ELT journal july 2009

An international journal for teachers of English to speakers of other languagesTraversing the lexical cohesion minefield Critical reflection in a TESL course: mapping conceptual change C

An international journal for teachers of English to speakers of other languages

Traversing the lexical cohesion minefield Critical reflection in a TESL course: mapping conceptual change

Challenges in teaching ELF in the periphery:

the Greek context Why and how textbooks should encourage extensive reading Point and counterpoint

Process-oriented pedagogy Text messages

A tale of two songs Comment ELT and the challenges of the times

ReviewsTeaching Other Subjects through English Cross-Curricular Resources for Young Learners Uncovering CLIL

Developing and Assessing Intercultural Communicative Competence

To Get to Know Each Other Leads to Better Mutual Understanding Teaching Modern Languages to Young Learners

Teaching Foreign Languages in the Primary School Literature and Stylistics for Language Learners Oxford Learner’s Thesaurus

Building a Validity Argument for the Test of English as a Foreign Language™

Form-focused Instruction and Teacher Education Websites review

IATEFL Cardiff Online 2009

More support for your

Young Learners

classroom

Grammar for Young Learners

nReady-made grammar activities

to suit different learning styles

nStrengthens grammatical

accuracy in a fun and purposeful

way

nFocus on grammar through

drawing, storytelling, songs and

games

Storytelling with Children

nFully revised with new ideas and stories

nGuidelines on combining stories with drama, poems and music, cross-curricular studies and personal development

nSelection of ready-to-tell stories, photocopiable worksheets and easy-to-draw pictures

Every effort has been made to trace the owners

of copyright material in this issue, but we shall be

pleased to hear from any copyright holder whom we

have been unable to contact If notified, the publisher

will attempt to rectify any errors or omissions at the

earliest opportunity

© Oxford University Press 2009

All rights reserved; no part of this publication

may be produced, stored in a retrieval system, or

transmitted in any form or by any means, electronic,

mechanical, photocopying, recording, or otherwise,

without either the prior written permission of the

Publishers, or a licence permitting restricted copying

issued in the UK by the Copyright Licensing Agency

Ltd, 90 Tottenham Court Road, London W 1 PS 9 HE ,

or in the USA by the Copyright Clearing Center,

222 Rosewood Drive, Danvers, Massachusetts 01923,

USA

UK ISSN 0951-0893 (print); 1477-4526 (online)

Typeset by TnQ Books and Journals Pvt Ltd.,

Oxford Journals Advertising

60 Upper Broadmoor Road

Article titles, abstracts, and Key concepts appear free

online through the ELT Journal website:

http://eltj.oxfordjournals.org

Abstracting and Indexing

British Education Index covers ELT Journal

The Advisory Board

IATEFL Graham Hall Northumbria University Éva Illés

Eötvös Loránd University, Budapest Barley Mak

The Chinese University of Hong Kong Jonathan Marks

Leba, Poland Alice Oxholm Sheffield Hallam University Annamaria Pinter University of Warwick Barbara Skinner University of Ulster Jane Spiro Oxford Brookes University Melinda Tan

University of Central Lancashire

Key Concepts Editor

Alan Waters University of Lancaster

Text Messages Editors

Jill and Charles Hadfield

Editorial Front Office

E LT Journal is a quarterly publication for all those involved in the field

of teaching English as a second or foreign language The journal linksthe everyday concerns of practitioners with insights gained from relatedacademic disciplines such as applied linguistics, education, psychology,and sociology

E LT Journal aims to provide a medium for informed discussion of theprinciples and practice which determine the ways in which the Englishlanguage is taught and learnt around the world It also provides a forumfor the exchange of information among members of the professionworldwide

The Editor ofE LT Journal is supported by an Editorial Advisory Panelwhose members referee submissions Their decisions are based upon therelevance, clarity, and value of the articles submitted

The views expressed inE LT Journal are the contributors’ own, and notnecessarily those of the Editor, the Editorial Advisory Panel, or thePublisher

Developing

Countries Initiative

What is the Oxford Journals Developing Countries offer?

This offer represents our commitment to providing free and greatly reducedonline access to many of our journals for low income countries

What does it include?

Institutions within qualifying countries can apply for the full DevelopingCountries collection, the Humanities and Social Science subset, or theScience, Technical, and Medical subset

Which institutions qualify for this offer?

This offer is available to established not-for-profit educational institutionsfrom qualifying countries based on country incomes as established bythe World Bank Report 2006 Access is either free or greatly reduceddepending on which list they appear in

For full details see: http://www.oxfordjournals.org/access_purchase/developing_countries.html

Back issues Subscribers toE LTJournal have access online to articles, features, and

reviews from 1996 to the present—together with ‘advance access’ toforthcoming articles You can search the archive by author, title, or by anykeyword: http://eltj.oxfordjournals.org

Some institutional subscribers (e.g university libraries) may also havesubscribed to the Oxford Digital Archive, which gives online access to everyissue ofELTJournal from Volume 1/1 (1946) to the present

Individuals, or institutions not wishing to subscribe to the whole DigitalArchive, can get access to articles and features from 1981 to 2006, and

to reviews from 1998 to 2006, by purchasing an archiveC D - R O M This

is searchable by author, title, or by any key word: publications.co.uk

ELT Journal Volume 63 Number 3 July 2009

Articles

Jean Wong and

Hansun Zhang Waring

‘Very good’ as a teacher response 195

John Bitchener and

Point and counterpoint

William Littlewood Process-oriented pedagogy: facilitation, empowerment, or control? 246David M Bell Another breakthrough, another baby thrown out with the bathwater 255William Littlewood OBE: a coin with two sides or many different coins? 263

Steve Darn Teaching Other Subjects through Englishby S Deller and C Price,

Cross-Curricular Resources for Young Learnersby I Calabrese and S Rampone,and Uncovering CLILby P Mehisto, M J Frigols, and D Marsh 275Silvija Andernovics Developing and Assessing Intercultural Communicative Competence: A

Guide for Language Teachers and Teacher Educatorsby I La´za´r et al.(eds.), and To Get to Know Each Other Leads to Better MutualUnderstandingby M Bedynska et al (eds.) 277

Simon Smith The TeMoLaYoLe Book: Teaching Modern Languages to Young Learners

by M Nikolov et al (eds.), and Teaching Foreign Languages in the PrimarySchoolby C Kirsch 280

Amos Paran Literature and Stylistics for Language Learners: Theory and Practiceby

G Watson and S Zyngier (eds.) 284Stephen Coffey Oxford Learner’s Thesaurus: A Dictionary of Synonymsby D Lea

(chief ed.) 288

Jesu´s Garcı´a Laborda Building a Validity Argument for the Test of English as a Foreign

Languageäby C A Chapelle et al (eds.) 291Darren Elliott Form-focused Instruction and Teacher Education: Studies in Honour of

Rod Ellisby S Fotos and H Nassaji (eds.) 295

Websites for the language teacher

Diana Eastment IATEFL Cardiff Online 2009 297

Correspondence 300

IATEFL 302

Please visitELT Journal’s website athttp://eltj.oxfordjournals.org

‘Very good’ as a teacher response

Jean Wong and Hansun Zhang Waring

Much scholarly and pedagogical attention has been devoted to correctivefeedback In this paper, we turn to positive feedback, and in particular, call for

a reconsideration of teachers’ use of explicit positive assessments such as ‘verygood’ Based on examples from an ESL classroom, we show that utterances such

as ‘very good’ may have the potential of inhibiting learning opportunities withinparticular pedagogical contexts We then broaden our discussion by offering

a range of suggestions for managing the complexities of positive feedback in thelanguage classroom

Introduction An integral part of language teaching is giving feedback As Fanselow (1987:

267) writes, ‘to teach is to provide feedback’ Over the past three decades, wehave made great strides in understanding the various facets and strategies

of feedback in language teaching Much of the scholarly inquiry, however,has been devoted to feedback giving when something goes wrong, i.e.negative or ‘corrective’ feedback (Gass and Mackey 2006) In this paper, weask what kind of feedback teachers should give when nothing appears to begoing wrong What do we say when a student has just produced a correctresponse? To many, the answer may be obvious, uninteresting, orunimportant We argue otherwise To that end, we will first introduce somebackground on positive feedback and its related practice of ‘praising’ Wewill then briefly show how the use of ‘very good’ may inhibit learningopportunities in a particular pedagogical context Finally, we will offer someteaching suggestions on responding to correct student contributions inways that possibly promote learning

Background In contrast to the large body of literature on corrective feedback, work on

positive feedback is difficult to find Allwright (1980) categorizes positivefeedback such as ‘fine’ or ‘good’ as part of the ‘quality judgements’ integral tothe guidance we give as teachers in the language classroom Some empiricalwork on positive feedback has addressed how it is done Based on datagathered from English language classrooms, Seedhouse (2004: 206–7)claims that positive evaluation is often implied in the absence of feedback inthe initiation–response–feedback sequence (Sinclair and Coulthard 1975)

By examining 25 hours of classroom discourse, Hellermann (2003: 88)shows that positive assessments done in teacher repetitions arecharacterized by:

1 rhythmical placement synchronized with student response,

2 falling pitch contour,

E LT Journal Volume 63/3 July 2009; doi:10.1093/elt/ccn042 195

3 mid-level pitch,

4 longer duration than student responses

Others have considered what positive feedback really accomplishes.According to Mehan (1979: 64), positive evaluation is a ‘terminal act’ thatmarks the final boundary of a sequence (cf Schegloff 2007) Utterances like

‘very good’, for example, can merely signal that it is time to move on to thenext person (Fanselow op cit.) or the next activity (‘transition ritual’ inBrophy 1981: 18)

