[PDF] OutSpoken ELA Principles of instructed language learning

View - Download OutSpoken ELA

Previous PDF

Next PDF

Principlesofinstructed languagelearning

q

RodEllis

DepartmentofAppliedLanguageStudies andLinguistics,Univers ityofAucklan d,

PrivateBag92019, Auckland,NewZealand

Received10October 2004;receivedin revisedform12 December2004;accepted 13Decembe r2004

Abstract

Thisarticlerepresents anattempt todrawtogether findingsfroma rangeofsecond language principlesaddress suchissuesas thenatureof secondlanguage(L2) competence(asformulaic andrule-basedknowledge), thecontributionsof bothfocuson meaningand onform,the need todevelopboth implicitand explicitsecondlanguage knowledge,theproblems posedbythe learner!s"built-insyllabus!,therolesofinput,outputandinteractionin learning,theimportance ofbothfree andcontrolledproduction. Theprinciplesdraw onavariety oftheoreticalperspec- tivesandare o!eredas "provisionalspecifications!foralearning-centered languagepedagogy. !2005ElsevierLtd. Allrightsreserved. Keywords:Instruction;Secondlanguageacquisition; Provisionalspecifications

1.Introduction

SecondLanguageAcquisition (SLA),assub-disciplineof appliedlinguistics,is stilla veryyoung fieldof study.Whileitmaynot bepossibletoidentify itsprecise

0346-251X/$-seefrontmatter !2005ElsevierLtd. Allrights reserved.

doi:10.1016/j.system.2004.12.006 q Theresearchon whichthisarticle isbasedwas fundedbythe MinistryofEducation, Wellington, New

Zealand.

E-mailaddress:r.ellis@auckland.ac.nz.

www.elsevier.com/locate/system

System33(2005) 209-224

SYSTEM

startingpoint,manyresearch erswouldagree thatthelate60smarkedthe onsetof anintense periodofempiricaland theoreticalinter estinhow secondlanguages are acquired.Muchofthis researchhasbeendirected atunderstanding andcon tribut- ingtomore e!ectiveinstructed languagelearning. Inaddition tothenumerous studiesthathaveinvestiga tedthee !ectsofinstructi onon learning(Norrisand Ort- ega!smeta-analysis publishedin2000 identified79studies),much ofthetheori zing aboutL2instructionhas beenspecifica llyundertakenwithlanguagepedagogyin mind,forexampl eKrashen !sMonitorMod el( Krashen,1981),Long!sInteraction Hypothesis(Long,1996),DeKeyser !sskill-learni ngtheory(DeKeyser,1998), VanPatten!sinputprocess ingtheory (VanPatten,1996, 2002)and myowntheory ofinstructed languagelearning(Ellis,1994)alladdress theroleof instructionin

L2acquisition.

However,theresearchand theorydonot a!ordaunifor maccount ofhow instructioncanbestfacilitate languagelearning. Thereis considerablecontroversy (seeEllis,2006).Inparticular ,thereis noagreementastowhet herinstructio nsho uld bebased onatraditionalfocus-on -formsapproach,involv ingthesyst ematicteach- ingofgramm aticalfeaturesin accordancewithastructuralsyllabu s,or afocus- on-formapproach,involvingatten tiontolingui sticfeatures inthecontextof communicativeactivitiesderivedfroma task-basedsyllabusor somekindofcombi- nationofthetwo.Nor isthereagreem entabout thee"cacyof teachingexp licit knowledgeoraboutwhattype ofcorrective feedbacktopro videor evenwhenexp li- citgrammar teachingshouldcommence .Thesecontroversiesreflect boththecom- plexityoftheobject ofenquiry(instruct edlanguageacquisiti on)andalso thefact thatSLAis stillinits infancy. Giventhesecontroversies,it mightbethought unwisetoattempttoform ulatea setofgen eralprinci plesofinstructedlanguageacq uisition.Hatch!s(1978a) warning -"applywithcaution !-isas pertinenttod ayasit wassomethirtyyearsago . Nevertheless,Ithinkthereisaneed totryto drawtogether aset ofgeneralizat ions thatmightserve astheba sisforlanguage teachereducation, andI amnotalonein this,forLightbown(1985,2000)hasfelt andrespo ndedtoasimilarne ed.IfSLAis too !erteachersguidancethere isaneedtobite thebulletand pro!eradvice,so long asthisad vicedoes notmasqueradeasprescriptions orproscriptions(andthere isal- waysadan gerthatadvice willbesoconstrued)an dsolong asit istentative, inthe formofwhat Stenhouse(1975)called"provisionalspecifications!.Iha vechosento presentmyownprovisional specifications intheform of"principles!.Ido notexpect thatallSLA researchersor alllanguage teacherswillagree withthem.I hope, though,thattheywillprovide abasis forargument andforreflect ion.

