Teachers of English to Speakers of Other Languages, Inc.

(TESOL)

Comments on Rod Ellis's "The Structural Syllabus and Second Language Acquisition". Implicit/
Explicit Knowledge and Language Pedagogy
Author(s): Rod Ellis
Source: TESOL Quarterly, Vol. 28, No. 1 (Spring, 1994), pp. 166-172
Published by: Teachers of English to Speakers of Other Languages, Inc. (TESOL)
Stable URL: http://www.jstor.org/stable/3587206 .
Accessed: 20/06/2014 22:38

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Teachers of English to Speakers of Other Languages, Inc. (TESOL) is collaborating with JSTOR to digitize,
preserve and extend access to TESOL Quarterly.

http://www.jstor.org

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
Krashen, S. (1985). The inputhypothesis: Issuesand implications.
Oxford: Pergamon
Press.
Larsen-Freeman, D., & Long, M. H. (1991). An introduction to secondlanguage
acquisitionresearch.New York: Longman.
Logan, G. (1988). Toward an instancetheoryof automatization.Psychological Re-
view,95, 492-527.
Mathews, R. C., Buss, R. R., Stanley,W. B., Blanchard-Fields,F., Cho, J. R., &
Druhan, B. (1989). The role of implicitand explicit processes in learning
fromexamples: A synergistic effect.Journal ofExperimentalPsychology,
Learning,
Memoryand Cognition,15, 1083-1100.
McLaughlin, B. (1987). Theoriesof secondlanguage learning.London: Edward
Arnold.
Pienemann, M., Johnston,M., & Brindley,G. (1988). Developing an acquisition-
based procedure for second language assessment.Studiesin SecondLanguage
Acquisition, 10, 217-243.
Reber, A. S. (1989). Implicitlearningand tacitknowledge.JournalofExperimental
Psychology: General,118, 219-235.
Reber, A. S. (1993). Implicitlearningand tacitknowledge: An essayon thecognitive
unconscious. Oxford: Oxford UniversityPress.
Robinson,P. J. (1994). Problemsof knowledgeand the implicit/explicit distinction
in SLA theoryconstruction.University ofHawai'i Working Papersin ESL, 13(2),
90-128.
Schmidt,R. (1992). Awareness and second language acquisition.AnnualReview
of AppliedLinguistics,13, 206-226.
Servan-Schreiber,E., & J. R. (1990). Learning artificialgrammarswith
Anderson,,
competitivechunking.Journalof Experimental Learning,Memoryand
Psychology:
Cognition,16, 592-608.
Sharwood Smith,M. (1991). Speaking to manyminds: On the relevanceof differ-
ent typesof language informationforthe L2 learner.SecondLanguageResearch,
7, 118-132.
Shiffrin,R., & Schneider,W. (1977). Controlledand automatichuman information
processing: Perceptual learning,automaticity, attendingand a general theory.
Psychological Review,84, 127-190.
Vokey,J., & Brooks, L. (1992). Salience of item knowledgein learning artificial
grammars.Journalof Experimental Psychology: Learning,Memoryand Cognition,
18, 328-344.
White,L. (1989). Universalgrammarand L2 acquisition. Amsterdam,Netherlands:
John Benjamins.

ImplicitlExplicitKnowledge
and Language Pedagogy
ROD ELLIS
TempleUniversity

I would like to considerfourissueswhichPeterRobinsonraisesin

his critiqueof myarticle:

knowledgedistinction
1. The statusof the implicit/explicit
2. The natureof implicitknowledge

166 TESOL QUARTERLY

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
3. The relationship and implicit
betweenexplicit knowledge
between
4. The relationship andlanguagepedagogy.
research,
theory,