One function of utterances such as ‘very good’ is to praise—a way ofreinforcing a student’s giving of a correct response, which, in the context oflanguage teaching, means reinforcing correct comprehension or

production of a language structure, for example Notably, the correctness of

a student’s response is not necessarily a key consideration in whether

a teacher offers praise Brophy (op cit.) maintains that teachers sometimesoffer inappropriate praise, lauding students for incorrect answers as well ascorrect ones In citing O’Leary and O’Leary, Brophy (ibid.) indicates thatthree features must be present in order for praise to function as

reinforcement First, the praise must be contingent on the actual execution ofthe behaviour that is being reinforced Second, the praise must be specificabout the behaviour being reinforced Third, the praise must be sincere andaddressed to the particular context in question

For example, one problem with the use of ‘very good’ in a second languageteaching context, according to Fanselow (op cit.), is that if a teacher uses thephrase ‘very good’ in response to a student’s utterance, ‘I extremely happy’,

to what does the teacher’s praise refer? Fanselow (ibid.) argues that theprecise target in this case may be ambiguous It may be that the teacher ispleased that the student is happy, or the teacher may be overjoyed that thestudent has produced a response at all Alternatively, it is conceivable thatthe teacher is only responding to the portion of the utterance that is correct,despite the fact that the utterance produced by the student is not fullyaccurate (Fanselow ibid.: 281)

Clearly, feedback giving, and in our case, positive feedback giving, is not

a straightforward task More experienced teachers, however, may be betterequipped to manage its complexities Forgas and Tehani (2005), forexample, report that experienced feedback givers are mindful of the impact

of mood on feedback and, accordingly, they give more positive and politefeedback when they are in a sad mood They remain alert and compensatefor their sad mood in giving proper feedback

In sum, even a simple item like ‘very good’ has its many faces A plethora ofissues surround its use The cases discussed below are used as a point ofdeparture for rethinking how a language instructor should respond tostudents’ correct answers or responses, at least, on some occasions like theones displayed in the ensuing discussion

unintended effect of shutting down learning opportunities by signalling notonly sequence closing but also ‘case closed’ In particular, we show a fewinstances of classroom data in a form-focused check-homework context,where the focus is on checking learners’ ability to use ‘present perfect’ or

‘present perfect progressive’ The brief analysis given below is derived from

a more detailed conversation analytic treatment of a much larger amount ofrelevant data (see Waring 2008) The transcripts presented below have beensimplified for readability The only notation unfamiliar to the reader may bethe two sets of vertically aligned brackets, which indicate simultaneous talk

or overlapping non-verbal conduct (indicated in double parentheses) bydifferent participants

In the first instance, the relevant exercise item is:

Wow, I didn’t know you were married

How long _?

(Purpura and Pinkley 2000: 73)

In Extract 1a below, Miyuki raises a question regarding this item:

Extract 1a 1 Miyuki I have one [ques]tion,

2 Teacher [Yes.]

3 Miyuki Number three is if without ‘be’ is not good?

4 Teacher How long you’ve been marrie[d?

5 Miyuki [Have you married

6 have you married

This sequence spans 75 lines of the transcript and lasts two and a halfminutes It turns out to be the most complicated error correction sequence

in the two-hour class Briefly, Miyuki has treated ‘marry’ as a verb, in whichcase its correct present perfect form would be ‘have married’, except that thepunctual aspect of ‘marry’ is ill-fitted to the duration query of ‘how long’(that is ‘marry’, like ‘find’ or ‘explode’ and unlike ‘sleep’ or ‘work’ are verbsthat entail no duration) Since the form of ‘married’ may be either a verb

or an adjective, Miyuki’s confusion is not surprising One wonders,however, why Miyuki did not raise her concern much earlier when the

‘married’ item was first being dealt with Here is what happened fourexercise items and 66 lines of transcript earlier:

Extract 1b 1 Teacher Number three Kevin

2 Kevin ‘Wow I didn’t know you were married’

3 ‘How long have you [been married’ ]

4 Teacher [((encouraging nods))]

5 ((emphatically)) Very good How long have you been

6 married ((smiley voice)) Very good Number four

Note that from line 4 onward, the teacher’s ‘very good’ is deliveredimmediately and emphatically, along with the encouraging nods, the smileyvoice, a repetition of the response as well as a second ‘very good’ Aside fromthe fact that her turn components emerge one after another without leavingany space for others to come in, this dramatic combination of verbal andnon-verbal expressions not only accepts Kevin’s response as correct but alsoputs it on a pedestal, so to speak One might argue that against the backdrop

of this finale-like, heavily advertised ‘perfect’ answer, any attempt toarticulate understanding problems or explore alternative answers, as inMiyuki’s case, would appear less than expected or desirable

In the next segment, we see that the ‘case-closed’ quality of ‘good!’ is sostrong that even when the teacher offers more space for learner contributionimmediately thereafter, no uptake emerges The relevant exercise item isthis:

In fact the team (5) (won) _ 98% of the games they (6)(play) so far

(Purpura and Pinkley op.cit.: 32)

Extract 2a 1 Teacher Okay? Who’s next? I think Jae? Is that you?

2 Are you next? Alright

3 Jae ((reads)) ‘In fact, the team has won uh

4 ninety eight percent of the games’

5 ((pause))

6 Teacher ((th[ree consecutive nods)) ]

7 Jae [they have played so far.]

8 Teacher Good! ((in excited tone)) In fact the team has won

9 ninety eight percent of the games ((in staccato

10 tempo)) they ((pause)) have ((pause))

that is important, salient, and worthy of remembering It serves to establishthe singular, irrefutable correctness of Jae’s response, thus implicitlyproposing that the case is now closed.

Despite this finale-like interactional state, in the ensuing space, the teacherdisplays no urgency to move on The long gap during which she walksaround provides an opportunity for learners to ask questions about the twojust-completed verb forms—‘win’ and ‘play’ The subsequent ‘Is everybodyokay?’ makes available another window of opportunity Yet, no questions areraised; all seems well That is, until 418 lines of transcript and ten itemslater:

Extract 2b 1 Marie number five uh ‘the team has very good players’ In

2 fact, the team is winning or

3 Teacher has won

Clearly, Marie has not fully grasped what the correct answer is or why it iscorrect as opposed to any alternatives, and the earlier ‘very good’ closing didnot seem to present a favourable environment for voicing her

understanding problems

There is, of course, always the issue of whether Miyuki or Marie had anyconcerns to voice earlier on in the first place One might argue, for example,that their questions emerged over time Since we are not privy to whatwas going on in their heads at the time, what we are proposing is that hadthere been a more ‘inviting’ space for student concerns when each itemwas initially dealt with, there might have been more room for thoseconcerns to be developed and articulated, and that the uses of ‘very good’ inthese particular contexts have not been conducive to creating that space

In sum, there is some evidence that the use of ‘very good’ delivered in

a particular tone and/or package may be inhibiting learning opportunities atleast in a form-focused context This outcome or by-product may be acutewhen the context is a language learning setting, one in which directspeaking opportunities in class and the frequency of them may contribute toand impact students’ developing mastery of the target language In whatfollows, we expand the discussion to the use of explicit positive assessmentssuch as ‘very good’ in general and propose some suggestions for teaching

Suggestions for the

a more general call for awareness, reflection, and action research

Use ‘very good’

sparingly

Arguably, in some circumstances feedback tokens such as ‘very good’should be used sparingly or even hardly at all especially with higher-levellearners who may need less reinforcement or ‘stroking’ in the first place(Brophy op cit.) In fact, learners typically assume that an answer given is

correct unless teachers tell students otherwise (Brophy ibid.; Seedhouse op.cit.) Herein might be a small time-saving mechanism, i.e not offeringpositive feedback after every student response, particularly with moreadvanced learners.

Produce ‘very good’

with ‘non-final’

intonation

Teachers might say ‘very good’ using a mid-rising intonational contour,which has the effect of functioning as a continuer, soliciting ‘more’ orfurther responses from the students In other words, utter ‘very good’ withaccompanying appropriate non-verbal cues so that the feedback gives off

a ‘non-final’ rather than a ‘finale-like’ tone

Accept with less

evaluative tokens

The teacher may accept the student’s correct response with less evaluativetokens such as ‘okay’, ‘alright’, and the like In fact, there is some evidence inWaring (op cit.) that when ‘okay’ is used instead of ‘very good’, students mayproceed to ask follow-up questions about the just-completed item

Ask ‘permission’ to

move on

The teacher may wish to give a simple, quick nod of the head up and down,which implies approval of the student’s correct answer in a non-verbalmanner and immediately follows up by saying ‘Okay to move on?’ If theoriginal respondent to the item does not have any problems with moving on,then the teacher turns to the whole class and asks again ‘Okay to move on?’Providing feedback in this manner is akin to ‘opening up closings’ whichgives added interactional opportunity spaces, if needed, for anyone in theclass to put forth ‘unmentioned mentionables’ (Schegloff and Sacks 1973).This technique may be important particularly for those students who arereticent to speak up and may need extra encouragement or interactionalspace in which to do so In contrast, note that in Extract 2a above, theteacher’s production of ‘good’ in an excited manner and her repetition ofthe correct student’s answer with staccato tones and pauses served toclose off further student questions even though she also asked ‘Is everybodyokay?’

Problematize correct

responses

Teachers might help students become more actively engaged with thelearning by problematizing a correct answer We do all kinds of things inresponse to an incorrect answer, such as silence, hesitation or delay,questioning certainty (‘Are you sure?’), asking for repetition or clarification(for example, ‘Can you say that again?’) If we use these same strategies for

a correct answer, chances are students will try harder to reach anunderstanding of not just what a correct answer is, but why it is correct

Ask ‘pursuit’

questions

Teachers might respond to a student’s correct answer by pursuing withquestions such as: ‘Why do you say that?’ ‘How did you get or arrive at thatanswer?’ ‘Go deeper into why this is a correct response Can you explain?’