2.Principle 1:Instructionneedstoensure thatlearnersdevelo pbotharichrepertoir eof

formulaicexpressionsanda rule-basedcompetence ProficiencyinanL2 requiresthat learnersacquire botharich repertoireofformu- laicexpressions, whichcatersto fluency,andarule-b asedcompetenceconsistingof knowledgeofspecificgramm aticalrules,whi chcaterto complexityandaccuracy

210R.Ellis/ System33(2005) 209-224

(Skehan,1998).Thereis nowwidespreadaccepta nceofthe importanceplayed by formulaicexpressionsinlangu ageuse.Nativespeakers havebee nshowntousea muchlarger numberofformulaicexpressions thanevenadvanc edL2learne rs(Fos- ter,2001).Formula icexpressionsmayalsoserve asabasisforthe laterdevelopmen t ofarule- basedcompetence. N.Ellis(1996),for example,has suggestedthatlearners bootstraptheirwaytogramm arbyfirst internalizingan dthen analyzingfixedse- quences.Classroomstudies byEllis(1984),Mylesetal. (1998,1999) andMyles (2004)demonstratethatlearnersoften internalizerote- learnedmaterialaschunks, breakingthemdown foranalysis lateron. Traditionally,languageinstructionhas beendirect edatdevelopingrule- based competence(i.e.,knowledgeof specificgrammatical rules)throughthesystemat ic teachingofpre-selectedstruc tures- whatLong(1991)hasreferredto asafocus- on-formsapproach.W hilesuchanapproachc ertainlyreceivessupportfromthe researchthathasinvestigated directinterventio nininterlanguag edevelopment, curriculumdesignersa ndteachersneedtorecogni zethatthistypeofinst ructionis aslikelyto resultin studentslearni ngrote-memor izedpatterns asininternalizing abstractrules(Myles,2004 ).This neednotbeseenasan instructional failure howeverassuchpatterns areclearly ofvalueto thelearner.Itpoints insteadto anacknowledgemen tofwhatcanberealisticallyachieved bya focus-on-forms approach,especiallywithyoung, beginnerlearners. Ifformulaic chunksplayalargerole inearlylanguageacquisition,it maypay to focusonthese initially,delayingthe teachingof grammaruntil later, asIhave pro- posedinEllis(2002).Anotional- functionalapproachlends itselfperfectlytothe teachingofprefabricatedpatte rnsandrou tinesandmayprovideanideal foundation fordirectinter vention intheearlystages.Clearly,though,aco mpletelanguagec ur- riculumneedstoensure thatitcater stothe developmentofboth formulaicexpres- sionsand rule-basedknowledge.

3.Principle 2:Instructionneedsto ensurethatlearnersfocuspredom inantlyonmeaning

Theterm"focusonmean ing!issomewhat ambiguous.Itisnecessary todistin- guishtwodi !erentsenses ofthisterm. Thefirstrefers totheidea ofsemantic meaning(i.e.,themeani ngsoflexical itemsorof specificgrammaticalstructures). Thesecond senseoffocusonmeaning relatesto pragmaticmeani ng(i.e., the highlycontext ualizedmeaningsthatarise inactsofcommunication). Toprovide opportunitiesforstudentsto attendto andperformpragma ticmeaning, atask- based(or,at least,atask-sup ported)approachto languageteachin gisrequ ired. Itisclear lyimportant thatinstructionen suresopportunitiesforlearnersto focus onbothtypes ofmeaning but,arguably, itispragma ticmeaningthatis crucial tolanguagelearning. Thereisan importantdi !erenceinthe instructional approachesneed edfor semanticandpragmaticmeanin g.Inthe caseofsemanticmeani ng,theteacher andthestudent scan treatlanguageasanobjectand functionas pedagoguesand learners.Butinthecase ofpragmatic meaning,they needtoview theL2 asatool

