Citation 北海道教育大学紀要. 教育科学編, 72(1): 173‑187

s by EFL Students

Author(s) KIM, Bogyeong; ISHIZUKA, Hiroki

Citation 北海道教育大学紀要. 教育科学編, 72(1): 173‑187

Issue Date 2021‑08

URL http://s‑ir.sap.hokkyodai.ac.jp/dspace/handle/123456789/12042

Rights

ReceptiveandProductiveVocabularyKnowledge ofNegativePrefixesbyEFLStudents

KIMBogyeongandISHIZUKAHiroki

DepartmentofEnglishEducation,AsahikawaCampus,HokkaidoUniversityofEducation

EFL学習者の英語形容詞の否定接頭辞における 受容語彙知識と産出語彙知識

キム　ボギョン・石塚　博規

北海道教育大学旭川校英語教育研究室

ABSTRACT

ThisstudyintendstoexamineEnglishasaForeignLanguage(hereafterEFL)learners’

knowledgeofnegationalprefixesinEnglishadjectivesinboththereceptiveandproductive aspects.AlthoughnegativeprefixesdemandmuchworkingmemoryfromEFLlearnersand haveacrucialimpactonthemeaningofthesentence,notmuchresearchhasattemptedto analyzetheirawarenessofnegativeprefixessystematically.Influencedbystudiesonnative speakers’(hereafterNSs)knowledgeofnegativeprefixes,anexperimentwasconductedto revealEFLlearners’understandingandtheusageofEnglishnegativeprefixes.Toserve thispurpose,sixhighschoolstudentsinKoreaparticipatedinaself-pacedreadingandfill- in-theblanktest.TheresultsshowedthatKoreanEFLstudentsarenotsensitivetothe correctnessoftheformsoftheprefixestothebases.However,theydemonstratedintuition overetymologicaldifferences.Theydichotomizedun-tonativeEnglishwordandIN-(in-, im-,ir-, il-)totheLatinateoriginwordmoreseverelythanNSs.Theresultsalsorevealed that un-isthemost favored negational prefix among theEFLlearners andthey rely heavilyonorthographicalinformationwhenproducingit.Thesefindingshopefullymake contributiontodevelopingmoreeffectivevocabularyteachingmethodsinEFLclassrooms.

Receptive and Productive Vocabulary Knowledge of Negative Prefixes by EFL Students

Derivation isamorphologically complex process sinceitdoesnotdepend entirely onetymology, phonology,orsemantics(Katamba,2006).Amongmanyaffixes,asnegationalprefixesaffectthemeaning ofthegivensentencesignificantly(Sherman,1973),theexplicitteachingofprefixalnegationinEnglish

ishighlyrecommended.However,EFLlearnerswithlimitedexposuretothetargetlanguageareprone tosimplymemorizethemseparately.ThisrequiresdoubleworkloadforEFLlearnersconsideringthe capacityofworkingmemorytoprocessthevocabularyinapsycholinguisticaspect(Lincketal.,2009;

Segalowitz&Hulstjin,2009).TherealsoexistsagapbetweenEFLlearners’awarenessofprefixandthe actualfrequencyofNSs’useofit(Mochizuki&Aizawa,2000).Thus,asteachers,weneedtonotonly lightentheirheavyworkloadinacquiringvocabularywhenexpandingtheirvocabularysizebutalsohelp themlearnauthenticuseofEnglish.

Morphologicalawareness,whichcanbedefinedasan“awarenessofmorphemicstructuresofwords andtheabilitytoreflectonandmanipulatethatstructure”(Carlisle&Feldman,1995,p.194)contributes to learners’ reading ability (Ramirez et al., 2011). Before employing any methodological teaching techniques,there is a prerequisiteto diagnosethe currentstatus of EFL learners’awarenessand understandingofEnglishnegativeprefixes.Thus,thispaperinvestigatestheawarenessofKoreanEFL learnersonprefixalnegationofEnglishadjectives─howsensitivetheyaretothenegationalprefix─and theirpreferencesinproduction.

Literature Review

Prefixal negation in English adjectives: NSs and ESL/EFL learners

The interest in the English negative prefix has generated many studies by historical linguists, morphologists,andresearchersinthesecondlanguageacquisitionfield(Aronoff,1976;Baldietal.,1985;

Nagy&Herman,1987;Mochizuki&Aizawa,2000;Bowers&Kirby,2010).

Amongthem,Baldietal.(1985)isprominentinthattheyexaminedprefixalnegationofEnglish adjectivesinpsycholinguisticdimensionsofproductivity.Theyconductedtwotests:aproductiontest (fill-in-theblank)andapreferencetest(multiplechoice).Intheproductiontest,withcarefullycoined 60pseudo-words,30ofwhichwereetymologicallysimilartoLatin/French,andtheotherhalfwhich resembledwordsofsolelyEnglishorigin,theresearchersaskedparticipantstoaddanegativeprefix beforeeachpseudo-word(p.37).Thiswouldempiricallyverifythedifferenceinproductivityofthe negativeprefixesamongun-, IN- (in-,im-, ir-, il-),non-,anddis-.Etymologically,un-usedtobe attachedonlytothenativeEnglishorigin.Ithasalwaysbeenthestrongest“oustingIN-moreandmore”

throughoutEnglishhistory(p.34).Inthepreferencetest,participantswereaskedtodecidethebestand thesecond-bestnegativeprefixoftherootwith90genuinewordsand60pseudoadjectivesusedinthe firsttest.TheresultoftheproductiontestprovedthatNSsarenotonlyabletomakeuseofetymological informationaboutbases,butalsokeepthephonologicalassimilationruleforIN-prefixation.However, theydidnotstrictlyfollowthedistinction.Theyratherproductivelyandcreativelyappliedtheirintuition ofetymologyandphonology,whenattachingprefixes.Meanwhile,theresultofthepreferencetests demonstratedthatun-isthemostproductivenegativizingprefixandnon-isthesecondmostfavored choiceamongNSs.