‘Explain why this is a correct answer based on what we have just learnt (orbased on the grammatical rules we have just studied)?’ This kind offeedback affords the student an opportunity to support or defend his or heranswer and to display confidence that what he or she has just said is correct

does someone else have?’ ‘Do you all agree?’ ‘Does anyone have a differentanswer?’ These kinds of feedback questions are not intended to imply thatthe one who answered initially had an incorrect answer, although theteacher may need to do some initial work to change this perception or

‘habit’, given that it is a common practice in classrooms that when a teachercalls on more than one student regarding a particular exercise item, it ishighly probable that the one who originally responded was not entirelycorrect Alternative positive feedback questions such as those suggestedhere may open up the classroom floor for further learning opportunities,allowing students the chance to question, debate, or agree with answers given

‘judgement’ on the item first

Recognize the

potential negative

impact of ‘very good’

Based on our analyses of the classroom data displayed above, which may betaken as indicators of what does occur in real teacher–student interaction onsome occasions of form-focused instruction, we would caution teachers that

in praising students for giving correct responses by offering positivefeedback tokens like ‘very good’ (‘excellent’, ‘good’, ‘wonderful’, or the like),ironically, this may have a negative impact on the learning situation(potentially), shutting down the sequence, which may lead to closing off offurther student participation, for example, students’ further questionsand comments Teachers should use feedback tokens such as ‘very good’appropriately, being sensitive to the contexts in which utterances of thiskind may inhibit rather than encourage student learning and participation

Engage in

self-reflection

Just as Forgas and Tehani (op cit.) have noted that experienced feedbackgivers are mindful of the impact of mood, and how that may affect the kindand frequency of feedback they give, we would suggest that (language)teachers, particularly those who are novices but perhaps more experiencedones as well, engage in some form of self-reflection of their positive feedbackmethods and utterances, for example, paying attention to when they use, donot use, or even overuse feedback tokens like ‘very good’, ‘excellent’, or thelike Our analyses of the data shown above reveal instances of ‘very good’ aspossibly shutting down learning opportunities in form-focused instruction.Teachers might examine form-focused and other instructional contexts inorder to get a (better) sense of when they use positive feedback tokens such as

‘very good’ (if at all)

Conduct action

research

Teachers might engage in action research and have themselves videotapedteaching a (form-focused) lesson, and subsequently transcribe and analysethe videotape by themselves or with another colleague, focusing on how,where, and when they use feedback tokens like ‘very good’ or whatever elsethey offer to students in terms of positive feedback, praise, or

encouragement when students give a correct response (some other forms ofpositive feedback may fall in the domain of non-verbal behavioural cues).Another side of this coin would be to include a ‘coding scheme’ for positivefeedback, noting when, where, and what was provided as positive feedback

on teacher observational forms used by supervisors when conductingrequired teacher observations On the supervisor’s part, the goal would notmerely be to check off that positive feedback was provided, as if it were sometaken for granted ‘default setting’ but to examine more actively the kind offeedback utterances which a teacher employs, their use and appropriateness

in terms of furthering or possibly stifling students’ participation andlearning opportunities If this kind of notation of positive feedback isincluded as a dimension in teachers’ observational reports, teachers mostcertainly would (begin to) engage in self-reflection concerning how andwhat they provide as positive feedback Regular self-reflection and(required) observational reports along parameters such as those proposedmay enhance the quality of teacher performance and better serve theinterests of those whose continued educational growth is at stake, that is thestudents

What teachers might find as a result of their self-reflection and actionresearch is that ‘very good’ is not something to be avoided at all cost When

‘very good’ is used along with further ‘pursuit’ questions or the elicitation ofpeer contribution as discussed above, its ‘case-closed’ quality may besignificantly mitigated Moreover, in a less form-focused context where thetask is ‘open’ (Kahn 2008), ‘very good’ may just provide the exact amount ofencouragement students need in continuing their exploration Finally, ‘verygood’ may be necessary or even essential to encourage the participation andperformance of some children, lower- and intermediate-level learners, those

in special education, or any student in need of more ‘stroking’ or building measures The point is, there is a very important affect dimensionthat ‘very good’ affords, and our challenge is to take advantage of the positiveaffect that ‘very good’ brings without suppressing learning opportunities

confidence-Conclusion Some teachers may find that they already use alternative positive feedback

techniques such as those mentioned above And indeed, teachers, inobserving and being mindful of their own teaching style, may think of otherways of providing positive feedback, ones that would work for theirparticular classroom contexts, which have to take into consideration timeand other classroom management issues as well Overall, we are notimplying that categorically there is no room for ‘praise’ or feedback tokenslike ‘very good’, but that in a larger context, positive feedback should bemeaningful and authentic, in tune with what a teacher hopes to accomplish

in his or her teaching goal(s) We, as (language) teachers, must examine indetailed ways what feedback tokens such as ‘very good’ possibly do inclassroom interaction from the perspective of promoting and encouragingstudents’ continued learning and growth (or not), and in the data that we

have displayed, this involves the kind of learning activity in which studentsfocus on form Examining, altering, or varying the ways in which teachersprovide positive feedback to learners is another dimension in scaffoldinginstruction (Vygotsky 1978) and providing guidance or ‘knowledge ofresults’ (Allwright op cit.: 167).

Revised version received June 2008

Notes

Some of the suggestions for this paper came from

audience participants at a discussion session

facilitated by Waring and Wong (2008) entitled,

‘Conversation analysis and giving feedback in the

language classroom’, presented at the Teachers of

English to Speakers of Other Languages (TESOL)

Annual Convention

References

Allwright, R.1980 ‘Turns, topics, and tasks: patterns

of participation in language learning and teaching’

in D Larsen-Freeman (ed.) Discourse Analysis in

Second Language Research Rowley, MA: Newbury

Brophy, J.1981 ‘Teacher praise: a functional

analysis’ Review of Educational Research

51/1: 5–32

Fanselow, J.1987 Breaking Rules: Generating and

Exploring Alternatives in Language Teaching New

York: Longman

Forgas, J P.and G Tehani 2005 ‘Affective

influences on language use: mood effects on

performance feedback by experts and novices’

Journal of Language and Social Psychology

24/3: 269–84

Gass, S.and A Mackey 2006 ‘Input, interaction

and output: an overview’ AILA Review 19: 3–17

Hellermann, J.2003 ‘The interactive work of

prosody in the IRF exchange: teacher repetition in

feedback moves’ Language in Society 32/1: 79–104

Kahn, G.2008 ‘The social unfolding of task,

discourse, and development in the second language

classroom’ Unpublished EdD dissertation, Teachers

College, Columbia University

Mehan, H.1979 Learning Lessons: Social

Organization in the Classroom Cambridge, MA:

Harvard University Press

Purpura, J.and D Pinkley 2000 On Target Workbook

1 White Plains, NY: Pearson Education

Schegloff, E A.2007 Sequence Organization

Cambridge: Cambridge University Press

Schegloff, E A.and H Sacks 1973 ‘Opening upclosings’ Semiotica 8/4: 289–327

Seedhouse, P.2004 The Interactional Architecture ofthe Language Classroom: A Conversation AnalysisPerspective Malden, MA: Blackwell Publishing, Inc.Sinclair, J M.and M Coulthard 1975 Towards anAnalysis of Discourse: The English Used by Teachers andPupils London: Oxford University Press

Vygotsky, L S.1978 Mind in Society: The Development

of Higher Psychological Processes Cambridge, MA:Harvard University Press

Waring, H Z.2008 ‘Using explicit positiveassessment (EPA) in the language classroom: IRF,feedback, and learning opportunities’ The ModernLanguage Journal 92/4: 577–94

The authorsJean Wongis an Associate Professor at The College

of New Jersey Her work has appeared in AppliedLinguistics, Research on Language and SocialInteraction, International Review of Applied Linguistics,Issues in Applied Linguistics, and in edited volumes(Gardner and Wagner 2004; Richards andSeedhouse 2005; Bowles and Seedhouse 2007) Herresearch inquiries include how to use conversationanalysis (CA) as a resource for understandinginteraction and advancing issues and concerns inlanguage pedagogy

Email: jwong@tcnj.eduHansun Zhang Waringis a lecturer in Linguistics andEducation at Teachers College, Columbia University,where she teaches Conversation Analysis andSpeaking Practicum, among other courses Herwork has appeared in Applied Linguistics, Journal ofPragmatics, Research on Language and SocialInteraction, Discourse Studies, Text and Talk, andJournal of Sociolinguistics She is currently interested

in using CA to examine instructional practices andtheir relevance to learning opportunities

Email: hz30@columbia.edu

The value of a focused approach to written corrective feedback

John Bitchener and Ute Knoch

Investigations into the most effective ways to provideE S Llearners with writtencorrective feedback have often been overly comprehensive in the range of errorcategories examined As a result, clear conclusions about the efficacy of suchfeedback have not been possible On the other hand, oral corrective feedbackstudies have produced clear, positive results from studies that have targetedparticular error categories This article presents the results of a study that examinedthe effectiveness of targeting only two functional error categories with writtencorrective feedback in order to see if such an approach was also helpful forE S L

writers The ten-month study was carried out with 52 low-intermediateE S L

students in Auckland, New Zealand Assigned to groups that received writtencorrective feedback or no written corrective feedback, the students produced fivepieces of writing (pre-test, immediate post-test, and three delayed post-tests) thatdescribed what was happening in a given picture Two functional uses of theEnglish article system (referential indefinite ‘a’ and referential definite ‘the’) weretargeted in the feedback The study found that those who received writtencorrective feedback on the two functions outperformed the control group on allfour post-tests

Introduction In 1996, Truscott declared that the provision of written corrective feedback

onE S Lstudent writing was ineffective and harmful and that it shouldtherefore be abandoned He maintained that there was empirical evidence(for example Semke 1984; Robb, Ross, and Shortreed 1986; Kepner 1991) toshow that the practice was not worth continuing Ferris (1999), in herresponse, pointed out, among a range of arguments, that the research base

he was drawing upon was too limited and conflicting in its findings and thatrestraint should be exercised while further investigations were undertaken

Of the studies that have been conducted until fairly recently, most, in terms

of their design, execution, and analysis, were flawed to some extent (seeGuenette 2007; Bitchener 2008 for a review of these issues) so this hasmeant that firm conclusions about the efficacy of written corrective feedbackare not yet available

Another reason for the failure of earlier work to produce conclusive answers

to the question of efficacy is the unfocused approach that was taken withregard to the range of error categories treated Up to 15 different linguisticerror categories were sometimes included in these studies so it was likely toproduce too much of a cognitive overload for learners to attend to By

204 E LT Journal Volume 63/3 July 2009; doi:10.1093/elt/ccn043

comparison, oral corrective feedback research (for example Doughty andVarela 1998; Lyster 2004), by focusing on a single error category, has beenable to reveal clear, positive effects for the focused approach.