R.Ellis/ System33(2005) 209-224211

forcommunica tingandtofunctionascommuni cators. 1

Ine!ect,thisinvolv estwo

entirelydi!erentorientations toteachingandlearning. Theopportunity tofocusonpragmaticmean ingisimpor tantfor anumber of reasons:

1.In theeyesof manytheorists(e.g., Prabhu,1987;Long,1996),only whenlearners

areengagedin decoding andencoding messagesinthecontextof actualactsof communicationaretheconditionscreated foracquisition totake place.

2.Todevelop truefluencyin anL2,learne rsmusthave opportunities tocreate prag-

maticmeaning(DeKeyser,1998).

3.Engaging learnersinactivitieswher etheyare focusedoncreating pragmatic

meaningisintrinsicallymoti vating. Inarguingthe needfora focusonpragma ticmeaning, theorists dosonot justbe- causetheyseethisas ameansof activatingthe linguisticresourc esthat havebeen developedbyothermeansbut becausetheysee itasthe principal meansbywhich thelinguistic resourcesthemselvesarecreat ed.Thisisthetheoretical positionthat hasinformedmanyhighly successfulimmersioneducation programmesaround theworld (seeJohnsonandSwain,1998 ).However, inadvocatingthisprinciple, I donotwish tosuggestthat instruction needsto bedirectedexclus ivelyatprovidin g learnerswithopportunitiesto createpragma ticmeaning,onlythat, tobee !ective, instructionmustincludesuch opportunitiesandthat,ideally, overan entirecurric- ulum,theyshou ldbepredomi nant.

4.Principle 3:Instruction needstoensure thatlearnersalsofocusonform

Thereisnowawidespread acceptancethat acquisitionalso requiresthat learners attendtoform.Indeed, according tosometheoriesofL2acquisi tion,suchatte ntion isnecessary foracquisitiontotakeplace. Schmidt(1994),forexampl e,hasargued thatthereis nolearni ngwithout consciousatten tiontoform. 2 Again,though,theterm "focusonform !iscapableof morethanone interpreta- tion.First,it mightreferto ageneralorient ationtolangu ageas form.Schmidt (2001)dismissesthisglobalattention hypothesis,arguing thatlearners needtoattend 1 Itisalso possibletoteach pragmaticmeaningas an"object!.Thatis, specificpragmatic meanings(e.g., requestingor apologizing)can beidentifiedandinstructionalmaterials developedtoteach learnersthe linguisticmeansfor performingthesestrategies. SeeKasperandRose (2002)forexamplesof studiesthat haveinvestigatedthe e!ectivenessofthis approach.Suchan approachconstitutesa versionof"focus-on- forms!,discussedon p.3.Here, however,Iwish toemphasizethe needto creatematerialsthat allow studentstocreate theirownpragmat icmeanings throughcommunication. 2 Theextentto whichattentionto formisnecessaryforlearningremains controversialhowever.A numberof researchers(e.g.,Williams,2005)haveprovided evidencetodemonstrate thatsomelearning takesplacewithout awareness.Schmidt(2001)hasmodifiedhis positionsomewhatto allowforthe possibilityofnon-consc iousregistratio noflinguisticform,arguingonly that"moreattention resultsin morelearning!(p.30).

212R.Ellis/ System33(2005) 209-224

tospecificforms.Second, itmightbetakentosuggest thatlearners needtoatte nd onlytothe graphicorphonetic instantiationsof linguisticform s.However, theorists suchas Schmidtand Longareinsistentthatfocus onformrefer stoform -function mapping(i.e.,the correlation betweenaparticu larformandthemeaning(s) itreal- izesincommuni cation).Third, "focusonform !mightbe assumedto refertoaware- nessofsome underlying,abstract rule.Schmidt, however,iscarefulto arguethat attentiontoformrefers tothe noticingof specificlinguisticitems,asthey occurin theinput towhichlearnersare exposed,not toanawar enessofgrammaticalrules. Instructioncancaterto afocuson formina numberofways :

1.Through grammarlessonsdesignedto teachspecificgrammati calfeaturesby

meansofinput- oroutputprocess ing.An inductiveapproach togrammar teach- ingisde signedtoencourage "noticing!ofpre-selected forms;adeductive approach seekstoestabl ishanawareness ofthegrammaticalrule.