TherearenotmanystudiesonEFLlearners’prefixalnegationknowledgeexceptforafewworks dealingwiththeirunderstandingofaffixesingeneral.MochizukiandAizawa(2000)revealedthat Japanesestudents’knowledgeofEnglishaffixeswascorrelatedwiththeirvocabularysizeandtherewas

astableacquisitionorderasinre>un>pre>non>anti>semi>ex>en>post>inter>counter>in>ante.

ThisorderappearstoalignwiththeresultsofBaldietal.(1985)’sproductivityandpreferencetests whenonlynegativeprefixesaresortedout.However,itisinconclusivethattheacquisitionorderreflects prefixproductivityinEFLlearners.Moreover,thesubjectswerehalfcorrectintheaffixesknowledge test,forexample7.24outof13pointsontheaverageintheprefixsection(Mochizuki&Aizawa,2000).

ThisindicatesthatEFLlearnershavearelativelylowunderstandingofprefixesinEnglish.

Double workload on EFL learners’ working memory

Manypsycholinguisticstudieshaveaddressedsecondlanguagelearners’workingmemoryintheir performance.Asworkingmemoryisstrictlylimitedincapacity(Gathercole&Baddeley,1993),ESL/

EFLlearnerstendtochoosethebestplacetodirecttheirattentionto,whentheyfacewithheavytasks.

Takingthisintoaccount,tomemorizebasewordsandderivednegativeformsindependentlyrequires doubleworkloadinlearners’workingmemorytoprocessthem.Thisisnotthemostefficientwayfor learnerstoexpandtheirvocabularysize.Infact,affixknowledge,especiallyderivativeknowledge, bestowslearners’abilitytoreadnewwordsandinfertheirmeanings(Nagyetal.,1993).L1learners’

dramaticexpansionoftheirvocabularysizefromthefourthgradeisattributedtoincidentalvocabulary acquisitionandincreasedknowledgeofaffixedwords(Nagy&Herman,1987;Mochizuki&Aizawa, 2000).

Ithasbeenattestedbymanyresearchersthatmorphologicalknowledgehadaninfluenceonefficient acquisitionofvocabulary,wordrecognition,andevenreadingcomprehension(Laufer,1990;Qian,1999;

Bowers&Kirby,2010).

Psychological Analysis on Bilingual Lexical Selection: the Bilingual Interactive Activation Model, and the Revised Hierarchical Model

AprecisemechanismofhowL2orthographicformsareactivatedinalearner’scognitioniswell explainedthroughtheBilingualInteractiveActivation(hereafterBIA)modelbyDijkstraandVan Heuven(1998).TheBIAmodelspecificallytriestorevealthewordformationprocessinbilingualism (Kroll & Tokowicz, 2005). It is based on the interactive activation model, which claims that comprehensionprocessingisinitiatedbyvisualinputfromthetext,takingabottomupapproach,from lettertolettertoformaword.Then,languagenodesareappliedtodecidewhethertheL1orL2willbe selectedintheoppositeapproach,top-downway.ThistheoryexplainsthatbothL1andL2lexiconare integratedandthelexicalaccessisnonselective(Dijkstra&VanHeuven,2002).AccordingtotheBIA model,EFLlearnersshouldrespondsensitivelytothederivedwordwithprefixedletter(s).Itisbecause theirlexiconisactivatedinbothL1andL2sothattheyareabletoaccessthemeaningofthederived wordoncetheyrecognizetheletterattachedtothebase.Howelaboratelyandsensitivelytheyreactto theletter(s)hasnotbeenyetresearchedsystematically.

Anotherpsycholinguisticmodelconsideredascomplementationforthepitfallsofpreviousmodels (Potteretal.,1984)istheRevisedHierarchicalModel(Kroll&Stewart,1994,hereafterRHM).TheRHM ismeanttoexplainthedevelopmentoflexicalandconceptualrepresentationinthesecondlanguage.

ThismodeldescribesthatL2haslinkstobothL1andconceptsbutthestrengthoftheassociationis

different.WhilelexicalconnectionsofL2toL1arestronger,thelinksofL2toconceptsareweakerand theirbondinggetsrobustastheproficiencylevelofalearnerimproves(Kroll&Tokowicz,2005).The implicationofthissequenceofdevelopmentallowsteacherstopredictthatbeginnerandintermediate levelofEFLstudentsmayaccesstheinputsuperficiallybydrawingtheirattentiontotheorthographic orphonologicalfeatureswhileadvancedlearnerswouldbemoredeeplyinvolvedinmeaningaspects, makinguseofsemanticinformation.

Psycholinguistic tools to measure EFL learners’ comprehension and production

Typesoftasksrequirelanguageuserstotakedifferentapproachestovocabularyprocessing.For example,comprehensioninitiateswithwordrecognition,stimulatedbypropertiesofinputandadvances towardstheconceptsthewordconveys,whileproductionbehavesintheoppositeway.Itstartswiththe conceptsandreceivesfeedbackfromsemanticsandembodiesthemeaningintoawordform(Kroll&

Dijkstra, 2002; Hermans, 2000). Thus, the time language selection occurs is also different. In comprehension,languagenodesarenotactivateduntilrelativelylatewhilelanguagecueinproductionis encodedaspartoftheconceptualrepresentationoftheevent(Kroll&Tokowicz,2005).