Although positive findings have been reported in three recent writtencorrective feedback studies (Sheen 2007; Bitchener 2008; Bitchener andKnoch 2008) that were conducted over a two-month period, anothershortcoming of the existing research base has been its primary focus on textrevision Little attention has been given to investigations of the extent towhich written corrective feedback can facilitate accuracy improvement inthe writing of new texts

In order to address both of these issues, this article presents the findings of

a ten-month longitudinal investigation of the extent to which a targetedfocus on two functional error categories resulted in improved accuracy infour new pieces of writing

The study

Introduction

The study investigated the following research question: does accuracy in theuse of two functions of the English article system improve over a ten-monthperiod as a result of written corrective feedback?

Accuracy was measured over a ten-month period by means of a pre-test test design (a pre-test after one week; an immediate post-test following thecorrective feedback treatment after two weeks; three delayed post-tests aftertwo, six, and ten months)

post-Participants The study was conducted in the English Language Department of

a university in Auckland, New Zealand Students from four existing intermediate classes were invited to take part in the study Fifty-two studentswere available for the ten-month data collection period Students who werenew to the university were assigned to a proficiency level after taking

low-a stlow-andlow-ardized grlow-ammlow-ar test, low-a writing test, low-and low-a one-on-one interview.Students who had previously been studying at a lower proficiency level wereplaced in the low-intermediate level on the basis of earlier competency-based assessments The English Language Department describes itsapproach to the teaching of English as communicative and gives an equalfocus to reading, writing, speaking, and listening Most of the students weremigrants who had settled in New Zealand within 18 months of

commencing study at the low-intermediate level Four hours of instructionwere provided five days a week The students (19 males and 33 females) werepredominantly from East Asian countries: Korea (15 per cent), Japan (11 percent), and the People’s Republic of China (18 per cent) Other countriesrepresented were Vietnam, Yemen, Russia, Switzerland, Saudi Arabia,Chile, Brazil, Serbia, Turkey, Somalia, Romania, Iran, Sri Lanka, India, andIndonesia The average age of the students was 31.7 years The majority(78 per cent) claimed to have had formal instruction though their length ofearlier study varied across a seven-year period The four classes werearbitrarily assigned to one of three written corrective feedback groups(n ¼ 39) or the control group (n ¼ 13) that did not receive corrective feedback

Target structures Compared with earlier studies on the value of written corrective feedback

(see Ferris 2003), where sometimes as many as 15 linguistic forms and

structures had been examined, this study investigated the effect of targetingtwo functional uses of the English article system: the referential indefinitearticle ‘a’ for referring to something the first time (first mention) and thereferential definite article ‘the’ for referring to something alreadymentioned (subsequent mention) Other functional uses of the definite andindefinite articles were not targeted in the study.

These structures were targeted because students across English languageproficiency levels experience difficulty in the use of the English articlesystem (Butler 2002; Bitchener, Young, and Cameron 2005) For example,they may experience difficulty deciding whether an article is required and, if

it is required, whether it should be the definite or indefinite article So thatsecond language learners are not stigmatized as a result of incorrect usagewhen communicating with native speakers of English and so that doubts donot arise about which items they may be referring to, it is important thatcorrective feedback be provided on the use of articles when students revealrecurrent difficulties with correct usage The occasional error may notnecessarily impede the overall coherence and cohesion of a text but frequenterrors may well do so The extent to which written corrective feedback, asone form of input, can facilitate the acquisition process is investigated in thisstudy Accuracy in the use of these functions in the pre-test revealed a meanscore of 59.41 per cent, thereby indicating that students at a low-

intermediate level have only a partial mastery of the functions

Treatment Each of the three groups within the wider written corrective feedback group

received different combinations of written corrective feedback These arepresented in Table 1 below

Written corrective feedback group Group one—received direct error correction,

written, and oral meta-linguistic explanationGroup two—received direct error correctionand written meta-linguistic explanationGroup three—received direct errorcorrection

No written corrective feedbackgroup

Group four—received no correctivefeedback

nUse ‘a’ when referring to something for the first time

nUse ‘the’ when referring to something that hasalready been mentioned

e.g A man and a woman were sitting opposite

me The man was British but I think the womanwas Australian

Oral meta-linguisticexplanation

nThe 30-minute mini-lesson

nAbove rules and examples explained

nAdditional examples illustrated on whiteboardand discussed with class

nStudents completed five-minute controlled practiceexercise, filling gaps in each sentence with ‘a’, ‘the’,

or neither, and answers were then discussed.table2

Feedback procedures

Instruments Each of the five pieces of writing required a description of what was

happening in a given picture (settings at a beach, a picnic, a campsite, a familygathering, a sporting event) Picture descriptions were chosen because therange of people, objects, and activities illustrated had the potential to createobligatory opportunities for the use of both English article functions Thirtyminutes was given for the writing of each description

Procedure The procedures of the study were administered according to the timeline

Timetable for procedures

The sequence of activities for the immediate post-test varied as followsacross the groups:

Group oneThe immediate post-test was completed after the students had been givenfive minutes to consider the error corrections and the written meta-linguistic explanation and had received the 30-minute lesson (oral meta-linguistic explanation)

Group twoThe immediate post-test was completed after the students had been givenfive minutes to consider the error corrections and the written meta-linguistic explanation

Group threeThe immediate post-test was completed after the students had been givenfive minutes to consider the error corrections

Group fourThe immediate post-test was completed as soon as the uncorrected pre-testpiece of writing had been returned

Analysis The analysis of the data involved several steps:

1 Obligatory uses of the targeted features were identified

2 Written corrective feedback or no corrective feedback was provided (asdescribed above)

3 Accuracy was calculated as a percentage of correct usage For example, inany one script, three correct uses of the targeted features from tenobligatory occasions meant a 30 per cent accuracy rate.

4 Inter-rater reliability calculations with a trained research colleaguerevealed a 95 per cent agreement on the identification of targeted errorsand a 98 per cent agreement on the assignment of errors to the targetedcategories

5 Descriptive statistics for the pre-test and the four post-tests werecalculated separately for the written corrective feedback groups and the

no feedback group

6 Because no statistically significant differences on the pre-test scores werefound between the groups, a two-way repeated measures analysis ofvariance (AN O VA) was chosen to address the research question

Results Table 4 below shows the descriptive statistics for the treatment group and

the control group at the five different testing periods

Group Time 1 Time 2 Time 3 Time 4 Time 5

N M SD M SD M SD M SD M SD

1 CF 39 59.15 17.71 84.05 11.70 79.31 11.49 81.44 12.20 86.21 11.15

2 Control 13 63.23 17.51 67.08 21.45 56.62 22.29 62.46 18.97 58.92 16.16

table4

Descriptive statistics for

mean test scores by

group and testing period

SD ¼ standard deviation CF ¼ corrective feedback

Figure 1 provides a visual representation of the mean percentages for the fivetesting periods for each group Table 4 and Figure 1 illustrate that whilst bothgroups scored around 60 per cent on the pre-test, only participants in thetreatment group were able to increase their accuracy after the pre-test andkeep that gain in accuracy over the following testing periods

To determine the most appropriate data analysis technique, an independentt-test was conducted to determine if there were any statistically significantdifferences between the two groups on the pre-test As this was not the case,

t (50) ¼ –.721, p ¼ 474, a two-way repeated measuresA N O VAwas chosen toanalyse the data (Table 5 below)

figure 1

Mean percentage

accuracy for treatment

group and control group

over time

Mean percentages over time

50 55 60 65 70 75 80 85 90

Time 1 Time 2 Time 3 Time 4 Time 5 CF

CF = corrective feedback Control

Source df F pBetween subjects

Group 1 1293.29 ,.001Within subjects

To investigate if the accuracy of the two groups differed on any of the tests, independent samples t-tests were conducted All these tests weresignificant as can be seen in Table 6 below, indicating that the treatmentgroup consistently outperformed the control group

post-Post-test t df pTime 2—after 2 weeks 3.618 50 001Time 3—after 2 months 4.783 50 ,.001Time 4—after 6 months 4.196 50 ,.001Time 5—after 10 months 6.796 50 ,.001

table6

Independent samples

t-tests for post-tests

Discussion Students who received written corrective feedback outperformed those who

received no feedback in all four post-tests even though all groups developeddifferently over time This means that the provision of written correctivefeedback on a single occasion had a significant effect, enabling the learners

to use the targeted functions with greater accuracy over the ten-monthperiod These results corroborate and extend those of three recent studies(Sheen op cit.; Bitchener op cit.; Bitchener and Knoch op cit.) thatexamined the effect of written corrective feedback on new pieces of writingover a two-month period The enduring effect on accuracy over a ten-monthperiod is clear evidence of the potential for focused written correctivefeedback to help learners acquire features of a second language Whereasearlier research has focused on a comprehensive range of error categories,the results of this study demonstrate the value of focusing on a single errorcategory In this respect, they corroborate the findings not only of the writtencorrective feedback studies referred to above but also of those on oralcorrective feedback (for example Doughty and Varela op cit.; Muranoi 2000)

Excerpts 1 and 2 below, from the pre-test and immediate post-test texts ofone student, illustrate the accuracy with which the two article functionswere used in the second text after written corrective feedback had beenprovided.