THE STATUS OF THE IMPLICIT/EXPLICIT

KNOWLEDGE DISTINCTION
It is notentirely clearto me whether Robinsonis objecting to the
distinction betweenexplicitand implicit knowledge or merelyto my
presentation of it.It iscertainly misleading tostatethat"Ellisreplaces
Bialystok's (1982) +/- analyzed distinctionwith a distinction between
twotypesofknowledge, and
explicit implicit." In fact,Bialystok began
byusingtheexplicit/implicit distinction(seeBialystok, 1978,1981)and
laterabandonedthisin favorof +/-analyzed(seeBialystok, 1982).In
on
drawing Bialystok's work, I have opted for the earlier distinction
becauseI finditclearerandbetter suitedtothepurposeofconstructing
a theory of secondlanguageacquisition (SLA).Whereasthelaterdis-
tinction allowsonlyforthedevelopment of unanalyzedknowledge
intoanalyzed,whichrestricts itsrelevance to SLA (a pointRobinson
acknowledges), the earlier distinctionpermits a two-way interface, with
explicitknowledge into
converting implicit as well as implicit ex-
into
plicit.
The distinction betweenexplicit/implicit knowledge did notorigi-
natewithBialystok, of course,althoughshe was one of thefirstto
applyittoSLA. Itiscommon incognitive psychology andepistemology
(see Bialystok, 1981, and Robinson's own helpful of recentrefer-
list
encesfromcognitive psychology). However, as Robinson rightlypoints
out, the distinction is not uncontroversial. To sortout someof the
problems, itis usefultodistinguish explicit/implicitknowledge and ex-
plicit/implicit learning. Robinson tends to blur the two. The former is
muchless controversial thanthe latterif onlybecauseit does not
necessitate anyreference to "conscious" and "subconscious" learning
processes; the distinction between andimplicit
explicit knowledge can
be operationalized in termsof whether thelearneris awareof what
s/heknowsandcanverbalize it.In mydiscussion oftheexplicit/implicit
distinction, I have taken care to consider onlyknowledge in orderto
head-off someofthecriticisms thathavebeenleveledatthedistinction
whenitis appliedto learning(see McLaughlin, 1990).
In short,Robinson's claimthatmy"invocation of a distinction be-
tweenimplicit and explicitknowledge to replaceone of Bialystok's
constructs, thedimension ofanalysis, leadstoconfusion andtheoretical
incoherence" is a reflection of hisownignorance of Bialystok's early
work(his confusion) and a failureto recognizethe importance of
distinguishing and
knowledge learning in discussions of the explicit/
implicitdistinction(his theoreticalincoherence).

THE FORUM 167

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
THE NATURE OF IMPLICIT KNOWLEDGE
Robinsonis also criticalof myclaimthatimplicitknowledgeconsists
of formulaicand rule-basedknowledge,although,again,hiscriticisms
become confusedbecause he failsto make a clear distinction between
knowledge and learning. However, I accept that I am obligatedto
justify the claim and to relateit to discussionsof implicitknowledge
in the literature.
The main point of contentionappears to lie in whetherformulas
(i.e., ready-madechunksof language)shouldbe consideredto be part
of implicitknowledge.Mydecisionto includeformulaswithinimplicit
knowledgewas motivatedby mygeneralcharacterization of implicit
as
knowledge knowledge thatis intuitiveand unanalyzed and explicit
knowledgeas knowledgeof whichthe holderis aware and which,as
a consequence,is analyzed.To quote fromN. Ellis (1993):

Somethingswejust can do, likeour nativelanguage(L1), ridinga bike,

or dreaming.We havelittleinsightintothenatureof theprocessing in-
volved:likeswallowsfly,wejust do them.Otherof our abilities
depend
on ourknowing howtodo them,likespeaking or
pigLatin,multiplication,
cooking froma recipe.(p. 290)

Now, it seems obviousthatformulaicchunkslikeI don'tknowor Can

I havea ____?constitutethe kind of knowledgeinvolvedin "things
we canjust do" ratherthanthekindof knowledgeinvolvedin abilities
likemultiplication.Bialystokand SharwoodSmith(1985) reacha simi-
lar conclusion,treatingformulaicchunksas part of the unanalyzed
knowledgesystem.To claim thatformulaicknowledgeis partof im-
plicitknowledgeis not so controversial.
The questionarisesas to whetherimplicitknowledgeis to be charac-
terizedentirelyas formulaicin natureor whetheritcan also takesome
other form.I have chosen to argue thatimplicitL2 knowledgecan
also be abstractand rulebased-a positionthataccordswiththatfound
in earlyinterlanguagetheoryand whichcurrentlyinformsSLA re-
searchin the fieldof UniversalGrammar.I am aware,however,that
this claim is controversialand that certaincognitivetheories(e.g.,
parallel distributedprocessing;see Rumelhart& McClelland, 1986)
challenge the assumptionthat implicitlinguisticknowledgeis rule
based.
The claim,then,thatimplicitknowledgeis bothformulaicand ab-
stract/rulebased has considerablepedigree,although,likemanytheo-
accepted.Now,Robinson's
reticalissuesin our field,itis notuniversally
objectionsto thischaracterization of implicitknowledgeappear to rest
on the factthat I have not specifiedthe natureof the relationship

