‘how’ of language skills and strategicuse can also bc misleading: thcrc is every reason to suppose that knowing ‘what’ can informand support knowing ‘holv’.Munby (1978) has a simi , though much more tlctailctl, specification of discoursejiutures (cohcsion, initiating, dcoping thc tliscoursc, ctc) mixed in with textual operations(‘reading between the lines’, extracting salient points, skimming antl scanning the text,etc), lvhich, among man: other things, form a continuum from basic phonemic andgraphemic discrimination through to macro-planning, all ler the heading of ‘languageskills’. ‘Discourse lcvel units’ (Munhy 1978: 27) arc still , though, as separate fromlanguage micro-functions antl grammatical/lcxical rcalizat , and discourse is a level orlayer of language rather than integral to its cntirc operation.Although, as we shall see, kvays of implementing the notion of a discourse element inthe syllabus vary considerably, thcrc docs seem to I)c witlcspread agreement that the ideaof discourse cannot he ignored; syllabus tcmplatcs antl check-lists as offered by appliedlinguists such as Munly (1 978) antlYaldcn ( 1 983) have a discourse clement built in. But wemust now consider how more integrative vic\vs of discourse influcncc the nature of thesyllabus and the tcaching that evol\-cs from it.One problem with thc vie\vs of comniunicativc competence as implied by the syllabusspecifications that we have looked at so far is that thcy havc assumed that language use canhc analyscd and described as a sct of components of various kinds. This assumption oftencrcatrs difficulties in that the separation of’componcnts can produce a false picture oftheirrole in creating the overall message. A good example of this is the sort of list often foundin syllabus specifications of speech acts or functions, such as promising, directing, enquiring,apologizing, etc. As Candlin (1 976) points out, an inventory of speech-acts of this kind ‘cannotscrvc any more than sentences as thr direct endpoint of a communicative syllabus’. Anysyllabus consisting solcly of such a list lvould fail in two directions simultaneously: it wouldfail to provide the learner with a clear vicw of thc interrelated and structured nature ofelements of the language system such as modality antl mood, and it would fail to show howapologies, enquiries, promises, and so on arc actually realized in interaction and as part ofa .sequence of utterances antl how such realizations depcndcd on higher-order constraintsof genre. In other words, we \vould hc guilty of dealing bvith (some of> ‘the components ofdiscourse, not with discourse itself’ (Widdowson 1979: 248). Widdowson and Candlin bothcome at the problem from the other direction: communicative compctcncc is not a list oflearnt items, but a set of strategics or proccdurcs ‘for realizing the value of linguisticelements in contcxts of use’ (Widdowson 1979: 248), and, just as learners may be expectedto pcrccive grammatical regularities in scntcn , so thcy should be given the opportunityto interpret pragmatic clues for the attachment of value to utterances in discourse, andbecome themselves analysts of discoursc (Candlin 1976).One highly innovative approach to incorporating an integrative view of discourse intothc syllabus is provided hy Aston (1 988). One ofAston’s concerns is to redress the imbalancetowards transactional language common in much language tcaching (which wr commenton in section 4) and to get to grips with the problem of creating the contexts for interactivediscourse in thc classroom. Aston too moves away from simply adding discourse as an extracomponent in the syllabus and effectively builds his syllabus around central and fundamentalfcaturcs of interactive discourse. For Aston (as we havc argued) interactive discourse isconcerned not only with illocutionary uptake (the realization of speech acts), nor just with‘cognitive convergence’ (achieving shared knowledge and pcrlocutionary effect), but alsowith affective convergence (an essentially humanistic notion), with the processes of creating

DESIGNING THE DISCOURSE SYLLABUS 59and hvith the global and local strategies negotiated in individual contextsfor achieving them.Aston recognizes the problems created by analysis and classification as thc precursor ofsyllabus specifications: any analysis claiming to describe competence and to itemize it for asyllabus \vi11 fail to capture the fact that discourse is realized by the crcative exploitation ofthe resources that constitute competence (Aston 1988: 1634). In this sense, the learnercan engage properly with discourse only ly Cioind it.This would seem to be a strong argumentin favour of thc task-hascd approach as expounded by Prabhu (1987). In the task-lnscdclassroom, languagc is tised in the process of solving preordained tasks, \vith the purpose ofpromoting and enhancing uptake antl Icarning, rather than presented and lcarnt in orticr tobe used later in cxcr . or outside in the real world.Aston, however, sees many problcms arising from more extreme views of the taskbasedapproach (the extremest form of which would be the completely negotiable syllabus,with nothing preordained and everything open to negotiation among lcarncrs and teachers,which Clarke (1 991) claims would be unworkable anyvay). Aston seeks to build a syllaliusXvherein the learning process is not just left to gct on with itsclfin unpredictable ways, hutin which teaching can operate as a guidance.lb this end, it is not sufticirnt just to specify aset of tasks for learners to undertakc. For one thing, many of the task-types advocated bytask-based syllabus designers fall into the same trap as the information-gap activities ofcommunicative approachcs, in that they cncouragc a transactional vimv of language at theexpense of the interactional. Furthermore, simply specifying tasks ignores the fact thatlcarncrs can be guided in the procedural knowledge (the ‘ho\.r, things are done’ in particularspeech communities) as well as the declurutive kno\vledge of‘what is clone’, both of whicharc essential to the creation of coherent discourse. Aston, therefore, favours a task-liasetlapproach that does not shy axvay from specifying the discour. ,tratcgics that thc lcarner\vi11 need; these will lx specified in a strategic pre-syllabus, which hc sees as a ‘contcntlmd’one (Aston 1988: 188). But c n with this pre-syllabus, tasks involving the learnelincreating discourse as the main syllabus are not enough. For Aston, the main syllabus istwo-stranded, and the second strand involves the learner in hecoming a discourse-analyst,or indeed a sort of anthropologist (1988: 184), observing and cieconstrLrcting how discourse iscreated.Aston’s final model therefore, looks like this:contcxt-lmctl syllabus construction tlcconstruction(Aston 1988: 188)Aston’s viem- of the syllabus seems to recognize that discourse is a process rather than aproduct (which tends to be the view of those who see ‘discoursC-as-a-layer’ in languageuse), but, sensibly, he sees the value both of an analysis and classification of discoursestrategies as a precursor to selecting tasks for the classroom antl of making the learner standliack a little from language and become an observer of it, though as a tliscoursr-analyst ratherthan as the sentence-parser and rule-discoverer of some approaches to traditional grammarbasedsyllabuses.Ours is also an intcgrativc vicw, whcrcin the ovcr-arching pcrspcctivc of languagcas-discoursewill affect erwy part of thc syllabus, including any conventional ‘system’(lexico-grammatical) components and functional/specch-act components, however they

60 MICHAEL McCARTHY AND RONALD CARTERarc treated, whether as a series of layers of language, or as realizations within generalspecifications of discourse strategies.4 Analysis as the precursor of tasksAston’s programmc favoured a ‘ lire-syllalius’ oriented tolvards strategic issues in discourse.Specifying strategics is something that can lie done in different ways and at different levels.Some syllabuses (for example ICC 19x6) specify a gcncral set of strategies, but it is alsopossiblc to conceive of a highly detailed set lvhich translate some of the more traditionallyconccivrd ‘features’ of language use into the strategic domain, and this is Ivhat we wouldlike hriefly to consider in this wction. What \z.c propose are a sct of strategy-headings thatcan act as a sort of filter lictn n the learning group antl its need and the specification oftasks. Each heading is follon.ctl by example qucstions that arc raised liy each onc and thesort of practical issues that arc likely to l x encountered in the detailed specification oftliscourscfiattrrcs that might bc encountcrcd in the subscquent tasks. We say ‘likely’ because\re cannot always guarantee what thc outcome of a task will lie. It will he noted that theglobal sct of strategy headings can subsume what has previously hen seen as a scparatediscourse ‘layer’ by some svllalius designers. Thc most gcncral heatlings are as follows.1 Genre-related strategiesWhat are the mctlia and modes that the learners \I ill encounter?What genres arc likclj to be most useful?What patterns of interaction arc most useful (c.g. narratix e, lirohlem-4olution),2 Coherence-related strategiesWhat aspects of topic managcment, turn-taking, ctc, \vi11 hc involvctl?What types of cohesion (c.g. stronger emphasis on across-turn lexical cohesion forintcractionally oriented tasks; tlil’fcrcnt types of ellipsis in tliffcrcnt media)?3 Politeness strategiesWhat aspects of facc \vi11 need to Iic atltlrcssctl?What forms of address will I>c itnolvctl (c.g. pronoun systems, mood systcms)?Holv important will reciprocity be (c.g. very important in interactional tasks)?4 Planning strategiesWhat sorts of anticipator! strategies will he idul (c.g. cnuniei-ativc labelling, cataphoricuses of articles)?Will special conditions for i-efcrcncc apply (c.g. anaphora across paragraph tiountlarics inwritten metli um) ?What scquences oftmsc, aspect and voicc arc likely to 1~ involvcd (c.g. con\rntions rclatcdto genre)?What degree of crccitiri~. antl risk-taking with language is fcasihlc antl appropriate?5 Convergence strategiesInformational or cogniti~e con\crgcntc: \I hat aspects of categories such as theme, moodand modalit! \I ill be in\ol\ctl?

DESIGNING THE DISCOURSE SYLLABUS 61Affective convergence: \\,hat adjacency-pair typcs arc likcly (c.g. solidarity routines, Ixoblcmsharing,agrccmcnt~disagrcemcnt)?What transaction-boundary features are likely (pitch-sequcncing, markcrs)! What role willrepetition play in creating convcrgrncc in diffcrcnt modes and genres? What tlcgrcc ofcultural convcrgcncc will be required? How will ‘knowing almut’ language anti culture ain solving convcrgcncc problems?6 Repair strategiesWhat arc the risks ol communicational prolilems or cultural misunderstandings?Is repair likcl! to lie largcly self repair, or morc global, ncgotiablc repairs?Stratcgirs invarialily overlap. Iyor example, repair strategies may involve politeness, whichin itself involves cultural awareness and the problem of convcrgcncc, antl so on. Rut giventhe practical exigencies of dividing the discourse process, n-c \vould arguc that thc stratcgiclist I-epresents a manageable antl reasonably faithful framework for syllabus and task design.What one docs with a list of strategies for a particular learner group dcpcnds on onc’sphilosophy concerning methodology. The discourse-had approach (i.c. whcrc \ve slur[with discourse as the overall driving force of our syllalius) lends itself best, \vc have implied,to a task-liasetl methodology, in that, in this way, language is not atomized antl treated asproduct, thus destroying the basic notion of discourse as engaging Ivith language as processantl meaning as ncgotiatcd and contextual. Ho\ve\er, in the real world, teachers often haveto \vork within clear and restrictive constraints Lvhere they arc expected to \vork to explicitlystatcd classroom input and to achicvc explicitly measuralilc output, in other \\-or&,syllaliuses that say rihar is to lie learnt and in \vhat ordcr.We see no contradiction het\vecn our proposed list of tliscour tratcgics and the sulisequcntspccification of the syllalius in terms of a set of specific performance goals, only thatwe start from a different premise: that all such goals can, and should he, expressed as discoursegoals rathcr than as Iexico-grammatical or notional-functional oncs. Ibr example, \\e mightcnvisagc a ‘learners should be able to . . .’ lcaturc including something like the follo\ving:Ask significant favours of othcrs in appropriate secluences in\ olving1 signals of opening2 explaining thc problrm3 asking4 minimizing5 reinforcing6 acccding7 thanking.The asking of a la, our is thus conceived of as a genre rather than as a function or spccch-act,and inlolves not only sp h-act realizations at the micro-lc 1, hut also a strategic Icvclinvolving politcncss strategies (l’acc), planning (opening), convergence (reinforciny), andso on. At the Icxico-grammatical interface, onc could specify modalitv and (drlirnding onlevel) use of idioms. The point is that the conventional syllat,us~as~inventoi-y view can stillbe meaningfully adapted to a languagc~as~tliscourse approach \vithout just atitling discourseas a layer upon the other layers. t..qually, such an in\-cntory, in our opinion, tloex notsarily preclutle additional use of well-choscn tasks in class that can subscribe to Aston’s(1988) conditions of construction and deconstruction, nor docs it necessarily ~ireclude somc

62 MICHAEL MCCARTHY AND RONALD CARTERsort of proportional syllabus approach such asYalden (1983) advocates. For us, it is theanalysis of‘ language needs through a discourse perspective which is most important as aprecursor to tasks and activities, whether such tasks are additional to a more conventionalcommunicatively oriented syllabus or whether the analysis is merely a pre-syllabus for theselection of open-ended tasks that will form a whole task-based syllabus in themselves.If analysis from a discourse point of view is to the pre-syllabus for a task-based one,then we would strongly support Aston (1 988) in his view that an analysis based oninteractional language is just as important as one based on transactional uses of language.Real data show that the two types of language use rarely occur discretely (see McCarthyand Carter 1994: 117-24; Iklton 1988; McCarthy 1991 : 136--7). For an interactional viewof language to have an input into task dcsign, the understanding of how natural conversationworks, how speakers/writers orient towards rcciprocity and convergence, how they do sousing systcmatic resources such as lexical cohesion and how features such as topicmanagement are realized arc all central. It is hcrc, we feel, that syllabus designers have mostto learn from what discourse analysts can offer.Designing tasks is no easy matter, and much useful literature exists which treats withmore rigour than space allows us hcrc thc factors which can make or break tasks (seeespecially Nunan 1989). It does seem worth underlining here, howcvcr, that tasks whichpromote only or mainly transactional uses of languagc (e.g. information-gap tasks) areunlikely to engage learners in a full range 01’ discour. trategies. Discourse strategies, wehave argued, are concerned with human heings presenting a picture of themselves, not justconveying information to one anothcr.Thcrcforc, if‘gaps’ or ‘problems’ arc the core featuresof tasks which motivate their completion, thcn we need to build in much more than justinformation or ‘opinion’ gaps (see Aston 1988: 192 -9 for a critique of information- andopinion-gap approachcs). Gaps in rapport, prohlcms of sensitivity, convcrgrnce towardsacquaintance or friendship, gaps in self-image, problc~ms of face, all of these will assume asmuch importance as gaps in placcs on a map, or gaps in agrccing on where to spend aSaturday night, the stock-in-trade of many prcscnt classroom tasks. Tasks can fulfil some ofthese interactional criteria by dclibcrately ‘designing in’ unpredictable reactions, ‘diflicult’participants, goals where conversational well-king is morc important than informationaltransaction, and so on.An example of an attempt at building into a task interactional constraints demandingpoliteness and convergence strategies, taken from the International Certificate Conference’steacher-training programme for teachers intending to use their discourse-strategy and taskbasedsyllabus (ICC 1986), involl participants in a consensus activity to agrce on thearrangement of furniture for a school opcn-day. Much of the task is transactionally oriented,culminating in leaving instructions for the school caretaker to execute the furniture plan.However, the person who role-plays the caretaker is required to take offence at the tone ofthe instructions and the task therefore cannot be complctcd until oil has been poured ontroubled waters and ‘affective’ convergence has been achieved, even though cognitiveconvergence is already prcscnt in the written instructions for the furniture plan.This is onlyone small example, but it shows how task dcsign can attempt to replicate a wider range ofdiscourse conditions, and how the ‘pre-syllabus’ might fecd into thc constructional syllabusin a more controlled way, if the dcsirc is to follow a task-based approach.5 ConclusionWe hope that the discussion in this chapter has pointed to the following conclusion: thatawareness of discourse and a willingness to take on board what a language-as~discourse view

DESIGNING THE DISCOURSE SYLLABUS 63implies can only makc us better and more efficient syllabus designers, task designers,dialogue-writers, materials adaptors and evaluators of everything \ve do and handle in thcclassroom. Above all, the approach we have advocated enablcs us to be more faithful to whatlanguage is and what people use it for. Thc momcnt one starts to think of language asdiscourse, the entire landscape changes, usually, for ever.BibliographyAston, G. 1988 Learning Comiy. Bologna: Editrice CLUERRclton, A. 1988 ‘Lexical naturalness in native and non-native discourse’. English LangnugcResearch Journal (ns) 2: 79 105British Council 1983; 1986 English Teaching Prof;rIe on .h‘ala,vsia. London: British CouncilCanale, M. 1983 ‘From communicative competence to Communicative language pedagogy’. InRichards, J. C., Schmidt, R. (eds) Language and Communication. London: Longman, pp.2-27Canale, M., S\vain, M. 1980 ‘Theoretical bases of communicative approaches to .language teaching and testing’. Applied Linguistics 1 : 1 47Candlin, C. N. 1976 ‘Communicative language teaching and the debt to pragmatics’. InKameh, C. (ed) Georgetonm Universiy Round Table on LungLiuges ant/ Linguistic

couldChapter 5Guy CookTHE USES OF COMPUTERIZED LANGUAGECORPORA: A REPLY TO RONALD CARTERIntroductionOMPUTERIZED LANGUAGE CORPORA have inspired somc of the mostC important insights in rcccnt linguistics. They have sho\vn us, for example, that actuallanguage us(’ is less a matter of coml)ining alistract grammar rules vith individual lexicalitems, and more a matter of collocation; that thcrc arc grammatically possible utteranceshich do not occur, and others lvhich occur ivith tlispro1,ortionatc frcqwncy; that in.tcmatic descriptions of occurrcnccs, grammar and lcxis cannot lie as casily separated ashave Iiccn traditionally, either in pedagogy or in linguistics. Ronald Carter is right tosuch insights ‘cxciting’, antl his o\vn \vork with Michael McCarthv on the CANCOIlEcorpus has added to them. As his articlc (1998) illustratcs vcry \vcll, the grammaticalconstructions \ve find in actual con\.crsations arc not al\vay accountctl for in traditionalgrammars.Clearly all thcsc findings arc important, and thcy do have implications for languageteaching. The problcm is, howcvcr, that somc corpus linguists (c.g. Sinclair 1991, Stubbs1996) ovcrrcach thcmsclvcs.Thcy talk as though thc cntirc study of languagc can I x replacedby the study ofthcir collcctions, antl as though all important insights \vi11 emerge only fromautomatic scarchcs of their data antl no\\ hci-c clsc. Clearly such solutions to thc study ofcomplex human phenomena cxcrt a good deal cductivc pvcr. If the tratlitional concernof linguistics language in all its cultural and 11 ological complexity ~~ be replacedhy a neat computer hank of data, life \vould Iic much siinplcr.Yet thc leap from linguistics to pcdagogv is as Carter realizes ~~ far from straight-forward. He is not one ofthc extremists, antl his paper is, for that rcason, a \vorth\vhilc andinteresting eontriliution to language teaching. He proc (1s cautiously, providing someinteresting ‘rcal’ data, antl pointing out significant diffcrcnccs Ixt\vccn actual antl texthookEnglish. He docs not say one should replace the othcr. In his view., materials should lieinfluenced by, hut not slaves to, corpus lintlings. (In this hc sccms to agree with the view ofSummers antl Rundcll (1995) that pedagogic matcrials should be ‘corpus based not corpushound’, and to disagree lvith the CORUILL) slogan that they should lie ‘corpus driven’(Stuldis 1997).) This is eminently rcasonalilc, though for that very rcason not particularlyradical. My problem with \vhat Carter says is that he seems a little hcsitant ~ or perhapsunwilling ~~~ to say \\here he stands. Ilocs he reject the fundamentalist views ofthose linguistsand languagc tcaching theorists for \z hom corpus findings arc the only source of truth?

THE USES OF COMPUTERIZED LANGUAGE CORPORA 65My first aim in this rcplv is to pursue some of the shortcomings of corpus-tlri\enapproachcs which I think Carter avoids confronting. I shall also consider some of the moreextreme applications of corpus findings to language teaching. My argument is that there isan important difference bet\\ een the hard antl soft linc approachcs, that the former, byappcaring to offer yet another easy ‘scientific’ solution, can do immense damage, and that\vc all, including Carter, \voultl do \cell to consider more precisely \vhether \z e think corpustindings mcrcly add a new dimension to earlier approachcs, or replace them.Uses and abuses of corporaA number 01’ Ialse conclusions can lie reached about corpora. It is often assumed, forexample, that as a description of language brhaviour, thcy arc the only valid source of factsabout language; the same as a description of language in the mind; provide a goal and a routefor language 1carning.Thei-e is much in computerized corpus analysis to makc us reconsiderreceived ideas about the learning, representation, and use of language. But whcrc pcdagogyis concerned, corpus statistics say nothing about immcasurablc but crucial factors such asstudents’ and teachers’ attitudes and expectations, the personal relationships lxt\\their o\cn wishes, or the diversity of traditions from which thcy comc. Conscqucntlycomputer corpora ~ cchile impressive antl interesting records of crrtain aspects of languageuse can ncvcr bc mort- than a contribution to our understanding of cffcctivc languagcteaching.Corpus asfactEvcn as a rccord of‘facts’ computer corpora are incomplete.They contain information alioutproduction but not about reception.Thcy say nothing about how many people have read orheard a text or utterance, or how many times.’Thus a memo hastily skimmed by one personantl consigned to the wastepaper Iiasket counts equally Icith a tabloid headline read bvmillions, or with a text, such as a prayer or poem, which is not only often repeated but alsodeeply valued. Occurrencc, distrihution, and importancc, in other \vords, are not the same.This applics to whole tcxts, hut also to shorter units. Some phrases pass unnoticed precisclybecause of thcir frequency, othcrs strikc and stay in the mind, though they may occur onlyonce. And becausc tliffcrcnt intlividuals noticc tlil’ferent things, such saliency can ninclutletl in a corpus.The same is true ofa whole host of aspects of language use: metaphors,speech acts such as apologies or compliments, interactive cvcnts such as intcrruption orIs of formality. They arc not ‘facts’ but matters ol‘is a truism to obscrvc that there is no straightfor\vartl correlation tictvuse, the intentions thcy had in thcm, and the interpretations which other pcoplc put uponthcm. If this were not so, there would be no disputes ovcr the mcaning of what people say.Corpus as recordCorpora are records of language lichaviour. The patterns which rmc’r-gc in that lichaviourdo not ntxcssarily and directly tell us how people organizc antl classify language in thcirmvn minds? and for thcii- o\vn use, or how language is best systcmatized for teaching.Linguists’ analyses ol’thcse data are not ne arily users’ analyses, or thosc ivhich arr mostuseful to teachers and Icarncrs. Thcy arc just one kind of fact. The ways in \vhic.hb nrainmarians antl pcdagogucs ha\ c organizctl thcir matcrial in grammars, syllaliuscs, antldictionaries ~ are also facts about language. So are people’s emotional lielirfs that one type

asveryremainsof language use is better than another-. We should not promote some kinds of facts at theexpense of othcrs.Corpora are only partial authorities. The cumulative languagc experience of anindividual, though lcss amenable to systematic ac 1, far larger anti richer. Even athree hundred million word corpus is cqui\ alent to only around thrcc thousand books, orperhaps the language experience of a teenager. This is why our intuition (in cffwt ourrandom antl incomplctc, access to our total cxpcricmce of the language) can still tell us factsabout the language which cannot 1)c evidenced by a corpus (Witltlowson 1990). Forexample, the canonical forms of sayings antl provcrI)s occur 1 cry rarely in corpora, thoughthey are obviously well knoum by ~icoplc (Aston 1995). Such omissions, hobvevcr, arc notmerely a quantitative issue; they cannot lie rcmcdicd simply by making corpora larger andlarger. They arc inrvitalilc in an approach lvhich accepts only one of the three sources offact about language: observation; and ignores or villainizes t\vo othcrs: introspection andelicitation. For there arc aspccts of language which arc knolvn but not used. Corpus linguistsare fond ofobscrving that thc commonest uses of words are not the same as their standarddefinitions. ‘I bet’, for example, is more rarely used in the sense of‘wagcr’, and most oftenin the sense of ‘suppose’(Sinclair 1987: xi i). Rut this unsurprising olxcrvation does not atall invalidate thc view that ‘\vagcr’ is a ccntral prototypical meaning for many speakers toIvhich more colloquial uses arc attachcd. (And indeed, thr ‘wager’ meaning is still given asthe first meaning of ‘bet’ in the Coliuiltl dictionarv.)Description and prescriptionBut let us assume for thc sake of argument that corpora arc accurate records of languagebehaviour, that they do catalogue antl I-cwal all the important ‘facts’ ahout the language.Thequestion then arises as to lvhosc language hchaviour is accurately recorded ~~ antl thequestion takes on a particularly sinistcr signifcancc whcn the corpora in question start beingused not as data for dcscriptiw linguistics, but as sources of prcscription for TESOL. Forthc answ-er to the question is (as Cartcr scctns painfullv awarc) that corpora arc primarilyrecords of native speakers’ languagc Iichaviour. ‘Kcal’ languagc in cffcct means nativespeakerEnglish, and the only language excludcd from this category (apart from the invrntetlcxamplcs of linguists antl textbook writcrs) is that usetl to antl by language learners. To hiscrcdit, Carter confronts this issue, and intends to remedy it. But the proposed addition of‘a wider variety of international Englishcs’ will not solve the problem. This will only addother standard Englishes as spoken by their o\vn native speakers.And then a second question arises to which Cartcr explicitly refers, hut does notanskvcr. Why should the attested language usc of a native-speaker community bc a modelfor learners of English as an international language? If a certain collocation occurs frequentlyamong British or American English speakers, must it also lw used by the Japanese or theMexicans? This is where lve encounter an easy slippage from description to prcscription,in cffcct making the former into the latter.Thc English which is used by one or more nativespeakercommunities, it is implied, ought to lie the English learned for internationalcommunication.The ready-made lexical phrases \zhich corpora reveal to he so frequent in native speakeruse are moreover ~ Carter readily recognizes ~ often culturally specific and loaded.In deploying such units, thc foreign speaker is very likely to produce corpus-attested butcontextually inappropriate language. (This is why attempts to teach set phrases arc likelyto he as tragicomically disastrous in lexical syllabuses as they \vere in functional ones.) Inthe terms of Hymes’s (1 972) four parameters of communicative competence, corpus-driven

THE USES OF COMPUTERIZED LANGUAGE CORPORA 67language teaching always risks stressing what is actually done at the expense of what isappropriate in a particular context.Pedagogical issuesIn an extensively quoted, and in itself excellent, essay liy Pa\vlcv and Syder (1 983) on nativclikesclcction and fluency, corpus-liascd language teaching finds a source of inspiration,providing a potential link lietwccn the facts of language Iieliaviour antl a theory of howlanguage is acquired and processed in the mind. Here is the claim that mature native spcikcrs(for this is whom the essay is explicitly about) have ‘hundrcds of thousands’ of institutionalizedlexicalized or semi-lexicalizctl units in memory. Though many of these units can beanalysed grammatically, the likelihood is, so the argumcnt goes, that they arc often producedand understood holistically. Nativc speakers acquire, represent, and process language inlexicalizcd chunks as \\ell as grammar rulcs and single \vords.Yet it hy no means follows that foreign lcarners must do the same.Thcy may not \rantto study language in this way; they may live within culturally tli\ ersc pedagogic traditionsnot compatilile with this approach; they may not aspire to or nccd native-like English; thcymay not have as much time available as native-speaker children; above all, as adults withconscious learning strategies available to them, they can choose. And \vhy should thchoosr to continue vicwing the language as grammar structures and slot-filling wordmay not lead to native-like English, but it may lead to communicative and expressive English.It may be learnt morc quickly. And it will avoid the tedious rote learning of mundanephrases, or the bc\z.ildering refusal to teach grammar, which arc the inevitable consequencesof an overemphasis on ‘lexical chunks’.Yet cvcn if appearing native-likc were accepted as the goal of languagc learning, it \voultlstill not follo\v that frequency and tlcsirability arc the same. Thcre is a hidtlcn irony in thedogma that frequcnt native-likc collocations are the best model to imitate. It is that e\en\i.ithin the native-speaker community it is often the infrequcnt \vortl or expression which ismost pon-cdd antl most communicatively effective, antl therefore most sought after. Thisis also \\ hy foreigners’ speech is often expressive and striking. tbth for native and non-nativespeakers there is an altcrnati\e goal to seeking the most usual, the most frequent or, in short,the most clichktl exprcssion. It is the goal of rich, varied, antl original language. Amongnative speakers it is unusual language kvhich is valued. Should non-native spcakers lie trcatcdtliffercntl y ?This lcatls to the important point that not all types of language arc cquallp valued, eitherby native spcakers or foreign learners. Something is not a good model simply liecauseit occurs frequcntly. A good dcal of actual language use is inarticulate, impoverished,and inexpressive. Inevitably, because onc cannot teach cvcrything, part of the job ofteachcrs and coursc designers is to sclcct the languagc use which they \4.ish their studentsto emulate. Many foreign language students have strong feelings about this too. They tlonot Lvant to learn just any English because it occurs in a corpus, and it is patronisingto overrule them. In advocating selection and modelling of corpus data, in the use of literaryrather than transcribed dialogues, and in his recognition that one of the topics in his authenticdata (‘straggly hair’) may have a limited topic life in many classrooms, Carter seems toagree.To be corpus driven, in short, deprives cvcryone (native antl non-native spcaker alike)of the opportunity for choice antl to make their own impact on the language. Corpora areinevitably records of what has happened rather than what is happcning.They prcscnt us witha,fuit accompli, a fixed product rather than an open process.

moclcratc,to68 GUY Cool

‘modclling’ivhichTHE USES OF COMPUTERIZED LANGUAGE CORPORA 69all the culturally various pedagogic traditions in hvhich they \vork antl study, arc, as Ixwis\vould put it, ‘hvrong’.Conclusion1 have contrasted throughout this rcply what I scc as the soft and the hard line \ ie-\vs of therelevance of corpus findings to language teaching. In thc one, \IC have the wicc ofmoderation urging a limited application ~ as Carter calls it ~ by \ii-tucof its very reasonableness does not amount to anything very radical. In the other, \vc havethc stronger view: cvangclical, authoritarian, antl dismissive of tradition, assuming that alittle of the latest linguistics theory is all that is needcd to changc- the coursc of languagcteaching.I believe that if Carter lvcrc to follo\v his arguments through to their conclusion,he too \vould explicitly reject, as I do, the morc extreme versions hth of corpus linguisticsantl of corpus-driven language teaching, Rut it is b: no means clear whether he docs so.Notes1 This point has bcrn made hy corpus linguists thcmsclves (Francis 1979, Stul)lw1996: 1 I), hut the point is not adequately taken on board, either in corpus constructionor analysis.2 Stubbs (1996: 21) tells us that thr. ‘deep patterning’ rcvcalrd by corpuh anal!‘bcyond human ol)servation antl memory’,3 This issue is clouded by snolhish and chauvinistic claims that a particdar national orsociolcct is lxttcr than anothcr. But this is not a necessary componcnt of the notion thatcrrtain usages ~ literary, xvrittcn, or simpl\ eloquent and elegant one^ arc morcdesirable models than others.ReferencesAston, G. 1995. ‘Corpora in language pdagogy: matching theory antl practice’ in G. Cook andB. Scitllhofcr (cd.;.). Principle cind Prcictic-e in Applied I.inguittic.>. Oxti)rd: Oxfo~-tl Uni\ crhityPress.Cat-trr, I

70 GUY COOKSummers, D. and M. Rundell (etls.). 1995. Longman Dictionary of Contemporary English. London:Longman.West, M. P. 1926. Learning to Read a Foreign Language. NcwYork: Longmans, Grccn.Widdowson, H. G. 1990. ‘Discourses of enquiry and conditions of relevance’in J. E. Alatis(etl). Linguistics, Language Teaching and I.angnup Acq~iisition. Washington DC: GeorgetownUniversity Press.Willis, D. 1990. The Lexical .$llabns. London: Collins.

Chapter 6Ann Hewings and Martin HewingsAPPROACHES TO THE STUDY OFDISCIPLINARY VARIATION IN ACADEMICWRITING: IMPLICATIONS FOR SYLLABUSDESIGN1 IntroductionN RECENT YE A R S , S Y L LA B U S E S for academic writing in higher educationI have increasingly focused on teaching students about the features of differing writtengenres. So, for instance, we find published material on laboratory and technical reports (forexample Dudley-Evans, 1 985), expcrimental rcscarch reports and other research papers(for example Wcissberg and Buker, 1990), theses and dissertations (for example Andersonand Poole, 1994) and essays (for example Roberts, 1997). The gcncral motivation of thisapproach is thc need to offrr appropriate dcxcriptions and models of generic texts so thatthc students’ ability to understand and produce them is improved. More specifically, studentsarc taught about thc textual features, both tcxt structural and sentence-level, that arecharacteristic of‘ each gcnrc.While this represents a valuablc development from earlier approaches which treated‘academic writing’ as an undifferentiated, homogeneous entity, it is important to recogniscthat variation is found not only from genre to genre, but also within genres. Evidcnccis accumulating that single genres vary over time (Bazerman, 1988; Dudley-Evans andHenderson, 1990; Selager-Meyer, 19959, vary from one cultural context to another (Taylorand Chcn, 1991), and varv from discipline to discipline (Berkenkottcr and Huckin, 1995;Prior, 1998).This essay is primarily concrrncd with the third of these and, in particular, thc methodsthat have been adopted for the study of disciplinary variation and the implications of findingsto date for syllabus design. Knowledge of disciplinary variation is liccoming especiallyimportant with the growing trend towards inter- and multi-disciplinary study in highereducation so that students may be required to work within a number of disciplines whichhave different views on the naturc of academic writing. We begin by reporting threc arcasof applied linguistic investigation which have explored the question of disciplinary variationin rather different ways and with rather different implications for syllabus design. First, wepresent Swales’s approach to genre anahis and discuss studies of disciplinary variation basedon this approach, in particular those which have explored variation in thc academic rmcarch

72 ANN HEWINGS AND MARTIN HEWINGSarticlc. Second, we outline lvork \vhich has cxamincd metadiscourre in academic \vritingthat is, the part ofa tcxt Lvhich helps thc rcadcr organisc, classify, cduate and react to thepropositional contcnt (Vandc Kopplc, 1985) and the \lay this varies in texts taken fromdifferent disciplines. Third, i\c report studies of one clause-level feature of text, thr~qrammatical siihjcct, that have tlcmonstratcd its significance in reflccting how m.ritcrs reprcscntdata, previous research antl thcmsclvcs in the tcxt, and how this varies across disciplines. Adiscussion of the implications of the findings of such work for academic writing syllabusesconcludes the essay.Throughout, our attention is primarily on gcm-es Ix-oduccd within an academic context,either clussroom genres those produced IIV students for purposes of asscssmcnt, such asessays, dissertations and thcscs, 1al)oratot-y and case study reports, and litcraturc reviews ~or prof;ssional presthc tcxts by which scientists antl scholars communicate \vith otherscientists antl scholars, such as conference papcv-s, research articles, monographs, technicalreports, working papers, and grant proposals.2 Approaches to the study of disciplinary variation2. I GenreWithin the context of English for Spccific I’urposcs, the teaching of academic lvriting hasIxcn greatly influencctl by the apliroach to gcnrc arising from work by John Skyales (forexample 198 1, 1984, 1990). ’ This approach considers a non-fictional gcnrc to be:a rccognizahlc communicati\ c cvcnt charactcrizctl by a sct of communicativepurposc(s) idcntificd antl mutually understood by mcnil)crs of thr professional oracademic community in \vhich it regularly occurs. Most often it is highly structuredand conventionalized with constraints on allo\vahlc contributions in terms of theirintent, positioning, form and functional valuc.(Bhatia, 1993: 13)The primary critcrion, thcn, by \vhich tcxts arc consitlcrctl to lie of tht. samc genre iscommunicative purpose. If tcxts have tliffcrcnt communicative purposes, they are likely tobe of a tliffcrcnt genrc, antl it is this shared communicative p~irpse that produces theconvcntionalisctl form of the gcni-e antl its chai-actcristic linguistic features. The analysis ofa genre for pedagogical purposes involves the identification of thcsc regularities in tcxtorganisation and Icxico-grammatical features antl, in addition, an attempt to explain howthcv relate to the tlisrource cornmnniy (Hcrzbcrg, 1986; S\valcs, 1990; Bizzcll, 1992) Tvithin\vhich the genre is produced. Such analysis can thcn bc converted into syllabuses andmaterials that aim to teach students aliout tcxt organisation antl relevant language forms.S\valcs’s (1981 , 1990) pionccring lvork itlcntifictl a set of ‘movcs’, and ‘steps’ withinthem, which were rccurrcntly found in the introductions to research articles in order tocontcxtualise an author’s o\vn research. A move is a unit \\ hich is rclatrd both to the purposewriters haw antl to the contcnt they nish to communicate, \vhilc a stcp is a component ofa move which is a morc tlctailcd option availaldc to the kvritcr in setting out a move (Dudlcy-Evans and St John, 1998: 89). Swales proposed (1990: 141) a three-movc model for articlcintroductions (modifictl from four in his 1981 lvork):

DISCIPLINARY VARIATION IN ACADEMIC WRITING 73Move 1Move 2Move 3Establishing a territoryStep 1 Claiming centralityunti/or Step 2 Making topic gencralisationsclnti/or Step 3 IievicLving items of previous researchEstablishing a nicheStep 1A Counter-claimingor Step IB Indicating a gapor Step 1C Question-raisingor Step ID Continuing a traditionOccupying the nicheStep 1A Outlining purposesor Step 1B Announcing prescnt resrarchStep 2 Announcing principal findingsStep 3 Indicating research article structure.Typically, academic discoursc communities are bound togcther by subject matter, antlprofessional academic writing is seen as adding to the body of kno\vlcdge Lvhich is at thccore of the discipline. In addition to disciplinary knowledge, the way subject matter isdiscussed the genrc conventions used is also of importancc. To hecome ‘good academicxvl-iters’, students need to become a\varc ofthesc conventions, that is, how the tcxtual formsand communicative functions arc related to the expectations of the academic communityto which they belong.Swalcs’s movc and step approach has been used not only to identify the characteristics ofparticular genres, Iiut also to compare texts of the same genre but from different disciplines.For example, in the prcliminarics to an invcstigation of active and passive vcrb forms in t\voastrophyics journal articles, laronc et a1. (1 998) notc that S\valc,s’s (1990) overview of theorganisation ofthc rcsearch article as having an ‘hourglass’ shapc is not applicable to articlesin astrophysics. ‘I Iourglass’ articles hegin with a broad overview of the ficltl, narrow thefocus tlo\vn to a specific area of interest Lvhich is then expcrimentcd on in somc way, andconclude with a widening-out of the discussion to rclatc findings to t,roadcr issues relcnntto the ficltl. Astrophysics papers, however, are consitlcretl hy‘l-arone et (11. to have an ‘invertedpyramid’ construction in which the focus of‘ the paper is gradually narro\\ccl down,beginning with general physics, through the particular phcnomcna to explain, thc specificphysics of relevance, spccific equations, to a specific solution.Thc reason, thcv argue, is thathvhile thc hourglass is a satisfactory rcprcsentation of reports of cxperimcntal studies,astrophysics attends to suliject matter \vhich cannot lie cxpcrimentcd on, so that papers inthe discipline prescnt logical arguments rather than expcrimcnts.A number of studics have examined how scctions of rcscarch articles vary acrosstlisciplines.Thc typical sections of research articles arc an introduction, a mcthods sectionwhich explains the procedures undertaken (often experimental procedurcs in the casc ofscientific research articles), a report ofthc results ofthc procedurcs, antl finally a discussionof these results antl their significance. Brett’s (1994) starting point is Swales’s (1990: 175-6)proposal that tlisciplinary differences in research articles are likcly to lie in methods andresults sections rather than introductions antl discussions. He examines results sections inresearch articles from sociology and observes certain communicative categories withinthem, such as his Substuntiation of Findings antl .Yon-vu/itlation of’ Findings, not prcviousl)

74 ANN HEWINGS AND MARTIN HEWINGSdocumented as appearing in rcsults or discussion sections (for example in Relanger, 1982;Dudley-Evans, 1989; Hopkins antl I>utllcy-Evans, 1988; Weissberg and Kukcr, 1990). Herelates thcsc to the methods of quantitati ,ociological research in which ahstract conceptssuch as ‘ethnic identification’ or ‘satisfaction’ arc prcscnted as numcrical data. These dataarc then manipulated using accepted statistical techniques, and the resulting statisticsintcrprcted to produce deductions about human tiehaviour. Not only are the effccts on thedependent variahle of the most significant indcpcndcnt variables tliscussetl, but so arethe effects of other variables. The second of thcsc, I3rctt found, took up more space thanthe discussion of the most significant findings, kvhilc thc effects of secondary variables arcassessed as either supporting (Substantiation (rf Fintiings) or lcssening the validity of ( AJonvalidationo_f‘Fzntfings) the main findings. As in Taronc et al.’s work, then, Brctt suggests thatthc suliject matter and the mcthotlologics deployctl intluencc the constitution of writtentext within the tlisciplinc.Holmcs (I 997) undertakes a comparative study of the organisation of the discussionsections of articles presenting original research from the disciplines of political science,sociology and history using a Sudcs-typc mow analysis. He identifics a total of eight moves(Background information, Statement OJresult, (Iln)expectcd outcome, Rcfirence to previous research,Explanation of unsatisfactoy result, Gcnerali.sation, Recommendation, and Outlining parallel orsubsequent derdopmcnts). In comparison with similar \vmk on articles in the natural or hardsciences (Pcng, 1987) Holmes (1997: 332) finds that discussion sections in the socialsciences, as represented bv political scicnce and sociology, are less complex, employing arestricted repertoire of moves, jet arc less predictable. On the other hand, social scienceintroductions have ticcn found to display ‘qreatcr complexity than those in the hard sciences(Crookcs, 1986; Holmcs, 1995). On the basis of such cvidencc, research articles in thesocial sciences \vould seem to display greater complexity and elaboration at the bcginningthan at the end while the revers

DISCIPLINARY VARIATION IN ACADEMIC WRITING 75Such studies, thcn, suggest that the nature of a particular discipline, as indicated in itssubjcct matter, its methods of investigation, the amount of previous research in the ficld,the level of consensus on agrerd knowlcdgc, and the degree of bureaucratization in thediscipline may he reflected in its generic convcntions.This has direct implications for syllabusdesign. It reinforces the need to move away not only from academic writing as ahomogeneous entity, but also from homogeneous genres. While information on, forexample, the sections of the research article may be useful as a prcliminary, students alsoneed to be made aware of the specifics of what to include, what to cmphasisc, antl what tocxclude within each scction for their own particular disciplinc.2.2 MetadiscourseA rather different approach to the in tigation of the relationship lictwcen disciplinarycommunities and their tcxts is found in studies of metadiscourse in academic writing. Adistinction can be made bctlveen the propositional content of a text, its information or subjectmatter, and metadiscourse, that part of the text which helps the reatlcr organisc, classify,evaluate and react to the propositional content (Vantle Kopplc, 1985). The elemrnts ofmetadiscourse have heen divided (scc, for example, Hyland 1999a) into those which, in theterminology of systemic functional grammar, serve a textual function and those hvhich scrvcan interpersonal function. Halliday (1 973: 66) descrilxs the textual function asan enabling function, that of creating text . . . It is this component that enables thespeaker [or writer] to organise what he is saying in such a w’ay that it makes sense inthe contcxt antl fulfils its functions as a messagr,while the intcrpcrsonal function is said to includeall that may he understood by the expression of our olvn personalities and personalfeelings on the one hand, and forms of interaction and social interplay with otherparticipants in the communication situation on the other hand.Mctadiscoursc, thcrefore, allows hvriters not only to show ho\v a text is organised and ho\vdifferent parts of the text arc rclatetl, but also to express thcir attitude towards the subjectmatter of thc text and towards the intcndcd readership. Academic disciplines haveconventional ways in which writers are allowed both to present thcir arguments and toreprcscnt themselvcs, and this is achicvcd mainly in thc metadiscoursc in text. It is throughthe study of metadiscourse in the texts of a particular disciplinary community, therefore,that the characteristics of that community can bc explored. Studies of metadiscourse inacademic text have looked at cultural antl gender variation (Crismorc et a]., 1993; Mauranen,1993) and the use of metadiscourse in particular academic genres (Hyland, 1999a; Hcwings,1999). Howc\w, it is Hyland’s (1 99911) work on metadiscourse and disciplinary variationthat is of main concern here and reported below.In a study of fifty- six research articles takcn from eight disciplines (microbiology,physics, marketing, applied linguistics, philosophy, sociology, mechanical engineering, andelectrical engineering), Hyland considers variation in the writers’ stance. Stance is part ofthe interpersonal component of metadiscourse and defined asthe ways that writers project themselves into thcir texts to communicate theirintegrity, credibility, involvement, and a relationship to thcir subject matter and thcirreaders.(19991-3: 101)

76 ANN HEWINGS AND MARTIN HEWINGSIt iy considered to hale fi~c main components:hedges (c.g. po.ssib/e, might, perhaps, belicr.c), through which the Ivritcr Ivithholds fullcommitment to a proposition;emphutic.~ (c.g. it is ohriotis, chfinitclv, o/‘cotir.sc), through \vhich the writer cmphasisesthe force of a proposition;attituck markers, concerned \\ ith the \vr-itcr’s attitutlc to \\.hat is said and signalledbv such devices as uttitutle r~erhs (c.g. I ci

DISCIPLINARY VARIATION IN ACADEMIC WRITING 772.3 The grammatical subjectA third approach to the stud! of tlisciplinar; \ariation has de\elopcd from the vork ofMacDonald (1992, 1994) on the p~~mmaticill ruhlect Thc subject, or grammatical suhlt.ct(G$), has all? a)\ been rccogniwd a4 a signhcant component of English in both ti aditionaland functional grammars. The t! pica1 English sentence pattern IS 5ul1~ect Vcrli Oblcct,lor exampleSu7anne playcd her I iolinSirhlect T‘erh Objectand the subject is an obligatory clement in all sentences with the exccption of inipcrativcxs.MacDonaldk rationale for focusing on the GS \vas that it isthe constituent defining the topic of the scntcnce ~ that \vhich the sentence is ‘about’antl which it presupposes as its point of departure.(Quirk et ill., 1985: 79)Thus the GS is important for determining \\,hat a \vriter is writing ohorit antl ho\v theyrepresent data, previous research and themselves in their texts.In her invcstigation into disciplinary differences in professional \vriting in the humanitiesand social sciences, MacDonald developed a mcthod of classifying GSs on the his ofIvhether the); contriliutc to the content of a text or hvhethcr they are more concerncd lvithconsolidating knowledge construction in a discipline ~ that is, the huilding of knowledgeon foundations laid hy other rcscarchcrs. For some disciplines, the consolidation ofknowledge is of such importance that it is reflcctcd not only at text level hut also in scntcncelevelchoices. Analysis of the GS is used to demonstrate the rhetorical practices of suchdisciplines. For example, acadcmic texts with sentences beginning ‘Work 1,: Jamcs . . .’ or‘Experimental cvitlencc . . .’ are clearly liuiltling on a foundation constructctl for thcdiscipline hy those within it.The GSs can thus lie said to have an epistemic focus, one whichis concerned \vith methods of study and the validity of knowlctlge claims. bor otherdisciplines, the people, things antl events that constitute the phenomena or content that arcstudied arc foregroundcd. For example, in literature studies, ‘Shakespeare’s plays’ might\vel1 lic the content or subject that is being ivrittcn about, and where thcsc \?-or& occur asthe GS thcrc is said to be a phenomenal focus. Within this Iiroad two-fold division, MacDonaltlrccognises a further disciplinary characteristic. Disciplines such as those in the humanitiesare more concerned with specific ~icople, placcs and events, whereas those in, for examplc,the social sciences rely more on gencralisations and abstractions. This is again reflected inthe choice of GS, with a cline existing hetlveen those GSs Lvhich represent phcnomena attheir most specific and individualistic through to those which are most abstract.On this basis, MacDonald (1992) has developed a classification system for GSs, amodified version of tvhich is summarised inTable 6.1 .Thc \-due of this \vork for the study of disciplinary variation becomes apparent if \vccompare findings from psychology, history and literature (from Macl)onald, 1992) antlwildlife hchaviour, conservation biology and legislative history (Samraj, 1995). Macnonald’s(lata are hascd on an analysis offour journal-length articles in cach discipline, \vhile Samrajanalysetl six student papers from thrcc different courses which formed part of a US Mastcr’sprogrammc in environmental studies. The figures inTablc 6.2 show the percentagc averagedistribution of cach of thc seven classcs of GSs in the texts.

78 ANN HEWINGS AND MARTIN HEWINGSTable 6. IA classification system [or thr grammatical subjectSummary of grammatical subject classificationPhenomenalPhenomenal grammatical subjects are thosc which deal u ith the material that a researcher or writerstudics or writcs about.Class 1 (Particulars) contains nouns rcfcrring to spccitic people, placcs, or ohjrcts, usually named(e.g. the USA, Stalin, Europe, the storm).Class 2 (Non-Spcxcific Groups) contains gcncralisctl or grouprd nouns (..e. large ureas ofthe country,gullies, downland soils, the tsetsepy poptilotion).Class 3 (Attributes) contains the most abstract, least material nouns in the phenomenal category.They arc nouns cxprcssing properties, attributes, actions or motivations of the people or things in1 and 2 (e.g. agricultural practices, rainfall intensities, planning).Epi,stemicEpistcmic grammatical subjects rcprcscnt thr concepts, catcgorics, abstractions or methodologicaltools the researcher uses to reason ahout the subject.Class 4 (Research Matters) contains rctrrcnccs to the constitucnts of rcsrarch and the activities ofresearchers such as data collection and analysis, gcnrrating itlcas and comparing and contrastingtiitfcrcnt theories (c.g. the eyoations, the,Fnal hypothesis, it has been argued that).Class 5 (Researchers antl Studies) contains rcfercnccs to published research and to writers in thefield (e.g. Pinch and Story, /YY2. I).Class 6 (Audience) contains suhjccts like thr gcncraliscd we (hut not thc actual we which refers toauthors) and one oryou.Class 7 (Discoursal) contains suhjccts which refer to the text itsclf, cithrr in whole or in part, andrhctorical questions used to organisc the discourse (c.g. this cssoy, Figure 2, Why are mobile homespopular uith pensionen?).An examination of the tablc shows some very large variations in the types of GSsfavoured by different disciplines antl the grouping of these into phenomcnal or cpistemiccategories. Disciplines with more epistemic subjects (classcs 4-7) foreground researchmethods, inferences and findings rather than the phenomena that are being studied orwritten about (classes 1-3). Some disciplines are clearly more phenomenal in their focusand some more epistemic. Psychology, for example, has a total of 62 per cent epistenlicgrammatical subjccts, whereas literature has 84 per ccnt phenomenal. At the levcl ofindividual classes, legislative history and literature usc more class 1 (particulars) than theother disciplines. This undcrlines thcir conccrn with specific pcople, places and objects. Incontrast, psychology uses less than 1 per ccnt of class 1 GSs. Instead, it favours non-specific,more generalised referenccs to phenomena as found in class 2. Most disciplines, exceptliterature, but especially the two histories, have fairly high numbers of non-specific groupsof people, places or things. Class 3 GSs are frequent in all disciplines except psychology.They arc nouns which express properties, attributes, actions or motivations of thc peopleor things in classes 1 and 2.Hewings (1999) has used MacDonald’s techniques to examine writing developmentamong undergraduate students within the discipline of geography at a British university.

DISCIPLINARY VARIATION IN ACADEMIC WRITING 79Tuble 6.2 Average distribution (Yo) of GSs in tliffcrcnt disciplines antl sub-disciplincs (data fromMacDonald 1992 and Samraj 1995)GS Psychology History Litcraturr Wildlife Conservation I egislati\ cclass hchaviour' hiology history1 01 6 30 1 7 352 27 44 10 30 28 303 I1 26 44 29 44 284 49 15 7 18 9 I5 12 6 5 18 8 26 1 3 4 ~/Comparing essays written by students in the first and final (third) years of their programme,she found a substantially higher proportion of phcnomenal to epistcmic GSs in first-ycxessays (76 per ccnt and 24 per ccnt respectively), while the proportion of epistcmic GSswas higher in final-year essays (56 per cent phenomenal and 44 per ccnt epistrmic). Firstvearessays showcd a greatcr focus on the real-world phenomena that geography invcstigatrs,as in the following example in which GSs, all phenomenal, are underlined:Bus and rail are the obvious components of a mass transit system. The 1x1s is thccheaper of the two options, but even with designated bus lanes it still adds to thcproblem of congestion antl has a lower capacity. on the other hand is much moreexpensive but has a higher capacity and takes travcl away from the roads. There arc anumber of examples of mass transit systems around the world, ~ ~ are m highly csuccessful others arr not.These clearly foreground the phenomena that are bring studied and indicate a priorityof content over rhctorical motivations. While third-year essays were also concerned withrcal-\vorltf phenomena, thcsc were oftcn displaccd from thc subject position by cpistemicGSs, underlined in the following examplc:ExDerimcnts by Morpan et al. (1982) on 'detachment of soil particlcs from a sandysoil by raindrop impact in storms of 50mm/hr and 61 mm/hr for 5 minutcs durationshowed that the rate of erosion untlcr a cover of hrusscl sprouts dccrrasetl as thecanopy cover increased from 0-1 5 25%, but erosion increased if the canopy coverincreased any more antl at 50% cover the erosion rate equalled that of bare soil'(Morgan 1986). Similar experiments were done on potato crops with similar findings.These results add to thosc ofvis which show under certain circumstances plant coversare associated with high rather than low ratcs of erosion mainly due to their influenceon the kinetic energy of intercepted raindrops. Othrr cxiieriments such as bv DePloev et al. (1 976,) recorded an increase in soil erosion with an incrrasc in grass cover.. . . Morpan (1980) showed in his study on soils in Silsoc that sandy soils inBedfordshire are ten times more erodible in summer than in n-intcr.On a gcneral lc\cl, then, the model is uscful for indicating certain disciplinary trendswithin writing. It provides a way of focusing on a particular linguistic fcature and uses it totease out aspects of the disciplinary culture which would otherwise be obscure. Bv this

80 ANN HEWINGS AND MARTIN HEWINGSmeans, advice on writing can he tlircctctl morc specifically to those features which reflectthe underlying culture of a particular disciplinary community. GSs in successful and lesssuccessful student writing could lie coniliared and stuclcnts encouraged to classify the GSs,initially using their own criteria antl later using the I)roatl phenomenal/cpistcmicclassification given aliovc.3 Implications for syllabus designThc starting point of this essay \vas that the recent trend to\r ards genre-based approachesto the teaching of academic lvriting \\as a positive tlcvelopment when compared Ivith thosewhich rrprcscntctl a homogeneous viclv of academic M riting, undifferentiated across genrcs.Indccd, in current thinking on tertiary academic literacy it is now taken almost as axiomaticthat an understanding of generic comcntions, particularly in terms of moves but also stepsand their lcxicogrammatical rcalisations, is essential in achicving academic succcss. Ihcarchis shobving both how genrcs differ antl how kno\vlctlgc of one gcnre may tic inadequatepreparation for the production of another. For cxamplc, Hylantl’s (1 999a) work on themetadiscourse of textbooks leads him to concludc thatstudents need to tic steered any from using textbooks as models. Too close afamiliarity with the lvays that textbooks address readers, organise material and prcsentfacts may mean that learners arc poorly pi-clm-cd \vhcn assigned research articles bytheir subject lecturers or kSP tcachcrs or when asked to write argumentative prose.(Hylantl, 1999a: 2 1-2)Ho\\-c-ver, in the design of syllabuses for academic writing programmes, it is necessaryto rccognisc the lesson ofthc research rcportcd in this essay: that helping students to developa knowledge of gcnrcs is insufficient in a number of \vays. First, it is important to guardagainst teaching gcnres as a set of templates to lie copictl unsxvervingly. Razerman’s (1 988)investigation of the cxprrimental article in science leads him to offer thc following caution:the largest lesson that this study holds is not that thcrc are simple gcnres that must beslavishly followed, that \vc must give students an appropriate sct of cookie cutters fortheir anticipatctl careers, hut rather that the student must understand and rethink therhetorical choiccs embedded in each generic haliit to master thc genre.(l3azcrman 1988: SO)Second, it is ncccssary to dcvelop students’ sensitit ity to the fact that gcnres vary,particularly across disciplines. This is incrcasingly important given the growing number ofstudcnts in multi- or inter-disciplinary academic programmes whcrc success is dependentboth on being abvarc of disciplinary variation in communication practices and on developingsufficient flexihility to producr writing that reflccts the predilections of a particulardisciplinary community. Such Ilcxibility is unlikely to he achieved simply by learning theprcfcrred conventions of a discipline, but must be untlcrpinnctl hy a deeper understandingof how this reflects such matters as thc dcgrcc of consensus within the disciplinc on thedefinition of prolilcms and appropriate methodologies to address thcse problems, theamount of prc-vious research that it is convrntional to ackno\vlcdge, and the cohesivenessof the hody of agreed kno\vlcdgc within the discipline.Third, we need a reassessment of ‘common-core’ and ‘discipline-specific’ componentsof academic writing programmes (see also Rhatia, 1 999). Common-core teaching, focusing

DISCIPLINARY VARIATION IN ACADEMIC WRITING 81on the ‘language and conventions related to the general requirements of the academiccommunity’ (Ihdlcy-Evans, 1 995), has the advantage of being addressctl to studcnts fromacross tlisciplines and is thercfore efficient both in terms of thc numhcr of students taughtand, often, financially. €Io\z.ever, it has the disadvantage ofhcing relatively insensitive to thcpi-c~t‘crrctl wavs of writing in particular disciplines. Discipline-specific components providcthis sensitivity but usually have the disadvantages of addressing smaller numbers of stutlcnts,needing specially focused research antl being less cost-efficient, and will often haw theatltlitional complexity of requiring the input at some stage of a subjcct specialist. Perhapsour goal is that of common-core tcaching mediated through a disciplinary filtcr. Whilchelping students develop an awareness of the general significance of certain of thccharacteristics of writing in particular genres text organisational patterns, metadiscoursalfeatures, grammatical subjects, for example ~ at the same time we need to provide themwith the strategies for examining how thcse operate and why this should be so ti? reflectingon the subject matter, working practices, valucs and idcologies of the discipline or disciplineswithin which they are working.Note1 Whilc the ESP approach to genre analysis has been particularly influential in pedagogicalapplications, other perspectives exist, antl Hyon (1996) has itlentifed t\vo additionalbroad areas of scholarship rcsearching non-literary genres: North American Rhetoricstudies and work within Australian systcmic functional linguistics (for cxxample Martin2000).2 Samraj had a further category, ‘miscellaneous’, which is not included here. Ilcnce thefigures (lo not add up to 100Yo.ReferencesAnderson, J. and Poolc, M. (1994) T~CSIS and ignment Il’riting. 2nd edn. Brisbane,Queenslantl: Jacaranda Wiley.Bazcrman, C. (1988) Shaping I1~’rirten Knorvletlge: The Genre and ActiviLy $[he Experimental ’4rticlein Science. Madison: University ofWisconsin Press.Bclanger, M. (1982) ‘A preliminary analysis of the structure of thc discussion sections in tcnneuroscicnce journal articles’ (mimeo).Rcrkcnkotter, C. and Huckin,T. ( 1995) Genre Knowletige in Disciplinuy Communication. Hillsdale,NJ: Lawrence Erlbaum.Bhatia,V. K. (1 993) Ana!y.sjng Genre: Longt~agc Use in fr&-sional Settings. London: 1.ongman.Bhatia, V. K. (1999) ‘Disciplinary variation in husiness English’. In M. Hewings antl C.Nickerson (ctls) Business English: Research into Practice. Harlow: Longman, 12943.Kizzell, P. (1992) Academic Discourse antl Critical Consciousness. Pittsburg: University of PittsburgPress.Brett, P. (1994) ‘A genre analysis of the results section of sociology articles’. English,fbr SpecificPurposes 13 : 47 59.Crismore, A., Markkanrn, R. and Steffenscn, M. (1993) ‘Mctadiscoursc in persuasive writing:a study of tcxts written Iy American antl Finnish university students’. IWrittenCommunication 10: 39 71.Crookes, G. (1986) ‘Towards a valiciatcd analysis of scirntific text structure’. Applied Linguistics7: 57-70.Dudley-Evans,T. (1985) Writing Laborator// Reports. Melbourne: Nclson.Dudlcy- Evans, T. (1989) ‘Genre analysis: an invcstigation of the introduction antl discussion

82 ANN HEWINGS AND MARTIN HEWINGSrtations’ , In M. Coulthard (etl.) Talking about Ext. English Languageof Birmingham, Birmingham, UK, 128 45.I)udlcy-Evans, T. ( 199 3) ‘Variation in communication pattcrns bctlveen discoursecommunities: the case of Highway Engincct-ing and Plant Biology’. In G. Blue (ctl.)Language, Learning and Siiccess: Stiit+ing through English. Inndon: Macmillan, 141 -7.Dudley-Evans. T. (1 995) ‘Common-core antl specific approaches to thc teaching of academicwriting’. In D. Bclchcr and G. Rrainc (ctls) ilcatlcmic CVriting in il Sccontl Language: E.s.sayson Research and Pctlagogj. Nor\\-ood, N]: Ahlex.Dudley-EI ans, T. and Henderson, W. ( 1990) ‘The organization of article introductions:evidence of changc in economics lvriting’ . InT Dudley-Evans antl W. L. Henderson (eds)The Lungiia‘qe qf’ Economics: The ,4no!p.s c$’ Lconomics Discourse. London: Modern EnglishPublications/ British Council, 67--78.Dutllcy-Evans, T. and St John, M. (1 998) Dei.elopments in English for Specific Purposes: .4 Multirlisciplinay‘4pprouch. Camhritlgc: Cambritlgc University Press.Halliday, M. A. K. (1 973) Explorations in the Functions of’l.anguuge. London: Etlward Arnold.Herzherg, B. (1 9S6) ‘The politics of discourse communities’ . P ap presented at Conferenceon College Composition and Communication, New Orleans, 1986. Cited in Bizzcll,1992.I Icwings, A. (1 999) ‘Disciplinary cngagemcnt in untlcrgraduatc writing: an investigation ofclause-initial elements in geography cssavs’ . Unpu1)lished Ph.11. thcsis,The University ofBirmingham, UK.Holmcs, R. (1995) ‘Genre analysis antl thc social sciencan investigation of theintroductions, background sections antl discussion sections of research articlcs in history,political science antl sociology’. Unpublishctl MA dissertation, Univcrsity of Surrey, UK.Holmcx, R. (1 997) ‘Genre analysis antl thc social sciences: an investigation of the structure ofresearch article discussion sections in thrcc disciplines’ . Engli.sh,for Specific Prirposes 16:321 37.Hopkins, A. and Dudley-Evans, ’I.. (I 988) ‘A genre-based investigation of thc discussionsections in articles and tlisscrtations’. English-fir Specjfific Purposes 7: I I 3 22.Hylantl, K. (1 999a) ‘Talking to students: metatliscoursc in introductory courscbooks’ . Englishfor Spec!$c Purposes 18: 3-26.Hvland, K. (199911) ‘Disciplinary discourses: writer stancc in rescarch articles’. In C. Candlinand K. Hylantl (eds) Il’riting: Exts, Procc.s.ses and Practices. I Iarlow: Addison- Wesley-Longman, 99-1 2 1.Ifyon, S. (1 996) ‘Genrc in thrcc traditions: implications for ESL.’. TESOI. Quarter(v 30:693 -722.MacDonald, S. P. ( 1992) ‘A method for analyzing sentence-level differences in disciplinaryknowledge making’. IVritten Communication 9: 533 -69.MacDonald, S. P. (1994) Pr~fe.ssiona1 Acarlemic Writing in the Numanitier. Carbondale: SouthcrnIllinois University Press.Martin, J. K. (2001) ‘Tcchnicality and ahtraction: language for the creation of specialisedtexts’. In A. Burns and C. Coffin (eds) Analysing English in a Global Contexr. London:Routlcdge.Mauranen, A. (1993) Cultural Differences in Academic Rhctoric:/l Textlinguistic Stu+. Frankfurt-am-Main: Peter Lang.Peng, J. ( 1987) ‘Organisational featurcs in chcmical cngineering research’. English LanguageResearch Journal 1 : 79 1 16.Prior, P. (1998) M/riting/Disciplinaritl~ /I Sociohistoric hcount of Literate Activity in the Academy.Mahwah, NJ: Lawrence Erlbaum.Quirk, R., Grecnbaum, S., Leech, G. antl Svartvik, J. (1 985) A Comprehensive Grammar of’theEnglish Language. London: Longman.