Excerpt 1 (pre-test

text)

At the kiosk, a woman buys two ice-creams A woman gives the ice-cream

to her child and keeps ice-cream for herself

Excerpt 3 (delayed

post-test 3)

Lots of children are playing in the room A baby is playing with blocks and

an old man is sleeping on a sofa The baby is putting the blocks beside thesofa while the old man is sleeping

While these excerpts illustrate the accuracy gains that can be made whenwritten corrective feedback is focused, further research is required todetermine the extent to which it is effective with other error categories inother linguistic domains It is especially important that it be tested withmore complex features to determine whether or not its optimal effect is withsingle rule-based functions such as those examined in this study

Conclusion Based on the findings of the study, a number of pedagogical

recommendations can be offered Teachers should feel confident aboutproviding direct written corrective feedback on their students’ linguisticerrors, providing it is based to the best of their knowledge on theirstudents’ ‘readiness’, both in terms of their proficiency level and theirunderstanding of the merits of focusing their attention on writtenaccuracy We believe that student motivation is more likely to be gained ifteachers negotiate with their students about which features they will focus

on, about how frequent the feedback will be given, about the type offeedback that will be given, and about what the students will be expected to

do in response to the feedback The study has also shown that a singlefeedback session can be effective in developing accuracy in the use of tworule-based features but if teachers are able to provide additional feedback

on more occasions, it may be possible to increase the accuracy rate andalso reduce the amount of time that is required to help learners achieve

a high level of mastery over recurrent errors Finally, and mostimportantly, we believe that there is clear evidence from the study torecommend that teachers provide selective, focused feedback on one ortwo linguistic error categories at a time rather than feedback on toocomprehensive a range of features Although it might be argued that thisapproach hinders good language learners from making more rapidprogress in acquiring features that have been problematic if they arerequired to proceed in a lockstep manner with the class as a whole, we

would suggest that once there are signs of accuracy development, teachersrenegotiate an additional feedback focus with such students In order toconfirm the advantage of this approach over the more comprehensiveapproach, future research is required to compare the two approacheswithin a single research design.

Final revised version received May 2008

References

Bitchener, J.2008 ‘Evidence in support of written

corrective feedback’ Journal of Second Language

Writing 17/2: 102–18

Bitchener, J.and U Knoch 2008 ‘The value of

written corrective feedback for migrant and

international students’ Language Teaching Research

Journal 12/3: 409–31

Bitchener, J., S Young,and D Cameron 2005 ‘The

effect of different types of corrective feedback onE S L

student writing’ Journal of Second Language Writing

9: 227–58

Butler, Y.2002 ‘Second language learners’ theories

on the use of English articles’ Studies in Second

Language Acquisition 24/3: 451–80

Doughty, C.and E Varela 1998 ‘Communicative

focus on form’ in C Doughty and J Williams (eds.)

Focus on Form in Classroom Second Language

Acquisition New York: Cambridge University Press

Ferris, D R.1999 ‘The case for grammar correction

in L2 writing classes A response to Truscott (1996)’

Journal of Second Language Writing 8/1: 1–10

Ferris, D R.2003 Response to Student Writing:

Implications for Second Language Students Mahwah,

NJ: Lawrence Erlbaum Associates

Guenette, D.2007 ‘Is feedback pedagogically

correct? Research design issues in studies of

feedback on writing’ Journal of Second Language

Writing 16/1: 40–53

Kepner, C G.1991 ‘An experiment in the

relationship of types of written feedback to the

development of second-language writing skills’

Modern Language Journal 75/3: 305–13

Lyster, R.2004 ‘Differential effects of prompts

and recasts in form-focused instruction’

Studies in Second Language Acquisition

26/3: 399–432

Muranoi, H.2000 ‘Focus on form through

interaction enhancement: integrating formal

instruction into a communicative task in

50/4: 617–73

Robb, T., S Ross,and I Shortreed 1986 ‘Salience offeedback on error and its effect onE F Lwritingquality’.T E S O LQuarterly 20/1: 83–93

Semke, H.1984 ‘The effects of the red pen’ ForeignLanguage Annals 17/3: 195–202

Sheen, Y.2007 ‘The effect of focused writtencorrective feedback and language aptitude onES L

learners’ acquisition of articles’.T E S O LQuarterly41/2: 255–83

The authorsJohn Bitcheneris an Associate Professor in theSchool of Languages and Social Sciences atAUT

University, Auckland, New Zealand He teaches onthe MA in Applied Language Studies programmeand supervises a wide range of Masters and Doctoralthesis students His research interests includewritten and oral corrective feedback and thediscourse patterning of academic genre He isPresident of the Applied Linguistics Association ofNew Zealand and Co-Editor of New Zealand Studies

in Applied Linguistics

Email: john.bitchener@aut.ac.nzUte Knochis a research fellow at the LanguageTesting Research Centre at the University ofMelbourne She completed her PhD at theUniversity of Auckland focusing on diagnosticwriting assessment Her research interests are in theareas of language assessment (particularly theassessment of writing, rating scales, rater training,and the assessment of languages for specificpurposes) as well as language pedagogy andlanguage and immigration She is a recipient of

a Spaan Fellowship in Second and Foreign LanguageAssessment from the University of Michigan in

2006 and 2008

Email: uknoch@unimelb.edu.au

Traversing the lexical cohesion minefield

Iain McGee

When teachers hear the word ‘cohesion’, they usually think of grammaticalcohesion—an aspect of cohesion reasonably well covered in student books andteacher materials However, occupying an area that straddles both lexis ‘proper’and cohesion lies ‘lexical cohesion’ In what follows, it is argued that the teachingand learning of certain aspects of lexical cohesion is problematic, and that thisstate of affairs may be behind the current neglect of this subject inEFLmaterialsand classrooms The paper begins with a brief overview of Halliday and Hasan’s(1976) classification of lexical cohesion, and then looks, in turn, at four types ofcohesive device Learners’ uses of these different cohesive ties are discussed, theobstacles to correct usage are noted, and suggestions are made as to how teacherscan help students to develop this aspect of their writing

Definition and

overview

While the terms ‘cohesion’ and ‘coherence’ tend to crop up together in theliterature, the relationship between the two is a contested one: for example,Carrell (1982: 486) argues that coherence leads to cohesion, whereasHalliday and Hasan (1976: 2) suggest that cohesion brings about coherence.One thing that all writers would agree on, however, is that the use of lexicalcohesive ties does not, necessarily, make a text more coherent, or ‘better’than another As Connor (1984: 308, 311) points out, a text lacking in lexicalcohesive ties may be better organized, or the points may have better supportthan a text with more lexical ties Having made this important

qualification—putting lexical cohesion in its place—we can now look inmore detail at the subject

In Halliday and Hasan’s (1976: 4) influential work Cohesion in English, theauthors explain that cohesion is a semantic concept, referring to meaningrelations in text.1They divide cohesion into two broad areas: grammaticalcohesion and lexical cohesion The former includes reference (for examplethree blind mice they), substitution (for example My axe is too blunt I mustget a sharper one.), ellipsis (for example Which hat will you wear? This is thebest.), and conjunction (for example use of the words but, yet, so, etc.) Thebulk of Halliday and Hasan’s book concerns itself with discussing thesetypes of cohesive ties, and books aimed at developing academic reading andwriting skills have given considerable attention to reference and

conjunction and their roles in helping texts hang together.2Even thoughlexical cohesion is the more pervasive in creating textual cohesion, it is

212 E LT Journal Volume 63/3 July 2009; doi:10.1093/elt/ccn040

neglected inELT(as noted, for example by Flowerdew 2006: 209).Halliday and Hasan classify lexical cohesion in the following way:

1 Reiteration

a same word repetition

b synonym or near synonym

c superordinate

d general word

2 CollocationRegarding the first of these two classes—reiteration—Halliday and Hasanprovide the following examples of how reiterative inter-sentential cohesiveties can be made (Figure 1)

Regarding the second means of achieving lexical cohesion (collocation),Halliday and Hasan (op.cit.: 284) define this as ‘cohesion that is achievedthrough the association of lexical items that regularly co-occur’ AsTanskanen (2006: 12) notes, collocation is not always considered to be a type

of lexical cohesion (cf McCarthy 1991: 65), and it will not be considered asplaying a role in creating lexical cohesion in this paper However, collocationknowledge will be referred to as a specific type of knowledge whichstudents need to have to enable them to use reiterative lexical cohesivedevices correctly

In the sections that follow, I look in more detail at the four reiteration devicesnoted above in the context of developingE F Llearners’ writing skills

Repetition While repetition is a standard way of achieving lexical cohesion, particularly

in science texts, the fact remains that there can be a lot of, what Ting (2003:6) calls, ‘redundant repetition’ in students’ writing From a marking point ofview, it may be that teachers are hesitant to draw attention to redundantrepetition in student writing: repeating a word does not impede

understanding, and neither is overuse necessarily misuse However, theeffect of such repetition on the reader can be quite negative To illustrate thispoint, I reproduce below a paragraph written by a Saudi intermediate user

of English, studying at a Saudi university, answering a question aboutstudent preparation for exams The text has not been edited

a seccesful study

figure 1

From Halliday and

Hasan (1976: 279)

The student’s use of break/breaks five times in the above extract certainlydoes give the text a rather tedious and monotonous feel to it, detracting fromthe otherwise well organized, though simply written paragraph So, why is itthat students ‘overdo’ repetition in their writing?