2.Thro ughfocusedtasks(i.e.,tasksthat requirelearners tocomprehendand pro-

cessspecificgramm aticalstructuresin theinput,and/ortopro ducethe structures intheperfor manceof thetask).

3.Bymeans ofmethodol ogicaloptionsthat induceattentionto forminthecontext

ofperforming atask.Twomethodological optionsthathave received consider- ableattention fromresearchersare(a) theprovision oftimeforstrategicand on-lineplanning(YuanandEl lis,2003 ;FosterandSkeh an,1996)and(b) correc- tivefeedback( Lyster,2004). Instructioncanseekto providean intensivefocusonpre-se lectedlinguistic forms (asina focus-on-forms approachorin alessonbuiltaroundafocusedtask) oritcan o!erincidental andextensiveattention toformthroug hcorrectivefeedbackintask- basedlessons. Thereareprosandconsfor bothintensive andextens ivegrammar instruction.Somestructures maynotbe masteredwithouttheopportunity forre- peatedpractice. Harley(1989),forexampl efound thatAnglophonelearnersof L2 Frenchfailedtoacquire thedistinction betweenthe preteriteand imparfaitpast tensesafterhour sofexpo sure(andpresumablysomecorrective feedback) inan immersionprogrammebut wereabletoimprove theiraccuracyin theuseof these twotensesafter intensiveinstruction. However, intensiveinstruction istimeconsum- ing(inHarl ey!sstudythe targetedstructures weretaught overan8weekperiod!) and thustherewill beconstraintson howmanystruc turescanbe addressed. Extensive grammarinstruction,on theotherhand,a!ordstheoppor tunityforlarge numbers ofgrammatical structurestobeaddressed.Also, morelikelythan not,manyof the structureswillbeattended torepeat edlyover aperiodoftime.Further, becausethis kindofinstructio ninvolves aresponsetotheerrorse achlearnermakes,it isindivid- ualizedanda!ordstheskilledteacher on-lineopportunit iesforthekindofco ntex- tualanalysis thatCelce-Murcia(2002)recommendsasabasis forgrammarteachin g. Ellisetal. (2001)reportedthat extensiveinstructi onoccurredrelativelyfrequently in communicativeadultESLlessons throughbothpre-emptive(i.e., teacherorstudent - initiated)andreactive(i.e., correctivefeedback) attentiontoform.Loewen(2002) showedthatlearnerswho experiencedsuch momentaryform-focusedepisodes

R.Ellis/ System33(2005) 209-224213

demonstratedsubsequentlearning oftheformsaddressedin bothimmediateandde- layedtests. However,itis notpossibleto attendtothose structuresthatlearne rsdo notattemptto use(i.e., extensiveinst ructioncannot dealwithavoidance). Also,of course,itdoesnot providethein-depth practicethatsomestructuresmay require beforetheycanbe fullyacquired. Arguably,then, instructionneed stobe conceived ofinterm sof bothapproaches.