DISCIPLINARY VARIATION IN ACADEMIC WRITING 83Roberts, I). (1 997) The Student :s Gt~ide to I?.iiting Essuys. London: Kogan Page.Samraj, B. T. R. (1995) ‘The nature of academic Lvriting in an interdisciplinary field’,Unpuldishetl Ph. D. thc , University of Michigan.Sclager-Meyer, F. (1999) ‘Kcfercntial bchaviour in scientific writing: a diachronic study(1810-1995)’. Englishfbr Speczpc Purposes 18: 279-705.S\vales, J. (1981) :Ispects pf;4rticle Introductions. Birmingham:The University ofAston, LanguageStudies Unit.Swales, J. (1984) ‘Research into the structure of introductions to journal articles antl itsapplication to the teaching of academic writing’. In R. Williams, J. Swales antl J. Kirkman(ctls) Coinrnon Ground: Shilretl Interests in ESP and Communicution Sttidies. EIT Documents117: 77-86.Swales, J. (1990) Genre ,Anu!ysis. Cambridge: Cambridge University Press.Tal-one, E., Dwyer, S., Gillettr, S. antl Ickc,V. (1998) ‘On the usc ofthe passhe antl active voicein Astrophvsics journal papers: with cxtcnsions to other languages antl other ficltls’ .Englishfbr Spec,i/;c Purposes 17: 1 13 32.Tavlor, G. antl Chcn, T. (1 991) ‘Linguistic, cultural antl subcultural issues in contrastivediscourse analysis: Anglo-American and Chinese scientific. texts’. Applied I inguistics 12 :319 36.Vande Kopple, W. J. (198 5) ‘Somc exploratory discourse on metadiscourse’ . College Compositionand Communication 36: 82-93.Weissbcrg, R. and Buker, S. (1990) lVriting up Research. Engle\vood Cliffs, NJ: 1’1-cnticc-Hall.

PART TWOPolitical and institutionalconstraints in curriculumdeve I op me nt

Chapter 7Ronald CarterPOLITICS AND I

88 RONALD CARTERto persuade teachers to a return to thc 1940s antl to the kinds of practiccs of languageteaching illustrated by this cxamination paper.What are the practi Tvhich arc illustrated by this examplc? Why do governmentministers wish to see the reinstated? What do tcachcrs think of them? What is the viekvtaken by linguists of such practic AnsLvcrs to such questions may begin to explain whythe materials for teachers protlucctl by the Language in the National Curriculum(henceforth, LINC) project \vert not only rcfuscd puldication by the British government,but also hccamc the centre of contesting iic\vs almut languagc and education.1.2 Views of language and language teachingThe different \iews of language and language tcac hing in rcspcct of this reprcscntati\cexamination paper held ti) go\ crnmcnt, English teachers antl 1): linguist5 mal be lx-oatll!wmmarisctl under three hcatlings (1) go\crnmcnt 1 IC\$ 5, (2) teachers’ liens, and (3)linguists’ \ ICWSGOT ernment i leii 71 . The examination papcr illustrates a manifest conccrn with measurable knoMledge. Abocly of linguistic facts can hc taught, learned by pupils and thcn tested. Answers areeither right or wrong, the liody of knowlctlgc taught is dcfinite and measurable, andteachers can even be assessed lw ho\v \vel1 they teach it.2. The learning \vhich cnsucs is tlisciplincd antl takes places nithin a clear framework.It contrasts vividly with what is felt to he the vaguc and undirected concern withcreativity and personal cxprcssion which charactcrises ivork in many English lcssonsat the prcscnt timc.3. Such practiccs will help to guarantee corrcct grammar antl standard English. Theywill remove sloppiness in expression antl eradicate a climate in which errors areviewed only in relation to a process of language dcvelopment and thus not alwaysimmediately corrected.7iacherc’ rrerisUntil recently, teachers’ views have bccn rcgularlv dominated by what are describedas ‘romantic’ conceptions of hglish as a subjcct (. Christie, 1989; Carter, 1988).Romanticism in English teaching involves a classroom emphasis on languagc use \vhich isperson-centred, lvhich str . the capacity of thc individual for originality and creativity,and a concern that strict rulcs and conventions may be inhibiting to pupils anti, in the, restrict thcir capacitirs for using the language. There is a particular stress on theprimacy of speech, even in writing where indivitluals arc cncouragctl by thc teacher to findtheir olvn personal voice. During the coui-sc of thc LINC project shifts in tcachcrs’perception of’ formal language study were recorded, but strong rc tance remains, on thcabove grounds, to the tlccontcxtualised study of language, to aching practiccs andpedagogics which arc ncc .aril! transmissive and narrolvly kno\vletlgc-bascd, and whichallow little or no scope for an emergence of the pupil’s owm ‘voice’.

POLITICS AND KNOWLEDGE ABOUT LANGUAGE 89Linguists have taken a prominent role in the shaping ofthe National Curriculum for Englishin England and Wales. Most take the follohving main viclvs of grammar-haset1 tcaching andtesting of linguistic knowlcdgc:1 They point out ho\v examination papers from the 1940s and 1950s are prc(~ccupietlwith the written rather than the spoken language.2 They point out that the analysis is invariably decontcxtualisetl since the dcfinitionsrequired of pupils arc formalistic. Examinations such as thosc above arc exercisesin the naming of parts.3 They point out that such examinations arc concerned lvith scntcnccs rather than texts.In fact, the text here is genuinely incidental. The focus is on a bottom-up analysis ofthe smallest units of language with little or no interest in eliciting from pupils howsuch units might combine to form larger functional mcanings and effects.Accordingly, thosc linguists lvho advised the government did not recommend a return tothe 1940s antl to a tcaching of grammatical forms by mans of decontextualisctl drills. Butthey did not reject a formal study of language. Instead, they strongly advocated programmesof study for pupils in knowledge about languagc (KAL), based on a wider range of analysisthan grammar, and [. . .] clearly rooted in theories of language variation, both spoken andwritten. The government \vas quick to rccognise that knowledgc about language, based ona variety of texts, includes discussion of language in context, and that discussion of contextis often necessarily social. Such an orientation served only to rcinforcc for the goxrnmentthe desiraliilitv of dccontcxtualiscd drills and exercises.2 LINC: An in-service teacher education projectLINC is designed to make the theories and descriptions of language in the nc\r NationalCurriculum accessible to teachers, antl to assist them with thc language components ol’theNational Curriculum lor English. [. . .]In basic outline the main projcct team \vas asked to produce study units for tcachcrswhich were to be used in in-service courses, in school-based follow-up and (tiand in sclf-stud! sessions. The resulting training package is therefore activity-lnscd antlopen-entletl. It contains many linguistically hascd tasks with accompanying commentaricsso that teachers can work on the material in a range of‘ contexts.2.1 The LINC banThe LINC project assumed political prominence \\hen the government dccidctl that it didnot wish to publish the materials produced by the project. Neither would it allowcommercial publication in spite of interest on the part of. ral international puhlishcrsin publishing the complete training package. Although the projcct \vas allowd to continueanti although the LINC training package could be made available in photocopied form forpurposes of in-service training courses, such dccisions amounted to an effective ban on\videspread publication and dissemination of LINC materials.[. ’ .Il>cIiatcs surrounding thc LINC ban ccntrc on certain key\vortls. They are tht. satnckeywords which recur repeatedly at times of social and cultural change when questions oflanguage and the nature of English as a subject are always central.

90 RONALD CARTERIt is no semantic accidcnt that words such as standard, correct, and proper areamong the kcy\vortls. Debates ahout the state and status of the English language are rarelydebates about language alone. The tcrms 01’ thc deliatc are also tcrms for defining socialbchaviour.The term English is synonymous with Englishncss, that is, with an understandingof who the proper English are. A vimv of one English with a single set of rules accords witha monolingual, monocultural version of society intent on preserving an existing order inwhich everyone knows their place. A view which rccognises Englishcs as well as English andwhich stresses variable rules accords with a multilingual, culturally divcrse wrsion of society.Both positions include politically extrcmr versions. Thcsc range from a view that standardEnglish is correct English and must tic uniformly enfbrccd in all contexts ofuse (with dialectsextirpatcd) and that children not drilled in thr rules of stantlard grammar are both deviantand discmpo\vercd (strong right-wing position) to a view that standard English is a badgeof upper-class power, antl that to require children to learn it is a form of social enslavement(strong leftning position I) to a view that standard English must be taught to working-classchildren so that they can \vrcst linguistic power from those more privileged than themselves(strong left-wing position 11). It is striking how political positions converge in certainrespects and how the pedagogical positions arc oftcn identical.2.2 LlNC and grammar[. ’ .IIn the LlNC training materials therc is no advocacy ofa return to the dccontextualiseddrills and exerciscs of the 1950s. Instead therc is systematic exploration of grammaticaldifferences lietwecn spccch antl writing, Iietwwn standard and non-standard forms of thelanguage, and between diffcrcnt varieties of English. In spite of tieing described in certainnational newyapcrs as a dialect project, 97 per ccnt of the examples is a LlNC materialsare of pupils speaking, reading and lvriting in stantlard English. They also dcmonstrate thatone of the most effective 1%-ays of learning standard English is for pupils to compare andanalysc diffcrences between their o\vn dialects and thc tlialcct of standard English, discussingexplicitly how and when differcnt forms are appropriate. [. . .]Here is an example of LINC’s approach to grammar taken from some local trainingmaterials. The example is liascd on a text in the form of a postcard delivered through theletterbox of customers of a water company.The following tcxt communicatcs information; in this case the information concernsthe interruption to water supply. Whenever instructions arc given, a ‘modality’enters the relationship between the writcr and reader of a text. ‘Modality’ takes anumber of diffcrcnt forms in English liut the presence of modal verbs is particularlysignificant. Hcrc are some of the main modal vcrlx in English:can; could; will; izould; must; should; shall; mayWhat is the function of modal verbs in thc tcxt that follows?What other verb forms work, in particular, to establish a relationship between theWater Company and the customcrs to whom it has distrilmted this notice?CommentaryThis tcxt is in a curiously mixed mode. The Water Company has to inform itscustomcrs that repairs are unavoidable. It has to give its customers instructions which

POLITICS AND KNOWLEDGE ABOUT LANGUAGE 91Notice of interruption to supplyWe are wrry to inform you that nccc55ar) mains repair5 in the area ma) cause aninterruption to Jour 13 atrr supply hetuccn the hour5 olerleaf.1. Every effort will be matlc to keepinconvenience and thc duration ofthe shut-off to a minimum.2. Do not draw more water than yourminimum requircmcnts.3. If the water docs go off, do notleave taps open or flooding mayresult when the supply is restored.4. You may use water from thc hottzater systrm but it must be boiledbefore drinking.5. Even if the domestic hot \vatersupply runs dry there will bc norisk of damage to the system, liutas a precaution kccp a lo\v firewhere a back hoilrr is installed antlturn or switch off other sources ofheating thc water by gas, oil orelectricity.Central heating systcms cancontinuc to be used at moderatetempcratures.The main \vi11 be flushed before thesupply is restored butdiscolouration and or chlorine maypersist for a short time. Allo\v yourcold tap to run for a frw minutesto clear this watcr from yourscrvicc pipe.Do not use your washing machineor other appliances during thediscolouration.We apologise again for any inconveniencc this may cause you and request yourpaticncc and co-operation. In case of any difficult! please contact theNottingham District Office.Please remember neighbours who may be older or disabled - they mayneed your helpthey nerd to follow both in their own intcrcsts antl in the intcrcsts of other consumers.At the same time the company ncctls to reassure its customers that a more or I txnormal service is still available, that, in spite of the interruption to supply, the companystill providcs a good service and, above all, that there are no safety or hcalth risksinvolved for its customers so long as they comply with the guidelines and instructionsissued with the notice. It is important therefore that the company is clearly seen tolie in control.This ‘mixed mode’ is inscribed in the different modal verbs in the textsalong the following gencral lines:Mode qf rcassurance/possihi/it/c.: may cause an interruption; may persist for a shorttime; they may nccd your help; every effort will be madc; flooding may result;any inconvcnicncc this may cause you.Mode ofcontrol: must be boiled before drinking; the main will be flushed; cancontinuc to be usd.Notice that some modal wrbs can signal possibility ant1 control, depending on theother words which surround thcm as well as on the context in which thcy arc used.For cxample, ‘you may use water’ (primarily control); ‘they may nccd your help’(primarily possibility).

92 RONALD CARTER‘Control’ 15 also established through an cxtensne uw of imperatne forms of theherb \.z hich unambiguouslj inform us u hat to do and M hat not to do. For example.Do not lea\ c taps openAllow your cold tap to runDo not uw )our 1% ashing machinePlease remember neighboursAc tivi tvCollcct examples of further texts in lvhich you would expect modal verbs to be usedquite extensively. For example,horowopcsmcather forccaytsproblem pagesschool noticesrecipeslegal tcxtsWhat other examples can ~ ou find? Wh) arc modal Lcrbs concentratcd in some tcxtsbut not others?It is one key fcature of the IJNC approach to grammar that teachcrs and pupils should,where possible, explore grammar in complete texts, in relation to social and culturalcontexts and \vith reference both to forms and functions. It is primarily concerned withhow grammar works to construct meanings in the kinds of literary texts with which manyEnglish teachers arc familiar antl, as in the ahovr example, in the everyday texts we allencountcr in our daily lives.2.3 KeywordsWhat was effcctivcly a ban on the puldication of I SNC training materials probably shouldhave hcen expcctctl.The emphasis on language variation antl on language in context led toa too frequent rcfercnce to social theory antl an emphasis on sociolinguistic persprctives.For governments of a particular political persuasion the word social is directly equitablewith the word socialrst.’l’hc training packagc itself was tlcsignctl, it was said, in too activitybasedand open a manner.The govcrnmcnt cvcntually made it clear that it had preferred allalong training matcrials \vbich cmphasiscd right and wrong uses of English, reinforcing suchan emphasis with drills antl cxerciscs for teachers and pupils to follow, antl with a printedappcndix containing the correct answers to the exer .The emphasis should be on factualknowledge which is measurable and tlctcrminahlc, and which can be transmittctl from aposition of authority rather than be discoverctl through activity-centred processes. A kcywordhere is thc wort1 drill. Finally, it \vas said that certain keyvords do not appear in asufficiently unambiguous \Yay. In the training package words such as correct, standardand proper arc always rclativiscd to specific contexts and practices of teaching.In respect of such key\vortls, linguists antl teachers tlo, in fact, need to find a w.ay oftalking about language M h Iwttcr controls and engages \vith the cxisting public discourses,especially those of most ions of thc prcss antl mctlia. In this connection, English teachershave to apply thcir knowlctlge about language to a major problem of communication. Thevery \mcabulary currently available to talk about language variation offers only apparentlynegative or oppositional tcrms which play neatly into the hands of those with the mostsimplistic notions of language and education. Thus, to talk about non-standard English canbe seen as a departure from standards; to talk about the dangers of absolute rules ofcorrcctncss is sccn as an endorsement of incorrcct English or as a failure to correct pupils’

POLITICS AND I

94 RONALD CARTERcorresponds to a report lvhich is one of impersonal classification rather than personalohscrvation. Such impcrsonality is rcinforccd by the use of the passivc voice (‘theyare covered in scales’).3.1 Reactions to genre-hased teaching1 . LINC tcams have lieen convinced by the strcngth and depth of arguments for makingthe language structure of texts tnorr visililc on the grounds that genuine interventionby thc teacher and conscycnt tlevclopnient in pupils’ language use arc not possiblcunless the relevant patterns of language arc identified. [. . .]2. LINC teams have acccptcd that a primary concern with personal shaping of expericncehas resulted in classrooms in \vhich thcrc is an ovcr-concentration on narrative to theexclusion of other gcnrcs.3. In a related way LINC has adopted a morc inclusivc 1 ic\v of authorship, especially inthe writing classroom. It a pts the view of Pam Gilbert (1990: 70) that: ‘Authorshipis but one of the ncncst of a long line of discursive devices which serve to entrenchpersonalist, individualist, SIJ 11-oriented theories of writing in schools.’ Althoughsuch a position obscures important developmental connections between spcech andM-riting, it establishes a basis lbr inore impersonal Lvriting motlcs, and thus a \viderrangc of generic types of ivriting on lzhicli LINC has built.LINC’s introduction of a morc gcnrc-liasctl approach to lvriting has provoked somehostility on the Inrt of British tcachcrs. A major concern is that such writing practicesarc inherently conservative antl are designed to produce unreflective operatives \vhowill bc able to do no morc than sustain a market economy for a conservative society.The concern of gcnrc thcorists for a \vitlcr rangc of Mriting typcs which arc in turncloser- to the rcquircnicnts 01’ thc \vorltl of \vork is intcrprctcd as a narrowvocationalism. What has hclpcd to change this pcrccption is the notion of criticalliteracy, \vhich augments functional litci-acy to cnablc learners not only to comprehendantl produce society’s discourses, but also to criticisc antl rcdircct them, if necessary.As Michael Hallidav (I 996: 357) has put it:To be literate is not only to participate in the discourse of an information socicty;it is also to resist it . . . it is rathrr pci-vcrsc to think you can engage in discursivecontest without engaging in the languagc- of the discourse.Such mark underlines that gcnrc-Imcd teaching is both revolutionary antl reactionary.British tcachers havc bccoinc increasingly impressed by thc precise analytical work\vhich has cnablcd central, prototypical featurcs of particular genres to be identilied.It is the samc explicitncss of anal . which has helped both pupils and tcachcrs todevelop a critical linguistic literacy.LING tcams have valuctl the overt, cxplicit and rctricval)le arguments advancedin particular I)? Martin (1989) antl KI-css (1989) but also 1w others. Taking suchstrong, clear argumentative lines cnablcs others to ai-guc with or argue against in asystematic \lay.

POLITICS AND I

1 of96 RONALD CARTER6.the workplace. For example a rcport gcnrc in a junior school is markedly differentfrom a rcport genre in industrial or business work settings. Text-intrinsic accounts ofgenre need to take fullcr cognisance of the audiencc, purpose and context in whichparticular genres operate.Encouragement to pupils to rcllcct on language has tendcd to lie restrictcd to thepatterns of IanguagcT in the gcnrc in focus. Instead, a gcncral classroom climatr needsto he established in which talking and writing almut language leads to [. . .] languageairureness ~ that is, gcncral scnsiti\ it\. to different styles and purposes of language use.These include tliffcrcnces lxtmwn spoken antl written language, cxplorations of thelanguagc of literature, thc language of jokes, advertising, pop fiction, and politicalrhetorics, and investigations of’ the continua I>ct\vccn different accents and dialects,including standard English. Such cxplorations arc a nc ary habit-forming preludeto looking more closely antl analytically at the linguistic patterns \vhich make updifferent genres. Analysis is not al\vays best fostcrcd by practising analysis of andreflection on language solely within the context of individual genres.Several of these observations arc hardly ncw, and many of’thcrn have been advanced by genretheorists themselves. Teachers in Britain intcrestctl in lzriting development arc bcginningpositively to emlracc xzork on gcnrc-theory and on gcnrc bvithin a functionalist pcrspcctivcin particular. Thcsc observations should I>c vicwwl in a correspondingly positive light.I. . .I4 Conclusions: the lessons of LINCA project of the scale antl complexity of LlNC cannot escape criticism. It is important thatthe lessons of both succcss and failure arc ix~ot-dcd. 1:or cxamplc, for all their SUCC~SSCS\vith tcachers, LINC materials ncctl to he further adapted in three main ways. First,materials on reading should Iic tl lopctl to cxcmplify in grcatcr detail what a mixedmethods approach to rcding entails. Morc examples antl case studies ~-ould illustratc howreaders use a rangc- of different cucs and cluc-s, syntactic antl semantic, phonic and visual,learning to read. Morc action rescarch \voultl illustrate when to mixvhcn to concentrate on a xinglc teaching ptwccdurc. Future LINC materials) must also cnablc teachers Iictter to analysc the linguistic tliffcrcnccsbetween real books and hooks from gr-adctl reatling schcmcs. Sccontl, supplements toexisting units arc ncccletl on diffcrenccs bct\vccn spoken and Lvrittcn English, particularlyin relation to the teaching of punctuation, which depends crucially on the relationshipIiet\vccn grammatical structure and the rhythms and contours of spccch. More examplesarc also nccdcd of how standard English varies across spoken and written modcs hvhile stillremaining standard English. Third, inorc cxxaniplcs arc nccdctl to show how literary textscan stimulate enhanced kno\vlcdgc about language, especially the history of the language,antl how greater linguistic kno\vlcdgc underpins literary appreciation.4.1 Negative conclusionsEven if the gcncral tlcvclopmcnts outlined ahovc take placc, they \vi11 take placc against acultural background in lvhich both positive antl negative factors arc at work. The mainnegative factors arc, first, that some teachers will continue to pcmist with the worst cxccsscsof romanticism in their view of language learning and tcaching.l’hcy will continue to makelinguistic processes invisildc antl regard language only in so far as it provides a \vindo\v on

POLITICS AND KNOWLEDGE ABOUT LANGUAGE 97to content, the expression of the individual self, the kvorltl of ideas. They will continuc torefuse to see forms of language as a powerful resource for creating significant domains ofmeaning. Second, governments may want to intervene more directly in the shaping of theEnglish curriculum. If so, and Lyhatevcr their political persuasion, governments may notwant to endorse classroom language study which explores relationships between languageand society, and which subjccts those relationships to interrogation. They are likely tocontinue to be especially disturbed by classroom KAL work which encourages childrcn toinvestigate such rclationships indcpcndently. Thcy may exert their powers to impose alanguage study which is ‘neutraliscd’ by being more decontextualisctl, formalist rather thanfunctionalist in orientation, and \vhich, above all, can be easily assessed and measured. Thecurrently very overt demands by thc British government for grcatcr attention to phonicsin the teaching of reading is but a signal of an increasing emphasis on the basics in so far as1% hat is ‘basic’ often involvcs a dccontextualiscd language focus.4.2 Positive conclusionsIt is a positive factor that governments are drawing attention to language, recognising it asboth medium and mcssagc, mounting arguments in relation to the ‘proper’ study of English,attacking the positions adopted by those with a profwsional interest in language.Although the battles will continue to be bet\\ n those who have thc power but notthe knowlcdgc, and those who have the knowledge but not the power, the \-cry fact thatgovernments arc forced to mount explicit arguments about language is healthy both lorproccsscs of public debate and for the cause which espouses the ccntrality of language tothe school curriculum. Increasing attention to language on the part of teachers, coupledwith high degrees of enthusiasm and conviction, will lead to pupils being progrcssi\-clyinterested in language. Increasing knowledge about language among pupils will producewithin a generation a society which is likely to be less prejudiced and ignorant and moreinformed and articulate about matters to do with language.Finally, a morc positi\e view of applicd linguistics emerges from projects such as thcLINC project. It is a view in which teacher and linguist work more collaborati\ely ton.ardscommon agendas. As a result, tcachcrs become more a\varc of the problems of linguisticdescription and, in turn, linguists begin to address problems identified by teachers, rathcrthan only those problems itlentifietl by linguists themselves. [. . .] Incrcasingly, all concernedwith language have come to appreciate how notoriously fascinating, complex and ultimatelydangerous language and language study arc. In a project inspired by thc work of MichaelHalliday, the final word must be left to Halliday (Halliday, 1982):. . . there is a real sense in kvhich linguistics is threatening; it’s uncomfortable, and it’ssubversive. It’s uncomfortable because it strips us of thc fortifications that protect andsurround some of our dccpcst prcju . As long as xve keep linguists at bay ~vcango on believing \\.hat ~ve \\,ant to hcli >out language, both our own andClSC~’S . . .More than any other human phenomenon, language reflects and reveals theinequalities that arc cnshrinctl in the social process. When we study languagesjrstcmatically . . . we see into the power structure that lies behind our ever);day socialrelationships, the hierarchical statuses that are accorded to different groups \vithinsociety . . .

98 RONALD CARTERBibliographyCarter, R. (1988) ‘Some Imvns for Kingman: language education and English teaching’, inGrun\vell, P. (etl.) .,Ipplied Linguistics in Socic:~ 3, British Studies in Applied Linpi.stics (CIIT,London) pp. 5 1-66.Christie, I;. (1 989) I unguugc Edricirtion (Oxford University Press, Oxford).Gilbert, P. ( 1990) ‘Authori7ing disadvantage: authorship and crcativity in the languageclassroom’, in Christic, F. (cd.) Literuy fi)r a Chunging Cihrld (Australian Council forEducational Research, Halvthorn, Victoria), pp. 54-78.Halliday, M. A. K. (1982) ‘Linguistics in teachcr cducation’, in Cartcr, R. (etl.) Linguistics andthe Zucher (Routledge, London), pp. 10 16.Halliday, M. A. K. (1 996) ‘Literacy and linguistics: a func.tional pcrspectivc’, in Hasan, R. andWilliams, G. (eds) Literacj, in Sociecl, (Longman, Lnndon), pp. 339 75.Krcss, G. (1 989) Liniqtiistic Procews in Sociocultural fructicc (Oxford University Prcss, Oxford).Martin, J. (1 989) Fuctuul IVriting (Oxford Univcrsity Press, Oxford).

Chapter 8Gary M. JonesBILINGUAL EDUCATION AND SYLLABUSDESIGN: TOWARDS A WORI

100 GARY M. JONESTcihle 8. ICompulsory and cxaminablc suhjccts in Rrunri priniarv antl sccontlarv schoolsEnglish LanguageEnglith LanguageMathematic5Mictor)ScienceGcograph:English I.anguagcMathematicsScienceGeographyI oii cr PrimoLy (age 5-8)Mala! I anguagcMathematicsGeneral StudiesIslamic Rcligious Kno\z.le.tlgcPhysical EducationArts antl HandicraftCivicsllppcr Primor) (ugc 9 I I )Malay LanguagcIslamic Religious Kno\z.lctlgePhysical EducationArts and HandicraftCivics1.o~ cr Seconduy (uge 12 I J)Malay LanguagrIslamic Kcligious KnowlcclgcHistorvEnglish I anguagcMathematicsScicncc/Art/T'echnical Subjccts(clcpcntling on stream)Although some Malay continues to be usctl in the RI-uncian system, at Primary 4 thcrcis an abrupt change, antl this occurs at a time when the pupils have only a limited proficiencyin English. Not only is the number of English-medium hours greatly increased, but theincrcasc is in some of' the most cognitivcly demanding subjects, subjccts which Cummins(1 984) would characterise as precisely those that require a well-tlcvclopcd L2 proficiency.BICS and CALPCummins (1984) has in fact distinguished bctween two sets of language skills: basicinterpersonal communication skills (BICS) and cognitive acadcmic language proficiency(CALP). He argucs that children will be unable to cope with the school curriculum unlcsstheir cognitive academic language proficicncy (CAW) is sufficiently dcvclopcd. A child'slanguage-cognitive abilitics nccd to be sufficiently ~.cll tlevclopcd to cope with the curricularprocesses of the classroom. This proficicncy could be tlevcloped in either of the tiilingualchild's languages or in both simultaneously. In Cummins's (1984: 143) opinion, CALPinvolvcs some universal undcrlying proficicncy which is shared across languages. Onceacquired in one language it can tic transferred to any other language. Thus, proticicncy ofthis sort acquired in Malay could hc transferred to English-medium classes antl vice versa.Cummins develops thc concept of RICS antl CALI' in the four quadrant model whichis reproduccd as Figure 8.1 .

BILINGUAL EDUCATION AND SYLLABUS DESIGN 101Cogillti\ cl) undcmantling(RICS)HBConCcxt-cmbctltlctl Contcxt rctluc c dc I)(CAI P)C‘ogniti\ cl\ tlcinantlingFlgtrre 8. I RICS ant1 CALI’Sorrrcc: Curnmins. 198 1In the Brunei context, subjects such as Mathematics and Science, which arc cognitivelydemanding and often context-retlucrd (based on abstract rather than concrete examples),would be placed in the fourth quadrant (D), whilc those such as Art and Physical Education,which arc cognitively undemanding and generally context-embeddetl, xvould be placed inthe first quadrant (A). What should be of some conccrn to curriculum designers in Bruneiis that Cummins (1 98 1 ) hclicvcs that it often takcs one or t\vo years for a child to acquirecontext-eml~edde~l second language fluency (the typ language that might IK din Art or Physical Education classes), but from fivr to years to accluirr context-reducedfluency (working with more abstract subjects). If thi case, then after only three yearsof English 1,anguagr as a subject at lower primary school, Bruncian children arc unlikely tohaw thc rcquired English proticicncy to study the type of cognitivel! demanding, contcxtreducedsubjects that the); arc currcntl); introduced to in Primary 4. [. . .]Cummins’s concept of BlCS and CALP has been criticiscd for being too simplistic.Romaine (1 989) argues that language skills cannot be compartmentalisetl as neatly asCummins suggests and that Cummins is guilty of equating semantic development withcognitive development. It is ccrtainly the case that not all subjects can be simply and easilyplaced in their rrspcctivc quadrants. Science will allvays be cognitively demanding, but itcould be taught in a context-cmbcdded as \vel1 as a contcxt-rcduccd style.’l’he same is trueof most subjects: much dcixnds upon the style and skill of the teacher. Nevertheless, whileit might he difficult to neatly place all school subjects into one of‘ the four quadrants,Cummins’s motlcl does provide insight into Lvhy pupils \vorking in a sccond language maystruggle in some subjects hut do wrll in othcrs. Most important is that Cummins hclicontext-reduccd, cogniti\rl!.-demantling communication capaliility develops intlependcntlyand can be promoted by either or both languages. [. . .] If kno\\lctlgc is transferable acrosslanguages, then thew is no nerd to begin the study of thcsc academically tlrmanding subjectsat an early age through the medium of‘ English to prcparc for an English-mediumexamination that will he takcn eight years later.1. ’ .IThreshold levelsThe present assuniption in Brunei is that subjects which will be examined in English at ‘0’level at age 16 should bc taught through the medium of English from as early an agc aspossihle. Subjects that arc eventually examined in Malay are therefore taught through themedium of Malay throughout. [. . .]This present division of Malay-medium/English-medium subjects, cspccially at thcprimary lc\-cl, is putting an unncccssary strain on pupils and the education system. Many

102 GARY M. JONESpupils have an insufficient command of English to prolxrly follo\v their new subjects,cspecially Mathematics antl Science. The result is that many pupils arc failing to acquireeither sufficient subject knowledge or to imlx-ove thcir language skills. 1. . .]The shift to cognitivcly and linguistically dcmantling tasks at Primary 4 is at presentmade on the assumption that pupils ha\ c sufficient mastcry of English to actually studythrough the medium of English (as \\‘as originally cnvisagetl in 1951). This assumes thatpupils have attained some miniinurn languagc aliility or threshold. [. . .]Although language thresholds \vert not discussed as such hack in the 1950s, rcccntclarification of what they involve helps to licttci- clarify the situation in Brunei and relatelanguage levels there to those attained elsc\vhci-e.Van Ek andTrim (I 991) dc rilie the threshold level in terms of the type of functionsthat a young European learner should be alilc to pcrfoi-m in the target language. Thesefunctions are incvitalily Euroccntric, rcflccting the nccds of Europcan stutlents.They includebeing alile to understand and use the target languagc as a medium of instruction antl as alanguage of social interaction in English classes antl among learners during breaks and atmealtimes; lwing able to report and discuss prolilcms relating to teaching, social conditionsand accommodation and also how to follow admission procedures to cntcr teachinginstitutions.Van Ek and Trim (1991) specify thc numlicr of tcaching hours that should be neededto attain the threshold level of proficiency: t\vo to thrcc hours per Lvcck, 35 to 40 \vccks awar over two to three years. A minimum of 140 hours of teaching and a maximum of 360hours. In Bruneian lower primary schools, prior to transfer to Primary 4 and the teachingof acadcmically demanding subjects through the medium of English, Hruneian pupils receiveapproximately 262 hours teaching in and on the English language (two and a half hours perweek, 35 weeks a year over three years). It should also Iic rcmcmbcrcd that these Bruneianchildren arc very young (oltlcr Icarncrs may not ncccssarily lie Iwttcr learners but they dounderstand the educational Iiroccss and arc thcrcforc faster Icarncrs, as Singlcton, 1 989,has oliservetl), that they may not ha1.c a vcrv supportive learning cn\ ironmcnt outsideschool, especially for the tlcvclopnient of Lnglish, antl that English is unrelated to any oftheir other languages. These arc condition5 \vhich must surely crcatc greater languagestudying difficulties than for their European peers.I. . .] In 1951 , with selected pupils following intcnsivc personal tuition, minimumproficiency levels \z-crc presumahlv consitlercd attainable. Ho\\ er, the Same is not truetoday of large mixed ahilitj classcs ofchiltlrcn. Many pupils arc failing to attain a minimumproficiency in English liefore the introduction of cognitivcly and thcrcby linguisticallydcmanding English-medium sulijccts. It is probably the case that they are only reaching sucha level whm they cntcr Lower Secondary school, at \vhich time thcv should really befunctioning at an intermediary level hcyontl the sccond threshold. This level, in turn, mayonly be reached at the point these pupils rcach upper secondary, at which timc the pupilshave to take their ‘0’ levels, which require an cvcn higher language Ic\cl.As a result of problems ticginning at primary school, pupils continue to lag behind theirrequired level or threshold of language proficiency antl the majority never really rcach thelanguage standard which their age might assume. Sornc c.vidcncc for this has been gathcrcdbj Lewis Larking, who tested the reading comprehension ability of Bruncian pupils inPrimary 5 and 6. He found that at Primary 6, 700/0 of pupils were helow their nativespeakerequivalent grade level in English reading comprehension (28% were one year belowgrade Icvcl; 38% two or more years Iiclow antl 4% three or more years hclow gradc level).Interestingly, only 7% of thcsc same pupils \\we hclow grade level in Malay readingcomprchcnsion (Larking, 1994: 58). 1. . .]

BILINGUAL EDUCATION AND SYLLABUS DESIGN 103Subject order: some considerationsGiven the mix of primar hool subjects and the various degrees of cognitive demands thcymake antl opportunities for language interaction they offer, perhaps the easiest and mosturgent changes would be to the subjects that are taught through the mcdium of English atupper primary and to the nature of thc transition from predominantly Malay-medium toEnglish-medium subjects.Instead of teaching the cognitivcly demanding, ‘contrxt-rcduccd’ subjects in Englishand the cognitively undemanding, ‘contcxt-cmbeclded’ subjects in Malay, the easiest proposalwould be to switch subjects. Thus Mathematics and Science would rcmain Malay- mediumsubjects (at lcast, until Lower Secondary) whilc Art and Physical Training would be taughtas English-mcdium subjects at the primary Icvcl. Support for retaining thc tcaching ofMathcmatics and Science in the mother tongue (at lcast until secondary school) can also beseen in Ramirez et al. (1991) where it is noted that students Ivho were abruptly moved intoalmost exclusive instruction in English ‘. . . experienced a marked dccrcasc in growth inmathematics skills over timc’ (Ramircz et a]., 1991 : 33).To avoid the sudden incrcasc in English-medium subjects at Primary 4, morc Englishmediumsubjects could be introduced at lower primary level. [. . .]Art and Physical Education arc not literacy but oracy and participatory activitics thatshould not endanger the acquisition of first language literacy. Thus these two subjects couldbe introduccd in English-medium at lower primary, together with English Language, lvithoutharming thtx pupils’ ability to first become literate in Malay.Age and language acquisition: some considerationsHarlev (1 986) antl Singleton (1 989) have shown that the question of age antl languageacquisition is complcx and does not lend itself to an easy and universal anslvcr. Most of theresearch supporting the ‘younger is better’ position does so with reference to phonologicaladvantages, while that supporting the ‘older is better’ stand is on the basis of syntax antlmorphological measures of ability. Hamcrs antl Blanc (1 989) and others question theevidence for there being a sensitive period and a biologically tlctcrmincd optimal age forL2 acquisition. [. . .]The conclusion that younger learners arc at an advantage because they have more timeto learn and are less likely to suffcr interfcrencc from their first language matches quiteclosely, though for different stated reasons, the conclusions that Genesee (1 987) has drawnconcerning the various immersion programmes in Canatla:Second language proficiency tends to increase the earlier immersion begins and themore second language exposure the learner has. Thus, early total immersion generallyyiclds higher lcvcls of second language pruficicncy than carlp partial immcrsion,delayed immersion, or late immersion.(Genesee, 1987: 191)Singlcton (1 989) argues that many factors arc involved in language acquisition and thatcxamples of age-related research have to be analpsed individually, noting the peculiariticsof each study. Singleton concludes that:there is a fair amount of cvidcncc suggcstivr of a long-term adbantagc for learnerswhoa(, c,xpcrirnce of the target language begins in their childhood yrarr. . . . with

104 GARY M. JONESregard to short-term attainment, the picture is more confused. Hobvcvcr, the Iialanceof evidence docs sccm to indicate an initial advantage for oldcr learners at least as faras grammatical dcvclopmcnt is conccrncd.(Singleton, 1989: 122)Dcspite his extensive research on the sulijcct, Singleton remains undecided about thebenefits of one approach over thr other:(1) The a\ailablc e\ idcncc doc5 not consistcntl: \upport the hJpothcsis that youngcrsecond language learner? arc gloliall~ more cfhcicnt and ruccc5sful than oldcrlearners.(2) Nor 15 it possililc to conclude from the (x\itlcnce that oldcr second languagelearners arc glol)all! more cfhc lent and 5ucccssfuI than Joungcr learners(Singleton, 1989: 138)Although Singleton docs not favour cithcr an carl! or late start, he does suggest thatCuinmins’s RICS/CALI’ distinction rcconcilcs contradictions in the cviclcncc aliout agerelateddifferences among second language learners, citing Cummins’s o\vn proposals:the older learners, \vhosc‘ CALP is Iwttcr tlcvelopcd, \voultl acquire cognitive/academic I .2 skills more rapidly than youngcr learners; however, this \vould notnecessarily he the case for those aspccts of L2 proficicncp unrclatccl to CAI,P(i.c. L2 BICS).(Cummins, 1979. In Singleton, 1989: 1 1 3)This lcntls support to the type of sul3jcct division rccommcndctl earlier for Brunei,with the study of cognitivcly untlcmantling, cont~~xt-crnl~c~l~lc‘tl suhjccts precedingcognitively demanding, context-rcduccd subjects.The majority of Bruncians u.ho voluntccrcd a reply to the question’ Are there anj.comments thatyou udd like to make ahout Dri,ihuhusci!’ in a national attitude questionnaire that\\-as concluded rccrntly suggested that more English should h c introduced at an earlier age.It \voultl lie \cry cas! to dismiss such suggestions as Iicing uninformed and subjective, butI think that this \vould bc \\rang and a misjudgcnicnt of‘ the rcspontlcnts.Rruncians havc hcen exposed to a varicty 01’ school tcms.’l‘hcrr can bc little doubt,ho\vevcr, that the Bruneians who arc most at case \vith the English language arc those whoattcndcd English-medium mission schools at an earl! age. Some of thcsc pcoplc continuedtheir education in the mission schools ivhilc others \vent on to government schools. In eithercase, an advantage seems to have Iiccn tlcrivctl from early cxposurc- to English. Of course,an ability with English \voultl also he tlcpcndent upon factors outsidr school, especiallylanguages iisetl in the home, and it can I>c assumed that many parents \vho sent their childrento English-medium mission schools \z oultl themselves very prohalily usc English at home.But this would not havc been the case for all familics. The English language abilitj ofgraduates from such schools is takcn as cvitlcncc Iiy Bt-uncians that early exposure to Englishresults in ticttcr acquisition of the 1anguagc.This conclusion ma! bc subjective, hut in Bruneiit is accepted as self-evident and is the most commonly cited reason given 1iy Rruncianparents \vho can afford it for sending their chiltlren to English-mctlium kindergartens.Although research \voultl obviously haw to lie undertaken to provc the point, my ownimprrssion fi-om oliscrvations in Brunei is that as \vcll as phonological atlvantagcs, earlyexposure to English also appears to result in a greater cor$dencc among learners in actually

BILINGUAL EDUCATION AND SYLLABUS DESIGN 105using thc language. Such confidrncc results in fewer inhihitions antl a willingness toexperiment with thc language. At its simplest, the Bruneians who haw acquired Englishearly at mission schools are gencrally more confident, and therefore more at case with thelanguage antl thus likcly to use antl experiment with the language, than their peers whowent to government schools.The relevance of The German ModelA wcll established motlcl of bilingual education that has given consideration to the timingand sequencing of the school subjects in its system is that no\v referred to as ‘The GermanModel’ (Masch, 1993). Bilingual Gymnasium (‘grammar’) schools in Germany make agradual transition from German to a second language, introducing the new languagethrough a sequcncc of subjects that are chosen to complement each other as \vel1 as to aidboth language and subject acquisition. [. . .] In sequence, the subjects chosen as 1-ehicles forbilingualism in The German Model arc: (1) Art; (2) Geography; (3) Politics (Civics); (4)History.Art is considered to provide a concrete situational base from which to develop languageskills (as well as subject ability). Geography performs a refcrcntial, information givingfunction and provides a relatively simplc start to description. Geography is also recognisedas pcrforming a second hut crucially important role in bilingual education in Germany.While some scicncc subjects are considered to bc language poor because of the spccialisetlcontent ofthcir subject-specific languagc (Masch cites Biology as an example), ‘Geographyalone sufficiently covers virtually all the necessary elements from thc natural sciencesthrough physical geography and geophysical phenomena, and from the application ofmcthodological skills through its work with figures, statistics, graphs and sketches’ (Masch,1993: 163).Politics (Civics) antl History arc included in the lilingual education system in Germanyas much for integration and a better understanding of the country’s European neighboursas for the language benefits of the subjects. Nevertheless, the study of politics does includea recognisablc languagc function:[. . .]The aims ofthe course in politics suit thc bilingual section: an ability to rccognisctliffcrcnt types of action antl a capacity to form an opinion.(Masch, 1993: 1634)Physical EducationPhysical Education is not mentioned in Masch’s dcscription ofthe German Model. However,this subject docs lend itself particularly to communicatiw activities because it emphasisesn language and physical movement. A physical education lecturer at theUnivcrsitv of Brunei Darussalam has noted the link lxtwern his suhjcct and languageacquisition in Brunei. He maintains that physical education creates a language richcnvironmcnt :[. . .] The most noticeable change in students’ attitude or lxhaviour \vas a readilydiscernible increase in confidence . . . Emanating from this incrcase in contidrncr anoticeablc improvement in fluency together with greater self-assurance whilst makingstatements was evident in the studcnts’ performance.(Austin, 1992: 25 -6)

106 GARY M. JONESRevised syllabusAs an alternative to the present distribution of suhjects and language media, and with dueconsideration to language acquisition antl age as ~.cll as RICS/CALP and the threshold levels,I would suggest that compulsory and examinable sulijects in Bruneian primary and secondaryschools might be more appropriately distributed as shocvn inTalile 8.2.Table 8.2 An altcrnatiLc distribution of subjects in the Kruncian education 9ystcmEnglish medium ,WO/UJ mediumEnglish LanguageArts (antl I Iantlicraft)Physical EducationEnglish LanguagcArts (and Handicraft)Physical EducationGrographyEnglish LanguageMathematicsGeographyScienceEnglish LanguageMathematicsScience/Art/TcchnicaI Sulijccts(depending on stream)I ower PrirnqMalay LanguagcMathematicsGeneral StudiesCivicsIslamic Religious Kno\\ lctlgcllpper Priniar)Malay LanguagcMathematicsCivicsScicncc1 IistoryIslamic Rcligious Kno\vlctlgeLuiwr SeconclucvMalay Languagef IistoryIslamic Rcligious Kno\vlctlgc[Jppcr Scconday.Malay LanguagcPhysical Training and Art Mould join English Language as English-medium subjccts inthe lower primary curriculum. Geography, givcn the expcricncc of the German Model,would also seem an appropriate subject to introduce at thc primary level. Mowcvcr, ratherthan tax the pupils with too many English-medium subjects at once, the introduction ofGeography might best lie postponed until upper primary. In this revisctl system, Historywould remain a Malay- medium subject throughout (there arc not the same integrativepolitical considerations operating in Brunei that encourage the bilingual teaching of thissubject in Europe) and the introduction of Scicncc and Mathematics \vould be dclaycd untilsecondary school. I. . .]The system described above is open to a number of permutations. For instance, shouldthree subjects bc introduced simultaneously from Primary 1 or would it be bcttcr tointroduce them consccutivcly, one year at a time? Is there any advantage to lie gained inswitching History from Malay to English-medium? Ideally, a number of permutations mightbe triallcd until the most appropriate model for Brunei is arrived at. Of course, this wouldtakr time and may not tic feasible because of the common national examinations that have

~ (1BILINGUAL EDUCATION AND SYLLABUS DESIGN 107to be taken at the end of primary school. However, givcn the research that has beenconducted into threshold levels and the timing of their acquisition, as well as considerationsof contcxt-embedded /context-reduced, cognitivcly demanding/cognitively undcrnandingsubjects, then this revised model for the introduction of subjects in the Brunei educationsyytem, in one form or another, would seem more appropriate than that currently employed.While I bcliew that the above \vould he a better system than the present distributionof subjects, there arc a number of practical considerations which would impede theimplementation of this proposal. As well as the problem of new examinations and syllabuscs,there would also be the huge task of supplying or retraining teachers for new media ofinstruction. Givcn that providing a sufficient number of properly qualificd teachers hasalways been a problem in Brunei, changing the language-medium of some subjects would,in the short term at least, further compound this problem. Malay-medium Art, Geographyand Physical Education teachers may not be willing or able to teach in English; EnglishmediumMathematics, History and Science teachers may not be able to teach in Malay.Ncvcrthclcss, the introduction and redistribution of school subjects at thc primary level isan issue that should eventually be addressed.ReferencesAustin, S. (1992) ‘Languagc development through education’. Paper presented at a seminar inthc University of Brunei Darussalam, December 1992.Cummins, J. (1979) ‘Cognitix e/academic language proficiency, linguistic inter-dependence,the optimum age question and some other matters’. Ilbrking Papers on BilingLiulism 19,198 203.___~(198 1) ‘Thc role of primary language developmcnt in promoting educational succcss forlanguage minority students’. In The California State Department of Education (ed.)Compendium on Bilingual-Bicultural Education. Los Angeles: California Statc Departmentof Education.984) Bilingualism and Special Education: lssues in Assessment and Pedagogy. Ckvedon:Multilingual Matters.Genescc, F. (1 987) Learning Through Eo Languages. Cambritlgc, MA: Ncwbury I Iouse.Hamcrs, J. F. and Blanc, H. A. (1989) Bilinguals and Bilingualism. Cambridge: CambridgeUniversity Press.Harley, B. (1986) Age in Second Language Acquisition. Clevedon: Multilingual Matters.Larking, L. (1994) ‘Reading comprehension ability of Primary 5 & 6 children in Malay andEnglish in Brunci Darussalam’. In M. L. Tickoo (ed.) Reatling and Research in Writing.Singaporc: Rcgional Language Centre.Lenneberg, E. H. (1967) Biological Foundations $Language. New York: Wiley.Masch, N. (1 993) ‘The German Model of bilingual education: An administrator’s pcrspcctivc’.In H. Ractcns Rcardsmorc (cd.) European Model.$ ?f Bilingual Education. Clcvcdon:Multilingual Matters.Ramirez, J. D., Yuen and Ramey (1991) Longitudinal Study of Immersion Earlj-exit and Late-exitTransitional Bilingual Education Programs .for Language Minoriy Children. Mountain View,CA: SRA Technologies.Romaine, S. (1 989) Bilingualism. Oxford: Basil Blackwcll.Singleton, D. (1989) Language Acquisition: The Age Factor. Clevedon: Multilingual Matters.Van Ek, J. A. and Trim, J. L. M. (1991) Threshold 1-eve/ 1990. Strasbourg: Council of EuropePress.

Chapter 9Kimberley BrownWORLD ENGLISHES IN TESOLPROGRAMS: AN INFUSION MODEL OFCURRICULAR INNOVATIONIntroductionT HAS BEEN SUGGESTED (c.g. Kachru, 1988)thatitistimeforaparadigmshit‘tI which takes into account the changing roles antl functions of English around thc worldin linguistic research antl in languagc pcdagogj. In spite of clearly articulated arguments andnd-dcfinrd calls for applied and theoretical research in the languagc education community,Kachru’s perspective on the role and functions of English as an international languageremains a minority perspective. Even though t\vo gcncrations of scholars have been rcfiningthe elements of what has comc to Iic termed thc World Englishes paradigm (Kachru, 1992a),there is little evidence of its infusion intoTcaching English to Speakers of Other Languages(TESOI.) preparatory programs (Vavrus, 1991 a).This paper examines possiblr explanations for the lack of incorporation of this paradigmin TESOL preparatory programs. In an attempt to find a \vaj to incorporate thc WorldEnglishes paradigm intoTESO1, programs, I kvill draw upon Hamnctt et al.’s (1984) threeprongedapproach to what they term ‘thc intligenization of social science research’ (78).This paper will also explore impediments to curriculum design and teaching practice thatmay hamper the inclusion of the World Englishes pcrspcctive into TESOL preparatoryprograms.A paradigm refcrs to a particular thcorctic framcmvork or perspcctivc. The WorldEnglishes paradigm (hereaftcr referred to as the WE paradigm) may lie charactcrizcd bythree elements (Kachru, 1988: 1):a ticlicf that there is a ‘repertoire of modcls for English’a belief that ‘the localized innovations [in English] have pragmatic Iiases’a belief that ‘thc English languagc now bclongs to all those who use it’.Paradigm shifting and diffusion of innovationWithin meta-theory research, i.e. theories about theories, I’atton (1 975) explores therelationship betwecn the context in which information is lcarnctl and the dcgrcc to whichpeople remain attached to that information. I IC statcs: ‘. . . paradigms are deeply embedded

WORLD ENGLISHES I N TESOL PROGRAMS 109in the socialization of adherents and practitioners telling them \\hat is important, what islegitimate, what is reasonable’ (1 975: 9). By implication then, any process of shiftingparadigms cannot be a simple proccss.Tollcfson (1991 ) suggests that the intellcctual beliefsystem an individual may adhere to is often not seen as a particular set of lenses; i.c.individuals may hold to particular ideologies without evcn recognizing that there issomething subjective about these idcologies. Thus before any shift in ideologics can comcabout, the first step must be to recognize and name the paradigm to which one has bccnintellcctuallp socialized. [. . .]It is possible to see the introduction of a new paradigm into the intellectual arena assimilar to thc diffusion of an innovation into a particular setting. In thc social scienccliterature that explores attributes of innovations, Rogers (1983) examines variahles affectingthe rate at kvhich innovations are adopted. One is of particular relevance to this discussionif the WE paradigm can be considercd an innovation. Rogers terms this variable ‘perceivedattributes of [an] innovation’ (1 983: 233).The five facets of this variablc are compatibility, relative advantage, complcxity,trialability, antl obscrvahility. Rogers suggests that compatibility ofthc new idea with currentidcas (or paradigms) antl with client needs (in this case teacher and learner needs) affectswhether thc new idea is adopted. An exploration of currcnt idcas in TESOL reveals twoframeworks incompatible with an underlying framework in thc WE paradigm IvhichVavrus(1 991 a) terms the ‘Dynamic Paradigm’ .To avoid confusion, ho cr, the term ‘perspective’will be used instead of paradigm.Much of the underlying theory in the WE paradigm belongs to what Vavrus (1 991 a,199111) calls the ‘Dynamic’ perspective, a framemnrk in which non-native varieties of Englisharc sccn neither as fossil-ridden examplcs of interlanguagcs, nor as inferior cxamples ofincorrect speech. She suggests that most language teaching fraine\vorks may lie characteriredas belonging to one of t\vo perspectives. The Deviational perspective supports the notionsthat all errors arc due to fossilization or to learners being at various stages of interlanguagetransfer.Thc Deficit perspective supports the notion that errors occur because learners arcdeficient in their command of English. Neither the Deficit nor Ileviational perspectives arccompatible with the Dynamic pcrspcctive. The lack of compatibility tictwccn these thi-ecpcrspecti\es thus affccts the rate at which thcWE paradigm may tic adopted. Until cducatorshccomc more aware of the reasons for adopting a WE paradigm or of thc conse‘quenccs ofnot adopting it, they may resist this innovation.The remaining elements in Rogers’ pcrccivetl attributes of innovations arc rclatiwadvantage, complexity, trialaliility, and ohscrvability. Relative advantage refers to a pcrceptionthat the new idea is better than prcvious ones. Complexity is sclf-explanatory. Rogerssuggests that if an idea is n as being too complex, it \vi11 not lie atlopted.Trialaliility rcfcrsto ‘the tlegrec to which a nnovation may be experimented with on a limited liasis’ (1983:231). Research suggests that if individuals can work with a new idea on a trial basis, theymay lie more likcly to adopt it. Finally, ohscruability refers to how visible an innovation is.If individuals arc quite familiar with an idca, they are more likclyAt the present time, the WE paradigm docs not clearly pocompatibility, and ob \-ability. I do not bclicve the WE pcrspcctivthan other perspccti . But much of the early work in World Englishcs is not in a userfriendlyformat for c room teacher educators. This does affect the trialability factor.(. . .] Hamnctt et a/. (1984) discuss thrcc elements that have a direct connection tonce of the WE perspective in TESOL prcparatory programs. Thc firstis thcoretic indigenization ‘in Lvhich the social scientists of a nation are involved in constructingdistinctive conceptual framc\vorks and mctatheories that reflect their om-n world vie\\ s,

theoretic indigenization would involve the crcation and refinement of thcory. At thc presenttime, these parameters arc well developed (scc Kachru, 1992a).Therc does appear, however,to be a problem with access to antl availability ofinformation.This problem will be discussedmore thoroughly in a later section of the paper. Another aspect of the theoretical dimensionof the WE perspective is also presented by Lincoln, who looks at the politicization of theresearch process and suggests that at the present time ‘some groups and research models[are] favored over others, with some definitions of problems morc acceptable than otherswith avenues to funding and support clcarly discriminatory’ (1 990: 70).The second is structural indigenization, defined ‘with rcfcrence to national institutionaland organizational capabilities for social science knowledge [including capabilities for]educational and research institutions, a community of indigenous scholars, and locallyproduced social science literature’ (Hamnrtt et al., 1984: 78). Within the WE perspective,in Outer Circle countries, i.e., those countrirs once colonized by England or the USA andwho use or have used English for intra-country purposes (Kachru, 1988), structuralindigenization would involve thc development of institutions which sponsor a particulartype of research, the development of scholars committed to working within the WE paradigm,and the development of locally produced WE literature and empirical or qualitativestudies. In Inner Circle countries, i.e. the USA, UK, Australia, New Zcaland, or Canada(Kachru, 1988), the structural challenge is to support the development of young scholarsfrom Outer Circle and also Expanding Circle countries (Kachru, 1988), i.c. those countriesin which English has been taught as a foreign language, who will return home to promotestructural changes in how English is taught, antl to support Inner Circle scholars developingcollaborative frameworks with Outcr and Expanding Circle colleagues for their teaching.For cxample, the cooperative small group antl pair work in communicative language settingsis a standard concept promoted in most TESOL Methods courses. In many Outer andExpanding Circle countries, large classes and a standardized Ministry of Educationcurriculum which prepares students for college-level entrance exams are common. Smallgroup or pair work may be proscribed. Having successful teachers from these largeclassroom settings prepare lcssons on how to teach large classes, which could then he infusedinto current Methods cow. , would hclp Inner Circle teachers learn from Outer andExpanding Circle colleagues.The third element in Hamnett et al.’s text involvx substantive indigenization, which is‘concerned with the content focus of the social sciences [such that] the main thrust ofresearch and teaching in a country be toward its own society and people and their economicand political institutions’ (1984: 78). Within the WE perspective, substantive indigenizationwould call for the development in Outer and Expanding Circle countries of’ their ownresearch and teaching focus. The challenge in Inner Circle teacher preparatory programswould be to encourage Outer and Expanding Circle students to return home to conductresearch on topics and with agendas that may not have been those suggested in basic researchdesign courses. A further challenge would be to prcpare Inner Circle students planning onteaching overseas to understand and appreciate the integrity of the possible alternativeplanning frameworks they would be working under.Shifting paradigms in TESOL programs is a difficult task. Individuals who havecompleted their own profcssional preparation under one paradigm may not see a reason toshift. It is necessary for those scholars who have called for the paradigm shift to see themselvesas change agents and to actively engage in effective promotion efforts so that teachereducators and practitioners in the held can understand the perceived attributes of the WEperspective. This promotion process may involve contributing to the development of WE

WORLD ENGLISHES IN TESOL PROGRAMS 111theory and collaborating with colleagues in Outer and Expanding Circle countries tosupport the tlevelopmcnt of indigenous institutions, scholars, literature, and empirical orqualitatkc studies. Further, it may involve bringing to the attention of scholars in InncrCircle countries details of the current research and teaching focus of English languagecducation programs in Outer Circle and Expanding Circle countries.Many efforts on US campuses to ‘internationalize’ thc curriculum have drawn upon aninjusion model in which supplementary units on particular topics are worked into cxistingcurricula. [. . .]In addition to the overall difficulties with respect to any paradigm shift or curricularinnovation, there arc at least five other possible impediments: amount of time necessary fortruc diffusion of scholarship; student and instructor background schemata; text availabilityand level of difficulty; supporting material availability; and workshop and short-termcourscwork availability for Methods instructors.Time for diffusion of scholarshipThere are currently two major centers in the USA \\here scholars are conducting extensiveresearch in World Englishcs, at the University of Illinois at Urbana-Champaign (UIUC) andat the East-West Center (connected to the University of Hawaii at Manoa) in Honolulu. Byextension, Methodology courses taught in these centers are likely to incorporate the WEperspective. A second generation of scholars who have graduated from these institutionsare teaching a third generation of teachers and scholars to incorporate such a perspectivein institutions from upstate New York to Indiana to California. Other scholars who havcbegun to publish extensively in this ficld havc come to places like UIUC antl the East-WestCcntcr for short periods of time antl have returned to teaching positions throughout theworld. Even in Outer Circle countries, though, it has only been recently that master’sstudents in language education have been encouraged to conduct their research from hzithinthe WE perspective. For example, Robert Raumgardner, an cxpcrt in thc ficld of PakistaniEnglish, writes that in Pakistan, the first master’s degree paper to look at a WE issue waspresented in 1985 (personal communication).Research being conducted around the world may not make its way into mainstreamjournals for three to five years after its coniplction. Anthologics of matcrial publishcd byplaces like the Regional English Language Centre (supported through SEAMEO) arc notpublicized in materials catalogues that most teachers in US TESOL programs have readyaccess to. When multiple copies of texts arc requested from halfway around the world, manyInner Circle bookstores balk at filling orders because of currcncy exchange difficulties andpolicies which prohibit them from ordering materials from other countries.Student and instructor background schemataMany stutlcnts cntcring TESOL programs havc not had any prior cxpcricnccs identifyingparadigms and ideologies. They may not have been trained in identifying their underlyingassumptions ~ cultural and paradigmatic, pedagogical, or personal. They arc not criticalreaders in thc scnsc that they haw not bccn asked to read material in linguistics and languagecducation in order to categorize the ideology of thc authors. One explanation for this isthat much of the writing students cncountcr is heavily centered in what R. Paulston (1976)calls the ‘equilibrium paradigm’, e.g., structural-functionalist and systems theories ascontrasted with theories in the ‘conflict paradigm’, e.g. Freirian, Marxist, or nco-Marxisttheories. The rhetoric in structural-functionalist and systems theories is less readily

112 KIMBERLEY BROWNidentifiable. To ask students to consider making a paradigm shift when they cannot yetidentify paradigms is problematic.International students in TESOL programs may sometimes possess low self-esteemregarding thcir o\vn language proficiency and, as Ihxtcr (1 980) points out, may never havehad the chance to reflect upon antl respond affirmatively to the question ‘Arc you a speakerof English?’ To identify thcir oxvn idcologics and move to a point of grcatcr acceptanceof whatever their variety of English may bc comes at the expense of the amount of timenecessary to absorb such a perspective.Just as it may lie difficult for students to name the ideologies and planning frameworksthey work under, it may also lie difficult for their teachers. Paradigm shifts cannotbe made when people do not overtly idcntify paradigms which currently dominate thefield. Both Tollefson (1 991) and f’hillipson (1 992) deal extensively with this issue in theirtexts.Text availability and level of difficultyThree of the most popular Methods texts, Cclcc-Murcia (1991), Brown (1987), antlLong and Richards (1 987), devote one chaptvr or less to the concept of World Englishes.In Celce-Murcia, there is no mcntion at all ofanything related toworld Englishes. In Brown(1987), while there is rcfermce made to Kachru, and the institutionalized vcrsusperformance varieties of English, the total number of words is less than 200. In Long andRichards (1987), one chapter by Judd defines the term ‘English as an International andIntranational’ language. There is no chapter dealing specifically with the concept ofinstitutionalized versus performance varieties of English, the role of native and non-nativespeakers of English, or any of the points highlightctl by Kachru (1 992a) as bcing central toWE theory.Five other concepts Ivhich need to 1)c rc-examined within traditional Methods coursesif a WE perspectives is infused into the curriculum arc the presentation of instrumental antlintegrative functions of language without introducing the expressive (Pride, 1979) function oflanguage; the presentation of interlanguage continuum inlormation and its relationship tosociolinguistic continuum information; terminology used to refer to the teaching we do;thc role ofnative spcakcrs in instruction in the Outer Circle (i.c. EX countries like India,Nigeria, Malaysia, and Ghana) antl the Expanding Circle (i.c. EFL countries like Japan,Germany, antl Brazil); antl approaches antl techniques for helping students from OuterCircle countries versus students from Expanding Circle countries improve the internationalintelligibility of their varieties of English.Most introductory Mthocls texthooks cover instrumental and integrative reasons forlearning a language (Gardner and Lambert, 1972). Choosing to learn a language for thepurpose of expressing one’s identity or choosing to \\.rite in Filipino English or MalaysianEnglish to express certain \vays of being and knolving arc not cxamined in traditionalMethods texts. [. . .]In units on second languagc acquisition, the interlanguage continuum concept which,as Vavrus ( I 991 a) noted, is part of the Deviational perspective is usually not presented inconjunction with information \vhich is part of the Dynamic perspective.Terminology referring to the teaching of English still falls most consistently into thepolar terms English as a Second Language (ESL) and English as a Foreign Language (EFL).Until the time when more inclusive alternative terms are used, lie they ‘English as an InternationalLanguage’ or, as Canadian colleagues have long practiced, ‘English as an AdditionalLanguage’ , attitudes toward Inner Circle, Outer Circle, and Expanding Circle varieties of

WORLD ENGLISHES IN TESOL PROGRAMS 113English are not likely to change. Speakers of Outer Circle and Expanding Circle varietiesof English in the program where I teach continuously remark on the lack of relevance ofsome material in stantlard Mcthods courses to their ncctls in their countries.Kegarding materials xvhich focus exclusively on World Englishes, Giirlach states ‘thebooks published in 1982-84 make up a particularly impressive list: it is no cxaggcration tosay that the following ten books more or less sufticc to teach a full academic course on thctopic [ofWorld Englishes]’ (1 991 : 1 l).Yct one ofthc Kachru texts, The Indianiyation ofEnglish(1 983), would not be ordered as a class text by some US collcgc bookstores because thetext is published outside the USA; and Kachru (ed.), The Other Tongue: English Across Cultureswas out of print until just recently, and a revised cdition with substantial changes has justbeen published (see Kachru, ed., 1992b). Another Kachru text, The Alchemy $ English(1 986), is too difficult for most undergraduate Methods students, although with stud:questions and referral to Richards et d .’s 1985 Dictionaiy $Applied Linguistics, currently alsoout of print, students can manage. Platt et a/.’s The New Englishes (1 984) as m.cll as Pride’s1982 New Englishes arc out of print. Gorlach’s 1991 volume Englishes: Studies in Varieties pfEnglish, 1984-1 988 is published by John Benjamin; one of its representativcs at the 1992American Association of Applied Linguistics meeting in Seattle commented that thispublishing tirm docs not exhibit at international TESOI, meetings as it is not Lvorth theirfinancial while.Teacher educators attending a key meeting in their lield would thus not haveaccess to this text to even peruse for potential course adoption. Finally, even chapters dealingwith World Englishes in fine texts such as Rerns (1 990) have been termed ‘too difficult’ bysome teacher educators for use with undergraduate students.Supporting material availabilityOften, instructors mill assign research papers or annotated bibliographies on WorldEnglishes topic. Over the past three years, when such assignments have been made in ourMcthods classes, at least one or two students per quarter have decided to research topicssuch as the ‘Non-native English speaking teacher’. Resources such as ERIC list few referenceson WE. Difficulty in obtaining articles reflecting a WE perspective deters studentsfrom pursuing these topics. A cursory comparison of US library holdings via an on-linesearch through the Online Computer Library Ccnter (OCLC) and Research LibraryInformation Network (RUN) of eight journals, six of which routinely publish articlcsreflecting the WE perspective antl t\vo of which sometimes publish articles in this arca(TESOI. Quarterk and journal $Applied Linguistics), r als significant diffcrcnccs in thenumber of journals availahlc throughout the USA. In addition to the TESOI. QuarterbJ antlthe journal ofApplied Linguistics, the other journals include English Toduy, English I~/orldivicic,World Englishes, the RELC journal, the Indian Journal of Lingtiisrics, and the Indian Journal ofApplied Linguistics (sccTablc 9.1).As Table 9.1 illustrates, there is a very evident accessibility prohlcm. Journals whichconsistently carry WE perspective articlcs are much lcss availahlc to teachers and studentsinTESOL programs than arc other journals.This portion of the paper has examined structural impediments to infusing a WEpcrspcctivc in current TESOL programs. The next portion of thc paper documents \vhy itis critically important to \vork to infuse a WE perspective in the Methods sequence.