Firstly, it has been argued that the L1 may play a role in L2 repetition(McCarthy op.cit.: 67, 68) Typical text structure patterns and stylepreferences in the L1 may well transfer to L2 usage, and Mohamed andOmer (2000) have noted the prevalence of repetition as an Arabic cohesivedevice in text

Secondly, it may be that students do not appreciate the importance of variety

in academic writing Repetition is particularly common in speech, andShaw and Liu (1998) have noted the tendency ofEFLlearners to write in

a spoken register Clearly, students need to be made aware about academicwriting norms in English, which eschew, to a large extent, repetition

An additional reason may be found in the texts thatE F Lstudents typicallyread Edited and simplified texts are not rich in their lexical cohesive ties andCox, Shanahan, and Sulzby (1990: 60) argue that exposure to ‘contrived’texts has a negative effect on the development of a learner’s use of lexicalcohesive ties in writing Ironically, while texts are simplified to helpstudents, such texts are, implicitly at least, also training the readers in ‘sub-standard’ lexical cohesive device usage

Fourthly, and finally, it may be that a student only knows of one way ofreferring to a concept This is a typical explanation for the lack of variety instudent writing (see for example Ting op.cit.: 6); however, it may well be thatthis factor plays a more significant role in elementary students’ writing than

in intermediate/advanced students’, as argued below

The student who wrote Text 1, above, was given some grammatical feedback

on his text, and he then attended a class in which the writer of this papertalked about and exemplified the importance of the use of synonyms to addlexical variety to academic writing The students were asked to rewrite theirfirst draft paragraphs, paying particular attention to the avoidance ofrepetition and the use of synonyms The student’s second draft isreproduced below Grammar and spelling mistakes are retained

Text 2

In order for a studant to have good and healthy studing is having healthybreaks Breaks are very benefical even though they are short A studantought to have a five minutes break at least every one hour He can spend[it] his leisure time waching Television, eating, drinking, relaxing or eventaking a bath to destress Why should you have a time out? Simplybecause the mind’s efficiency goes down after constant study To sum up,having continuous rests is an important method for secessful study.Even though the student has the same break ‘cluster’ towards the beginning

of the paragraph, it is noteworthy that he attempted to reduce its use Hemakes three changes in this regard Firstly, he substitutes ‘it’, present in thefirst draft, and included in square brackets in the above (originally referring

to ‘break’) with ‘his leisure time’.3Further on he substitutes ‘break’ directlyfor ‘time out’ and ‘continuous rests’ It should be stressed that the student

rewrote the second draft within a day or two of producing the first text: thereal problem, therefore, seems to have been one of awareness (the secondpoint noted above).

The next point to consider with regards to repetition is complexrepetition—the use of a derivational form of a word, rather than an exactrepetition—in a text to effect lexical cohesion Examples in the student textabove would be ‘studant’ (used twice), ‘studying’ used once (incorrectly),and ‘study’ (twice) as a noun While word tables showing derivational formsare still fairly standard items found inE F Lbooks, they are typically used inexercises to highlight the grammar of the language, rather than

highlighting how different word forms may be used as cohesive ties in a text.However, tables of this kind can easily be adapted to such usage It isimportant to include information about the frequent or strong collocates ofdifferent word forms because if this information is not provided, it is quitelikely that student attempts to use certain word forms may produceunidiomatic combinations The two examples below, from the same class ofintermediate students referred to earlier, illustrate this problem

Example 1: They need to catch some rest while they are reviewing Example 2: it also helps if he arranges a schedule so that he could getsome breaks in between studying

The collocations ‘catch some rest’ and ‘get some breaks’ are both untypical innative speaker usage According to the Oxford Collocations Dictionary(2002), ‘get’, ‘have’, and ‘take’ are typical collocates of ‘rest’, and ‘have’ and

‘take’ typically collocate with ‘break’ Not only, therefore, do students need toknow different word forms to engage in complex repetition, they also need

to know the typical collocates of the different word forms, and this is a heavylearning burden

Synonyms Inkpen and Hirst (2006: 224, 225) note three types of differences between

synonyms/near synonyms: denotational differences (i.e where there is

a difference in meaning, for example ‘lie’ is deliberate, ‘misrepresent’indirect), attitudinal differences (for example ‘thin’ is neutral, ‘skinny’pejorative), and stylistic differences (for example in formality: ‘cops’ and

‘police’) True synonyms are few and far between, and research, particularly

in corpus linguistics, has helped us discover the different distributions of

‘apparent’ synonyms in different genres, the semantic prosodies that thesewords have (for example ‘bring about’ tends to be used in positive contexts,

‘cause’ when the consequence is negative), and the different collocationpatterns in which synonyms occur Given such a state of affairs, simplyencouraging students to use synonyms for key words in their writing, ratherthan repeating them, is, in effect, an invitation to commit semantic suicide

We would not usually expect our students to be sensitive to the above notedpoints, and yet such knowledge is required to use synonyms successfully.The following examples, from the same class of Saudi intermediate levelstudents noted above, are from essays looking at what society can do toreduce road traffic deaths Before the students embarked on this task, it wasstressed that they should try to use synonyms, rather than repeat key words

in their writing At that stage in the course, they were not warned about thepotential dangers in trying to use synonyms as cohesive ties.

Example 3: In the last year, my school made that day to introduce us thelaws and to respect the regulations

Example 4: accidents To decrease the rate of bad events

Example 5: When you become an adult or have a permission to drive youshould behave as big guy with good thinking

Example 6: (In talking about people who die in car accidents)

A huge number of souls Many innocent human Example 3 is a successful use of synonyms: ‘laws’ (of the road) and (traffic)

‘regulations’ work well as synonyms within this sentence Turning to thefourth example, while ‘accidents’ are ‘bad events’ we can recognize that thestudent’s attempt at synonymy is not idiomatic The fifth example is a goodillustration of stylistic insensitivity While women cannot drive in SaudiArabia, i.e adults in the Saudi driving context are all male, ‘big guy’ would bemuch better substituted with ‘grown up’—‘big guy’ is too informal for thistype of writing With regards to the sixth example, the use of ‘souls’ in thiscontext is untypical in current English language usage A human is notusually referred to as a soul, except within a religious context (for example

‘Have you ever thought about your soul?’) When it is used in other contexts,its typical collocates (for example ‘brave’, ‘hardy’) give the word a ‘tough’aura which is not appropriate for a victim, as in the context of the essay Inaddition, while we could speak of ‘innocent humans’, the more typicalcollocations of ‘innocent’ are ‘people’ and ‘victims’

The above commentary is by no means meant to belittle the student efforts:they are trying, as encouraged by their teacher, to use different vocabulary,rather than simply repeating words or using more ‘run of the mill’ frequentvocabulary items However, the task is not an easy one It may be thatteachers and teacher materials are partially to blame in at least two areashere: the use of synonym lists for example, may give students the falseimpression that certain words are (always) interchangeable In addition,simplistic answers to student vocabulary questions can easily suppresssynonym sensitivity appreciation, rather than enhance it

Superordinates Other than their use in definitions, superordinates (i.e words which

‘contain’ other words, for example ‘vehicle’ is a superordinate of ‘car’) receivevery little attention in the classroom While many students have heard of theword ‘synonym’, ‘superordinates’ (also called hypernyms) and

‘subordinates’ (hyponyms) are not words typically heard in theEFLwritingclassroom

It is usually the case that the more specific word is used first in a sentence ortext, and then superordinates are used later on, as they contain lessinformation This being so, as Salkie (1995) points out, subordinates andsuperordinates cannot be simply switched round in a text For example, inthe text below, Example 7, where ‘Brazil’ and ‘country’ have been switchedaround from the original text, Brazil seems to refer to a different country tothat referred to in the opening words

Example 7: The country, with her two-crop economy, was even moreseverely hit by the Depression than other Latin American states andBrazil was on the verge of complete collapse.

(Salkie 1995: 16)One useful teacher resource available from the internet is WordNet 2.1(available from http://wordnet.princeton.edu) which is a lexical databasedeveloped at Princeton University This program provides a wealth ofinformation about words and their typical hyponyms and hypernyms

To encourage the same class of intermediate students referred to earlier touse superordinates in writing a paragraph about the Titanic disaster, thefollowing figure (Figure 2) was provided The students were given nofurther help about how to use the words, just encouraged to use them

Below I provide two examples of the students’ writing The lexical chainusage is noted below the student text, followed by a short commentary Thetexts have not been edited

Example 8: Titanic was one of disastrous stories It was a nice and mostbeautiful vessel in its century It travelled to America but when it was in itsvoyage, a dire event was waited the ship Actually, the ocean liner struck by

a mountain of ice and then it sunk

ChainsTitanic–vessel–ship–ocean linerTravel–voyage

Event–struck

It is interesting to note here how the student uses ‘travel’ before ‘voyage’where ‘travel’, as a verb, is a complex superordinate of ‘voyage’ (noun), i.e.the more general word is used before the more specific The student followsthe same pattern (i.e more general word used before the more specific) inhis use of ‘event’ before ‘struck’ As such, these uses, while not wrong, arenot so typical, and it may be that this particular student would have benefitedfrom some more input on typical superordinate usage, and the usualpattern in ordering

Example 9: In 1912, one of man made disaster, which was the great vessal

in that time, called titanic titanic was a voyage from Southampton inEngland to New York City.the ship had prominent people from American,British and European families titanic began the travel and it impact with

an iceberg That event was responsible for sinking ocean liner

figure 2

General and specific

vocabulary

Chains(the great) Vessel–titanic–titanic–the ship–titanic–ocean liner(a) Voyage–(the) travel

(It) Impact–(that) event

Of particular note here is the omission of ‘the’ before ‘ocean liner’ in the lastsentence, which, if interpreted uncharitably, could be seen as referring to

a different ship, i.e Titanic’s impact with an iceberg led to the sinking ofan(other) ocean liner Students must be made aware that the omission of thedefinite article (or a determiner) before the superordinate can lead to