Commonly traditional cross-modal priming and masked priming tasks are used to observe morphologicalprocessinginpsycholinguisticdimensions.Thesemethodsaretomeasurethetimespent onaparticipant’s word recognition when avisual target word ispresented with auditory priming stimulusorviceversa(Roberts,2013;Ahnetal.,2009).However,theyarenotappropriatetoinvestigate learners’sensitivitytowardthemarkednessofill-formedprefixedderivationsastheypresupposethe primingeffect.Ontheotherhand,aself-pacedreading(hereafterSPR)hasbeenadoptedtoexaminethe placewherecognitivedifficultyarisesinlearners’languageprocessingthoughitusuallydealswith syntacticstructure(Jegerski&VanPatten,2013).Thelongertimeasubjectspendsonacertainword, themoredifficultyisassumedtoberaised.ThisbasicassumptioniswellapplicabletomeasureEFL learners’cognitiveprocessincomprehension.

Recently,on-linemeasurementforproductionhasthelimelightinsteadofoff-linetests.Itisbelieved toplainlyreflectwhathappensinlearners’cognitionandproductionprocess(Milleretal.,2008;Van Patten,2013).EFLlearners,however,feelpressuredbyon-linetasksbecausetheyhavehardlybeenin thesituationtobeforcedtoproduceL2onthespot,andtherefore,tomeasuretheirgenuinecompetence, productiontasksneedtobemodifiedaccordingtothecharacteristicsandlearningenvironmentofthe participants(Newton&Nation,2020;Panahzadeh&Asadi,2019;Yuan&Ellis,2003).

Purpose of the Study

Basedonthestudiesandavailableresearchtoolsdescribedinthepreviouschapter,thisstudyseeksto examine actual EFL learners’ receptive and productive knowledge on negational prefix in English adjectives.IthasthepurposeofdevelopingconcreteteachingtechniquestolessentheburdenofEFL learners’workingmemoryinexpandingtheirvocabulary.Accordingly,thispaperattemptstoaddress thefollowingquestions:

1.TowhatextentareEFLlearnersawareoftherelationsbetweenthenegationalprefixesandtheir

bases?

２.WhichnegationalprefixdoEFLlearnersproducethemostwhentheyencounteranewbase?

Methods

Participants

TheparticipantsweresixofthefirstgradestudentsatapublichighschoolinSeoul,Korea.Interms ofproficiency,twowereinbeginner,twoinintermediate,andtwoinadvancedlevel.Thisdivisionwas basedonthefirstsemesterEnglishgradesattheschoolandthevocabularytestdesignedforSPR.In thisprerequisitevocabularytest,thebeginnerlevelstudents’averagescoreswereintherangefrom30 to40outof100,thoseoftheintermediatelevelstudentswereinbetween50and60,andtheadvanced levelstudents’averagescoreswereinbetween80and100.Eachgroupconsistedofonemale,andone femaleandallofthemdidnothaveanyexperienceoflivinginthecountrieswhereEnglishisthe dominantlanguage.

Materials

Self-paced Reading Sentences

Toobservehowquicklyandaccuratelytheparticipantsrespondtothewell-formedandill-formed prefixesintheSelf-pacedReading(SPR)test,14Englishadjectiveswithnegativeprefixeswereselected, withtwowordsforeachofsevencategoriesofnegationalprefix,un-,IN- (in-, ir-, im-, il- ),dis-,non-.

Theywerechosenaftercross-checkingCOCA(Corpusof ContemporaryAmericanEnglish)with frequencyrangedfrom1106to5535and2067essentialvocabularyrecommendedforhighschoolstudents bytheKoreanMinistryofEducation.Thenusingthese14adjectives,14simplesentenceswereprepared (withoutembeddedclauses),andanother14sentenceswithonlyprefixesreplacedwithill-formswere alsoprepared.Inaddition,10fillers,whichdidnotincludeanynegativewerealsoprepared.Allthe sentenceswereexcerptedfromtheexamplesintheelectronicdictionaryandcarefullymodifiedtoavoid anysyntacticambiguityasbelow(Appendix1indetail).

e.g. ir-adjective1a.Hevisitedhisparentsatirregularintervals.

ir-adjective1b.^＊Hevisitedhisparentsatunregularintervals.

Fill-in-the-blank Test Questions

Thefill-in-theblanktestquestionsweredevisedtoinvestigatetheparticipants’preferencesamong negativeprefixesandtheiraccuracywhentheyattachedtheprefixestothegivenbases,especiallyin relationtoetymologicalinformationandphonologicalassimilation.Thirtyadjectiveswerechosento generatethetest.Twentywerefromcrosscheckingtheaboveessentialvocabularyandthewordsused inBaldietal.(1985)’sclassification(Twoforeachcategory:Native un-, Latin un-, Latin in-, im-, ir-, il- withfournon-adjective, andfour dis-adjective).Tenwerefromthelistof60pseudo-adjectivesin thesameresearch.Amongthem,fivewereLatinatepseudo-wordandtheotherhalfwerenative-pseudo word.Thesefivewordsfromeachgroupfulfilledphonologicalconditionsoffivecategoriesintheprevious study:vowel,neutralconsonant,/l/,labial(/m/or/p/),and/r/asfollows(Appendix2indetail).

e.g.pseudoLatinatewith/l/condition:limoral pseudonativewith/l/condition:loofish

Procedures

Preparation (Prerequisite Vocabulary Test)

AllsixstudentstookthevocabularytestaboutthewordstheywouldencounterintheSPRtest.In thispriortest,12targetbaseformswereused.Among14basesexplainedintheabovesection,‘stop’and

‘popular’wereexcludedsince‘stop’belongedtotheelementarylevelessentialvocabularybytheKorean MinistryofEducationand‘popular’wasintroducedineverymiddleschooltextbookandcoveredinthe lessonwhentheexperimentwasconducted.Thetestguaranteedthatparticipantshadlittleknowledge aboutthetargetwordsorderivedformofthemintheirlexicon.Theywerealsogiventhelistofthe vocabularytopreventanydifficultyinunderstandingthemeaningsofthesentencesintheSPRtest.