114 KIMBERLEY BROWNTable 9. IJournals publishing articlcs on World Englishcc topics]ournu/ A'wnhcr ofstates .Ytirnhcr ofinytitutionsTESOL QiarterlyJournal of'.4pplicd Linguistic5English TodayIlbrld EnglishrsEnglish IlbrldwidcREI C Journallndian Jotirnal of LinpisricsIndian Journal ?f,lppliecl I inguistict464244322626119384I23646857551314Student responses in courseworkThe following excerpt is from aTESL certificate student paper; the student was exposed toless than four hours of information rclatctl to World Englishrs in ninety hours of Methodscourses.[Sample A]The lack of TEFL-specific materials may lie a result ofa hclicf that most EFI. teachersshould themselves be non-nativcs [i.c. non-native speakers of English] with the resultbeing that thc majority of tcaching matcrials are in the teachers' native languagcs, notEnglish. But ifa person who has learned English from a non-native then teaches EFLto yet someonc else, the English of the last learner may lie far removed in quality fromthat of a native. . . .cr, more EFI. teachers kvcre native speakers, then more TEFLspccificmaterials in English \voultl be tlemandcd and produced. Also, the matcrialsmight bc of better quality, giving thc quality of EFL tcaching more of a chance toimprove . . .TEFL materials oricntcd toward native English speakers will help thcm to give clearexplanations of what they already know. A native speaker is better able to combinehis or her inherent knoivlcdgc with the information provided in an EFL textbook togive a more comprehensive grammatical explanation. However, this is hest achievedif the textbook is written \vith the cxpcctation that the user will be a native.Another lienefit from having a nativc speaker teach English is that students get moreexposure to English, especially when interacting with the teacher outside theclassroom. A teacher whose native languagc is also the students' native language ismore likely to converse hvith them in that language when class is not in session,whereas practice with a native English-spcaking teacher outside of a stressfulclassroom, where performance will not be graded, can be most helpful.The second sample was written by an individual who had complctcd ninety hours in aMethods course and thirty hours in a World Englishes class.[Sample B]As a result of the reading and thought involved in doing this paper, I now believe thatas part ofthc shift being made in ESLIEFL teaching to accommodate needs for Englishas an international language rather than a language used only to communicate with

WORLD ENGLISHES IN TESOL PROGRAMS 115native or very proficient speakers from countries whcrc English is learncd as a firstlanguagc, \ve need to shift or perhaps, more properly, expand our views of reading.We nccd to extend learncrs’ knowledge of literacy antl reading rather than changc it,lvhich I believe lve often implicitly if not explicitly attempt to do through ourmethodology antl ethnocentric view as middle-class, gcncrally white educators for apost industrial countrj. . , .As a result of this paper, I have also come to realize that in my own teaching ofreading, I have too heavily emphasized U.S. mainstream reading styles antl strategieslvhich may lie of littlc usc to students learning English as a world or internationallanguage, and who arc most likely to read knglish written by writcrs not usingdiscourse or newspaper styles Ivhich are predominant in the U.S.The comments made in the first papcr are geocentric, i.e., focused on Inner Circlecountries’ English, antl rcflcct an attitude of linguicism. Skutnabb-Kangas and Phillipson inI’hillipson (1 922: 47) dcfinc linguicism as:idcologier, structures, and practices n hich are used to Icgitimatc, effectuate, andreproduce an unequal cli~imn of pomer and resources (both material and immaterial)liet\\ ccn groups \\ hich are tlcfincd on the basis of language.The comments in thc second papci- are much less ethnocentric and reflect an attitudc oftolerance antl respect for multiple varieties of English. Both individuals are leaving the sameTESL certificate program. In the first case, I held an cxtensivc conference with the studentto indicate \\.hat I found prohlrmatic in many ofthc statements and lines of argument of thcpaper. 1 lielicw the confcrcnce was useful, but as the culture contact litcrature revcals,siinply having individuals with two different pcrspectives mcct to talk about ideas may notresult in long-term attitude changc. [. . .][. . .] It is possible for a paradigm shift to occur. However, necessary resources tofacilitate such shifts ncctl to find their way into Methods textbooks and Methods courses.Much work remains to be (lone at thc structural and substantive levels. An infusion modelof curricular revision is the most practical means to diffusc the innovation inherent in theWorld Englishes paradigm. The folloxving recommendations arc also key elements inpromoting the diffusion of this perspective within Methods courses and ivithout.RecommendationsLanguagc education preparatory programs must name the paradigmatic frameworkslve work under.We must activcly comllat linguicisrn and gcocentrism, particularly institutionallinguicism, charactcrixed by structures which promote inequality.Wc must help promotc a diversity of perspcctives, not only a perspective whichsuggests that the native speaker of an Inncr Circle variety of English is the mostappropriate professional language educator.Professional language education organizations and their respective journals mustcontinue to provide a forum for oral and \vrittcn dialoguc, e.g. within TESOL: theTESOL Quarterh:, the TESOL journal, the TESOL international conference, and theTESOL summer institute. In like manner forums should be provided with IATEFL:International Association ofTcachrrs of English as a Foreign Languagc, and NAFSA:Association of Intcrnational Educators.

116 KIMBERLEY BROWN5678Professional language education organizations should work to crcatc a resource bankof World Englishcs scholars antl materials.Professional language education organizations should promote mid-carccr professionalcxchange to bring World English scholars as tcachers to Inner Circle teacherpreparatory programs tor one or tlvo tcrms.I’rofcssional language education organizations should support activities which hclplireak down structural Imrricrs to promoting an infusion model of curricular reform.Puldishing houses and authors of kc? texts in English language education programsshould liroadcn thcir rcpcrtoirc of citations.ReferencesRaxtcr; Jarncs (1980) ‘Ho\v shoul(l I s p k English! American-I!, Japanese-ly, orinternationally!’ ].ILTjournul, 2, 3 1-61 .Rcrns, Margic (1990) Conlexts of Competence: Sociill anti Culrurul Consitlerirtion.~ in CommunicativeL ang tiuge Euc h ing . N civ Yor k : Plcnum .HroTvn, H. Douglas (1987) Principlec clf’Lungticigc I.ccirning and Teciching. 2nd edition. EnglewoodCliffs, NJ: Prenticc-Hall Kcgcnts.Cclcc-Murcia, Marianne, cti. (1991) Teuching English us u Second or Foreign lunguuge. NenYork:Ncdury House.Gai-dncr, Kichartl and Lambert, Wallacc (1972) .il/itridcs antl .Ilotirotion in Second Lungtiagelearning. Rmvley, MA: Ncdiury HOLISC.Gijrlach, Manfrctl (1991 ) Encq/i.shes: StriJicr in I;lrJe~ie.s o/’ Englirh, 1 Y81 1988. Amsterdam/Philadelphia: John Benjamin.Hamnctt, Michael, Porter, Douglas, Singh, Amai-jit and Kurnar, Krishna (1984) Ethics, Politics,ant/ lnternotionul Socrcil Science Rcceurch: From Critiyue to Pruxis. Honolulu, Hawaii:University of Hawaii Press.Kachi-u, Braj R. (198 3) The Inditini/ution o/’En!ylish: Thc English I.ung~icyc in India. NC\V Dclhi:Oxtord Univcrsity I’rcss.Kachru, Rraj B. (1986) The ,4/chcn;i. ?f’ English: The Spread, Function.r, ant1 Models ?f Non-l\ativeEnglirher. Oxford: I’ergamon Press. (I

WORLD ENGLISHES IN TESOL PROGRAMS 117Pride, John (1979) ‘Communicatirc needs in the usc and learning of English’. In .YewVarieriec ?fEnglish: I.s.sucs and i1pproachc.s. Etlitcd by Jack Richards. Singapore: SkAMEO RcgionalI.anguagc Centre, pp. 33 72.Pride, John, ed. (1982) New .knglishcs. l

Chapter 10Numa MarkeeTHE DIFFUSION OF INNOVATION INLANGUAGE TEACHINGIntroductionHE LAST TWO DECADES IN APPLIED LINGUISTICS -whichroughlvT coincide with thc evolution of the communicative approach in language teaching -have seen the tlevelopmcnt of a numhcr of languagc tcaching innovations, including thenotional/ functional syllabus, the process syllabus, the Natural Approach, the proceduralsyllabus, and task-based languagc teaching. All of these proposals have contributed inimportant ways to an understanding of theorctical issues related to designing innovativelanguage syllabuses. Rut it is onl? rather rcccntly that applied linguists have hrgun toinvestigate the problcms associated nith implcmcnting thcsc innovations.Indeed, with the exception of such pionecrs as White (1988), Henrichsen (1 989), anda number of other hvritcrs, not many language tcaching spccialists haw developed anyfamiliarity with the voluminous literaturc that already exists in a number of disciplines onhow and why innovations tliffuse.This is unfortunate Iwcausc, as Kenncdy (1988) and Beretta(1990) demonstrate, a ‘tliffusion~of~innovations’ perspective on syllabus design providescurriculum spccialists, materials dewlopcrs, and tcachcrs mith a coherent set of guidingprinciples for the development and implemcntation of language teaching innovations.Furthermore, it supplies evaluators with critcria for retrospcctive evaluations of the extentto which thesc innovations have actually been implemcnted. In other words, this perspectiveprovides a unified framework for conceptualizing both the development and evaluation ofinnovations in language teaching. Thus, although the terminology used may at first soundcxotic and unfamiliar, a diffusion-of-innovations pcrspectivc on syllabus design, for example,addresses concerns that arc central to all language tcaching specialists.In contrast to applied linguistics, education alrcady po cs a well-cstablished traditionof innovation rcsearch and practice (Fullan 1982, Miles 1964, Nicholls 1983, Kudduck1991), as do such disciplines as sociology (Rogers 1983), urban planning (Lambright andFlynn 1980), and language planning (Cooper 1989).7’hus, a rcviekv ofthc issues that defineinnovation in thc specific contcxt of language teaching nil1 draw on these academicspecializations to dcvelop a multi-disciplinary framework, inspired particularly by Cooper’swork on innovation in language planning. The framework for this discussion consists of thefollowing composite qucstion: “Who adopts what, where, when, why and how?” (Cooper 1989),with responses to each individual componcnt of the question. In this wy, the basic issues

THE DIFFUSION OF INNOVATION 119that are of intcrcst to practitioners may be defined. This framework should be sufficientlyencompassing to account for practitioners who wish to engage in any innovation related tolanguage education.On defining innovation: “who”Teachers arc kcy playcrs in any attempt to promote innovations in syllabus design. At thcsame time, other individuals will also be involved in the innovation process (Fullan 1982).The actual participants who become involved in deciding whcthcr an innovation will beadopted vary from context to context. Whatever the specific context of implcmcntation,however, participants tend to assumc certain social roles which define their rclationshipswith othcr participants. The urban planners, Lamhright and Flynn (1 980), have suggestedthat individuals relate to each othcr as adopters, implementers, clients, supplicrs, orentrepreneurs (also known as change agents).Kcnncdy (1 988) suggests that, in thc context ofa materials project inTunisia, ministryof education officials, deans, heads of dcpartmcnt, and others play the role of adoptcrs;teachers are implementcrs; students are clients; curriculum and materials designers aresuppliers; and the expatriate curriculum cxpcrt acts as the change agent. As Kennedy pointsout, in practice these roles arc not mutually exclusive. Indeed, it is quite likely that the samcperson will play different roles, somctimes simultaneously, somctimes at varying timesduring the course of a project. Thus, teachers may at times also be regarded as adopters;furthermore, in some cultures, they may take on the roles of change agents and suppliers.With the exception of thc change agent, any of these individuals may also adopt thc role ofresisters who oppose an innovation. Thus, a broad range of people playing out differentsocial roles is always involved in the design and implementation of any innovation.“Adopts”Adoption has been conceptualized in terms of individuals or institutions engaging in adecision-making process which may be divided into a number of different phases. Rogers(1 983), a rural sociologist who is onc of the leading authorities on the diffusion ofinnovations, suggests that there are five steps in this decision-making process.Thesc involvcpotential adopters 1) gaining knowledge about an innovation, 2) bcing persuaded of itsdue, 3) making a preliminary decision to adopt the innovation, 4) implementing theirdecision to adopt, and 5) confirming their decision to continue using thc innovation. Theeducator Fullan (1 982) proposes a slightly different sequence of four steps which he callsinitiation, implementation, continuation, and outcome.From an evaluator’s perspective, adoption may also be conceptualized in terms of“leve1sof implementation,” a measurc which specifics thc depth to which any changes haveoccurred. Berctta (1 990), for examplc, in his evaluation of the procedural syllabus (Prahhu1987), uses this notion to evaluate thc extent to which tcachers actually implemented thrtask-based methodology associated with that syllabus. Lack of space precludes detaileddiscussion of Keretta’s rcsults, hut it is noteworthy that only 47 percent of the teachersinvolved in implementing the procedural syllabus reached what Beretta catcgorizied asan “adcquate” level of implementation; only 1 3 percent reachrtl what ma! I)r considered an“expert” levcl of implementation.This result shows how difficult it is to promotc innovationat a fundamental level. It is salutary to remember that all innovation is a risky business andthat close to three quarters of educational innovations arc likely to fail over time (Adamsand Chcn 198 I), either because they arc never fully adopted or else do not survive theconfirmation stage posited by Rogers (1 98 3).

120 NUMA MARKEE“What”Innovation itself, as a concept, is central to the implementation and/or evaluation ofncw ideas and ne\v procedures. Synthesizing what she claims are basic characteristicsof innovations, Nicholls states:An innovation is an idea, object or practice perceived as ne\v by an individual orindividuals, which is intended to bring about improvement in relation to desiredobjectives, which is fundamental in naturc and which is planned and deliberate(1983: 4).However, in language teaching contexts, her definition is somewhat problematic. For thepurposes of this paper, innovation \vi11 be defined as proposals for qualitative change inpedagogical materials, approaches, and values that arc perceived as new by individuals whocomprise a formal (language) education system.Nicholls’ idea of“ncwncss”bcing a subjcxctivc matter of users’ perceptions is importantin languagc teaching contexts.This perspective correctly permits the inclusion of the NaturalApproach as an innovation despite the fact that Krashen antl Terrell (1983) view thisapproach simply as a rcdi )very of the underlying principles of traditional “natural” ordirect methods popular earlier in this century suitably reformulated and updatcd in lightof current second language acquisition rcscarch findings. While Krashen and Terrell’sasscssmcnt of the absolute innovativeness of their proposals is accurate, there is no doubtthat from a user’s perspective, the Natural Approach was regarded as an innovation byteachers in the early 1980s. It continues to be viewed in this light by new teachers who areintroduced to it for the first time today.r, in terms of the ti\ c examples of language teaching innovations \vc have alrcadyidcntiticd in the introduction (the notional /functional syllabus, the proccss syllabus, theNatural Approach, the proceclural syllabus, and task-based language teaching), the remainingcomponents of Nicholls’ definition arc either too restrictive or clsc omit defining criteriathat are important for languagc teaching situations. Therefore, the alternative definitiongiven above is more appropriate to language teaching contexts.The need for this alternativedefinition is suggested by a critique of Nicholls with respect to the follokving four issues: 1)the systemic context of innovations; 2) the fundamental naturc of innovations; 3) thc cxtcntto which innovations actually improve on the status quo; and 4) the cxxtcnt to whichinno\-ations are necessarily dc~libc-rate antl planned for.First, the systemic contcxt in hvhich an innolation is implemented scc‘ms to be animportant dctermincr of whether or not the innovation will IIC adopted. As Prabhu (1987)points out, the fact that a procedural syllabus uas implcmcntcd in primary antl secondaryschools in India placed some major constraints on the project since it was decided that thcprocedural syllabus should not be used ivith students who hvcrc due to take various statematriculation exams. By omitting any specific mention of the systemic contcxt ofinnovations, Nicholls lays herself open to bcing interpreted as saying that individuals arefrcc to innovate as they wish. Clearly, individuals do not enjoy such a degree of freedom.This observation suggests that the relationship lietween individuals antl systems must beconsidered in a dctinition of innovation.Second, it is only through a modification of pedagogical values that innovation can besaid to involve “funtlamental” change. At the less complex levels of using new materials andapproaches, teachers can adopt ncw practices with littlc or no undcrstanding of why theyarc using these new materials antl approaches ~ which hardly counts as a fundamental

THE DIFFUSION OF INNOVATION 121alteration in behavior. This limitation docs not mean that such surface changes arc not inany sense innovative nor that they cannot lead to deeper change later.Furthermore, the “inno\.ativencss” of an innovation decreases o\ er time as it lieconicsinstitutionalixtl antl more familiar to usc~s. Iior example, notional/functional syllabuseswcrc initially claimcd to be 1Lintlamrntally diff‘crcnt from structural syllabuses bccauselanguagc contcnt was organized in semantic rather than syntactic tcrms. Ho\\.c\ cr, it \vassoon recognized that, although notional /functional syllabuses arc indccd qualitatii elydifferent from structural syllabuses, all the criteria for organizing the contcnt of instructionin both types of syllabus (i.e., notions, functions, grammatical structures, wcabular!; rtc.)are still linguistic (Long and Crookes 1992). Consequently, it is difficult to argue that there-orientation implied by notional/functional syllahuscs is fundamental in any meaningfulsense of this w-ord.‘l‘hus, it is liettcr to dctinc innovations in tcrms of qualitative change, aterm ivhich conveniently covers all three lei-cls of innovative Iiehavior (materials,approaches, and values).‘I‘hird, although innovations arc certainly intentled to improve on the status quo, it isnot the casc that thev al\vavs do constitutc an improvement on previous practicc (Idlan1982). For example, Urumfit (198 1 ; 1984a) has sharply criticized notional/functiorialsyllaliuses as bcing in some instances an untlcsirahlc innovation. He argues that ivhcn thcsyllabuscs arc uncritically implemented, they dcprivc learners of the generative potentialof grammar (i.c., the ability to use sjntactic rules to crcatc nciv sentences), \vhicli Bruintitindispensable resource for learning. This criticism suggests that improvcmcntrily a defining characteristic of innovations in actual practice. Indccd, in somecases, innovations should be rcsistctl rathcr than promoted because their adoption may liemore harmful than bcncticial.Fourth, and finally, the notion of “delihcratc planning” is problematic for languagcteaching in t\vo \Yay. I;irst, although the notional/functional syllabus is indeed a productofcxtensive planning, the only aspect ol‘a project that can hc planned is \vhat is to be taughtor tcstcd, not lvhat is to he learned (Urumfit 198411). Second, it is doubtful that thearticulation of the principles of thc procrdural syllabus, ivhich \vas achieved through aprocess of trial antl error (Prabhu 1987), can rcally count as an rxamplc of tlclibcr-ateplanning. [. . .]“Where”The question of where an innovation is implemented is conceived in sociocultural tcrms(Cooper 1989). That is, the concrrn is lvith specifying thc sociocultural contcxt of aninnovation rather than its geographical location. Practitioncrs who \z ish to introduce innovativesyllabuses into an educational em must rccognizc the potential impact (lvhcthcrpositivc or negative) of nrious soci tural constraints on their acti\-itics. For example,Markee (1 986a; 1986b) identifies cultural, ideological, historical, political, economic,administrative, institutional, and sociolinguistic factors that affected the implementation ofan aitl-f‘undctl project in the Sudan. [. . .]In addition, some attempts have becm made to address thc issuc of when thoscsociocultural constraints should lie considered in thcCooke 1982, Munbv 1978; 1984). The relati\e imI anct’ of thesc constraints \vi11 \arvfrom one contcxt of implementation to another. [. . .] Expcrience suggests that theseconstraints cannot Iic accounted hr in a discrete, linear fashion; rather, they \vi11 impingeon all aspects of innovati Ilalius design, implcmcntation, and evaluation.

122 NUMA MARKEE“When”While some adopters will implement a given innovation relatively quickly, others will nredmore time to carry out the same innovation.Thus, if one knows whcnA adopts an innovationand when B, C, or D adopt the same innmation, it is possible to specify the ratc at whichan innovation diffuses among a group of potential adopters and also to distinguish bctwdifferent catcgorics of adopters.Diffusion may be expressed as the perccntagc of adopters who implement an innovationover a giwn period of time (Kogers 1 98 3). Figure 10.1 shows a typically S-shaped diffusioncurvc.Thc lazy slope of the toe of the curve sho\vs that adoption at first occurs very slowly;if a critical mass of approximatel: 2 5 percent of potential adopters accept the innovation,it may take off. At this point, the slope in the midsection of the curvc bccomcs stccpcr(i.c., the ratc of adoption accelerates) as people “jump on the band\vagon.” Finally, the curveplateaus as diffusion slo\vs down and cvcntually tapers off, either because every potentialadopter has adopted or else I~ecausc the innovation stalls.%ofadoptionFigure 10. IAn S-shapctl diffusion curvcWith respect to diffusion ratc, fivc catcgorics of adopters haw been idcntified(Huberman 1973, Rogers 1983).These include innovators, early adopters, early majority,late majority, and laggards; as already noted in thc scction cntitlcd “Who,” people who neveradopt a particular innovation are known as resisters. In terms ofthe S-shaped diffusion curveshown in Figure 10.1, innovators antl early adopters occupy the first 25 percent ofthe curve.Early antl late majority occupy the steepest portion of the curve. Laggards occupy the lastpart of the curve as it flattens out to form a plateau.This information has at least two kinds of potential applications. First, it allow programdesigners who introduce an innovative syllabus to focus on those teachers most receptiveto the innovation. Since each category of adopter tends to be associated with personalcharacteristics bvhich are either conducive or not conducive to innovation, recognition ofearly adoptcrs would hc important to the innovation process. Second, such data allowevaluators to determine how successfully and how quickly an innovation has sprcad amonga group of potential adopters.“Why ”The reasons 1% hy innovations are adopted or rcjectctl arc many and varied. The sectionentitled “Where” already addressed a number of the sociocultural constraints that come intoplay. In addition, there arc individual psychological factors with respect to the personsinvolved, and innovations themselves possess various attributes that influence adoption.

THE DIFFUSION OF INNOVATION 123Rogers (1983) notes that individuals with particular psychological profilrs tend todisplay specific adoption behaviors. For example, individuals who adopt early tend to travelwidely and are usually \vcll-cducatcd antl upwardly mobile; they tend to seek out and beopen to ne\v idcas, and they tend to have a high degree of exposure to mass met1ia.Theircontacts with other people are often extensive, antl they are usually able to tolerate highlevels of uncertainty. Laggards, on the other hand, tend to display diametrically oppositecharacteristics while the pcoplc in between exhibit intermediary traits.Finally, innovations themselves po attributes which tend to promote or inhibit theiradoption. A number of writers (Bricknrll 1969, Henrichsen 1989, Kelly 1980, I.e\ ine 1980,Zaltman and Duncan 1977) havc proposed different sets of attributes of innovations. Theattributes proposed by Rogers (I 987) arc usctl here because they are derived from some1,500 empirical and/or theoretical studics on innovations across disciplines antl also becausethey are the most well-knolvn. Thcse attributes include the following:the relative advantage to potential adoptccs of adopting an innovation (i.c., the coor benefits);the compatibility of the innovation \vith previous practice (i.c., how different orsimilar the innovation is to \\.hat the potential adoptcr already uses);thc complexity of the innovation (i.c., how difficult the inno\ation is to understandor usr);the trialability of the innomtion (i.c., ho\v easy it is to try out in stages);antl the observability of thc innovation (i.e., hou visible the inno\-ation is).“How”In Henrichsen’s (1989) account of the extent to kvhich audiolingualism diffused in Japaneseschools in the aftermath of the Second World War, he notes that scveral difrercnt theoriesexist bvhich seck to account for how change occurs. These include equililirium thcory,evolutionary theory, conflict theory, rise and fall theory, and diffusion theory. Only the lastof these is directly relevant to language teaching. Within a diffusion-of-innovationsperspective, the cducator I lavelock ( 1971 ) distinguishes bct\reen three basic models ofinnovation. He laliels these three the Research, tlcrdopment and tlifjiusion (RD anti D) model, theproblem-solving model, and thr social interaction model rcspcctivcly, from Lzhich hc synthcsizcsa hybrid linkuge model. Similarly, the social scientists Chin antl Benne (1 976) idcntifv thrrrfamilics of innovation strategies which they respecti\ ely call empirical-r~itionul, normutive-reeciucutirz,antl power-coercive strategies of innovation. These models antl strategies “pair up”and haw lieen used, mostly unconsciously, by developers of various language teachinginnovations.Empirical-rational innovation strategies assume that pcople are rational and \I illtherefore be persuaded to adopt an innovation if it can he demonstrated that it is in theirrational self-intcrcst to do so. Such stratcgics tend to be used by pcople who subscribe toan RD and D model of innovation. A good example of this combination is the initialdevelopment of notional/ functional syllahu. by scholars associated \vith the Council ofEurope (Wilkins 1976).This model is rational, systematic, and theory-based. It depends vcry heavily on longtermplanning and involvcs a division of labor among teams of highly trained specialists \\ ho1vot-k on separate phases of an overall project. Thc planning process is basically linear(although fcxdback loops may be built into the framcwork) antl assumes that thc cnd productwill be used by a passive, though rational, consumer. Planning hcgins with basic research,

124 NUMA MARI

~ 19841).THE DIFFUSION OF INNOVATION 125ConclusionIt has been argued that the adoption of a diffusion-of-innovations perspective by practitionersis crucial to the development of language teaching theory and practice. Such a point of\ ie\\provides a unified framework for conceptualizing both the dcvclopmcnt and evaluation ofinnovations in languagc teaching. In ordrr to illustratc Ivhat issucs arc relevant tounderstanding a diffusion-of-inno\ ations pcrspcctivc on language teaching, innovation hasbeen analyzed in terms of Cooper’s (1 989) question: “Il’ho udopts what, where, when, rvhj, andhow.?” This framchvork providcs an appropriate set of criteria for analysis:1 a profile of participants’ socially defined roles and their adoption behaviors;2 a ctelinition of innovation in the context of language teaching;3 an account of the sociocultural factors which constrain innovations;4 a definition of diffusion;5 an overview of the prrsonal factors which constrain innovations, as \vel1 as theattributes of innovations which either promote or inhibit thcii- adoption; and6 a synopsis of various innovation models and stratcgics which ma); lie used to promotechange in language education.The most important characteristic of emerging “post communicative” approaches tocourse design ~~ approaches which are explicitly based on a diffusion-of-inno\.ationspcrspcctivc is or will be their focus on t\ro issues: 1) the extrnt to \z hich teachers actuallyuse new materials and approaches, antl 2) the dcgrce to which they actually reconstructthrir pedagogical values. This shift of cmphasis from design to implcmcntation antl evaluationis both dcsirable and also long ovcrduc.BibliographyAtlama, K. and D. Chcn. 1981 . The process of’ educational ~nnovation: An internutionul pcr.spcctir.c.London: Kogan Page in association with the UNESCO Press.Kcrctta, A. 1990. ‘Implrmcntation of the Bangalore projcct’. Applied Linguistics. 1 1 . 32 1-37.Brickcll, H. M. 1969. ‘Appraising thc cffccts of innovation in local schools’. In K. W.’Ijler (cd.)Educutionul evaltiotion: XCM. roles, new means. Chicago, IL: National Society for thc Stud) ofEducation. 284-304.Rrumk, C. 1981 . ‘Notional syllabuses revisited: A response’ . .4pplietl Linguistics. 2. 90 92.~- 1984a. Introduction. In C. Brumfit (ed.) Gcnerul English y,llabu.s design. Oxford:Pcrganion. 1-4. [ELT Documents 118.1‘Function and structure ofa state school syllalms for learners of second or fweignlanguages with hetcrogcncous needs’. In C. Brumfit (cd.) General English yllahus desi. Kcnnc, R. Chin and K. E. Corcy (cds.) The plunning of’changc, 3rdetl. Ne\vYork: Ilolt, Rinehart and Winston. 22 45.Cooper, R. L. 1989. I ungna,cle plunning and social change. Cambridge: Cambridge UniversityPress.Fullan, M. 1982. The meaning oJed~c~7r1onal chonge. NenYork:Tcachcrs College Prcas.Havclock, R. G. 1971 . ‘Thc utilization of educational research and tlcvclopmrnt’. Britishjournal ?f Educational Echnologj,. 2.84 97.Henrichscn, L. E. (1 989). Dijfiision !finnovations In English language teaching: The ELEC effort inJapan, 1956-f 968. New York: Grecnwood Press.Holliday,T. antl C. Cooke. 1982. ‘An ecological approach to ESP’. In A. Waters (etl.) /.ss.sues in

~ 198611~ 1989.126 NUMA MARKEEESP. Oxford: Pcrgamon Press. I24 143. [ Lancaster Practical Papers in English LanguageEducation 5.1Huberman, A. M. 1973. llndersranding change in cducation:iin introtluction. Paris: OECD.Kelly, P. 1980. ‘From innovation to adaptability: The changing perspective of curriculumtlrvelopmcnt’ . In M. Galton (etl.) Curriculum change: Thc lessons of a ticcade. Izicester:Leicester Univcrsity Press. 65--80.Kennedy, C. 1988. ‘Evaluation of thc management of change in LLT projects’. Appliedlinguistic.^. 9(4). 329-42.Krashen, S. and T.Terrell. 1983. 7he ncirurcil approach. Nc\vYork: Pergamon.Lambright, W. H. and P. Flynn. 1980. ‘Thc rolc of local burcaucracy-centered coalitions intechnology transfcr to the city’. In J. A. Agile\\ (ed.) Innovation research and public poliy.Syracuse, NY: Syracusc University Press. 243---282. [Syracuse Geographical SeriesNo. 5.1l.evine, A. 1980. LIyhy innorrrtion fiiils. Alhany, NY: State Llnivcrsity of NclzYork Press.Long, M. I I. and G. Crookes. 1992. ‘Three approaches to task-liasctl syllabus design’. TESOI.QiiarterLv. 26.27-56.Male?, A. 1984. ‘Constraints-bascd syllabuscs’. In J. A. S. Read (cd.) Trends in language .yllahnsdesign. Singapore: SEAMEO-RF1.C. 90 11 I.Markee, N. 1986.3. The importancc of sociopolitical factors to communicative course design.The ESP Journal. 5.3 16.‘Toward an appropriate technology model of communicative course design’.EnglishJor SpccJfic Purposes. 5.16 1 - 172.Milcs, M. B. 1964. ‘Educational innovation: The naturc of the problem'. In M. R. Milcs (cd.)Innovution in education. Ne\\ York: Teachers College Press. 1 48.Munhy, J. 1978. Coinmunicutii,c ~i.llcihus tlcsign. Cambritlgc: Cambritlgc University Press.1984. ‘Communicative ~vllalius design: I’rinciplcs antl prohlcms’ . In 1. A. S. Read (et!.)Trentls in language .;r//ahos design. Singapore: SLAMLO-RLLC. 5 5 67.Nicholls, A. 1983. Manc7ging eJucationul innoimions. I onclon: Allcn & Un\vin.Pralihu, N. S. 1987. Second lanpu

Chapter 11Zakia SarwarADAPTING INDIVIDUALIZATIONTECHNIQUES FOR LARGE CLASSESN H IS RES E ARC H F I N DIN G S hascd on responses from nonnative teachers ofI English from Indonesia, Japan, Nigeria, Senegal, antl South Africa, Coleman (1989) listsfour prol)lcms faced by teachers of large classes (1 OOf). First, thcy feel self-conscious,nervous, antl uncomlortalde; it is indeed tiring to lie the constant focus of 100+ pairs ofeyes for three to four periods a day. Secondly, large classes pose disciplinary antl classtnanagementproldems, in ivhich the noise level must be kept down so as not to disturbothers. Thirdly, it is difficult to evaluate the oral or \vrittcn \fork of so many learners;teachers of large classes seem to lie Iiuried under an endless pile of hoinc\vork. And lastly,teachers feel that liccausc intlivitlual attention cannot lie given, very little learning takcsplace.From 1983 to 1985, as a part of my studies for a master’s degree at Sydncy Unhcrsity(Sarwar 1983 85), I dcsigncd and cxccutcd a r rch project cntitlcd “li~aching Englishas a Forrign Language with Limited Resources. ne of the aims of the research \vas toexperiment with communicative language techniques and activities that \I oultl be effectivein large classes of loo+. Communicative techniques \voultl naturally include the Ixoadconccpt of individualization. Finding effective techniques for large classconcern in order to cxaniinc the teachers’ popular belief that in large clnominal and the interactive approach, relying on group/pair \vork, is not possiblc.Thc conccpt of individualizationBefore outlining my r arch and describing the tasks antl actilities that encouragedindividualization in my rners, I want to clarifj my terms of rcfcrcncc for thc conccptof individualization. The uinbrclla title, indirriduulizution, covers “such sc~mingly tlivcrsetopics as one-to-one teaching, home study, intlivitlualized instruction, sclf-aself-directed learning, and autonomy, I)ccausc thcy all focus on the learner as an individual”(Geddcs and Sturtritlgc 1982). It also encompasses a learner-centred approach to languageand takes special note of ethnolinguistic aspects of language lcarning, in which theautonomous rolc of the learner- is colourcd by their “second language self-image” and theteachcr/lcarner roles prevalent in thcir sociolinguistic sphere (Riley 1988). There are alsocertain underlying basic assumptions regarding learning whcn \vc talk of“individua1ization .”According to 1,ogan (1980):

in128 ZAKIA SARWARPeople lcarn even the same material ~ different ways (this implies acceptingdifferent learning styles).People can lcarn from a varicty of sour , cvcn if the final goals arc thc same ~implying that the instructional matcrials can vary.Direct teaching liy a tcachcr is not essential for lcarning; it is only one of many possibleexpericnccs ~ which mcans that a tcachcr can be a facilitator instead of a preacher.A variety of learning activities can take place simultaneously referring to integrativelanguage-learning activities.Pcoplc may have a variety of goals or objectives lor learning a second language ~implying that learners learn lor tliftcrcnt reasons.Another pcrspcctive is addctl I>\ Altman (1 98O), who clcarl) talks of three liasic tenetsthat charactcri7c Individuali/ctl language teaching:a syllabus that meets the nccds, aliilitics, and interests of cach lcarncrpersonalized goals, mcans, antl expectations for learnersteaching methods tailored to thc ncds of the learnersLogan’s assumptions and Altman’s tcncts \vcrc cxamincd for my research in gcneral,and lor the sclf-learning programme in particular, to dctcrmine how the conccpt ofindividualization could be exploited for lar-gc classes, \\.here learners needed (u) exposureto language learning, (h) activities for confidcncc building, and (c) a lcarner-centredapproach to build rapport bctween the tcachcr and the learners. Obviously, a tailor-madesyllabus antl teaching methodology for cach learner was out of the question for my largeclasses of 1 OO+. All the same, the learners \vcrc still considered to be the focal point of thelearning programme, with realistic appraisal that they \nm~ld all Ihllow their own pace oflearning and reach achicvcmcnt levels congrucnt with the goals they set for thcmsclvcs. Itwas also accepted that if Logan’s five assumptions lvcrc applicable in small classes, they couldvery well be applied to large classes, so long as thc lcarning programme offered thc learnersa variety of optional actil ities.The four Rs of individualizationThe \\.orking definition that emerged from these deliberations was the acceptance ofAltman’s “Thrcc Rs of Individualization: Reeducation, Responsibility, and Relevance” (Williamsand Williams 1979) ~ but mith thc addition of one more R, signifying Rapport, which canlie taken for granted in one-to-one instruction or in a small class. This rapport is difficult,though just as (if not more) essential, to achieve in a large class. In the context of myresearch, the meaning of thcse four lis of indivitlualization arc as follows.ReeducationThis means reconstructing thc role of the teacher as facilitator and the lcarncr as the activeagent in the process of learning. In the Pakistani context, this change nccds to be emphasizedall the more, since the teacher and the taught are both used to the lecture pattern oftcachingin which thc student is a passive learner as the teacher “talks” without any interaction orIireak for the whole teaching period.

ADAPTING INDIVIDUALIZATION TECHNIQUES 129ResponsibilityThis implies that learners take charge of thcir own learning, For the Pakistani learner thisis a conceptual leap as they are used to rote learning and lack confidence in their owncognitive capabilities. It also implies the teacher’s responsibility to set up clcarly stated tasksthat can be monitorcd by learners on their ovvn antl ensure the availability of self-learningmaterials for learners.RelevanceA? most of the glossy and readil) a\ailalde material 15 tleiisctl for the nonnati\e lcarncrstud)ing EFL in the West, we need matcrials rclclant to our learners. Alw, rrlr\anw meansfinding contexts of learning that arc mcaningful for our lcarncrs.RapportA class of 100+ is a class-managcmcnt chalknge for any teacher. It is only through the properrapport that an atmosphere conducive to learning can bc built up. Also, “humanizing” a lai-gcclass is perhaps thc only way to motivate learning.The research programmeI devised a 50-hour rclmcdial English course that focused on reading comprchcnsion andwriting skills. It was a voluntary, non-credit course, taught for two hours three times a week,and lasting approximatcly right and a half weeks. The learners (104 volunteers, who 1vei-eselected on a first-comc, first-scrved basis) agreed to stay after their regular classes for thiscourse.Thcy were young fcmalc adults between 16 antl 20 years of age, the majority comingfrom a middle-class background. Thesr students \vere false beginners of English, havingstudied it For approximately n ycars. They had little or no exposure to English in theirday-today lives except for studying it as a “subject”in the Faculty of Humanities.Whilc discussing thc lcarncrs’ expectations for the programmc on thc first day, it \vasmutually agreed that since 50 hours of class work would not he sufficient for any tangiblcimprovement, the learners ~vould supplement their bvork by following a self-monitoredlearning programme that included listening, reading, and writing skills. Learners ivere givena three-part qucstionnairc hcforr and at the end of the project to cvaluatc thcir progress.The questionnaire was devised to find out (1) thc Icarncrs’ background, (2) thcir attitudetowards learning English, and (3) their proficiency level in English, through a readingcomprchcnsioncheck and paragraph writing.The response performance as \vel1 as feedback from the project group ~ \vas veryencouraging. For the purposes of this article, hom r, I shall only focus on the steps uscdfor putting the concept of individualization into practice. Thc four Rs were taken as arcfcrcncc point in a two-pronged thrust: (u) individualization in largc classcs, and (b)individualizing learning tasks (see Figure 1 1.1).Individualization in the classroomThis section deals with the last R of individualization: rapport. It focuscs on activities that“humanized” this largc class for me by helping me familiarize mysclf with the learners aspersons. It also helped to a great extent in class management.

130 ZAKIA SARWARVoluntary learningThe 104 students mho enrolled tor the language project (LP) clitl so \oluntaril) Thc) meretold that the aim of the programme mas flucnc! rather than accurac!, antl that they \~ouldbe taught skills rather than prescribed textbook4 The\ were under no pressure to pin thecourse ~ especial$ as it \%a\ a non crdit"unollic ial"course, carricd out a? part of a researchprojcct. In the lollo\\ ing >cars I hac used the concept of \oluntar) lcarnmg ti) consultingthe student\ at thc lxynning ot each academic !car betorc setting up the >car's teachingprogramme tor compulsor\ classesBackground questionnaireLearners \vere given an hour-long questionnaire on their first day in class. Learning abouttheir background, attitudes, and pcrcci\wl ncctls, as \vel1 as their proficiency level, helpedme a great deal in undcrstanding my learners. It also clariticd thcir course expectations.The responses gave me information about the socioeconomic antl ethnolinguistic communityof the learnel-s, Lvhich proved uscf~il in organizing their group activities and outside classprojccts.GroupingOn the \-cry tint clay, after introducing thc co~irsc, I askctl the learners to form groups ofthree or four \vith their friends. As a numlwr of activities \\ere to use up time outside theclass, \\e figured that it \vouId I)c casicr for Irarncrs to do their group tasks together in theirfree time in friendly groups. Forming thcir o\vn groups also ga\e the learners moreresponsilility in sharing the class-managcmcnt issues. Groups of four \\ere then given anumhcr to identify their group. The) \vcrc also askctl to sit togcthcr in class so they couldsharc thcir group activitics. Instcad of moving ai-ountl, \ve had pcrmancnt groups xvith apcrmancnt scating arrangcmcnt.Name tagsOn thc second day Icarncrs ci-c rcclucstcd to make namc tags for thcmsclvcs by \vritingtheir name antl group number with a thick marker- on a card mcasuring 3" X 4".Thcy wcreto wear them as part of the class uniform throughout the duration of the coui-se.This put aname to a face, which is otherwise impossildc in a class of 104! It also made it casicr for meto address everyone by their first names during activities antl discussions. The magic of thefirst namc also I)rought a more congenial atmosphere to the classroom, as classmates liecamemore familiar with each othci-.Thcrc \\as definitely a better rapport tict\vccn \ arious groupsas well as with mc.Projle cardsa ici-4 intci-c sting 1% a! to get to knoxz the learners lxtteiF~rst, the points that \I ere consitlerctl M orth kno\\ ing alwut an) onc \T crc clicitcd in abrainstorming session and put on the I~lacklioartl 5ccontllj, thew points mere catcgorvedantl put in an order acceptable to r)onc. I hen students wcrc askctl to prepare their ownprohlc cards, complete ith thcir photograph antl thr tlctail\ listed on the ldackhoard. It\\as announced that thrcc pri/cs \zoultl be gncn to prohlc cards \\ ith good handwriting,correct spclling, and a neat, attract11 e prcseritation.Crcating profile cards pro\ctl to IK

ADAPTING INDIVIDUALIZATION TECHNIQUES 131I learnt a lot ahout m! learncrs’ aspirations, strong points, and mcak points, and ha\ ingtheir photograph.; on the profile cards certainl) helped me tamiliari/c m! self \\ ith theirfates The students cnjo!ed making thew cards, too Another ad\antage of ha\ing their namesantl addrc5ses on their prolile card \\as that I \\a\ able to reach mj learners b! mail after asutltlcn closing of all educational institutions (a frequent happening hercl) antl send thcmguidance on hov to continue working on their nLending libraryAs the majority of the 1.1’ learners did not have access to English hooks, magazines, andnebvspapcrs, I gathered from my f‘riends and brought to the class used, simple stoi-y books,glossy magazines, and the magazine section of daily nc\vspapcrs.T\vo or threc of these \vcrctlistributed to each group, lvho Lvere to he (u) responsiblc for rotating antl exchanging them\\ ith different groups, and (h) returning thcm to me at the end of the course. My purpose\vas siinply to provide materials for extensive rcading. I did not check to see if all the studentsused these, hut they \vcre film, fashion, antl sports magazines, generally liked by tccnagegroups. I sa\v a brisk exchange before and after the class, so 1 presume a number of studentsdid use these hooks antl magazines. The responsibility of keeping track of these magazinesgave the learners a sensc of importance.Individualizing learning tasksI \vi11 discuss here selr-learning programmes (SLP) aimed at improving students’ languageoutput as well as encouraging thcm to becomr independent learncrs. Training learners tomonitor their o\vn learning is as important in a large class as in a small one ~ in fact, moreimportant, bccause in a small class, work can he supervised 137 the teacher, but in a largeclass this is virtually impossiblc. Hence, thc best chance that a lcarncr in a large class has isto take responsibility for his o\vn learning. For the SLI’, all the three Rs of individualizationmentioned by Altman \vcrc considered. Students \\ere given guidelines to (u) monitor theiro\vn scoring, (h) do pcer correction, and (c) Lvork independently on group projccts. As aReadinglwritingcloze exerciseRadio newsquestionnaireWriting picturestorieslpeercorrectionlearning tasksLetters to:-Visitors-Teacher- Each otherGroup projects:- Making readingcards- Tableaus/songsVoluntarylearningin classResponsibilityclass libraryGroupingwith friendsProfile cards

132 ZAI

~ Handwriting~ GrammarProofrcaciingPunctuationReadingADAPTING INDIVIDUALIZATION TECHNIQUES 133to copy a paragraph a day from their prescrihed textbooks, hying out words, filling themin later, antl then checking with the textbook again (see Figurc 1 1 . 3).Please usc your prcscrihetl English tcxtlxiok tor the cxcrciscTr\ to \\rite at least one paragraph cvcry day.Use the attached handout as a model tor your hantlxlriting.The aim olthis cxcrcisc is to impro\c your~~~skillscomprchcnsionStep I(Week\ 1 anti 2)uhcdeEclcct a paragraph from )our Fnglish tcxtlmokMark or undrrlinc c\ cr\ 7th \\ ortl.Copy the passage in your best handwriting, Icaving out the markc.. $5 ordsinstead.Cloar. the I)ook. 'l'akr a I)rrak.Fill in the hlanks.ka\v a blank linc(Wccks 3 antl 4)I ea\ e out c\ cr) 6th \\ ortl in Ctcp 1 u(Wcck 5 onuards)Lea\ c out r.~ rr! 5th \\ ortlStep 2Check your \vork:Havc you put in J margin?Havc you put in thr tlatr?Havc you indrntrd thc paragraph?Lhcs thc writing look ncat and tidy?Oprn your tcxtbook and chcck if you punctuatcd your work corrcctly.Chcck your I-csponscs in the hlanks.Chcck thc numl)cr of blank.; antl give yoursclf onc mark lor cach correct ans\vrr.Figure 11.3 Worksheet 2: self-created clozcThc usual practice is to give an unseen passagc for clozc to teach/test comprehensionor itcmizctl grammar. Rut in the pilot testing of materials I discovered that my learners facedgreat difficulty if they wcrc unfamiliar with the text. Copying from familiar texts made thecxcrcisc easier for them. The feedback confirms that a number of' thcm improved in theirscores with practice ofthis adapted version of clozc.Thcy also Iiccamc more confident whenthey attempted regular cloze cxcrciscs. Furthcr, comparing their writing with thc prcscribcdtext, they got training in proofreading thcir own work, which highlighted thcir omissionsand careless mistakes.

signifying134 ZAI

ADAPTING INDIVIDUALIZATION TECHNIQUES 135Implications for teaching/learning in developing countriesMy rcscarch started with the basic assumption that cia in countries like Pakistan arc notlikely to be reduced in size in the foreseeable future. Hence, solutions have to be realistic,within the limited constraints of the present teaching/lcarning situation.The acceptance ofthis reality can help a teacher to overcome the psychological barrier that the interactiveapproach/activitics cannot be used in large classes.This assumption led me to seek out newways of managing the class and individualized activities.ptancc of reality also led to setting up rcalistic, mcasurablc, short-term achievablegoals, which had a reinforcing effect on the tcachcr and learners. What and how much cana teacher/learner achieve, given the Icarning conditions that prevail in large classes? Ateacher obviously cannot meticulously correct a huntlred papers every day. In thc same my,a lc-arncr cannot learn flahvless English with limited exposure to the languagc. Therefore,the initial targct was fluency rather than accuracy, providing learners with an occasion to“use” the languagc in real life.[. ’ .1The broad concept of individualization and the whole structurc of the project dcmantletla drastic change in the tcachcr/lcarner roles. Again I started with thc assumption that directtcaching or lecture is only one form of learning experience (Logan 1980), and that adultstudents are capable of taking their learning into their own hands.The transition from learnerdependcncc to independence was not an easy process ~ especially in a system of educationwhcrc spoon-feeding and rote learning are common teaching/learning stratcgics. But theskill-based approach tlemantling cognitive interplay \vas a challcngc to a numhcr of students,It moved them to\z.artls relying on their own judgements and conclusions, so that theybccamc- gradually indcpcndcnt. On the othcr hand, as a teacher, relegating learning tasksand responsibilities to studcnts involved an clement of risk and ensuing frustrations. For ateacher used to complete control of the class, this \vas initially not an easy task, but thcstudents’ responses antl enthusiasm lent a lot of support. In the last stages, their increasedoutput and productivity liecame a rc\vard in itself.I. ’ .IConclusionBy incorporating individualization tcchniqucs my classroom rcscarch addressed three majorELT prohlcms: large classes, the dependent learner, and lack of exposure to real-\vorldEnglish. Now what is needed most is its replication so as to e\aluate the variables involved.Ideally, this replication should be done in Pakistan as well as in countries whcrc similarteaching/learning conditions pre\ ail. In contemplating such research, the follolvingsuggestions should be kept in view.The hasic materials and outline of the rcscarch done so far should be picked up, withadaptations and changes suitable for the age and 1 1 as well as thc socio~cthnolinguisticbackground of the learners. The rationale behind the hroad concept of individualizationshould he atloptcd as the basis of thc approach used in handling large classes, and the focusof the rcscarch should he on activities and techniques tha uld lie effective in large classes.Abovc all, more classroom-based resrarch in large cl involving practicing teachersshould he encouraged by institutions, organizations, and drvelopers of syllabi and materials.No doubt the picture of a large class of loo+ appears sad to those who have never hadthis experience, yt it is a condition faced by more than half the world’s population ofteachers antl learners. Hcncc it is of vital importance that action rcscarch involving largeclasses hc givcn high priority. 1. . .]

136 ZAI

I.Chapter 12William Savage and Graeme StoverAN EMERGENT LANGUAGE PROGRAMFRAM EWO RI

138 WILLIAM SAVAGE AND GRAEME STORERIThe AIT aquaculture outreach projectFunded by the Overseas Development Agency of the United Kingdom, the AquacultureOutreach Project ofthe Asian Institute ofTcchnology (AIT) is liasetl in Udornthani, a majorcity in the resource-poor northeast. Taking an interdisciplinary research approach tofreshwater aquaculture for small-scale farmers, the project's main objective is to dctcrmineaquaculture strategies which are sustainalile, using inputs (feed, nutrients and matcrials)which arc readily available. Kecommcndations for fish-farming practices arc generated bya methodic flow of' information among the project, the participating farmers and on-stationresearch at AIT.A delicate aspect of the project concerns the manner in which farmers are recruitedand participate in the project. Thus, Iiroject staff, king from the northeast regionthemselves, are crucial to the project's efforts in that they are of the same culture and speakthe same first dialect as the farmcrs.Thc project staff in the Udornthani main office and twosub-offices in other northeastern provinces were the group of learners for whom a languageprogram \vas requested.Our initial contact with the project's manager, a native English spcakcr, highlightedseveral general reasons to learn English. I'hc funding agency is from an English-speakingcountry and the project often receives visitors from that agency and others, as \vel1 asEnglish-spcaking rescarchcrs who often stay for weeks or months at a time. Aaquaculturc research information is essential for many staff' members and all the data whichare collected in the field are ultimately reported in English. Finally, it seems likely that theproject will be extended to other countries in the region, itlcally drawing on the cxpertiseof the Thai staff.Thcsc reasons established the motivation tor a language program but did not give muchinformation about lvhat needed to lie learned. To that end, a one-day visit to Udornthani\vas arranged during kvhich one of the trio teachers met the learners to discuss with themtheir work-related nccds. The findings from the site visit \vcrc usctl to set up a two-dayplanning workshop at the AIT campus in the northern outskirts of Bangkok. Sevenrepresentatives of the staff carried out work-rclatccl tasks alongside scvcn counterparts whohad participated in language programs at the AIT campus. In this \Yay, information was builtup from the site visit through thr planning \vorkshop to provide the content for the 2-w~ekintensive workshop.I1 The language program frameworkOne of the aims of the language program, as stated liy the project manager, was to gct theparticipants to a point where they would bc aldc to continue to develop thcir English abilityThat is, the conclusion ofthc program \vas not to I>(- view-cd as an end point,but as another starting point in itsclf. For this reason, ive have avoidctl usink" 7 course" in ourterminology as, for us, it strongly suggests a discrete end point. We also feel that a courscimplies a set of content which is presented in hasi he same form each time it is regularlygiven. The content of the language program bed here is unique, not only to itsparticular situation but also to the group of learn 're lve to go through a similar proccssof developing another language program at the same location, hut for a different group oflearners, the content would be quite different. Therefore, \vc will refer to the two-weekphase as a workshop or simply as the two-\\ k intensive. We vimv thc program and itsdevelopment as threr-phase: the one-day site visit, two-clay planning workshop and two-\vcck intensive workshop.

AN EMERGENT LANGUAGE PROGRAM FRAMEWORK 139I ,site visit ---- - > planning workshop --- -- > intensive workshop/\ // /\I1I1I1I1I1I1I1I 1I1I 1I 1I1I1I1I1I1discussionsto identifywork-relatedneedsParticipant:working on tasksreporting backexpandingevaluatingIITeacher:observingassistingunderstanding learning needs______Figure 12.1 ‘I‘hc initial languagcprogram framcworkIn approaching the language program, the teachers began \vith the framework illustratedin Figure 12.1 . The rolc of the participant in this initial framework was to work on tasks,rcport back, expand and e\ aluate. The role of the teacher was pcrccivcd as observing,assisting and understanding the learning needs as the participants were working.Identification of needs \\.odd not lie confined to one phase, but would occur throughoutthe program.The one-da,v site visit (Udornthani)During the site visit, one of the teachers met with the project staff. Through discussion,they were able to identify the following work-related needs:1 EXPLAINING(a) The work of the project (methodology antl recommendations) to visitors(11) Figures antl graphs2 DESCRIBING experiment results and analyzing data3 REPORTING from farmer (lata collection forms4 SPEAKING and LISTENING5 WRITING(a) Monthly reports of work progrcss(b) Summaries of the monthly reports(c) Subproject reports6 READING antl WRITING(a) Scientific project reports(b) Office memos(c) Farmer report formsThis information was inadequate in that it merely prescribed a set ofcontcnt to be taught.(Should we now offer a course callrd “Writing Office Memos”?) It did not tcll us what thelearners could already do in English and what language learning concerns they thoughtneeded to be addressed. This led us to the two-day planning workshop.

140 WILLIAM SAVAGE AND GRAEME STORERThe two-day planning workshop (AZT campus)Thc site visit information tlcf‘inctl the tasks filr the planning \vorkshop when scvcnreprescntativcs of the Iirojcct stall‘ \vcrc pairctl \vith seven AII’ cainlius staff. The tasksconcerned writing (monthly reports, suinmarirs of monthly reports antl translating reportsfrom Thai to English); tlescrihing antl presenting information ahout field w.ork, includingphotographs of a farmer’s fish ponds; pi-eparing captions for project extension media; anddescrihing office I>roccdurcs. The i-cliortliack sa\v cach staff mcmlwr speaking to the wholegroup about the task, Lvith the partners stepping in to help out \vhcn necessary. Because thcpartners’ work also rclatcd to aquaculture, there \vas a gi-cat deal of support and sharing ofideas. For some, it was their first occasion to use English to discuss their 1vol-k. During theion, thc tcachcrs \vcrc ahlc to olxcrvc the participants using English antlplaced to comment on thcii- needs; the participants themselves wcrc betterablc to discuss their lcarning ncctls. Mr Vorapong, for example, finishctl his reportback byannouncing “Wc want to write (thc farmer report forms) in English.”Whcn asked \z hy hedid not do so, he began to talk about his limited vocabulary and his lack of’cxpcricncc inconstructing sentences. ’l‘hat is, he I qan clarifving his nccds for himself antl Ihr us.On the second morning, the AIT campus staff spokc of lvork they hatl done whilestudying English, expanding on language learning conccrns cxIircsscd hy the project staffthe previous afternoon. Oncc again, thv focus of the discussions moved from simply talkingabout work-related nectls to more specific learning concerns. Foi- example, Mr Pirat said:“I \\.ant to use English c ryday, but I [am] shy. 110 you have this problcm?”Mr Supong, oncof the AIT partners, ans\vcrctl by talking aliout ho\v he hatl ovci-conic his shyness when hefirst hegan to learn English; he \vent on to explain about strategies he had tlcvrlopetl forlearning vocabulary.This widening of focus in the i-cportlmcks meant that the tcachcrs heganto rethink their initial program framcw.ork as the tcachcr and participant roles as originallyperceived xz-crc no longer distinct. Thc final afternoon \vas devoted to a spoken evaluationof the tLvo-tfay planning \vorkshop, during n hich cach participant clatmi-atctl on areas thcvthought ncwlecl tlcvclopmcnt.The two-week intensive workshop (Udornthani)The tcac hci s met \z ith thc scicn I cpr hctorc thc intcnsi\ c u (11 kshop to diwussthcir plans tor thc trio necks and tot in the oi icntation On the Iii-stmorning, each ot thcponsibilit) tor orienting the othcrsc\entccn staff mcmbcrs to the program and about M hat might lie cxpcctctl the! discussedthe iinpoi-tance of tr! ing to use English, ot asking tor help \\ hen “stucY antl the use of thetirst language, the! talked about keeping a loghook, the! spokc ahout deleloping thetechnical tcrms tlictionar! (an idea \I hich hatl come up during the planning workshop), andthe\ shmed photogi aphs takcn during the planning norkshop and explained 1% hat washappening in cach one, thcrcln inti oducing thc othcr statt incmlicrs to the itlea of orki kingon tasks and reporting hack.The first task had hccn dccitlctl on I>\ thr tcachci s and the groups had bccn prc lormetlIn the second task, though, the participants \zcic asked to select an aica fiom the ofhcc’smeekl) suli-projcct sheet to talk allout and the) ncrc encouraged to form their OM n interestgroups 1 his mol ement from tcachcr tlchnccl tasks to tasks idcntihed 11) the participantsthemsclx cs \%as integral to thc progi-am approach B! thc sccontl \z eck, thc participants\\ere lorming their ov n groups antl dchning ai cas of intci-est to nork on for the final postersession

AN EMERGENT LANGUAGE PROGRAM FRAMEWORI< 141I11 Identified needs and their realization in the programDuring thc planning M orkshop, the participants and teachers \T ci e able to clarif! andelaborate on the nerds that hatl been identified during the one-tla! site \isit 5ome need5had been immaturcl: defined, tor cxample, the reading and mriting of farmer report formsWhile the teacher5 underitood that the report form5 hatl to be written in English, theparticipants made it clcar thatParticipant 1 :Participant 2:The quantitative data on the forms are not translated into English butarc coded antl then entered into the data basc.The qualitative data on the forms can be in notr form, not completesentences.After the two-week intensilc, the teacher5 decided to trace the needs through the programantl to scc hov the) had been rcali/etl and handled through acti\ities in the tno-meekintcn5i\ eDuring the planning \corkshop, both teachers had kept tletailcd notes The sesiions hatlalso been \idco-tapetl In the two-meek intensiTc, a tlail:, log nas kept which dctailcd each(la) 's acti\ ities Thc notes, the \ideo record and the dail) log sen et1 as data In re\ i e ing ~the data, the teachers looked for learning needs directl! exprcsscd b:, the participants; antllearning needs ohscr\ ctl b> the teachers as the participants were using English The learningneeds tell into four groups(A)(B)(C)(D)interacting ~ includes such acts as explaining, describing and discussing, as \\-ell as thefrequently mentioned listening and speaking;langtiage use ~ represents mcchanical language abilities;writing and reading ~ contains all references to materials to be \vritten and read; antl,rnetacognitive ~ refers to commcnts about managing thc learning process itself.Ihc learning needs in each ot thcie groups and ho\\ the! \\ere rcali,xd in the t\\o-\\clrkintensire appear inTableq 12 1 4IV Rethinking needs analysisAt the beginning of the paper, hvc asked se\eral questions kvhich \ve consider to be basic tothe work done in ncctls analysis. Looking hack on our approach, it is apparent that \IT havedealt in practical tcrms with these questions, ones still being discu. d in the literaturc. Intheir approach to ESI' language ncccls analysis, Hutchinson antl Waters distinguish targetneeds from learning needs (1987: 53 ~63). In doing so, thry move beyond the cattlgorizingof linguistic features lvhich results from instruments such as Munby 's CommunicativeNeeds Processor (1978). But what they describe as learning needs can, in fact, lie stxcn asinstructional logistics needs. For examplc, the learning needs rclatc to questions of thcpurpose of the course, background of the learners, types of instructional resources, andlocation and time of the course. When Hutchinson antl Waters focus on targct needs, theyview lcarncrs as being short of the mark, or lacking (1987: p. 58), rather than as peoplewho bring their own cxpcrience and expectations to a language program.It was this neglect Lzhich prompted us to develop an approach which \vould moreactively involve lcarncrs in the needs analysis antl design of the language program. Jacobson(1986: p. 173) approached ncctls analysis"in terms ofthc strategic competcncc that students

142 WILLIAM SAVAGE AND GRAEME STORERTable 12.1 Interacting1 Explaining(a) Extension matrrial dcvclopmcnt(I)) Concepts antl ohjcctivcs inprojcct mctlia(e) Project recommendations(d) Connections bct\z ccn variousstaff duties(e) Figures antl graphsInteracting \vith visitors;rqmrtback: tlcscription of radio station surveyfor vxtcnsion, evaluation of extension mctlia;“muscov! duck” videolicportl>ack on 1st 8( 2nd tasksInteracting \vith L isitors; student vitlcos; poster session:“Horv to get farmc’rs to grokv fat tish”Reporthack: discussion about Khmer and Souay dialects;vitlcosI istening posts2 Describing(a) Physical features o1’pontlsystems(1)) I’rocedurcs*I’ostcr sessionI’ostcr session: tish fry transfer, how to deal withI isitorsreportback: new criteria tor villagc~ sclcction3. Discussing Lvork dutieslicporthac,k: recruitment antl follo\v-up in one suh-officc;\vcckly meetings; farmcr visit forms4. Interacting on the telephone5. Speaking antl li5tcningI-ocus of all reporting Imck; listening posts; studentvitlcos; \\ cckly meetings held in English; discussions ofmeek I evaluation to set up Lveck 2; defining antlclarifying tasks* Thesc procedures arose during the 2-1zcck intcnsivc; onc othcr, related to ofticc procedure, \vas notatltlrcsscd t)ccause of a lack of time.need in order to successfully carry out the work required in the [university physics] lab.”His task-based approach primarily involved direct observations in the lab environment inwhich the students were working, and interviews with the lab instructors. Includingobservations of what the learners actually had to do with language marked a major additionto what had typically becn put forward as methods for collecting information about languagelearning needs, for example, with questionnaires antl interviews (Mackay, 1978: p. 2 1). Butin the end, what ensued was the delivery ofa prescribed syllahus whose purpose was to fillin the gaps identified.Widdowson’s discussion of needs analysis wends its waj through the inadequacy ofrcgister-liascd analyses to arrive at the desirability of considering “aspects of discourse”(1983: p. 85). In order to do this, he argues, we nccd to devise way5 to engage learners in“procedural work” which will convert items of knowledge about language into “actualized

AN EMERGENT LANGUAGE PROGRAM FRAMEWORK 143Table 12.2 Language uscIdenr!f;cd needsHow reo/i/cd clnring 2-week intensii eIVocabulary(a) Improper choice (c.g.rccommcnd for collect)(I)) Inatlcquatc to coniplctcforms(c) A\ oitling circumlocution(tl) Technical termsA focus of the tcchnical terms dictional-!Teacher input antl some student student corrcctionFarmer visit \vorkshcctsRcportback antl meekly mcetings: usc of media and rediaRecording \vords in logbooks; contcxtualizing \vords fordictionary; larmrr visit \\ orkshects2 SIntax(a) Connwting ideas(b) I inking ivithin paragraphs(c) Consti-ucting sentences3 'l'cnnc4 Introducing a topic in \vriting5 Writing clcai-1)'1 cachcr input olconncctorsWriting memos, minutes of \\ cckl?- meeting antl videoscriptsAs abo\ cPreparing for reporting back; \ ideo scriptsPreparing for \vcckly mcctings; recorders in mccting;minutes; video scripts; contcxtualizing vocabulary;cxplanations foi- dictionary and for setting up listeningposts6 Gi\ ing details Focus of "cxpantling"Tohle 12.3 Writing antl reading1dentlf;cd needsHon reu/i/ed ciiiring 2-n cek intensive1 Writing(a) Farmer \isit l'orms(I)) Intcrnal memos(c) Mcmos to report unusualdata(d) Monthly reports*(c) Monthly report summaries*(t) Report outlines(g) Abstracts of books and articlcsFarmer \ isit \\ orkshcctsManager reported an increase in thr number of internalmemos \\ rittcn in English; teachers askctl to chcc kI>id not arise1)iscussctl ivith manager but not follo\vctl through2 Reading(a) Incoming memos(b) New sub-projcct tasksMrmo from sub-oftice in tnglish: discussed antlre\\ rittcn at Meekly mcctingInformed second taskPrcparation and rollo\v-up for \vcckly mccting* Note that \\e had intcndcd to hold a tvriting \vorkshop in the second \vcck. This plan \vasabantlonctl as it was felt that there \vas just too much else going on. Writing was addressctl in othcrareas, c.g. memos antl mccting minutes, though this \vas only at the scntencc/paragral'll Icvcl.