‘unfortunate’ ambiguity

Overall, the student attempts to use the hypernyms/hyponyms were quitesuccessful, despite the limited input from the teacher It would seem thatsuperordinate usage is less of a minefield than synonym usage: the Titanic

is always an ocean liner and an ocean liner is always a ship; however, ‘slim’ isnot always ‘thin’ or ‘skinny’ (as noted above in the Synonyms section)

General words The ‘general word’ class of Halliday and Hasan overlaps, to a certain extent,

with more recent research on nouns, for example Flowerdew’s (op.cit.)signalling nouns These nouns (for example ‘achievement’, ‘problem’,

‘situation’) can be used in a number of ways and they are a useful way forstudents to refer back to a particular event/state of affairs referred to earlier

on (i.e anaphorically) in their writing I have exposed more advanced levelstudents to the use of such words in corpus data, and analysed their abilities

to use these nouns In one such exercise, the students were asked to use

a number of general nouns anaphorically in a third draft of a report, and tohighlight their presence by underlining them to enable me to check on theirusage The students’ use of the nouns was, by and large, successful.However, one of the nouns in particular (‘situation’) was quite regularlymisused Two student extracts are provided below

Example 10: Mass transit is very uncommon in Saudi Arabia; only privateautomobiles cruise the streets, most of which are air-conditioned All ofthese situations are dependant on burning of fuel which produces morecarbon dioxide lead

In this example, ‘All of these situations’ refers back to the lack of masstransit, the use of private cars, and the use of air conditioning This phrase isnot a particularly common one in standard English: it is interesting that thestudent felt that he had to use the plural (‘situations’) to refer to thepreceding information An additional concern in this text is that ‘situations’does not seem to be the best noun to use: ‘uses’ or ‘means of transport’would seem to be more appropriate

Example 11: (Referring to reports about the confiscation of certainpassenger items by Saudi Arabia airlines staff)

Because of that, many items were taken away by Saudi Airlines andpeople change their thought and prefer not to go toK S A These unhelpfulsituations happens in a bad time, especially whenS CTwanted to increasethe number of foreign tourists

The collocation ‘situations happens’ is not typical in native speakercorpora—more frequent collocates of ‘situation’ would be ‘arise’ and ‘occur’,and the plural ‘situations’, as already noted, is not very common As with theprevious example, the choice of the word ‘situation’ seems

inappropriate—‘developments’ would seem to fit better in the text Finally, itshould be noted that the student has used an inappropriate colligate, i.e.grammatical partner, in the phrase—‘in’ rather than ‘at’ In his study of

a learner corpus, Flowerdew (op.cit.) found colligation errors to be byfar the most common mistake made by learners, in their use ofsignalling nouns This being so, it is probably worthwhile introducing thesegeneral nouns to students in some typical phrases, for example the noun

‘situation’ could be provided in the frame ‘this situation has arisen at

a time’

The next step Having noted the above issues, there follow a number of brief suggestions

for teachers to consider when thinking about the teaching of lexicalcohesion to their students in their reading and writing classes

1 Raise awareness of the role of lexical reiterative devices in creatingtextual cohesion Texts can be analysed for different reiterative devicesand comparisons made between published texts and students’ writing.Salkie’s (op.cit.) workbook contains some exercises specifically aimed

at encouraging students to identify reiterative cohesion in texts.Though a useful resource, it should be noted that this is not a bookaimed atE F Llearners, and the texts used contain some rather difficultvocabulary

2 Warn students against adopting a simplistic attitude towards the use ofsynonyms Altered texts can be given to students asking them to identifyinappropriate uses of near synonyms, and students can also bechallenged to choose from a variety of options which word (from a list of

‘synonyms’) is missing from a stretch of discourse Such exercises willhighlight the point that synonymy is a slippery concept

3 Give students practice in using hyponyms and hypernyms of key words

in their writing Students are sometimes asked by teachers to use certainspecific words in their writing, and it is not too difficult to develop thiskind of exercise to work on this specific skill Wordnet 2.1, as notedearlier, is a useful resource for teachers to refer to

4 Be aware of the problem of collocation As noted above, collocationerrors are pervasive in student attempts to vary their lexis As much aspossible collocation knowledge must be developed alongside reiterationskill development Collocation dictionaries or corpus data can be used byteachers to help give students the most typical or strongest collocates ofimportant words

5 Increase student awareness of redundant repetition in their writing.Highlighting overuse may well provide the required stimulus forstudents to begin thinking about lexical ties in their writing It isimportant to encourage effort here, otherwise students may well justrevert to ‘default’ repetition in the face of difficulties

Conclusion The use of lexical cohesive ties has been found to be a significant

differentiating factor between native speaker and non-native speakerwriting (Connor op.cit.: 307), and while it is tempting to postpone a focus ongood writing style in the classroom to advanced level classes, this is probablynot the best course of action to take While lexical cohesion is a complex area,and fraught with difficulty, it has been suggested in this paper that there arecertain exercise types and awareness-raising activities that can make thesubject a rewarding one for students and teachers to explore together inclass: there are ways to traverse the minefield of lexical cohesion—indeed,some quite interesting ways

Final version received April 2008

Notes

1 The writer is aware of further work by these

authors in which they refine this taxonomy, for

example Hasan 1984, 1985, and Halliday 1985

However, the basic types of lexical reiteration,

while renamed or reclassified, remain

2 As Connor (1984: 302) points out, substitution

and ellipsis are not as common in written

discourse as in conversation

3 It should be noted here that ‘leisure

time’—typically referring to a long period of

time—is not a particularly appropriate synonym

for ‘break’ (a short period of time) in the text

References

Carrell, P L.1982 ‘Cohesion is not coherence’

Connor, U.1984 ‘A study of cohesion and coherence

in English as a second language students’ writing’

Papers in Linguistics 17/1–4: 301–16

Cox, B E., T Shanahan,and E Sulzby 1990 ‘Good

and poor elementary readers’ use of cohesion in

writing’ Reading Research Quarterly 25/1: 47–65

Flowerdew, J.2006 ‘Use of signalling nouns in

a learner corpus’ Lexical Cohesion and Corpus

Linguistics (Special issue) International Journal of

Corpus Linguistics 11/3: 227–47

Halliday, M A K.1985 An Introduction to Functional

Grammar London: Edward Arnold

Halliday, M A K.and R Hasan 1976 Cohesion in

English Harlow: Longman

Hasan, R.1984 ‘Coherence and cohesive harmony’

in J Flood (ed.) Understanding Reading

Comprehension Newark, DE: International

Reading Association

Hasan, R.1985 ‘The structure of a text’ in

M A K Halliday and R Hasan (eds.) Language,

Context and Aspects of Language in a Social Semiotic

Perspective Geelong, Victoria: DeakinUniversity Press

Inkpen, D.and G Hirst 2006 ‘Building and using

a lexical knowledge base of near-synonymdifferences’ Computational Linguistics 32/2: 223–62.Lea, D.(ed.) 2002 Oxford Collocations Dictionary forStudents of English Oxford: Oxford University Press.McCarthy, M.1991 Discourse Analysis for LanguageTeachers Cambridge: Cambridge University Press.Mohamed, A H.and M R Omer 2000 ‘Texture andculture: cohesion as a marker of rhetorical

organisation in Arabic and English narrative texts’

Tanskanen, S.-K.2006 Collaborating TowardsCoherence: Lexical Cohesion in English Discourse.Philadelphia, PA: John Benjamins PublishingCompany

Ting, F.2003 ‘An investigation of cohesive errors inthe writing ofP RCtertiaryE F Lstudents’ Available

at http://www.stets.org.sg/

vol2N2_2003FengTing.P D F Last accessed 12February 2008

The authorIain McGee’s research interests include investigatingdifferences between teachers’ intuitions aboutlanguage and corpus data, and the teaching ofcollocation He taught in Kuwait and King FahdUniversity of Petroleum and Minerals, Saudi Arabia,before moving on to Taibah University, Madinah,Saudi Arabia

Email: idmsjm96@muchomail.com

Critical reflection in a T E S L course: mapping conceptual change

Thomas S C Farrell

How can teacher educators gauge what participants have learnt after taking

a course in teaching English as a second language (T E S L)? One method that canhelp both language teacher educators and their students trace conceptual changes

as a result of taking a course is the use of concept maps This paper examines theconceptual changes of a group of MA participants in Canada as a result of taking

aT E S Lcourse Pre-course and post-course concept maps were elicited from theparticipants who were also asked to write short descriptions of changes (and thereasons for these changes) they observed between their pre- and post-course maps.Participants were also interviewed about the contents of their individual conceptmap and their perceptions of the course Results indicate that the course had someimpact on the participants’ prior beliefs and that a concept map may be a usefultool for tracing conceptual change

Introduction Participants come to teacher education courses with prior experiences,

values, and beliefs and with specific expectations about the subject matterthey will learn These beliefs have been accumulated from a variety ofsources including their past experiences as students in the school systemand may act as filters to what they have been exposed to in the teachereducation programme (Lortie 1975) Hence, differences are likely to existbetween what teacher educators may think is important for the participants

to learn and what they actually learn as a result of taking a course Bearingthis in mind, it is crucial then for educators to be able to establish a reliablemeans of gauging the effectiveness of their courses One method available

to language teacher educators interested in tracing participants’ conceptualchanges, or any changes in participants’ preconceptions or initial intuitiveideas as a result of taking a course, is the use of concept mapping Conceptmaps are diagrams that show relationships and understandings amongconcepts within a specific topic (Novak 1990) This paper outlines a small-scale study of how a language teacher educator used concept mapping totrace the conceptual changes of seven participants (the total number ofstudents who took the course) as a result of taking a course in teachingEnglish as a second language (T ES L) in a Canadian University