168 TESOL QUARTERLY

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
betweenformulaicknowledgeand abstract/rule-based knowledge.He
argues that I am obligatedto discuss this in termsof distinctions
proposedbyLogan and Reber.In fact,I tooka consciousdecisionnot
to enterintoanydiscussionof therelationship betweenformulaicand
rule-basedknowledge,notbecause I do notconsiderthisimportant-
it clearlyis-but because I wishedto maintaina focuson the primary
distinction.
explicit/implicit In a singlepaper,it is simplynot possible
to deal fullywitheveryissue raisedby a theory.
In my view, the relationshipbetween formulaicand rule-based
knowledgeneeds to be consideredin a furtherelucidationof implicit
knowledgeand how thisdevelops. There are, in fact,a numberof
possiblepositionsregardingthisrelationship.One is thatformulaic
knowledgeand rule-basedknowledgeare distinctand thattheformer
playsno partin thedevelopmentof thelatter(see Krashen& Scarcella,
1978). Anotheris thatformulaicknowledgeis graduallyunpackedand
fed intothe developingrule system(see WongFillmore,1976). These
and otherpositionsare reviewedin R. Ellis (1994).
To sum up, I have argued thatthe claimthatimplicitknowledge
consistsof formulaicchunksand abstract/rule-based representations
is fullywarrantedby the existingliterature.I have acknowledged,
however,the need for a fulldiscussionof the relationshipbetween
these two typesof implicitknowledge.

THE RELATIONSHIP BETWEEN EXPLICIT

AND IMPLICIT KNOWLEDGE

The main focusof myoriginalarticlewas the relationshipbetween

explicitand implicitknowledge.Unfortunately, Robinsonmisrepre-
sentsmyposition.He characterizes on
myposition L2 developmentas
"the processwherebyexplicitdeclarativeknowledgebecomesimplicit
proceduralknowledge."WhilethetheoryI have proposeddoes allow
forsuch an interface,providingcertainstringentconditionsare met,
it also advancesa numberof otherhypothesesregardingthe role of
explicitknowledgein the developmentof implicitknowledge.These
hypothesesare clearlystatedon page 98 of the originalarticleand
are reiteratedin summaryformhere:

1. Explicitknowledgecan be used to monitoroutput,which,in turn,

servesas a source of input.
2. Explicitknowledgecan help learnersnoticefeaturesin the input
and understand the meanings they realize.

THE FORUM 169

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
3. Explicitknowledgecan help learnerscomparetheirexistingrepre-
sentationof a grammaticalfeaturewiththatactuallyobservedin
the input.

Robinsonignoresthesehypotheses, preferring insteadtoconsideronly

the claim thatimplicitand explicitknowledgecan interface.
In fact,though,my proposalsfora structuralsyllabusreston the
hypotheseswhichRobinsonfailstoconsider,noton theinterfaceclaim.
Indeed, as Robinsonpointsout,I advancea numberof reasonswhythe
interface claimcannotserveas a basisforteachingimplicitknowledge.I
then go on to suggestthata structuralsyllabusmightprove useful
in two otherways,both relatedto the idea of consciousness-raising:
(a) helpinglearnersnoticeand comprehendgrammaticalfeaturesin
the input and (b) helpinglearnersdevelop explicitknowledgeof a
grammaticalfeature(which,indirectly, mayfacilitatethe subsequent
of
acquisition implicitknowledge). It is these proposalsthatlie at the
centerof myarticleand whichI had hoped tosee discussed.Robinson's
failureto considerthemleads me to concludethathe has missedthe
mainpointof thearticle-thata structural syllabusshouldnotbe used
to teach deplorableimplicitknowledge'butmightserveas a meansfor
contributing, partiallyand indirectly,to itsacquisition.

THE RELATIONSHIP BETWEEN THEORY, RESEARCH,

AND LANGUAGE PEDAGOGY
Finally,I would like to challenge Robinson'simplicitview of the
relationshipbetweentheory,research,and language pedagogy.Rob-
inson argues that

proposalsforpedagogy between
basedon distinctions andexplicit
implicit
knowledge andwhich aim to the
specify pedagogicconditions
whereby one
canbe convertedintotheothershouldbe motivated research
byempirical
intothe processthathas takenplace in cognitivepsychology. ... In fact,
thatexistsin the
research
proposalsforpedagogybasedon theempirical
area are probably
premature.

The underlyingassumptionappears to be this:Researchersdo re-

search,and onlywhentheyhave arrivedat sufficiently robustconclu-
a
sions (i.e., theory), can proposalsfor pedagogy be advanced. How-
ever,this is an assumptionthatmany would not agree with;it is based
on a particularconceptionof teachingand teacherdevelopment,one
thatFreemanand Richards(1993), followingZahorik(1986), charac-
based." An alternativeconceptionis to view
terize as "scientifically
teaching as an "art/craft."Whereas a scientificallybased conception of