144 WILLIAM SAVAGE AND GRAEME STORERTable 12.4 Mctacogniti\c1 Acking others ahout \\ortl\ notunderstoodl-xainplc\ gi\cn allout ho\\ to ask lor helpMan) example\ ol \tuclcnt stutlrnt and \tutlcnt- teacherrequest\ lor hrlp2 Lrarncrs assisting each other Participanta used to jvorking as a team (a l’cature of\vay the projcct is set up)Moat apparcnt in preparing lor reporting hack, mcctingsantl postrr session3 Overcoming shyncss to spcak This \z as commcntctl on by 6 of thc participants in their(conliticncc)Ivrittcn evaluations4 U\ing L 1 to explain L2 Scrn in larnicr visit lorms and tcchnical terms dictionary“Muscovy duck” video* All spoken to in the orientation by the participants from the planning workshopcommunicatile behavior” (1 983: p. 87). Kenny’s (1 985) re\iem of Witldowson’s LanguagePurpose antl Language IJ\e (1 98 3) ad(lcd thi5:An analysis by the teacher ofthe learner’s conccptual requirements in the defined fieldwill point us in the direction of thc required tliscoursc . . . The selected discoursebecomes in the lesson the ohjcct we respond to, dissect and discuss, and thecommunication we share. Ho\v is it conccptually organized? It is all right?What exactlyis meant? Do wc agree? Might WT add to it? Should \c-e elaborate this point? Cansomeone explain this?(1985: p. 177)Inherent in these questions is negotiation and through such a process an understanding oflearners’ language needs can bcgin, as lcarners arc cngaged in Widdowson’s “proccduralwork.”At the same time, \ve arc forced to consider methodological issues as bcing at onewith finding out Lvhat learners know and what they need to know. Does the methodologyallow for previously unidentificd needs to be addrcsscd or is the content of the program setin stonc beforehand! Does the methodology allow for futurc needs to be handled by thelearners “to achieve their o\vn aims after the course is over 1) applying the procedures theyhave used in learning to the continuation of learning through language usc” (Widdowson,1983: p. 91)?This concert of needs anal and mcthotlology is central to the approach presentedin this paper; by engaging the lcarners in these conccrns of program design, the learningexperience is readily accessible to the participants in terms of the content and their abilityto participate. Nccds analyzctl concurrcmtly with the program and embedtled in themethodology must tie of immcdiatc relevance. A methodology which fosters learnerautonomy sustains momentum to continue learning; it becomes “a catalyst for learning”(Folev, 1991 : p. 69).The validity of any approach to identify antl addrcss the language needs

AN EMERGENT LANGUAGE PROGRAM FRAMEWORI< 145of learners is ultimately cstahlishctl 1: “how effectively it achieves its declared purposc ofdefining the content of purpose-specific language programs” (Widdowson, 1983: pp.85 86). In the previous section, we gave cxamplcs of needs identified by the learners andhow these \\-ere realized as content. Let us now turn to placing the program’s aspects intoa hackground.Working on tasksWe depart from the types of‘tasks discussed in /.angnage /.earning Tusks (Cantllin and Murph!,1987) in one vital respect: the tasks are derived from and dcfincd I y actual work situationsin Lvhich the learner needs to use English. As one of the AIT partners observed in thcplanning workshop, the work content can scrvc as the language lcarning content. Work-rclatctl tasks arc suitahlc for dctcrmining lcarning needs hccausc thc use of tasks allo\vstcachcrs to cstahiish “the rules [the learner] is using and the . tcms and categories he isworking with” (Corder, 1981; in Im-sen-Freeman and I,ong, 1991 : p. 41).The advantagefor the learner is that it allows him to focus on what he can do, to locate his starting point.It is important that thc first task bc apIiropriatcly sct up so that, on reporting liack,language antl content arc generated to allokv the participants to procccd. In the case of theplanning workshop, this \vas achieved by pairing the participants \\ith counterparts \\ hocould advise and assist them. In fact, they were helping each other. MrTanin, an AIT partncr,commented that although he had helped his partner with vocabulary, his partncr had alsoexplained aquaculture concepts to him.Reporting backReporting hack comes from the \vork donc at the Language Ccntcr of AIT in thedevelopment of its prr-sessional master’s program,Talkbasc. A reporthacka focus on mcthod, a sharing of information and reciprocal curiosity allout what others arcdoing or haw donc, and a first attempt at narrowing tlohvn a \vide and unmotivated topicto one which is both managcahlc and of personal interest to the students” (Hall and Kenny,1988: lip. 21 -22).‘livo rclatctl points need to be emphasizrd hrrc. First, mcthod is takcnto mcan thc lvay in lvhich the task \vas accomplished. In our approach, bccausc Icarners aredealing with work-based tasks, the mcthod for doing thc task during the language programantl for doing the task for work arc one and the same. Second, it follows that thc topic isalready of interest as it is dcfined hy the learner antl involves the attainment ofa work goal.The “narrowing down”1ircoines a process of managing the topic within the learners’ currentability Ic\~cl.ExpandingAn increased abilit! to deal mith the content of the task at a more challenging leiel usinglanguagc lust bc!ond the cui rent lelel of abilit) comcs allout through expanding what thelearner has to sa) during a rcportliack Thc participants’ current knov lctlgc of language usetl in the accomplishmcmt ot the task, upon 1% hich can be huilt a greater aliilit)to report hack Problems atldressetl in the tasks arc naturail! centered on the leal ncr \z hoIlcncfits from guidance, not on11 lrom the tcachcrs hut also from other learners, thusexpanding the scope of the learner’\ task. Or put another \\a!. “student5 [hac] a plan torfurther action u hich might iniollc exploration of further sources ot data, a rcdchnition oirefinement of topic ara or a search for mol e detailed information” (Hall antl Kenn), 1988

146 WILLIAM SAVAGE AND GRAEME STORERp 22) 1 his pointing tonartl an cxpandctl, elaborated goal is at the le\el of the content ofthe ta5k and reportliack antl language I\ dciclopcd to reach thc ncxt point Thus, ourundcrstanding of expansion of language is that it occurs lxxausc of a need to discussexpandcd content and not as it has Iieen litnitetl in Wid(lo\z son’s formulation.generuli/ution + clarification + elaboration + cxempl$cution(Widdon son, 1978: p. 1 33)A language expansion sequence such as this seeins more communicatively useful whenapplied to the content which language is king used to transmit and not simply to thelanguage’s structural representation.EvaluatingEvaluating is seen as “a rcgulai- and continuing process” (Rea, 1987: p. 165). lhe kejcharacteristic of such cvaluation is that it is integral to lcarning antl teaching. In practice,this means that the language program participants (learners and teachers) arc explicitlyaware that whatever is going on is ultimately open to evaluation; they question how a givcntask \vas accomplished and how it could tic improved. For example, such improvementsmight concern the need to develop a greater ability to talk aliout a certain task during areportback session.Thcn the language nccdcd can be input to the learner, from the teacheror from other learners; the outcomc of that particular evaluation can lie acted uponimmctliatclj. As Watcrs points out, the negotiation about what is required to act on a taskprovides an actual situation to discuss \vhat is to lie communicated and how it will be done.Participatory evaluation highlights the jobs to lie done in thc ESP classroom and the bestmeans of carrying them out (Watcrs, 1987: pp. 7-8).I.et us now return to thc language program framework in a rc-vised form which betterreflects the itlcas we have forwarded and makcs cxplicit the manner in which the program’saspects operate on each othcr (see Figurc 12.2).V Concluding remarksWhat we have tlcscribcd hcrc is the design of an cmcrgcnt language program, throughoutwhich the learners were invol\ed in defining the content antl how it \vould he addressed.Some will argue that the \vay in which \ye have proceeded here is singular to the situationand not transfcrralile. Certainly, \ve were helpcd liy thc fact that the aquaculture projectstaff \vere already a cohesive team before \ve began working \cith them and that they sharedthe same first language; antl, that the tlvo-week intensive took placc on-site.However, we wish to make explicit certain situational constraints. First, the participantswere at widely different levels of ability in English, from beginners to those who werereading (and writing) research papers in aquaculture. Second, we conducted the programwith a limited amount of media technology two snappy cameras, four portable cassetteplaycrs antl one video camera. Third, work demands meant that some participants werecalled away during the two-wcek intensive.We have de1il)cratcly not dealt in detail with the practical instructional features of theprogram because individual teachers \vould respond to the learners’ work-related contentin thcir own way. An area which could hc developed is team teaching in an emergentprogram.

AN EMERGENT LANGUAGE PROGRAM FRAMEWORI< 147site visit - - - - > planning workshop - - - - > intensive workshop---_- > working on tasksIIIIIIIIIIobservingIIclarifyingandre-definingIIreporting backIIIIIIII . - - - - - - expandingIIIIIIIIIII- - - - - - - - evaluatingassistingunderstandinglearning needsIn concluding, wc cite a particularly challenging passage from Clarkc’s discussion ofthe ncgotiatcd svllabus which sums up the targct \ye \vould likc to sce reached in languageprograms:The syllabus as derivcd from and generated by specific groups oflcarncrs . . . \Till liemeans - rathcr than ends - drivcn insofar as the ends cannot in fact he accurateljprcdictctl. Thc v hole discussion about “design” becorncis somewhat solipsistic whenit is realizcd that the Ncgotiatcd syllabus tlocs not in fact cxist bcforc thc learncrsmeet with the teacher in a particular cn\ ironmcnt in order to cstablish its parametcrs.Dcsign is therefore no longer extcrnal to, or prior to, thc irnplemcntation of thesyllabus and in fact becomes its most essential pedagogical component, Iicing itsclfpart of the learning process.(Clarkc, 1991 : 1’. 14)ReferencesCantllin, C. N. and Murphv, D. F. (1987) I-anguage Learning Tusks. London: Prcnticc-Hall.Clarke, I>. (1991) “The negotiated syllabus: what is it and hohv is it likrly to work?” iipplicdLinpistics 12, 13-28.Corder, S. (1981) Error Anulysis und Intcrlanpqge. Oxford: Oxford University Press.Folcy, J. (1 991 ) “A psycholinguistic framework for task-bascd approaches to languageteaching.” Applied Linpistm 12, 62-75.Hall, I). and Kcnny, R. (I 986) “An approach to a truly comrnunicativc methodology: the AITpre-sessional course.’’ English fir Specrf;c Purposes 7, 19- 32.Ilutchinson, T. and Waters, A. (1987) English,fbr Spectf;c Purposes: rl Lcurning-Centred .4pproach.Cambridge: Cambridgc University Prcss.Jacollson, W. H. (2986) “An assessment of the communication needs of non-native speakers ofEnglish in an undergraduate physics lab.” English for Xpecipc Purposes 5, 173-87.

148 WILLIAM SAVAGE AND GRAEME STORERKenny, B. (1985) “RcvieLv: learning purposc and languagc use.’’ The ESP Journal 4, 171-9.1.arscn-Freeman, D. ant1 Long, M. (1991 ) An introduction to Second Language Acquisition Research.Ncw York: Longman.Mackay, R. (1978) “Itlcntifying the nature of the learner’s necds.” In Mackay, R. andMountford, A. (cds) Englishhfbr Specijic Purposes. London: Longman.Munby, J. (1978) Cornrnunicatir~e Sj/luh~~Design. Camhridgc Univcrsity Prcss.Kea, P. (1987) “Communicati\e curriculum \alidation: a task-based approach.” In Candlin,C. N. and Murphy, D. F. (cds) Language Learning Tasks pp. 147 65. London: Prentice-Hall.Waters, A. (1987) “Participatory coursc evaluation in ESP.” Englirh,fbr Spec!$c Purposer 7, 3-1 2.Widdowson, 11. (1978) Teaching Langiiacqe us Communication. Oxford University Press.Widdow.son, H. (1983) I.eurning Purpose ant/ Languuge Ure. Oxford University Press.

Chapter 13Defeng LiTEACHERS’ PERCEIVED DIFFICULTIES ININTRODUCING THE COMMUNICATIVEAPPROACH IN SOUTH KOREAECENTLY, EDUCATIONAL INNOVATIONS IN L2 education hawR received consitlcrable attention (Bailey, 1992; Freeman and Cazden, 1990; Kcmncd\;1988; Markee, 1997; White, 1987).The literature on this topic includes studies of languagecurriculum dedopmcnt, language tcaching methodology, and the process of innomtionthat occurs in tcachcr tlcvchpmcnt contexts (Bailcy, 1992).Attempts to introtlucc communicative languagc teaching (CLT) into EFL contcxts onEI;L countrics’ ohvn initiativcs and through international aid projects haw prompted man!innovations in 1,2 education. In general, such innovations havc had a low ratc of success(Urindley antl Hood, 1990), and implcmenting CUI’ uw-lchziclc has often pro\.ed difficult(Anderson, 1993; Chick, 1996; Ellis, 1994,1996; Gonzalcx, 1985; Kirkpatrick, 1984; Sano,Takahashi, antl Yoneyama, 1984; Shamin, 1996; l’ing, 1987; Valdcs and Jhoncs, 1991 ).Difficult as it is, many EFI, countrics arc still striving to introducc CI -1‘ in the hopc that itwill improve English teaching there.Why has CLT Ixcn so difficult to implcnicnt in EFI. classrooms? How appropriatc isCUI’ for EFI. contexts? I bclie\c teachcrs’ perceptions of the feasibility ora CLT inno\ationin a particular contrxt are crucial in determining the ultimate su ’ or failure of thatinnovation (Kc,Ily, 1980; Markcc, 1997). For this reason I undcrtook a caw study of SouthKorean secondary school English teachers’ understanding of the uptake of CUf in SouthKorea. As many EFL countries sharc somc of the characteristics of English tcaching inSouth Korea, for cxamplc, traditional tcaching methods antl large classes, this study haswitlcspread implications.CLT: one definitionCLT starts Lvith a throry of language as communication, and its goal is to develop learners’communicative competcnce. Canalc and Swain’s (1 980) definition of communicativccompetence is probahly the bcst kno\z n. The)- identified four dimensions: grammatical,sociolinguistic, discourse, and strategic compctcncc. This definition has untlergonc soincmodifications over thc years, perhaps bcst captured in Bachman’s (1 990) schcmatization of\vhat he calls lungtiage competence. The most significant differcnce l)ct\veen the two moticls

150 DEFENG LIis that Bachman takes a far broader vic.12 of thc role of strategies than Canalc antl S\vain doand separates strategic compctcnce completely from \vhat he calls language competencies(Bachman, 1990; North, 1997).In CLT, meaning is paramount. Wilkins (1 972) classifies meaning into notional andfunctional categories and vicws learning an L2 as acquiring the linguistic means to performdifferent kinds of functions. According to Larsen-l:recman (1 986), thc most obviouscharacteristic of CLT is that “almost everything that is (lone is done lvith a communicativeintent” (p. 132). Teachers select learning activities according to how well they engage thestudents in meaningful and authentic. language use rather than in the merely mcchanicalpracticc of language patterns.Another dimension of CLTF is “its learner-centered antl experience-based view ofsecond language tcaching“ (Richards and Kodgcrs, 1986, p. 69). According to CUI theory,individual learners possess unique interests, styles, necds, and goals that should he reflectedin the design of instructional mcthods (Savignon, 1 99 1 ).CLT is characterized by123456a focus on communicative functions;a focus on meaningful tasks rather than on language per se (e.g., grammar or vocabularystudy) ;efforts to make tasks antl language rclcvant to a target group of learners through ananalysis of genuine, realistic situations;the use of authentic, from-life materials;the use of group acti\-ities; andthe attcmpt to create a secure, nonthreatening atmospherc.I stress that thc description abovc reflects just one definition of CLT, what Holliday (1994)terms the weak version of C1.7. According to Holliday, the strong version is actually quitediffercnt:The focus is not on language practicc but on Icarning ahout how language worksin discourse.The Icsson input is language data in the form of text, and communicative relatesmore to the \lay in \vhich the student communicatcs \vith the text. Also, students collaboratefor the purpose of helping each other solve language problems rather than for the purposeof communicating with each other. Because the aim is not to practice language forms,teachers do not need to monitor group and pair work closely, antl in fact activities do nothave to be carried out in groups or pairs. As long as students are communicating with richtext and producing uscful hypotheses about the language, \vhat they arc doing iscommunicative, according to Holliday (pp. 171- 172).CLT in EFL contextsA number of reports in thc literature deal with CLT innovations in EFL contexts. Whcrcassomc accounts have emphasized the local needs antl the particular English tcachingconditions in the EFL countries and the importance antl success of traditional languageteaching methods (Bhargava, 1986; Sampson, 1984, 1990), others have strongly advocatedthe adoption of CLT in EFL countries (Li, 1984; Prabhu, 1987). However, the majority ofaccounts have recognized the difficulties EFL countries lace in adopting CL,XBurnaby and Sun (1 989) report that teachers in China found it difficult to use CLT.Theconstraints cited include the context of the wider curriculum, traditional teaching methods,class sizes and schedules, resources antl equipment, thc low status of tcachers who teachcommunicative rather than analytical skills, antl English teachers’ drficiencics in oral English

THE COMMUNICATIVE APPROACH IN SOUTH I

152 DEFENG LImagazines, English nchvs on the radio, antl English TV programs. Thc curricula reflect thebelief that “CLT is characterized l~y learner-ccntrcdncss” (p. 1 Sl), antl teachers arccncouraged to organize materials based on students’ needs.Accompanying the release of the nc\v curricula \vas the publication of a series of newtextbooks. Over 10 sets of English textbooks arc now a\-ailahle to secondary school Englishteachers, who arc free to choosc any set provitlctl that the \vholc school adopts it.The newtextbooks incorporate a communicative perspective and more listening and speakingmaterials and activities relative to the older ones.Will the shift in the government’s policy result in an improvement in students’communicative competence? Is Korea prepared to implement CLT in English instruction?To ans\vcr these questions, 1 investigated Korean teachers’ perceptions of the difficulties inusing CLT.DesignThe analvsis consisted of a pilot study, a \vrittcn questionnaire, antl interviews. To developan appropriatc survey instrument for this study, in summer 1994 1 administered a pilotsurvey to 21 South Korcan EFL teachers studying in a teacher education program ata Canadian univcrsity. Thc final qucstionnairc includctl both open-ended questions andquestions with tixcd altcrnativcs gcncratcd from the data collcctcd in the pilot study (seethe Appendix).In summer 1995, the questionnaire \vas administered to 18 South Korean secondaryschool EFL teachers studying at thc same Canadian univcrsity.To ensure that the participantsfully understood the questions, I distributed the questionnaires at the end of a class. Theparticipants were urged to read the clucstionnairc, antl they asked questions for clarification.All 1 8 questionnaires distributed \vcrc handed hack. Following the survey, I conductedin-depth inter\,iews with 10 ofthc participants to explore further the teachers’ back-ground,their understanding of English teaching in South Korea, antl their tlillicultics in using CLT.The interviews \vcrc scmistructurctl, contluctcd in a systematic antl consistent orderbut allowing me as the intervic\z et- sullicicnt frccdom to digress and probe far beyond theannvers to thc prrpared antl stantlartlizctl qucstions (Berg, 1989, p. 17). The interviewsa\varc that the teachers’ imperfect Englishmight limit thc information they provided, I made certain that they u ere ablc to cxprcsstheir itleas fully ti) prcpariiig anti sending a numlxr of questions to them ahratl of time.1. . . Ixvcrc conducted in English. Although I was n ~ l lParticipantsSurv5r partrcipuntsThe participants in the formal qucstionnairc survey \vcrc 18 South Korean.English teachers who \vert‘ studying in the Korcan li~achcr Education Program (KTEP) ata Canadian university in the summer- of 1995. [. . .IThe 9 inale and 9 female participants ranged from 30 to 50 years in age, \vith the majorityin their 30s; thc avcragc age \vas 36.5,’fhcir experience in teaching English varied from 5to 25 years, with an avcragc of over 1 1 ycars. At the time of the study, 8 participants wereteaching in mitltllc schools, and 10 were teaching in high schools. Many had taught at bothmiddle and high schools, as secondary school teachers in South Korea must transfcr schools

THE COMMUNICATIVE APPROACH IN SOUTH KOREA 153every 5 years; high school teachers quite commonly transfcr to middle schools antl \ ice versa.Half of the participants were teaching in rural secondary schools and half in urban settings.A representative 10 ofthe 18 survey participants ww-e also given an in-depth intervicw.Data analysis is not a simple description of the data collected but a proccss liy which therescarchcr can bring interpretation to the data (Povmey and Watts, 1987).Thc themes antlcoding categories in this study emerged from an examination of the data rathcr than kingdetermined beforehand and imposed on the data (I3ogdan antl Biklen, 1992). [. . .]ResultsThe South Korean tcachcrs \vcrc interested in the methods thcy used in teaching English.Fourteen of the 18 participants rcportrd that they were wry concerned, and the other 4reported that they bvere fairly conccrncd. All reported that the grammar-translationmcthod, the audiolingual method, or a combination of the two characterized their teaching.However, 12 rcportcd having tried CI,T before attending thc tcacher education program inCanada and having encountered difficulties in such attempts.The difficulties reported liy the Korean tcachcrs fall into four categories: those caused(a) by the teacher, (b) by the students, (c) liy the educational systcm, and (d) by CI,T itself.Among them, difficulties falling into the first category were mentioned most often, almosttwice or three times as much as those in thc other three categories (seeTablc 13.1).Tuhle 13.1 Krportrd difficulties in implcmcnting CLTSource und (liff;c~ilt,i.TeacherDeficiency in spokcn EnglishDeficiency in stratcgic anti sociolinguistic coinpctcnccLack of training in C1.7’Fc~v opportunitica for rctraining in CLTMisconcrptions about CLTLittlc tinic lor developing materials for communicative classesStudentsI OLV tinglish proficirncyLack of motivation for de\ cloping communicativr coinpctcnccResistance to class participationEducational systcmCLTGrammar-based examinationsInsufficirnt fundingI.ack of supportlnadcquatc account of LFL trachingI ack of rffrctivc and cfticicnt assrssmcnt instruments.Yo. of’rnentionc”99181818161514501817156118181312341816“ The number of timer the rcscarch subjects referred to a thcmc in either the qucstionnairc or theintervie\.\. as a constraint in using the CLT in their onn contcxt. Thr maximum number of mentionspossible for each ofthc themrs included Lvithin thc four major categories is 18.

154 DEFENG LIDificu1tie.s caused by the teucherDeficiency in spoken EnglishAll 18 participants considered that their own deficiency in spoken English constrained themin applying CLT in their classrooms. As rcportccl by thc Korean teachers, the South Koreangovernmcnt wanted CLT implemented I)ecausc of disappointment about students’ oralproficiency in English.The governmcnt as \vel1 as the teachers hopcd that CLT would helpstudents develop lxtter oral English. Although the tcachcrs gcncrally felt that they werehighly proficient in English grammar, rcatling, and writing, they all reported that theirabilities in English speaking and listening were not adequate to conduct the communicativeclasses necessarily involved in CLT. ’l’he following comment was typical.1. I am good at English grammar, reading, and lvriting. Rut my oral English is verypoor. Since 1 can’t speak English \vcll, how can I teach it to my students?(Dong-Soon, July 3 1, 1995)Surprisingly, even respondents who spoke English fluently and communicated wellthought thcir English was “too poor to use communicative language teachings” (Jin-Kyu,July 17, 1995). Deficiency in spoken English apparently prevented some teachers fromapplying CIrr, but for others lack of confidencc was more likely to have been the reason.Deficiency in strategic and sociolinguistic compctcnccAll 1 8 participants reportctl that thcir low strategic and sociolinguistic competence inEnglish would limit their use of CLT. As teachers’ sociolinguistic antl strategic competencemust lie much greater in a communicative classroom than in a traditional grammar-focusedclassroom, the participants gcncrally felt incompetent to conduct a communicative class.2. Studcnts askcd more qucstions in the class. I was happy when they asked mequestions related to the English grammar. Rut those questions that are related tothc sociolinguistic aspccts of English arc really hard for me. . . . In Korea, whenyou can’t answer all of the students’ questions right any, you can’t be a teacher.(Young-Chcol, July 26, 1995)The fear of losing face hccausc of not Iwing ablc to answer students’ questions all thetime discouraged teachers from using CLT.3. I once tried communicative activities with my Grade 10 kids.The kids enjoyed it.In tact I cnjoyctl it too, cxccpt they asked so many questions related to the Englishculture.They kvere interesting questions. Some of them I could answer, and someof them I could not. That made me very much embarrassed. . . . If your kids findthat you cannot always answer their questions very confidently, you arc going tolose thcir respect antl finally lose them. In our culture, teachers are supposed toknow everything and be always correct.(Jin-Kyu, July 17, 1995)

THE COMMUNICATIVE APPROACH IN SOUTH KOREA 155Because of their deficiency in sociolinguistic competence in English and fear of losingthe respect of their students for being unable to give prompt ans\vers in class, teachers“chose to stick to the traditional grammar-centred, text-ccntred and teacher-ccntrctlmcthotls so that [they] always had a good idea about what was going to happen in every classand made adequate preparations for it” (Dong-Soon, July 3 1, 1995).I ack of training in CLTAll 18 participants named lack oftraining as one ofthc main obstacles they faced in applyingCLT. As reported by the teachers, thcy had learned about CLT in different hvays ~ inuniversity methods courses, English teaching confcrcnccs, antl English teaching journals ~but thcy all agreed that they had not practiced it much.4. Like many of us, I Icarncd CLT when I \vas studying at uni\-crsitv. But it \vas taughtas a piece of knowlcdgc for us to remember, not to use. I did not practice using itwhile at university, though I did try it a fe~.timcs latcr when I became a teacher.(f..om-Mi, July 25, 1995)5. I learned the term CLT at a teachers’ conference. To be honcst, I did not quiteunderstand hov it norks.(Mwng Sook, Julv 30, 1995)This lack of systcmatic training led to a sketchy antl usually fragmented understandingof CI,T and made it difficult for the teachers “to leave thc sccurity of the traditional methodsantl take thc risk of trying new unfamiliar methods”(Tack-Soo, July 20, 1995).Fc\v opportunitics for rctraining in CLTSixteen teachers reported that fe\v in-scrvicc opportunities for retraining in CL‘I’ lvereavailable. Most of the respondents hac1 not had such opportunitics t)cfore the teachereducation program thcy \vert attcnding at that time. Mi-Ju cxpressed her frustration whcnasked about her in-service education.6. This is the first time I participatc in an in-service teacher cducation program. Ittook me 18 years to gct such an opportunity.(Mi-Ju, July 28, 1995)Even after the publication of the government’s ncw communicatiw curricula, feiv inservicetcachcr education programs offered training in CLT. Without proper retraining,teachers will inevitably misunderstand some elements of CLT.Misconceptions about CLTFifteen respondcnts referred to teachers’ misconceptions about CLT as one of the principalobstacles. A typical misconception was that by concentrating on appropriateness antl fluency,CLT docs not teach form at all and thus totally neglects accuracy.7. Before attending this tcachcr education program, I thought that communicati1 elanguage tcaching does not teach grammar antl only teaches speaking. I did not

156 DEFENG LIthink that \\as a good ma> to teach our kids Engli5h. I think grammar should bepart of it, at least for our kitis After all, thcy ha\e to pass a lot of exam4 antl therei\ a lot of grammar in them(Myong-Sook, July 30, 1995)Such misunderstantlings led thc teachers to Iiclic\ c that CLT contradicted thcir beliefsabout language learning and did not allon them to prepare students for the harious examsthat arc critical to their future careers. For that reason, the teachers refused to accept CLT.Littlc time tor antl cxpcrtiw in material tlc\clopmentFourteen tcachcrs reported that lack of time for and lack of cxpcrtisc in tlcvelopingcommunicative materials had been constraints for them. All thc English textbooks availalile(before the publication of the ncw series of textbooks accompanying the publication of thecommunicative curricula) had lieen developed under thc influence of‘ the grammartranslationand audiolingual mrthods, so teachers had hatl to write their own materials anddesign thcir oxvn activities if they wanted to use C1.T. [. . .] This prohlcm was particularlyserious for female teachers liecause they also hatl to deal with housework.8. I teach in a high school. I haic to hc at school from 8:00 in the morning to 6:3O inthe afternoon. When I go homc, I hale to take care of my tmo kidr. Because mjhusband teaches ana) from our home in Seoul, I hale to take my kids there at\Teekends to see him. I really do not hale time for an> extra work.(In-Ran, July 24, 1995)Lack of expertise in designing communicati! c actii itics \vas also a concern among thetcachcrs.1. ’ .ILon tngli\h proficiencyAll 18 respondents reported that one important difficulty preventing them from using CLT\vas their students’ low English proficiency. Korean students do not start to learn Englishuntil after thcy cntcr middle school (Gradc 7), antl thcy haw only four 1 -hour English classeseach week, making progress slow.Thcy usually have a small English vocabulary and a limitedcommand of English structures. Recause students did not have the necessary proficiency inEnglish, the teachers found it hard to do any oral communicative activities with them.9. Thc avcragc secondary school students have a very small English vocabulary. Theyknow limited number of English structures. So thcy haw great difficulty to expressthemselves in English when they are assigned to do communicative activities.Gradually they lose interest in trying to speak English and liecome too discouragedto spcak English any morc.(In-Ran, July 24, 1995)

THE COMMUNICATIVE APPROACH IN SOUTH KOREA 157As pointed out earlier, the Korean teachers believed that CLT necessarily involvedspeaking activities. Therefore, Lvhcn oral activities were not possible or appeared to hedifficult, the teachers became frustrated with CLT and in most cases galc it up.10. In such activities, I often see the kids struggling to express themselves in English,only to make each other more confuscd. . . . I do not know whether I am doingthc right thing with thc kids. To be safe, I prcfcr to use the method I am l‘amiliar\vith to help the kids icarn.(Eom-Mi, July 25, 1995)Little motivation for communicativc compctenccSeventeen participants identified students’ lack of motivation to work on their communicativecomprtence as a great limitation. Although an increasing number of pcoplc inSouth Korea have realized how important it is to be able to communicate in English rathcrthan to know English grammar well, students in secondary schools still care much morcabout grammar.1 1 . My students know it is wry important to learn to use English for communication.But since their goal is to enter the university, they prefer to work on Englishgrammar because thc National University Entrance Exam is grammar based.(Joon-Suk, July 26, 1995)Because grammar still plays a dccisive role in all English examinations in South Korca,“tcachcrs who teach communicativc competence arc not liked as well as thosc who tcachgrammar” (Mi-Ju, 28/07/95), Students complained that “they [were] not learning anythingif they [did] not learn nebv words and grammar in a class” (Na-Yun, July 26, 1995).Rcsistancc to class participationFifteen respondents cited the students’ resistancc to class participation as a primaryconstraint in trying CLT. As students have already been in school for at least 6 years by thetime they enter middle school, they haw become accustomed to the traditional classroomstructure, in which thcy sit motionlcss, take notes while the teacher lectures, and speakonly when they arc spoken to. After so many years of schooling in traditional settings,students rely on the teacher to give them information directly, making it very difficult toget thc students to participate in class activities.The inconsistencies among teachcrs in their expcctations of studcnts also discouragedstudents from participating in class activities,12. Especially when English class is thc only place whcrc participation is encouraged,it can bring about confusion for the students as most tcachcrs of other subjects willprobably ncvcr toleratc, not saying encourage class participation.(Jin-Kyu, July 17, 1995)To play it safe, students usually chose to behave traditionally in English class. When studentswere not willing to participate in class activities, teachers saw little chance of fulfilling theirgoal of using CLT, rendering it pointless to adopt CLT in their class.

158 DEFENG LID@culties caused ty the educational ysternLargc classcsAll 18 respondents referred to large classes as one of the principal constraints on theirattempts to use C1.T. In South Korea, a sccontlary school class usually contains 48-50students.The teachers found it very difficult, if not entirely impossible, to use CLT with somany students in one class hecause they Iielieved that oral English and close monitoring ofclass activities Lvere essential in CLT.13. With that number of students in one class, first of all, it is very difficult for classmanagement if UT use the communicativc method. For example, when everyonestarts to talk, the class can be very noisy.Tcachers and students in nearby classroomswill complain ahout the noise in the English class. Secondly, it is not possible forthe tcachcr to give cach of them [individualized] attention as required by thecommunicativc method. Thirdly, \vith so many students in one regular classroom,there is not even enough space for the students and the teacher to move around tocarry out the communicativc activities. Especially when the desks and stools arcfixed to the floor, you cannot even move them.(Jin-Kvu, July 17, 1995)Grammar-basctl cxaminationsGrammar-based examinations were namctl Iiy all 18 rcspondents as another importantconstraint. Among the many English cxaminations in South Korea, the National UniversityEntrance Examination (the English section) is thc most important one because other formaland informal English cxaminations are motlcled on it. Until 1994 it consisted mainly ofgrammar, rcading comprehension, and translation items. Now it has an additional part called“Listening Comprehension,” Iiut its grammar-hased nature has remained unchangcd.Teachers, under pressure to make their students do well on such tests, often devote valuableclass time to teaching test-taking skills and drilling students on multiple-choice grammaritems.14. This exam [the National University Entrance Examination] has had tremendousinfluence on the English teaching in South Korea. As soon as students start middleschool, they have a clear goal in mind - to pass the National University EntranceExamination.Teachcrs also have a clear goal in mind - to help students succeed inthe Examination. Because it only tests students’ grammar knowledge and readingability, both students and tcachcrs are interested in grammar and rcading in Englishclasses.(Young-Cheol, July 26, 1995)Such an attitude leaves little room for CLT for 110th tcachcrs and students. As Savignon(1991) observes, many curricular innovations have been undone by a failure to makecorresponcling changes in evaluation.

THE COMMUNICATIVE APPROACH IN SOUTH I

160 DEFENG LICLT’s inadcquatc account of EFL teachingAll 18 participants reported that CLT has not given an adcquatc account of EFL tcachingdcspitc its initial growth in foreign languagc teaching in Europc.The teachers saw importantdifferences bctwccn teaching EFL and teaching ESL. Thcy expressed frustration at the factthat the research community, cspccially many Western language education rcscarchers, hasrarelv differentiated EFL from ESL.20. In my opinion, EFL is very tlilfci-ent from ESL. But many people tend toconfuse them and often ignorc the special clcmcnts of EFL. situations. I think that’swhy wc EFL teachers usually find Western language teaching methods difficultto use.(Joon-Suk, July 26, 1995)The significant differences that thc teachers saw bctwccn EFL and ESL included thepurposes of learning English, learning environments, teachers’ English proficiency, and thea\-ailability of authentic English materials.[. . .I2 1 . For cxamplc, in ESL situations, studcnts usually haw a very supportive learningenvironment outside school. Thcy have many chances to hear and speak Englishoutside class, which can reinforce VI hat they learn in class. Resides, they havethe motivation to work on oral English txcausc they need it in their lives. Inour situation, the classroom is the only place hvhcrc studcnts can hear and speakEnglish.Thcy do not nccd to usc the language in their livcs but only in pretendedsituations.(Jin-Kyu, July 17, 1995)Lack of effective and cllicicnt assessment instrunicntsUsed to traditional discrete-point tcsting of grammatical knowledge, the teachers foundit disconcerting that there \vcrc no prescrilied, ready-made assessment tools for communicativecompetence and that they would have to design their own. The Korean teachersbelieved that one ofthe liest \vays to test students’ communicative competence was to givethe studcnts oral tests. In general, they each taught four classes of approximately 48 students.Finishing even one round of individual oral tests would take a long time, and there wasnobody to supervise the other students while the teacher was conducting the tests.22. When you teach four classes and each has nearly 50 students, you are dealing with200 students. If I have to do oral examinations to assess their communicativecompetence, it would take me dozcns of days to finish just one round.(Mi-ju, July 28, 1995)Resides, the Korean teachers generally did not support these sulijectivc tests.

THE COMMUNICATIVE APPROACH IN SOUTH I

THE COMMUNICATIVE APPROACH IN SOUTH I

164 DEFENG LICanale, M. and Swain, M. (1980). “Theoretical basis of communicative approaches to secondlanguage learning antl testing.’’ Applied Linguistics, 1 , 1 47.Chau, L. antl Chung, C. (1987). “Diploma in education graduates’ attitude towardcommunicative languagc teaching.” Chinese Universiy Education journal, 15(2), 45 5 1 .Chick, J. K. (1996).“Sat‘-talk; Collusion in apartheid etiucation.”In H. Coleman (Ed.), Societyand the language classroom (pp. 21 ~ 39). Camlx-itlge: Cambridge University Press.Daoud, M. (1996). “English language development inTunisia.” TESOI Quarterly, 30, 598-605.Deckert, G. (1987). “The communicative approach: Helping students adjust.” English TeachingForum, 25(3), 17-20.Development Committee of the Sixth Curriculum for High School English. (1 992). The reporton the revision ofthe English curriculum /or high school. Seoul, Korea: Author.Edge, J. (1996). “Cross-cultural paradoxes in a profession of values.’’ TESOI. Quarterly, 30,9-30.Ellis, G. (1994). “The appropriateness of the communicative approach in Vietnam: Aninterview study in intcrcultural communication.” Unpublished master’s thesis, La TrobeUniversity, Bundoora, Australia.Ellis, G. (1996).“HOWculturally appropriate is the communicative approach?” ELTjournal, 50,2 13-218.Enright, I). S. and McCloskcy, M. L. (1985). “Yes, talking! Organizing the classroom topromote second language acquisition.” TESOI Qiarterly, 19, 431453.Fotos, S. S. (1994).“Integrating grammar instruction and communicative language use throughgrammar consciousncss~raising tasks.” TESOL Quarterk, 28, 323 35 1.Freeman, D. antl Cazdcn, C. €3. (1990). “Learning to talk like a profcssional: Somc pragmaticsof foreign languagc teacher training.” Prugmcitics and Lungtiage Learning, 2, 225-245.Frymier, J. (1987). “Bureaucracy and the neutering of teachers.” Phi Delta Kappan, 69, 9-1 4.Fullan, M. (1993). Change forces: Probing rhc depths ?f etiucational yform. London: FalmerPress.Goetz, J. P. antl LcCompte, M. D. (1984). Ethnography and qualitatire design in educationalresearch. NelvYork: Academic Press.Gonzalcz, A. (1985). “Communicative language teaching in the rural areas: How docs onemake the irrelevant relevant?” In R. K Llas (Ed.), Communicative lanpap teaching (pp.84-1 05). Singapore: Singaporc University Prcss.Harvey, P. (1985). A lesson to be learncd: Chinese approaches to language learning. ELTjournal, 39, 183 186.Holliday, A. (1994). Appropriate methodology and social conlest. New York: Cambridge UniversityPress.Kclly, P. (1980). “From innovation to adaptability: The changing perspective of curriculumdevelopment.” In M. Galton (Ed.), Curriculum change:The Ic.s.sons .fa decade (pp. 65-80).Leicester, England: Leicester Univcrsity Press.Kennedy, C. (1988). “Evaluation of thc management of change in ELT projects.” AppliedLinguistics, 9, 329-342.Kirkpatrick,T. A. (1984).“The role of communicative language teaching in sccondary schools:With special reference to teaching in Singapore.” In R. K. Das (Ed.), Communicativelanguage teaching (pp. 171 ~ 191 ). Singapore: Singapore University Press.Larsen-Freeman, D. (1986). Techniyries and principles in language teaching. New York: OxfordUnivcrsity Press.Lee, C. (1990). “Korean high school seniors’ oral antl literate comprehension and productionskills in English.” Unpublished doctoral dissertation, Harvard University, Boston.Li, X. J. (1984).“In defense ofthe communicative appruach.”EL?‘/ournal, 38, 2-1 3.North, €3. (1997).“Perspectives on language proficiency and aspects of competence.” LanguageZaching, 30(2), 93-100.

THE COMMUNICATIVE APPROACH IN SOUTH KOREA 165Penncr, J. (1995). “Changc antl conflict: Introduction of thc communicative approach inChina.” TESL Canada Journal, 12(2), 1- 17.Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford Univcrsity Press.Powney, 1. and Watts, M. (1987). Interviewing in educational re.search. 1.ondon: Routledge.Prabhu, N. S. (1987). Second language pedagogj. Oxford: Oxford University Press.Price, R. F. (1988). “The politics of contemporary cducational rcform in China.” In E. B.Gumbcrt (Ed.), Making the Juturc: Politics and educational rfform in the IJnited States,England, the Soviet Union, China, and Cuba (pp. 99-1 14). Atlanta: Georgia StateUniversity.Richards, J. C. and Lockhart, C. (1994). Rejlective teaching in second languugc classrooms.Cambridge: Cambridge University Press.Sampson, G. P. (1984). “Exporting language teaching methods from Canada to China.” TESLCanadojournal, 1 (l), 19-32.Sampson, G. P. (1990). “Teaching English literacy using Chinese strategies.” TESL Talk, 20( l),126-1 38.Sano, M., Takahashi, M. and Yoneyama, A. (1 984). “Communicative language teaching andlocal needs.” ELTjournal, 38, 170-1 77.Savignon, S. (1 991 ).“Communicativc language teaching: State of the art.” TESOL Quarter!,,, 25,26 1-277.Savignon, S. and Rcrns, M. (1984). Initiatives in commtinicarive language teaching: A hookreadings. Reading, MA: Addison-Wesley.Schiin, D. A. (1 983). The rejlective practitioner: How. pr$ssionals think in action. London: TempleSmith.Shamin, F. (1996). “Learner resistance to innovation in classroom methodology.’’ In H.Coleman (Ed.), Socict,~ and the language classroom (pp. 105 121). Cambridge: CambridgeUnivcrsity Press.South Korean Ministry of Education. (1 992a). The sixth curriculum.fbr high .schools. Seoul, Korea:Author.South Korran Ministry of Education. (1992b). The sixth curriculum,fir middle schoo1.s. Seoul,Korea: Author.lilleman, H. H. (1994). Training and professional expertise: Bridging the gap bctween ne\+information and pre-existing bzliefs of teachcrs. Euching and ‘leacher Educution, 10,601 615.Ting,Y. K. (1987). “Forcign language teaching in China: I’roblcms antl perspectives.” Canadianand International Education, 16, 48--61.Tomlinson, B. (1990). “Managing changc in Indonesian high schools.” ELT]ournal, 44, 25-37.Valdes, A. I. and Jhones, A. C. (1991). “Introduction of communicative language tcaching intourism in Cuba.” TESL CunudaJournal, 8(2), 57 63.White, R.V. (1987). “Managing innovation.”ELTlotirnal, 41, 21 1-218.Wilkins, D. A. (1 972). The linguistic and situational content ?f the common core in a mit/credit.y.~tcm. Strasbourg, France: Council of Europe.

Appendix: questionnairePlease complete the following question\ as appropriate.1 Age2 sex3456How many years have you lieen a tcachcr of English?Are you teaching in a middle school or high school?0 Middle School 0 High SchoolWhich grade(s) are you tcaching?Are you teaching in an urban or rural middlc/high school?0 Urban Rural789101112Are you concerned about the methods you use in tcaching English?DYES0 N OWhat methods are you using now?Haw you tried Communicative Language Teaching (CLT)?DYES0 NOWhy did you or why didn’t you try CLT?Hoiv did you like using CLT in your classroom?The following are some difficulties that other EFL teachers had in adopting CLT.Did you comc across thcsc difficultics or do vou think thcv might lie difficulties foryou in adopting CUI’ in South Korea?123456789Teachers’ deficiency in spoken English?Teachcrs’ deficiency in strategic and sociolinguisticcompetence in English?Tcachers’ having little time to hvritc communicativematerials?Students’ low English proficiency?Students’ passive style of learning?Lack of authentic tcaching matcrials!Grammar-hased examinations!Large classes?‘l’he differences hctwccn EFL and ESL?UYES O N 0

PART THREEPlanning and implementingcurriculum change

Chaptev 14Adrian HollidayACHIEVING CULTURAL CONTINUITY INCURRICULUM INNOVATIONIntroductionN THIS PAPER I AM GOING TO LOOK AT thc issue ofcultural continuitvI in curriculum innovation. My major point will be that a major obstacle to true culturalcontinuity is our own professional discour.which prevent us from seeing the real worldsof the people we work with. We therefore need to be critically aware of ourselves as culturalactors antl learn how to see the people wc work with in their own terms instead of in ourterms.I shall begin with the principle of cultural continuity and why it is important both inthe classroom and the wider domain ofthc curriculum and curriculum projects. I shall thcndemonstrate how professional discourses create obstacles to cultural continuity, and howthis might tie avoided.The principle of cultural continuityCultural continuity is achieved when meaningful liridges are built bctwcen the culture of‘the innovation and the traditional expectations of the people with whom w-c work. Thenotion of ‘cultural continuity’ is taken from Jacob (1996), who is interested in the way inwhich the teacher mediatcs hcthveen a ‘foreign’ lesson content and the ‘local’ orientationof her or his students. Homever, it can be used to refer to a broader aim which has becomecommon inTESOL in the last ten years ~ to be sensitive to the cultural expectations of the‘rccipients’ of innmation, whcther they bc students or tcachcrs encountcring new tcachingmethodologies, or stakeholders in curriculum projects. Phillipson (1992) antl Pennycook(1994) havc drawn our attcntion to the dangers of cultural or linguistic imperialism \\hendominant forms of professionalism in TESOL arc transported from one place to another,as has my own work on how to make classroom and curriculum project methodologies‘appropriate’ to social context (Holliday, 1994). The now influential phrase, ‘appropriatemethodology’ was introduced into TESOL by Rowers many years ago (Rowers andWiddowson, 1986).The plea for more attcntion to the sociopolitical environment of TESOL,w-as made by Swales several years earlier (1980). Coleman’s (1996) work on the influenccof socicty on what happens in the classroom is a more recent part of this movement, as isrecent critical thinking about how the paradigms of TESOL profcssionalism haw been

170 ADRIAN HOLLIDAYsocially constructed (Beaumont and Wright, 1998). The basic idca of cultural continuity isthat a particular innovation is adjusted to enable the hcst possiblc fit with a host environment(Figure 14.1). It is a two-way process in that the innovation needs to be informed by datafrom the host environment.INNOVATIONattcmp to fit ho\t ~tiltiircc.g. needs muL\ms, act ion rcsenrch. cthnogruphj,inioli.ing ‘insiders’, CI alriulionf 10ST ENVIRONMENT(classroom or institution; stucicnts,tcachcrs or othcr stakeholdrrs)Figure 14. / Cultural continuit!The dominant discourses of teachingThere is a strange iron); here, which involves the prolhmatic nature of learner-ccntredncssand skills-based education. At first sight, lrarncr-ccntredncss and the teaching of skills wouldscem to support the possibility of cultural continuity in that they follow the principle thattcaching should connect with the perceptions antl needs of the student. Indeed, lcarnercentredncssreprcsents an admirable attcmpt in cducation, sincc the 1960s, to allow studentsa more interactive, participant role in the classroom. At thr same time the teaching andlearning of skills implied that the content of education had to be useful to the needs of thelearner and the environment in which she or he \vas to operate.However, with deeper analysis, various kvritcrs in education, such as Usher and Edwards(1994), following thc critical sociology of Foucault, are Ileginning to argue that Iearncrcentredncssand skills-based education might lie having the opposite effect. The 1970s and1980s brought an increased need for accountability; and a skills-based education lent itselfwell to the measurement of stutlcnt progress through the achievement of discrete learningobjectives. The breaking down of skills into compctcncies was instrumental in this. Theoutcome is a bureaucratisation of lcarner~ccntredncss. Usher and Edwards (ibid.) argue thatin swing the student in terms of a set of pre-defined, mcasurablc compctcncies anti skills,she or he is reduccd to a learning automaton. Thus, the ‘learner’ at the centre of learncrcentrednessis no longer a real person, but a product of mcasurablc educational technology.T\vo things are going on here. First, what claims to bc a sensitivity to the ‘learner’ ~learner-centredness ~ has liecome a lircaking up of the student into teachable skills. Second,the terminology with which education speaks about the ‘learner’ has become highlytechnical. Hence, learner-ccntredncss becomes what Fairclough (1995) calls a‘technologiscd discoursc’ which appears ideologically neutral but in fact reprcsents thebureaucratic and idcological needs, not of thc ‘learnrr’, but of a particular professionalgroup. Clark antl Ivanic assert that: ‘“Skills” [. . .] suggests a set of neutral technologies ortechniques that are somehow separate and separable from the social context. [. . .] It hasled to the vieiving of language and language activities as consisting of discrete, apparently

ACHIEVING CU LTU RAL CONTINUITY 171manageable and “teachable” components, and so appears to facilitate teaching and learning.It implies a normative and prescriptive view of communication’ (1997: 84).This perception is bcing confirmcd in research into the ideological basis for TESOLprofessionalism . Raxter (in process) has noted that in teacher training programmes, despitethe liberal rhetoric of learner-centredness, the real concern is with the technology ofteaching, which is presentcd as ideologically neutral, in which the ‘learner’ becomes anaccessory ~ for the purpose of accountablc professionalism.concern IZ ith learning need5 ant1 ob~ecti~cs,skill\, cornpetencec and uccountabili~~someit here elsetcchnologisrtlprofcssional discourscFigure 14.2 Profcssionally constructrtl imagc of ‘the learner’Hence, although we might claim learner-ccntrcdness, we construct an image of ‘thelearner’ within our own powerful, technical discourse of professionalism.This is illustratedin Figure 14.2. If the innovation is a new classroom rncthodology claiming ‘learnerccntrcdness’, the surface implication is that there will therefore be a concern with thc needsof ‘the learner’. However, the tcchnologised professional discourse of learner-centredncsstakes attention away from the real student. Learning needs and objectives, skills andcompetencies serve the accountability required by the discourse rather than the real student.The outcome is a control of ‘learning’ through planned tasks, again, serving thc technicalneeds of the discourse rather than the rcal student.The dominant discourse of project managementI will now move on to the morc macro issue of curriculum innovation and arguc that asimilar process is taking placc. Thc problem here concerns the way in which the so-calledrecipients of curriculum innovation are perceived, accommodated and managed.There has been much rcccnt concern that curriculum innovation should hc scnsitive tothe local setting. This has resulted in what has been called a more person-sensitive processapproach to curriculum project management. The process project claims adaptation tosituational needs. And in what might be called a stakeholder-centred approach, groups oflocal people arc quite rightly identified as representatives of these needs; and strategiesarc dcveloped to satisfy their intcrcsts and maintain their ownership of thc innovation.Stakeholders can be broadly defined as all the people who have a stake in the innovation.Several examples of this can be seen in Hayes (1 99711). In projects in Indonesia andThailand,

172 ADRIAN HOLLIDAYAmlirosc-Yeoh reports how eighty-scvcn secondary school teachers arr consultcd ‘in afcasibility study’, antl in the resulting training:A generally friendly and interactive style \vas adoptcd to counter any sense of isolation.[. . .] To pcrsonalise the materials and to cstalilish rapport ivith the teachcrs, passivelanguage was generally avoided and there was also deliberate choice of pronouns suchas ‘we’, ‘1’ antl ‘you’ over pronouns such as ‘they’ or ‘he’ or ‘shc’.(1 997: 89-90)In Malta, Jarvis and Cameron (1 997) monitor the changing roles of teachcrs as they adoptand interpret innovation. Also, Martin antl Ralahanis (1995) describe how in Egypt, ‘workingparties’ are set up to involve senior rqx-escntativcs from USAID, the Ministry of Educationand the language centre where the innovation \vas to take place, and ncgotiate conscnsus.Similarly, Weir and Roberts (1994) tlcscribc ho\v ‘insitlcrs’ Iiecomc involved in the evaluationof the innovation process, in, for cxamplc, the cstahlishmcnt of ‘liasclinc’ data, and howformative evaluation liecomcs integratcd with self-directed tcachrr development.er a problcm \vith this stakeholder-centred approach, similar to thcarncr-centred approach which I have already described. As with theclassroom, there is a strange irony. As bvith learner-centrcdness, a tcchnologised professionaldiscourse has bccn created. Weir and Roberts (1994) rightly note that as the concept offormative evaluation inTESOL maturcs, it takes on the role of quality control. Indeed, itfalls in line with the growing dominant ideology of late modern society in which everythinghas to be accountable to the client. Even the pro project has to lie commodified alongwith the other aspects of cducation ant1 other institutional practices such as medicine noted117 Usher and Edlvards (1994) and Fairclough (1995). Thus, we have a professionallyconstructed image ofthc ‘stakeholder’, as lvc do of the ‘learner’(Figure 14.3).As with the technologised discourse of learncr~ccntredness, the technologiscd discourscof stakeholdcr~centredness has an emphasis on control (right hand bubble). Here the controlis situated in a prolifcration of highly technical project documrnts, at the centre of whichare the current log-frame and time-lines for resource input. Although these documcnts are,quite rightly, intended as the product of ‘agreement’ with key stakeholders, thcy are verync\v curriculumIitcmattempts to bc ‘stakcholtlcr~ccntretl’~’concern with institutional needs, oiimmhip,management ckr /Is undo car in tu hi /i

ACHIEVING CULTURAL CONTINUITY 173much the product of the technologised discourse itself. This is wry clear whcn stakeholderswho do not belong to this discourse find them incomprehensible.In my experience, therc is an interesting puzzle here. Thrrc is often the appearance of’n the various parties with regard to project documentation. Thehyperrational project discourse takes this as evidence that there is real agreement.Stakeholders coming from different discourses of their own might see the situationdifferently, and be prepared to go through the motions of agreement for their o\vn ulteriormotives.This type of phenomenon is \vel1 documented in Holliday (1 994), where I describethe operation of informal ordcrs and deep action within the host environment, which havestakeholders pursuing their o\vn ends in their own ways. The following hvarning by Tayloragainst naive notions of mutuality rings true:Establishment and maintenance of sustained dialogue between all those involved[. . .] [is] not casy [. . .]There arc barriers antl inequalities of language, culture andeducation, and frequently even the need for such dialogue is not recogniscd by eithersidc. There is unfamiliarity on both sides with the use of common management toolsfor the sharing and analysis of information (from project frameworks, analysis, tables,grids and diagrams to statistical methods and computrr programmes).(1997: 116-17)Something similar was found by Smith (1 991 ), who noted that notions of ‘control’,‘predictability’,‘generalisation’ and ‘objective’ \vert constructed differently by differentparties in educational innontion in a numhcr of scenarios.An important extension of the discourse of stakchO1dc.r-centretlness is the equallytechnologiscd discourse of evaluation, \vhich bases itself very much on the carefullymeasurcd and verified consensus of stakeholders. If this consensus is only apparent, thenthe technology of evaluation cannot be as sound as it appears. According to Fairclough(1995), a political, though tacit antl pcrhaps unconscious motivc, of tcchnologised discoursesin late modern societies is to create a false image of consensus as \vc find ourselves graduallyconsumed by thc hchavioural technicalities which they demand. As we try to get our hcadsaround the discourse of quality control \ve find ourselves more and more taking part in it,eslxcially as the tliscoursc takes on the appearance of’lnr’ltlng us to participate in our o\vncvay. I do not somehow think that the local participants in many curriculum projects aretaken in in this way. They haw other tliscour of their o\vn to get on with.Empowerment and ownershipThis statc of affairs throws an intcrcsting light on the tray in which the behaliour ofstakeholders is perccivcd and constructed by the tcchnologised discourse of stakcholdcrccntrctlness.In much of the literature on stakeholtlrrs, there is a tacit polver distinctionn those parties who somehow instigate, manage, fund, design, and possess thetechnology of innovation, antl those who do not. InTESOL projects this distinction can beexpressed easily in terms of expatriate, ‘nati pcakcr’ ‘experts’ on the one hand, and ‘local’personnel on the other. This also corresponds \vith the ‘insider-outsider’ distinction, wherecxpatriatc agencies antl pel-sonncl comc from outside in every sensr of the kvord, antlinsiders arr local not onl! to the innovation scenario but to thc country within which ittakes place.It is important to stress that I am talking here about perccptlons created within thediscotirse of a particular innovation methodology, which constructs the reality of inno\-ation

174 ADRIAN HOLLIDAYscenarios in a particular way, rather than the realiv. It is also important to stress that thewriters of literature within the discourse might themselves he unaware of the itlcologicalprinciples they are perpetuating. Fairclough (1995: 36) makes the point that people arcoften ‘standardly unaware’ of the itlcological meanings which have hecome normalisedwithin their own language. Clark antl Ivanic (1997: 176) confirm this lvhcn they cite a studywhich shows that many people are often not aware of the tlccpcr ideological meanings ofwhat they rcad.Thus, it is the discourse, rathcr than individual actors within it, which revealsan ‘us’-‘them’ distinction found in the litvraturc.Clark and Ivanic (1997) make the point that the act of writing is itsclfa struggle withina world where competing discourses vie for hegemony. Such a struggle can be seen in theway in wrhich Smith (1995) lvritcs about a kcy stakeholdcr group Lvhich falls into the localinsidercategory of ‘counterparts’ the people who work alongside ‘expert’ cxpatriatccurriculum developers hvith whom thcre is somc form of transfer to enable the innovationto continue after the ‘expert’ has left.Smith suggests that it cannot lie denied that there may lie a pohver diftercnce in manydeveloping world locations, \vhen thc expatriatc ‘expert’ has lhe ‘privileges [. . .] grantedto (or assumed by) the foreign gucst’ which enable access to budgets, key locations, eventsand people, and the counterpart docs not, and is then expected ‘to sustain project impactafter the aid has been withdrawm’ on ‘US$25 pcr month’ (1995: 67- 8). Discussion ofwhether or not this is altvays thc case involvcs looking more deeply at the whole rclationshipbetween insiders and outsiders; but here one can suspect that thc problem might not somuch be one of power per se, but of the nature of the technology which the counterpart isexpected to carry on. Might it bc that what thc ‘expert’ is considered to be expert in is notsufficiently compatible, or too ethnocentric to thc discourse ofscukeholder~cencredness from theoutset? Smith acknowledges that a morc ‘humanistic approach’ to project ‘sustainability’must get ‘closer to the ways of the rccipicnt’ antl that the po\ver required to sustain theinnovation may not be something thc counterpart simply docs not have, hut somethingwhich she or he might ‘refuse to accept’ (1995: 67). Here, as in so much of this literature,there is a concerted e@rt to get to grips with and understand the viewpoint of the ‘local’,but the outcome, the insistence that ‘empowerment’ of the ‘local’ is the answer, is still deeplyrooted in the ‘us’-‘them’ perception, in which ‘they’‘don’t know the technology’ antl are‘easily dominated’ .Although analysts do try to get under the surfacc at the deeper social issues, and reallydo try to understand the viewpoints antl predicament of other partics in innoyation contexts,they tend to consider large cultural factors as the overriding issuc. Hence, Smith puts‘cultural’ at the top of his list of ‘ohstacles’ to empowerment. Speaking about Cambodiahe suggests that local personncl:will have to push hard to bring about any changcs.This will tie difficult where culturallyone defers to and is not assertive towards someone higher in the hierarchy. [. . .]Othcrs have noted the ‘cultural nature of management’ [. . .I antl the ‘differing culturalconcepts as to the appropriate roles for professionals employed in the public sector’.(1995: 71)I IC continues to state the ‘need for a thorough understanding ly outsiders ofthe host cultureinto M-hich the innovation is king introduced’ (ibid.: 74 citing Leach). He thus alludes tothe model of cultural thinking seen in Hoftstetle, who looks at ‘the consequenccs of nationalcultural differences in the way people in a country organisc themselves’ and how ‘organisationalpractices and theorics arc culturally dependent’ (1 991 : xiii).

ACHIEVING CU LTU RAL CONTIN U I TY 175The rational, systematic nature ofthis national culture model fits \vel1 with the technicalneeds of the discourse of stakcholtler-ccntretlness, as it does with many activitics, such asmanagement, which seek to commodify human difference efficiently. Follohving this line ofthinking, Flew sees ‘counterpart training’ as essentially an ‘interpersonal interaction acrosscultures’. She quite rightly shrinks from the perception of a one-kvay transfer from culturallysuperior rxpatriate curriculum developer to culturally inferior countcrparts as ‘potentiallypatronising’ (1 995: 76) and recommends ‘mutual learning between people from tliffercntcultures (1 995: 81). One Lvondcrs, however, whethcr ‘trust and esteem (1995: 78) \vi11 hesufficient to lircak the ‘us’-‘them’ paradigm and stand in the Lvay of a potentially damagingof mutual othcrisation. On the onc hand, one would not nowadaysrecommend a professional exchange of virws on the hasis of a sharing of gender or racialdifference. On thc other hand, the headings ‘training’ and ‘empowerment’ under which theexchange takes place seem to indicate the ideology of only one side.Overall, the literature on stakrholders seems to crcate the ‘us’-‘them’ distinction in avery particular way (Figure 14.4). On the one hand, ‘they’ are deficient, mainly in tcrmsofthe technologised discourse itself; on the other hand, they arc classified as such vcry muchin the same way, perhaps regartllcss of their so-called national culture. Onr implication hereis that the major agent of difference is not the national culture at all, but the power of thetcchnologiscd discourse. A colleague of mine in a project in India commented that theproject created thc notion of ‘all Indians together’. Perhaps it is not just Indians, but anyonewho docs not conform to thc discoursc. Again, an important implication here is what doesit all mean if the ‘Indians’ do not really want to conform to thc project after all?S’I‘A K E H 0 L1)k J

176 ADRIAN HOLLIDAYOne teacher recorded hcr experience. ‘When you spcak English everybody will (sayto you) “What language you do?” Other teachers (will say) “You arc strange . . . youtry to show off like this” ’. [. . .] It is in relation to their position in society, the cultureand traditions of thcir schools and accepted norms of behaviour within theirclassrooms that teachers in Thailand have to ‘re-interprct (INSET activities) in theirown tcrms’.(Hayes, 1997a: 80)Similarly, Barmada, revisiting the curriculum project at Damascus University in which I wasinvolved in the carly 1980s, reveals an insight unnoticed by me in five years of projectmotivatedinvestigation:But sometimes I feel as if1 represent the West in the classroom and as if1 were tellingmy students that our methods of learning and thinking are not good and should liereplaced by those of the West I. . .] ‘unpaid soldiers of the West’. This made my [sic]very nervous. I should pay attention to what I say in the classroom.(1994: 175)Understanding ourselvesSomething else ~ ve need to do is to liccomc aware ofthc fact that what we do as profossionalsis not ideologically ncutral, but that it is part of a pomuful, dangerous, ideologicaltechnologised discourse. We must come to tcrms mith the fact that our discourse makes ussee othcrs in our own terms, antl not in thcirs. We must not be naive to assume thattechnologies of investigation, cvaluation, quality control and management created withinour own discourses arc equally meaningful to other people. Wv must come to trrms withthe fact that thc l,i-idgcs \ve build to rcach other cultures might only Iic meaningful to ourculturc. The concepts of Icarncr--ccntre(lncss and stakeholtler-centi-edness arc products ofour own discourses, and may not ticlong to the differently constructed worlds of those wewish to reach. We thus need to look tlccply and critically at our ohvn discourses beforcjudging those of others.BibliographyAmbrose-Yeoh, A. ( 1997) ‘LXstance education and in-scrvice language teachcr devclopment’,in Hayes, D. (et].) 86 89.Barmada, W. (1994) ‘Ikveloping an institutional self-cvaluation scheme in an ESP Centre inthe Arab world: rationale, experimentation and evaluation’ , unpublished PhL) thesis,Department of Linguistics, University of Lectls.Baxter, A. (In process) ‘The reproduction of professional culture through teachcr education forELT’, unpublished paper, Department of I.anguagc Studies, Canterbury Christ ChurchUniversity Collcgc,.Beaumont, M. antl Wright,T. (1998) ‘ELT and paradigm shifts: in from the cold or out on alimb’, unpul,lishcd paper prcscntctl at thc IATEFL conference, Manchester.Bolvers, R. and Widdo\vson, H. (1 986) ‘A tlcbatc on appropriate methodology’ in Alhott, G.antl Beaumont, M. (ctls) The Jcrrlopmenr ELT. the Dunf;ird Seniinur.~ I978 I993, ELTKcviebv, Hcmel I Icmpsteatl: Prcntice Hall antl thc British Council 141-5.Clark, R. and Ivanic, R. (1 997) The pollrics ofwriting London: Koutlcdgc.Coleman, H. (etl.) ( 1996) Sociey md the Ionpage classroom Cambridge: Cambridge UniversityPress.

