-vlso published In

Oxford Handbooks for Language Teac h e rs

Teaching American English Pro nu ncia rio n The Oxfo rd ESOL H an d boo k
Peter Avery and Susan Ehrlich Philida Schellekens

Des igning and 'Analyz ing Language T~srs Exploring Learner Language
Nathan T Carr Elaine Tarone and Bonnie Sunerzbin

ESOL: A Critical Gu ide Teach ing the Pronunciat io n afE ng lish

Melan ie Cooke and[ames Simpson

Success in English Teaching

as a Lingua Franca
Robin Walka How Languages

Paul Davies and Eric Pearse

Doing Seco nd Lan g uage Research

Doing Task-based Teac h ing
Jane 'X'illis and D.we \'V'illi s are Learned

[ames Dean Broum and Theodore S. F.vdgen Exp laining Eng lish G ramma r
Ceo~fJ" rule Fourth edition
From Exp eri en ce [0 Knowledge
[uit.m Edge .tna )lIt ";'II'ron

Teac hing Bus.n ess English

.\L lrk Ellts 'lJld Chn.rincfot-nron

Intercultural Business Co rnrn u n icarion Patsy M Lightbown andN ina Spada

Robar Gibson

Teaching and Learning in the Lan guage

Classroom
Tricia Hedge

reach ing Second Langu;lge Reading

Tbom Hudson

Teachi ng English Overseas: An I nrroducr ion

Smldm La McKay

Teac hi ng En glish as a n l rue rnarlo nal La n guage :'

Sandra Let AfcK.IJ

Comm u nic ario n in the Language Classroom

Tony Lynch

Teac h ing Second L.Jr1guage Listening

Tof/.v (ync!;

Teaching Youn g Langu.ige Learners

A n n.lnUln,1 PItJt~r
OXFORD
VN IV ERS I TY P R ESS
OXFORD
U"'IVER.Sfn l'RES S
Great Clarendo n Str eet , Oxfo rd. OX2 6DP.
United Kingdom
Oxford Un ive rsity Press is a department of the
Uni ve rsity of Oxford. It furthers the University's
objective of exce llence in researc h. scholarship.
and ed ucatio n by pub lish ing wo rldwide . Oxford
is a regist e red tr ade ma rk o f Oxford Uni ve rs ity
Press in the UKand in cert a in other countri es
~ Oxford University Pres s 2013
The mo ra l rig hts of th e author have been asserted
Firs t publis hed in 2013
2017 2016 20 1 5 20 1 4 2013
10 9 8 7 6 5 4 3 2 I
All righ ts reserved. No part of this pu blicatio n
may be re prod uced. stored in a re trieval sys te m .
or tra ns rrurted . in any form or by an y means.
without the prior pe rm issio n in wri ting of Oxfo rd
Un ive rsity Press. o r as expressly permitted by
law , by licence or under term s agr e ed with th e
ap pro p riate re progr a ph ics rights o rgan iza tio n .
Enquiries concerning re prod uct io n o utside the
scope o f th e above should be sent to the ELI
Rights Depart m e nt. Oxfo rd Un iver sity Press .
at the add re ss above
You m us t no t circu late this wo rk in an y other
form and you m us t impose this same conditi on
on any acquirer
Links to thi rd party we bsite s are provide d by
Oxfor d in good faith and fo r in formar ion on ly,
Oxfo rd disclaims any respons ibility for th e
materials con tained in any third party web site
re fere nced in thi s wo r k
Photocopying
The Pu blisher grants permission for the
photocopying of those pages ma rket!
' photocopi able ' acco rd ing to the fo Uo wing
conditions . Ind ividual pur chasers ma y make
copies for the ir own use or for use by classes
that th ey teach. School purchasers m ay m ake
copies for use by staff and stu den ts. but th is
permission d oes no t exte nd to additiona l
schoo ls or branches
Under no circu mstan ces may an y pa rt:of th is
book be pho tocopied fo r resale
JSB ~ : 9780 19 454 106 8
Prin ted in China
Th is boo k is printed on paper fro m certified
and we ll-managed sou rces .
ACKNOWLEDG EM liNTS
TIlt' lJuthol''S una pubJisht!T ..rrc grateful to thost"who how:
gtven p ernussion to reproduce tile follOWing exrrcrts ana
adapranons of capynghr ma<enaJ : p.17 Extract from
Language Development and Language Disorders by
Lois Bloo m an d Margare t Lahey (1978), Macmi llan
Pub lis hers: p,47 Figure fro m 'So me issues relating
to the Mo nitor Mode r by Stephe n Krashen , On
TESOL (1977). Repri nted by perm ission "fTESOL
Int e rn at ion a l Associa tion : pA9 Extract from
'Constru ctin g an acquisi tion-bas ed proce dure
for se co nd language assessme n t ' by Manfred
Pienema nn . Malcolm Jo hn st on . and Geoff Brindley To the teachers and students from whom
in Scudies in Second language AcqUisition. Volume 10/2.
pp.117-43 (19881. Repr od uced by permiss ion of we have lear ne d so much
Cambridge Uni versity Press: p.53 Extract fro m
'Spee ding up acq ui siti on of hIS (her. Exp licit LI/U
co n trac ts hel p ' in Second Langu age AcquiSition and
the Younger Learner: Child's Play? by Joanna Wh ite
120081pp .193 -1 28. With kind permission of John
Benjam in s Pu blishi ng Com pa ny. Arnsterd arm
Philadelp hia : p.54 Extrac t from 'Second language
instruction doe s rnake a difference' by Catherine
Do ug hty In Studk s ill Second Language AL'quisitiun ,
Vo lum e 13{4. pp.431- 69 (1991). Reprod uced by
per rmssion of Cambridge Un iversity Press; p.136
Reprinted fro m Inlernational Journalof Educational
K<s<arch. Volume 37 by Merrill Swai n and Sharon
Lap kin 'Talki ng it through: two Fre nch immersio n
learn ers' res po nse to refo rm ulati ons ' pp.28S- 304
12002) with permission from Elsevi er: p.l39 Extract
fro m 'Co rr ective feedback and le arn e r u ptake '
by Roy Lyst er and l eila Ran ta in Scudies in S<eond
Language ,\cquisltion. Vo lu me 19{1 pp .37 - 66 11997).
Reproduced by permiss ion of Cam bridge
Univer sity Press .
Ca rr oons by: Sophie Gri ller to Oxfo rd University
Press 1993 . 2005 , and 20 12.
.
,; _.
CONTENTS

Acknowledgements xi
Preface to the fourth edition xiii

Introduction
..,
Before we begin ...
Language learning in early childhood 5
Preview 5
First lan gu age acquisitio n "i
The first th ree years: Milestones and developmental sequences b
The p re-schoo l years I'::
The school years 13
Explai ni ng first lang uage acq uisi cion 1-1
The beha viour isr per spe ctive 15
The innatisr per spective 20
Inreracrionisr/developrnenral per specti ves l'!
Language di sorders and dela ys 29
C hild hood bilingualism 30
Summary 33
Suggestio ns for further read ing 34
,
2 Second language learning .3 5
Prev iew 35
Learner characteristics .36
Learning conditions . 38
Studying the langu age of seco nd la nguage learners 40
Contrastive analysis. error analysis. and inre rlanguage 41
Developmental sequences 45
More about first language influence 57
Vocabulary 60
Pragma tics 65
Phonology 68
Sampling learners' language -..,
Sum ma ry -.,
Suggesrions lor further reading - .3
viii Contents Contents IX