5.Principle4: Instruction needstobe predominantlydirectedatdeveloping implicit

knowledgeoftheL2while notneglecting explicitknowledge ilyamatt erofimplici tknowledge.Explicitknow ledge"isthede clarativeand often anomalousknowledgeof thephonological,lexical,gramm atical,pragma ticandso- cio-criticalfeaturesofan L2togetherwiththemeta languagefor labellingthis knowl- edge!(Ellis,2004 ).Itis heldconsciously, islearnable andverbal izableandistypicall y ticdi"cultyintheuseof theL2.A distinctionne edstobe drawnbe tweenexpli cit knowledgeasanalysedknowl edgeandas metalingualexplanation.The formerentails aco nsciousawarenessofhowa structuralfeatureworkswhile thelatterconsis tsof knowledgeofgrammatical metalanguageand theabilitytounderstandexp lanations ofrules. Giventhatitisimpl icitknowledge thatunderlies theabilityto communicate fluentlyandconfiden tlyinan L2,itisthistypeofknowledgethat shouldbe theultimate goalofanyinst ructionalprogramme.How thencanit bedeveloped? Thereareconflicting theoriesregardi ngthis.Accordingtosk ill-buildingtheory (DeKeyser,1998),implicit knowledgearisesoutofexplicit knowledge,whenthe latterisproceduralizedthrough practice.In contrast,emergentisttheori es(Kra- shen,1981;N.Ellis,1998)seeimplici tknowledge asdevelopingnaturallyout ofmeaning-focu sedcommunication,aided,perhaps,bysomefocus onform.Irre- spectiveofthesedi!erenttheoretical positions,thereisconsensus thatlearners needtheopportunityto participatein communicativeactivityto develop implicit knowledge.Thus,communicative tasksneedtoplayacentra lroleininst ruction directedatimplicit knowledge. Thevaluein teaching explicitknowledge ofgrammarhasbeen andremainstod ay oneofthe mostcontroversia lissues inlanguageped agogy.Inordertomakesenseof thedi!erentpositions relatingtotheteachingof explicitknowled geitis necessary to considertwoseparate questions:

1.Isexplicit knowledge ofany valueinandofitself?

2.Isex plicitknowl edgeofvalueinfacilitating thedevelopmentofimplicit

knowledge?

214R.Ellis/ System33(2005) 209-224

Explicitknowledgeisarguabl yonlyofvalueif itcan beshownthatlearners are abletoutil izethistype ofknowledgeinactual performance.Agai n,there iscon- troversy.Onepositionis thatthisis verylimited.Krashen(1982)arguesthatlearn- erscanonly useexplicit knowledgewhen they"monitor!andthatthisrequires that theyare focusedonform (asopposedto meaning)an dhavesu "cienttimetoac- cesstheknowl edge.Otherpos itionsarepossible.Itcan beargued thatexplicit knowledgeisusedin boththe processofform ulatingmessages aswellasinmon- itoringand thatmany learnersare adroitinaccessingtheirexpli citmemoriesfor thesepurposes, especiallyiftherules are,toadegree,automatize d.However, this doesrequire time.YuanandEllis(2003)showedthatlearners!grammaticalaccu- racyimproved significantlyiftheyha dtimefor"on-lineplann ing!whileperforming anarrativetask, aresult mostreadily explainedin termsoftheiraccessing explicit knowledge. Irrespectiveofwhetherexpli citknowledge hasanyvalueinandof itself,itmay assistlanguage developmentby facilitatingthedevelopment ofimplicit knowledge. Thisinvolves aconsiderationofwhat hasbecome knownasinterfacehypothesis, whichaddresses whetherexplicitknowledge playsaroleinL2 acquisition.Three positionscanbeidenti fied.According tothenon- interfaceposition(Krashen,

1981),explicitand implicitknow ledgeareentirely distinctwiththeresul tthatex-

plicitknowl edgecannotbeconvertedinto implicitknowl edge.Thisposition issup- portedbyresear chthat suggeststhatexplicitand implicitmemor iesare neurologicallyseparate(Paradis,1994).Theinterface positionarguesthe exact opposite.Drawingonskill-learni ngtheory(DeKeyser,1998),itargues thatexplicit knowledgebecomesimplicit knowledgeif learnershavetheopportunity forplenti- fulcommunicative practice.Theweakinterfaceposit ion(Ellis,1993)claims that explicitknowledgeprimes anumberofkeyacquisi tionalprocesses,inparticular "noticing!and"noticingthegap!(Schmidt,1994).Thatis, explicitknowl edgeofa grammaticalstructuremakes itmorelikelylearnerswill attendtothestructure inthe inputan dcarryoutthecognitive comparisonbetweenwhatthey observe intheinput andtheir ownoutput. Thesepositionscontinue tobeargued atathe- oreticallevel. Thethreepositions supportvery di!erentapproachesto languageteachin g.The non-interfacepositionleadstoa"zerogrammar !approach,i.e.,onethatpriori tizes meaning-centeredapproachessuchastask-based teaching.Theinterfacepos ition supportsPPP-theidea thatagramm aticalstruc tureshould befirst presentedexpli c- itlyandthen practiseduntil itisfully proceduralized.Theweakinterface position hasbeenused toprovide abasis forconsciousn ess-raisingtasks( Ellis,1991)thatre- quirelearners toderivetheirownexplicit grammarrules fromdata theyareprovided with. Thisprincip le,then,assertsthatinstruction needsto bedirecteda t developingbothimplicitandexplicitknowl edge,giving prioritytotheformer. However,teachersshould notassumethatexplicitknowl edgecanbeconverted intoimplicit knowledge,astheextent towhichthisispossible remains controversial.