ACHIEVING CULTURAL CONTINUITY 177Crooks, T. and Crcwes, G. (cds) (1995) Language and deidoprnent Bali: 1AI.F.Fairclough, N. (1995) Critical discourse anab.sis: the critical stu+ ?f language London: AtltlisonWesley Longman.Flclv, A. (1995) ‘Counterpart training and sustainability: effecting an exchange of skills’ inCrooks,T. and Crewcs, G. (eds) 76-82.Hayes, D. (1997a) ‘Articulating the context’ in Hayes (ed.) 74-85.IHayrs, D. (ctl.) (199713) In-rervice teacher development: international perspectives ELT RevielzLondon: Prenticc Hall.Hofstctle, G. (1991) Cultures and organisatlons: s.ftivare oj‘thc mind Maidenhead: McGraw-Hill.Holliday, A. R. (1994) Appropriate methodology und social conrext Cambridge: CambridgeUniversity Prcss.Jacob, G. (1996) ‘The CDS co-ordinator’, unpublished paper, Department of English,University of Punc, India.Jarvis, J. and Camcron, L. (1997) ‘Kole shifting in INSET: an exploration of a primary Englishproject’ in Hayes, D. (cd.) 37 49.Martin, W. M. and Balabanis, L. P. (1995) ‘Team dcvclopment in ELI- projects: a caxe study’ inCrooks,T. antl Crewes, G. (ctls) 16 30.Pennycook, A. (1994) The cultural politics of English as an international language London: AddisonWesley Longman.Phillipson, R. (1992) Linguistic imperialism Oxford: Oxford University Press.Smith, H. (1995) ‘Power antl sustainability in language-related development projects’ inCrooks,T. and Crcwes, G. (cds) 65-75.Smith, N. L. (1991) ‘Evaluation rcflections: the context of investigations in cross-culturalevaluations’ in Studies in Educational Evaluation, 17, 3-2 1 .Swales, J. (1980) ‘The educational environment and its relevance to ESP programmc tlcsign’ inProjects in Materials Design, ELT Documents Special. London:The British Council, 6 1 70.Taylor, G. (1997) ‘Management issues in INSET: a practical perspective’ in Hayes, D. (cd.)116-1 27.Usher, R. and Edwards, R. (1994) PostmoJernism and etlucation: djifcrent voices, djffcrent worldsLondon: Routledge.Weir, C. J. and Roberts, J. (1994) Evaluation in ELT Oxford: Rlackwell.

Chapter 15Kathleen GravesA FRAMEWORK OF COURSEDEVELOPMENT PROCESSESU R R I C U L U M DES I G N SPEC I A L I S TS have developed various frameworksC that break down the process of curriculum and course development into componentsand subprocesses (see, for example, Dubin and Olshtain 1986; Hutchinson andwaters 1987;Johnson 1989; Nunan 1985, 1988a, 1988h; liichards 1990; White 1988). A framework ofcomponents is useful for several reasons: It provides an organized way of conceiving of acomplex process; it sets forth domains of inquiry for the teacher, in that each componcntputs forth ideas as well as raises issues for thc teacher to pursue; it provides a set of termscurrently uscd in talking ahout course tlevclopmcnt and thus a common professionalvocabulary and access to the ideas of othei-s.Thc fi-amcwork described hcre, while drawingon the work ofothcrs, is cast in terms ol’my own work with teachers. It is not a frameworkof equal parts: Each individual’s context clctermincs which processes need the most timeand attention. Furthermore, the proc cs are not necessarily secjucntial hut may be carriedon in the planning, tcaching, and rcplanning stages ol‘ course development.InTalile 15. I, each componcnt is identified and rephrased in question form to clarifyits meaning.Needs assessmentWhat are my students’ needs? HOM. can I assess them so that I can address them?What is nccds assessment,’ and whv does a teacher undertake it? At its most basic,nccds assessment involves finding out what the Icarners know and can do and what theyneed to learn or do so that the course can bridge the gap (or some part ofit).Thus nccdsassessment involves seeking and interprcting information about onc’s students’ nceds sothat the course will address them effectively. However, how one defines a student’s needsis a complex issue open to interpretation. One \vay of conceptualizing needs is to distinguishbet\veen “ohjectivc” and “subjective” needs (Richterich 1980). Rrindley (1 989: 70) definesobjective needs as “derivable from different kinds of factual information about learners, theiruse of language in real-life communication situations as well as their current languageproficiency and language difficu1ties”and subjective needs as “the cognitive and affective needsof the learncr in the learning situation, derivable from information about affective andcognitivr factors such as personality, confidence, attitudes, learners’ wants and expectations

~ countryCOURSE DEVELOPMENT PROCESSES 179Table 1 i 1brame\\ ork component\Needs assessment: What arc in) stutlcnts’ nectls? I lo\\ can I asscss them so that I can adtlrcsathem?Determining god5 and objectives: What are thc purpow and intcntlctl outcoinc< of thccoiirw’ What \\ill in! students ncctl to do or lcarn to athic\c thew goal\/Conceptualizing content: What 13 111 l x the backhone ot\\ llal)u

particularly,teachers180 KATHLEEN GRAVESmay have to work with a conccption of nccds dctc.rmincd by their institution or other partyand conduct their assessment accordinglv.When does one conduct a nccds assessment? Ikpcnding on one’s context, needsasscssmcnt can lie conducted in stagc 1, the planning stagc; in stagc 2, thc teaching stage;and also in stage 3, the rcplanning stagc, if one determines that the assessment must bemodified in some \.ray. Teachers \vho have contact with their students prior to teaching thecoursc can undertake a precourse nectls assessment. In many cases, however, a formalprccoursc ncctls asscssincnt is neither necessary nor appropriatc. Some teachers arc ableto make fairly accurate assumptions ahout their students’ nccds with respect to the courscon the basis of prior cxpcricncc \vith the coursc or with thosc particular studcnts. In manycases, precourse asscssmcnt is simply not fmdilc lx~ausc~ the tcachcr docs not have contactwith the students until the first day of class.Another important factor in tlcciding \vhcn to assess needs is the teacher’s view of thepurpw of nccds amcnt can also IK a teaching tool because it canhelp students bccomc more aware and more purlioseful in their learning. Many teacherssee it as an ongoing part of teaching, on the one hand, because it may take time to establishthe kind of rapport with students that allo\vs fbr a clcar understanding of needs and, on theother, because they view it as a teaching tool that enables thcm to \vork in partnership miththcir students to determine ncctls antl ensure that the course mccts thosc nccds. lcachcrswho use nccds assessment as an ongoing part of their classes develop activities that helpstudents clarify and focus their needs. Such activities can include mindmapping (crcating\vortl maps liascd on, for cxxamplc, the \vortl rvritrng) antl student-generated questionnaircs(Grant and Shank 1993). [. . .]How docs onc conduct a ncctls ssmcnt? Teachers use a variety of methods.Questionnaires are a common needs a mcnt tool. They can bc written in English or,when appropriate and feasililc, in thc native languagc of thc students. One of the challengesin tlcsigning a questionnaire is choosing qucstions that will t)c interpreted correctly and willprovide the information sought, especially if one is seeking subjective data. Interviews withstudents antl others (such as employers or professors) arc another common way of findingout students’ needs. Other means include observation of or, in some cases, participation inthe situations in kvhich students will use English. Teachers may obtain samples of writtenmaterials, such as manuals or tcxtlmoks, that students will have to use. Stern (1992) cautionsagainst gathcring so much data that one cannot analyze and put it to use.Tests and intcrviclvs that mcasurc proficiency arc also a part of needs assessmentbecause they help dctcrmine what students already know and where they are lacking. Manyinstitutions administer proficiency tests for placement purposes. Teachers ma? also designin-class activities for the first days of class that mcasurc stutlcnts’ proficiency in reading,xvriting, speaking, or listening.Hutchinson antl Waters (1987: 54) make a distinction hctivccn target needs (“whatthe learner needs to do in the target situation”) antl /earnin

COURSE DEVELOPMENT PROCESSES 181problem because many of their students have no target needs, no clearly anticipated use forthe skills gained through study. English may be a requirement for an exit or entrance cxam.It may lie viewed as a subject like math or science, or it may be a social undertaking like thestudy of music. For these students, thc notion of needs outside the classroom is tenuous.The focus of the needs assessment shifts to the learning needs or subjective needs of thestudmts so as to increase motivation and to help students find purpose antl interest in whatthey are doing in the coursc. For example, Gorsuch (1 991) dcscrilies a technique for helpingstudents in a conversation class in Japan articulate thcir needs antl set periodic antl achicvablcgoals to meet those needs.IssuesNeeds assessment is not a value-free process. It is influenced by the teacher’s view of whatthe course is about, thc: institutional constraints, and thc students’ perceptions of what isbeing asked of them. For example, one teacher of immigrants might ask thcm to listsituations in which they us? or expect to use English, with the aim of providing instructionin the language and behavior necessary to deal Lvith those situations. Another teacher mightask the same students to articulate or enact problems they face in adjusting to the nmvculture, \vith the aim of helping thcm exert control over the acculturation process.For many students, needs assessment is an unfamiliar procedure, and they may ha\-cdifficulty articulating their purposes or nerds. The process itself may cngrnder uncertaintyin the students, as knowing thcir nceds is presumably the responsibility of the teacher orinstitution. Questions may be interpreted differently by different students or may not elicitthe anticipated answers. Studcnts’ perceptions of needs may not match those of the teacher.The teacher’s view of the students’ needs may conflict with those of the institution.Thc content and method of needs assessment should he e\ aluatcd as to appropriatenessantl effectiveness in achieving thcir purpose of identifying the nccds of the students. It mayral tries to dcvc>lop ef‘fectivc needs assessment tools. Those tools should not heviewed as “one time only” pro ’es. Needs asscssment should he 1,icwctl as an ongoingprocess, both in its development antl in its use.Determining goals and objectives1Vhat arc the purposes and intentled outcomes of.the coiirxe?IVhat n,ill m,r’ students need to c/o or leurnto achieve these goals? What arc goals and ohjcctives antl what is the relationship betwcrnthcm? Goals arc general statcments of the overall, long-tcrm purposes of the course.Objccti\es express the specific \.rays in xvhich thc goals \vi11 be achieved.Thc goals of’a courscrcpresent the dcstination; the objectives, the various points that chart the course to\vartlthe destination.To arri\-c at thc destination, one must pass each of these points. [. . .]Why sct goals antl objectives? Setting goals and chjcctives pro\-idcs a scnsc of directionand a coherent framework for thr teacher in planning her course. Breaking goals down intoobjectives is very much like making a map ofthc territory to be explored. It is a Lvay for theteacher to conceptualize her coursc in terms of teachable chunks. Clear goals and ohjcctivesgive the teacher a basis for determining which content antl activities are appropriate lor hercourse. They also provide a framelvork for evaluation of the effectiveness or worth of anactivity : Did it help students achieve or make progress to\vard the goals and objectives?Clearly, thcre are many routes (objcctives) to a givcn destination, some more circuitous thanothers, and the length antl nature of the route will depend on one’s departure point.How does one choose appropriate goals and olijecti There is no simple ans1vc-r tothis question. To arrive at the goals, one asks the question, “What are the purpost~ and

182 I

COURSE DEVELOPMENT PROCESSES 1835 Critical thinking objectives articulate which lc-arning skills students 1% ill develop.Examplc: Sttidents will he able to determine characteristics of u good paragraph and .say whythe). think a paragraph is good.Tension often exists between coverage objectives and mastery objectives hccausc thctimc it takes to master skills or knowlcdgc or to develop awareness may not corrcspond tothe timc allotted in a syllabus.This tension can create dilemmas for teachers Lvho must coverantl test the material in the syllabus yet wish to ensure that students ha\c mastered thcmaterial prior to moving on. The tension can also put tcachcrs at odds with their studentsor the institution if thc teacher believes that success is achieved through demonstratedmastery but the studcnts cxpect coverage to mean mastery.IssuesThe main issue is that many teachers do not formulate goals and objcctives at all or do soonly after having thought about what they will teach and how. Studies on teacher planningundcrscore this fact (Clark and Peterson 1986). My own work xvith teachers has shown thatthey consider the setting of goals and objcctivcs a valuablc process but one that they finddifficult to articulate and organize.They feel that they must first be clcar about what theyarc teaching and how they vicw thc content.They report from expcricncc that they cannotclearly formulate their goals and objectives until after they haw taught thc course at leastoncc. (Returning to the map analogy, one cannot map a route until onc has traveled it.)Thus for many teachers, this is not thc entry point into the process of course dAnother issue involves clarity with respect to students’ nccds. It is easier to set goalsin situations where these needs are clcar; otherwise, the goals of the coursc may shift andbe redefined as thc course progresses. Finally, goals and objcctives are a statement of intent,subject to reexamination and change once the course is under way.Conceptualizing contentIzl’hhat will he the backbone ofwhat I teach? What will I include in my yllahris?When a teacher conceptualizes content, she is figuring out which aspects of languageand language learning she will include, emphasize, and integrate in her course. This is notthe relatively simplc pro ’ it once mras.Two dccatlcs ago, language tcaching was still heavilyinflucnccd by a structural vicw of language (Richards and Kodgers 1986). This influcnceresulted in a“one size fits al1”approach to content and methods, meaning that, for example,an EFI, teacher could use the same textbook and the samc drills or pattern practice forfactory workers, college students, and housewives. There \vas not much question aboutcontent: It was grammatical structures and vocabulary.Much has changed in rcccnt wars in the ficlds of applied linguistics antl languageacquisition and in approaches to language teaching. The proficiency movement, the conceptand various modcls of comrnunicativc competence, the advent of ESP (English for specificpurposes)), the proliferation of methods of language teaching, and the diversification of thepopulation of English learners have all provided the teacher with many more options toconsider in deciding what will be the backbone of her course (Canale 1983; Hutchinsonand Waters 1987; Omaggio Hadlev 1993; Richards 1990; Savignon 1983;Yaldcn 1987).Now the choices a teacher makes are much more contcxt-dependent antl so involve anumber of factors such as who the students are, their goals and expectations in learningEnglish, the teacher’s own conception of what language is and what will best meet thestudents’ needs, the nature of the course, and the institutional curriculum. A course for

184 KATHLEEN GRAVESimmigrants in an English-speaking country will likely stress different content than a coursefor high school students in their own country.Let us look at somc ways of conceptualizing and categorizing content. The boundarieslietween categories arc permcablc; they ovcrlap conceptually antl are not exclusive of eachother.The teacher’s challenge is to figure out \vhich ones are appropriate for her course antlhow she will integrate thcm. They \vi11 he described and then outlined in a syllabus grid,lvhich \vi11 be adtlctl to with each successive component. In my experience, teachers do notusually use syllabus grids to la! out the content of a coursc but a grid is a graphic way toillustrate possible categories.The traditional wy of conceptualizing content, which many teachers have experiencedin thcir own learning of languagc, is as grammar structures, sentence patterns, andvocabulary. These aspects of language arc relatively . tcmatic and rulc-govcrncd and arcoften the basis of content found in tcxthooks. Th includc rules of word formation(morphology), rules of pronunciation (1’ nology), antl grammatical structures andrelationships among \vortls at the sentence IC 1 (syntax). A syllahus grid that includes theseaspects of language might look likc this:GrammarPronunciationVocabularyFor language teachers, the possibilities for \vhat to include in a syllabus opened up cviththe advent of what has come to I)c called the communicative approach (Larsen-Freeman1986).Thc work of sociolinguists such as Hymcs (1972) antl Hallitlay ( 1973, 1975) antl ofapplied linguists such as Wilkins (1976) and Van Ek (1975) has helped reorient thinkingabout the nature of language.‘l’he communicativt, approach is liased on ideas about language,on the one hand, antl about the purposes 01‘ languagc learning, on the other. Language isused in a context, hvhich clctcrmincs and constrains the choices that language users makewith respect to purpose, stvle, register, antl topic. Learners must use the language and havepui-poses for using it. From the point of view of‘ conceptualizing content, the communicativeapproach added s ral dimensions. First, it adtlctl the dimension of language functions,such as to apologi to persuade, to con\ information. It also atltlcd the dimension ofnotions, kvhich form a continuum 1rom general concepts such as time, space, andrelationship to specific topic-related notions such as house and home, \veather, antl personalidentification (Van Ek 1975). Language was sccn as being used for communicative purposesin situations lvith other people, which call on the learner to pay attention to both the contentof the languagc and its appropriatcncss tvith respect to formality, non-verbal behavior, tone,antl so on. Communicative situations might include ortlcring food in a restaurant, buyingstamps at the post office, extending an invitation to a social cvcnt. Thus \.re can add thesecatcgorics to our syllabus grid:I bunctions I Notions antl topics I Communicative situations IGrammarPronunciationVocabularyThc proficiency movcmcnt antl thc dcvclopmcnt of proficiency guidelines havecmphasid a four-skills-based approach to syllalius tlcsign (Chaggio Hadley 1993). Forsomc teachers, these skills arc a givm, as studcnts have to use some combination ofspeaking,

COURSE DEVELOPMENT PROCESSES 185listening, reading, and \vriting in class. However, because hecoming proficicnt in each ofthese skills cntails mastery of a set of subskills and processes, many teachers choose toemphasize certain skills or find ways to integrate them. For example, to become proficicntin writing, a student must learn ho\v to structurc paragraphs, holv to use cohesi\e &\ices,the rhetorical styles ofwritten English, editing techniques antl so on. Thus bvc can add thcfolhving categories to our syllabus grid:Listening skills Speaking skills Reading skillsI Writing skillsFunctionsNotions and topicsCommunicative situationsI Grammar I Pronunciation I Vocal>ular!IThe emphasis on communicative competence as based on antl brought almut byinteraction has prompted a vielv of language as not just something one lcarns hut somethingone tlocs.Thus teachers may conceive oftheir s);llal)us in terms of what thc students will doin the classroom as activities or tasks.Tasks ha\e been \ariously defined. Prabhu (1 987: 24)defines a task as an activity that requires learners “to arri\e at an outcome from givcninformation through some process of thought,” such as deciding on an itinerary based ontrain timctablrs or composing a telegram to send to soincone. Tasks have also been definedas projccts in which learners Lvo1-k together to ~-)roducc. something, such as a putting togethera ncLvspapcr or conducting a survey (Hutchinson 1984). Nunan (1989) proposes a taskcontinuum, lvith real-izorld tasks at one end and pedagogic tasks at the othcr. Real-worldtasks ask students to use languagc in ways that they might outside thc classroom, such aslistening to thc radio, reading the ncn.spapcr, or using a train schedule. Pedagogic tasks arcones that \vould not occur outside ofthc classroom hut help stutlcnts tl lop skills necessaryto function in that world, such as information gap actiyitics.Thc competency-bascd approach to syllabus design was dc lopcd in the Unitctl Statesin response to thc influx 01’ immigrants in the 1970s antl 1980s. It is a comlination of thecommunicative and task-lmctl approaches antl has bccn used in courses for teachingimmigrants, who have immediate ncctls with respect to functioning in English in thccommunity and in the workplacc. Competencies are “task-oriented goals written in termsof behavioral objecti\ es that inclutlc langu Ixhavior” (Center for Applied 1,inguistics 1983:9).Thcy arc the language antl Ixhavior nc x-y to function in situations related to living inthe community antl lintling and maintaining a job. Competencies related to living in thccommunity haw also been called ljfi-skilf.7. Those related to jobs have been called rmmioncdskills. (See, for example, the California ESL Model Stantlards for adult cducation 1993.)However one dcfincs them, tasks can 1)c gcarcd to one’s spccific group of learners. Forliusiness personnel, tasks might include giving a business presentation or lvriting a report;for university students, tasks might include \vriting a rcscarch paper or preparing a reportfrom notes takrn at a lecture. We can add t\vo othcr categories to our spllahus grid:Tasks and activiticsCompctcnciesI Listening skills I Speaking skills I Reading skills I Writing skills II 1,unctions I Notions and topics I Communicative situations 1GrammarPronunciationVocabu 1 ai.)

186 I

COURSE DEVELOPMENT PROCESSES 187Participatory processesExamples: problemposing, experientiallearning techniquesCultureExamples: c u It u reawareness, culturebehaviour, cultureknowledgeLearning strategiesExamples: self-monitoring,problem identification,note takingTasks and activitiesExamples: informationgap activities, projects,skills or topic-orientedtasks such as giving aspeech or making apresentationContentExamples: academicsubjects, technicalsubjectsCompetenciesExamples: applying for ajob, renting an apartmentListening skillsExamples: listeningfor gist, listening forspecific information,inferring topic,choosingappropriatereponseFunctionsExamples: apologizing,disagreeing, persuadingSpeaking skillsExamples:turn-taking,compensating formisunderstandings,using cohesivedevicesIReading skillsExamples:scanning forinformation,skimming for gist,understandingrhetorical devicesNotions and topicsExamples: ti me, quantity,health, personalidentificationWriting skillsExamples: usingappropriaterhetorical style,using cohesivedevices, structuringparagraphsCommunicativesituationsExamples: ordering in arestaurant, buying stampsat the post officeGrammarExamples : structures(tense, pronouns),patterns (questions)PronunciationExamples: segmentals(phonemes, syllables),suprasegmentals (stress,rhythm, intonation)VocabularyExamples: word formation(suffixes, prefixes),collocation, lexical setsF~pre I:.1 ’l’hc complctctl syllabus gridIssuesTeaching involves making choices. It is not possible to teach a syllalius that cxplicitlyencompasses all the areas mentioned here so teachers must tlccidc lvhich catcgorics makesense to them for a givm coursc. The categories also overlap, both conceptually antl in theclassroom. For example, pronunciation is an important part of speaking skills. Vocabularydevelopment is a part of notions antl topics. Ixarning strategies can bc linked to specificskills. Some of the categories arc vast and can lie divided into several subcategories. Manyrcatlcrs will find that they ~vould label or define the categories differently or that certaincategories are missing. For example, some teachers conceptualize content thcinatically.Teachers of courses whosc content has already hcen specified will face different issues.They may find that thr breadth of content is unrealistic for the amount of time they have toteach it or that the way content has been defined is inappropriate, in their view, for thepurposes of the coursc. The ovcrlapping nature of the categories may be an aid in findingways to adapt the existing content to their vision of the coursc.Selecting and developing materials and activitiesHopi ant1 crith nhut will I tcuch the coiirrc~LYhut ir mj ro1eiVVhat lire my \tutlent\’rolec~For many teacher\, coui \c dc\ clopment \tarts not with determining objccti\ e\ or

188 Icyond the constraints of the text. Das (1988: viii) pointsout that matcrials should not “pre-specify learning outcomcs or attempt to control orsubstantially guide learning: their function is primarily to provide opportunities for learningthrough interaction.”The question “How will I teach?” also encompasses a teachcr’s approach and how sheiievvs her role and that of the learners. How much initiative will the students be expectedto take, and toward what end? Ilow will the students be asked to interact?Thc emphasis onlearner awareness and concern for extending learning beyond the classroom have made therole of the learner a central focus of how a course is taught. Teachers design courses withactivities and materials that have the students take a more active role in reflecting on theirlearning, determining the content ol’the course, antl pursuing projects of interest to them.Such an approach ma! tacilitatc the search for materials in that the emphasis is not on thematerials themselves hut on what the students (lo with them.

COURSE DEVELOPMENT PROCESSES 189IssuesFor some teachers, the lack of materials is a challenge; for others, it is an opportunity.loping materials requires time before, during, and aftcxr the course ~ for preparing,using, antl modifying them, rcspcctivcly. Yet having to use certain materials may producethe dilemma of coping with a tcxt that does not meet students’ nerds or docs not promotethe teacher’s view of the roles of learners antl teachers. Other aspccts of coursedevclopmcnt, such as needs a. ment antl objective setting, may help thc tcachcr .to adapt unsuitahle materials and to what extent. Evcntually, all materials arc adapted ormodified in some \vay. Even materials that have been drvelopd hy teachers for specificcourses \vi11 be modified o\er time.Organization of content and activitiesHow will I organize the content anti activities? What ystems will I ckidop?Regardless of whether one follows a fixcd sequence or adopts a morc fluid approach tothe order in which one teaches the content, part of course de\elopmcnt is figuring outtems lor organizing the course. Systems can focus on the lesson level (the organizationof each lesson) antl on the course Icw.1 (thc ovcrall organization of the course). We will lookfirst at specific considerations in sequencing inatcrial and then at eonsiderations of the overallorganization of the course.Two general, complcmentary principles of sequencing are building and redeciding ho\v to sequence material, one considers hiltling from the simple to the complex,from more concrete to morc open-ended or so that unit or activity A prepares stutlcnts forunit or activity B. Building fi-om the simple to the complex in a writing course may meanlearning how to write narratiw prose before developing an argumentative paper. In anintroductory language course, it may mean learning the nurnlicrs 1 to 9 to use telephonenumbers antl then learning the numlxrs 10 to 60 to tell time. Ruiltling froin morc concreteto niore open-ended in a writing course may mean that students first unscramhle and discussa sample paragraph before writing their o\vn paragraph. In an introductory language coursc,it may mean talking about a family in a textbook picture using prescrihetf ocahulary Idoretalking ahout one’s on-n family.Conceiving of activities as Iiuilding Mocks puts them in a “leeding” relation xvhcrc oneactivity feeds into another “if it provitles something that is needed for the second onc . , .or the second exercise could not be donc unless thc first had already heen completed” (1.o~1989: 145). For example, in a reading unit, students predict the content from pictures orheadings that accompany the tcxt before actually reading thc tcxt. Or prior to a rcstaurantrole-playing activity, students learn mcnu itcms antl the language for ordering f(~od.The principle of recycling matcrial means that studcnts encounter previous matci-ial inne\\ wys: in a new skill area, in a differcnt type of activity, or with a new focus. For example,material encountered in a listening activity may lie recycled in a 1% riting exercise. Matcrialencountered in an individual reading acti\ity may be rccyclcd in a role play with otherstudents. Material about the target culturc may be recycled in an activity about one’s o\vnculture. This approach to recycling material assumes that each new encountci- with thematerial provides a challenge to students, therely maintaining thcir interest and motivation.Recycling has the effect of integrating material and thus augments students’ ability to useor understand it.1. . .ITwo complementary ways to approach thc overall organization of a course arc as a cycleor as a matrix. Both approaches suggest a core of material to be learned and activities to be

190 KATHLEEN GRAVESconducted within a givcn time frame. In the cyclical approach, a regular cyclc of activitiesf'ollo\vs a consistent sequrncc. In a matrix approach, the teacher xvorks with a set of possibleactivities for a givcn time frame and, as thc course progresses, dccides which activities tolvork with. For an EAP course, I3lyth (1 996) dcscrilm such a situation, in which shecompiles a list of possible activities and materials and then tlccidcs which to use, dependingon her studcnts' interests as \vcll as the availability of the materials.The cyclc and the matrix arc not mutually exclusive; inany teachers use elements ofboth. Certain features in a course ma! lie predictable, augmented by other elements drawnfi-om a matrix, dcpcnding on the situation. Teachers who work with a fixed syllabus, suchas that in a textbook, may nevertheless follow a cjclc in thc way they lvork with the material.Adapting material oftcn means approaching it as a matrix from which to select, dependingon onc's students. Many teachers also set up certain tlaily or weekly rituals. For example,somc teachers hegin cach session \vith a \varm-up or review. Some tcachrrs liegin eachweek with a student presentation or end cach \vcck with an oral feetlback session. All ofthcsc methods of organization permit a teacher to gi\c a shap to her course.IssuesAlthough thc order in \vhich thc content and materials arc taught may hc determined priorto teaching the course, it ma! also bc dctcrniinctl antl modified as the coursc progresses.For some teachers, a negotiated syllabus, in which teacher antl students decide togcthcrwhat they will learn, is prcfcralile. In such ca~ , a predetermined scqu'nce is seen as ahandicap as it docs not alhv teachers to takc into account the particular group of' studentsin thcir course. In such a course, the sequence is not dctcrminetl beforehand. Rather, theteacher has a map of the possihlc tcrritor! antl \vorks \vith the students to dctcrmine whereit is most useful for them to go and in bvhat order. Whet-c a syllabus is provided, achicvingflexibility is an issue.EvaluationHow will I u.s.sc.ss ithat stiitlents have learned? Him. iv711 I assess the $fictivcnc.s.s of'the course?For most teachers, evaluation tncans evaluation ivithin the course; assessing students'proticiency, progress, or achie\micnt. Ho\v proficient arc students in listening? Arc studentsimproving their \\.riting skills? Have the); Icw-iied to I'unction in English in the workplace?Teachers liuild in some form of studcnt evaluation when developing a coursc, ranging fromformal tests to informal mcnts. Hughes (1 989) tliscu ' four purposcs for testing: tomeasure proficicncy, to ose specific strengths antl w csscs, to place students in acourse or program, antl to assess their achicvvment in a course or program.Thc same testinginstrument may be used for- more than one purpose. For cxamplc, thcTOEFL test is usedby graduate programs in the United States as a proficicncy test, but it is sometimes used asan achievement test ifstutlcnts show a gain on a7'OEFL posttcst. Ho cr, tests are not theonly means teachers have to assess their students. Tcachcrs may structure their classroomactivities so that they can assess their students whilc the students participate. They may usea portfolio approach, in which students put togcthcr a portfolio of their work (Fingeret1993).They ma); involve their students in deciding what should be assessed and how (Hull1991).Evaluation in course development also includes evaluation of the course itself'. Was thecourse effective? In what ways? Where (lid it fall short? Such an evaluation may not bedirectly linked to assessment of student progrcss, although student evaluation and test results

COURSE DEVELOPMENT PROCESSES 191can provide feedback on the effectiveness of the course. If the students do \vel1 on tests orarc judged to have made progrcss, presumably the course has been effective. Rut if studentsdo not make progress or do not demonstrate a certain level of achievement, the effectivenessof the course may be questioned. Finding \vhcrc the fault lies \vould lie onc of the purposesof course evaluation antl could involve having students suggest why they did not make theprogress expected.Why docs one evaluate? Gcncrally speaking, a course is evaluated to promote andimprove its effectiveness.This may be an internal matter, as when the teacher is concernedwith developing the best course possible, in \vhich case the evaluation is done largely forthe Ixncfit of the students and the tcachcr. However, courses are also evaluated to providedocumentation for policy reasons, such as continued funding or retention in the curriculum.In such cases, evaluation is an external matter, and the teacher may he rcquircd to use certainmethods of evaluation or to document the effectiveness ofthe course in a manncr prcscrihetlby an outside party. This in turn may influence the development of the course.What can be evaluated? Any part ofthe process of course development can lie evaluatcd,including thc assumptions about antl analysis of students’ needs or backgrounds, goals antlohjcctivcs, materials and activities, means of ing students’ pi-ogress, studentparticipation, student roles, and the tcachcr’s role. each element of the frame\vork isitself subject to evaluation. Was the needs assessment cffcctivc? Did I seck the right input,antl did it enable me to make appropriate decisions about the course? If not, \vhy not? Wcrcthe goals and objectives appropriate and achievable? Should they be changed? Did studentsfind the material appropriately challenging, or was it too easy or too difficult? Wcrc thractivities appropriate? I>id all students participate easily? Did I find suitable \rays to evaluatestudents’ progress? Did the tests test what had heen learned?When does one rvaluatc? In curriculum design, a distinction is usually made bctwcrnfbrmcitivc ei.uluotion, which takes place during the development antl implementation of thccurriculum for purposes of modifying it as it is Iieing developcd, antl summutire evaluation,which takes place after thc curriculum has heen implemented, for purposes of evaluatingits success antl improving it for future implementation (Brown 1989). A teacher \vho isinvolved in each stage of coursc design can think of cvaluation as an ongoing part of theentire proccss.’l‘hus evaluation can occur in the planning and teaching stagcs of the course,after it is over, and \vhcn it is replanned and rctaught.Who evaluates? At the course IC\ el, the teacher and thr students are the principalr, administrators, funtlers, parents, and clients may hac a role incxvaluation, antl thcir role may influence the shape or existence of the course.How does one evaluate? A variety of ways arc available. A tcachcr’s most importantmeans is close observation of’what students do in class and how they do it. If’students ha\?great difficulty performing certain tasks, on

192 KATHLEEN GRAVESnecessary. Teachers must become familiar with thc various purposes and types of tcsting,but thcy must also tlevisc their own systems and arcas of inquiry. As with needs assessment,teachers must experiment kvith different methods ol’evaluation and monitor the success ofeach so as to maximize the effectiveness of thcir courses.Consideration of resources and constraintsWhat are the girwx ofmy situation.?Resources and constraints arc thvo \yay of looking at the same thing. A required courscliook may bc a constraint for one tcachci- and a resource for another. A class of fe\ver thanten students may be a rcsource lor one tcachcr antl a constraint for another. Though thesegivens may seem secondary to the processes just descrihcd, in fact thcy play a primary rolein the devclopment ofa coursc because it is in considering the givens that a teacher beginsto makc sense of processes such as ncctls assessment and material selection. I have referredto this elsewhere (1 996) as prolilematizing: defining the challenges of one’s situation so thatone can makc decisions about what to do. In the absence of prohlcmatizing, a teacher mayseck to graft solutions appropriate to another unique situation onto her situation. Thishecamc clear to me in the case of an EFL tcachcr who faccd an extraordinary challenge:designing a conversation class for 140 studcnts in a space meant for half as many. She feltthat having examples of nccds analysis questionnaires \vould lie a key to drveloping hercoursc.To me, this was an example ofa tcachcr seeking answcrs from outside without havingfirst specifically defined the challcngcs of her o\vn situation. Such prolilematizing couldeventually result in an examination of hou. others approached needs analysis as an aid indeveloping her owm. Hci-c is a skctch of one Ivay of prolilematizing this teacher’s situation:This is a coni-ersation class, brit there are 140 students in u .space thnt.fi:ts 70. I nccd to lookat \vays of hvorking within thc constraints of the classroom such as ways to group orrotate students.Il’hut kinds o_f‘coniwsations can 1 4 stridcnts possib[r hare? I ncctl to assess their languageability (At what Icid can thy cary on a conrersation?) antl find out about thcir hackgroundand interests (Il’hat can the), haiz coniw-sations about?). Ho\v will I go about doing that?What kinds of questions should I ask them? If thc assessment shows that their abilityis lo\v, I nccd to focus on the kind of prqm-ation and foundation work necessary forconversations to take place.How cc7n I get them to work together ro hare these conversations? Classroom managcment isan issue. I need to look at available matrrials \vith carefully structured activities as ameans of classroom management. Or perhaps I could ask other teachers what hasworked for them in this situation.How can I monitor their uctirq? I nccd to examine my rolc in the classroom. I also ncctlto think almut thc types of monitoring antl evaluation mechanisms I will usr in theclass.Il’hut has worked in the post? I need to think allout thc acti\ ities or classes in which I feltthat things went wcll. Why did they gc) wcII?Wliat can I takr from thosc succcsscs andhuild into this coursc!Thcsc arc questions that I propose. Were thc tcachcr to go through a similar process,shc might ask different ones or respond to the same ones in different ways becausc of herintimate knowledge ofhcr context antl her rolc in it. For cxample, how students arc graded,whether there is a required text, and attendance patterns \vould all influence the kinds of

COURSE DEVELOPMENT PROCESSES 193questions she would ask. I inclutletl the question about past succcsscs because tcachcrscarry their experiencc ovcr from one context to the next, and being able to understandwhat has been successful antl why can provide a foundation for planning a coursc. In thecontext under discussion, the teacher had already taught the course and thus could berealistic in her expectations about lvhat she could hope to accomplish mith this group ofstudents.‘The constraints and rcsourccs of one’s situation take many forms, some tangible, othersnot. Teachers work with or without physical antl material resources such as lmoks,technology, a classroom, and furniturc. The lack of physical rcsourccs may cncouragc ateacher to use available resources in creative ways. The availability of technology may allowa teacher to have groups of students work indcpcntlcntly. Time is another importantconsideration in designing a course. How often, how long, and o\er what period of timewill the class meet? How much time is available to the teacher to prepare for the courseantl the classes? A teacher may adjust her teaching priorities according to the length of thecourse. The kinds of activities she designs may bc affectcd by the amount of timc shc has,hoth in class and before class.The institutional philosophy, policy, and curriculum are important givens. Having towork within existing curricular guidelines is both a constraint antl a resource; so is ha\ ingto devise one’s own syllabus. The type of administrative antl clerical support provided bythe institution affects a teacher’s choices. For example, lack of clerical support will suggeststreamlining paperwork and materials. Support from the administration for innovation willencourage experimentation.The numbers, levels, antl cultural backgrounds of the students are both a constraintantl a resource. For example, a large class may cause a teacher to focus on classroommanagement. A multile\d class may influence the teacher’s selection of material or activities.The teacher herself is the most important given. Her background, experience, andbelicfs play a significant role in the choices she makes. For example, one teacher will focuson certain content becausc she dccms it csscntial to successful language learning, whileanother will ignore the same content. A teacher who usually devclops hcr own materialsmay choose to use published materials whcn teaching a course' whosc contcnt is new to her.The givens of a situation cover a broad range of factors and affectteacher makes. Teachers plan antl teach courses not in the abstract but in the concrete oftheir constraints and rcsourccs. For cxamplc, an ESL teacher hvho teaches in an intensiveEnglish program, whosc students change from one program to the next, may need toinvestigate the background antl proficiency of her students, whcrcas for a high school EFLteacher, this may be a given because she knows the students. The teacher in the intensiveEnglish program might begin with a question such as “IIow can I find out the culturalbackground and needs of my students so that I can address those needs effectively in thc sixweeks of the coursc?”Thc high school teacher’s initial question might be quitc different,say, “How can I keep my students motivatcd in a rcquircd course?” Course development,like teaching, is not a neatly organized process t a complex one in which teachers arcconstantly considering multiple factors and pro ding on many fronts.IssuesThe givens of one’s teaching situation, both tangible antl intangibly, cannot lie ignored.Effecting change rcquircs both recognizing what can be changcd and accepting what cannot.The“If only . . .”syntlromc (if only \ve had the technology, if only we had quieter classrooms,if only our students were more motivatcd) can obstruct change as firmly as the “Yes,

545.194 KATHLEEN GRAVEShut . . .” 5>ndrome (Yes, hut that \\ill rimer mork in m> wtting.) Problemati/ing enablesa teacher to tlecidc \that shc can changc, \I hat she can’t, antl cc hcrc to startConclusionThe components discussctl in this chapter and summarim-d in Table 1 5.1 should serve notas a checklist for thc teacher but rather as a set of tools for talking about, understanding,antl directing thc process of course tlc lopmcnt. Each component is contingent on everyother component. For cxample, assessment dcpcntls on how one conceptualizes content oron how. she interprets students’ needs. Conccptualizing content in turn influences the coursegoals antl objectives. Thus whcrcvcr one starts in thr Ixocess, each component willeventually come into play. Each componcmt is, in many respects, one \vay of\vorking withthe whole.Note1 The tcrms needs tina$,.si.i antl nerds usicrsmcn~ arc often used intcrchangcabl!. Rut as SusanPomcroy oncc suggcstcd to me, thcy rcfcr to separate processcx: Asscssmcnt involvesobtaining data, \vhercas analysis involves assigning value to those data.ReferencesAucrllach, E. 1993. “Putting the p back in participatory.”’l’ESOL Qiurterb. 27 (3): 543 ~Auerbach, E. and N. Wallcrstcin. 1987. LSL fir Action: Problem Posing at ICbrk. Reading, Mass.:Addison- Wesley.Blvth, M. (1. C. 1996. “Designing an LAP course for postgratluatc students in Ecuador.” In K.Graves, ctl., Teuchers as Course Derdopers. Carnhritlgc: Cambridge University Press.Rrindley, G. 1989. “The rolc ofncccls analysis in adult ESL program tlesign.” In K. K. Johnson,ed., Thr Secontl Language C~~rrictil~~ni,1111. 63-78. Camhritlgc: Cambridge University Press.Brinton, D. M., M. A. Snon., antl M. 13. Wcachc. 1989. Content-based Seconcl lungriagc Instruction.Brown, J. D. 1989. “Language program c\-aluation: A synthesis of existing possibilities.” InR. K. Johnson, etl., The Second Langrrage Crirriculum, pp. 222-243. Cambridge:Cambridge University Press.California Drpartmcnt of Education. 1993, English (7s (I Second ~.on~qtrugc .l)fodel Stuntlartl.sfbr;ltluZtEducation. Sacramento.Canale, M. 1983. “From communicative compctcnce to communicative languagc pedagogy.”In J. Richards and R. Schmidt, eds., Lungmqe ~7ntl Communicntion, pp. 2-27. I ondon:Longman.Crnter for Applied Linguistics. 1983. From [he Clasiroom to the Ilbrkpluce: Teuchin

COURSE DEVELOPMENT PROCESSES 195Gorsuch, G. I991 . ‘Helping students create their own Icarning goals.”LangLiaiqeTeucher I 5 (12):3,9.Grant, S. antl L. Shank. 1993. “Bcyond questionnaires: Engaging learners in ntwls assessment.”Presentation at the TESOI. conference, Atlanta.Graves, K. 1996. “Teachers as course devrlopers.” In K. Graves, cd., Teachers us CourseDerelopers. Cambridge: Caml)ritlgc University Press.Halliday, M. A. K. 1973. Explorutions in the Functions pflungtrage. London: Arnold.-. 1975. Lecirning How to ,Venn: Explorations in the Der,elopmenr $Lc7nguuge. London: .\mold.Hughes, A. 1989. Testing.for Languuge Teacher,-. Cambridge: Cambridge University Press.Hull, I.. 1991 . “Self-monitoring and self-evaluation: A guide for facilitating intkpcndcnt andautonomous learning.” Unpul)lishetl master’s thesis, School for Intcrnational Training,Rrattlcboro, VT.Hutchinson, T. 1984. Project English. Oxford: Oxford Uiiiwrsit! Press.Hutchinson, T. and A. Waters. 1987. English for Spcc$f;c Purposes: .I Lccirning-Centered ..lpprooch.Cambridge: Camhritlgc Uniwrsity Press.Hymes, D. 1972. ‘On communicativc competence.” In J. Pride antl J. Holmes, rds.,Sociolinntiistics, pp. 269-293. I-larmontlsnmth, England: Penguin.Johnson, R. K. cd. 1989. “A tl ion-making framework for the coherent languagecurriculum.’’ In R. K. Johnson, ctl., The Second Lungiiczge Curriculum, 1111. 1 23.Cambridgc: Cambridge Univrrsity Prvss.Kramsch, C. 1993. Contest onJ Cultrire in Langtiap Teaching. Oxford: Oxford Univcrsity Press.Larsen-Freeman I>. 1986. Techniyucs and Principles in Language 7iaching. Oxford: OxfordUnivrrsitv Press.Loiv, G. 1989. “Appropriate design: The internal organisation of course units.” In R. K.e Second I,unpige Curriculum, pp. 1 36-1 54. Cambridge: CambridgeNunan, D. 1985. Lungtiuge 7koching Course Detign: Eentls antl Irsuex. Adclaidc, Australia: NationalCurriculum Resource Centre.. 1988a. The Learner-Centred Currimltim. Cambridge: Caml)ritlgc Uniwrsity Press.-. 198811. Sjdluhtis Design. Oxford: Oxford University Press.-. 1989. Designing Eilsks fir the Corninunicatir.e Cluvcroom. Cambridge: Cambridge University1’1-Omaggio Hatlley, A. C. 1993. Teuching Lungtrugc in Context. Boston: Hcinlc antl Hcinlc.O’Malley, J. M., antl A. U. Chamot. 1990. Learning .Ytrute

196 I

Chaptev 16David NunanACTION RESEARCH IN LANGUAGEE DU CAT10 NIntroductionN THIS PAPER, I HOPE TO PROVIDE A RATIONALE for the uw ofI action research in recond and torcign language education Quc\tions addrcwd in thepaper includeWhat is action research in languagc education?Is action rcscarch 'real' research?What arc some of thr problems confronted by teachers doing action research?What are somc of the solutions to these prolilcms?What arc the views of the tcachcrs on the action research process?The palier 11 ill lie illustrated \z ith data from a longitudinal action rewarch project.Action research: description and rationaleUntil comparatively recently, the focus of concern in much of the writing on sccontl andforeign language cducation was at the level of method. Mcthodological prescriptions \vcrcgcncrally argued logico-tletluctivel>. and prescriptions fbr practice wcrc gcncrally devoidof data.'l'his tcndcd to reinforce the gap hctwccn theory, research and practice, a gap which,according to van Licr, is due in part to the obstacles which pre\ent teachers from doingrcscarch:Thosc of us \\rho hvork in teacher education knoxv that one of the most difficult thingsto lialance in a coursc is the tension between theoretical and practical aspects of theprofession. . . . Theory and practice arc not perceived as integral parts of a trachcr'spractical professional life. . . . This situation is the rcsult of communication gapscaused by an incrcasingly opaquc research technocracy, rcstrictiw practiccs incducational institutions and liureaucracics (c.g. not validating research time, or notgranting sabliaticals to tcachcrs for professional renovation), and ovcrhurdeningteachers \vho cannot conceive of\vays of theorising and researching that come out ofdaily work and facilitate that daily \vork.(van Lier, 1992: 3)

198 DAVID NUNANDespite the difficulties referred to by van Iier, thcrc is some evidence that the picture isbeginning to changc.The changc has Ixxn prompted in part hy a grokving sensitivity on thepart of many researchers to the complexities of the tcacher’s task. I’ractitioncrs, on theirpart, seem to have grown tired of the s\vings and rountlabouts of pedagogic fashion, antl arelooking for evidence before cmlx-acing the latest tt-end to appear in the educational marketplace. This is not to suggest that a revolution has taken placc, however.While position papers, and logico-tlcducti\ c argumcntation have not tlisappcared fromthe scene (antl I am not suggesting l~or a inomcnt that thcy should), thcy arecounterhalanced hy empirical approaches to inquiry. I believe that these days, \vhcnconfronted liy pedagogical questions antl I)roldcms, researchers and teachers are morelikely than \vas the case ten or fifteen !cars ago, to seck relevant data, rithcr throughtheir own research, or through the rcscai-ch of othcrs. IXescarch activity has increasedto the point whcrc those \Tho fa\mur logico-dctluctivc solutions to pedagogic problcmsarc lieginning to argue that thcrc is too much i-cscarch.(Nunan, 1992)An important concept underpinning action rcscarch (AR) is that ot reflcctilc practiceIn his exccllcnt liook on rcflcctne teaching, Wallace (1 991 ) argues that rcflectilc teachingpro\ ides a \T a! of de\ eloping prolcssional c otnpctcncc 11) integrating two wurces ofkno\i ledge, rcccxi\ cd knon lctlgc antl cxpcricntial kno\z lctlgc, \\ ith practice Wallace’sconccption I\ captured in Figurc 16 1Trainee’sexistingconceptualschemataor mentalconstructsExperientialknowledge‘Reflective cycle’COMPETENCESTAGE 1: (Pre-training)STAGE 2: (Professional education/development)GOALHe links this with action research, arguing that:‘action research’ can he attractivc for two reasons:1 It can havc a specific antl iinmcdiatc outcome \vhich can I)(- directly related topractice in the teacher’s o\vn context.2 The ‘findings’ of such rcscarch might lie Iiriniaril? specific, i.c. it is not claimcdthat thcy are necessarily of general application, antl thcrcforc the methods might hemorc free-ranging than those of conventional rcscarch.. . . ‘Kcsearch’ of this kind is simply an extension of the normal t-cflcctii e practice ofmany teachers, Iiut it is slightly more rigorous antl might conceivahly lead to morceffective outcomes.(Wallace, 1 99 1 : 56-- 7)As we can see from the selcctcd extracts presented ahvc, action rcscarch is justifiedon the grounds that it is a valuable professional dcvclopment tool. It rcpresents what I would

ACTION RESEARCH IN LANGUAGE EDUCATION 199call an ‘insidc out’ approach to professional development. It rcprcscnts a departure fromthe ‘outside in’ approach (i.e. one in lvhich an outside ‘expert’ brings the ‘good news’ tothe practitioner in the form of a onc-off workshop or seminar). In contrast, the inside outapproach begins with the concerns and interests ofpractitioners, placing them at the crntr-cofthe inquiry process. In addition to being centred in the needs and interest of practitionrrs,and in actively involving them in their own professional development, the inside outapproach, as realiscd through action rcscarch, is longitudinal in that practitioners arcinvolved in medium to long-term inquiryI believe that the Ixnefits to professional d lopment are justification enough for theent of an action research agenda. However, I believe that a further rationale forpmcnt of such an agenda comes from the research process itself, and I shall dealwith this in section three of my paper. First, ho er, I should like to look at the stepsinvolved in the action research process.Steps in the research processThe action rcscarch process is generally initiated by the identification by the practitioner ofsomething which they find puzzling or problematic. This puzzle or problem may, in fact,have emerged from a period of obscrvation and reflection. Thc second step is the collcctionof basclinc data through a preliminary investigation which is dcsigncd to identify what iscurrently happening in thr classroom without trying to change anything. Rased on a rcvicwyielded by the prrliminary investigation, an hypothc is formetl.The next steplopment of some form of intervention or change to sting practice, along witha way of evaluating the effects of this change.Thc final step is reporting on the outcomes ofthe interaction, antl, if necessary, planning furthcr intcrvcntions.T\vo examples ofthc actionresearch cycle are prcsentcd inTahlcs 16.1 and 16.2.Action research and ‘real’ rcsearchIn thc first part of this paper, I argucd that action research can IICjustified on professionalTable 16.1 ‘I‘hc action rcscaarch cyclc: an FSI exampleProb lcni / pur~lc +intlcntiticationIPreliminary +in\cstigationIHypothesis +IPlan intrrvrntion +Outcome +A teacher identifies a Imhlcni/purzlc. ‘MY students don’t sccniintci-cstrd or moti\ atctl.’IWhat’s going on7 Rccmrtling and oliscrving class over sc\ craldays.IContent doesn’t secm to stimulate students. Exclusive usr otdisplay questions.I . . .Increase UYC of rctcrcntial questions. Makc links bct\vccncontent antl learners.IMorc complcx intrractions. Morc in\ ol\ cmcnt and interestMorc ‘natural’ discoursc, c.g. students nominate topics,Ss disagree \vith tvacher, S--S interaction.IStaff‘tlc\ clopmcnt session.

200 DAVID NUNANTable 16.2 ’l’hc action research cvc,lc: a hrcign language cxainplcI I’rohlcm + A tcachrr idcntifics a prolAm in hcr classroom. ‘My stutlcntsitlcntiticationIaren’t using thc targct language [(;rrman].’2 I’rcliminary -+ What’s going on? Ilcccirding and oliacrving class ovcr scvrralin\ cstigationIin English.4 Plan intc-rvcntion + Tcachcr incrmscs target Ianguagc iisc.ITcachcr usrs (kriiian hr clasarooni managrmrnt ctcIdcvelopmcnt grounds. Hout AK can also be justified on research grounds.In fact, I believe that there is sntially patronising in the vimv that, Mhile ARmight be good for profcssiona , it hardl? counts as r arch. [,et us, to use acurrently fashionable term, ‘ his vie\v. First of all, \\.hat (lo \\re mean by‘research’? What is the functiElscw-here, I havc defined research as ‘a sjstcmatic proc ’ of inquiry consisting ofthreeelements or componcnts: (1) a question, Imhlcni, or hypc csis, (2) data, (3) analysis ant1intcrprctation of data’ (Nunan, 1992: 3). Action research incorporates these three clcmentsantl thercforc qualifies as ‘rcal’ 1-csearch. For mc the salient distinction hctwecn AK andothcr forms of research is that in .4K the I- arch process is initiated and carrictl out I,>. thepractitioncr. As l’ar as I am concerned, the opposition is not lict\vccn action research and‘rcal’ research, but Iiet\vc.cn good rcscarch and bad rcscarch. A further characteristic,pcrhaps differentiating AR from othcr forins of practitioner rescarch, is that it incorporatesan clcmtwt of intervention antl change.Fundamental to any discussion of‘ research is a consideration of the rcscm-cher’sconception of notions such as ‘truth’,‘ohjcctivity’, and thc status of kno\vlcdgc. I recentlyattemptctl to deal \vith the tensions of objective antl subjective kno\vletlgc by suggestingthat they represent t\vo altcrnati\.c \vays of looking at thc world:T\vo ah-native conceptions ofthc nature of research provide a point of tension withinthe book.Thc first view is that external truths exist ‘out there’ somewhere. Accordingto this view, the function of rcscarch is to uncover these truths. The second vicw isthat truth is a ncgotiablc commodity contingcnt upon the historical context within\vhich phenomena arc observed antl interprctcd. Further ‘[rescarch] stantlards arcsubject to change in the light ofpracticc [lvhich] \vould sccm to intlicatc that the searchfor a substanti\c universal, ahistorical nicthotlology is futilc.’ (Chalmcrs 1990: 2 1). . . This second, context-bound attitude to research entails a rather differcnt role forthe classroom practitioncr than the first. If knmvledgc is tentative antl contingcnt uponcontext, rather than alisolutc, thcn I liclicvc that practitioners, rather than beingconsumers of othcr pcoplc’s rcscarch, should adopt a research orientation to theiro\vn classrooms.There is cvidcnctx that the tcachcr-researcher movement is alive and

ACTION RESEARCH IN LANGUAGE EDUCATION 201mcll and gathering strength. Howcicr, if the momentum mhich has gathered 15 not tofalter, and if the teacher-researcher molement 17 not to becomr >et another fad, then5ignilicant numbers of teachers, gracluatc students, antl others \z ill ncctl skllls inplanning, implementing and e\ aluatmg research(Nunan, 1992)There are thosr hvho \vould arguc that my definition of research as a systematic processof inquiry involving formulating a question, collccting relevant data, antl analysing antlinterpreting that data is inatlequatr, that in order to count as research, the process shouldalso meet the twin structures of reliability antl validity. Key questions for establishing thercliahilitv and validity of research arc set out inTable 16.3.Tuh/e 16.3 Questions for establishing thr rcliahility and validity of a stud!‘Ijpek’cy YllCStlOflInternal reliabilityWould an intlepcntlcnt researcher, on re-analysing thedata, conic to the same conclusion?Extcrnal reliabilityWould an indcpcndcnt researcher, on replicating thcstud\, come to the same conclusion?Intrrnal validityExternal \alidit>Is the rmcarch design such that \vc can confitlcntl> claimthat the outcomes arc a result of thr experimental treatment?Is thc research tlcsign such that \z c (‘an grneralisc Ixyntlthc subjects untlcr inkcstigation to a witlcr population?Soirrcc: Nunan. 1992While I would argue that any rcscarch needs to tic rclialde, the issue of validity is moreproblematic. If one is not trying to establish a relationship Iwt\vcen variables, hut (forexample) to dcscribc and interpret phenomena in context, does the imperative todemonstrate that one has safeguarded one’s research from thrcats to internal validity rcmain!By the same token, if onc is not trying to arguc from samples to populations, then it wouldnot be unreasonable to a, rt that external validity is irrclmant. I would argue that as mostAR is not concerned mith arguing from samples to populations, external validity is not atissue. (For an cxccllcnt discussion of issucs to do nith reliability and validity in qualitativercscarch, see LeComptc and Goctz, 1982.)It is popularly assumed that the purpose of research is to test theories. For examplc,‘That communicative language teaching is more Ffective than audiolingualisrn.’ Allwright and Baileyhaw pointcd out that there arc problems mith this proposition. In the first place, sometheories are untestable (for example, Krashen’s attestations on ‘subconscious’ acquisition).Secondly, classrooms are too complcx for us to control all thc variables in the mannerprescribed by experimental research. Thry propose an alternativc purpose for research,namclj to try antl understand and deal with immcdiatc practical problrms facing teachersand learners (Allwright and Bailey, 1991). If \vc accept this alternative purpose, MC art’drawn immetliately into embracing AR, because it makes no sense for an outsider to arhitratcon the practical problems facing teachers and learners. This does not mean that outsiders,

202 DAVID NUNANsuch as unhersity-based researchers, hahe no role to play in practitioner-based research.However, the role is one of collaboration and ad\ icc rather than direction and control.Problems and solutions in doing action researchI would now like to reassure those who might feel that I am looking at teacher researchthrough rose-colouretl glasses. It is certainly not the case that everything is rosy in the AKgarden.The principal problems identified liy teachers with whom I have Morked in a numberof different contexts include the following:Lack of timeLack of expertiseLack of ongoing supportFear of being rclvealcd as an incompetent tcacherFear of producing a public account of their rcscarch for a wider (unknolzn) audienceWe have cxperimcntcd tvith a number of solutions to the problems. I believe that thcchances for an action rcscarch agenda to succccd will be maximiscd under the followingconditions:There is somcwne ‘on thr ground’ to ‘okvn’ the project.One or more individuals with training in research methods arc availablc ‘on tap’ toprovide assistance and support to tcachcrs.‘Teachers are given paid relcasc time from face-to-facc teaching during the course oftheir action research.Collaborative focus teams arc established so that teachers involi et1 in similar areas ofinquiry can support one another.Teachers arc given adequate training in methods antl techniques for identifying issues,collecting data, analysing and interpreting data, and prcscnting the outcomes of theirresearch,In order to facilitate the process, colleagues antl I haw del elopcd an in-serviceprogrammc.This programme \vas initially devised for the IdPT project (Languages InserviceProject forTeachers) in South Australia, antl has bccn further moditietl and refined in Sydney,where a project has bccn establishctl hinging together mainstream teachers, ESL teachers,and teachers of LOTE (Languages Other than English). InTalilc 16.4,l have provided asummary of the professional development programme as it currently exists.Evaluating action researchFrom what has already Iwen said, it is clear that action research is difficult, messy,protilematic, and, in some cases, inconclusive. It consumes a grcat deal of time, and oftenstrains the goodwill of the teachers invol\etl, as well as those with whom they work.Ho\vever, evaluative data from teachers themselves suggests that teachers who have hccninvolved in action research are overwhelmingly in favour of it. For example, Mickan, whocollected data on the reaction of outside teachers to his LII’T project, writes:Teacher4 hahe welcomed the article4 from LIPT. The) ha\e found them particularlyuseful and relaant hcxausc the! depict the complex circumstances of classroom life

ACTION RESEARCH IN LANGUAGE EDUCATION 203in an honest and direct way.They have found them a rich source of ideas and valuablefor informing their own practice. Thc warts and all descriptions (including failuresand successes), thc research techniques used, the analTsis of' results and the contextualdetail are all rlcments hvhich readers relate to and understand. As such they posscssa validity which derives from the detailed narration of classroom ecology. Thecxpcriential reports giw other practitioners models and ideas for their ow-n practice.They also suggest topics and procedures for classroom investigations in diffcrcntcontexts.(Mickan, 1 99 1 )Table 16.4 The insen icc programme in outlineSession IaI)An introduction to classroom olxc.rvation and rcscarchA series of reflective activities dcsigncd to gct tcachcrs thinking about their o\vn teaching stylr.Rcllccting on the teaching of others: teachers examine anti critique cxtracts from a range ofclassrooms identifying those aspects olthc extracts thcy liked and tlislikctl.c Identification of ideological beliefs and attitudes underlying critiques.Rctwccn session task: teachers record and rcllcct on their o\vn teaching.Session 2 An introduction to action researcha Teachers I-cport back on the hetween wssion task.I> Introduction to issues and methods in action rrsearchcIntroduction to thr action research pro'Session 1 Focus groupc antl action plansaI-,cRefining qucstions.lop a draft action plan.Brt%vc.cn session task: baseline ohscr\ation, focus group meetings, preliminary data collection.Session 4 Analysing dataaIictwcrn session task: ongoing data collcction antl analysis, focus group meetingsSession 5 Writing upahFormation of focus groups antl appointment of facilitators.Sharing of draft action plans.Participants devclop ways of analysing and making scnsr of thcir (lataI'articipants rcccivc input on prcscnting their research.Development of draft rrporting outlines.tlctwccn session task: production of draft reportsSession 6 Rching reportsParticipants rcccivc fccdback on anti discussion of their reportsSrssion 7 EbaluationParticipants c\ aluatc thc LIPT process and provide fccdback on how their involvement changcd them.An evaluation by Lewis (1 992) is also favourable. She reports on a study conductedwith a group of teachers of French immersion programmes in British Columbia.The focusof hcr research was the effect on the professional practice of the teachers of engaging inAK. She drew the following conclusions from her research.