However,thislistwascomprisedofallthebaseformswithoutprefixes.Participantsweretoldthatthey weregoingtoreadsentencescontainingthosewordsintheSPRtestandgiventhreedaystomemorize themifneeded.

Self-paced Reading Test

ThestudentstooktheSPRtest,withnon-cumulativelineardisplayofwordbyword,followedbya Yes/Nocomprehensioncheck-upquestionaboutthesentenceshown,suchas“Theroomisinadequate forourlargegroup,”followedbythequestion,“Theroomwastoosmallforagroup.”(Yes/No).Either well-formedorill-formedwordwaschosenrandomlyfromthepair-sentencesbyLingerprogram,aself- pacedreadingsoftware,andpresentedtoeachstudentsothateachstudentencountered24sentences(14 targetsentenceswith10fillers).　

Fill-in-the-blank Test

Theparticipantswereaskedtowritedownanegativecounterpartofthegivenwordbyaddingan appropriateprefix.Iftheyalreadyknewtheprefixalnegationformofthegivenadjective,theyhadto tickinthecheckboxnexttothequestion.Itwastoruleoutthecasethattheyalreadymemorizeda prefixednegationadjectiveasanindependentwordintheirlexicon.

In Depth Interview

AftertheSPRtestandfill-in-the-blanktest,eachstudenthadafivetoseven-minute-longinterview session. They were asked how they had felt in the SPR test and why they had chosen a certain negationalprefixoverothersinthefill-in-theblanktest.Allinterviewswerevoice-recordedwiththe participants’consents.

Data Treatment

For the SPR test, all data was collected regardless of the correctness of the answer to the comprehensionquestionfollowingthesentence.Thenthemeantimeandstandarddeviation(hereafter SD)ofwell-/ill-formedwordswerecalculatedandcomparedacrossparticipants’proficiency.Themean

timeandSDpereachtypeofprefixeswasalsocalculatedbyusingtheautomaticfunctionofMicrosoft Excelprogram.Inthefill-in-theblanktest,theproportionofeachprefixthatparticipantsusedwas calculated.ThiswastosortoutthemostpreferrednegationalprefixamongKoreanEFLlearners.With pseudo-wordsofLatinateandnativeEnglish,eachstudent’sanswerwascollectedandcategorizedbased onetymologicalgroundsasinBaldi,etc.(1985).Theproportionwasalsocalculatedtocomparewiththe dataofNSs.　

Results

Self-paced Reading Test (Comprehension Test)

ThemeantimeinTable1showsthatslightlylongertimewasspentonill-formedadjectivesthan well-formed.However,theill-formedSDvalueintotalisalmostasdoubleasthatofwell-formed adjectives.Thiswascausedbytheintermediatelevelparticipantslingeringonincorrectadjectivesina widerangedependingontheitems.

Figure1andTable2demonstratethemeanreadingtimespentonthenegatedadjectivesdepending ontypesofprefixesWhiletheresultshowsdifferencesacrossproficiencylevels,participantslingeredon well-formedwords,exceptthecasesofun-, il-anddis-.

Amongtheexceptions,learnersspentmuchmoretimeonthewronguseofil-prefix.Itiseasyto Table 1

Mean(M) and StandardDeviation(SD) of Reading Time on Target Adjectives

M SD

Well-formed Ill-formed Well-formed Ill-formed

Advanced 1465.93 1509.14 658.58 659.64

Intermediate 5054.13 6974.25 5152.87 12137.28

Basic 4134.47 3581.55 3711.57 3175.44

Total 3598.9 3897.73 3947.30 7305.8

Note.Themeasurementofthetimeismillisecond(ms)

Figure 1

Average Reading Time onTarget Adjective per Types of Prefix

attributethisprominencepartiallytothelimitedconditionofil-whichshouldbefollowedbyonly‘l’

startingbase.However,ir-,whichalsoimposeshighconstraintontheonsetofthebaseshowsthe completelyoppositeresult.

Fill- in-the Blank Test (Production Test)

Outofthe20genuineadjectives,advancedparticipantsshowed50％ofaccuracyinattachingaprefix tothebasecorrectlyandbothintermediateandbasiclevelreached30％ofaccuracy.AccordingtoTable 3, the preference of the prefix un- was revealed regardless of the participants’ proficiency levels (average38％ ).ThisresultconformswiththeresultofNSsinBaldietal.(1985).

However,incontrasttothesecondstrongestchoiceofnon-byNSs,theparticipantsusedmoredis- andin-prefixationthannon-.Therestoftheprefixessuchasim-, ir-,andil-weredistributedalmost evenlyalthoughir-wastheleastfavored.