3 In dividual differences in second language learning 7 5

5 Observing learning and teaching in the second language classroom 1_J

)~

Preview 7 5
Preview 123

Research on learner cha racteristics 77

Natural and instru ctio nal sett ings P_J'

Intellige nce 79
In natural acq uisition settings 124

Language learning aptitude 80

In structure-based instructional settings 126

Learning sty les 83

In co mm unicative instr uctio nal sett ings 127

Personali ty 84
Observatio n sche mes 129

Atti tudes and motivation 87

C lassroom comparisons: Teacher- student interactions 129

Mo tiva tio n in the classroom 88

C lassroom comparisons: Studenr-scuden t interactions 13 5

Identity and ethnic group affiliation 89

. Corrective feedba ck in rhe classroom 139

Learner beliefs 90
Q uestio ns in the classroom 14 5

Ind ivid ual di fference s and classroo m instruction 92

Erh nograp hy 149

Age and second language learn ing 92

Summary 151

The critical period: More than JUSt pro nunciarion? 94

Suggestions t~r further reading 152

Intuitions of grammaricality 95

Rate oflearning 96
6 S~cond language learning in the classroom 153

Age and seco nd language instructio n 96

99
Preview 153

Summary
Suggesrions for further reading 100
Proposals for teaching 153

1 Get it righ r from the beg inning 154

2 jus t listen . . , and read 159

4 Explaining second language learning 103

3 Lee's talk 16 5

Preview 103
4 Get rwo for one 171

The beh aviourisr perspective 103

5 Teach what is reachable 177

Second language ap plicarions : Mimicry and me morization 103

6 Ge r it righ r in the end 182

The innatisr perspective 104

Assessing the prop osals 194

Second language app lications: Krashe n's 'Mcn iror Model' 106
Sum m ary 197

The cogn itive perspective 108

Suggest ions for further reading 198

Info rm ation processing l 08

Usage-based learn ing 110

7 Popular ideas about language learning revisited 201

The competition model I II

Language and the brain 113

Preview 201

Second lang uage ap plications: In teracting, no tici ng,

Reflecting on rhe popular ideas: Learning from research 201

processing, and practising 113

Concl usion ;. , 212

The sociocu lrural perspective 118

Second language applications: Learn ing by talking 1 19

Glossary 213

Summary 120
Bibliography 22 7

Suggestions fo r fu rther readi ng 12 1

Index 249

 ACKNOWLEDGEMENTS

We wish first to thank the readers who responded so pos it ively to the earl ier
editions of this book. W ith each edition. we have benefited fro m sugges­
tio ns and feedback offered by colleagues and studen ts. Our thanks to Ahlem
Ant mar. Alexander Ary, Philippa Bell. Luz Celaya. Laura Collins. Maria
Frohlich , Randall H al ter. Z haohong Han. Marlise Horst. Jim Hu, Phillip
Hubbard, Youjin Kim , Roy Lyster. Alison Mackey, Kim McDono ugh. Shawn
Loewen , Paul Meara, Imrna M iralpeix, Vicki M urphy, Carmen Munoz, H elke
Neumann . Howard Nicholas. Paul Quinn. Katherine Reh ner, Mela Sarkar,
Raquel Serrano, Younghee Sheen. Wa taru Suzuki. and Yasuyo Tomita . Leila
Ranta. and Jude Rand made essent ial con tr ibu tio ns to th e first ed ition .
At Oxford Unive rsity Press. we owe a de b t to H en ry Wi d dowso n for h is
early enco urage ment and to C ristina Whitecross, who was o ur editor for
the first three ed itions. We are grateful to Cather ine Kneafsey, Julia Bell,
H azel Gearches, an d An n Hu nte r who have wo rked with us thro ugh the
development o f this new edition. We thank the English Speaki ng Union tor
conferring the 1993 Duke of Edinburgh book prize for Applied Linguistics
on the book.

l~ .. r

II
PREFACE TO THE

FOURTH EDITION

How Languages Are Learned (H LAL) started ou t as a series of p rofessio nal

develo pment workshops for teachers in Q uebec, C an ada, where we bo th
worked for many years. Three editions of the book have now travelled far
from those origins. When we were working on the first edition in the 1980s
an d 1990s we were still in the early days of rem arkable growth of research in
seco nd language acquisition. In updating th e research for each new ed itio n,
the decisions abo ut what to include have grown more difficult. Keepi ng the
book to a reasonable length has often meant choosing between classics in
the field and im po rtant new st udies, of which th ere are now so many. In th is
edition, we have annotated some 'Suggestions for further reading' ar the end
of each chapter. We encourage readers to follow these readi ngs and the refer ­
ence list to deepen their understanding of topics that we can only introduce
here.
In this four th edition of HLAL, we have ad ded 'Questio ns for reflectio n' at
th e end of each cha pter, an d we have included some.new 'Acnvirles' thar give
readers opportunities to explore some of the topics. Another new feature
of this edition is a companion website which contains additional activi­
ties, read ings, and other we b-based material an d resources to enhance yo ur
reading and understanding of the contents of thebook. It will also provide
oppo rtu nities for readers to interact wi th others ~nd to share th eir ideas for
re~ching and learning lan gu ages. .. ~ ,:
The website for How Languages are Learned can be accessed at
www.oup.com/eltlreacherlhlal. !..
We are cu rrently working on a new series o f boJks for reache rs, the Oxford
Key Conceptsfor the Language Classroom . Each volume, written by a d ifferent
author, will focus on a specific topic (such as assess'ffi'ent, content-based lan­
guage reaching, literacy, and oral interaction), reviewing the relevant research
and linking the findings to classroom practice. We hope tha cthe books in this
series will encourage teachers to continue learning about some of the topics
that are introd uced in HLAL.
We hope that both new readers and those who have read the previous edi­
tions ofHLAL will find ideas and in formation that will challenge and inspire
them to make the ir own contributions to second language learning, reach­
ing, and research .