204 DAVID NUNAN1 ’l’hrough thc pro of systematically iniplcmcnting thcir o\vn choice of action projectf the students in particular, each tcachcr learned more about thciro\vn thcories, or frames for teaching, antl motlifictl thcsc frames to a ccrtain extent.2 The frames for tcaching of the participants in this study arc related to the biggerquestions of second language education antl education in general. Practice cannot lieunderstood thoroughly nithout aplircciating how educational theory is expressedlvithin teachers’ frames and neither can theor? bc uscful without rccognising that\\hat counts is how theory Ixwmies cxIircsscd \Tithin practice.3 The ‘tcachcr as r archer’ or ‘rdlcction in action’ approach to tcachcr education canlie a \cry powcrtul \vav of facilitating changc in curriculum.In elaluating the last ofthc I IPT proicct\, \\e asked teachers to complete the folio\\ ingstatements:Action research is ................................................................................Action research is carried out in order to ...................................................Wc also a\ked thcrn to respond to thc folio\\ ing:I23What are the most significant things you have lcarncd in carrying out your classroomresearch?What qucstions/issucs has your classroom research raised for vou?What further arcas/itlcas arc you interested in pursuing?Sample responses to the til-st of thcsc prolxs on the most significant outcomes for theparticipants arc set out in Appendix 1. It can be sccn that thcsc are ovcrwhelminglyfavourable, the participants choosing to focus either on the sulistanti\c content outcomes(‘By collecting antl analying data on ni! children, I found that they wcrc more highlymotivated than I had given thcm credit for’), learning procrss outcomes (‘The activeinvolvement of the children in thc learning process facilitates learning.’ ‘I discovered thatkids know ho\v to learn ~ the project taught me to listcn to thcrn’), or reflections on theresearch process itself (‘In norking through the action research ~iroc‘ess, I discovcrctl \vhichmethods of data collection arc most suited to my research question ~ next time I will hclietter prcparctl as I \vi11 lic more awarc of\vhat I am looking for, and \\ill be hettcr able tomatch my questions antl data.’). The enthusiastic validation of learner-ccntrcd approachcsto instruction, cvcn though this \\‘as not a primary aim of most research, is also Lvorth noting.Finally, participants \\‘crc asked to complete a chccklist to indicate how thcir trachinghad changed as a result ofthcir involvement in the projcct. Iksults are set out inTablc 16.5.It can Iic sccn from thc sur\.ey that, if sell’-rcports are to lie believctl, the expericncc \vas,for most teachers, an over\\ hclmingly positi\.e onc.ConclusionIn this Ijalier, I have argued that the atloption of an action research orientation can bc justifiedlopmcnt terms antl rcscarch terms. Despitc thc Iw-caucratic difficultiesand obstacles Ivhich arc placed in thc \vaT of tcachcrs, thc clitism of a certain cadre ofresearchers (somc of 1% hom ~vcrc once classroom teachers themselves!), and the suspicionwhich is sometimes directed at academics Lvho arc trying to promote a closer rrlationship

ACTION RESEARCH IN LANGUAGE EDUCATION 205‘ruble /6.5 How has your tcaching changeti? Complete the follo\\ing: ‘Since I havr hecn doing actionresearch, 1 tint1 that \\ hen I tcach I no\\ . , .’I234567891 01112131415Mot-c Abouttcntl to be &recti\ ctry to itsc a greater \aricty ofI)cha\ iourspraise studentscriticise studentsam axvarc of studrnts’ teeling.\gi\ c directionsani conscious of mv now\ crbal communicationuse the targct language in classam conscious of non-verhal cuc\ of studentstr\ to incorporate student itlcas into my teachingspend more class time talking mysclf116150184111912201the ,amc1461011h161461259try to gct students \\-orking in groups158ti-! to get tiivcrgcnt, open-cndcd student responsesdistinguish I)ct\vc.cn cnthusiasm and lack of oi-dcr14910I5tr\ to get \tutlcnt\ to participate 18 7LtY\100013050000150000bet\ccen theory, rcscarch, antl practice, there is c\ idence that things are beginning to changeI can otter no more fitting conclusion to this paper than the follo\sing extract from the vorkot tno ot the prolession’s foremost adlocates of the dc\clopmcnt ot hai-monj bct\\ccntheor), rcwarch and practice, ~ hhac o strixen in thcir OM n teaching, Lcriting antl research,to enhance thc status ot both practitioner and r archer \\ ithin language cducation\lo\\ I>, thc profcs5ion as a \\ hole is reahsing that, no matter hou much intellectualcnergj is put into the inicntion of ncv methods (or of ne\\ approaches to s!Ilahusdcsign, antl so on), \\hat rcall; matters is \\hat happens 1% hen tcachcrs antl learnersget together in the classroom This 4hitt in emphasis from concentrating onplanning decisions to concentrating on looking at M hat actuall) happens in thecla\sioom, has led r archers to hac much greater respect for claw oom tcachingI he more me look, the more \\e find, antl thc moic \\e rcahw hov complex theteacher’s 101) is And tcachcis, in thcir turn, faced at last 111th rrsrarchcrs mho ha\cat least some itlea ot the cnoi mous complexit) ot e\erjdaj clas\rooni litc, archeginning tu he more r pti\c to thc \\hole research enterprise Bcing a gootlclassroom teacher means being all\ e to \z hat goes on in the claswoom, all\? to theproblems ot rorting out \\hat matters, momcnt bj momcnt, from \I hat tlocs not Ant1that 15 \\hat classroom rcscarch is all about gaining a better undcrstanthng of \\hatgood teachers (and lcai ncrs) do instinctncl; as a matter of course, so that ultiinateljall can hcncht(Alhright antl Railej, 1991)Appcndix 1What are the most significant things jou ha\c lcarned in carr)ing out !our cla\woomresearch!The active inyol\ emcnt of the chiltlrcn in the learning process facilitate5 learning.

206 DAVID NUNAN............Children have differcnt learning prcfcrences antl teachers nccd to allow for this intheir instructional practices.Children find it difficult to express feelings and opinions on papcr.It is casy to ‘spoon fectl’ children, but this leads to ineffcctive learning.Teaching problems only go alvay if they arc recognized and tackled.The most important outcomc for mc was that I learnctl how to do action research,To benefit, I thcreforc have to do it again!Working with the children together (c.g. finding their thoughts/feelings and actingon them).In working through the action resrarch process, I discovered which methods of datacollection are most suited to my research question ~ next time I will be bctterprepared as I will be morc aware of what I am looking for, and will be bctter ablc tomatch my questions and data.The process removcd my tunnel vision to teaching.It helped me to make links xvith other teachers of Mandarin, as well as parents antlthe community.The proccss dramatically cnhanced my rapport with students.I found that by careful, step-by-step direction of students, I was able to give themtools to manage their oivn learning.Ry collecting and analyzing data on my children, I found that they were morc highlymotivated than I had given them credit for.The most important outcome for me was that I tliscovered the children enjoy (andrespond well) to bcing consulted about their learning and tieing given some say inwhat they learn.There \vas a negative outcomc for me I’ve learned not to expect children tohavc complctcd tasks or to value something just because they’re important tomc.I found thatyear 7 learners still need lots of structure and guidance, even whenindependent skills arc cncouragcd.I was disappointed. I expected too much in my initial project ~ book flood! Only booktrickle is possible in such a short time.The most important discovery for me \vas that my students nccd morc time andopportunities to work in groups as they need to learn to work on their own withoutteacher directed lrssons all the time.The need for informed input in this proccss ~~ one ncctls to read ctc., recent researchand thinking in order to maximize value of one’s obvn research, and move beyondone’s own ‘lilinkercd’ vision.The positive bcnefit of concentrating on one particular area because the attitude/approach of opennrss and inquiry carries ovcr into one’s teaching in gencral.I have learned that students with a very limited knowledge of the target language areprepared to try to hvrite morc than I expcctcd, and that in future I should try to fosterthis willingness in my classes.Contrary to my exprctations, I found that the chiltlrcn lverc keen to be part of a‘project’. This led to increased motivation (maybe Hawthorne Effect?).I have learned that one ncctls to undcrtake classroom research. One needs to interveneobservation alone isn’t a good enough indicator of how much children areIcarning.In my research, I tlelvctl into how my lessons were arranged and the effectiveness (ornot) of my teaching. I looked closely at my learning strategies. It allowed me

ACTION RESEARCH IN LANGUAGE EDUCATION 207to construct a unit that mas tlesigncd for junior primar: 5tudent5’ needs and intercstsand m) research allowed m> to construct strategic’s accordinglj.I disco\cred that kids kno1.r ho\t to lcarn ~ thc project taught me to listen to thcmAcknowledgementsGrateful acknondcdgcment is made to thc British Council for financial support. Figurc 16.1,‘Reflecti\-e practicc model of professional cducation development’ from Training ForeignLanguage Teachers:A Rcjlectlve ,4pprouch (M. Wallace, 1 99 1 ), is reproduced hv kind permissionof Cambridge University Press.ReferencesAllwright. D. antl K. M. Bailey. 1991. /,OCLIS on the Language Classroom. Cambridge: CUP.Chalmers, A. 1990. Science and its Fabrication, 21. Milton K nes: Open University Press.LcComptc, M. antl J. Goet7. 1982. ‘Problems of rclial it? and valitlity in cthnographicarch.’ Review ?fEducarional Research, 52/ 1 .Le\vis, C. 1992. Action research with French immersion teachers: a pilot study. Unpuhlishcdmonograph, University of British Columbia: Canada.Mickan, 1). 1991. LIPT: Languages lnscrvice Program for Tcachers Stagc 3 1990. ActionIcesearch Keports Volume 6, March I991 . Adelaide: Languages and MulticulturalCentrc.Nunan, D. 1 992, Research Methods in Langiiage Learning. Cambridge: cui‘.Van Lier, I.. 1992. Not the nine o’clock linguistics class: investigating contingrncy grammar.Unpublished monograph, Monterey Institute for Intcrnational Training, Montcrey :California.Wallace, M. 1991. Training Foreign /.anguage Eachers:A R$ecrive,Ipproach.Cambridge: CUP.

Chapter 17Susan FeezCURRICULUM EVOLUTION IN THEAUSTRALIAN ADULT MIGRANTENGLISH PROGRAM1 Introduction1.1 The Adult Migrant English ProgramHIS CHAPTER TRACES THE DEVELOPMENT OF curriculum and syllabusT design in the Australian Adult Migrant English Program (AMEP). The AMEP is anEnglish-languagc programme offered by the Australian government to all immigrantsof non-English speaking background. Many people believe that the AMEP has hccn animportant elcmrnt in the succcssful integration of the thousands of peoplc from diverseInckgrountls 1%-ho have migratctl to Australia sincc thc Sccontl World War.1.2 Fifyyearr of curriculum developmentOler the fift) )car\ ot the AMEP, teachers hac interpreted dc\clopmcnts in appliedIinguistiLs in order to customisc the curriculum to thc need5 ot non-tnglish 5pcakingimmigranta '1 heir intcrprctation5 ha\ c rcflcctcd thcii- bchcfs about language and languagrlearning, both conscious and unconscious Thew I,clict\ ha\c shaped the \x a) tcachera in theAMEP ha\ e di\ idcd language up into chunky ot content and then sccIucnLcc1 the wntrntinto claw-oom act11 ities L)c\clopmcnt\ in applied linguistic5 olcr the last hft) ycars hacresulted in three distinct \xa\es of tcac hing approa~hcs in the AMFP These arc.I structural approache\2 lcarncr-centrccl, ne bad, conimuniati\c approa~he\3 trxt-bascd approaches2. I Structural approaches2. I . 1 The origins ofstructural (ipproachecAt the bcginning of the tlvcnticth century the learning of a foreign language in formaleducational settings was limited to the privileged fen,. Students learnt the language bystudying grammar rules and using thcsc rulcxs to translate literary texts, a method known

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 209as grammar-translation. As the century unfolded antl more people had the opportunit? orneed to trawl, there was a demand for approaches which taught people how to communicatein a wider range of contexts nith speakers of other languages. Applied linguists in Britainantl Amcrica responded to this demand in different ways. British applied linguists developedsituational language teaching, lvhilc in America audiolingual methods \\ere dan account of these devclopments scc Howatt, 1984; Richards and Rotlgers, 1986).The AMEP looked to British situational language teaching, \vhich linked structures ‘tosituations in which they could be usctl’ (Richards and Rodgcrs, 1986: 35). British situationallanguage teaching had emcrgctl becausc a group of British linguists, in particular Firth andHalliday, \yere rxploring holv structure and meaning were relatcd to contrxt and situation(see Richards antl Rodgers, 19x6).2 I 2 Srr irctirrcil upproat he\ and hehai roural eclucatronal py rhologTeachers using situational/structural approaches taught Icarncrs the component parts oflanguage. They used tcchniques dcvclopetl liy hchavioural psychologists to teach ‘correct’language habits and accurate forms. Words and structures \vci-e taught in a fixed sequencethrough response, rcpctition and mcmorisation using, for example, imitation drills andsubstitution cxcrcises.Situational language teaching \vas used in the AMEP becausc non-English~speaking migrantsneedcd to be alile to use ‘real-\.vorld’ English as quickly as possible on arrival in Australia.All learners in the AMEP mw-e taught the samc dialogues in a fixed sequence from a commontextbook. These dialogues \vert ‘situated’ within an everyday ‘setting’ such as a restaurantor a railxvay station. The dialogues introduccd lexical items antl grammatical structureswhich were then practised by the learners in follow-up activities. The dialogues used at thetime no\v seem very contri\-ccl and inauthentic nrvcrthcless, they drew learners’ attentionto language use inThe linking of structures to situation in the AMEP curriculum was an earl) examplc ofthe close link \vhich continues to this day between the academic discipline of applicdlinguistics antl the de\ elopmcnt of curriculum and expertise in the AMEP When comparcdwith equivalent language learning approaches of the day, thc situational approach stands outfor its responsiveness to the needs of learners. Through this syllabus AMEP trachers \vcrctlevrloping expertise in linking thr language learnt in the classroom and the languagelearners nccdcd to use in rcal life. This expertise became the foundation on which futuretlevclopmcnts in AMW curriculum were based.Other innovations during this period no\v taken for granted everywhere asTESOL Iwstpractice includc:a concern with all the macroskills of language ~ listening, speaking, reading and writingthe usc of thcmcs antl topics as a basis for coursc designclassroom management stratcgiex in\ olving clicitation techniques, pair antl grouplvorkpresentation and practice techniques \vhich incorporated rcalia, concrete and visualmaterials, gcsturc and mimethe identification of teaching ohjcctives in tcrms of \\.hat thc Icarncr should bc ableto do with English at the enti of the course.

210 SUSAN FEEZ2.1 .-! The challenge to structural upfr@i~he~Despite the innovations outlined almvc, t.hc situational approach retained the follotvinglimitations:Language forms \vert learnt in isolation antl in a fixed progression irrespective of thelearner’s necds and goals.Language learning was f,roduct-orientecI, teacher-centred, concerned with accuracymore than fluency, antl atomistic, that is, concentrating on individual isolated phrasesand structures.Grammar and vocabulary \\ere taught in isolation from thc \vay language \vas used inreal life situations (sccYaltlen 19873).2.2 Leaner-centred, needs-based communicative approachesFrom the end of the 1970s the AMEP lqan to move akvay from a ccntraliscd, structuralapproach towards an indivitlualiscd syllahus in \\-hich classroom teachers were responsiblefor syllabus dcsign. Teachers constructed their syllahuscs from a diverse repertoire ofsyllabus elements and methodologies. These can be roughly grouped according to whetherthcy were informed by:12second language acquisition antl progrcssiw pcdagogiescommunicative and social theories of languagc. antl language learning.2.2. 1.\aturn1longtiage learning i7nd progrewre pcclqogie.sApproaches Lvhich were dcscrild as more ‘natural’ lvays of lcarning a language cmergedafter thc American linguist Chomsky claimed that language use was ‘not imitated bchaviourbut . . . creatcd anew from untlcrlping kno\vlctlge of alxtract rules’ (Richards and Rodgers,1986: 59). Following Chomsky, sccontl language acquisition (SLA) theorists began todescribe language learning as a proccss in lvhich learners actively test their emerginginterpretations of the new language.Influenced by SLA, AMEI’ teachers began to understand that a learner’s non-standardapproximation of the target languagc,, or interlanguage, was not mcrelp ‘incorrcct’ but ratherrevealed how the learner \vas progressing (Corder, 198 1 ; Sclinker, 1991). AMEP teachersfelt that, if language learners were in an environment rich in language input just beyondwhat thcy 1vei-e able to produce themselves in a stress-frcc cnvironmcnt, they would acquirethe target language unconsciously, effortlessly and flucntly (Krashcn, 1 988).A stress-free learning environment m-as achicved by drawing on the progressivepedagogics which had emcrgcd in Wcstcrn education by the end of the 1960s. Progressivcpcdagogies cncouragcd teachers to abandon their traditional authoritarian role in ordcr to:tlcvclop more equal antl respectful rclationships with learncrsfacilitate humane, interesting and interactivc educational settingsrecognise and respond to the individual needs, interests antl motivations oflearnersencourage learners to takc responsibility for their omm learning, to takc risks and todiscover knowledge as they need it.AMEP teachers were especially concerned with learning principles appropriate to adultsincluding, for example, self-directed antl contract learning (Kno\vlcs, 1 990). Some AMEP

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 211teachcrs also became intrrested in approaches which highlighted social justicc, and politicaland personal freedom (for example, Freire, 1972).By the end of the 1 980s many AMEP teachers had developed the following Iiclicfs aboutlanguage learning:Fluency is more important than accuracy.Intervention in thc learning process is countcr-productive because it diminishes theindividual's motivation, self-expression and personal development and takesrcsponsibility for learning avay from the learner.Learners should 'own' thcir learning so teachers should negotiate the curriculum withlearners based on learner need.Classroom materials should he based on authentic languagc use and tcachcrs shoulddra\v on a wide range of methotlologics.--- -. -.-,.______I. ', -' ..r :,.(. , * -7"' ,.,/ ,"

212 SUSAN FEEZstretches of meaningful language, as \vel1 as in tcrms of grammar and vocabulary (Ingramand Wiley, 1984).As the 1980s progre d, AMEP teachers Iiccamc increasingly aware of criticism beinglevrllcd at inventories of notions antl functions, for example hy Candlin (as cited in Mclrose,1991 : 19- 20) and Widdo\cson (1979: 248 as quoted in Yaldcn, 1987a: 77). These criticspointed out that, although these \vcrc lists of units of meaning rathcr than lists of words andstructures, they remained inventories of isolatctl components and so were of questionablevalue in de\ eloping communicative competence.2.2.3 The challenge to learner-centrcJ, ncctlc-hacd comrnrrnicc7ti1.c approachesBy the mitlL1980s, for each new class they taught, AMEP teachers \vere cxpcctcd to:analyse the nccds of learners (13rindIcy, 1984)negotiate language learning objcctivcx with thc learnersuse functional-notional inventories to select antl sequcncc syllabus elementsimplement learner-ccntretl, communicative methodologies based on self-dircctcdlearning antl the development of fluencyprovide opportunities for tlcvcloping thc skills antl strategies learners needed if theywerc to become effective listeners, spcakcrs, readers and writcrs.In practice the task of designing coherent courses from all these componcnts proved to brvery difficult. As teachers tried to tlcvisc situations in which the studcnts could practisedifferent bvords and structurcs in ‘natural’ contcxts of use, fluency \vas cmphasised at theexpensc of accuracy. Tcachcrs tended to sclcct and sequence syllabus elements accordingto their perception of xvhat ‘worked’ in the classroom, forgetting that communicativepractice alonc \vould not dcvclop communicative competence. As Hymes had pointed outfrom the Iieginning, learners also ncctlcd to dcvclop kno\vledge ahout language.By the end of the 1980s AMEI’ tcachcrs hvere grappling with ‘the complexities ofdesigning intlivitlualiscd programs lvith little institutional support’ (Brindlcy and Hood,1990: 223 -4). A consequence ofthis was that AMEI’ learners had little sense of progressionor achievcmcnt (Brintllcy, 1985; Nunan, 1987). To ovcrcomc these difficulties, teachcrsrequested ‘non-mandatory curriculum guidelines to assist thcm in planning courses to meetthe nccds of commonly recurring Icarncr groups’ (Nunan, 1987: 59).One commonly recurring learning group in the AMEP wcrc ‘stabilisecl’ lcarncrs nrhohad achieved a non-standard intcrlanguage xvhich allo\ved thcm to survive in Australia.Thesclearners, howe\rer, did not progress to more standard forms of English. Ikcausc teacherswere focusing on fluency in classrooms and were trying not to intcrvcne in languagelearning, learners were interacting with each other, not with the teacher. Consequently thcopportunities for learners to engage with stantlard English wcrc limited and theirinterlanguage hecamc established as the means of communication. Non-standard forms ofEnglish limited students’ opportunities in Australia, especially in further cducation andemployment.AMEP curriculum (Icvclopcrs realisctl that it \vas tiinc to 1-cvisit the idea of plannedintervention in the process of language Icarning, but this time from thc persprctivc of:an interactive model of grammar and discourse, one that demonstrates the necessityand importance of both levels of languagc to the languagc learning process and to theattainment of communicative competence.(Celce-Murcia, 1990: 146)

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 213Developments in the wider field of English-languagc teaching wcre supporting this shift.Widtlo\vson (1 990: 164), for example, strongly challcnged the ‘natural’ approach tolanguage learning.The early 1990s initiated a period of rapid social, economic and industrial change inAustralia. AMEP learners ncctlctl increased levels of English-language proficiency if thcywere to access community services, find work or participate in further education in thischanging environment. If their English-language skills were to bc recogniscd in this environment,AMEP learners would have to demonstrate their language skills in tcrms of explicitoutcomes of a rccognised curriculum (for an account of this period, see Hagan, 1994).The \\ riters of the AMEP curriculum were faced with sevcral challengcs.’I’he needs ofAMEP learncrs in the changing social environmcnt tlemandctl that the AMEP curriculumcomprise:a flexible framework to accommotlatc thc diverse ncctls of learners across a \viderange of tcaching contexts across Australiaa clcar learning pathwa?a common language for describing learner ncds and goalsexplicit statements of what learners could do at the end of each stage of the pathway,that is, explicit outcomesa communicative approach which reflected real-life language use in tcrms of discourserathcr than isolated elements such as words, structures, functions or notionsan explicit and tematic approach to the tcaching of language structures antlfeatures.ro meet the challcngey, the AMkP curriculum tlc\elopers again turned to the vork olIlallida) They began with Hallida>’s itlea of a text being a nholc unit of language useWorking mith \\hole texts in real contexts of language use, AMEP learners nould hac theopportunit) to dc\elop communicati\ c fluenc) as well as accuracy in tc-rms ot tcxt structure,grammar, lexis and surface features such as pronunciation and spelling The 4MEPcurriculum tic\ eloper? also turned to a petlagog> tle\elopcd ti) Hallida! ’$ collcagues inAustralia Thi.; pcdagog?, the genrc approach, pro\ ided a model for explicit intcxr\ention inthe process of languagc learning2.3 Genre-based approaches2.3.1 Language ilc tcvtThe foundation ofthc gcnre approach is thc study ofn-hole tcxts in contcxt. A tcxt is a unitof discoursc (spoken or written) in which related meanings are \vovcn togcthcr to make aunified whole hvhich achicvcs a social purposc (see Halliday in IIalliday and Hasan, 1985:10). Halliday has shown that there is a n a tcxt and thc contcxt inwhich it is used. Systcmic functional ling s explore register variation in language; in other\vords thcy explore the 1% tcxts \ ary from social situation to social situation. Considcr,for example, the diffcrcnn the following tcxts:a telephone Id1 antl a lcgal billan exchange bctm n a parent and a child anti an exchange bct\vcen a tcacher and astudenta story pulilishcd in a book antl a story told in casual conversation.

214 SUSAN FEEZAt the same time they look for the underlying patterns in texts which remain relativelystablc in order to achieve similar purposes across registers and which makc texts culturallyand socially meaningful.2.3.2 The genre approachA group of educators tlra\ving on systemic linguistics developed a pedagogy based on thenotion of‘genre’. (For more detailed accounts of the genre approach in schools, see Cope,Kalantzis, Krrss and Martin, 1993; Krcss, I99 1 ; Martin, 1993; Rothery, 1996). A genre isa relatively stable pattern which recurs in tcxts used to achieve the same general socialpurpose. For example, linguists are able to describe patterns which recur when people uselanguage to:build a relationship through casual conversationrecount a series of events to share what happened with someone elseentertain liy telling a suspenseful story or an amusing anecdoteexplain how somcthing worksgive instructionspersuade someone to your point of vieworganisc informationmakc a story newsworthy for the electronic or print media.This information is vcry useful to language teachers and learners because it makes learningto use language a much less hit-andmiss affair. It makcs it possible to identify what peopleneed to lie able to do with language in order to be successful in education, in the communityand in employment. Learners work with individual tcxts which excmplify different genresin order to learn:the overall patterns of different genres of communicationspecific language features used in examples of that genre most relevant to theirindividual learning needs.Genre pedagogy is a teaching approach in which students work systematically andcomprehensively with the language of texts belonging to the genres relevant to theireducational goals. Genre pedagogy is usually presented to teachers in the form of a cycleof teaching and learning. (For a diagram and tletailcd description of the teaching-learningcycle initially used in schools, see Kothcry, 1996: 101 4.)2.3.3 A language-based theory of learningThe cycle of teaching and learning designed to teach about texts reflects Halliday’s (1 992:19) view of learning as a process of ‘learning to mean and to expand one’s meaningpotential’. Research into first language dcvclopment hy Halliday and his colleagues revealedlanguage learning to lie a social process (for example, Halliday, 1975; Painter, 1985; 1996).Halliday (1 991 ) describes educational learning as an organised social process in which theconstruction of meaning takes place systematically.2.3.1 Scaffblded learningIn a genre-based cycle of teaching and learning:Lvhat is to he learned and assessed is made explicit to studentstcacher-learner interaction is valued as much as interaction between learners.

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 215The design of the gcnrc-based teaching-learning cycle draws on the theories of the Russianpsychologist Vygotsky (1 934/ 1978; 1986). Vygotsky’s work suggests that instructionpreccdcs learning. Drawing on both Halliday’s and Vygotsky’s ideas, the genre approach isused to construct a social context in which languagr learning can occur. In that context:teachers and learncrs collaboratetcachers interact with learners to guide them towards their potcntial level ofpcrformancethe ttxchcr’s role is an authoritative one similar to that of an cxpcrt supporting anapprenticelanguage is used interactively to mediate learning.The genrc-based cycle of tcaching and lcarning has two key characteristics:scaffoldingjoint construction.Scaffolding occurs when the teachcr contributes what learners arc not yet able to do aloneor do not yet know.Teachers adjust, and strategically diminish, thcir contribution, supportinglearncrs as they progress towards their potcntial level of independent performance.Joint construction occurs when tcachrr and learner share the responsibility forfunctioning until the lcarncr has the knowledge and skills to perform independently andwith sole responsibility. (For further discussions ofscaffolded learning scr Gray, 1985, 1987,Gibhons, 1998 .)The genre-based teaching-lcarning cycle was initially del elopcd for primary schools,but it has been adapted for ESL (see Burns, Joyce and Collin, 1996: 88; Cornish, 1992: 17;Hammond, 1989; 1990; Hammond, Burns, Joyce, Rrosnan and Gerot, 1992; Hood,Solomon and Burns, 1996; Joyce, 1992: 44). All variations of genre pedagogy, cspeciallythose designed for more advanced students and adults, provide lcarnrrs with opportunitiesto extend and customise their knowledge about text into spccific contexts which areimportant to thcir own educational or personal goals.2.3.5 Critical litcracjAll variations of genre pedagogy also emphasise the tlevelopment in learncrs of a criticalapproach to what they arc lcarning (see Rothery, 1996: 1 16--20). By making the languagepatterns of different types of texts more visible, genre pedagogy also makes morc visiblethe values and worldviews embodied in those patterns. Thew values and worldviews arcthen open for discussion, negotiation, criticism and challengr (Christie, 1991 : 1 1 ; Christie,1995).3 The Certificates in Spoken and Written English3.1 A national curriculum frameworkThe genre approach provided the writers of the national AMEP curriculum with two keydesign principle\:1 a diqcoursc-oriented unit of language around which to write gencraliscd curriculumoutcomes2 an intcractive pedagogy for intervening in and supporting language learning.

in216 SUSAN FEEZThe writers of the curriculum used these principles to develop a curriculum frameworkcalled the Certificates in Spoken and Written English (CSWE) .This frametvork was writtenin terms of a pathway of language outcomcs organiscd across four levels:1 Beginner2 Post-beginner3 Intermediate4 Advanced.Thc CSWE language outcomes arc written in terms of genrcs to make the languagedescriptions general enough to lie a common language for planning courses and monitoringand assessing learner progress across the AMEP. At the same time Halliday’s languagc modelallows teachers to use the notion of register to customisc the very general genre descriptionsof the framework in order to meet the specific language-learning nccds of individual AMEPlearners.3.2 From curriculum to syllabus3.2. I D$ning tcrrncThc \\ riters of the CSWF differentiated Iictn ccn tv o IC\ el\ of English-language provision,one gencral, the curriculum le~el, and thc othcr spccihc, the syllabuy leic,l.3.2.2 Genercil otrtcornes: the curricirlurn lerdThe CSWE is written at the general Icvcl of curriculum. 1.carners \vork through the CSWEpathway at a pace related to their educational liackgrountl. As they movc from level to level,they nvrk in increasingly spccialiscd contexts, from a gcncral lcarning context at Iicginncrand post-hcginncr 1 1 to morc spccialiscd contexts rcla to employment, further studyor community access at the intcrnictliatc antl atlvanccd 1The discourse-oi-icntctl learning outcomes arc \vi-ittcn in tcxrms of \-cry general gcnrecategories, for example, description, rccount, instructions or information tcxt. Thesecategories are thcn linked to a macroskill listcning, speaking, reading or writing ~ orderto dcscrihc what a learner should he ablc to do with language at thc cntl ofa course of studyat that level, for cxamplc:Can dcmonstratc understanding of a spoken information tcxtCan tell a recountCan rcad written instructionsCan write a tlcscription.The outcomes arc grouptd into language-learning domains:listrning antl speakingreadinghvriting.This organisation makes it possihlc to Iircak the curriculum into smaller modules forstudents tvho need intensive \vork in listening antl speaking or rcading and writing lvhereone of thcsc areas lags behind the other. The complctc lcarning path\vay is illustrated inFigure 17.1.

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 217ASLPR 0 to 011ASLPR 12 to 1*General contexts forlanguage learningStudents may begrouped by learningpace (Bands A, B, C)General contexts forlanguage learningStudents may begrouped by learningpace (Bands A, B, C)ModulesOrientation to Learning* Speaking and ListeningReading and Writing* Mixed Language Skills* Numeracy (optional)ModulesOrientation to Learning* Speaking and ListeningReading and WritingMixed Language SkillsNumeracy (optional)ASLPR 11 to 21ASLPR 2 all skills +*Syllabus strandsIVocationalEnglishFurtherstudyMixed focusModulesOrientation to LearningSpeaking and ListeningReading and Writing* Mixed Language Skills* Numeracy (optional)Orientation to LearningSpeaking and Listening* Reading and WritingMixed Language SkillsFiprc 17. I Ccrtificatcs in Spokcnand Writtcn English: curriculumstructurc

218 SUSAN FEEZ12Can untlcrtakc the roles antl rcsponsil)ilities of a learner in a foi-mal learning cnvironmentCan use a range of learning strategies and resources3456Can clrmonstratc untlcrstantling of a spoken intorination tcxtCan provide personally relc\ant information using spoken languageCan request information/gootls using spokcn languageCan tell a short rccount789IOCan read rocial sight signsCan rcad simple written instructionsCan rcad a short information tcxtCan rcad a short narrativc/recount11121 3Can complete a \implc formatted tcxtCan w ritc a short clcscriptionCan 1% ritc a \hart rccountFigure 17.2 Outcomes for Certificate I in Spokcn and Writtc-ri EnglishThe way curriculum outcomes arc organisetl into domains within a level is illustrated bythe list of outcomes for Certificate I in Spoken antl Written English, the beginners’ level,in Figure 1 7.2.Each outcome is written in terms of a gencralisrd tcxt type, or gcnrc.The key languagefcaturcs of each text type are written as clcmcnts of the outcome. Performance criteria forassessment are based on the clcmcnts. Thus the performance criteria for each outcome drawon what the gcnre approach tclls us almut thc prcdictahlc language features of that type oftext. The elements, and their related performance criteria, are organised, using Halliday’slanguage model, into:fratures relating to the structure and texture of whole textslexical and grammatical featuresphonological or graphological featurcs.(For overviews ofthc stratified systemic functional language model see Eggins, 1994: 1-24;Fcez, 1998: 8; for an introduction to functional grammar see Butt, Fahey, Few, Spinks andYallop, 2000).The number and complexity of the pcrformance criteria for each outcome depend onthe learner’s level. The range within which stutlcnts will be assessed against those criteria,and an evidence guide, is also indicated for each outcome.Figure 17.3 shows an example of a writing outcome for Certificate 1, the beginncrlevel. Because the outcomes of the CSWE are explicit, learners studying within theframcwork know what is expectcd of them at any point in the learner pathway. They arealso able to map their own progress. In addition, teachers working within the AMEP, andin other contexts wherr the CSWE is used, share a common framework for course designantl for assessment.

Elements Pcrformancr criteria Range statements €t idence quideDiscourse structure1 can use appropriate staging2 can use conjunctionsGrammar and vocabulary3 can use vocabularv appropriateto topic4 can construct simple clauses5 can construct simple noungroups6 can use action vci-bs in thepast tense7 can use pcrwnal pronouns8 can express time and/orlocationuses appropriate staging, i.e.beginning, middle and end(orientation, sequence of went3and optional reorientation)joins simple clauscs withconjunctions, for example‘and’, ‘then’, ‘bccauw’uses vocabulary appropriate totopicconstructs simple clauscsconstructs simple noun groupsand uscs adjcctivcb, forexample ‘I had a good weekend’,‘My Lvcckcnd \vas good’uses action verhs in the past tenseuses personal pronouns asrequiredindicates timc or location as required,for cxamplc by using advcrhs and/orprepositional phrasesat least five clauscs Lrithcorrect past tense formsuses at least one conjunctionfamiliar/rclc\ant topicrccourse to dictionarymay include a fc\zgrammatical, punctuation andspelling ci-rors, hut errors shouldnot interfere nith meaning ordominate textmav redraftSample taskLearners recount sequenceof past cvcnts, for exampleexcursion, \reekend activitiesGraphologyIt is assumed that:thrre may be inaccuracies in letter formation, spelling and punctuationteaching programmes \vi11 pay attention to graphological featuresIn CSWE I the punctuation focus \vi11 I,c on capital letters, full stops and qucstion marksFigure 17.3 Compctcncv 1 3: Can Lvrite a short rccount

220 SUSAN FEEZ3.2.3 Spec!fic objectives: the .y~l/crbu.s leielThe general CSWE framework is common to all AMEP programmes. Teachers reportlearner achievement at the end ofa course against the performance criteria of the commonframework. The design of indivitlual courses of study, however, is carried out at the levcl ofthe syllabus. At this level teachers address the needs of specific groups of learners.Course objcctivcs arc statrments about \vhat is planned for a particular course of study.They bridgc thc gap between the general outcomes and a specific learning context. Courseohjcctivcs arc a distillation of an analysis of Icarncrs’ ncetls and goals and other variables ofthe lcarning contcxt.To design a systematic plan of course content customiscd to the learners in their class,teachers kvork through the following steps:1 Analysing lrarner nerd and set specific course objectives, including language-learningobjectives related to the immediate contexts where learners need to use English.23456Linking the specific objccti\cs to the general curriculum outcomes, identifying theimmediate contexts in \vhich Icarncrs will be using tcxts hclonging to the genre ofthat outcomc.Identifying and selecting what nceds to I)c learnt to meet the course objectives.Sequence the syllabus clcmcnts into an cffcctiw progression of teaching and learning.Planning how to monitor learner progress during the course and assess learnerachievement at the end of the course against the specific course objectives.Planning how to report learner achicvcment against the general curriculum outcomes.3.2.4 From curr~cu/um to s)illabu.s: from genre to regi.stcrWe haw already scen that language outcomcs in thc CSWE framework arc based ongcncraliscd text pattcrns, or gcnrcs. To customisc thc general curriculum outcome to aspecific course, teachers identify the immrdiate context of use in which these text patternswill be used. In other words thcy identify the register or variety of language learners willbe working with.For the selected context of use, teachers identify the social activities and topics whichrelate to the chosen situation of use (the field), as well as the role of those involved in thesituation (the tenor). Thc CSWE outcomcs arc already written in terms of whether thetcxts will be spoken or written, that is, the role language is playing in the situation (mode).For contcxts of use relevant to specific students, teachers may need to refine the modedescription of some outcomes. For cxamplc, thcy may need to teach a particular spokentext in the context of using the telephone rather than speaking facc-to-face. (For moredctailctl introductions to Halliday’s model of register see Eggins, 1994: 49 80 and Fccz,1998:75 81).If at the level of curriculum, lcarncrs arc working towards thc general outcome Canwrite a short rccount (CSWE I, Competency 13), they might work with tcxts such as thcfollowing:a recount of an excursion to a place of interest (fcltl) for a class book (mode) beingprepared for visitors to the teaching ccntrc on open day (tenor)a recount ofa traffic accident (field) onto an insurancc form (mode) as part of a claimto an insurance company (tenor)a rccount of a mishap with an clcctrical appliance (field) in a letter of complaint(mode) to a manufacturer (tenor)

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 221a recount of a visit to a tourist destination (field) on a postcard (modc) to a friend(t cn or) .If learners arc pi-eparing for employmcnt, they might work with this text:a recount of an incident (ticltl) on a shift hantl-ovcr rcport (mode) for the foreman(tenor).Learners could produce any of these texts at the end of a cycle of teaching and learning tomeet the requirements of thc competency.The elements and performance criteria of each CSWE outcomc are listed in two catcgoriesof language featurcs:tliscoursc structure, lvhich relates to the rccognisable parts of the genre pattern andare linked to construct a cohesive, uniticd tcxtgrammar and vocabulary.As learners work with specific texts, the elements and performance criteria ofthr outcomcguide what thcy learn about the structural, grammatical, lexical and phonological orgraphological language features of texts of that type. Each element is atldressctl within thespecific context of situation in bvhich the learners are learning to write the tcxt typc.3.2.6 lUni~r o/’rcorkThe process of syllahus drsign also involves linking the tliffcrcnt types of texts bcing taughtin the course into related units of work. For example, a unit of work on lvriting a trxtlielonging to thc genre of recount at CSWE Level 1 can lie linked to units ofwork on spokenaccounts and written descriptions. ‘l‘hese units of \vork might he based on related contcxtsof use or students might lie shown how to transfcr what they have lcarnt to completely newcontexts of USC.The approach to assessment which underpins the CSWE is described liv Mackcn and Slatlc(1 993: 205-6, 207) in the folloxving way:an eflkctivc language assessment program must IIC linguistically principled, explicit,critcrion-refercnccd, and must inform different types of assessment, includingdiagnostic, formative and summativc assessment. . . . Shared critcria based on a soundkno\vledge of language and its varieties nil1 cnahle teachers to rcflcct on the strengthsand to diagnose \veakncsses in the texts Iiroduced by their students.The gcncral curriculum outcomes of the CSWE provide general statements and relatedperformance criteria against which all AMEl’ teachers can assess learncr achievement withina common framework. Specific course objectives providc a syllabus-lcvel focus forassessment of individual learncr progress. Teachcrs can use the data collected at both Icvclsto evaluate their course dcsign.Instead of grading and ranking learners against \ague notions of general languageproficiency, assessment within the CSWE frame\vork enables learners to demonstratcx,

222 SUSAN FEEZagainst the explicit criteria of the curriculum and syllabus, what they have learnt duringtheir course.3.3 Text-based syllabus designTexts, these stretches of unified, meaningful and purposeful natural language, arc thc corecomponent of a text-based syllabus. Each trxt is a single instance of language use in aparticular context of situation. lkxts, howcvcr, draw on prcdictablc gcncric text patterns,or genres, which relate to social purposes and thc tmiadcr cultural context. Syllabuselements and syllabus design principles focus on thc use of wholc tcxts in context.3.3.1 Syllabur element\The key elements ofa tcxt-based syllabus in the context ofthc CSWE are:texts, identified according to the type, or genre, they belong to, and therefore, to thecurriculum outcome to which they relatetopics, organiscd according to whether they relate to community access, employmentor further studylanguagefeatures, related to the text type of the text in which they arc usedskills antl strategies, organisctl according to situation/registcr, especially macroskillactivities and tasks, including teaching activities antl assessment activitics, whichdetermine the materials antl resources recluiretl.Figure 17.4 is an outline for a unit of’llork on casual conversation to illustratc how syllabuselements have been selected to customise a gcncral curriculum outcome to the needs of aspecific group of learners.3.3.2 Design principleslabus is what Ur (1996: 178) tlcscrilm as a ‘mixed or multi-strand’ syllabus.rent syllalius elements such as topics, tcxts, structures, Icxis, skills andstratcgics ~ ‘in order to be maximally comprehensive’ (Ur, 1996: 178). In summary, a textbasedsyllabus is a mixed syllabus in which the organising principle is the study of lvholctcxts in context. In the context of the CS WE, coursc objectives, antl therefore learner nccd,determine the selection of syllabus clcmcnts. These elements arc then incorporated into atcxt-based cycle of teaching and learning.3 3 3 from yllubus to rncthodologj n teht bared yclc of teaching and leorningGenre pedagogy, and its intcractivc cycle of teaching and learning, provides teachers witha framelvork for selecting, organising and sequencing the comprehensive mix of text-basedsyllabus elements in a principled way, supporting lcarncrs as they gradually mow toincreasingly independent language use.The process of sequencing syllabus elements, that is, what is to be taught, involvcsteachers in deciding how to teach them; in othcr words, it involves teachers in choosing amethodology. The text-based methotlology designed to support learners working towardsCSWE outcomcs is rcprcscntcd graphically in Figure 17.4.

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 223Figure 17.4 Mcthotlology to support lcarncrs Lvorking to\vartls CSWF outconicsSource: Fccz \\ith Jo)-c.c, 1998:28, atlaptctl from Callaghan antl I

224 SUSAN FEEZPhase 2: Modelling and dccon5tructing tcxtPhase 2 is teacher-directed. Learners arc introduced to model texts Ixlonging to the targetgenre in the context they cxplorcd in Phase 1. Lcarncrs use the model tcxts to study thestructure and language features ofthc tcxt typc.This is the phase in which second-languagelearners learn the grammar ofthe target languagc, hut in the context of purposcful languageuse.For example, activities relevant to writing rccounts at lieginner level would focus onthc structure of simple tlcclarativc clauscs with past simple tense forms of action verbs, aswell as the usc of conjunctions and prcliositional phrases to scqucncc thc clauses in time.Students would also learn the structure of noun groups. Many traditional ELT grammaractivities can be effectivcl? uscd during this phase, although, in the context of Halliday’sgrammar, learners’ attention is always drawn to meaning as \vcll as form.Phasc 3: Joint constructionDuring this phase the teacher begins to hand over rcsponsiliility to the 1carncrs.The learnerscontribute to the construction of a tcxt helonging to the target tcxt typc with thr teacheracting as guide, antl if ncccssary, scrilic. Teacher and lcarncrs discuss and negotiate themcanings they are making as the! go.Phaw 4 Intlcpcndcnt con\tr uctionDuring this phasc the scaffolding is taken ama) antl lcarncrs research the context and nark\\ ith their 01% n tcxts intlcpcdcnth, consulting 12 ith othcr lcarncrs and the teacher onlj asneeded Achie\cmcnt asscssmcnt 15 carried out at thi5 stage of thc L~LIC.Phase 5: Making links to rclatcd tcxtsle, links are niatlc to r-clatcd tc.xt typcs, so learners have thc opportunityy have lcarnt in othcr contcxts of use, comparing and contrastingtliffcrcnt tcxts antl thcii- uses antl effectiveness. Some lcarncrs ma); bc ready to adapt thetcxt type they have lcarnt to control to a specific IiurIiosc, adjusting thc stages antl thelanguage fcaturcs as necdcd.3.3.4 dleeting thc ncccls ?f’clf;tfcrcnt Ictirncr~Whenever necessary the teaching-learning cycle is modihcd to suit the ncctls of differentlearners. Learners can cnter at an! phasc of the cycle, returning to an earlier phase forreview as needed or skipping phascs if thcy arc not ready or if they do not need them. Inmost adult ESL classcs thc context-building phasc is essential. Some beginning learners withminimal formal Icarning in their first language, however, may not go beyond the jointconstruction phase for some more challenging tcxt typcs. In contrast, many tcrtiaryeducatedadult lcarncrs with sophisticatctl study skills find the joint construction phaseunncccssarv.3.3.5 .I lun

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 225possiblc for teachers to selcct, modify and locate a mricty of mcthods in a principled antlstrategic way. In othcr words, they select from thc language-teaching repertoire mcthodswhich make it possible to build the type of classroom intcraction required hy thc tliffercntphases of the cycle.4 ConclusionOverall the AMEP has bcncfitcd greatly from basing its new curriculum on genre pedagogy.AMEP learners now can track thcxir progress against general descriptions of Englishtext patterns \vhile, at the samc- time, learning to customisc thcir own texts to mcct thcdemands of thcir immediate situations. Learners can consciously build the cultural andlinguistic knowledge which will help them make the most of the new community thcy arcentering. The CSWE is a framework within which teachers plan courses antl map learningpathways.To support curriculum change in the AMEP, classroom materials antl resourcesmodelling a tcxt-based approach to syllabus design have been published (for examplc Brownand Cornish, 1997; Clemens antl Crawford, 1994; Cornish, 1992; Dclaruelle, 1998; Fccz,1998; Jovce, 1992; NSW AMES writing team, 1997). Professional development has includedcxtensive training in educational linguistics, course design and assessment. Sincc thcimplementation of the new curriculum framework, there has also been an ongoing cycle 01’national classroom-based collaborative action rcscarch. The action rcscarch model providesteachers with a useful tcchniquc for rellecting on and rrnovating classroom practice (seeBurns and Hood, 1995; 1997; 1998; Burns and Joyce, 1999).Genre pcdagogy, like language pedagogy generally, is evolving and changing. The \vaytcachers in the AMEP arc working with the pedagogy is also changing as different tcachersintcrprct it in different ways. When teachers first applied genre pcdagogy, many s upimposcdstructural approaches onto the generic descriptions of text structure and languagcfcaturcs.This resulted in teachers teaching text patterns as fixcd rulcs and forms rathcr thanin terms of meaning and function. It also resulted in some teachers fceling that they had toabandon the lcarncr-centred mcthotls developed as part of needs-bad, communicativcapproaches and return to teacher-ccntrcd classrooms. As tcachers have adaptctl to thenew curriculum environment, gaining knowledge, skill and confidcncc and adjusting thcirbclicfs about language and language learning, they are increasingly integrating thc best ofsituational, learner-centred antl communicative approaches into a text-based framework.Teachers are also beginning to identify kvhich aspects of the text-bascd approach needreviewing or developing.Thc AMEP is currently experiencing a period of rapid change. Changing political andeconomic ideologies are moving the AMEP away from being a stable, unified, public-scctorprogramme to a more fragmented market-oriented programme. It remains to be seenwhether this new oricntation will bc able to dclivcr a service of’ comparable quality. Itccrtainly is not clcar whether thc new environment will continue to support the principleddevelopment of AMEP curriculum and cxpertise in tandem with dcvclopments in thc ficltlof applied linguistics.While the future for migrant education in Australia is unclear, it is clcar that text-hascdapproaches provide language educators with a strong foundation for further developmentsin language teaching. The key clcments of this foundation arc:an understanding of what constitutes a whole unit of language in the context of itsuse

226 SUSAN FEEZa methodology for providing learner? Lvith experience of whole unit.; of language usein context while they are still only ablc to manage language fragments.From this foundation, language educators havc thc opportunity to develop increasinglyeffective ways of‘ teaching explicitly and systematically about text, grammar and lexis inorder to makc it possible for learncrs to build skills in spoken and written English whichthey would not be ablc to tlcvclop on their owm.ReferencesRrindley, G. (I 984) ‘The role of nccds analysis in adult ESL programme design’. In R. K.Johnson (etl.) The Secontl Lungooge Curriculum, Cambridge: Cambridge University Press.Brindley, G. (1985) The Asse.s.sment of Second Language Prqpcicnc).: Issues and Approaches, Adelaide:NCRC.Brindle);, G. and Hood, S. (1 990) ‘Curriculum innovation in adult ESL’. In G. Brintlley (ed.)The Second Language Curriculum in :lction, Sydney: NCELTK.Brmvn, K. and Cornish, S. (1997) Reuch Street \:An English Course_fbrAdults, Sydney: NSW AMES.Rrumfit, C. (ctl.) (1986) The Prncticc oj Communicatire Eaching, Oxford: Pergamon Press.Burns, A. and Hood, S. (eds) (1995) Euchers’ Lbices: Exploring Cotir.se Design in a ChangingCurriculum, Sydney: NCEI TR.Burns, A. and Hood, S. (cds) (1997) Euchcrs’lbiccs 2: Zuching Di.spuratc Leurner Groups, Sydney:N C E LTR .Burns, A. and Hood, S. (cds) (1998) Tcuchcrs’lbices 3:lcoching CriticalLiterucy, Sydney: NCELTR.Burns, A. antl Joyce, H. (etls) (1999) Tcachers’lbices -I: Stuying Learner-centred in a CompctcncybasedCurriculum, Sydney : NCELTR.Burns, A., Joycc, H. antl Gollin, S. (1996) ‘I See what you Mean’: Using Spoken Discourse in theCla.s.sroom:A Hundhook,fbr Euchers, Sydney: NCELTR.Butt, D., Fahey, R., Fccz, S., Spinks, S. antlYallop, C. (2000) Under.standin

AUSTRALIAN ADULT MIGRANT ENGLISH PROGRAM 227Gibbons, P. (1998) ‘Classroom talk and the learning of new registers in a second language’.Language and Education 12 (2): 99-1 18.Gray, R. ( 1985) ‘Helping children become language learners in the classroom’. Paper given atthe Annual Conference of the Mcanjin Reading Council, Brisbane, May 1987, in M.Christic (etl.) Aboriginal Pmpectives on Experience and Learning: The Role tf’ Language inAboriginal Education, Gcelong, Victoria: Dcakin University Press.Gray, R. (1987) ‘Ho~v natural is “natural” language teaching employing wholisticmethodology in the classroom’ The Australian journal ofEar(v Childhood 12 (4): 3-1 9.Green, J. (1992) Making the Links. Melbourne: AMES Victoria.Hagan P. (1994) ‘Competency-basctl curriculum: the NSW AMES experience’, Prospect: Ajournal ofAristralian TESOI~ 9 (2): 3040, Sydney: NCELTR.Hallitlay, M. A. K. (1975) Learning Hov. to Mean: Explorations in the Development of Language,London: Edward Arnold.Halliday, M. A. K. (1 976) System and Function in Language. London: Oxford University Press.Halliday, M. A. K. (1978) Langtiage as a Social Semiotic, London: Etlward Arnold.Hallitlay, M. A. K. (1991) ‘The notion of“contcxt” in languagc education’, in 1‘. Lc and M.McCausland, Language lnteraction and Development: Proceedings of’ the InternationalConference, Vietnam, 1991, Launccston: Univcrsity ofTasmania.Halliday,’M. A. K. (1992) ‘Towards a language-based theory of learning’. Paper prepared forthe Phonctic Society of Japan in the context of the Symposium on Language AcquisitionTokyo 3, Octobcr 1992.Halliday, M. A. K. and Ilasan, K. (1985) Language, Contest and7ixt:Aspcct.s oflanguage in ~7 SocialSemiotic Perspective, Geelong, Victoria: Deakin Universit? Press.Hammontl, J. (1989) ‘The NCELTR literacv project’, Pro.spect: ’4 journal clfZ4u.stralian TESOL5(1): 27-70, Sydney: NCELTK.Hammond, J. (1990) ‘Teacher expertise antl learner responsibility in literacy development’. InPro.spect:A journal ofAustralian TESOL 5( 3): 39 5 1 , Sydney NCELTR.Hammond, J., Burns, A,, Joyce, H., Brosnan, D. and Gerot, L. (1992) English for Social Purposes,Sydney: NCELTR.Hood, S., Solomon, N. antl Burns, A. (1996) Focus on reading, Sydney: NCELTR.Howatt, A. (1984) 11 Histoy ofEnglish Languuge teaching, Oxford: Oxford UnivHymes, 1). (1972) ‘On communicative competence’. In J. 13. l’ridc antl J. Holmra (rtls)Sociolinguislics, Harmonch orth: Penguin, pp. 269-93.Ingram, D. and Wilcy, E. (1984) Aristralian Second Language Proficieny Ratings, Canberra:Australian Government Printing Service.Joyce, 1-1. (1992) Workplace Exts in the Language Classroom, Sydncy: NSW AMES.Knowles, M. (1990) The Adtilt Learner - A .h‘eglected Species, Houston: Gulf.Krashcn, S. (1988) Second I.ungtia~cleAcytiisition and Second Language learning. NeIvYork: Prentice-Hall.Krcss, G. (1991) ‘Texture antl meaning’. In W’orking with Genre, Papers from the 1989Conference, University of‘lkchnology, Sydney, Sytlncy Common Ground.Mackcn, M. and Slatlc, D. (1993) ‘Assessment: a foundation for effective learning in the schoolcontext’. In R. Cope and M. Kalantzis (eds) The Power.s oJ‘Literacj.:A Genre Approach to the‘Itaching !f Writing, London: Falmcr Press.Martin, J. R. (1993) ‘Genre antl literacy modelling context in educational linguistics. AnnualReview ofApplied Linguistics 13, 14-172.Melrose, R. (1991) The Communicative Syllabus: A Sjxtcmic-functional Approach to Languqeteaching, London: Pintcr.NSW AMES writing team (1997) LVanjarri: Indigenoux A~istralia in the ESL Classroom, Sydney:NSW AMES.Nunan, D. (I 987) The Teacher or Curriculum Developer, Sydney: NCELTR.

228 SUSAN FEEZPainter, C. (1 985) Learning the Mother Tongue, Geelong, Victoria: Deakin University Press.Painter, C. (1996) ‘Thc dcvclopment oflanguagc as a resource for thinking: a linguistic view oflearning’. In R. Hasan and G. Williams (cds) Literuy in Sociey, London: Longman.Richards, J. C. and Rodgcrs, T. S. (1986) j1pprociche.s and Methods in I.anguap Teaching: itDescription and Anahsix, Cambridge: Cam1)ritlgc Universitv Press.Rothery, J. (1996) ‘Making changes: tlc~eloping an educational linguistics’. In R. Hasan andG. Williams (eds) Literacy in Socict,i., London: Longman.Sclinker, L. (199 I ) Rediscovering Interlanguqqe, London: Longman.Ur, P. (1996) ,4 Course in Lungucige Eciching: Practice and Theory, Cambridge: CambridgeUniversity Prcss.Vygotsky, L. S. (1934/ 1978) /141ntl in Socicy: The Development qj’ Higher f.ychoJogical Processes,Cambridge: Cambridge Univcrsity Prcss.Vyptsky, ,‘. L. S. (I 986) ThoLight and Lunguu‘qc, revised and cditcd by A. Kozulin, Cambritlgc,MA: MIT Prcss...Widdowson, H. (1990) Aspects of’longuuge Icciching, Oxford: Oxford Univcrsity Press.Wilkins, D. (1976) Notional Syllabuses, Oxford: Oxford University Press.Yalden, J. (I 987a) Principles ?f’ Cowse Design ,fir Language Eacbing, Nen. York: CambridgeUnilersity Press.Yalden, J. (19871,) The Cornmunisatit.e S~~llahus: Evolurion, Design and Implementation, London:Prentice-Hall.

Chapter 18David R. HallMATERIALS PRODUCTION: THEORYAND PRACTICEIntroduction: learning a languageEFORE PLANNING OR WRITING MATERIALS {or language teaching, thcrcB is one crucial question \ve need to ask ourselves.Thc qucstion should be the first itemon the agenda at the first planning meeting.1hc qucstion is this: How do we think people [earnla ng uap?Nobody knows ho\v we learn language. Of course, thcrc arc plenty of theories around.Many people makc: a distinction between first language learning and second or foreignlanguage Icarning. Others distinguish clearly between “learning” a language and “acquiring”a 1anguagc.Thcrc is a further distinction between “learning to spcak”, “learning to understandspokcn language”, “learning to read”, antl “learning to write”, with some thinking that thcarc similar proccsscs, others that they are clifferrnt but rclatcd proc YS, and still othersthinking that they involve completely unrrlatcd skills. Some think that hvc learn by studyingand internalising rules, othcrs that xvc learn by rcpcatrd practice of common patterns, vetothcrs that \YC learn by mcmorising hvords antl collocations, by simulating real situations,by negotiating, by hearing and reading things we can understand [. . . I‘l‘hc list is long.No-one knows how \vc learn language, so the \vortls we think in the original qucstionare very important. What is our own opinion?What can the writing team agrcc on? It is ourview of how we learn language that will underpin wcrything else that we do in planningand writing our materials. There arc, of course, other important considerations, some ofwhich will be mentioned here, and there are always constraints the straitjacket of accntraliscd syllabus, the need for pupils to pass public examinations, the lack of liliraryrcsourccs, a cultural prefercnce for the teacher being the holder of all knowledge, the needfor an orderly and industriously quict classroom, and so on. But thcsc should not governwhat we must do.Thcy should bc acccptcd as fcaturcs of the context in which \\-e attemptto do what we think is the right thing to do.tion of this chaptcr, I will put forward somc of my own liclicfs aboutlanguage learning and teaching, suggesting criteria liy which materials might lie evaluated.The discussion is conducted under four headings: the need to communicate; the need forlong-term goals; the nced for authcnticity; antl the ncctl for studcnt-ccntrcdncss. In thethird and final section, I will examine materials from projects in which I have been involvedantl evaluate thcm in the light of the criteria.

230 DAVID R. HALLTheoryThe need to communicateThere are three things about language learning that arc fairly obvious and uncontroversial,but which are not often enough stated together. First, just about everyone learns tocommunicate fluently in at least one language. This is normally thc L1 or mother tongue.Second, very few people learn to communicate fluently in a new language learned entirelyin formal language classes. Third, most people who learn to communicate fluently in alanguage which is not their L1 do so by spending a lot of time in situations where they haveto use the language for some real communicative purpose.The conclusion I draw from thcsc three facts is that the need to communicate is at theheart of learning a language. If lvc want our pupils or students to learn English, we mustput them in situations where they ncctl to communicatc in English. For rcal communicationto take placc, there are three conditions:123We must haw something that Lve want to communicate.We must have someone to communicate with.We must have some interest in the outcome of the communication.In most languagc-teaching materials, and in most languagc-teaching classrooms, thcsc threeconditions do not exist. It is, of course, unrealistic to expect that they could exist all thetime. A sccondary school class of forty pupils and thcir teacher cannot all have the needto communicate in a forcign language for thcir owm purposes all the time. Nevertheless,given the lack of success of conventional classroom teaching, it is perhaps worth makingthe attempt to approximate as closely as possiible to the three conditions.The need for long-term goalsIn devising a syllabus, materials or methodology for a language-teaching program, it hasalways heen necessary to conduct some sort of needs analysis. This tends to concentrate onaspects of language structure, language function, situational features, technical content, orbehavioural outcomes (competencies).We should, however, look at the longcr-term ncctls of students. The language-teacheroperates within fairly tight limitations. What, in oliservahlc terms, can lie achieved in aprogram of one school year when you see the students three times a week and they neverhave to use thc target language outsidc those three hours? By the end ofthe year, the studentsmay be able to perform more or less satisfactorily in a formal test, but only the reallycxceptional will have progressed to anything approaching fluency. No matter what excitingmethodology you use, the results are always going to he disappointing.It is only in the longer term that some students will cxperience the need to communicatein the targct language and mill havc the chance to become fluent. In looking atneeds, perhaps we should look more closely at this longer term. If we are going to help ourstudents succeed in learning a language, M’C must prepare them for the opportunities whichwill come outside the classroom. In other words, we must teach them how to learn. By‘teaching horn to learn’, I do not mean to refer to learning in classroom settings, whereconsiderable work has been done on learning styles and thcir applicability to the languageclass (Hawkey, 1982; Willing, 1989), but rather to the ahility to take advantage of anyopportunity to learn outside the classroom.

MATERIALS PRODUCTION 231If a tcachcr is to provide students with the tools to learn language outside the classroom,the main aim has to be to give students confidence in their ability to communicate dcspitedifficulties, to the point where they can: (a) initiate communicative cvcnts, and (b) persistwith the attempted communication even when it becomes difficult. The use of gestures,facial expressions, rewordings, questions, guesswork, and approximations is a crucial partof such communication, and the ability to use such strategies becomes an essential feature,in my view, of student needs.I believe that in a natural setting, people learn language by having to communicatesomething that they do not know how to say. While more elegant or more accurate or morcverbal ways of expressing the point may be developed as similar situations are repcatcd, thereal initial learning takes place when a solution has to be found to thc problem of notknowing how to communicate something. It is the long-term goal of our language teachingto provide our students with the confidence and ability to do this.The need for authenticityIn the light of the above assumptions about language learning and the long-term needs oflanguage learners, the question ofthc use of authentic materials can lie refocused. It is not,primarily, the materials themselves which have to be authentic. It is, rather, the response tothe materials what is donc with them ~ that should be authentic.Many materials are impeccably authentic, by which I mean that they arc ‘found’matcrials originally written for some other purpose than languagc tcaching. But becausethey deal with topics from the students’ particular specialisation, they often deal with topicswhich arc already very familiar to students. The reading of such rnatcrials becomes merelyan exercise, not involving an authentic need for reading it is neither thr seeking ofspecifically needed information nor the exploration of a new topic. in many casc-s, it isnothing more than the reading ofa text for the purposes of being tested on it through variousforms of comprehension tasks and linguistic manipulation.An authentic response depends on thc existence of an authentic need. In the classroomcontext, this need may only be an approximation and may be artificially created. It can behelped, in ESP matcrials, by close cooperation with the content teacher, so that the kS1’matcrials complement and support what is happening in the content class. At the very least,an authentic response dictates the atltlressing of content rather than form, and discussionfor clarification or expansion rather than for the mere checking of understanding. (I am notsaying that teachers should not check understanding. Checking can be donc just as easilyand is more natural during genuinely communicative events.)Kenny (1989) classifies student response to content into threc catcgorics:thc empiricalthe interpretationalthe socially validated.By ‘empirical’ is meant the addressing of the content as a context-free, isolated entity. Theempirical response involves working out the meaning of a text within the boundaries of thetext. The language tcachcr’s typical tools of comprehension questions, structuremanipulation exercises, summaries, vocabulary explanation, and mode-switching (e.g., textto table, graph to text) all stay firmly at the empirical level of response.An ‘interpretational’ responsr addresses the meaning of a text in relation to the individual.It involves assimilating new knowledge into the structure of information in theindividual’s head. Learning styles arc obviously important in this area, as thcy rclatc: to the

thethe232 DAVID R. HALLway in which knowledge and experience arc storcd and retrieved. Thr content is examinedin relation to existing kno\vledge structurcs and belief systcms.Traditiona1 essays (“Compareand contrast the vicws of x andJ”) might include interpretational responses to a number oftexts, as might examination of parallels and contradictions between different texts.The ‘socially validated’ response involves exposing the individual’s response to a textto group evaluation. In other words, it is not enough to assimilate new knowledgeindividually. The undcrstantling of a text antl the validity of that understanding need to bctested through group interaction, and the intcrprctation defended in a process of criticalscrutiny. Public presentation of ideas through postcr sessions, tlcbatcs, presentations, andso on, is an opportunity for socially valitlatctl responses. In this context, the opportunityfor cross-disciplinary communication in ESP classes with a heterogeneous studentpopulation should be seen as an advantage rather than a disadvantage (Hall, 1994).To put thcsc three categories of response anothcr Lvay: the empirical has a singledimension ~ contcnt; the interpretational has two tlimcmsions ~ contcnt and theindividual; while the socially validated adds a third dimension that of society, representedin the classroom 1y the group. Combining the nccd for authentic response with the nccdfor drveloping confidcncc to initiate and pcrsist \vith communication, we can see that it isdesirable to aim for a socially validated response to materials in class. It is also clear thatmost materials used in thc language-teaching classroom approach neither social validationnor the prerequisites for communication. What sometimes looks like a social validationactivity is often no more than an exercise in which real beliefs are not cxplored, the contcntbeing dictated Iiy either thc tcacher or the matcr-ials. An example of this might be wherea student is given notes on arguments for better public transport and asked to “role-play”,by giving a presentation as a lobbyist to a group of other students playing the role ofpolicy-makcrs.The need for student-centrednessThe language classroom may bc thought ol’as ha\-ing thrcc components ~the teacher, thelearners, antl the materials. Traditionally, all of the actual contcnt of the class, i.e., \vhat istalked about, comes from the teacher or thc materials. The potcntial for learners toparticipate in generating materials has long been neglected. I \vould suggest that studcntsthemselves arc in a unique position to look for rclcvant resource matcrials.Thcy know whattheir owm needs antl interests arc.The process of learners searching tor materials and then bringing thc materials back toclass where they arc prcscntcd to other students involvcs morc than simple selection. Theprocess changes student status from passive receivers of information to active accountability(see Kcnny, 1993). Their selection of materials not only has to be pi-

3)(xMATERIALS PRODUCTION 233Sample materialsI have Iieen involvcd in a number of materials and curriculum development projects. I willhcrc illustrate four of them and give an example of representative classroom activity fromeach, analysing the activity in terms of the above discussion.A notional-structural approachThe development of the materials which became the “Nucleus: English for Science antlTechnology” series arose out of the demands of the tcaching situation in the early scyentiesat the University of Tahriz in northern Iran. I’rachcrs and students alike Lvcrc unmotivatedby the general knglish textbooks then in use and wanted something more rclevant to theactual purposes to which students were going to put their English.Thc new materials were arrangcd under chapter headings labcllcd with scientific“concepts” such as Measurement, Description antl Process (Bates, 1978; Dudley-Evans etal., 1976), in a similar xvay to a Notional Syllabus (Wilkins, 1976). There is no doubt thatthe materials were very innovative in a numhcr of ways, but it is also clear that the ostensiblynotion-based framelyork for thc syllabus disguised an undcrlying structural approachusing pattern practice and traditional guided writing trchniqucs. The series proved tobe very popular when it was released commercially, and thc syllabus framework waswidely imitated, both in other commercially produced textbooks and, more significantly,in hundreds of indi\idual materials-writing projects in different institutions around thcworld as ESP became the catchword of the late 70s and earl! 80s. Hcrc is a rcprcscntativcexerciseA quadratic cquation has two solutions, called roots. If the factors of’ a quadraticequation can lie found easily, then \vc can find the roots by factorising.Example: 1:aLtorisation ofx’ + 1-1 20 gi\es (A ~+ 4) = 0.The roots of the equation arc therefore 3 antl 4.No\\ make similar rentences about thc follo\z ing:a) x2 + 7x + 10 = 0b) x’- 91r + 18 = 0c) x2 100 =0ti) x2 + 5x 6 = 0(Hall, 1980: 51 52)In tcrms of expected student responsc, it is clcar that thcrc is nothing here beyond theempirical 1evel.The student may bc motivatctl by thc partial relevance of the subject-matter,but thcrc arc no dcmands made on student inventiveness and nothing is contributed by thestudent. All language production is controlled entirely by the textbook, to the extent thatconceptually correct answers that arc not in conformity with the prescriptiveness of thetextbook author are deemed to bc incorrect. In tcrms of the prerequisites forcommunication, this text would only prowke authentic communication if students disagreedon some aspect of the content and the teacher allowcd thc discussion to go beyond thcdcmantls ofthc tcxt.The view of languagr learning is essentially behaviourist that learningtakes place through exposure to language patterns.