Table4consistsoftheactualprefixesthatparticipantsattachedtothepseudowordswithcertain Table 2

Mean Reading Time(ms) onTarget Adjectives per Types of Prefix Types of

Prefix Markedness Advanced Intermediate Basic Total

un- well-formed 1012 1714.5 4124 2283.5

ill-formed 1042.5 2707.5 3444 2398

in- well-formed 1371 5403 1859 2877.16

ill-formed 1046.5 4519 1630 2560.5

im- well-formed 1672.5 6051.3 4994.3 4560.2

ill-formed 1232 1473 7891 2948

ir- well-formed 1127.5 8752.5 2536 4137.16

ill-formed 910.5 3792.5 2531 2411.3

il- well-formed 1506 7457.5 2014 3659.16

ill-formed 2330 22956 6716 10667.3

non- well-formed 2232.5 1823 3623.3 2711.5

ill-formed 920.5 4712.5 1446 2670.4

dis- well-formed 1790.5 3678.5 3724.5 3064.5

ill-formed 1311.5 4789 10691 4578.4

Note.Themeasurementofthetimeismillisecond(ms)

Table 3

Proportionof the Use of Each Prefix

un- in- im- ir- il- non- dis-

Advanced 45％ 10％ 7％ 7％ 8％ 5％ 18％

Intermediate 37％ 20％ 10％ 5％ 10％ 8％ 10％

Basic 33％ 13％ 12％ 5％ 10％ 13％ 13％

Total 38％ 14％ 9％ 6％ 9％ 9％ 14％

Note.Alldatawasroundedofftothenearesttenth,whichresultedin99％totalinthebasicgroup.

phonologicalconditions.Alltheparticipantsprefixedil-toLatinatepseudowordwith/l/sound.They also showed the strong tendency to attach un- to the native English origin words. However, the participantsdidnotstrictlyfollowthephonologicalassimilationrulesastheygeneratedwordssuchas inrepuliant, nonmonarial, dismonarial, or unplankity.

Discussion

Receptive Vocabulary Knowledge on Negational Prefix

Themeanreadingtimespentonwell-andillformsintotaldidnotshowmeaningfulimplicationdueto thehugeSDvalue(Table1,Table2andFigure1).Inotherwords,L2learnersreacteddifferentlyto eachprefix.Theystayedlongeronwell-formedadjectivesexceptun-, il-,anddis-.Theywereexpected tostaylongeronill-formedadjectives,becausetheywouldunconsciouslyfeeloddnesswhenreadingthe sentences,iftheyhadsensitivitytowardcorrectnessoftheformforeachbase.Throughthisitcanbe interpretedthatL2learnersdonothavesensitivitytotheappropriatenessoftheprefixtothebase.For theexceptionofun-, anddis-,participantsexplainedintheinterviewthattheyfeltmorefamiliarwith un-anddis-prefixtoothers.Thisispresumablybecauseofitsfrequencyinthetextbookusedinthe classroom.Regardingil-,thestudentsshowedparticularsensitivitytothemarkednessofill-formed words(Figure1).Atfirst,itwasassumedtobeduetotheconstraintof‘l’startingroot.However,they hardlyreactedtotheill-formsofir-whichrequiresthe‘r’onsetconditionaswell.Severalstudents answeredintheinterviewthattheydependedonorthographicalfeaturestodeterminetheadequate

Table 4

Participants’ use of Prefix to thePseudo-word

Pseudo-Latinate phonological

condition vowel Netural

consonant /l/ labial

/m/,/p/ /r/

Words exliar Sempient Limoral monarial repuliant

Advanced in Un Il im in

Advanced un In Il im ir

Intermediate Il Non Il non un

Intermediate im Im Il im ir

Basic dis Dis Il im ir

Basic Ir Il Il dis in

Pseudo-Native(un-) phonological

condition vowel netural

consonant /l/ labial

/m/,/p/ /r/

Words arnful Finkled loofish plnkity rimbled

Advanced in Dis Un Un dis

Advanced dis Dis Un Un ir

Intermediate un Dis Il Im un

Intermediate dis Ir Non Un il

Basic ir In Un Un un

Basic un Un Dis Un un

prefix,andtheybelievedthat“l”isvisuallyprominentincontrasttomanyofother“short”characters suchasm-, r-orn-.Theythoughtthattheprefixil-shouldbeharmoniouswith“tall”letterl.

Still,theparticipantscorrectlyunderstoodthemeaningoftheprefixedadjectivewithlittleregardfor typesofprefix.Themomenttheyrealizedthatitreferredtotheoppositemeaningofthebase,their focusmovedtothesemanticpropertyofthewordasthegoaloftheiractivitywastoanswerthe followingcomprehensionquestion.Theyreactedtothestimulusbutnotinanextremelyelaborateway.

Inotherwords,EFLlearnersfirstlyrecognizedtheletter,atypeofnegationalprefix,butfailedtojudge thegrammaticalappropriatenessofaffixation,andjustprocessedthemeaningoftheprefixedadjective.

Throughthisoperation,EFLlearnerscanprocesstheinformationtoservethepurpose(answering comprehensioncheck-upquestions)asefficientlyaspossible.

Productive Vocabulary Knowledge on Negational Prefix

Un-wasthemostproductiveandfavorednegationalprefixamongtheparticipantsinthisstudy.The secondmostfavoredprefixwasdis-(Table3).Theresultsareverymuchrelatedtothefrequencyof exposuretoeachprefix.Infact,un-isthestrongestproductiveprefixusedbyEnglishnativespeakers aswell.ItsusagehasbeenexpandedtotheloanwordsthroughoutEnglishhistorysothatitnaturally attachestobothnativeEnglishandLatinatewords.ThevocabularyinthetextbookthatKoreanEFL studentsareusingcanbesaidtoaccuratelyreflectthecurrentprefixstatusinEnglish-spokencountries.