R.Ellis/ System33(2005) 209-224215

6.Principle5: Instructionneeds totakeinto accountthelearner's'built-in syllabus'

Earlyresearchintonaturalistic L2acquisi tionshowed thatlearnersfollowa "natural!orderand sequenceofacquisition(i.e., theymasterdi!erentgrammatical structuresinarelatively fixedand universalorder andtheypassthrougha se- quenceofstagesof acquisitionon routetomasteringeachgramm aticalstructure). Thisledresearchers likeCorder(1967)tosuggestthat learnersha dtheirown "built-insyllabus!forlearninggramm arasimplicit knowledge.Krashen(1981)fa- mouslyarguedthatgrammar instruction playednorole inthedevelopment of implicitknowledge(what hecalled"acquisition!),aview basedonthe conviction thatlearners (includingclassroomlearners)wouldautoma ticallyproceedalong theirbuilt-in syllabusaslongasthey hadaccesstocomprehensible inputand weresu"cientlymotivated.Grammarinstructi oncouldcontributeonly toexplicit knowledge("learning!). Therefollowedanumber ofempiricalstudiesdesigned to(1)compare theorder of acquisitionofinstructedand naturalisticlearne rs(e.g., Pica,1983),(2)compare the successofinstructedan dnaturalistic learners(Long,1983)and(3) examinewhet her attemptstoteachspecificgramm aticalstruc turesresultedin theiracquisition (Ellis,

1984).These studiesshowedthat, byandlarge,theorderand sequenceofacquisition

wasthesame forinstructed andnaturalistic learners,a findingsupportedbylater re- search(e.g., Ellis,1989;Pienemann,1989);thatinst ructedlearners generally achievedhigherlevelsof grammaticalco mpetencethannaturalisticlearne rsand that instructionwasnoguarantee thatlearners wouldacqu irewhattheyhad beentaught. Thisledtotheconclusi onthatit wasbeneficial toteachgramm arbutthatitwas nec- essarytoensureit wastaught inawaythatwascompat iblewi ththenatural pro- cessesofacqu isition. How,then,caninst ructiontake accountofthelearner!sbuilt-in syllabus?There areanum berof possibilities:

1.Adopt azerogrammarapproach, asproposed byKrashen. Thatis,employa

task-basedapproachthatmakes noattempttopredetermine thelingui sticcontent ofaless on.

2.Ensure thatlearnersaredevelopmental lyreadyto acquireaspecifictarge tfeature.

However,thisisprobablyimpr acticalasteacher shaveno easyway ofdetermin- ingwher eindividualstudentshave reachedanditwouldnecessitate ahighlyindi- vidualizedapproachtocaterfor di!erencesinde velopmentallevel amongthe students.Also,aswe notedearlier,suchfine-tuning maynotbe necessary. While instructioninatarget featuremaynot enablelearners to"beat!thebuilt-in sylla- busitmay servetopus hthemalong itaslong asthetarge tstructureisnot toofar aheadoftheirdevelopmental stage.