234 DAVID R. HALLA communicative approachThe University of Malaya Spoken English Project of the early 80s (Hall, 1985; Khong, 1984),set up with British Council help, used the Munby ‘communicative’ needs analysis approach(Munby, 1978), and was perhaps thc only major project to attempt to do so with any rigour.A major problem in the project \vas that the first 18 months of the 3-year project weredevoted to discussion of necds analysis and theoretical considerations, with the predictableresult that by the time it came to actualll; writing, the team members wcrc so entrenchedin different antagonistic theorctical positions that consensus writing had become almostimpossible. Despite the warnings in this chapter to consider theoretical positions seriouslybefore planning materials, writing should not be delayed too long. The acts of‘ writing andtrialing cannot be delayed until a fully worked out thcorctical position has been established.In fact, the dcvclopmcnt of theory and practicc go hand in hand. See Figures 18.1 and 18.2for a representativc example.1 You are looking lor a scholarship to stuly ovcrscas.Complete this table with the dctails of the scholarship you \voiild likc to gctAmount pcr monthIhrationSubject of studyCountryExtra alloLvanccs2 You have some accommodation to rent.Fill in this table \vith thc details of the accommodation you can offcrNature of accommodationNumber of roomsRcntFacilitiesLocationConditionsFigtire 18.1 Workshect 1 : stutlcnt A only

theandMATERIALS PRODUCTION 2351 You are looking for accommodation to rent.Complete this table \vith dctails of the accommodation you would like.Nature of accommodationNumhci- of roomsIkaciliticsLocation2 You have a scholarship to offer.Fill in thr table with the details of the scholarshipAmount pcr monthDurationCountryConditionsIIFigtire 18 2 Work\hcct 2: \tudent B onl)No linguistic structure is prcscribcd in the example given, although the matcrials didin fact include a language support section in each unit in an attempt to anticipate the languagcneeds of the activity. One consequence of using the Munby approach was that oftenmore time was devoted to setting up a situation than actually doing the activity. It \vasnot unknown for a couple of pages of input to produce only a line or two of linguisticoutput.The above excrcise is fairly typical of the sort of information gap cxercisc frequentlyfound in ‘communicative’ textbooks. Despite an outward appearance of social validation(opinions have to be exchanged), roles are assigned, and content relatidy tightly controllcd.The cxcrcisc docs not havc thc thrcc prcrcquisitcs for genuine communication. The contentis not the student’s own ~ role-play attempts to create ownership artificially ~ inthe end it is a mattcr of indifference to the student whether the outcome of the financialnegotiations is advantageous or not.The student is not enpgk, is not involved to the extentof having a personal stake in the outcome.

236 DAVID R. HALLA genre-based approachThe approach to reading and writing tcchnical texts tlcvclopcd at thc Asian Institute ofTechnology inThailand in thc mid-80s can hc classified as a gcnre-liasetl approach (Hall etal., 1986) in that it attcmpts to analpsc tcxt in terms of thc typical discourse featurcs andlanguage functions to I>(- found in tliffcrcmt kinds of technical writing, particularly thoserelating to the research article antl the student dissertation. Unlike the approach takcn byAustralian genre-tmcd theorists (Dcrclvianka, 1991 ; Martin, 1993), it tlors not attempt toassign grammatical fcatures to particular functions. It aims, rather, to provide analyticaltools to students so that they can approach rcatling in a critical \vay, transferring this skillto a critical rcatling of their o\vn \vriting. Hcrc is an example of an activity, chosen morebecause it is short than because it is rcprcscntativc. A more representative sample wouldstretch to many pages and normally involvc thc analysis of part of a text in the context ofa whole article or thcsis.Predict how thc follo\z ing cxtracts might continue. All extracts arc takcn from thejournal “Solar Energ! Materials”,Vol. 19, 1 antl 2, 1989.In fig. 6 the dependence of thc optical transmission antl sheet resistance on theannealing time at 620°C for two tlil’fcrcnt coatings are shown. The liehavior oftransmission T and shcxct rcsistancc R at this temperature in thc investigated timcinterval is different for diffcrcnt stabilizer matcrials. For Ni (see fig. 6) we oliservctlat thc beginning an incrcasc of the transmission and a decrease of the shcet resistivity.Aftcr two minutes,T\vo diffcrcnt unconstrained optimization approaches \\rere implemented to evaluatelayer thicknrsscs and Ag optical constants. In onc approach all unknmvns (threethicknesses and 36 pairs of n antl k) \vcrc evaluated simultaneously using theMarguartlt algorithm [ 12, 131.Thc merit function, F, \vas the following:where the whscript I denotes ith mcasurcmcnt, the superscripts cxp and cal referto the experimcntal and calculated \dues, rcspecti\elj, and rn i\ the numlier ofmeasurement\. We M 111 call this the “one shell” approach.Fig4a shows that in the casc ofthick coatings (60 C/dm’ and more) the well knowncmission characteristics of intcrmcdiatcly absorbing dielectric media are obtained.The dashcd curvcs wcrc calculated with the optical functions of fig. 3 and agreesatisfactorily with the emittance mcasurcmcnts.

1 throughMATERIALS PRODUCTION 237This is a small part of a unit for students of Encrgy Tcchnology on sequencing. Itconcentrates on using linguistic antl contextual clues for efficient reading. Students discusspossible completions to the extracts in small groups, so that there is a limited tlegrcc ofgenuine interaction, provided that the texts are ones that arc of real interest to the students.In terms of the criteria for communication and response to text, this third example seemsto me to lie approaching authenticity within the constraints of the classroom. In particular,of text very often involved discussion of the intention ofthe authors and a criticalexamination of their arguments, often cxtcnding to students’ o\vn knowledge and henceinvolving social validation.As the course developed, we began to rely less and less o uhlishcd texts antl to getstudents to bring their own work to class. This appeared to lie> more motivating, giventhat the aims of the Information-Structuring course were to help students develop theirown writing for immediate deployment in assignmcnts and a research-based thesis. In thishvay, the course also fulfilled the criterion that it should provide students with the tools tocontinue improvement after the coursc. Ncverthclcss, some of the tasks in the course stilllooked like exercises designed for practice rather than the occasion for genuine involvement,antl it is \vorth noting that the course for which these materials were originally developedat All‘ has itself now cvolved into something which integrates language and other skills \viththe initial stages of carrying out research and writing a thesis.A studen t-generated, experiential approachThe ‘Talkbase’ approach, also tleveloped at the Asian Institutc of Tcchnology in the 8Os,(I Iall, 1994; Hall antl Kennv, 1988; Hall and Kenny, 1995), was devised for an intensi\,e,full-time course. No detailed timetal>le or content is specified. Only a general syllaliusoutline is given, Iiased on a repeatcd pattern of Plan, Do, Report Rack, Evaluate, and PlanAgain. Studcnts carry out a major piece of intlcpentlent Lvork during the coursc, using allthe resources of the immcdiatc environment including teachers and other students. Worka series of report-back sessions in various modes poster sessions,tions, individual consultations, interviews, and so on.‘l.he course, unfortunately for the purpo . of this chaptcr, does not use teachingmaterials as such, so representative examples are difficult to find. The syllahus is a set ofprocedures rather than a set of materials or a set of linguistic, functional, Iiehavioural orsituational categories. Hohvever, a description of thc first week of opcration may help togive an idea ofwhat the course is like.On the tirst morning of the course, the only teacher-provided “material” ot’the firstwcck is given to studcnts.’l’his consists of a slip of paper, on which arc written the lvortls:Wclcomc to thcTalklmw courw. We mould no\\ like >ou to lea\? the classroom antlto Lome hack again this afternoon ready to talk for a fcn minutes aliout X.‘X’ is a single wort1 or a phrase chosen by the teacher. Examples arc: Drying; UncxpcctcdOutcomes; Autonomy; Water; Technology ; Saving.First presentations by students arc normally short anti not particularly coherent, butthey are discussed by the teacher and all the other students, normally in groups. At the endof this, students have to plan again, informed nolv hy feedback from others antl hy thcirexperiencc ol‘\vhat others hale tlone.They then go off and report hack a second time. 011the third occasion, thcy report in writing, and writing is passcd around among the groupfor comments. As the first lveek develops, students begin to find personal meanings in thcir

238 DAVID R. HALL“word”and gradually the very wide area covered by the original word is delimited to a topicwhich is of personal intcrcst to the student.As the coursc develops, and students hcgin to analysc published and unpublishedacademic discourse produced hy others, both form of presentation antl organisation ofcontent improve markedly, antl communication within the classroom, as well as outside it,becomes committed and almost totally student-dominated. Except at a very few places,such as the example from the first day of the first week, texts (recorded interviews, journalarticles, etc.) are found and brought to class by the students themselves, so that coursecontent is generated by students, not by teachers.Students find themselves engaged in research in their own field of study, research whichmany of them will go on to develop further as part of their Master’s or Doctoral dissertation.They struggle to communicate their research not only to others in the class who share theirtechnical specialisation, but also to those \vho ncctl more detailed background information.At the end of the course, students’ scnse of achievement at being able to present complextechnical information to various differcnt audiences givcs them precisely that confidencementioned in the section of this paprr on “The need to communicate” to initiatecommunication and to persist with it when there are difficulties.In terms ofthc prerequisites for communication, they are all present: there is a genuinecommitment to communicate, there is a genuine audience, and students care about whetherthey have made their point. It is intercsting to watch the effect that this has on weakerstudents, who in many language classes would never open thcir mouths unless forccd to bythe teacher. In this course, the desire to take the floor and to make a point does not dependon linguistic ability or a forceful personality; it depends on having something to say. In termsof Kenny’s three categories of interaction with text, activities fall clearly into the socialvalidation category. Students present their work, their ideas and their opinions for publicscrutiny.Concluding remarksThe principles and opinions given in this chapter are based on personal experience andreflect my own development as a teachcr and materials writer.Thc ‘social validation’ of myvalues has ultimately been through presentations in journal articles and conferencepresentations, but initially they have been discussed in the hothouse context of materialsand curriculum development teams and tested in actual use in the classroom.For materials writers, it is worth bearing this in mind.You do not write to conform tosomebody else’s model.You look at other people’s models and you read current theory, butin the end your materials and the writing of the materials will not be the simple passiveimplementation of someone else’s ideas.Thcy will be developed in the interaction betweenthe writcm, the teachers and the studcnts.They will contribute to the sum total of materialswriting experience. Both your own and other people’s beliefs about effective languagelearning will be modified and enriched by your experiences.ReferencesBates, M. 1978. “Writing ‘Nuclcus”’. English for Spectfic Pnrposes, ed. by R. MacKay and A.Mountford, 78-96. London: Longman.Clayton,T., Shaw, J., Le,T.T. M., Nhan, C. H. and Pham,T. 1993. “Discovering resources in HoChi Minh City: preparing the ground”. Lanpage Programs in Development Projects, ed. by W.Savage, 33 1-341. Bangkok: Asian Institute of Technology.

MATERIALS PRODUCTION 239Drrelvianka, B. 1991. Explorin

Chapter 19Simon SergeantCALL INNOVATION IN THE ELTCURRICULUMIntroductionH E C 0 M PL EX IT Y 0 F I N FO R M AT I ON T EC H N 0 LOGY (IT) innovationT and the speed of diffusion antl technological advance sccm to haw lcft the Englishlanguage teaching profession searching for ways of integrating IT usefully into thrcurriculum. While there seems to be little doubt ofthc potential of 1'1; it is difficult to specifythe nature of the new learning opportunititx Papert (1987) and l'erkins (1985) highlightthe fact that there is much still to 1~ tliscovcrctl about the placc of computer-assisted learning(CAL) in education, and this is still the case totlay.'I'his article does not claim to produceanswers, but I hopc it \vi11 contriliutc to awareness of the problcm. The aims of the articlearr:to cxamine thc nature of CALL (coniputcr-a tee] language learning) innovation ant1its potential as a force for curricular change hvith cxamplcs tlrawm from my work ina language ccntrc in Singapore;to invrstigate reasons fbr the shorttall Iictwccn thc potential ot' CALL and actual use,and discuss reasons why CALL opportunities arc not taken;to indicate strategies by which a changc agent may add valuc, to a CALL facility.Computers in commerce antl industry arc associated with higher cfficiency.This assumptionhas been carried into the educational arena, antl into language teaching in particular, withvarying degrees of succcss. CALI. as a discipline is cstablishing a rcsearch basc after severalyears' local trial and error supportcd by anccdotc. Hohvevcr, rcscarch is oftcn carried outunder ideal conditions \vhich arc only partially realizable \vithin the constraints of everydayuse. These local constraints arc informed tiy attitudes of the major stakcholdcrs in CALL:managcrs (usuallj non-users), CALI. pcrsonncl (initial users), and tcachcrs antl studcnts(entl-users). Students, who arc thc recipients of CALI,, arc thc least consulted during thcdwision-making proccss.Thcy arc also thc ones lvho arc most tlisadvantagetl if CALL is noteffectiwly implcmcntcd.The full potential ofintcgrating computers into the ELT curriculum has not yet beenreached and their use is still limited. CALL is trcatcd as a separate entity antl boltrd on tothe existing curriculum. I will suggcst in this articlr that duc to the additional complexity

CALL INNOVATION IN THE ELT CURRICULUM 241of the computcr medium compared with normal classroom activitics, a high standard ofteacher expertise is csscntial. Without this cxpcrtise not much useful learning takes placeantl CALL becomcs a form of ‘electronic baby-sitting’.BackgroundCALL facilities have lxcn available at my teaching institution for many years, starting Lvithan exploratory project to in tigatc the pedagogic value of microcomputers in the ELTclassroom. Since then, informal e\-aluation based on the observation of teachers and studentsusing computers, positive comments in student questionnaires antl informal discussions allsuggest that on the whole, despite a small number of negative reactions from students, usingcomputers to learn English can be enjoyable as well as educational.CA1.L facilitics have grown so that computers feature throughout our course structure.The main computer room houses a network of computers. Students usually work in pairsor groups of thrcc. Timctabling is Ilexihle. Slots are booked, usually a meck in advance byteachers when they feel their class would most benefit. On a 100-hour full-timc course, astudent may spend ten hours using the computer. In terms of a quantitative evaluation,CALL in our centre has had an extrcmcly high adoption rate. Ovcr a ten-year period therehave been between 300 000 and 350 000 half-hour lessons hooked.Thc actual time studentssprnd in front of a computer and the high degree of adoption by teaching staff is animportant visible sign of success, especially as use is discretionary, but it conceals theimportant dimension of quality, bvhich I shall rcturn to later.Teachers arc trained in a number of ways. Each teacher has a short induction giving thema broad overview of CALL and how to use the most popular programs. The teacher is thensupported by written information which offers more detailed help. A CALL co-ordinator(CC) is on hand to respond to questions as they arise, while more experienced teachers passon their cxpcrtisc. A special four-clay training course, the CALLTeacher Education Course(CALLTEC), was also designed. CALLTEC aims to give teachers the theory and practicalcxpericncc necessary f‘or effective CALL use and materials development.The fascination of the computer as machineThe introduction of computers into the culture of language learning is a complex change.When we think of CAI,I., the first impression is of the computcr itself, apparently doingsomething sophisticated mith students peering intently at thr screen. Then \vc may reflectthat the apparent sophistication is a stitchcd-together product of people and systems withtheir inherent flaws. Less obvious is the enthusiast working late behind the scenes trying toensure that the stitchrs arc not obvious and that the thing does not suddenly get out ofcontrol, by making the hardware, software, pedagogy, communications and infrastructurerobust. We then need to add the reactions ofthe users and managcrs: enthusiastic, accepting,indifferent, cynical, nervous or rejecting. Finally, we step back and look at the whole picture,and reflect on how all these interacting elements constitute a new sukulture of languagelearning.It is clear that, together with a fascination for computers, many students rank acquisitionof computing skills alongside thc acquisition of English language as essential for survi\ al inthe modern world. Given thc holding pow-cr of the computcr, it is hardly surprising thatlvc tend to foreground the computrr antl computrr- applications, when we shouldconcentratc more on the interaction between the technology antl the culture of learning.Papert (1 987) calls this tendency technoccntrism ~ making an object the centre of our

242 SIMON SERGEANTattention.Technocentrism is endemic in CALL research and evaluation as well as in the wayteachers, students and managers perceive computers in cducation. It often leads to thcassumption that having provided the opportunity to use computers, lcarning happens liyitself.The ecology of CALL innovationCALL, like any classroom innovation, takes place at many levels. ‘The first important thingis that change is systcmic, that is to say it takes place in an cnvironmcnt which consists of anumber of interrelating systems’ (Kennedy 1988).Kcnnedj employs a ‘wheels within wheels’ diagram in which classroom innovationforms the centre of the whcel, and institutional, educational, administrative, political andcultural levels form progressively outer circles. Chin and Benne (1 976: 33) discuss theproblems of introducing new ‘thing’ technologics (for examplc, audio-visual devices,television, computers) into school situations:As attempts arc made to introduce these new thing technologies into school situations,the change problem shifts to the human problems of dealing with the resistance,anxieties. threats to morale, conflicts, disrupted interpersonal communications andso on, which prospective changes in patterns of practice evoke in the people affectedbv the change.Paisey (in White 1988: 1 16) reminds us that. . . it is people who inhabit an institution, and an organisation consists of networksof relationships tietween people acting and reacting on each other ~ thus organisationscontain rational as \vel1 as non-rational clcments . . . Most crucially, an educationalorganisation is operated by the persons who arc themselves thr instruments of changc.Without their willingness and participation, thcrc will be no change.These writers give some idca of the dynamics of introducing ‘thing’ technologics intointeracting systems and sulisystems, although they fall short of providing a detailcd modelof the curriculum in a state of flux.Innovation or change?White (1 988) defines innovation as ‘a deliberate effort, perceived as new and intended tobring about improvement’. It is distinguished from change, which is any difference betweenTime 1 and Time 2. Delano et al. (1 994) define innovation more narrowly for the ESLcontext in terms of change, development, novelty and improvement. An innovation in asecond language teaching programme is an informed change in an underlying philosophyof language teaching/learning, brought about by direct experience, research findings, orother means, resulting in an adaptation olpedagogic practices such that instruction is betterable to promote language lcarning.Kemmis et al. (1 997) make a distinction between minimal and maximal curriculuminnovation. Minimal innovation occurs when there is a change in the way a particular aspectof the syllabus is presented to students.Thc course will be altcred to accommodate the newidca. Maximal innovation would be evident in a massive reorientation of a course influencedby the CALL aspect of the course.

CALL INNOVATION IN THE ELT CURRICULUM 243First order and second order innovationPcrkins (1985) sheds light on the \I ay in which innovations are minimally adopted ineducation. I IC distinguishes bctvontl order ‘fingertip rffects’ of informationsing technology. First order fingertip cllects are t h c o I I\ .’ IOU\ : c I‘ff I crcncrs an innovation, thr iinmctliatc advantagc put at one’s fingcrtips, such as Iwing able to c‘onvcrsc withfricmds overseas (tc,lrcommunications), or casicr typing (the lvord procrssor). Second orderfingertip effkcts arc the deeper rcpcrcussions of thr innovation. ’l‘he use of the wordor for instance puts a pvcrful tool at thc fingertips of the L2 student of \vriting.lit? to crcatr antl manipulate text easily to move, insert, copy or tlclctc hlocks bel‘oi-cdcciding how the completed document \vi11 appear, libcratcs the \vritcr from linearconstraints and from the chore of reivriting in long-hand.l‘hc ‘opportunitirs get taken’ hypothesis (Pcrkins ibid.) suggests optistudents will recognize the opportunity of large-scale editing. The deeperel‘fects involving a restructuring of the cogniti! e skills underlying thr \vi-itiIic ‘soaked up’ liy assigning \vriting tasks on the word processor. In other words, theopportunity tloes the teaching liy itself. Hou.evcr, Pcrkins claims, ‘Most typically . . .the opportunities arc not taken.’The nature of missed opportunitiesSomrthing mill always be learned when a stutlcnt engaucs in a CALL activity but this may?not be cvrn at Pcrkins’ first ortlcr 1 1. C>pportunities lor the derper second Irvel learningmay also be missed. Considcr the rcsults of a sur\ of perceived program usc among full-time students on a 100-hour intensive gcnrral English course. At thr end ofthc course 200students \vcrc asked \vhich programs they had used and to estimate how many times theyhad used them.Thc results arc shonm in Figure 19.1.One of I’erkins’ criteria lor transfer of learning is a varicty 01’ \vide-ranging practice.This is not occurring since almost 57 pcrcvnt of perceived CALL use is accounted for hvtwo programs: Storyboard’ and Gapmastcr.’Teachcrs arc not exploring tliffci-cnt programs.The popularity of Storyhoard antl Gapmaster may IIc accounted for by the case of enteringtexts into thc programs, or ‘authoring’ .To author Storylioartl, teachers typr in a tcxt (authorthe program) antl save it. The samc applies to Gapmaster. Teachers place the words they\vant to blank out in square hrackets. The trxts iwd are usually rxtracts from studenttextbooks or grammar/vocahularv practicc books.Another v ay ot‘gctting closcr to thr nature ofmissed opportunities is to rclatc the le\-clof actual program use to types 01’ learning genei-atcd by CA1.L. Krmmis ct al. (1 997)distinguish five lcarning stvlcs for CAL, which Phillips (1 985) uses to map the types of’lcarning naturally arising from a particular program typcx. These arc recognition, recall,comprchcnsion, experimental and constructive understanding, In thc first style, the studentis required merely to recognize prcviouslv presented language forms. In thestudent is required to reproduce previously acquired knowledge. Neither rccognition norrecall involve the active construction of ncw kno\vlcdge. The third tvpc, comprchrnsion,involves a more active role antl entails the ability to operate on a Iiody of content andtransform it in some \Yay. Experimcntal learning may involve thc active exploration of asimulation. Languagc production is less consti-aincd by on-screen tcxt. Constructiveunderstanding involves using the coinputrr as a tool to discover nThe most common use of Storylmard is for students to retricpreviously encountered in their textbook. Storyhard contains a ‘cheat’ feature \vhich means

244 SIMON SERGEANTStor) boardGapmasterWord procc-ssingVocal) GamesTcstmastcrGrammar GamesClarity GrammarPinpointFast FuodWord st o reFCF FxcrciscsMatchmastci-London Adventure12666145773221501169449353422131038.318.617.59.84.53.52.81.51.11 .o0.70.40. 3Instructional /conjecturalInstructionalEinancipatoryInstructionalInatructionalInstructionalInstructionalInstructional/conjecturalKcvelatoryInstructionalInstructionalInstructionalKcvelatory* Instructional: Icarncrs irecall \I hat has Iiecii taughtRevelatory: Icarncrs take part in a rclati\cIv struc.turcd learning situation, c.g. a simulationConjectural: lcariicrs cngagr in tasks \\ ith opcii-cn(lcd, unprcdictalilc solutionsbmancipatory: Icarncrs cng"gc in authentic, real ad\itivsFigtrrc /'I. Il'crccivcd program usethat at any time a student ma: see the entire tcxt again lvithout a penalty.Thr same appliesto individual words. Both thcsc stratcgirs arc uscd liy studcnts to reduce learning load.Though teachers intcnd this activity to improvc comprchcnsion, the type of learning arisingfrom this activity is usually at thc I 1 of recognition or rccall. Copying a text verbatimmay help students to rcmcmlier words or syntactic structures, spelling may improve, andit is probably more fun than copying a text using pen and paper.If they work on a Storylward acti\-ity collaborativcly, students may learn somethingfrom the language they use to complctc the text. though research on the nature of' talkgenerated in front of CALI, programs summarized by Nicholls (1992) and Nicholls' ownresearch on Storyboard in particular suggest that conversational spin-off is limitrd. Thediscourse produced is impoverished in terms of lcxical and syntactic varicty, vith manysingle wort1 utterances and repetitions of screen text, and it is of limited pedagogical value.Gapmaster is most frcqucntly uscd in the drill-andpractice mode. An cxcrcise from agrammar textbook is typed in, for instance to practise question tags. The outcomc is fixedand non-ncgotiablc. Thc facility of the program to accept more than one correct answerrcquircs more effort hy a teacher to author thc altcrnatives (enter the tcxt rcquired) and isoften not uscd. The off-screen interaction is limited and the learning is at the level ofrccognition or rccall.The problem of opportunitics for learning not tieing taken deepens when thr mode ofthe CAI,L cxpcricncc is considered (Figure 19.1). CALL in the instructional mode accountsfor 8 1 percent of total use, ivhercas CALL in the revelatory mode accounts for 1.4 prrccnt.ing accounts for the total use of CALL in the cmancipatory mode at 17.5percent. CALL in thc instructional mode involves no negotiation of outcome. The aim ofactivities is for the student to produce tcxt which has hccn prc-determined before the

C A L L INNOVATION IN THE ELT CURRICULUM 245activity liegan .They involve the manipulation of language in ways which do not involvc anyexchange of meaning.Transformation exercises antl controlled pattern practice are activities\vhich involvc thc production of language but not the use of language (Willis 1990). Thisapproach is thcrcforc at odds with current communicative language teaching mcthodologyrts that people learn a language best by using the language to achieve realmeanings and outcomes. Underwood (1 984) commcnts that CAIJ, in this modc trics tosimulate what the tcachcr docs in the classroom to be exact, thc least interesting things.It tends to be authoritarian, evaluative and overly structured. The shortfall between thrpotential learning opportunities that could be rcalizcd and the reality of the way programsarc frequently undcruscd is obvious.Summary of the problemTeachers with a low level of CALL, expertise arc less likely to be ahvarc of the rangc ofopportunities offered by authoring and using \%-ell designed CAI,[, activities, though somenovices make up for this by being enthusiastic and crcative liecause they do not havrpreconceived ideas. During a four-hour lesson, migrating to the computcr room is awelcome change of ne which tends to giw thc- tcachrr and students a break from eachother. Once a task has hcen set, the teacher can take a more passive rolc, offcring guidanceonly when requircd, sometimes not at all. Tcachers take a technocentric viewpoint andassume that thc minimal task imposed by the program, whcthc~r gap filling, testreconstruction, or intcracting at a basic level with a simulation, constitutcs a worthn.hiletask.Each timc teachers use a new CALL activity, it represents a micro~innovation.Tcacherswill usually make a cost/benefit calculation based on how much benefit their classes \vi11reccivc from half an hour in front of the computer offset against the amount of effort andrisk involved.The effort teachers need to put into lcarning a new program and training theirstudents to take part in an activity will be calculatcd.The risk of failurc is an important partof thc calculation, based on perception ot the reliability of the hard\varc antl the complexityof the program.The more complex a program is, the more a tcachcr will fear thc failurc ofthe activity due to someone pressing thc \rrong key or cntcring part of the program thatthe teacher has not yet discovered. A number of personal failures, for example with Lvortlprocessing, may disco gc a tcacher from using valuable activities, and staffroom anccdotcsabout such experien will discourage other teachers. As a consequence, thc safer, lesscomplex activities tc o lie favoured by the majority.Certain factors militate against the use of more time-consuming integrated activitiessuch as simulations, which involve the class and the teacher in learning how to use a programthat has less repeat value than a text reconstruction activity. This type of program involvesmore preparation and time in terms ofpre-CALL and post-CALL activities in the classroom.There is the need to complctc the textbook material prescribed for the leespecially the case where classes are shared by more than one teacher, so ‘extra’ activities,which are less obviously related to course content, may be less used. There is, thcrcfore, adanger of over-using a small number of programs and requiring students to use the sameprogram (with different texts) repeatedly.A similar cost benefit calculation applies to the creation of materials. Certain programssuch as Storyboard arc easy to author (enter text) and highly productive in tcrms of theratio of authoring timc and effort to the amount of student use. Storyboard has a consistenthistory of almost 100 percent relialilc use, so there arc few lost lessons. Under conditionsof teacher ownership, materials are Lvrittcn into courses using these easily mastered packages

246 SIMON SERGEANT\vhich produce as much student ‘liusyncss’ as Iiossililc tor the least effort in materials writingor lesson preparation.Most materials exist only as texts. Thcy arc supplcmcntary to the tcxtlmok matcrials.They arc easily authored materials ~1rittc.n into courscs, so that a particular unit in atextbook may tic supplcmcntctl \\it11 a text reconstruction activity, a vocabulary activitvand/or a gapfilling acti\ it!. Thcy arc \\ rittcn into the teacher’s notes, antl Iiecomcinstitutionalized, fixed supplcnicntarv clcmcnts Ibr a Iiarticular coursc. The syllalius thenIiecomes resistant to more integrated acti\ itics in cniancipatorv or rc\clatory motlcs, suching or simulations. At this stagc it is tlifl’icult to alter the materials orintroducc a \vitlcr variety of prograins.The Iireliondcrancc of CAI.1. inaterials in the instructional mode (see Figui-c 19.1)reflects the nature 01’ the \\ idcr syllalius, pritnarily tlcri\-ctl from textbooks with astructural/functionaI ordering ol‘itcnis. In the \vitlcr syllalius thcsc structures and functionsarc supplemented \vith further materials 01’ the same nature. A nunilicr of communicativeactivities are also a\ ailalilc, but arc considered secondary to the process ol teaching thesubject matter of the syllalius. This is also rcflcctctl in the Iialancc of CAIl materials. Theprc\alcncc of the supplementary usc 01‘ CAI I. tends to tldinc the normal level of CALLuse, \vhich is the typical Icvcl of’atloption of‘ the majority of’tcachers.CALL implerncntation strategiesCALL expertise is a complex skill \vhich can lie acquired liv various means. In an idealsituation, the CC (CAI.1- co-ordinator) gains cxlicrtisc hv studying the ticltl intcnsivrly, bytalking to other practitioners antl by everyday olisci-vation antl practicc. A selection of thcscskills are simplitictl antl translcrrctl to teachers through in-service training in various formsantl through dealing mith c\~ct-v(Iay Iiroblcnis and qucrics. A similar sclcction, simplificationand transfcr of skills takcs placc Iict\vc.cn tcachcr and stutlcnts.‘I‘hc CC, in his/hcr efforts to cnsurc cfl(xtivc CALL Icssons, is in thc position of coordinatingthe interaction of two highly complex systems: net\\ orkctl computers and thestaff within the organintion (scc Appendix Ix-lo\~ ).XI cnsurc adoption, the CC can thereforeivoi-k at the ‘thing’ 1 1 or at the ‘person’ Ic\-c.l. Working at thc ‘thing’ Icvcl lcatls to case ofaccess for all users: students, tcachcrs antl niatci-ials \vritcrs. Working at the ‘person’ Icvclin\-olves creating antl maintaining a flo\v of information t,ct\z n all stakeholders \z ithin theinstitution, both users (students antl tcachcrs) antl non-uscrs (managers, technical staff,atlniinistrativc staffOn a day-to-tla: Iiasis, thc CC makes decisions about the most cffcctivc lint. of work,ivhrthcr to focus acti\ity from thc bottom-up of the system (c.g. materials development.bvr i ting cl ear ins ti- uc ti ons / tlocunic n t a t ion) or from t he top - don n ( c . g. t cac he r training,maintaining the goodwill of the managcmcnt). E\.crytlay priorities usually in\-olvc thcbottom-up approach, dealing \\ ith Iiroblcms as they arise, \\orking under the assumptionthat if things arc running smoothly, thc good\vill of the managcmcnt is assurd.Impro\ cnicnts made to the system, materials antl instructions arc permanent, \z hcrcastraining antl retraining is a constant rcquircnicnt lbr nc\v staff or for those requiringupdating. Most of thc timc it is more Iirotital)lc to locus on Iicrmancnt improvements. Forexample, something can hc made easier for teach-s to usc, Iivrhalis simplifying a procedurebv a single key press, or \vriting clearer instructions. Il‘this is multiplied hy 40 staff or 2000studcnt users, it means that far less training is rcquirctl.Thc ldl implementation of CALL is a lengthy Iiroccss. Five ycars \vci-c nccdctl in ourccntrc for the institutionalization of a minimal Icvcl of CAI.1.: to set up system structurcs,

CALL INNOVATION IN THE ELT CURRICULUM 247to source soft\\ arc, to providc instructions to teachers, to author high cluality materials antl~vcavc them into the structure of courses. Eight years m ~ recluii-et1 ~ c before our ccntrcti thc standard of' implcmcntation antl expertise rcquircd to gcncratc a teachertraining course such as CALLI'EC.Summary of value-adding activities'I'hc summary in the Appendix to this articlc illustratcs thc ua!s in \\.hich a teacher Lvithrcsponsilility for CAI .I ma: add \-aluc to a CALL facility hy \\-orking \vith managers,tcachcrs and studcnts cithcr at the thing (system) IC 1 or at the person 1cWhcrc only a sniall amount of non-teaching time is made availal)lc, the main focus ofactivity tcntls to shift alvay from teacher training to other consitlcrations which, Tvhilc thcyarc morc mundanc, are the sine clua non oi' CALI,: trouldc shooting, software acquisitionantl installation, timctabling, maintenance of materials (printing, cataloguing, dclction, filchackup) antl therefore the lcvcl of use tleci-cases in scope antl quality. The pro\-ision of morctime allo\vs the CC to focus on value-adding activities 1% hich arc lcss conccrnctl nit11 theday-to-(lay running of the facility such as teacher training, tlc\-cloping interesting matcrials,c\.aluation and self-education,ConclusionWith insufficient management, thc Icycl of CALL use is likclv to tlcclinc.Thc change agent,in this casc the CA1.I. co-ortlinator (CC), is ccntral to the ~~~wcess of ensuring that CALLoperates smoothly. The CC deals lvith practical problcms, antl ensure5 that the innovationis at least minimally implcmentetl.Thc CC can minimize problcms faced I>! users of' CALL.by 'atltling \due' tc tcm at iarious lcvcls. On a larger scale, the actions ofthc CC arcpivotal to thc pi-ot igniiicant curriculum change. 'I'hcsc actions are rcsponsihlc torfacilitating conscious learning opportunities by ensuring that CALL learning exists, antl thatmanagers, tcachers antl students rccognizc these opportunitics antl take them.To crcatc antl maintain the CALL facility in good working ortlcr requires a pi-ofcssionalchange agcnt: the CAI.1 co-ortlinatoi- (CC) or a team of' protcssionals with a high tlcgrecofcxpcrtisc in CAL.I..Thcy can intcrprct CALL use in tcrrns of current mcthotlology, define,crcatc antl maintain high quality learning structures antl communicatc their potentials tomanagers antl users simply antl cffectivclv. The CC is rcsponsihlc for the creation andmaintenance of a student learning niche within the cui-riculum. Ideally, thcrcfoi-e, thecxpcrtise of the CC as change agcnt should include at least a rudimentary appr-cciation ofho\v CALL is cmhcddctl in the curriculum antl ho\v to tnanagc the innovation.In this article, I havc cxplorctl the nature of' day-to-day CALI value-atltling activitywithin the context ofthc CALI, facility \vhcrc I work. I put for\vartl possildc strategies lotdealinguith prohlcms arising from the institutionalization of a minimal Icvrl of CALL us~~.The resolution of these problems is sccn as a precontlition for maximal benefit to the ELI'curriculum.

248 SIMON SERGEANTAppendix: value-adding activities1 Working with non-users (managers)InfuencingInfluencing thc private evaluation ol CALL I)>hardware/softm arc.managers, on the level of finance andEchnica/ mattersReassuring managers that the technical performance of the system is robust and reliableMaterials developmentEncouraging managcrs to have an active interest antl investment in materials dcvchpmcntfor CALL.Eacher trainingkncouraging the management to initiate antl dc\clop various forms of teacher training.ConimunicationrImprol ing the information flou bctu ccn CALL personnel and manager?.EvaluationEvaluation b? managers of CALL on thc lcvcl of consumcr satisfaction, observableorganizational change antl flow of communication.2 Working with users (teachers and students)Mbrking at the y5tem laelDesigning and programming the s)stcm to make it reliable and transparent to use, andidcntifying thc nrcd tor new harth arcSoftware evalciation c7nd acc~uititionInitiating the purchase or design of ne\v software antl submitting it to materials writers forevaluation.GLlidlng C/ILL ureAdministering the timetable. Writing instructions and manuals to support CALL use.Cataloguing and publishing materials in a form that teachers find useful when planninglessons.,Ilaterialc tlevelopmentWriting materials and model lessons antl supporting teacher5 1'1ho arc authoringcoursemare.Eachcr trainingInitiating and developing tcachcr training, ranging from presentations and markshops toresponding to the day-today questions of individual teachers. Training may be cithcr in thcuse of existing activities or in thc creation of materials.

CALL INNOVATION IN THE ELT CURRICULUM 249E I a Ilia t I onE\aluating thc lc\el of CALL usc and thc contribution CAI I can make to diffvrcnt cour5c-5.InpuencingChanging the may teacher5 think ahout CALL.This le~cl is the prilate domain of thc teacher,concerned \\ ith how teachers relate to CALL and the \\a> CAI I is integrated into a lessonat the planning stage.Notes1 In Storylmartl, students have to rcasscmblc a tcxt which has 1)ct.n deleted from the2ScI-cCn.Gapmastrr is a form ofcloze procedure, \vith students filling in missing uortls in a text.References and bibliographyKrurnfit, C., Philips, M. antl Skehan, 1’. (eds) 1985 ‘Computers in English LanguagcTcaching:AVien from the Classroom’. In British Council Occasionu[ Papers no. 122 (British Council/Pergamon)Chin, I. 1976 ‘General stratcgics for cffccting changes in human systems’. InBennis,W. G., Rcnnc, K. D., Chin, R. and Corcv, K. D. (cds.) 1976 The Planning of’Chnnge(Holt, Rinchart antl Winston, Nc\vYork)Delano, I.., Riley, I,. and Crookcs, G. 1994 ‘The meaning of innovation for ESL teachers’. InSyxtem vol. 22 no. 4 (I’crgamon)Kcmmis, S., Atkin, R. and Wright, E. 1997 ‘I low do students learn?’ (Occasional Paper no. 5,CAKE, University of East Anglia)Kcnncdy, C. 1988 ‘Evaluation of LL‘I‘ projrcts’. In .+plied Linguistics i.01. 9 no. 4: 329 42(OW)Nicholls, L. 1992 ‘Computers as a stimulus for talk: the nature of talk gcncratcd by pairs ofstudcxnts using StoryIx)ard’. In 0.L-CXL vol. 9 no. 2: 19-29 (University of Queenslantl,Australia)Papcrt, S. 1987 ‘Computer criticism vs. technocratic thinking’. In Educutionnl Reieorch 17:22-30 (NE‘EK, Koutlcdgc)Pcrkins, 1). N. 1985 ‘The fingertip effect: hon information-processing technolog? shapesthinking’. In Etiticutionul Reseurch 14: 1 1-1 6 (NFER, Routlctigc)Phillips, M. 1985 ‘Logical possibilitics and classroom scenarios for thc tlcvclopincnt of CALI,’.In Brumfit, C., Phillips, M. antl Skehan, P. (e&) 1985Untlcr\vootl, J. I I. 1984 Linguistics, Cunip~iters und the Lcinguuge Eucher: ci Cornrnunicatir,e .,lpprocich(Newbury HOLISC)White, R. 1988 The El T Curriculum: Desip, Innoration and .Ilunuiyerncnl (Black\vell)Willis, D. 1990 The Le~iciil Syllabus (Collins)

PART FOUREvaluating curriculum change

ofChapter 20Pauline Rea-Dickins and Kevin GermainePURPOSES FOR EVALUATION1 IntroductionN U M B E R 0 F D I F F E R ENT P U R P 0 S E S for evaluation can lie identified. TheyA can he divided into two broad categories:1 general purposes2 specific, topic-rclatctl purposes.The main gentd purposes arc examined first.2 General evaluation purposesE\ aluation ma) bc undertaken for thrcc principal rcawns1 accountability2 curriculum de\ elopmcnt and betterment3 self cle\elopmcnt tcac her\ and othci language teaching profc\\ional\Evaluation for purposes of accountabilityEvaluation for purposes of accountability is mainly concerncd with determining whetherthere has bccn value for money, in other words \z hethcr something has been both effectiveand efficient. The main aim is to report on a product and givc an evaluative judgt,mcnt,whether something is intrinsically a ‘good thing’ or not. Generally the information tlcrivcdfrom evaluation for purpo 1 accountability is not used in any niajor \vay to improve thcfunctioning of the curriculum or classroom practice. Kathcr it informs decisions as towhether something is to continue or be discontinued. If, for example, sponsors or heads ofinstitutions are not satisfied with thc implementation of a particular project, then cuts maybe made.Thus, if a particular reading schcme is introducctl, cvaluated a year later, and thenjudged to be ineffccti\-e, it is highly likely that a school \\ill discontinur supporting thisventure.E\ aluations of this type are largely, although not exclusively, the domain of policymakcrs or pro\-itlrrs of resources. There is a close link bctwccn pmver and cduation foraccountability. There arc other points to notice. Usually, such evaluations arc carried outafter an innovation has been running for some time, or at the end of a project. This type of

254 PAULINE REA-DICKINS AND KEVIN GERMAINEevaluation, knoivn as summcitii’c L‘I illtrotion, has also tended to in\ olvc testing and mcasurcmcnt,antl anal! scs of thc statistical significance of results olitainctl. It has focuscd on theo\.crall outcomes, i.c. end product ol’an inno\ation, antl has consistently failed to takc intoaccount tcachcrs’ cvaluati! c comtnc‘nts. Sumniati\-c cvaluations arc liinitctl I)! their focuson outcomes at the cntl of an educational inno\ a t. ion.Evaluation for purposes of curriculum developmentTcachcrs have a kc? rolc to pia! in thc curt-iculum r-cnc\\.nl and dc\.clopmcnt process. Theideas of Stcnhousc (I 975) \\.crc piiota1 in placing classroom practice at thc fot-cfi-ont of’curriculum cnquirics:Fi-otn the first set oftrials it \vas learned that information coming from children’s testresults \vas tentati1.c antl not readily usable for guiding re\\.riting tvithout beingsupplcmcntctl liv other data. Thc rcsults pla!ctl a USC~LI~ part in confirming that thcgctwral approach ofthc materials \\as cttccti\.c in promoting achievement of its statedobjccti\q antl the dcvclolimcnt of tests also had side-lmdits for the production ofUnits. Rut for indicating changes which \zouId make thc Units morc effective the?\\ere of much less use than information from other sourc(~ . . .Whilst it could not be said that the test information \vas without value for thisProject, it can I)c said that \vhcrc resources arc limited antl it is necessary toconccntt-atc itlion gathering inli)rtnation to give thc greatest rcturn on mom?, timcantl human energy, then the choice nm~ild lie for tcachct-s’ reports and tlircctolxcrvations in the classroom antl not for tests of short-term changes in children’stwhaviour.(Harlcn 1973: 91 92 cited in Stcnhousc 1975)t val u a t i on for CLI r r i cu lu ni t l eve I o p in c n t pur 1x ’ \vi11 involvc information fromtcachcrs an(\ other relevant El :I‘ l)rofcssioiials. It portant in the management ofevaluation to inclutlc all relevant partics. From this it follo\vs that tcachcrs have majorcontributions to make in the c\.aluation of‘ classrooms. It is the tcachcr, rather than the‘trstcr’ or the evaluation ‘expert’, ~ h has o most information alwut specific classroomcontcxts. This information may lie rcportctl at various times antl in various f’orms, forexample as rcymnscs to qucstionnairc,s, intcrvic\vs, records, or diary kccping- It ma? Iwlargcly descriptive antl qualitative, and ncctl not entail tests, measurements, and intcrcnccsabout curriculum quality from statistical data. In contrast to summative ?valuation forpurposcs of accountability, cvaluations intended to improve thc curriculum will gatherinformation from tliffei-ent people ovci- a period of timc.‘l’his is known as,fhrrnutive evuluution.Such evaluations arc ongoing antl monitor tl lopmcnts hy identifying the strengths antl\vcakncsses of all aspccts of teaching antl learning. As opposed to mcrcly passing an cvaluatil-ejudgement on the cnd product of a tcaching programme (summativc evaluation), formativce\ aluation is drsignctl to pro\-itlc information that may lie used as the basis for futureplanning and action. It is formative sincc it aims to strengthen and improvc the curriculum.Evaluation for purposes of teacher selfdevelopmentA third antl major rolc that evaluation has to play is in formalizing and cxtending a tracher’skno\vletlge ahout teaching antl learning in classrooms. This is sometimes referred toas illurninatii,e evuluution (Parlctt antl Hamilton 1987) Iiecausc it involves raising the

PURPOSES FOR EVALUATION 255coiisciousncss of tcachcrs antl otlic-1- tL'1' practitioners as to \\.hat actually happens (asopposed to what is suppscd to happen) in the language teaching classroom. This t) pv ofcvaluation is also tlcvclopmcntal antl formative.Evaluation of this kind is tlcf'initc~lT not conccrncd \vith mcasurcnicnt. 'I'hroughalvarcncss-raising activities, tcachcrs arc involved in dcscrihing antl Iwttcr understantlingtheir o\\ n contexts \vith a \ic\\ to improving thc teaching ant1 learning procc\.aluations arc both illuminative ant1 formati\e in purp)sc.Thc! focus more on thantl less on the product, of tcaching antl learning antl have a tnajor role to play in tcachci-SummaryIn this section \\ c hac examined the general purpows foi e\ aluation (accountal)ilit>,lopmcnt, antl tcachrr self dc\clopment). Accountabilit: 15 us\\itli sumniati\c e\aluation 1% hilc cui riculum tle\clopmcnt and teacher sclt tiarc Ixttcr informed 11) c\ aluation as a formati\c pro~ccc1 . 13 Specific topic-related purposes for evalu a t' iontion, ivc t-xaniinc \va!s in \vhich evaluation is important to classroom tcachcrsantl ho\v their a\varcncss can 1)c raised by evaluation acti\-ities. We arc concerned 1% ithformative and illuminative evaluation, associated Tvith curriculum drvc~lopmcnt antl tcachcrself-development.What is meant by thc cni.ironmen/ ofthc classroom since it is this that pro\.itlcs the focusfor our cduations? Thr cmvironnicnt is macle up of man! things including the socialorganization of thc classroom, the tcxtlmoks, the inodc(s) of teacher presentation, antl theresourcrs available to the teacher. Thus, \vhcxn \vc ask the question 'Do our teaching antlItm-ning Ixogrammcs \vork?' (i.e. arc they cffcctivc?), \\e need to itlrntify clearly the focusof OUI- question. Are \vc intcrc~stctl in evaluating the classroom organization, aspccts ofntation, or is thc focus on the textbook or thc \Yay tcach grammarlThcc\aluator has to he clear as to \\.hat is hcing cxamincd.Kcv factors arc that:123c\aluation is not restricted to the testing of Icarncrs' alditicsmore than just thc end product is important \\ hcn cvaluating a learning Iirogramnicthcrc arc diftercnt conditions that may explain, or contrihutc to\\ artls an cxplanationof, why a prugrammc is su4 other information, rclatcd Ilkrcnt aspccts of the teaching antl learningpi-ocess, should be incluclctl in an evaluation of the. curriculum, to complcmcnt (latatvpically tlcrivetl lrom a test analysis of learner pcrformancc.In other ~vords, the varied processes of teaching arc as important as the outcomc of learning,antl in 01-der to improve learning outcomes, there is a need to examine more closclv thoseconditions that ma? contribute to succcssful language learning. What is ncctlcd is a d edctlexamination of the cnvironmcnt created by the teacher antl Icarncrs to promote cffcctiwlanguage learning. Evaluation is thc means hy lvhich \vc can gain a Iwttcr understanding of\&at is cffcctivc, \\-hat is less cttc.cti\.c, antl xvhat ai1ptw-s to he of no use at all. In ortlcr todo this, evaluation must focus on both the means antl thc product of the learning pro .

256 PAULINE REA-DICKINS AND KEVIN GERMAINE4 MaterialsBefore anal) sing the extent to \\ hic h gi~en tcaching and lcarning matcrials arc suitable,there arc preliminary questions to address rhc matcrials selected tor cla5sroom ure can hcdefincd in a numbcr of c\ a! 5What do materials meanforyou?1 110 you refer cxclusivelj to textbooks, or do you includc teachers’ guides, teachingmanuals, supplcmcntarv units, rvatlcrs, audio and visual materials, etc?2 L)o you make a distinction hctwccn materials dcsigncd spccilically for first and secondlanguage teaching, and also I)ct\vccn those targeted specifically for use in school andmaterials that arc non-pedagogic but authcntic?3 Do you include materials produced by thc teachers and the learners?The role of materials within your teaching and learning context12What rolc(s) arc the) cxpcctcd to play?What goal(s) arc they expcctcd to achie\e?How are the materials to be used?12Are the) to be uwtl as thc sole 5ourcc and rcwurcc for teaching?Arc the\ one of scvci-al a\ ailablc resources?There has hccn a tendency for overreliance on classroom teaching materials, withunrealistic expectations made of thcm. However, the cff‘ectivcncss of teaching and learningis not cxplained solcly in terms of how good or bad the learning materials are. As Allwright(1 98 1 ) suggcsts, materials are only purr of thr. co-operative management of languagelearning. It is also crucial not to ovcrrniphasizc the importance of learning materials.Evaluation of classroom learning materialsThe first xvay in Lvhich materials inay I)c cvaluatcd is in terms of’ how wcll they reflect theprinciples by M hich they havc Ixxn writtcn. In the case of class textbooks, the evaluationcriteria will be those used lvhcn tlcciding \vhich hook is best for your teaching context.When it comcs to teacher-made materials a specification, i.c. a list of criteria against whichto evaluate the materials, is indicated at thc outset ~ or accumulated during the proccss ofmaterials writing and is thus ‘known’ to the tcachtxr. In both cases, \ve arc rcfcrring to thetheoretical worth of the materials.Examining the materials as they stand, that is Lvithout rcfcrence to their actual use inthe classroom, gives us no information about how these materials actually work with a class.This distinction between the theoretical (i.c. construct validity) and empirical value ofmaterials has becn explored by Rrccn (1989), who distinguishes three phases in theevaluation of materials: materials~as~\I.orkplan, matc,rials-in-proccss, and outcomes frommaterials.We can generalize from the notion of ‘tasks’ to the notion of teaching and learningmaterials in the following manncr. ‘Materials-as-workplan’ refers to the theoretical valueofmatcrials, taking up the range of points covered in comprehensive checklist. Rut, as Breen(1989: 189) statcs:

PURPOSES FOR EVALUATION 257Workplans can only provide opportunitics for change in knowledge and capabilityand for successful outcomes in relatively unpredictable antl broad measure.‘This brings us to an evaluation of ‘Materials-in-pro .’. This stage gcncrates informationabout the \rays in \vhich learners antl teachers actually use and rcspond to materials, thusproviding indicators as to whether the materials are ‘succcssfuI’ or not.Thc ‘outcomes fi-ommaterials’ rcprcscnt the relative achievements of learners.Who evaluates the inaterials is the final point to he considcrcd here. Lon. (1987)comments on the range of individuals connected \vith a language learning course andexamines both the nature antl purpose ofthe evaluations they arc likely to make. For I.o\z-,ten tliffei-cnt perspectives on materials evaluation could be offered:The Lcarnei-The ParentThe TeacherThc Head or College PrincipalI’hc Teacher TrainerThe Curriculum Committee MrmbcrThe InspcctorThe Educational Rcscarc hei-The IkhlishcrThe Materials Designer(Low 1987)U) examining the role(\) of these p i ticipating groups in the materials cx\aluationprocess, Lon (1987) shou s hon each group \r 111 hac different interests and differentquestions to be ansirered For examplc, a materials writer might carr: out a formati\ee\ aluation designcd to indicate 1% hcthcr the texts arc appropriate to the targct learners, antlthen make the nccessar! changes A parent, on the othcr hand, might onl) be intereyted inexamination results, M hic h imphes a summati\ c asscyymcnt ot learner pcrformanccOn the premise that ‘teachers ma; lieneht greatl) in the eialuation, design and usc ofmaterials b\ cngaging the help and r IC\\ s of learners’, Rrccn and Cantllin suggest a! s in\r hich learners ma) participate in the e\aluation of materialsOn the procedurcs for tiorking with tasks and actir.itie.7 in the clossroomWhat do you find are the most useful wa to learn a new language?What arc the Iwst kinds of language lcarn tasks and activities?What arc the reasonsfor your choice?What can a tcachcr do which \z-oultl hclp you most \\hen you arc learning a nc\vlanguage?What can other learners in thc class do which would hclp you most whcn you arclearning a new language?What is your favouritc kind of language lesson?What are the reasons for your choice?What are the good things and the had things about learning a language in a classroom?What can materials best provide you with to hclp you learn a new language!What arc thc best kinds ol’ language learning materials?What do they look like? Why do you think they’re best?

258 PAULINE REA-DICKINS AND KEVIN GERMAINEWhat is good and not so good a hit the tnatcrials you arc \vorking Lvith no\v? Whatdo you think is missing from them?What changes \vould you make to them?(13rcc.n antl Candlin 1987: 27)As with most in\ cntorics ol’this kind, the qucstions can Iic reformulated to make themmorc relevant to individual contexts. It is important to recognize the diffcrent and relevantcontriliutions to materials evaluation. As I.o\v (1 987: 27) maintains: ‘the evaluation of alanguage learning programme, or the materials used to tcach it, invol\cs morc vie\vpointsthan that of the “intlclien(lcnt” outsi(lc olxervei-’.Summa iyIn evaluating materials it is ti ai-\ to cxaminc the Ivays in \vhich teaching and learningmaterials arc sensitive to the uagc Icai-ning lira< . Evaluation criteria should relatenot only to thc aims antl contents of language Icarning, hut also, and importantly, to theprocctlurcs lor lvorking with texts and Iicd;)rming tasks in the classroom. It is necessary toanalyse learner outcomes, hit not to the exclusion of cvaluating other aspects of the teachingantl learning process. From this \\.e mav conclude that a comprchcnsivc evaluation of‘ ourteaching antl learning materials \vi11 cntail a thcorctical (i.e. \vorkplan) and cmpirical (i.e.process) analysis of materials, the data I’rom lvhich \vi11 bc augmented Lvith details of learneroutcomes. The importance 01‘ ohscrvational (lata, derived from an analysis of materials inuse, should not he untlcrvalucd.5 Teachers and teachingPurposes for classroom observationF,Yaluation is a crucial part ol’ teaching, Iiut how is it done \vcll!Testing knowledge of theoryis not cnough to judge clfectivc teaching. Wc need to ol)scrvc tcachci-s in action using thrirkno\vletlgc in the real sctting of thc classroom. Classroom oliscrvation givcs us thropportunity to see tcachcrs putting theory into prac.ticc: it shmvs us what tcachcrs tlo ratherthan \\.hat thrv kno\v.Grading teachersYour o\vn tcaching context \\ill influence your view on the ~iuqioscs of‘ classroomol)scrvation. The cxpcricncc of many tcachcm suggests that it is primarily summative inpurpc, incxtrica1)ly Imund up \vith reporting a grade, accountability, and jutigcmcmtalstatements.The focus \\.lien grading tcachcrs using a checklist is mainly on the product of teachingantl Icarning, for example, ‘ability to cstalilish rapport’, ‘suitalility of matrrials andmethods’, ‘use of aids’. Also, thc chccklist is used by an cxtc-rnal oliset-vcr.T!’pically, tcachcrsarc \.kited by inspectors \vho check thcir class rccor~ls and lesson plans, obscrvc a Irsson,antl then comment on the lessons using a chcc.klist as a guide. Sometimes this evaluation iscar-rictl out without much participation hy thc tcachcr ivho is actually Iicing ohscrvctl.I. ‘ .I

PURPOSES FOR EVALUATION 259Teacher developmentUsing observation merely to grade teachers, for example, with a vicw to promotion, isextremely limiting. It is important to use observation to provide information that teacherscan use as a basis for future action. Here we refer to the formative value of classroomobservation where the feedback from evaluation will he used to further develop or improvcan aspect of classroom practice, or as part of curriculum bettrrment or tcacher selfde\elopment. Consider this following way of evaluating teacher performance:The classroom assessment process should consist of three stagrs:1 prc-lesson matcrial2 the lesson3 the trainee’s post-lesson evaluation1 Thr pre-lesson material includes:(a) information about the class: descriptive and evaluativc(b) the scheme of Lvork and the place of the assessed lesson in it(c) the lesson plan (normally accepted form)2 The Icsson; attention to the following:(a) classroom pcrsonality(b) classroom management(c) awareness of learners(d) what is being sought(e) how it is being sought(f) what thc learner is doingThe above should be gone into in detail.3 Post lesson evaluation; written self-assessment on:(a) the lesson plan(b) the major headings on the asse5smcnt schedule(c) any additional releiant points(James 1983)In this approach not only is the tcachrr formally included at stage 3 by means of awritten self-assessment, but also there is an attempt to examine the process or teaching andlearning. The category ‘what the learner is doing’ could highlight, for example, the natureof the interaction (teacher to learner, learner to learner, learner to teacher) or the type ofwriting that the learners are doing: copying from the blackboard, filling in a gapped passage,reordering words and sentences). An item on a checklist which focuses on ‘how’, i.c. whatthe teacher is doing, can also identify a wealth of information about the teacher and teaching,for example, ‘What are the different question types that the teacher uses?’ ‘How are visualaids used at the different stages (presentation, practice, or production) of the lesson?’Checklist items such as these focus attention on details of the teaching and learning processand provide information that is useful in terms of modifying and improving classroompractice. It is, therefore, an example of formative teaching evaluation.Peer teaching is an alternative mcthod of evaluating teachers in training. Here trainees‘teach’ a lesson to their colleagues.Tutor and learner observers look out for specific pointsin the teaching practice. Feedback can come both from the trainer and fellow learners.Another way is using microteaching. In its simplest form a trainee teaches a group of learnersfor a short period of time, for example, fifteen minutes covering a specific topic or skill

260 PAULINE REA-DICKINS AND KEVIN GERMAINE(apologizing, reading for specific information, ctc.). Again, peers and/or a trainer observethis performance and comment on it using a checklist as a guide.The observation involved in the above practiccs can he used for improving thc teachers’techniques, monitoring their progress, antl counselling thcm on relevant aspects of theirteaching, However, in many ca. , they arc primarily geared towards training and grading,in other words, used to determine Lvhcther thc training institution will qualify a teacher,the syllabus is being covered, the teacher uscs the appropriate methodology, and so on.Additionally, not only is the olxcrvation largcly controlled by someone other than theclassroom teacher, but checklists may reflect an cxtcrnal observer’s judgement on what iseffective teaching. Thcrc is a need to consider lvays in Lvhich teachers themselves maybecome more involvctl in the proccss of evaluation.Teacher self-developmentA more participant-orientated cvaluation through observation is important in raisingteachers’ awarcncss, a key feature ofthc teacher development process.Taking microteachingas an example, \vc can consider ways in hvhich evaluation may be made more illuminative,collaborative, and useful in terms of tcachcr self-dcvclopmcnt.The proccss can be examinedat three levels: self-evaluation, peer evaluation, antl collaborative group work. We shallexamine these in detail.Teacher selfevaluationSelf-Evaluation is simply thc practice of tcachcrs reflecting on what has taken place in theI... tsson with a vie\v to improving their performance. It can lie very informal, for example inthe form of brief notes written immctliatcly after the lesson. Or it can be part of a written(such as the class record) or oral rcport on thc lesson itself. Alternatively, a checklist canbe used. One of the advantages of self-cvaluation inventories is that they can be designedby individual teachers to suit thcir own tcaching contexts. They are relatively simple to useand pet potentially they can providc a wealth of information about teachers, their teaching,and thcir learners.Peer evaluationPeer evaluation can be incorporated into microteaching where several trainees arepresent during the lesson or where they share the same microteaching session. Here it isimportant for there to be somc means of encouraging open and constructively criticaldiscussion.Now, consider the following procedurc (adapted from James 1983) which may involveboth teacher self-evaluation and peer evaluation.1 The teachers prepare an open profilc of themselves as teachers. It is in the form of asort of self-presentation which can precede the feedback session at the end of themicroteaching.2 Statements such as the following would make up a teacher’s own professionalprinciples:I always correct learner errors.I do not allow learners to use their first language.

PURPOSES FOR EVALUATION 2613I teach the rules of grammar to help lcarncrs use the language.I never ask a learner to use language which has not been previously presentedand practised.I always use authentic materials as a basis for teaching.I make certain that a large proportion of the learners’ time is spent in group work.I never ask learners to read aloud to the rest of the class.I always mark learners’ writtcn work.I believe that learner errors are the result of first language intcrferencc.I try to exercise a strong personality in the classroom.I adapt my teaching to suit what the learners say they want.For each of these statements, the teachers in discussion groups indicate whether theyagree or disagree. If thcrc is tlisagrccment then they are asked to rewrite the statementto reflect \vhat they think.Note that this self-cvaluation checklist [. . .] does not presupposc any external obscrwr.Nonctheless, in microteaching it can be used by both peers and tutors to discuss hvhatconstitutes elements of good teaching practice. Because peer evaluation is collaliorativc inapproach, thc teachers being observed might themselves suggest areas of their teaching thatthey feel need to be improved and ask their colleagues to concentrate on thest,.At this point cvaluation has moved away from the narrow summative functions ofevaluation for grading purposes and has taken on illuminativc and support functions andbccome formative in purpose.Collaborative group workThis is a fui-thcr cxtension of peer-evaluation where thc focus of the evaluation is agrecdon beforehand by the group. More control is in the hands of the peer group but it requiresgood lcatlership skills. Collaborative group work can offer an additional opportunity toevaluate the trainer and the programme.Kouraogo (1 987), in an article about Rurkina Faso, discusses the junction of teachers’self-help groups which can form the basis of a collaborative national teacher organisation.Kouraogo suggests that groups could meet on a monthly basis and discuss the practicalproblcms that tcachers have. At a later datc, these small groups antl their discussion topicscould be Ilrought together in a national conference. One of the purposes of these self-helpgroups is that they may not only help teachers resolve practical prohlcms, hut may alsocncouragc antl support teachers in difficult circumstances.SummaryWe havc moved from the narrow perspcctive of grading teacher performance to anevaluation of tcachers and teaching which can provide information of practical use totcachers for the development of their ttwhing. Evaluation through obscrvation is useful atall stages of a teacher’s ea]-eer to improve the quality of teaching for the benefit of theIearncrs. It may be a gradual process which is initially promptcd by an rxtcrnal olxcrvrrbut latcr moves tobvards self-evaluation. Since tcachers may find themselves in a situationwhere there is little or no in-service training, evaluation can bc the mans to understandingtheir own teaching bettcr, improving their performance, and adapting to the changing needsof the classroom. Evaluation in this broad scnse is an important part of teacher educationwhich teachers can use throughout their careers.