Theusageofnon-showedsomedifferencesbetweenEFLlearnersandNSs.AccordingtoBaldietal.

(1985),NSsusenon-asmuchasIN-.Theyexplainthatitispartiallyattributedtonon-’squestionable statusasaprefixandtotheeasinessofattaching,becausenon-isneitherinvolvedinphonological assimilationprocessnoraffectedbyetymology.ForKoreanEFLlearners,however,non-ishardlya productive prefix. In fact, it was the least favored prefix among them if in-,im-,il-, and ir- were categorizedasIN-prefix.Intheinterview,itwasrevealedthattheyhardlyregardednon-asatypeof negativeprefixasNSs.Theyratherconceiveditasaword,sothattheyfeltreluctanttogenerate derivativesbyattachingnon-tothebase.

Surprisingly,thetestusingpseudovocabularyaffirmedKoreanEFLlearners’sensitivitytoward etymologicalinformationaboutbases.Theyevendichotomizedun-andIN-(in, im-,ir-, il-)more severelythanNSs.Table5showstheprefixselectionofpseudowordsbyKoreanEFLlearnersin comparisonwithNSsbasedonthedatafromBaldietal.(1985).

Table 5

Percentage of prefixselection of pseudo words in the production

Korean EFL Learners Native Speakers

Native Latin Native Latine

un- 47％ 7％ 48％ 35％

correct IN 7％ 53％ 14％ 23％

incorrect IN 17％ 20％ 1.5％ 2.3％

non- 3％ 7％ 12％ 19％

dis- 23％ 10％ 13％ 10％

Other 0％ 0％ 11％ 10％

IncorrectIN-referstotheviolationofphonologicalassimilationconditionsuchasinrepuliant^＊instead ofirrepuliant.TheproportionofincorrectIN-ismuchhigherinKoreanEFLlearnersthanthatofNSs.

However,Table5 demonstratesthatKoreanEFLlearnersarepronetoattachun-tothenativeEnglish wordsandIN-(especiallyin-morph)toLatinateoriginwordsexclusively.Fifty-threepercentoftheir answerswerecorrectIN-incontrasttoNSs’mixeduseofun-andin-toLatinatepseudowords.Inaddition, thesametendencywasobservedconcerningtheusageofnon- inpseudowordsprefixation.Theparticipants frequentlyuseddis-prefixandtheleastfavorednon-whereasNSsproducednon-asmuchasIN-.

Withthefollow-upindepthinterview,theparticipantswereprovedtoorthographicallydistinguish nativeEnglishwordsfromLatinatewordsandlimitedtherangeoftheprefixun-’susagetoonlynative Englishwords.Itisprobablethattheycanuseetymologicalinformationwhengeneratingprefixed adjectivesalthoughthisclear-cutdivisionmaypreventlearnersfromattachingnegationalprefixas flexiblyasNSsdo.

TheirheavyrelianceonorthographicfeaturesofthebasecanbeexplainedbytheRHMproposedby KrollandStewart(1994).AllpseudowordsusedinthisstudyhadbeendevelopedbyBaldietal.(1985) on the ground of numbers of syllables (e.g.Latin-polysyllabic), suffixes, spelling and consonant peculiarities.Thus,thedifferencebetweennativeandLatinatewordswaseasilyspottedvisuallysothat EFL learners were able to precisely identify which base belonged to which origin. Moreover, the interviewrevealedthatalllevelsofparticipantsindeedtookadvantageoforthographicalinformation.

ThisisconformabletotheRHM,explainingL2hasstrongconnectionwithL1inlexicallevel,butnot withconcepts.Theparticipants’dichotomizationmayreflectthattheconceptionofL2prefixednegation doesnotexistinlexicon.Therefore,itisassumedthattheyconceivethewordinanonlysurfacelevel–

orthographical features not the meaning the word embodies, unless they are required to do comprehensiontasks.

Thedifferencesacrossproficiencyareworthwhiletobeaddressed.Basedontheirresultsinthe productiontest,advancedlearnerswerefoundtosuccessfullylinkL2andconceptdirectly.Itindicates thathighproficiencylearnersconceiveawellformednegationaladjectiveasawordandstoreitasitis intheirlexicon.ThistendencyofdevelopmentisalsoalignedwiththeRHM,arguingthatdevelopmental sequenceofbilinguallexiconisfromrelianceontranslationequivalentsbetweenL1andL2todirect conceptmediation.

Conclusion

Summary

Korean EFL participants in this experiment were not fully equipped with sensitivity to the appropriatenessofnegativeprefixforeachbase.Oncetheyrecognizedanynegationalprefixattachedto the base and if they were ready to process the meaning, they showed little effort to judge the appropriatenessoftheprefixusage.

Duetothefrequencyofexposure,KoreanEFLparticipantsfavoredandproducedun-prefix,andtheir secondchoicewasdis-.TheyalsoprovedtohaveclearsensitivitytoetymologicallyLatinateandnative Englishwordsdistinctiononthegroundofprimarilyorthographicalfeatures.Evidenceinthisempirical

datacontradictstoparticipants’assertionoftheirmorphologicalinterlanguagerules,thatistosay,reliance onphonologicalassimilation.Althoughtheyarguedthattheyreliedheavilyonphonologicalfeatures,all levelsofparticipantsmadeuseoforthographicalinformation,so-calledspellingoftheonsetofthebase.