Patsy M. Lighrbown, Harwich, MA, U SA

Nina Spada, Toronto, ON, Canada
INTRODUCTION

When new methods and textbooks for second and foreign language teach ­
ing are int rod uced . th ey are often said to be based on the latest research
in psychology, linguistics, or pedagogy. Teachers are told that the y will be
more effective than those that have gone before. In many cases, the new
approaches are prescribed for immediate implementation in a school o r
region. Sometimes, the new materials co me with opportunities for extensive
training in their implementation. Sometimes, they are simply ordered and
distributed to teachers who have co do their best to use them effectively.
Many approach es to language teaching have been proposed and imple­
mented . O ne approach req uires stu den ts co learn rules of grammar and lists
ofvocabulary to use in translating literary texts. Another emp hasizes th e
value of havin g st udents im ita te and p ractise a set of co rrec t sente nces a nd
me mo rize entire di alogues . Yet ano the r encourages 'natural' comm unication
be twee n stude nts as the y en gage cooperati vely in tasks or projects wh ile using
the new language. In some classrooms, the second language is used as the
medium to teach subject matter, with the assumption that the language itself
will be learned incidentally as students focus on the ' ~cademic content.
How are teachers to evaluate the potential effectiveness of diffe rent inst ruc­
tional practices? To be sure, the mos t important influence on teachers'
de cisions is their own experience with previ ou s successes or disappointments ,
as 'well as the ir understanding of th e needs and abilities of their students.
We believe that ideas drawn from research and theory in second language
acq uisitio n are also valuable in helping reachers-co evaluate claims made by
proponents ofvarious language teaching methods. The goal of this book is to
introduce teachers-both novice and experienced-c-eo so m e of the language
acq uisition researc h th at m ay help th em nor on lfm evalua te existing text­
books and materials but also to adap t them in ways that are more consistent
with ou r understanding of how langu ages are learned.
The book begins with a chapter on language learning in early childhood.
This background is im portant because both second language research and
second language teac hing have been infl ue nced by our understandi ng o f how
children acquire their first language. Several theo ries about first language
(l 1) learn ing are presented in th is chapter and they are revisited later in the
book in relation to second language (l 2) lear ning.
2 Introduction Introduction 3

In C ha p te r 2 we look at second lan gu age learn ers' developing knowledge, learning and teaching. Th ink about whether you agree or disagree with each
th eir abili ry to use th at knowledg e, an d how th is co mpares with L1 learning. op inio n. Keep th ese sraternents and yo ur reactions to them in mind as yo u
In Chapter 3, we rum ou r att entio n to ho w indi vidual learner characteristics read about current research and the o ry in seco nd lan guage learn ing.
may affect success. In C hapter 4, several theories that have been advan ced
to explain second lan guage learning are pr esented and discussed . Chapter
5 begins wi th a co mparison of natural and instructio nal environ ments for ACTIV ITY Give you r opinion on. t hese statements
seco nd lan guage lear nin g. We then examine so me d ifferent ways in which Indicate t he extent to which you agree wit h each statement by marking an X
researchers have observed an d described teaching and learn ing pra ctices in in the box associated with your o pinio n:
second language classrooms. SA-strongly agree
In Chapter 6 , we examine six proposals that have been mad e for second A-agree somewhat
lan guage teaching. Examples of research related to each of th e proposals are D-disagree somewhat
pr esented, leading to a disc ussion of the evidence available for assessing th eir SD-strongly disagree
effecti veness. The chapter ends with a discuss ion of wh at research findi ngs SA A 0 SO
suggest abo u t the most effective ways to teach an d learn a second language
in th e classroom. I Languages are learn ed mainly through imitation.

In C hap ter 7, we will provid e a general sum mary o f the book by looking at 2 Par ents usually correct youn g children w he n th ey

how research can inform our response to some 'popularopinions' abou t lan­ make grammatical errors.

r -
guage learning an d teach ing that are introduced belo w. 3 Highly inte llige nt people are good language

learners.

A Gl ossary pro vides a qui ck reference for a number of terms char ma y be new
o r have specific technical meanings in the context of language acq uisition 4 The mos t important predictor of success in

research. Gl ossary words are shown in bo ld letters where th ey first appear in second language acquisition is motivation .

the text. For readers who would like to find ou t more , an annotated list of
,
5 The earlie r a second language is introduced in
"

suggestions for further readi ng is included at th e end of each ch apter. Th e school programmes, the greater the likelihood of
Bibliography pro vides full refere nce info rm atio n for the suggeste d readi ngs success in learning.

and all the works th at are referred to in the text.

6 Most of th e mistakes that second language

We have tri ed to present th e information in a way that does no t assume that learners make are due t o interference from. the ir

readers are already famil iar with research method s or the oretical issues in , first language .

seco nd language learn ing. Exam ples and case studies are include d th rough­
7 The best way to learn new vocabulary is thro~gh

o ut the book to illustrate the research ideas. M an y of th e exam ples are tak en read ing. ~

from second language classrooms. W e ha ve also in clu ded a number of activi­

ties for read ers to practise some of th e techniques of o bservation an d anal ysis 8 It is essential for learners to be able to

used in the research that we review in this book. At the end of each ch apter pro no unce all the individual sounds in the ~~.J

are 'Q uestio ns for reflection' to help read ers consolida te an d expand th eir second language .

understanding of th e material. 9 Once learners know 1.000 words and the basic

structure of a language, they can ea sily participate

in conversations with native speakers.

Before we begin ...

It is pr ob abl y true, as some have claimed, th at m ost of us teach as we were
taug ht or in a way that match es our ideas an d p references about how we learn .
Take a moment to reflect on your views abou t how languages are learn ed and
what you th ink this means abou t how they should be taugh t. The sta tements
in the activity below sum m arize some po pular opi nions abo u t language
10 Teachers should present grammatical rules one at
a time. and learners sho uld practise examples of
each o ne before go ing on to ano t he r.
II Teachers should teach simple language structures
before com plex ones.
4 Introduction

12 Learners' errors should be corrected as soon as

they are made in order to prevent the formation
of bad habits. LANGUAGE LEARNING

13 Teachers should use materials that expose

students only to language Structures they have
IN EARLY CHILDHOOD

already been taught.

14 When learners are allowed to interact freely (for
example, in group or pair activities) , they copy
each other's mistakes.
15 Students learn what they are taught.
16 Teachers should respond to students ' errors by
co rrectly rephrasing what they have said rather
than by explicitly pointing out the error.
Preview
17 Students can learn both language and academic
con te nt (for exa mple, science and histo r y) In this chapter, we will look briefly at the langu age develo pm ent of young
simultane o usly in classes w here th e subject children. We wi ll then consider several theories that have been offered as
matter is taught in the ir second language . explan atio ns for how language is learned. There is-an immense amo u n t of
research on child language. Although much of this research has been done
18 Classrooms are good places to learn about in m iddle-class North American and European familie s. there is a rich body
language but not for learn ing ho w to use language .
of cross-linguistic and cross-c ultural research as well. Our purpose in this
Pho tocopiable © Oxford University Pre ss cha p ter is to touch on a few main points in this research, p rimar ily as a p repa­
ratio n for th e discussion of seco nd lan guage acquisitio n (SLA), which is the
focus of thi s book. "