E LT Journal Volume 63/3 July 2009; doi:10.1093/elt/ccn058 221

1987) In language education, many of these preconceptions about teachingand learning are usually influenced by the participants’ previous schooling

in that they spend many hours subconsciously observing their teacherswhile at the same time developing tacitly held images about teaching andlearning (Richards 1998) The result of this type of ‘apprenticeship ofobservation’ (Lortie op.cit.; Borg 2004) and other prior experiences is thatparticipants enter the teacher education programme already possessing

a vast array of tacitly held prior beliefs and assumptions about teaching andlearning which can, as Richards (1998: 71) has highlighted, ‘often serve as

a lens through which they view’ the content of such programmes Theproblem is that if participants find that any of the content they are presented

in the language education courses is in conflict with their prior beliefs, thenrather than restructure their beliefs, many may only fine-tune them a little(Richards, Ho, and Giblin 1996) In addition, as Burns (1993) has indicated,all this happened at the subconscious level and as such, teacher educationprogrammes must provide opportunities for their students to be able to

‘raise to consciousness the nature of the personalized theories which informtheir practice’ (ibid.: 63–4) so that they can become aware of any

inconsistencies between their prior beliefs and concepts they are presentedwith in these courses

Methodology Raising awareness of participants’ prior beliefs and gauging the impact

of a course on these beliefs are as much a methodological issue as

a substantive one From a methodological perspective, one means thatcan help raise awareness of prior beliefs while at the same time gaugeconceptual change as a result of taking a course is the use of conceptmapping Concept maps are ‘a visual representation of knowledge’(Antonacci 1991: 174) and show relationships between concepts in a type

of network system and are useful visual indications of what people knowabout a topic

Participants and

course

The seven participants were enrolled in a one-year programme, the MA inApplied Linguistics/T E S L, at a university in Canada Each participant wasassigned the capital letter ‘T’ and a random number (from 1 to 7) behind theletter ‘T’ so that identities remain hidden T1 was a female Canadian withover ten years teaching experience and a certificate inT E S Las was T7,

a female Canadian with similar teaching experiences; T2 was a female fromKorea with five years teaching experience as was T5, a female from Koreawho also had about five years teaching experience; T3 was a female fromChina and had no full-time teaching experience, but had a certificate in

T E S Las was T4, a female from China, who also had a certificate inT E S L; T6was a male Canadian with about five years teaching experience and

a certificate inT E S L So, not only had they all prior student experiences butalso many had prior beliefs based on certificate programmes they had taken

as well as prior experiences as teachers (five of the seven teachers) Thecourse they were taking emphasized the following curriculum: CurrentIssues in Applied Linguistics andT E S L, Issues in Language Learning(Second Language Acquisition), Issues in Language Teaching

(Methodology), Issues in Language Curriculum Development, Issues inCritical Pedagogy, and Issues in Professional Development

Data collection Following the work of Fischer, Bruhn, Grasel, and Mandl (2002) who

maintain that in order to ensure greater ownership of the learning process,students should construct their own concept maps, rather than have a pre-course map prepared by the course instructor, the seven participantswere each asked on the first day of class to construct a concept mapconcerning the topic (‘What does teaching English as a second language(TES L) mean to you?’) placed inside a circle with several nodes, or spokes,emanating straight from that circle like a bicycle wheel These pre-coursemaps would be used for diagnostic purposes by the instructor to gauge theextent of the participants’ prior knowledge and beliefs On completion ofthe maps, the participants were asked to share their answers during

a peer group discussion and reflection session At the start of the followingclass, and in order to clarify what they placed in their concept maps,

a class group discussion was conducted where the participants wereencouraged to further explore their prior experiences and beliefs about

T E S L

On the final day of class, the participants were again asked to constructconcept maps on the same topic and following the same written andexplained directions as on the first day of class When the participants hadcompleted their post-course maps, they were given their pre-course mapsfor comparison and asked to write comments about any changes theynoticed between the two maps and the reasons for these changes Eachparticipant was also interviewed in order to discuss and reflect on thechanges that had occurred in these maps and any further perceptions theyhad of the course they had just completed In addition, I also attempted tocontact each participant two years after they had taken the course to assess ifthey still held the same post-course beliefs aboutT E S L(I was only able tocontact two of the original seven)

Data analysis In order to analyse the data, a keyword method was applied to the database of

categories that were developed For example, as a result of taking the course,one type of category that emerged on the concept maps indicated that many ofthe participants interpreted course concepts in terms of critical reflection.Keywords from this category included ‘beliefs’, ‘teacher personality’, and ‘self-awareness’ among others Another interpretation was research and theoryand keywords from this category included ‘theory acquisition’, ‘research’,

‘corrective feedback’, and ‘link theory to practice’, among others Eachparticipant’s map was analysed as follows: each keyword was numbered and

a frequency count was noted along with any connections made to otherconcepts The number of keywords from the seven individual maps wastotalled and a pre-course group concept map was constructed after the firstclass and a post-course group map was constructed after the last class inorder to provide a visual of what the participants as a whole said they believedaboutT ESL In order to check for reliability of my coding of categories,

I trained two other coders and we had an intercoder reliability of about

(3), and another category which included many diverse items Teachingtheory/methods was present in six maps without any further explanation Inthe class discussion that followed the first class, they said that they thoughtthe MA programme would give them many teaching methods and that waswhatT ESLwas for them The next concept, Language learning/acquisition,was present in five maps and included learners’ differences especially interms of their personality, age, gender, and learning styles into this category.

Pre-course group

concept map (Number

in parentheses shows the

number of times

a concept was included

in the individual concept

maps.)

Culture was the next concept (present in four maps) and included issuessuch as culture difference, ethnicity effect on learning and teaching styles,and culture shock The group discussion that followed indicated that theparticipants were drawing on not only their prior experiences as students inthe school system but also their own language learning experiences (stated

by two of the seven) and from their experiences as students in their initiallanguageT ESLteacher certificate courses (stated by five of the sevenparticipants); in fact, many of the results of the initial map may be attributed

to some of the participants’ experiences in previous certificate courses asmany of the concepts seem to represent the subject matter of these courses.Thus, the pre-course group concept map gave the instructor someindication of these participants’ prior beliefs about, and experiences with,

Critical reflection was the most popular concept in the post-course groupconcept map and was subdivided into teachers’ personality, self confidence,self-awareness, self-assessment, knowledge of subject matter, classroomlessons, and evaluation Next came research and theory, further subdividedinto theory acquisition, applied linguistics—especially how, research theoryand practice are linked—can anything be proven, corrective feedback, andalternative assessment of students This concept was followed by anothernew concept curriculum design with subdivisions of textbooks, ideology,and materials

All participants wrote that they had noticed a major new concept of criticalreflection in the post-course maps In addition, critical reflection enabledthem to note that the post-course maps showed a different understanding ofthe concept ‘method’ that allowed for a move away from a focus on lookingfor the correct method when constructing the pre-course concept maps.These two findings are important because both concepts were new anddifferent from what the participants had said they ‘believed’ to be true for

T E S Lbefore they had taken the course

Regarding critical reflection, T3, a participant from China, said that shenoticed in her post-course map that she had ‘a new bubble called criticalthinking’ which she explained in the post-course interview as follows:You need your own thinking, not follow others like teaching is theirown voice in their teaching process Not just follows the administratorsthinking I think I need to raise my voice; to express my opinion of whatteaching is and how I should teach in my class including the kind ofmaterial to use in my class, not just the textbook

Later in the same interview she said that she would try to instil this kind ofcritical thinking in her students when she returned to China; she said:

Post-course group

concept map (Number

in parentheses shows the

number of times

a concept was included

in the individual concept

maps.)

‘When I go back to China I would be an English teacher I will let mystudents have this kind of thinking’ Another participant, T5 from Korea,also said that she would try to incorporate critical thinking into her teachingbecause she noted that:

Usually the tradition of education in Korea is we just obey We just followthe rules from the government or from the administrator, or someprincipal in the school But now I can think, I can decide, this is good, butthis is not, or I need to follow this, but I don’t think this is good I candetermine if it’s good for my teaching or for my students This is criticalthinking for me

However, she said that this process would not be easy for her to follow inKorea especially if ‘I want to work in the schools, so I must follow basic rulesotherwise I can be isolated from the other teachers’ In addition, T5 alsomentioned that this process of developing critical reflection was slow for herbecause of her past experiences and beliefs and that she only realized thatshe was becoming more critically reflective ‘almost at the end of the course’.Another participant, T2 also from Korea, voiced similar ideas about criticalreflection when she said: ‘In Korea, we cannot say anything to the professorlike bad things, I don’t think so, I cannot say this, but here I can say, I don’tthink so, and I can explain why I don’t think like this’ Similar to T5 above, itwas not until the end of the course that she realized her changed way ofthinking; she continued:

I realized this near the end of the course when I started to think aboutpolitics I never thought about it, the relationship between politics andeducation And when I saw, like, class ethics, or I thought just racism

I never thought it’s because it’s politics, but while I am reading thereadings, I can say yeah, everything is related to politics

T6, a male Canadian participant also realized that he may have just acceptedall he was presented with before in hisT ES Lcertificate programme withoutquestion; he said:

There were moments, in my teaching profession, until now, where I’vedone something ‘cause that’s the way I’ve done it And, if someone asked

me, I would say ‘well, this is this is, like the way to do it because it’seffective.’ But without really questioning the context that I was in andnot Without really questioning what was going on

Asked again two years after taking the MA course if he still held the sameideas he had expressed above, he said: ‘Yes’ Before taking the MA

programme and in particular the course reported on in this paper, T6 saidthat during hisT E S Lcertificate programme ‘while teachingE S Ltointernational students at the same university (where I got my certificate), Iwas nudged (in some cases harder than in others) towards a certain method

of teaching This helped me at first, as I needed some guidance But, in thelong run, it may have hurt me as well, for I allowed myself to be boxed in—toteach a certain way regardless of the teaching context’ Now, two years sincetaking the Foundations course reported on in this paper, where he said he

‘was asked to consider with a more critical eye the teaching concepts I wasreading about’, he suggests his own learning ‘continues today in my ownteaching experience’