170 TESOL QUARTERLY

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
teachingemphasizesthe importanceof basing pedagogic proposals
on theresultsof empiricalresearch,an art/craft
conceptionemphasizes
the importanceof teachersexploringand evaluatingtheirown teach-
ing throughinnovationand self-observation. Now, Robinson'sview
thatpedagogicproposals should awaitdefinitive
researchis compatible
witha scientificallybased conceptionof teachingbut not withan art/
craftview. Indeed, in an art/craftview,theoryor researchfindings,
no matterhow definitive, do not providea warrantfor pedagogic
practice.At best,theycan onlyprovidethe teacherwithideas which
have to be triedout in actualclassroomcontexts.
My own preferenceis for an art/craft view of teaching.In such
a view,the relationshipbetweentheory,research,and pedagogy is
somewhatdifferent fromthatassumedby Robinson.Theoryand re-
searchserveas one (butnottheonly)sourceof information and ideas
on whichteacherscan drawin makingdecisionsabout whatand how
to teach. These decisionsmust then be examined in the course of
actual teachingby teachersthemselves.One way in whichthis can
take place is throughaction research(see Nunan, 1990). In such a
conception,itis notnecessaryto waituntilresearchershave definitive
resultsor until a theoryis widelyaccepted. Teachers, on the basis
of their own intuitionand experience,may choose to select ideas
promulgatedby theoristsand experimentwiththem in theirclass-
rooms.The resultsof theirexperimentswilldeterminethe statusof
the proposals.The resultsmayalso contribute to the developmentof
the theorybecause teachingcan informresearchjust as researchin-
formsteaching.PerhapsRobinsonneeds to givefullerthoughtto the
relationshipbetweentheory,research,and pedagogy.
AlthoughI see everyneed to subjecta theoryto empiricaltesting,
I see no need to wait untilhypothesesare "motivatedby empirical
research"(as itis notclear to me whatthismeans)beforeteachersare
allowed to considerthem. I thereforerejectRobinson'scriticism.I
would also pointout thatmyown viewson the relationshipbetween
theory,research,and pedagogywere statedin the originalarticle-
albeit briefly-inthe finalparagraph.However,Robinsondoes not
appear to have consideredmyview of the relationshipbetweenre-
search/theory and teachereducation.

REFERENCES
Bialystok,E. (1978). A theoreticalmodel of second language learning.Language
Learning,26, 69-84.
Bialystok,E. (1981). The role of linguisticknowledge in second language use.
Studiesin SecondLanguageAcquisition, 4, 31-45.
Bialystok,E. (1982). On therelationshipbetweenknowingand usingforms.Applied
Linguistics,3, 181-206.

THE FORUM 171

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions
Bialystok,E., & Sharwood Smith,M. (1985). Interlanguageis not a stateof mind:
An evaluation of the constructforsecond language acquisition.AppliedLinguis-
tics,6, 101-117.
Ellis, N. (1993). Rules and instancesin foreignlanguage learning: Interactionsof
explicitand implicitknowledge.EuropeanJournalofCognitive 5, 289-
Psychology,
318.
Ellis, R. (1994). Thestudyofsecondlanguageacquisition. Oxford: Oxford University
Press.
Freeman, D., & Richards,J. C. (1993). Conceptionsof teachingand the education
of second language teachers. TESOL Quarterly, 27(2), 193-216.
Krashen, S., & Scarcella, R. (1978). On routinesand patternsin second language
acquisition and performance.LanguageLearning,28, 283-300.
McLaughlin, B. (1990). "Conscious" vs. "unconscious"learning.TESOL Quarterly,
24(4), 617-634.
Nunan, D. (1990). Action research in the language classroom. In J. C. Richards
& D. Nunan (Eds.), Secondlanguageteachereducation(pp. 62-81). Cambridge:
Cambridge UniversityPress.
Rumelhart,D., & McClelland, J. (1986). On learning the past tenses of English
verbs. In J. McClelland, D. Rumelhart,& the PDP Research Group (Eds.),
Parallel distributed Explorations
processing: in themicrostructure
of cognition:Vol.2.
Psychological and biologicalmodels(pp. 216-271). Cambridge,MA: MIT Press.
Wong Fillmore,L. (1976). The secondtimearound:Cognitiveand socialstrategies in
secondlanguage acquisition.Unpublished doctoral dissertation,Stanford Uni-
versity.
Zahorik,J. (1986). Acquiring teachingskills.Journalof TeacherEducation,27, 21-
25.

Teaching Issues
The TESOL Quarterly publishes brief commentarieson aspects of English
language teaching. For this issue, we asked two educators to discuss how
theoriesof language and cognitioncan informESL teaching.

Edited by BONNY NORTON PEIRCE

oftheWitwatersrandl
University
for Studiesin Education
OntarioInstitute

Language, Cognition, and ESL Literacy

and ESL LiteracyTeaching
Vygotsky
COURTNEY B. CAZDEN
Harvard GraduateSchoolof Education

MIn her book AppropriatingLiteracy:Writingand Reading in Englishas a

SecondLanguage, Rodby (1992) develops a "frameworkforESL literacy

172 TESOL QUARTERLY

This content downloaded from 188.72.127.69 on Fri, 20 Jun 2014 22:38:51 PM

All use subject to JSTOR Terms and Conditions