Oneoftheinterestingdifferencesacrossproficiencylevelsisthatadvancedlearnerswereableto access information from various sources such as phonology, syntax, and semantics along with morphologicalknowledge.TheRHMendorsesthisoutcomewiththedevelopmentalsequenceofforming strongerconnectionbetweenL2andconceptinalearner’slexiconashis/herproficiencyimproves.

Pedagogical Implication and Limitation

Itisanundeniablefactthatexplicitmorphological vocabularyteachinghasbeenexcludedsince consciousness-raising and noticing took center of the stage in SLA. Moreover, current vocabulary teachingseemstodisapprovemorphologicalmentioneitherinexplicitlyorimplicitlyinEFLclassroom.

Thisabsenceofmorphologyteachinginhighschoolleaveslearnersresponsibleformemorizingagreat dealofvocabulary.Theresultsofthisstudyrevealedthatlearnershavedifficultyinmemorizingthe baseanditsprefixedformseparately,andtheiraccuracyinmatchingthecorrectnegativeprefixtothe basedropssignificantly.Thus,toreducetheburdenoftheworkingmemoryandmenditsinefficiency, vocabularyteachingbasedonmorphologyshouldnotbediscarded.

ThisstudydiscoveredthatKoreanEFLlearnershaveintuitiononetymologicalinformationofLatin/

French word and native English word. It also found that the criterion they count on is basically orthographic features. These advantages can be maximized by drawing their attention to the orthographiccharacteristicoftheprefixexplicitlywithrichcontextprovided.Atthesametime,for productiveskills,practitionersshouldmakeeffortstoblurthedivisionbetweenun-andIN-,asNSs allowun- attachmenttonotonlynativeEnglishwordsbutalsoLatinateoriginatedwords.Teachers shouldfocusontheproductivenegationalprefixes,suchasun-ornon-tohelplearnersnottorefrain themselvesfromusingone-to-onematchforaffixation.Practitionerscandiscussmoreeffectiveand efficientwaysofteachingvocabularyinmorphologicalaspecttominimizetheredundantconsumptionof EFLlearners’workingmemory

Thelimitationofthisstudyisthesmallnumberofparticipants,whichbringsageneralizabilityissue.

Inaddition,asmosteducationresearchentailsasitslimitation,theresultofthestudydoesnotprovide evidencetoeasilyapplytooraccountforvocabularylearningprocessofEFLlearnersasthiswas conductedinaspecificcontextaswell.Afurtherstudycouldexpandthenumberofparticipantsto providemoregeneraldiscussion,andmorepsycholinguistictoolsandtheorieswouldbeimplementedto validatethefindingsofthisstudy.

References

Ahn,H.D.,Lim,I.J.,Jeon,M.K.,Han,J.I.,Hwang,J.B.,&Hong,W.P.(2009).

안희돈

임인재

전문기

한정임

황종배

홍우평

모국어

화자와

외국인

학습자의

한국어

접사형태

처리

통일인문학

Aronoff,M.(1976).Word-formationingenerativegrammar.LinguisticInquiry Monograph,1,Cambridge:MITPress.

Baldi,P.,Broderick,V.,&Palermo,D.S.(1985).PrefixalnegationofEnglishadjectives:psycholinguisticdimensionsof productivity.Historical Semantics–HistoricalWord-Formation,33-58.

Bowers,P.N.,&Kirby,J.R.(2010).Effectsofmorphologicalinstructiononvocabularyacquisition.Reading and Writing,23

⑸,515-537.

Carlisle,J.F.,&Feldman,L.B.(1995).Morphologicalawarenessandearlyreadingachievement.Morphological Aspects of Language Processing,189-209.

Dijkstra,T.,&VanHeuven,W.J.(1998).TheBIAmodelandbilingualwordrecognition.Localist Connectionist Approaches to Human Cognition,189-225.

Dijkstra,T.,&VanHeuven,W.J.(2002).Thearchitectureofthebilingualwordrecognitionsystem:Fromidentificationto decision.Bilingualism:Language and Cognition,5⑶,175-197.

Gathercole,S.E.,&Baddeley,A.D.(1993).Phonologicalworkingmemory:Acriticalbuildingblockforreadingdevelopment andvocabularyacquisition?.European Journal ofPsychology of Education,8⑶,259-272.

Hermans,D.(2000).Word production in a foreign language.KatholiekeUniversiteitNijmegen.TheNetherlands.

Jegerski,J.,&VanPatten,B.(2013).SELF-PACEDREADING.Research Methods in Second Language Psycholinguistics.

Routledge.20-49.

Katamba,F.(2006).Modern Linguistics: Morphology.London:FirstpublishedbytheMacmillanPressLtd.

Kroll,J.F.,&Dijkstra,A.(2002).Thebilinguallexicon.InR.Kaplan(ed.),Handbook of applied linguistics.301-321.Oxford, U.K.:OxfordUniversityPress.

Kroll,J.F.,&Stewart,E.(1994).Categoryinterferenceintranslationandpicturenaming:Evidenceforasymmetric connectionsbetweenbilingualmemoryrepresentations.Journalof Memory and Language,33⑵,149-174.

Kroll, J.F., & Tokowicz, N. (2005). Models of bilingual representation and processing. Handbook of bilingualism:

PsycholinguisticApproaches,531-553.

Laufer,B.(1990).Easeanddifficultyinvocabularylearning:Someteachingimplications.Foreign Language Annals,23⑵, 147-155.