3.Focusthe instructionon explicitratherthan implicitknowledgeasexpli citknowl-

edgeisnotsubject tothesame developmentalconstraints asimplicit knowledge. Whileitisprobablytrue thatsomedeclar ativefactsabo utlanguageare easier to masterthanothers,thisis likelyto reflecttheirco gnitiveratherthan their

216R.Ellis/ System33(2005) 209-224

developmentalcomplexity,whichcanmoreeasily betakenintoaccount indecid- ingthe orderofinst ruction.Trad itionalstructuralsyllabuses, infact,aregraded ontheba sisofcogn itivecomplexity. 3

7.Principle 6:Successfulinstructedlangua gelearningrequires extensiveL2input

Languagelearning,whether itoccursinanaturalistic oraninstructedcontext, isa slowandlabori ousprocess. ChildrenacquiringtheirL1take betweentwoandfive yearsto achievefullgramm aticalcompetence, duringwhich timetheyareexposed tomassive amountsofinput.EllisandWells(1980) demonstratedthatasubstantial portionofthevarian ceinspeed ofacquisition ofchildrencanbeac countedforby theamoun tandthequality ofinputtheyreceive.The sameisundoub tedlytrueof L2acquisition. Iflearnersdonotrecei veexposure tothetargetlanguage theycannot acquireit.In general,the moreexposure theyreceive,themore andthefast erthey willlearn.Krashen(1981,1994)hasado ptedaverystrongposition ontheimpor - tanceofinp ut.Hepoints tostudiesthathaveshown thatlength ofresi dencein thecountry wherethelanguageisspoken isrelatedtolanguageproficiencyan dother studiesthatthathave foundpositive correlationsbetw eentheamount ofreadingre- portedandproficiency /literacy.For Krashen,however,theinputmust bemade "comprehensible!eitherbymodifying itorby meansofcontextualprops.Research ers maydisagree withKras hen!sclaimthat comprehensibleinput (togetherwith motiva- tion)is allthatis requiredfor successfulacquisi tion,arguingthat learneroutputis alsoimportant (seePrinciple7below) buttheyagree aboutthe importanceof input fordevelopin gthehighlyconnectedimplicitknowl edgethatis neededto becomean e!ectivecommunicatorinthe L2. Howcan teachersensure theirstudentshaveaccess toextensive input?Ina "sec- ond!languageteachingcontext, learnerscanbeexpected togainaccesstoplenti ful inputoutside theclassroom,a lthough,asTanaka(2004)hasshownin astudyof adultJap anesestudentslearningEnglish inAuckland,notallsuch learnersare suc- cessfulinachieving this.Ina "foreign!languageteaching context(aswhenFrench or Japaneseistaughtinschools intheUnited KingdomorUnit edState s),thereare far feweropportunities forextensiveinput.Toensure adequateaccess ,teachersneedto:

1.Maximise useoftheL2insi detheclassroom. Ideally,this meansthatthe L2needs

tobecome themediumas wellasthe objectofinst ruction.A studybyKimand Elder(2005)revealedthatforeignlangu ageteachers ofFrench,German,Japanese andKoreanin Aucklandsecondary schoolsvaried enormouslyintheextentto whichtheyemploy edtheL2 intheclassroom(i.e.,between 88%and 22%of thetotalinpu t). 3 Agood exampleofwhere "cognitivecomplexity!and"developmentalcomplexity!canbe distinguishedis subject-verbagreementin English.Thisis typicallyintroducedvery earlyin structuralcoursesbut itis invariablyonlymastered atavery advancedstage ofdevelopment.

R.Ellis /System33 (2005)209-224217

2.Create opportunitiesforstudentstorecei veinputoutsidethe classroom.Thiscan

beachieve dmosteasilybeproviding extensivereading programmesbased on carefullyselectedgradedreaders ,suitedtothelevelof thestudent s,asrecom- mendedbyKrashen(1989).Elley(1991)reviewedstudies thatshowedthatL2 learnerscanbenefitfromboth readingandfrom beingread to.Als o,ideally, if moreresourc esareavailable,schoolsneed toestablish self-accesscenterswhich studentscanuseoutside classtime.Successf ulFLlearne rsseekoutopportunities toexp eriencethelanguageoutsideclass time.M anystudents areunlikelytomake thee!ortunlessteachers(a) makeresou rcesavailableand(b) providelearner- traininginhowto makee!ectiveuseof theresources. Itcanbe claimed withconfi dencethat,iftheonlyinput studentsreceiveisinthe contextofalimit ednumberof weeklyless onsbasedonsomecourse book,theyare unlikelytoachievehigh levelsof L2proficiency.