First language acquisition

Lan gu age acquisition is o ne of th e most im p ressive and fascinating aspects
of human development. We listen with pleasure to the sounds made by a
th ree-mo nth- old baby. We laugh an d 'answer' the co nversa tio nal ' ba-ba-ba'
ba bbling of olde r babies, and we share in the pr ide and joy of parents w hose
one-year-o ld has u ttered the first 'bye-bye'. Indeed, learning a language is an
amazin g feat-{)ne th at has attracted th e at tention ~Jin guists and psycholo­
gists for generatio ns. H ow do children acco m plish cl1is?What en ables a child
no t only to learn words, but to put th em togerher in meaningful sentences?
What push es child ren to go on developing complex grammatical language
even though their earl y simple com m unication is succ essful for mo st pur­
poses? Does child language develop sim ilarly around th e wo rld? How do
bilingual ch ildren acquire mo re than o ne language?
6 Language learning in early ch ildhood
Thefirst three years: Milestones and
'baby fall down'. These sentences are sometimes called 'telegraphic' because
developmental sequences
recogn ize them as sentences because, even though function words and gram­
One remarkable thing abo ut first language acquisition is the high degree of
similarity in the early lan guage of children allover the world. Researchers
have described developmental sequences for man y aspects of first lan guage
acquisitio n. The earliest vocalizations are simply the involun tary crying that
babies do when they are hungry or uncomfortable. Soon, however, we hear
the cooi ng and gurgling sounds of contented babies, lying in their beds
looking at fascinating shapes and movement around them. Even tho ugh
they have little control over the sounds they make in these early weeks of
life, lnfanrs are able co hear subcle differences betwee n the sounds of h uman
languages . Not only do they distinguish the voice of the ir mothers from
those of other speakers, they also seem to recognize the language that was
spoke n around their mother before they were born. Furthermore, in cleverly
designed experimenrs, researchers have demonstrated that tiny babies are
capab le ofvery fine auditory discrimination. For.example, they can hear the
difference between sounds as similar as 'pa' and 'ba'.
Janet Werker, Patricia Kuhl, and others have used new technologies tha t allow
us to see how sensitive infan rs are to speech sounds. What may seem even
more remarkable is that infants StOP making d istinctions between sounds
that are not phonemic in the language that is spo ken around them. For
example, by the time they are a year old, babies who will become speakers of
Arabic scop reacting to the difference betwee n 'pa' and 'ba' which is not pho­
nemic in Arabic. Babies who regularly hear more than o ne language in their
environment contin ue to respond to these differe nces for a longe r period
(Werker, Weikum, and Yoshida 2006) . One important finding is that it is not
enough for babies to hear language sounds from electronic devices. In order
to learn-or retain -the ability to distinguish between sounds, they need
to interact with a human speaker (Conboy and Kuhl 201l). The -Inrernet
abounds with remarkable videos of infams reacting to language sounds.
Whether they are becoming mo nolingual o r bilingual children, however,
it will be many months before their own vocalizations begin to reflect the
characteristics of the language or languages they hear and lo nger still before
they connect language sounds with specific meaning. However, by the end of
thei r first year, most babies understand quite a few frequently repea ted words
in the language.or lan guages spo ken around them. They wave when someo ne
says 'bye- bye' ; they clap whe n someo ne says 'pat-a-cake'; th ey eagerly hu rry
to the kitche n when 'juice and coo kies' are mentio ned.
At 12 months, most babies will have begun to produce a word or two that
everyone recognizes. By the age of two, mo st children reliably prod uce at
least 50 differ ent words and some produce many mo re. About this time, they
begin to combine words into simple sentences such as ' Mo mmy juice' and
Language learning in early childhood 7
they leave out such things as articles, prepositions, and auxiliary verbs. We
matical morphemes are missing, the word order reflects the word order of
the lan guage they are hearing and the co m bined words have a meaningful
relationship that makes them mo re than just a list of words . Thus, for an
English -speaki ng child , 'kiss baby' does no t mean the same thi ng as 'baby
kiss'. Remarkably, we also see evide nce, even in these early sentences that
children are doing more than imperfectly imi tating what they have heard.
Their two- and three-word sentences show signs that they can creatively
combine words. For example, 'more outside' may mean ' I want to go ourside
again.' Depending on the situation, 'Daddy uh-oh' migh t mean 'Daddy fell
down' or ' D addy dropped something' or even ' Daddy, please do tha t fun ny
thing where you pretend to d rop me off your lap.'
As ch ild ren progress through the discovery of language in their first three
years, there are predictable patterns in the emergence and development of
many features of the language they are learning. For some language featu res,
these patterns have been described in terms of developmenral sequences or
'stages'. To some extent, these stages in language acquisi tion are related to
chil dren's cognitive development. For example, chi ldren do not use temporal
adverbs such as 'tomo rrow' or 'last week' until they develop some under­
standing of time. In other cases, the develo pmental s€quences seem to reflect
the grad ual acquisi tio n of the linguistic e1emenrs fbi- expressing ideas that
have been present in children's cognitive understanding for a long time. For
example, children can distinguish between sing ular and plu ral long before
they reliably add plural endings to nouns. Correct use of irregular plurals
(such as 'feet') takes even more time and may not be'co rnp lerely under control
unuil the school years. .
Grammatical morphemes «
In the 196005, several researchers focused on how children acquire grammati­
cal morphemes in Eng lish. One of the best-known stud ies was carried o ur by
Roger Brown an d his colleagues and students. In a lqngitudinal study of the
language development of three ch ild ren (called Adam, Eve. and Sarah) they
found that 14 grammatical mo rphemes were acquired in a sim ilar sequence.
The list below (adapted from Brow n's 1973 book) shows some of the mo r­
phemes they studied.
presenr progressive -ing (Mommy running)
p lural -s (rwo boo ks)
irregular past forms (Baby went)
possessive -s (D addy's hat)
co pul a (Mommy is happy)
articles the and a
8 Langu age learnmg in early childhood
regular past -ed (she walkzz)
third person singular sim ple present -s (she runs )
auxil iary be (he is coming)
Brown and his colleagues found that a child who had mastered the gram­
matical morp hemes at th e bottom of the list had also mastered chose at the
cop. bu t the reverse was not tru e. Th us, mere was evidence fo r a 'develop men­
tal seq uence' or order of acquisition. However, the children did not acquire
the morphemes at me same age or rate . Eve had mastered nearly all me mor­
phemes before she was two-and-a-halfyears old . while Sarah and Adam were
still working on them when they were three-and-a-half or four.
Brown's longitudinal work was confirmed in a cross-sectional study of 2 1
child ren. Jill and Peter de Villiers (197 3) found tha t children who correctly
used the morphemes mat Adam, Eve, and Sarah had acquired late were also
able to use the ones mat Adam, Eve, and Sarah had acquired ear lier. The chil­
dren mastered me morphemes at different ages, JUSt as Adam, Eve. and Sarah
had d one, but the o rder of their acquisition was very sim ilar.
Many hypothes es have been advanced co explain why these gramm atical
mo rp hemes are acq u ired in the observed order. Researchers have stu died me
frequency with wh ich the morphemes occu r in parents' speech, the cognitive
co m plexiry of the meanings represented by each mo rphe me, and me difficulty
of perceiving or pronouncing them. In me end , mere has been no sim ple satis­
factory explanation for me sequence, and most researchers agree that me order
is determined by an interaction among a number ofdifferent factors .
To su pplement the evidence we have from sim p ly observing children , some
carefully designed procedures have been developed to further exp lo re chil­
dren's knowledge ofgrammatical morphemes. One ofthe first and best known
is me so-called 'wug test ' developed by Jean Berko Gleason (1958) . In this
'test' , children are shown drawings of imaginary creatures wich novel names
o r people performing mysterious acti ons. For example. they are told, ' H ere is
a wug. Now mere are rwo of them. There arc rwo _ _' or ' H ere is a man who
knows how to bod. Yesterday he did the Same thing. Yesterday, he _ _ '. By
co m p leti ng these sentences with 'wugs' and 'be dd ed' , children demonstrate
that they know the patterns for plural and simple past in English. By gener­
alizing these patterns to words they have never heard before, they show that
their language is more man JUSt a list of memorized word pairs such as ' boo k!
books' and 'nod i nodded'.
ACTIVITY Tryout the 'wug' test
A web search for 'wug test' will turn up many examples of the pictures and the
rext created for this landmark research. If you know some English-speaking
children unde r the age of five years, try using the test with them.
Language learning in earlv childhood 9
What similarities and diffe rences do you notice among the children at
different ages!
2 Which gra mmatical morphemes do they find easy and which ones are
more difficult!
The acquisition ofother language features also shows how child ren's language
develops systematically, and ho w th ey go beyond what they have heard to
create new forms and structures.
Negation
Children learn the functions of negation very early. Th at is, th ey learn to
comm ent o n m e disappearance ofobj ects, to refuse a suggestion , or to reject
an assertion, even at me single word stage . However, as Lois Bloom's (199 1)
longitudinal studies show, even though children understand th ese func ­
tions and express them with single words and gesrures, it takes so me time
before th ey can exp ress chern in sentences, using the appropr iate wo rds and
word orde r. The follow ing stages in' me develo pment o f negatio n have been
observed in th e acq uisi tion of En glish . Similar stages have been observed in
other lan guages as well (\Vode 1981).
~~1 ,
Negation is usually expressed by the word 'no', eith er all alone or as the first
word in me utterance.
No . No cookie. No co mb hair.
Stage 2
Utteranc es grow longer and me sent ence subject ma y be included . The neg­
ative word appears JUSt before the verb. Sentences expressing rejection or
prohibitio n often use 'do n't'.
'.'
Daddy no co m b hair. Don't touch mat! i
Stage 3 ~_,
The negative element is inserted into a more complex sentence. Children
may add forms of me negative other than 'no', including words like 'can't'
and 'do n't'. These sentences appear ro follow the correct English pattern of
attaching the negative to me auxiliary or modal verb. However, children do
not yet vary these forms for different persons or tenses.
I can' t do it. He don't want it.
Stage 4
Children begin to attach th e negative elem ent to th e correct form ofaux iliary
verbs such as 'd o' and 'be'.
 Languagelearning in early childhood 11
10 Language Learning in early childhood