Linck,J.A.,Kroll,J.F.,&Sunderman,G.(2009).Losingaccesstothenativelanguagewhileimmersedinasecondlanguage:

Evidencefortheroleofinhibitioninsecond-languagelearning.Psychological Science,20⑿,1507-1515.

Miller,K.S.,Lindgren,E.,&Sullivan,K.P.(2008).Thepsycholinguisticdimensioninsecondlanguagewriting:Opportunities forresearchandpedagogyusingcomputerkeystrokelogging.TESOL Quarterly,42⑶,433-454.

Mochizuki,M.,&Aizawa,K.(2000).AnaffixacquisitionorderforEFLlearners:Anexploratorystudy.System,28⑵,291-304.

Nagy, W.E., & Herman, P.A. (1987). Breadth and depth of vocabulary knowledge: Implications for acquisition and instruction.The Nature of Vocabulary Acquisition,19,35.

Nagy,W.E.,Diakidoy,I.A.N.,&Anderson,R.C.(1993).Theacquisitionofmorphology:Learningthecontributionofsuffixes tothemeaningsofderivatives.Journal ofLiteracy Research,25⑵,155-170.

Newton,J.M.,&Nation,I.S.P.(2020).Teaching ESL/EFL listening and speaking.Routledge.

Panahzadeh,V.,&Asadi,B.(2019).OntheImpactsofPressuredvs.UnpressuredOnlineTaskPlanningonEFLStudents’

OralProductioninClassroomandTestingContexts.Eurasian Journal of AppliedLinguistics,5⑶,341-352.

Potter,M.C.,So,K.F.,VonEckardt,B.,&Feldman,L.B.(1984).Lexicalandconceptualrepresentationinbeginningand proficientbilinguals.Journal of Verbal Learning and Verbal Behavior,23⑴,23-38.

Qian,D.(1999).Assessingtherolesofdepthandbreadthofvocabularyknowledgeinreadingcomprehension.Canadian ModernLanguage Review,56⑵,282-308.

Ramirez,G.,Chen,X.,Geva,E.,&Luo,Y.(2011).MorphologicalawarenessandwordreadinginEnglishlanguagelearners:

EvidencefromSpanish-andChinese-speakingchildren.Applied Psycholinguistics,32(03),601-618.

Roberts, L. (2013). CROSS-MODAL PRIMING WITH SENTENCES. Research Methods in Second Language Psycholinguistics,Routledge.212-229.

Segalowitz,N.,&Hulstijn,J.(2009).Automaticityinbilingualismandsecondlanguagelearning.Handbookofbilingualism:

Psycholinguistic Approaches,371-388.

Sherman,M.A.(1973).Boundtobeeasier?Thenegativeprefixandsentencecomprehension.Journal of Verbal Learning and VerbalBehavior,12⑴,76-84.

Yuan,F.,&Ellis,R.(2003).Theeffectsofpretaskplanningandonlineplanningonfluency,complexityandaccuracyinL2

monologicoralproduction.Applied linguistics,24⑴,1-27.

VanPatten,B.(2013).ThepsycholinguisticsofSLA.Research methods in second languagepsycholinguistics,Routledge.17-35.

Appendix A

Self-paced Reading Test

1.Itwasapainfuland(un/in)populardecision.

２.Shewas(un/in)fortunatetoloseherhusband.

3.Theroomis(in/un)adequateforourlargegroup.

４.Hewantsthepowertofire(in/un)competentemployees.

5.Thereisnothing(im/in)moralaboutwantingtoearnmoremoney.

6.Itseems(im/in)probablethatthecurrentsituationwillcontinue.

7.Hevisitedhisparentsat(ir/un)regularintervals.

8.Manypeopleconsiderpolitics(ir/un)relevanttotheirlives.

9.Heoncewas(il/un)literatebutnowheknowshowtoread.

10.Allpartiesregardedthetreatyas(il/un)legitimate.

11.Ithinkithasrained(non/un)stopsinceFridayafternoon.

12.Hygienewas(non/un)existentnorunningwaternobathroom.

13.Iwouldratherbepoorthangetmoneyby(dis/in)honestways.

14.Therehavebeenmanycasesofstudentsbeing(dis/ir)respectfultoteachers.

Appendix B

Fill-in-the blank Test

Filling-in-the-BlankActivity

※Chooseoneappropriateprefixamongthe<Options>andattachittoeachgivenwordtogeneratethe wordwiththeoppositemeaning.(Ifyoualreadyknowthederivedfromoftheoppositemeaning, pleasecheck(√)inthebox.)

〈Options〉un-, in- , ir- , im- , il- , non- , dis-

1.similar 　 　□ 16.arnful 　 　□

２.patient 　 　□ 17.selfish 　 　□

3.logical 　 　□ 18.natural 　 　□

４.responsible 　 　□ 19.limoral 　 　□

5.accurate 　 　□ 20.resistable 　 　□

6.exliar 　 　□ 21.trustful 　 　□

7.dairy 　 　□ 22.smoking 　 　□

8.loofish 　 　□ 23.rimbled 　 　□

9.expected 　 　□ 24.aware 　 　□

10.standard 　 　□ 25.plinkity 　 　□

11.monarial 　 　□ 26.sempient 　 　□

12.composed 　 　□ 27.white 　 　□

13.finkled 　 　□ 28.agreeable 　 　□

14.tolerant 　 　□ 29.repuliant 　 　□

15.measurable 　 　□ 30.legal 　 　□

＊

＊

＊

＊

＊

＊

＊

＊

＊

＊

＊

＊

＊

＊

(キム　ボギョン　旭川校教員研修生)

(石塚　博規　旭川校教授)