8.Principle7: Successfulinstruct edlanguage learningalsorequires opportunitiesfor

output ContrarytoKrashen !sinsistencethat acquisitionis dependenten tirelyoncompre- hensibleinput,mostresear chersnowacknowl edgethatlearne routputalsoplaysa part.Skehan(1998)drawingonSwain(1995)summarizesthecontributionsthat out- putcanmake:

1.Production servestogeneratebetterinput throughthefeedback thatlearne rs!

e!ortsatprod uctionelic it.

2.Itforces syntacticprocess ing(i.e.,oblige slearnerstopayattention togrammar).

3.It allowslearne rstotestouthypotheses aboutthetargetlangu agegrammar.

4.Ithe lpsto automatizeexistingknowledge.

5.Itprovides opportunitiesfor learnersto developdiscourseskills,for exampleby

producing"longturns!.

6.Itis important forhelpinglearne rstodevelopa"personalvoice!bysteering con-

versationontotopicsthey areinterested incontributing to.Ellis(2003)addsone othercontributionofoutput:

7.it providesthe learnerwith"auto-input!(i.e.,learners canattendtothe "input!pro-

videdby theirownpr oductions). Theimportance ofcreatingopportunitiesfor output,including whatSwain(1985) hascalledpushed output(i.e.,outputwherethelearne risstretchedtoexpress mes- sagesclearlyan dexplicitly), constitutesoneofthemain reasonsforincorporating tasksintoa language programme.Contro lledpracticeexercisestypicall yresultin outputthatislimitedin termsof lengthandcomplex ity.Theydo nota!ordstudents opportunitiesforthekindofsustained outputthattheori stsargueis necessary for interlanguagedevelopment. Research(e.g.,Allenetal., 1990)hasshown thatex- tendedtalkofaclause ormorein aclassroom contextismore likelyto occurwhen

218R.Ellis/ System33(2005) 209-224

studentsinitiateinteractions intheclassroomandwhenthey havetofindtheir own words.Thisisbest achievedby askinglearners toperform oralandwrittentasks.

9.Pr inciple8:Theopportunityto interactinthe L2iscentr altodevelopingL2

proficiency Whileit isusefulto considertherelat ivecontributions ofinputand outputto acquisition,itisalsoimpor tanttoackn owledgethatboth co-occurin oralinter- actionandthatboth computationaland socioculturaltheori esofL2 acquisition haveviewed socialinteractionasthematrix inwhich acquisitiontakesplace. AsHatch(1978b)famouslyputit"onelearns howtodo conversation,one learns howtointer actverbally, andoutoftheinteracti onsyntacticstructures aredevel- oped!(p.404).Thus ,interacti onisnotjustameans ofautomatizingexisting lin- guisticresources butalsoofcreatingnewresourc es.According tothe Interaction Hypothesis(Long,1996),interaction fostersacquisitionwhena communication problemarisesand learners areengaged innegotiatingformeaning. Theinterac- tionalmodifications arisinghelptomakeinput comprehensible,providecorrective feedback,andpus hlearners tomodifytheirownoutput inuptake.Accordingto thesociocultural theoryofmind,interaction servesasa formofmediation, en- ablinglearners toconstructnewforms andperfor mnewfunctionscolla boratively (Lantolf,2000).According tothisview,learningisfirst evidenton theso cial planeandonly lateronthe psychologicalplane. Inboththeori es,whi lesocial interactionmaynotbe viewedasnecessa ryforacq uisition,itis viewedas apri-quotesdbs_dbs9.pdfusesText_15

[PDF] pdf viewer android studio github

[PDF] pdf viewer android studio project

[PDF] pdf viewer visual studio 2015

[PDF] pdflatex use custom font

[PDF] pdfminer c#

[PDF] pdfminer htmlconverter

[PDF] pdfminer java

[PDF] pdfminer layout

[PDF] pdfminer python 3

[PDF] pdfminer python 3 documentation

[PDF] pdfminer python 3 tutorial

[PDF] pdfminer slow

[PDF] pdfminer textconverter

[PDF] pdfminer.pdfpage python 3

[PDF] pdt cocktail book pdf free