You didn't have supper. She doesn't want it. Stage 1

Children's earliest questions are Single words or simple two- or thr ee-word

Even though the ir language system is by now quite complex, they may still sentences with rising intonation:

have difficulty with some other features related to negatives.

Cookie? Mommy book?
I don't have no more candies .
Ar the same time, they may produce some correc t questions-correct because

Questions they have been learned as ch unks:

The challenge oflear ning complex language systems is also illustrated in th e Where's Daddy? What's that?
developmental stages through which children learn to ask questions.
Stage 2

There is a remarkable consistency in the way ch ildre n learn to fo rm ques­ As they begin to ask more new questions, children use the word order of the

tions in English. For one thi ng, there is a predictable order in which the declarative sentence, with rising intonation.

' wh- words' emerge (Bloo m 1991) . 'What' is generally the first wh- question
word to be used. It is often learned as part of a chunk ('Whassat?') and it is You like this? I have some?
some time before the child learns that there are variations of the form, such They continue to produce the correct ch unk-learned forms such as 'What's

as 'What is tha t?' and 'What are these?' . that?' alongside their own created questio ns.

'Where' and 'who' eme rge very soo n. Identifying and locating peo ple and
. Stage 3
objects are within the ch ild 's understanding of the world. Furthermore, Gradually, children notice that the structure of questions is different and
adults tend to ask children just these types of questions in the early days of begin to produce questions such as:
language learning, for example, 'Where's Mommy?' or 'Who's that? '
Can I go?

'Why' emerges around the end of the secon d year and becomes a favouri te Are you happy?

for the next year or two. Children seem to ask an endless number ofquestions
beginning with 'why', having discovered how effectively this lit tle word gets Although some ques tions at this stage match the adnlt pattern, they may be

adults to engage in conversation, for example, 'Why tha t lady has blue hair?' right for the wrong reason. To describe this, we need !co see the pattern from

the child's perspective rather than from the perspective of the ad ult grammar.

Finally, when the child has a better understanding of manner and time, 'how' We call this stage 'fro nt ing' because the child's ruleseems to be that questions

and 'when' emerge. In contrast to 'what' , 'where', an d 'who' questions, ch il­ are formed by putting some thing (a verb or q uestio n word) at the 'front ' of a

dren sometimes ask th e more cognitively dlfficult 'why', 'when', and 'how' sencence, leaving the rest of the sentence in its sraternent form .

questions without understanding the answers they get, as the following con­
versation with a four-year-old clearly shows. Is the teddy is tired ? Do I can have a cookie?

Why you don't have one? Why you catched it? ',~'

CHIL D When can we go outside?

PARENT In about five minutes.

Stage 4

C H I LD 1-2-3-4-5! Can we go now?

At Stage 4, some questio ns are formed by subject';'jHlXi.liary inversion. The

The ability to use these question words is at least pardy tied to children's cog ­ questions resemble those of Stage 3, but there is more variety in the auxilia ­

nitive development. It is also predicted in part by the questions children are ries that appear before the subject.
asked and the linguistic complexity of questions with different wh- words. Are you going to play with me?
Thus it does not seem surprising that there is consi stency in the sequence of
At this stage, children can even add 'do ' in ques tions in whic h there wo uld be

their acquisition. Perhaps more surprising is the consistency in the acquis i­

no auxiliary in the declarative version of the sentence.

tion of word order in questions. This develo pment is not based on learning
new mean ings, but rather on learni ng different linguistic patterns to express Do dogs like ice cream?
meanings that are already understood. Even at th is stage, howeve r, childre n seem able to use either inversion or a uih­
word, but not both (for exam ple, 'Is he crying?' bu t not 'Why is he crying? '