262 PAULINE REA-DICKINS AND I

Chapter 21David R. CarlessA CASE STUDY OF CURRICULUMIMPLEMENTATION IN HONG I

264 DAVID R. CARLESS2 Review of selected factors affecting the implementationof in novat ionsl‘hc litcrature on the management of change (c.g. Fullan, 1991 ; Markee, 1997) indicates anumhcr of different factors that may affect the implcmcntation or non-implementation ofcurriculum innovations. This , tion \z ill hrictly discuss just three factors that seemparticularly relevant to the caw study discussed in this paper, namely teacher attitudes,teacher training and teachers’ undcrstantling of the innovation.’l’rachcrs’ attitudes otniously affcct thcir xhaviour in thc classroom. Thcir attitudestend to be derived from their owm expcricn as learners, thcir training, their teachingexperience, their interaction \vith collcagucs and the values and norms of the society inkvhich they work. When teachers’ attitudcs arc congruent with the innovation, then theyare likely to be positivcly tlisposctl to\vards its implcmcntation. However, teachers who arcinitially enthusiastic about an innovation may easily become disillusioned if there is a lackof support for thc innovation, such as inadcquatc rcsourcing or negative sentiments fromthe principal or colleagues.If the innovation is incompatible \vith teachers’ existing attitudes, resistance to changeis likely to occur (Waugh antl Punch, 1987). Within EIT, for cxample, there arc a numbernt reviews of largely unsuccessful attempts to implcmcnt learner-centredcommunicative curricula amongst teachers \vhosc background and experiencc tendstowards more traditional tcachcr-centred methods. In some form this scenario has beendocumented in China (Hui, 1997; l’cnncr, 1995), Egypt (Holliday, 1994), Grcccc (Karavas-Iloukas, 1995), and Oman (IIarrison, 1996). [. . .]Teachcr training antl support are crucial issues in the prcparation of teachcrs toimplcment a ne\v curriculum [. . .]. Vcrspoor (1989), in a study of change in developingcountries, suggests four clements nccdcd for successful teacher training to supportinnovation:permanent and locally available in-service training, e.g. through a cascading model;establishment of effective tcms for supcrvision antl support of teachers;adjustment ofthc content oftcachcr training to the teachers’ o\vn Icvcl of knowledgeand experience; andencouragement of teacher motivation antl commitment, e.g. through improvedworking conditions or opportunities for professional dcvclopment.Training therefore needs to be ongoing antl tlcvclopmcntal rather than piecemeal (Brintlleyand Hood, 1990). Teachers need both on- and off-site training; the former to relate theinnovation to the realities of the spccific school context, the latter to permit the opportunityto reflect on the meaning of the innovation away from thc pressuresIf teachcrs arc to implcmcnt an innovation succcssfully, it is e.thorough undcrstantling of the principles and practice of the proposed change. It is desirablethat they understand hth the theorctical underpinnings and classroom applications of theinnovation, but it is the latter that tends to prove most essential, especially in contexts kvhcreteachers are not \ycll-traincd and/or lack sound subject knowledge. Fullan (1991, p. 199)\varns us of a cardinal fact of social change, that “p~oplc will al\vays misinterpret andmisunderstand some aspect of the purpose or practicc of something that is new to them.”For example, Karavas-Doukas (1 995), in an investigation of a communicative syllabus beingintroduced in Grcck secondary schools, found that teachers exhibited incompleteunderstanding of the innovation they were charged with implementing and that thesemisconceptions contributed to negative perceptions of’ thc innovation.

CURRICULUM IMPLEMENTATION IN HONG I

of266 DAVID R. CARLESSaged 6-7 years old. The case study approach seems particularly suitable to investigate acurriculum innovation becausc, as indicated earlier, relatively little is known about howinnovations are or arc not implemented in the classroom context. Case studies enableinformation to be collected from a number of sources and over a period of time. Theapproach enables the development of an understanding of thc phenomenon from theteacher’s view. This teacher perspective is crucial because teachers arc the key element inthe implementation process, in that they arc the individuals who will implement faithfully,reinvent or reject an innovation.The central focus of the study was to explore the nature of curriculum innovationthrough analysing thc pro 1 TOC implementation in the classroom. The researchquestions that guided the study focused mainly on the following issues:the teachers’ attitudes towards English teaching antl towards TOC;the teachers’ familiarity nithTOC principles, the extent to which they believed thatthey were carrying them out, whether they were actually implementing TOCprinciples and the strategies that they Lvere using; andthe extent and nature of change antl development in the teachers during the periodofthe study.Data collection methods used for the study comprised classroom observation, focusedinterviews and an attitude scale. Classroom observations were conducted for 5 -6 consecutiveEnglish lessons for each teacher in three separate cycles during the school year,totalling 15-1 8 audio-taped observations per teacher. I took the role of a participantobserver and was willing to take part in lessons; for example, I tried to encourage, assist ormonitor pupils during individual, pair or group activities. Both quantitative data in termsof a tailor-made classroom observation schedule and qualitative data in terms of lessontranscriptions and field notes were collected. This “compatibilist” stance (Lynch, 1996) ormixetl-method approach aimcd to facilitate triangulation through the use of both numericaland non-numerical data.A 26-item attitude scale was developed to measure the orientation of respondents toELT andTOC.Thirteen ofthe items (numhers 3,5,7,9, 10, 11, 12, 13, 14, 17, 18,20 and24) implied a broadly positive orientation towards TOC and related principles, the otheritems indicated a broadly negative orientation. The attitude scale was administered to thecase study teachers prior to the classroom observation period and again 6 months later atits conclusion. It was also administered to a wider sample of primary English teachers.A series of five scmi-structured interviews were conducted with each of the threeteachers. A baseline interview, prior to the commencement of classroom observation,collected relevant background information about the teacher and the school. Postobservationinterviews, carried out at the end of each cycle of observations, focusedprimarily on the lessons that had just been observed. Summative interviews were conductedin order to probe into some ofthe main issues, arising from the classroom observations andthe ongoing data analysis. All interviews were transcribed verbatim by the researcher.For reasons of space, this paper will focus principally on qualitative data from theclassroom observations and interviews.5 Background to the teacher and the schoolThe teacher involved in this case study, rrfcrrcd to as Carol Lee (a pseudonym), had 4 years’teaching experience at the commencement of this study. She has completed a teachingccrtificate as an English major from the Hong Kong Institute of Education, the main provider

CURRICULUM IMPLEMENTATION IN HONG I

268 DAVID R. CARLESSTable 21. IExcerpt from a Irsson transcript1510152025303540(She takes our an altractii e cnlurcycd plioro o/ ber,faini!r~)1': Look at me (points to herself). I am Miss I cc. M! name is Miss 1.c~. I amMiss Lee. My name is Miss Ire. Ant1 this (points to the phoro) is my family,And this is my family. Miss 1.(points)? Who is thi.;? Who is this?L1: This is Miss Lev.L2: This is Miss Lrc.L3: This is Miss Lee.'1': l'his is me (points). I am Miss I ce. This is my fhmily. This is my fathtxr(point\). 'l'his is my mother (point.\). ,4ntl I ha\c t\vo sisters. They arc mysisters (points). 'fhis is my sistcr (points). ficr name is Cclinc. I Icr name isCclinc. 'I'his is mv sister (points). Her namc is Stella. Her name is Stella.This is my father (points). 'l'his is my father. He is a man. fie. IIc is myfather. His name is h i . His name is h i . I IIS. I lis narnc is h i .(Then she recups and remind\ pupi1.s ofthc nume.s und then usks them questionsahorit the names antl relationships of her,fumi!i member,. She then usks them to takeout theirfami(r. photos; man) of them make comments in Cantonese, [~rcsiimub(~C.\CIMS/upolo~ps; she tells pupils iurh no photos to take oiit their hunch&hc71.e photos in them. She takes one pupil photo as cin crumple.)'1.: Look. Who's this (points)? Who's this? This is . . .L4: Irene.L5: 'I'his is Ircnc..I.: Good. This is Ircnc. I Icr narnc is Irene. Okay, this is Irene.LLL: This is Irene.T: Her name is Irenc.1.1.1.: I Icr name is Ircnc.T: And this one (points), \\bo's thi.;?(Further ifemon.strution and pructicc)family. This is m y family. Who is thisII hichT: Stand up. Stand up. No\v I \\ant you to gct \\ ith [sic] vour photo or yourhantll)ook (USES pmre). You can walk aroun(1 and then look at the othcrs,'What's this?' antl then you can ansmrr, you can answer okay no\v try, getyour hook antl get your photos readv. (Some get otit o/ their seuts.) Yes, goaround (cises gcstwe). . . . . . . . . .(f'upils leure their scot) mil mow toiwrds the Iron1 o/.thc clci

CURRICULUM IMPLEMENTATION IN HONG KONG 269the target language serves as the medium as well as the focus of instruction;the input is rich in directives;thcre is an adhcrcncc to the “here and now” principle;studcnts seem to be converting input into intake: andin the activity stage, learners have some intlcpendent control over the propositionalcontent: they havc some choice over what is said and there is some information gapbet\vecn speaker and listener.Notekvorthy is the quantity of comprehensible input to which the pupils are kingexposed and the use oftechniques such as repetition (e.g. lines 2 S), short simple scntences(e.g. lines 10 14) antl visual support (e.g. use of the photo) to facilitate pupil understanding.Interview data indicates that thc teacher is aware of Krashen’s (1 987) distinction betweenacquisition and learning and that she believes that acquisition is the most favourablc routcfor pupil language learning. “For primary school students, I think acquisition is importantfor them. I hclieve that it’s much more easy for them to acquire a language rather than learna language” (baseline interview, p. 9). In other words, she has a clear rationale for using thetarget language so as to lacilitatc language acquisition among the pupils.It is suggested that the task-liased approach of 1’OC puts a greater onus on teachers’language proficiency than more restricted form-focused tcxthook exercises. Carol’s lluentantl confident use of English seems to play a greater rolc in the choice of language mediumthan the pupils’ own limited knowledge of the language. In other words, she is able tomaintain English medium during the lessons mainly hccausc of her own high overallproficiency and her ability to use clear, simply English supported by pointing or gestures.This contrasts with a view, commonly expressed by Hong Kong primary tcachers, that thcynccd to use Cantoncse or mixed codr because ofthc low le\ el of language skills ofthe pupils.Carol explains thc I)encfits of using the target language as follo\vs:If thcy can try to listen to English more, it is easicr for them to learn a language. Ithink it’s strange if you learn, for example, Frcnch in a Chinese \ray \vith Chinese asa teaching medium as that’s why pupils like to go overseas to learn a language. I thinkit’s a kind of acquisition and I havc to give them an environment that English is thefirst language instead 01’ Chinese.(summativc inter\ie\v, p. 1)6.2 Interview dataHaling looked briefl> at an example of how Carol tarrieq outTOC in the classroom, I 1% 111nou proceed to discus, a number of themes from the inter\ IC\\ data.Thc extracts discussedhere relate to her attitudc touartls TOC, her understanding of TOC, the rolc ot theprincipal, change implementation, teacher \upport and teacher reflection6.2.1 .Ittirude tortar& TOC‘Her actions in thr classroom, her statcments in interviews and her attitudc scale responsesall indicate that Carol has a positive attitude to\rards TOC and associated principles. Forexample, in thc interviews shc describes hcr attitudes towartlsT0C as “positive” or “morethan positive hut I can’t say very positive.’’Initial analysis of the attitudc scale responses shows that she has a more positiveorientation to principles congruent ivithTOC than a lvidcr sample ol‘primary school English

270 DAVID R. CARLESSteachers. The following are the statements that she either strongly agreed with or stronglydisagreed with in both parallel administrations of the attitude scale used for the study.Theseprovide a sample of her attitudes.She strongly agreed with the following statements on both administrations of the scale:Item No.3791011131424StatementMaking errors is a natural part of the learning processThc main rolc of the tcachcr is to facilitate learning amongst pupilsPupils learn most when they arc actively involvedIt is important to give pupils the opportunity to learn at their own pacePupils learn through constructing their own grammar rulesIt is important for pupils to create their own sentencesIt is important for pupils to use a communicativc approach to teachingThe tcachcr would take into account pupils’ needs and interestsShe strongly disagreed with Item 22 of the attitudc scale, on both administrations.ltem No. Statement22 Under TOC pupils \vi11 be less motivated than beforeOverall, her expressed attitudes scem to be congruent with the constructivist view of‘learning cspoused in thc TOC framework (Clark et al., 1994, p. 1 5) and those linked tocommunicativc and/or task-based approaches to EIll: Interview data indicates that herattitudcs scem to derive mainly from her English language learning experiences as a schoolstudent, her prc-service training and her experience of“1anguage immersion” when studyingin the UK as an adult.6.2.2 Undersranding ? I TOCThe first four interviews all asked rcspondcnts to summarise their understanding of themain principle of TOC. Carol put different emphases on different aspects of TOC at varioustimes, but in general demonstrated a rcasonalile, though not full, understanding ofTOC,despite confessing to some confusion about the differences betwwn TOC tasks and associatcdterms, such as activities, exercises or worksheets. The following sample answer isquoted to illustratc clcmcnts of her conception of TOC:I think Lve should try to motivate thcm, try to increase their interest in Icarning, notjust copying. I think put the knowdcdgc in use is quite important in TOC. I think inTOC it should bc more livcly, not just a classroom situation, not just learn this butknow that it is useful and they can use it and they know that it is useful for the wholelifc, I think that isTOC.(post~obscrvation interview three, p. 7)Although she has not uscdTOC terminology directly, she has touched on a number ofTOC elements, for example, active involvement of pupils (first two sentences), task(“knowledge in use”), real-life context (“ not just a classroom situation”). Understanding ofthe principles and practice ofa curriculum innovation tend to cvolve over time and it is tobe expected that Carol will dcvclop hcr intcrprctation ofTOC further as she continues togain experience with it.

CURRICULUM IMPLEMENTATION IN HONG KONG 2716.2.3 The role $the principalInstructional leadership, staff’ development, the building of collaborative cultures, academic,administrative and resource support are some of the main means by which principals canfacilitate change. In Carol’s case the principal is supportive and willing to permit her a highdegree of autonomy. It is not clear the cxtcnt to which this is an informed managementstrategy or is indicative of a reluctance to be directly involved with‘I‘OC, a “wait and see”attitude prevalent amongst principals according to Morris et al. (1 996). Carol acknodcdgcsthe supportiveness antl flexibility of her principal:Even he thinks we shouldn’t do that [implemcntTOC] hut once we started, he givesa full support to me and if I want to take some courses he always mentions that Ishouldn’t worry about missing lessons. He thinks that it is good for me to take somecourses and he always asked me to encourage my colleagucs to go out antl take somecourses. He doesn’t control what I did, I can do it hohvever I like.(baseline intenkw, p. 5)This laissez-faire style is in contrast with more authoritarian leadership stylcs commonlyperceived to bc! found amongst many Hong Kong principals. In Carol’s case, it seems to beeffective as shr has the confidencc and ability to benefit from the autonomy granted by herprincipal.6.2.4 TOC and changeChange is often best effected gradually and as indicated by Clark et al. (1994) in theTOCcurriculum framework there should he flexibility over time-scales, with the developmentand implementation of TOC being aligned with the readiness of teachcrs and schools. AsCarol comments:Maybe it’s too rushed for the school to run thcTOC class, hve have to adapt it andchange bit by bit. First of all, wc have changed the time for each lesson, change thcformat of teaching, before kvc just adapt the ivholcTOC, theTOC matter lvcausc wchave to change the assessment task, the format of assessment, the format of rcportcard, too man! things at a go, so I don’t think it is a good way to change thecurriculum.(baseline interview, 1). 16)On the othcr hand, thc implementation of ‘I‘OC seems also to have brought somebenefit to Carol. As indicated by Morris clt al. (1 996) in their report on a major TOCrcscarch project, innovation can be used by principals 01- teachers as a vehicle fbr counteringinertia and lc-gitimising attempts to improve. The introduction of TOC provides teachcrswith a rationale for more activc antl innovative teaching approaches. Carol expresses it inthe following way:Because it is TOC I can do a lot of activities and prepare a lot of things. I have ancxcusc, because it is TOC class so I can make it different from the other class. Ifcvcrybody is doing a traditional class, maybe if1 do it in a tliffercnt way then the othersmay say, “why do you have to do so many things?Wc don’t do it so ifwe compare withyou, it seeins that we are lazier than you,”so because it’sTOC class, it’s a kind of excuseor reason why I change my way of‘ teaching.(summativc interview, p. 7)

272 DAVID R. CARLESSIn other \vortls, thcTOC initiativc pi-ovitlcs a theoretical and administrative backing forCarol to carry out the kind of learner-centred activitics that shc \vould likc to carry outany\vay. This enables her to teach in her preferred \vay, yet with less risk of facing negativepeer pressure from her more tratlitional~mintlcd colleagues.6.2.5 Teacher-researcher collaborationAs mcntioncd carlicr, school-bascd support is an essential component of inservice teachereducation provision for innovations (Vcrspoor, 1989).‘l‘his support can take different forms,for example, the collaboration bct\z n a tcachcr and a tcacher educator researcher asdescribed in this paper \vas mutually enriching. From my anglc, 1 have dcvcloped newunderstandings of the primary classroom antl of TOC, and in my o\vn teacher educationclasses, I now frequently use examplcs from Carol’s class to exemplify points that I ammaking. From the teacher’s point of vicw, Carol comments as follo\vs:You make my class a real English class, you make the classroom really English.Youmake me get used to having somcliotly watching my lesson, so now I don’t care ifanybody comes into my classroom antl \vatchcs ho\v I teach, and I have confidcncc inmy teaching and YOU have given me a lot of atlvicc in thc whole year, thank you verymuch. I think I’ve improvctl in some parts.(summativc interview, p. 10)It is also suggested that thc process of king intci-vic\vcd plays a role in clarifying ateacher’s understanding of thc innovation by prompting thought and discussion aboutrelevant issucs.6.1.6 Tcacher reflection and der cloprnentAt various points in the intcrvic\vs, Carol shows hcr open-mindetlness and interest in findingout more allout teaching as a means for prolkssional improvement. She comments on thevalue of peer observation in thc following extract:I think going to another classroom to \vatch how the others teach is important. I thinkit’s good because now I am doing the assignment [M.Etl. assignment] and I hale to gointo the classroom to \vatch the students. Even though I hvatch the studcnts, I canwatch how the teachers teach, I think I really learn a lot of things, many many things.I think my teaching skill is quite good already but I tind I can lcarn some more evenne\vcr things. So, I believe that if teachers likc to watch each other, I think the otherscan give you some comments so you can improve antl also improve by watching howthe others teach . . . but I think it is difficult Iwcause a lot of teachers don’t likc otherpcoplc to come in their classroom antl \vatch how they teach.(summativc interview, p. 5)Other responses also show an interest antl atiility in identifying and beginning toreflect on relevant teaching issues. Reflection and the ongoing consideration of alternativeteaching strategies is one of a numlier of factors identified hy Hopkins and Stern (1 996) asbeing characteristic of effective teachers. ’l’hc following extracts show evidence of Caroldeveloping a rcflcctivc orientation to her tcaching:

CURRICULUM IMPLEMENTATION IN HONG I cnjq doing thow actillties hut how much dit1 the!really learn? I just wonder(pmt obscr~ ation inter\ IC\\ 2, p. 1 )7 ConclusionThis paper tried to show how a well-qualified English teacher has responded to a curriculuminnovation. Reference has been made to her attitude towards the innovation, herunderstanding of thc innovation, her classroom teaching, her professional development andinterview comments on a number of issucs relevant to thc change process. The anal!of a case study, so extrapolating the findings is not possible but it is suggested that thediscussion has raised a numbcr of issues that may have \vidcr implications.It has been indicated that despite the challenges associated with successful curriculuminnovation, this teacher’s initial experiences withTOC have been largely positivc. A numberof her charactcristics have assisted her:hrr academic and professional training;her high standard of English proficiency;her positive attitudes ton-ards teaching and toivards the innovation; andher desire for further self-improvement and professional development.This discussion is not incant to imply that curriculum innovation can only lie fosteredbv teachers who have the abovc characteristics, but it is fair to sa\’ that such teachers arcprobably in a favourable position. Therefore, gcncral governmental initiatives that upgradrthe professionalism of teachers, in addition to being desirable in their o\vn right, do helpto pro\-ide a climate conducive to the development of curriculum reform. Such initiativesarc part of a long-term enhancement of primary education in Hong Kong (EducationCommission, 1992) of whichTOC is one componcnt.This reinforces Stenhousck \enerablcdictum that thcrc is no curriculum development without teacher tlevclopment.In addition to these wider initiatives, support for teachers at the classroom level playsa significant role in facilitating the implementation of innovations. In this case, thesupportivencss of the principal and fruitful collaboration bet\vccn the teacher and anexternal teacher ctlucator/rescarcher seemed to encourage a capable tcachcr in carryingout the innovation. In other cases, proactive involvcmcnt from principals or scniorcolleagucs and/or ad\ )ry visits from inspectors, tcacher trainers or experienced teachcrsmay be needed to facilitate implementation, Support and encouragement, in one form oranother, are an essential prerequisite for successful classroom implementation of acurriculum innovation.ReferencesBrindlc); G., Hood, S., 1990. “Curriculum innoIation in adult ESL.” In G. Brindle) (Ftl.). TheSecond Language Curriculum in Action. NCEUI’K, Sylncy.

274 DAVID R. CARLESSClark, J., Scarino, A,, Brownell, J., 1994. Improving the Qialiy $Leorning:A Framework for Target-Oriented Curriculum Rener4d in [long Kong. Institutc of Languagc in Education, HongKong.Education Commission, 1992. Report No 5: The teaching profession. Government Printer,Hong Kong.Education Department, 1996. Teocher Suri,e),. Education Dcpartmcnt, I-Iong Kong.Ellis, R., 1988. Classroom Second Longucige Derelopment. Prentice-Hall, London.Fullan, M., 1991 . The h’erz. ,Weaning of Etlucationul Change. Teachers College Press, NcivYork.Fullan, M., 1993. Change Forces: Probing the Depths of‘Educational Refbrm. Falmcr Press, London.Harrison, I., 1996. “Look who’s talking noLv: listcning to voices in curriculum rene\val.”In: K.Railcy, D. Nunan, (Eds.). Ibice.s,from the Language Classroom. Cambridge University Press,Cambridge, pp. 283 303.Holliday, A,, 1994. Appropriate .Clethodologj, antl Social Context. Carnhritlge University Press,Cambridge.Hopkins, D., Stern, D., 1996. “Quality tcachcrs, quality schools: international pcrspectivcsanti policy implications.” Teaching and Teacher Education 12(5), 501 -5 17.Hui, L., 1997. “New bottles, old \vine: communicative language teaching in China.” EnglishTeaching Forum 35(4), 38 41.Karal as-Doukas, E., 1995. “Teacher identified factors affccting the implementation of acurriculum innovation in Greek puldic secondary schools.” Language, Culture andCurriculum 8(1), 53 68.Krashen, S., 1987. Principles and Practice in Secontl Language .kyui.sition. Prcnticc-Hall,Englcnood Cliffs, NJ.Lynch, B., 1996. Language Program Eraluation: Theor). antl Practice. Cambridge University Press,Cambridge.Markcc, N., 1993. “Thc diffusion of innovation in language teaching.” :lnntial Review ofAppliedIinguistics 13, 229-243. See also chapter IO of this volumc.Markee, N., 1997. Managing Curricular Innoration. Cambridge University Press, Cambridge.Morris, P., 1992. Curriculum Development in [long Kong. Education Papers 7. Faculty ofEducation, Hong Kong University, Hong Kong.Morris, P., 1995. The Hong Kong School Curriculum. Hong Kong Univcrsity Press, Hong Kong.Morris, P. and 1 2 associates, 1996. ‘liirget-Oriented Curriculum Evaluation Project: Interim Report.Faculty of Education, University of I long Kong, Hong Kong.I’enncr, J., 1995. “Change and conflict: introduction of the communicative approach in China.”TESL Canatlajournal 12(2), 1 17.Stoller, F., 1994. “Thc diffusion of innovations in intensive ESL programs.” Applied LinguisticsIS( 3), 300-327.Verspoor, A,, 1989. Pathways to Change: Improi,ing the Qua/iy ?f’Etlucation in Developing Countries.World Bank, Washington DC.Waugh, R., Punch, K., 1987. “Teacher receptivity to systemwide changc in theimplementation stage.” Revieri ofEJucutionul Research 57( 3), 2 37- 254.Wong, Y. F., 1996. ‘‘To investigate the understanding of principals antl teachers of the kc?features of the Target-Oricntctl Curriculum (TOC) antl their perceptions of its impacton their teaching.” Unpublishctl master’s thesis, University of Hong Kong.

~ sourcesChapter 22Joan LesikinDETERMINING SOCIAL PROMINENCE:A METHODOLOGY FOR UNCOVERINGGENDER BIAS IN ESL TEXTBOOKSIntroductionHE MOST SIGNIFICANT INFLUENCE on an individual’s role is“the socialT prescriptions antl behavior of others,” according to Thomas and Biddle (1979, p. 4).The female and male characters in ESL textbooks have the potential to serve as those othersof social prescriptions antl behaviors for ESL students. According to some studies(e.g., Cole, Hill, and Dayley, 1983; Rem antl Rem, 1973), same-gender role models providestronger role identification for some people than models of a different gender. Thus, forexample, female characters in texts are stronger role models for some lvomen than theyarc for some men. Any one dcpiction of a female or male, however, may lx- inconsrquential;it is a particular bias sustained over time and through repetition which has a cumulativecffcct.Because of their ubiquitous presence in our schooling, cducational texts may have justthat effect. Content analysis of textbooks in a variety of educational subjects (e.g., socialstudics, Slectcr and Grant, 199 1 ; science, Powcll and Garcia, 1988; teacher education,Satlker, 198 1 ; children’s rradrrs, Weitzman and Rizzo, 1974; ESL, Porreca, 1984) has shownthem to contain gender bias. And \vc know that teachers in a variety of cducational subjectsusc tcxthooks about s nty pcrccnt ofthc time (Komoski, 1985); ESL tcachers arc probablyno exception.The potential influence ofgendered role models may haw particular weight in ESL inhighcr education. For many ESL collcgc students, ESL textbooks may be their first encounterwith the American educational system. ESL textbooks especially may have a morc profoundcffcct than texthooks used suhscqucntly in a collcgc student’s academic carerr.ESL text materials attempt ideological neutrality in order to appeal to a broad and oftencensorious educational market. Real and imagined pressures have often led writers antlpublishers to substitute or eliminate topics deemed objectionable to various constitucncics(lanner, 1988) .lcxt materials published for second language lcarning in particular prcscntlanguage in ways that reinforce the sense of ideological neutrality. These texts arc typically

276 JOAN LESII

anDETERMINING SOCIAL PROMINENCE 277Analyzing gendcr bias as social prominence in textsThe methodology is based on M. A. K. Halliday’s examination of the functions of language,specifically his concept of participant roles of nouns and pronouns and the division of themeand rhcmc from the Prague School of Linguistics and reconstituted by Hallitlay (1 985).I use these concepts to dctcrminc the relative social promincncc of females and males inwritten texts.Theme, rheme, and last stressed elementAccording to Quirk and Grccnbaum (1 973), “the thrmr is the most important part of’ aclause from the point of view of its presentation of a message in sequence” (p. 41 2).Themcis the psychological suliject. It is the first clement in a clause (lvith the exception of initialadverbs) ‘‘VI hich serves as the point of departure of the message; it is that with which theclause is concerned” (Halliday, 1985, p. 38). Rhcme is the remaining part of the messagc,which develops the themc.‘lhc last strcsscd clement of a clause is also important.Thc personor persons occurring as the last stressed element bear the information focus of the clauseand, like the, person or persons in theme position, have communicative prominence (Quirkand Creenhaum, p. 41 2).From communicative prominence to social prominenceA charactcr in the position of theme or as last stressed elcmcnt in a clause is the personin a position of communicative prominence. It is the person who is the center of conversation,the topic in writing, or the information focus until anothcr person is introducedor focused on or the communication or text ends. People who are centers of conversations,topics of writing, or the information focus ~voultl seem to have more social prominencethan pcoplc who are not. We communicate about people who in somc way intcrcst us orarc important to us; those Lvho are not of interest are not thc focus. Thus pcoplc can beperceived as having different degrees of social prominence hy the dcgrcc of intcrcst shownthem.Similarly, in a Ivrittcn text such as a textbook, if characters of one gcnclcr (rcprcscntcdly gcndcr-spccific nouns or pronouns) occur more frequently in the position ofcommunicative promincncc as theme or last strcsscd clement ~ underlying messageto rcatlcrs is that one gentler has higher social prominrncc than the othcr. Thc); most likclyprcsent a stronger, more alluring role model than those characters in rheme position.Participant junctionsThe division of gender-specific nouns and pronouns in clauses into theme, rhcmc, and laststressed element does not specify what those role models potentially are. A noun or pronounin a clause also has a participant role in terms of the ideational function of the clause, 1% hichhelps us to get at the ideological message. Since themes and last strcssed elements arc themost prominent and could be more intlucntial rolcs for students than those rolcs cmbcdtictlin rhrme, I examine their participant functions.According to Hallitlay, the ideational function of a clause is how it represents experiencein terms of meaning. The role of a noun or pronoun in a clause is how it participates in thcprocess of a particular experience expressed by that clause. Thus the noun or pronoun alsohas a participant role in terms of this ideational function.

278 JOAN LESIKINThe participant roles can be di\ided into two groups (seeTable 22.1). In Group I, thefunctions include doing, acting, sensing, saying, attributing, and rclating. Those in Group I1are the complcmcnts of the participants in Group I. Group I participants take a more activerole experientially than those in Group 11, whcre the role of actor is defined a3 the mostactive and direct participant in an experience.Table 2.2. IParticipant roles of nouns anti pronounsGroup 1Group IIActorBeha\ crSayerScnwrTokenRccipicnt, Client, GoalBrnchciai-?Rccci\cr, TargctI’hrnomrnonValucProcedureThe procedural steps to apply this mcthodology arc:Collect all clauses in the “unmarked” form (thosc that are not questions and/ornegations), containing at least one gentler-specific noun or pronoun (e.g., Anne is hereor Thy waitedfor Rob to come). ‘Categorize these clauses by gcnder antl by thcme/rhemc distinction. Simultaneouslyexamine each clause to see if it contains a last strcssed gender-specific noun or pronounin the same clause signaling a competitive focus of new information. If it does, notethe gender-specific focus. Eliminate themes or rhemes which have both female andmale nouns or pronouns (c.g., Anne and John arc here or Either June or Bob I$) since cooccurringforms offset each other.Tabulate the number of themes, rhcmes, and last strcssed elements according togender, and compare the numlicr of themes and last stressed elements to the numberof rhemes, by gender.To determine the roles of thosr gender-specific nouns antl pronouns labeled as themeand last stressed clement, re-examine them in terms of participant functions.Tabulate the themes and last stressed elements by participant function and gender.Lastly, incorporatc into the results the quantity of themes and last stresscd clt‘mentscompared to quantity of rhcmcs.FindingsTheme and rhemeI applied this methodology to a singlc chapter in each of scvcral tcxtbooks devrloped forthe ESL academic markrt. I will discuss thr findings from one of the textbooks, Grammar inUse (Murphy, 1989), to demonstrate thc application of the methodology.Out of 55 gender-spccific nouns and pronouns in clauses, 44 are theme, as in Ann in/inn telephoned someone (p. 94). Of these, thc ratio of females (n = 15) to males (n = 29) is5 2: 100 (34”h to 66%). Elcven gentler-spccific nouns and pronouns in clauses are rhcmcs,as in him inyou want him to get some stamps (11. 101). Of these, the ratio of females (n 3) to

DETERMINING SOCIAL PROMINENCE 279males (n = 8) is 38: 100 (27% to 73%). In addition, seven gender-specific nouns as rhemesare the last stressed elements in seven of the clauses, as in Torn in I’ve just seen Torn (p. 98).Of these, the ratio of females (n = 2) to males (n = 5) is 40: 100 (29% to 7 1 Yo). As themeand rheme (including the last stressed element) of a clause, female nouns and pronouns arepresent on average 33% compared to 67Yo for male nouns and pronouns (n = 37) or thcratio of 49: 100 (seeTable 22.2).Table 22.2 Grammar in Use: frequency of gender-specific nouns and pronouns as thcmc and rheme in“unmarked” clausesFemale Malc Totul RatioN YO N YO N I; to Ml’hcmc 15 34 29 66 44 52:100I‘\hcmc 3 27 8 73 II 18:100Total 18 33 37 67 55 49: 100Note: Thcmc = psychological subject o fa clause; rhcmc = noun or pronoun devrloping the subjcctincluding last stressed clcmcnt I)caring information focus.As theme and as the last strcsscd clcmcnt (n = 5 1) ~ the prominent forms in tcrms ofmeaning ~ 17 are females and 34 arc males (33% to 67%) or a ratio of SO: 100. (SecTablc22.3 .) Thus males dominate thc positions of communicative prominence in clauses in thischapter by double the numbcr of females. The total number of female to malr nouns andpronouns in the chaptcr is morc equitable: 44% to 56%.Table 22.3 Grammur 7n I1.w frequency of gender-specific nouns and pronouns as thcmc and laststrcsscd clcmrnt in “unmarked“ clausesFemale Malc Total RutioN 96 N Yo N 1 toMThcmc 15 34 29 66 44 52:lOOElcmcnt 2 29 5 71 7 40: I00I‘otal 17 33 34 67 51 50: 100Note: Themeinformation focus.psychological subject ofa clause; clcmcnt = last strcsscd clcmcnt in a clausc~ bearingParticipant rolcsBy examining those gender-specific nouns and pronouns in theme position in the sameclauses to determinc their participant roles, I found that they (n = 44) function in fivcparticipant roles (see Table 22.4). That is, they are actors, sensers, tokens, sayers, andbchavcrs. The 15 females occupy four of the rolcs while the males occupy fivc. Malesoutnumber females in all rolrs except that of senscr, where fcmalcs (n = 4) arc prcscnttwicc as often as males (n = 2). Howcvcr, males (n 1 1) arc actors, the strongest participantrole, more than three times as often as females (n = 3) or the ratio of‘ 27: 100. In addition,there are more than twicc the numbcr of mala (n = 10) than females (n = 4) as sayers or

280 JOAN LESIKINthe ratio of 40: 100. The roles of actor antl sayer hac the greatest numher of nouns andpronouns. Femalc5 (n = 4) antl malc5 (n = 5) arc most clcnly matched in the role oftokcn,the second largest role, in thc ratio of 80: 100.Tuhle 22.3 Grarnmur 7n Use: frcquc.nc.! of‘partic~il~aiit I-olea ofgcntlci--specitic iiouiis antl pronouns in“unmarkctl” clauscsRoles N (Yo N “0 N F to M8ctor 3 21 11 79 14 27: 100Scnscr 4 67 2 33 6 100: 50’l’okcn 4 44 5 56 9 80: 100Saycr 4 29 10 71 14 40: 100Reha\ cr 0 00 1 100 1 0:IOOTotal 17 3 3 34 67 51 50:100.\ole: Actor = a doer; acnscr = a pvrwn li,cling, thinking 01- sccing; token = a pcrson having anattribute or relation to another; saver = a vcrlialircr; Iwhavcr = a person cxhibiting physiological orpsychological Iwhavior.Summary of findings and implicationsMales dominate the positions of communicativc prominence (1 00: 50) in G‘rummar in Use.At the same time, males arc actors, the strongest participant role, more than three times asoften as females and dominate four of thc five roles in thcmc position in clauses, with anoverall prcscncc more than t\vice that of fcmalcs. Males arc primarily presented as actorsand sayers, suggesting males as thc docrs antl vcrlializcrs. Fcmalrs outnumlxr males asscnsers, suggesting females as feeling, thinking, antl sccing indivitluals. Both arc fairly evenlydivided as tokens expressing having an attribute or relation to another. While thc comparablel‘rcqucncy of females and males varies antl males occupy thc two strongest roles, all the rolespresented for lmth females and males arc activc cxpcricntially.The grcatcr quantity of males in the position of communicative prominence antl in thcparticipant roles in clauses in the chapter suggests that the males presented havc greatersocial prominence antl thcrcforc more pocver than the femalcs.Thc greater number of malenouns and pronouns in clauscs in general rcinforccs these finclings of dominancc.Thc findingsalso suggest that fairly traditional, stereotypical roles arc prescntcd for each gcndcr (themales as actors and sayers; the fcmalcs as scnscrs).Grammar in Use is onc ofthe largest sclling textbooks to thc academic ESL markct, basedon the opinions of ten marketing experts in pul)lishing and distributing ESL textbooks(hikin, 1995).Thc tcxtbook may influence how our students view their ohvn social powerrelative to that of others as thcy sort out a new gentlered identity in the acculturationprocess. Language learners arc “constantly organizing and reorganizing a smsc of who theyarc and how they rclatc to the social world.‘l’hcy are, in othcr \vords, engaged in identityconstruction antl ncgotiation” (Norton, 1997, p. 41 0). As part ofthe acculturation process,ESL students’ new identities arc shaped in part by what thcy read in our classes. When thetexts they read reflect biased assumptions about gentler, thc texts may transmit these hiascs,often reinforcing the lower prcstige antl p vcr ascrilxtl to fcmales.

DETERMINING SOCIAL PROMINENCE 281To my knowledge, no research has been done on the effects of gender bias in ESLtextlmoks. Studies on gentler and language have suggested that gentler bias and sexualstcrcotvping in written tcxts antl pictures and sexist behavior in classrooms have delrtcriouseffects for American femalcs. These effects include feelings of exclusion, devaluation,alienation, and lowered self-expectations. (See, for example, McArthur antl Eisen, 1976;Montemavor, 1975; MacKay, 1979; andTodd-Mancillas, 198 1 .) Macaulay and Brice (1 997)report on several empirical studies in education suggesting that “thr stcrcotTping ofmathematics as a male domain negatively affects females students’ attitudes toward,performance in, and perceived proficiency in the subject” antl that graduate students (femalesmost especially) who pcrccived gender-biased behavior in their classes wcrc negativelyaffected; in some cases they hvithtlrew from the discipline or graduate program(pp. 820 821).’l’hc results of these studies suggest that our female ESL students, like American females,may also construct less ponwful and prestigious identities than their male counterpartsfrom similar sources. The undervaluing of women potentially adds to the female languagelearner’s sense of alienation and worthlessness, making adjustments more problematic antlperhaps slowcr than for her male counterpart.ConclusionIn conclusion, I would like to offer some recommendations for countering the gender biasthat may I x present in ESL textbooks and elaborate on the reasons for classroom teachersn focus on ~ gentler issues with students. Teachers might begin by citingccs of gentler bias in the ESL class’s textbook.By discussing what ~ ve as tcachcrs perceive as gender hias, ~ vc may find that studentshave different pcrccptions antl vieLvs from our own. Discussion might lead to reflectivehvriting or to studcnts interviewing Americans or fricnds and family members on specifictopics raised about gendcrcd roles, behaviors, stercotypcs, or expectations. Students mayalso creatr alternative tcxts as language learning activities. For example, students might takea published tcxt containing generic masculine forms (c.g., he, mankind) and rewrite the textusing more inclusive languagc. A subsequent activity could have students applying this newkno\vlctlge to their o writing. Wc can also counter the gentler bias by supplcmc~~ting thematerial with more nly represented tcxt in instances where wc arc comfortable \vithother aspects of thc textbook or cannot change the textbook for programmatic reasons.Finally, we can write to publishers or speak to their representatives at professionalconferences to makc them a\vare of our disco\eries ofgendcr bias in their materials antl ourdistaste for these biases.In bringing gcndcr bias in ESL textbooks to our students’ attention, \IC immediatel)raise gcndcrcd behavior antl roles as issues in their ow-n acculturation process. Rehavior androle assumptions and cxpcctations arc changing, not only in the US mainstream culture butin the students’ cultures as \rcll. Students may be avwe of these forces of changc in theirown cultures but may not be a\varc of them in thc US. As their teachers, \ve help our studentsexplore, sort out, and construct their new roles and idcntitics in the new culture by makingthe unconscious conscious.Knowledge of gender bias in educational tcxts and in other aspects of schooling, suchas pcrccivcd academic strengths and differential conduct antl expectations of teachers inrelation to female and male students, ma): help stutlcnts reflect on their prior schoolexperiences, consider their present behaviors with teachers and students, and makekno\vlctlgeablc decisions about their future educational goals. In constructing ne\v identities,

282 JOAN LESII

DETERMINING SOCIAL PROMINENCE 283McArthur, L., and Eiscn, S. (1 976). “Achievements of male and female storybook characters asdetriments of achievement behavior by boys and girls.” journal of’ Personulicv and SocialP.ychology, 33,4677473.Macaulay, M., and Brice, C. (1997). “Don’t touch my projectile: Gentler bias and stereotypingin syntactic examples.” Langmge journal qfthe Linguistic Societl, Vfilmericu, 73, 798 825.MacKay, D. G. (1979). “Language, thought, and social attitudes.” In H. Giles, W. 1’. Kobinson,and P. M. Smith (Etls.), Language: Social psychological perspective.s (pp. 89-96). New York:Pcrgamon Press, 1980.Markstcin, L., and Hirasabva, L. (1981). Developing reading skills: Intermediate. New York:Newhury House.Mills, S. (1995). Feminist stl,’listt’cs. London: Routledge.Montcmayor, R. (1975). “Sexism in children’s books and elementary teaching materials.” InA. P. Nilsen, H. Bosmajian, H. L. Gcrshuny, and J. P. Stanley (Eds.), Sexism and langnuge(pp. 161 -179). Urbana, IL: National Council ofTcachcrs of English.Murphy, R. (1 989). Grammar in use: Rgerence and practice,for interrnetliute students oJEng1ish. NewYork: Cambridgc University Press.Norton, B. (1997). “Language, identity, and the ownership of English.” TESOL Quarterb, 31,409-429.Porreca, K. L. (1984). “Sexism in current ESL textbooks.” TESOL Quarterb, 18, 705-724.Powell, R. R., and Garcia, J. (1988). “What research says . . . about stcreotypcs.” Science nntlChildren, 25: 21-23.Quirk, R., and Grcenbaum, S. (1973). A conme grammar ~fcontemporay English. Fort Worth,TX: Idarcourt Brace Jovanovich College.Satikcr, M. (1981). “Diversity, pluralism, and textbooks.” In J.Y. Cole and T. G. Sticht (Eds.),The textbook in American society (pp. 4142). Washington, DC: Library of Congress (ERICDocument Reproduction Service No. ED 225 185).Sleeter, C. E., and Grant, C. A. (1991). “Race, class, gender, and disability in currenttextbooks.” In M. W. Apple and L. K. Christian-Smith (Eds.), The politics ofthe textbook(pp. 78-1 10). NcwYork: Routledge.Tanner, D. (1988). “The textbook controversies.” In L. N. Tanner (Ed.), Critical issues incurriculum (pp. 122-147). Chicago, IL: NSSE.Thomas, E. J., and Biddle, B. J. (1979).“The naturc and history ofrole theory.”In E. J.Thomasand B. J. Biddlc (Eds.), Role theory: Concepts and research (pp. 3-1 9). Huntington, NewYork: Robert E. Krieger.Todd-Mancillas, W. R. (1981). “Masculine generics = sexist language: A review of literatureand implications for speech communication professionals.” Communicafion Quarterk,Spring, 107-1 15.Wcitzman, L. J., and Rizzo, D. (1974). “Images of males and females in elementary schooltextbooks.’’ New York: National Organization for Women’s Legal Dcfcnsc andEducational Fund.

Indexabilities 16accountahilitj 2 53-4action research 197 207; tlcscription antlrationale 197 9; and 'real' rcscarch199 202; steps in pr(iactivities 185 6; organization of content antl189-90; selecting and tlcvcloping 188 9;see ulso tasksactivity ol)jcctives 182 3adoption ol'innovation 1 19, 122 3Adult Migrant English Program (AMEP)208-28; Certificates in Spoken andWrittcn English 21 5 25; communicati\capproaches 2 IO- I 3; curriculumdevclopmimt over 50 vcm 208; grmx-hawdapproaches 21 3 1 5; structui-al approachc%s208-1 0affective goals 182affects 10-1 1age 103 5Allwright, I). 201, 205Altman, H.H. 128analysis: antl classification 57 9; as thc prccursoioftask\ 60 2aquaculture outreach project 137 48art 103, 105, 106Asian Institute o1'Tcchnology, Thailand 1 38,7; Ace olo aquaculturc outreach projectcnt 190-1; CLTin Korca 160 I; CSWE22 I 2; 'I'OC 265Aston, G. 58 -9astrophysics papers 7 1attitude markers 76attitutlcs 161; students' 162; tcachcr\' 162, 264,269 70attrihutcs of innovations 123attrihutive hedges 76Austin, S. 105Australia: AMEP see Adult Migrant EnglishProgramAustralian Languagc Levels (ALI.) guidelines42 4authenticity 2 3 1 2availahilit) of material.; 112 14Ikhnian, L.F. 149 50Railcy, K.M. 201, 20513angalol-c Project 48Rarmatla, W. 176haw foi-m of vcrlx 51basic intrrpcrsonal communication skills (BICS)100 I, 104I3a/crnian, C. 80Iichavioural educational psychology 209Brnnv, K.1). 123, 242Ucrctta, A. 1 19Rhatia, V.K. 72Iilingiial education 9% 107Ih-ccn, M.P. 2, 257-8Krctt, P. 73 4Rricr, C. 276, 281Rrintllcy, G. 38, 178-9Hrown, H.D. 112Rrumfit, C. 121Rrunci 99 107burcaucrati\ation 74CALL (conipiitcr~a.;sistcd language learning) 4,240 9CALL co-ordinator (CC) 241, 246, 247CAI .I_ .l'cachcr Education Course (CALLTEC)24 1Canale, M. 149 50Candlin, C.N. 58, 257-8Cartcr, K. 64-9 powmcasc-stud) rcseai-ch 5, 265-73Cclcc-Murcia, M. 1 12, 2 12Certificates in Spoken antl Written English(CSWI:) 21 5-25change: liindamcntal 120- 1 ; innovation and 242

INDEX 285checklists 259-60Chin, K. 123, 242Cliomsk!,, N. 55, 210Clark, R. 174Clarkr, I). 147classification 57- 9classroom genres 74 5, 78 9classroom obser\ation 258, 267-9, 272clozc, self-crcatctl 132~ 3cognitii c academic language proticicnc! (c.4 I t')100 1, 104cognitive goal\ 182cohcrcnt c~rclatrtl sti-atcgics 60cohc\ion, lexical 57Coleman, H. 127collaboratii-c gi-oup \\ ork 26 1collocations 5 1common-core teaching 80 1communicatii e compctcncc 5 5--6, 149comniunicatii-c language teaching (CLT) 2,9-26;achie1ing23 5; .4MEPZl0-13;delining 14%50; tlilriculties due to 153,160 1 ; in I:l-L c'ontcxts 150 1 ; inadequatcaccount of EIL teaching 160; managing thelearning process 30-2, 33; materialsprotluction 2 34 5 ; and syllahus tlcsign184-5; tcachcrs' tlifliculties in introducing inSouth Korea 149-67; teachers'niisconcrptions ahout 1 55 655 7; syllahus tlc\ign 185- 6; targetcompctcncc 1 1-12compiitcr~assistctl language learning (CALL) 4,240 9computcri/ctl Iarigiiagc corpoi-a 2 3, 64 70consti-aintc and rcso~~rccs I92 4content: conceptualizing 18 3 7; lexis 46 54;organimtion of 189 90; and process 2; rolein CLT IC) 22content-l>asccl courscs 1 86contcnt q~ccilyirig lists 49 54context 24; formal antl infornial I6--17context~huiltling 223continuity: content antl 2 1 ; cultural 169 77contrilxitions, learner's 12 14, 15convci-gcncc htratcgies 60 1Cooper, K.1, 1 18core compctcncics 42-4corpora 2-3, 64 70countc%rparts 174-5course tlevclopincnt proc178-96(.ourscwork 1 14-1 5coverage objcctivch 182- 3critical literacy 94, 21 5critical thinking objectives 182-3c~ih-al c.ontinuity 169 77; pi-inciple 169 70culture 186Cummin\, J. 100 I, 104curriculum: rvaluation for piirpes ofcurriculum (lei clopmcnt 2 54; managing thclearning process 3 3--8; purI>o\c ol I0 1 1 ; wealso s! Ilabusciirriculum~planniri~ grid 44curriculum project managcmcnt 171 3cyclical-organisetl courscs 190tleconstructing text 224tlcticit pcrspccti\c 109de-lcxicalisctl item\ 5 Idescription 66 7tlei eloping countries 1 35tlc\iational perspecti\ c 109diffcrcntiation 15 16tliffusion olinno\ation 3, 118 26; paradigmshifting and 108 1 1diffusion of scholarship 1 1 1direction 15, 22disciplinar! \ariation 3, 71 83tlisciplinc-spccifc coinponcmts 80- 1discourse: AMEP 2 11-1 2; tcchnologicetldiscourse 170 3, 175, 176discourw compctencc 55 7(liscour\c s! Ilabus 55 6 3drills 92Dutllc! -E\ ans, I.. 74dynamic pcrspcctii c 109tast-Wcst Center, Hoiiolul~i 111cclcctic s! Ilahusc.; 56educational gron th, local I 6 3educational syllabus 46-7ctlucationd system 1 5 3, 158 9ctlucationi \aluc.s/attitud(~s 161Etl\vartls, K. 170EFI teaching: CLI"5 inadcquatc. account 01' 160Ellis, K. 267 9cmcrgcnt language pi-ogram fi-ainc\\ ork 1 37 48emphatics 76cinpirical~rational inno\ation strategies 121 4empirical response 231 2cmpoizcrnicnt 173 5cpistcmic grammatical subjects 77-9cquililxium paradigm 1 1 1-1 2evaluation 5, 173; action rcscarch 202 4,205-7; aquaculturr outwach Ixojcct 146;CLT 22 3; course tlevclopmcnt pro

286 INDEXexpantling 145-6Expanding Circlr 1 10 I I, 1 12 1 3expectations, learner 13-14experiential content domain 29-30experiential, student-gcneratcd material5 2 37- 8‘cxperts’, expatriate 174-5expressive functions 1 12facts, computer corpora as 65first order innovation 243Flew, A. 175focus 20formal learning context 16~ 17formative evaluation 23, 191, 254function \\ ortls 51functions 54, 2 1 1 ; syllabus design 184-5fundamental change 120-1funding, insufficient 159Gapmaster 243, 244gender bias 5, 275-83General Secondary Education rxaminationpapcrs 87 9Gcncsee, F. 103genre: AMEP and genre-based approaches2 13-1 5, 220-1, 225 6; disciplinaryvariation 72 - 5; gcnrc-based approach tomaterials production 2 36-7; LINC and genretheory 93-6genre-related strategies 60qeocmtrism I 14 I5gcography 105, 106Gcrman Model 1 05goals: long-term 230 I; setting 39 45, 181-3Gorlach, M. 11 3govcrnmcnt(s) 88, 89, 92, 97Gouer, R. 182- 3grading teachers 258 9grammar 32, 162, 184; LINC and 90 2grammar-based examinations 158Grammar in Use (Murphy) 278~-80grammatical subject 77-80group projects I34group work, collaborative 261grouping 39,41, 130, I34Hallitlay,M.A.K. 75,94,97,211,213, 214,277Hamnctt, M. 109 10Havelock, K.G. 1234Hayes, I>. 175-6hedges 76Henrichsen, L.E. 123, 124Hewings, A. 78 9high-structure teaching 32- 3, 34-5highway engineering 74Holliday, A. 150Holmes, R. 74I long Kong 263-74‘hourglass’ articles 73humanistic education 29humanistic/ps!.chological oricntation 38-9, 4 1Hutchinson, T. 141hvhrid models 124Hqland, K. 75-6, 80Hvincs, D. 21 1illuminativr evaluation 255implementation: factors affecting 264-5impro\emcnt, innovation and 121incluaions 47 9indcpcndent construction 224indigenization 109--10individualization 1 27 36informal lcarning 16--17information rcport 93-4initial contributions, learner’s 12-1 4Inner Circlc 1 10, 1 12-1 3innovation strategirs 123 4institutionalisd utterancrs 5 1instrumcntal functions 1 12integrative functions 1 12interacting needs 141, 142interactive tliscoursc 58- 9, 62interactive learning model 2 12 13intrrdependent participants 17-1 9interlanguagr continuum information 1 12International Ccrtificate Confercnce (ICC) 62intrrpcrsonal function 75interpretational response 23 1-2involvement objectives 182-3Ivanic, R. 174Jacob, G. 169Jacobson, W.H. 141-2Jamcs, G. 259Johnson, M. 53joint construction 2 15, 224Kachru, B B 108, 112, 113Kennedj, C 119Kenn) , R 144, 2 3 1ke!\vord\ 89 90, 92-3knowledge about language 3, 87-98Korea, South 149-67, educational system 158-9Kouraogo, P 261Krashcn, S 120Lakoff, G. 53language: knowledge about 3, 87 98; and thesyllabus 48 9; as tcxt 2 13-14; views oflanguage and languagr teaching 88-9language acquisition 269; agc and 103-5language awareness 96language competence 149 50language corpora 2 -~3, 64-70

INDEX 287language fcatures 22 ILanguage in thc National Curriculum (LINC)project 3, 87-98; ban 89-90; lessons from96--7language proficicncy orientation 38-9, 4 1language use needs 141, 143largc classes 158; intlivithalization and 127 36Larking, L. 102last stressed elcnicnts 277-8, 279learner-ccntredness 4, 170 I; AMEP 210-1 3;managing the learning process 27-9, 29- 30,38 -9, 40; materials production 232lcarncrs 248-9; actively involving in ncdhanalysis 137 48; and CI T 11-14, 15,18 19; contributions 12 14, 1 5; andintroducing CLT 153, 156-7, 162; mcctingnee& of different lcarncrs 224; antl paradigmYhifting 1 1 I 12; participation in c\aluatingmatcrials 257-8; roles 18-1 9, 28learning: language 229-32; language-basedtheory of‘2 14; lcarning-teaching context16-~17; scaffnldcd 214-15; styles antl CAL2434; tension of syllabus v language andlearning 48-9; ways of and TOC 265learning-ccntredness 29-30learning nectls 180 1learning process, management of 2, 27 45learning process goals 3 6 7learning stratcgics 41, 186-7learning tasks see taskslcnding (class) library 1 3 1 , 1 34Lewis, C. 203 4Lewis, M. 68lexical approach 2, 46--54, 68lexical syllabus 56Lincoln, Y. 110linguicism I I Slinguists 89, 97linkage models 124linking related texts 224LIPT action rrscarch project 2 024listening skills 185local educational growth 163Logan, G.E. 127- 8Long, M. 112long-term goals 230-1Lon, G. 257, 258low-structurc teaching 32--3, 34-5Macaulay, M. 276, 281MacDonald, S.P. 77 8, 79Mackcn, M. 221Malaysian Communicational Syllabus 5 5managcment of the learning process 2, 27-45Masch, N. 105mastery objectives I82 3materials: availability of 11 2-14; evaluation of256 8; pro1)lenis in using CALL 245 6;selccting and tlc\cloping 188-9rnatcrials protluction/drvclopment 229- 39; lackof rxpcrtise and time for 156; samplematcrials 233-8; theory 230 2mathematics 103matrix-organixd course5 190McCarthy, M. 52media I 5-1 6metacogniti ve nccds 141 , 144mctacommunication 14-15metadiscourse 75-6metaphorical patterning 5 3 4methodology: CLT 14- 17; CSWE 222 4Mickan, P. 202 3microteaching 260 1missed opportunities 243- 5modal auxiliaries 5 Imodal verbs 90-2modelling text 224models of innovation 123 4motivation 157moves 72-3Munby, J. 58Murphy, R.: Grurnmur in IJsc 278 80name tags 130, 134national curriculum: Australia 2 15 16; Englandantl Wales 89natural language learning 120, 2 10- 1 1needs, long-tcrm 230-1needs analysis 38-9; actively involving learners13748; course development proccsseyframework 178 8 1needs-hascd communicative approaches 2 10 1 3negotiated spllahus 147Nicholls, A. 120normati! e-re-cducati\ c innovation strategics1234notional/functional syllabuses 121notional-structural approach 2 33notions 2 11notions and topics 184-5‘Nuc1eus:English tor Scicncc and Technology’233Nunan, L). 28, 29, 32,40,48, 198, 200 Iobjective knowledge 200-1objccticc needs 178-9objectives, setting 3945, 181 3opportunities, missed 243 5oral skills 154, 162, 185organization of content and activities I89 90Outer Circle 110-1 I, I I2 13ownership 173-5paradigm shifting 108-1 I ; students’ difficultyIll 12

288 INDEXparticipant roles 277 8, 279 80participatory processes I86 -7I'atton, M. 108 9I'aulston, R. 111peer c\ aluation 260L 1, 272pccr teaching 260pcrccii ctl attributes of innovations 109Pcrkins, D.N. 243person markers 76phenomenal grammatical subjcrts 77 9physical education 103, 105, 106planning grids 44planning strategies 60plant hiology 74politeness stratqics 60politics 105; and kno\vlctlgc about languagc 3,87- 98Porrcca, K.L.. 276po\vcr-cocrcivc innovation strategies 123 4Prabhu, N.S. 48, 120prescription 66 7prc-s> Ilahus 59, 60pincipal, rolc of 271prohlcm-sol\ ing motlel 123 41)rohlcniatizing 192-4process 2; txaluation of curriculum proccs\22 3; management ofthr learning procc\s 2,27 45~I-OCCSS compctcncc 1 3process project 171 3profcwional dc\clopmcnt 197 9, 259 60,272 3professional genres 72 4, 75 6, 77 8, 79proficicncy: languag' prohcicncv orientation38-9, 41; lo\v antl introducing CI T 156 7;movement 185Iiroficicnc> goals 182proficiency tests I80~~i-otilc cards 1 10-~1 , 1 34progressivt. [mlagogics 2 10 1 1project managcmcnt 171 3pronunciation 184proportional syllahiis 56I)S)chological/humanistic orientation 38 9, 41purposc: of curriculum 10- 1 1 ; piirp(~scs forc\ aluation 25 3-62; specific purpoworientation 38 9, 41cfuc\tionnaii-c\ 1 30, 180radio ne\\ \ 1 32, 1 34Ramii-e/, J [I 99, 101iapport 129, 129 3 1reading learning need\ 141, 143, \kill\ 162, 185record\, corpora a\ 65 6r( education 128reflccti\r practlcc 198, 272 3rcgi\tci 220-1rclational markel-s 76relevance 129rcliahility 201repair \tratcgics 61rcpcrtoire: language teaching repertoire 224-5;target rcpertoirr 11 12, 19 ~20reporting hack 145, 237 8research: action research antl 199-202rcscarch articles (KAs) 72-4, 75-6, 77-8, 79lopmcmt and tliffubion (KD and U)n10dcI 123 -4rchcr-tcacher collaboration 272ancc: to change I 19, 122, 264; to classparticipation 157resources antl constraints 192 4I-csponsil)ilit~ 129rhcmc 277-8, 278-9Richards, J. 1 I2Kogcrs, E.M. 109, 119, 123romanticism 88routes 15. 22Samraj, R.I'.K. 77 8, 79Saphicr, J. 182 3scaffoltling 2 14 1 5scholarship, diffusion of I 1 1scirncc 101, 103second order inno\ ation 243sclf-created clcm' 132 3self-dcvclopmcnt, teacher 255, 260 1sclf-c\ aluation 260- 1sclt-learning programmcs (SI P) 1314scniantically densc itcyns 51wntcncc hcatls 51 2sequencing 20 I, 47 9, 189Singleton, I). 103-4situational language tcaching 208-10skills 54, 162, 185skills-hasctl rtlucation 170--1Sladc, D. 221Smith, II. 174social Iwhaviour 90social interaction motlel 123 4wcial prominence 275 83; analyhg gender biasas in tcxts 277 -80social sciences 74wcially validated response 2 3 1 -2wciocultural context 121socio-linguistic competence 56, 154 5speaking skills 154, 162, I85specific pui-posc orientation 38 -9, 41stakchol(lcr-ccntrctl approach 171- 3, 175stance75 6stanclard English 90Stcnhousc, I.. 254Stephens, J.M. 24steps 72 3Stern, H.H. 182

74INDEX 289Story1x)artl 243 4, 245strategic competence 56, 154-5strategic pre-syllal)us 59, 60stratcgics: tliscoursc 60 2; innovation 123-4;learning 41, 186 7structural approaches 183; AMEP 208-10structural indigenization 1 10student-gcneratcd, experiential materials 2 17 8stutlcnts see learncrssubdivision 2 1subjective kno\vledgc 200 1suhjectivc needs 178-9bilingual cducation and 99 107; SEC alsodisciplinar! \ ariation, tliscipline~.;pccificcomponentssuhstantive intligcmization 1 I0aluation 23, 191, 253-4support, teacher 159, 272, 273supra-scntcntial linking 52Swain, M. 149-50Swales, J. 72-3syllabus: AMI:P/CSWE 220, 222--5; design antldisciplinary variation 80 1 ; discourwsyllabus 55 61; Icxis in 2, 46-54; tension ofsyllabus v language antl learning 48 -9syllahus grids 184 7synonyms 52‘synopsiaing’ words 52‘Talkhasc’ approach 237 8targct competence 11-1 2targct need.; 180-1Target-Oriented Curriculum (I‘OC) 26 3 ~target repertoire 1 1 12, 19-20targets 265Taronc, k, 73task-bascd learning 2, 27-45, 59tasks 61 2, 265; aquaculturr outreach project145; individualizing learning tasks 1 31 4;syllabus design 185 6Taylor, G. 173teacher education 155, 162 3, 264teacher support 159, 272, 271tcachers 248 ~ 9; attitudes 162, 264, 269 70;grading 258 9; prrccived tliffjcultic\ inintroducing CLT in Korea 149-67;professional dc\elopmcnt 197 9, 2 59-60,272 1; purposes For evaluation 258 62;re.;earchcr-teacher collaboration 272; 1-01~ inCLT 17 18; sclf-development 25 5, 260 1 ;understanding of innovations 264 5, 270;view of needs assessment 180; views onlanguage and languagc teaching 88, 96-7tcaching: dominant discourses 170-1 ; highstructureantl low-structurc 32-3, 34 5;language teaching rcpertoire 224-5; textbased cycle 222 4Teaching knglish to Speakers of OtherLanguages (TESOL) pi-cparatory programs108-17technical texts 2 16-~7technocentrism 241 2tcchnologisctl tliscourse.; 170-3, 175, I76Tcrrell, T. I20tcxt: language as 2 1 3- 14; mctadiacoursc 75 6;tcxt l)ascd syllabus design 222 5theme 277 8, 278-9thcoretic intligcnization 109 10‘thing’ technologies 242training, tcachcr 155, 162- 3, 264transactional language 62transfer goals 182‘I.riin, J.L.M. 102understanding of‘ innovations, teachci-s’ 264 5,270units of nark 22 1University of Illinois at Urbana- Champaign(UIUC) 111University of Mala)a Spoken knglish Project234 5Ur, P. 222‘us’-‘thein’ configuration 175Usher, K. 170validity 201value-adding activities 247, 248 9valuc-synonyms 52values 161Van tk, J.A. 102Van licr, L. I97variation 2-3; disciplinai-v 3, 71 83Vaxrus, F. 109vc>rhs, hasc form of 5 1Vcrapoor, A. 264wcahlary 49, 184voluntary lcarning 130Vygotsky, L. 21 5Wallace, M. 198Waters, A. 141Witltlo\vson, 11. 48, 58, 142 4, 146Wilkins, 0. 49, 21 1Willis, 1). 48, 49-50, 68\+ords 49-54work-related needs 1 3940World Englishes paradigm 108 17‘would’ 5 1writing: learning ncctls 141, 143; skills 185Yaltlen, J. 571 .*-~-.‘”~.~‘-,...~-~,,.-.e,;-_