12 Language Learning in early childhood
Th erefore. we may find in version in yes/no questions but no t in wh- qu es­
dons, unle ss they are formulaic units such as 'What's that?'
Stage 5
At Stage 5, both wh- and yes/no qu est ion s are formed correctly.
Are these yo ur boo ts?
Why did you do that?
Does Daddy have a box ?
Negative questions may still be a bit too difficult.
Why the teddy bear can't go outside?
And even though performance on most questions is correct, there is still one
mo re hurdle. When wh - words appear in subo rdi nate clauses or embedded
questions, child ren overgeneralize the inverted form that would be correct
for sim ple questions and produce sentences suc h as:
Ask hi m why can't he go ou t.
Stage 6
Ar th is stage, childre n are able to cor rectly form all q uestion types, incl uding
nega tive and complex embedded q uestions.
Passage through developmental seq ue nces d oes no t always follo w a steady
uninterrupted path. Children appear to learn new things and then fall back
on old parrerns when there is added stress in a new situatio n or when they
are using other new elements in their language. But th e overall path tak es
them toward a closer and closer approximation of the language tha t is spoken
aro und th em .
The pre-schoolyears
By the age of four, most child ren can ask qu estions, give com ma nds, report
real event s, and create stories about im aginary on es, using correcr word order
and grammadcal m arkers most of th e time. In fact , it is generally accepted
that by age four, children have acquired the basic structures of the language
o r languages spoken to them in these early years. Three- and [our-year- olds
co ntin ue to learn vocabulary at the rate ofseveral words a day. They begin to
acq uire less frequent and more complex linguistic structures suc h as passives
and relative clauses .
M uch of child ren's language acquisition effort in the late pre- school years is
spent in developing their ability to use language in a widening social environ­
ment. They use language in a greater variety ofsituations. They interact more
often with unfamiliar adults. They begin to talk sensibly on the telephone
to invisible grandparents (yo unger children do not understand that their
telephone parmer can no t see whar they see). They acquire the aggressive or
Language learning in early childhood 13
cajoling lan gu age th at is needed to defend their toys in the playground. Th ey
show that they have learned the difference between how adults talk to babies
and how they talk to each other, and they use this knowledge in elaborate
pretend play in which they practise using these different 'voices'. In this way,
they explore and begin to understand how and why language varies.
In the pre-school years; children also begin. to develop metalinguistic aware­
ness, the ability to treat language as an object separate from the meaning it
conveys. Three-year-old children can tell you that it's 'silly' to say 'drink the
chair', because it doesn't make sense. However, although they would never
say 'cake the eat', they are less sure that there's anything wrong with it. They
may show that they know it's a bit odd, but they will focus mainly on the fact
that they can understand what it means. Flve-year-olds, on the other hand,
know that 'd rin k the chair' is wrong in a different way from 'cake the eat ' .
They can tell you that one is 'silly' but the other is 'the wrong way around'.
Language acquisition in the . pre-school years is impressive. It is also note­
wor thy th at child ren have spent th ou sands of hours int eract ing with
lan gu age- participatin g in co nversatio ns. eavesdro pping o n ot hers' con­
versations, bein g read to , watc hing television , etc. A quick mathematic al
exercise wil l show you just how many hours childre n spend in lan guage-r ich
enviro nme nts. If child ren are awake to r ten o r twelve hours a day, we may
or
estim ate that the y are in contact with the language their environment tor
20,000 hours or more by the time they go to school.,
Although pre-school children acquire complex knowledge and skills for
language and language use, th e school setting requires ne w ways of using
language and brings new opportunities for language development.
The schoolyears
Children develop the abili ty to use language to understand ot hers and to
exp ress their own meanings in th e pr e-school years, and in the school years,
this ability expands and grows. Learning to read gives a major boost to meta­
linguistic awareness. Seeing words represented by l~lJ ers and o the r symbo ls
on a page leads children to a new understanding that language has form as
well as meaning. Reading reinforces the understanding that a 'word' is sepa­
rate from the thing it represents. Unlike three-year-olds, children who can
read undersrand that 'the' is a word , JUSt as 'house' is. They understand that
'caterpillar' is a longer word than 'train' , even though the object it represents
is substan tially short er! Meralinguisric awareness also includes the discovery
of such things as amb iguity. Knowing that words and sentences can have
multiple meaning gives children access to word jokes, tri ck questions, and
riddles, which the y love to share with their friends and family.
14 Language Learning in early childhood
One of the mos t imp ressive aspects of language developmen t in the school
years is the astoni shing growth ofvocabulary. C h ild ren enter school with the
ability to understand and produce several tho usan d wo rds, and thousands
more will be learned at school. In both the spoken and written language at
school, words such as 'homewo rk' or 'r uler' appear freq uently in situations
where their mean ing is either immediately or gradually revealed . Wo rds like
'po pulatio n' or 'latitude' occur less frequen tly, but they are made impo rtan t
by their significance in academic subjec t matt er.
Vocabulary grows at a rate of between several hundred and more than a
thousand words a year, depending mainly on how much and how widely
children read (Nagy, Herman, and Anderson 19 85). The kind of vocabulary
growth required for school success is likely to' come from bo th reading for
assignments and reading for pleasure, whether narrative or non-fiction. Dee
Gardner (2004) suggests that reading a variety oftext types is an essential part
of vocabulary growth . His research has shown ho w the range of vocab ulary
in narrative texts ·is different from that in non -fiction . There are wo rds in
non-fiction texts that are unlikely to occur in sto ries or novels. In addi tion,
non-fiction tends to include more opport unities to see a word in its different
forms (fo r exam ple, 'm ummy', 'm ummies' , 'mummified'). The importance
of reading for vocabulary growth is seen when observant parents report a
child usin g a new word but mispronouncing it in a way tha t reveals it has
been encountere~ only in wri rren form .
Another important development in the school years is the acquisition of dif­
ferent language registers . Children learn how written lan guage differs from
spoken language, how the language used to speak to the principal is different
from the language of the playgro und, how the language of a scien ce repo rt is
different from the lan guage of a narrative. As Terry Piper (2006) and o thers
have doc umented, some children will have even mo re to learn if they come
to schoo l speaking an ethnic or regional variety of the school language that is
quite different from the one used by the teacher. They will have to learn tha t
another variety, often referred to as the standard variety, is requi red for suc­
cessful acad emic work. Other children arrive at school speaking a different
language altogether. For these children, the work oflanguage learni ng in th e
early school years presents additional opportunities and challenges. We will
retu rn to this topic when we discuss bilingualism in early childhood.
Explaining first language acquisition
These descriptions oflanguage development from infancy th rou gh the early
school years show that we have considerable kn owledge of what children
learn in their early lang uage development. More co ntroversial, however, are
questions abou t how this develop ment takes place . What abilities does the
child bring to the task and what are the contributio ns of the environment?
Language Learn ing in early childhood 15
Since the middle of the 20th century, th ree main theoretical position s have
been advanced to explain language development: behavlo urist, innatist, and
inte ractional/ developmental perspectives.
The behaviouristperspective
Behaviourism is a theory of lear ning that was influential in the 1940s and
1950s, especially in the United States. Wi th regard to language lear nin g,
the best-known proponent of this psychological theory was B. F. Skin ner
(1957). Traditional behaviourists hypothesized that when children imitated
the language pro duced by those around them, their attem pts to reproduce
what they heard received 'positive reinforcement'. This could take the fo rm
of praise or JUSt successful communication. Thus enco uraged by th eir envi­
ronment, children would continue to imi tate and practise these sounds and
.patterns until they formed 'habits' of correct language use. According to this
view, the quality and quantity of the language the ch ild hears, as well as
the con sistency of the rein forceme nt offered by others in the environment,
woul d shape the child's language behaviour. This theory gives great impor­
tance to the env ironment as the source ofeveryth ing che.child needs to learn.
Analysing children's speech: Definitions and examples
The behaviourists viewed imitation an d practice as the primary pro cesses in
language development. To clarif)rwhat is meant by these rwo ter ms, consider
the following defin itions and examples. J ,.
Imitation : wo rd-fo r-word repetition ofall or part ofsomeo ne else's utterance.
MOTHER Shall we play with th e dolls ?
LUCY Play with dolls
Prahice: repeti tive manipulatio n of form.
C IN D Y He.ear carrots. The oth er one eat carro ts. They both eat
"[
carrots. !
Now examine the tra nscrip ts from Peter, Ci ndy, and Kathryn. They were all
about 24 months old when th ey were recorded as ~y played with a visiting
ad ult. Using th e definitions above, notice how Peter im itat es th e adu lt in the
followin g dial ogue .
Peter (24 months) is playing with a dump truck wh ile two adults, Patsy and
Lois, look on .
PETER Get more .
LOIS You're gonna put more wheels in rhe du m p truck?
PETER Dump truck. Wheels. Dump tr uck.
(later)
PATSY What happened to it (the truck)?
 Ld nguage learning in early childhood J­
16 Language learning in early childhood
Cindy appears to be working hard on her language acquisition. She prac­

PET ER (loo king under chair for it) Lose it. Dump truck! Dump
tises new words and structures in a way that sounds like a student in some

truck! Fall! Fall!

foreign language classes! Perha ps most interesting is tha t she remembers the

LO [S Yes, the dump tru ck fell down .

' lan guage lesson' a week later and turns st raigh t to the page in the book she

PETER D ump truck fell down. D u m p truck.

had no t seen since Patsy's last visit, What is most striking is that, like Peter,

(Un published data from P. M. Lighcbown) her imitation and practice appear to be focused on what she is currently

If we analysed a larger sample of Peter's speech, we would see that 30-40 per 'wo rking on' .

cent of his sentences were imitations of what someone else had JUSt said. We The samples of speech from Peter and Cindy seem to lend some sup po rt to

would also see that his imitations were not random. That is, he did not simply the behaviourist explanation of language acquisition. Even so, as we saw,

imitate 30-40 per cent of everything he heard. Detailed analyses of larg e the cho ice of what to imitate and practise seemed determined by something

samples of Peter's speech over about a year showed that he imitated words inside the child rather than by the environment.

and sentence structures that were JUSt beginning to appear in his spontaneous
Not all children imitate and practise as much as Peter and Cindy did . The

speech. Once these new elements became solidly grounded in his language
amount of imitation in the speech of other children, wh ose development

system, he stopped imitating them and went on to imitate others.

proceeded at a ra te comparable to that of Cindy and Peter, has been cal­

Unlike a parrot who imitates the familiar and continues to repeat the same cula ted at less than 10 per cent. Consider the examples of imitation and

things again and again , children appear to imitate selectively. The choice practice in the following conversation between Kathryn and Lois.

of what to imitate seem s to be based on so meth ing new that they have just
Kathryn (24 months)

begun to understand and use, nor sim ply o n what is availabl e in th e enviro n­
ment. For example, consider how Cindy imitates and practises language in LO IS Did yo u see the toys I brought?

the followin g con versations. KAT H RY N I br ing toys? Choo choo ? Lois brought the cha o cho o train ?

LO IS Yes , Lois brought the choo choo train.

Cindy (24 months, 16 days) is looking at a picture of a carrot in a book and

KATH RYN (reaching for bag) I want play with choo choo train. I want play

trying to get Patsy 's attention.

with choo choo train. (taking out slide) Want play. What's this?
CINDY Kawo? kawo? kawo? kawo? kawo?
LO IS Oh you know what that is. "
PATSY What are the rabbits eating?
KATHRYN Put down on floor. This . I do this .
CINDY They eating ... kando?

(Kathryn putS the slide on the floor.)

PAT S Y No, that's a carrot.

C INDY Carrot. (po inting [Q each carrot on the page) The other ...
~TH RYN (taking out tWO cars of train) Do this . I want do this . (trying

carrot. The other carrot. The other carrot. to put train together) I do this. I do this .

LOIS OK. You can do it. You can do it. Look <
I'll show •vou how.
(A few minutes later, Cindy brings Patsy a stuffed toy rabbit.)
~

(Lo is putS it rogerher.)

PATSY What does this rabbit like to eat?
CIN DY (incom prehensible) eat the carrots. KATH RYN (searching in box) I get more. Get a msse . No more choo

choo train. Get truck. (taking out truck) Kathryn truck.

(Cindy gets another stuffed rabbir.)

Where? Where a more choo chao train?

CINDY He (incomprehensible) eat carrots. The other one eat carrots. LOIS Inside. It's in the box.

They both eat carro ts. KATH RYN A choo choo? (taking out parr of train) This is a choo choo

(One week later, Cindy opens the book to the same page .) trai n.

CINDY Here's the carrots. (pointing) Is that a carrot? (from Bloom and Lahey 1978: 135)
PATSY Yes. Like Cindy, Kathryn sometimes repeats herself or produces a series of related
practice sentences. but she rarely imitates the other speaker. Instead , she asks
(Unpublished data from P. M . Llghcbown)
and answers questions and elaborates on the other speaker's questions or
statements.
18 LanguageLearning in early childhood Language Learning in early childhood 19

Thus, children vary in me amount of imitation they do . In addi tio n, many
of me things they say show that they are using lan gu age creatively, not just:
repeating what they have heard . This is evident in the followi ng examples.
Patterns in language
Th e first example shows a child in me process of learning parrerns in lan­
so hard on performing me fascinating new gesture and me formulaic expres­
sion 'I'd like ro propose . . . ' mat he failed to realize mat the wo rd he though t
he kn ew- 'roast '- was not me same roast and could not be replace d wit h its
apparenc near-synon ym , 'a p iece of bread'.
Questio n formation

guage, in. th is case me rules of word fo rmatio n, an d overgene ralizing them Randall (2 years, 9 mo nths) asked m e foll owing questio ns in various situa­

to new contexts. tion s over m e course of a day.

Ran dall (36 months) had a sore on his han d. Are dogs can wiggle their tails?
MOTHER Ma ybe we need to take you to the doctor. Are those are my boots?
RANDALL Why? So he can doc my littl e bum p? Are this is ho t?
Ran dal l forms me verb 'doc' from the nou n 'docto r', by analogy with farme rs Randall had concluded m at me trick of asking questions was to p ut 'are' at
who farm , swi mme rs who swim, an d acto rs who act. me begin n ing of me sentence. H is questions are goo d examples of Stage 3 in
question development.

Order of events
Randall (3 years, 5 months) was loo king fo r a rowel.
You took all the rowels away because I can't dry my hands.
He meant 'I can't dry my hands because yo u w o k all m e towe ls away', b ut
he made a mistake abour which clause com es first. C hildre n at th is stage of

~)
language develop ment ten d to men tio n events in rp,e order o f th eir occu r­
ren ce. In th is case, me towels disa ppeared before Randall arternp red to dry
his hands , so char's wh at he said first. He d id no t vet.understand how a word
like 'before' or 'because' changes the order of ca~eind effecr.
/C:~~
'Si~oo~~;)
--/r­ These examples of chil d ren's spee ch provide us witha window on th e p rocess
of lan guage lear ning. Imi ta tion an d practice alone canno t exp lain some of
m~ forms created by chil dren . The y are not merely repetitions of sentences
that they have heard from ad ults. Rather, children'appear to pick out patterns
and generalize them to new contexts. They create new forms or new uses of
words. Th eir new sentences are usually comprehensible and often correcr.
Focus on meaning Behav iourism seems to offer a reasonable way o f un4,gstanding how children
Even olde r childr en have to wo rk out some puzzles , for example, whe n famil ­ learn some of the regular an d routine aspects of langifage, especially at m e ear­
iar language is used in unfamiliar ways, as in m e example below. \JVhen David liest stages. However, children who do lit tle overt imitation acquire language
(5 years, 1 m o nt h) was at his older sister's birthday parry, toasts were pro­ as fully and rapidly as those who imitate a lor. And although behaviourism
posed with grape juice in stemmed glasses: goes so me way to explaining me SOrts of overgeneralization mat childre n
make, classical behav iourism is not a satisfactory exp lana tion fo r me acquis i­
FATHER I'd like to p rop ose a to ast.
tion of m e more complex grammar th at chil d ren acquire. These lim itations
Several minutes late r, David raised his glass: led researchers to look for d ifferent explanations for lang uage acq uisition.
DAV ID I'd like to p rop ose a piece of b read.
O nly when laug h ter sent D avid slinking from the table did m e group realize
mat he wasn't intenti o nally making a play on words! H e was concen trating
20 Language learning in early childhood
7he innat~tperspective
Noam Chomsky is one of the mos t influential figu res in linguistics, and
his ideas about how language is acquired and ho w it is stored in the mind
sparked a revolution in many aspects of linguistics and psychology, includ­
ing the stu dy of lan guage acquisition . The innatist perspective is related to
C ho ms ky's hypothesis that all human languages are based o n some innate
universal principles.
In his 1959 review of B. F. Skinner's book Verbal Behavior, Chomsky chal­
lenged the behaviourist explanation for lan guage acquisition. He argued that
ch ildren are biologically programmed for language and that language devel­
ops in the child in JUSt the same way that other bio logical functions de velop .
For exam ple, every chi ld will learn to walk as long as adeq uate nourishment
and reasonable freedom of movement are provided. The child does not have
to be taught. Mo st childr en learn to walk at about the same age, and walking
is essentially the same in all normal human beings. For Chomsky, language
acq uisit ion is very simi lar. Th e enviro nme nt makes o nly a basic co ntri bu­
tion-in th is case. th e availab ility of peopl e w ho speak to the ch ild. The
child , o r rath er, the child's b iolog ical endowm ent , w ill do th e rest.
C ho msky"argued that the behaviourist theory failed to account for 'the logical
pro blem oflanguage acq uisitio n'- the fact th at child ren co me to kn ow mor e
abo ut th e structure of th eir language than they could reasonabl y be expected
to learn on the basis of the sam ples of language they hear. The language
children are exposed to includes false starts, incomplete sentences, and slip s
of the tongue, and yet they learn to distinguish between grammatical and
ungrammatical sentences. He co ncluded that chi ldren's minds are not blank
slates to be filled by imitating language they hear in the environment. In stead ,
he hypothesized, ch ild ren are born with a specific innate ability to discover
for themselves the underlying rules of a language system on the basis of the
samples of a natural language the y are exposed to . This innat e endowment
was seen as a SOrt of templat e, co nt ain ing the principles that are universal to
all human languages. This universal grammar (U G) would pr event the ch ild
from pursuing all SOrtS of wrong hypotheses about how language systems
might work. If ch ildren are pre-equipped with UG , then what they have to
learn is the ways in which the language they are acquiring makes use of these
princip les.
Consider the follow ing sentences, from a book by Lydia White (1989). These
Engli sh sentences contain the reflexive pronou n ' h imself' . Both the pronoun
and the noun it refers to (the antecedent) are printed in italics. (An aster isk
at the beginning o f a sentence ind icates tha t the sent ence is u ngrarn m atical.)
a John saw himself.
b "Him selfuuo J;hn.
Language learning in etzr/ychildhood 2l
In (a) an d (b), it loo ks as if th e reflexive pronoun mus t follow the no un it
refers to . But (c) disproves thi s:
c Looking after himse/fboresJohn.
If we co nsid er sent ences suc h as:
d John said th at Fred liked himself.
e "f ohn said that Fred liked himself.
f John told Bill to wash himself.
g «John told Bill to wash himself.
we might conclude that the noun closest to the reflexive pronoun is the anre­
cedent. However, (h) shows th at thi s rule won't work eith er:
h John promised Bill to was h himself.
And it's even more complicated than that. Usuall y the reflexive must be in
the same clause as the antecedent as in (a) and (d), but not always, as in (h) .
Fur the rmo re, th e reflexive can be in th e subject pos ition in (i) but no r in OJ.
i John believ es bimselfu: be intelligent (no n-fin ite clause).
j 'John bel ieves th at himself i s inte lligent (fin ite clause).
In so me cases, more than o ne antecedent is possib le, as in (k) wh ere th e
reflexive co uld refer to eithe r Joh n o r Bill:
k John ? showed Bill: a picture of himself.
When we loo k at thi s kind of complexity, it seems-it would be very hard
to lear n , and ch ildren d o make errors along the way. Yet, most schoo l-age
ch ildren would be able to co rrectly interpret the grammatical sent ences
and recognize the ungrammaticaliry o f the othe rs. Researchers who study
language acq uisitio n from the innatist perspective argue th at such com plex
gram m ar could never be lear n ed purely on the basis of imitating and practis­
ing sentences avai lable in the input. They hypothesize th at since all child ren
acquire the language of th eir environment, theY
must have some innate
me chanism or knowledge that allows them to discover such complex synt ax
in sp ite of limitations of the input. They hyp othesize furthermore that the
innate mechanism is used exclu sively for language acquisition.
The innatist perspective emphasizes the fact tha t almost all ch ildren success­
fully acquire their native lan guage- o r more than one lan guage if the y live
in a multilingual community. Children who are profoundly deaf will learn
sign lan guage if the y are exposed to it in infancy, and their progress in the
acquisition ofthat lan guage system is similar to hearing ch ildren's acquisition
ofspoken language. Even children with very limited cognitive abiliry develop
quite complex language systems if they are brought up in environments in
which people int eract with th em.
22 Language learning in early childhood
C hildr en acquire the basic syntax and morphology of the language spo ken to
them in a variery ofconditio ns, some of which wo uld be expected to enhance
language development (for example, caring, attentive parenrs who focus on
the child 's lan guage), and some which might be expected to inhibit it (for
exam ple, abusive or reject ing parents). Childre n achieve different levels of
vocabulary, creativity, social grace, and so on, bu t virrually all achieve the
abiliry to use the patterns of the language or languages spoken to them. Thi s
is seen as sup po rt for rhe hypo th esis tha t lan guage is somehow separa te from
other aspects of cognitive development and may depend on a specific module
of the brain.
The Critical Period Hypothesis
The innatist perspective is often linked to the C ritical Period Hypothesis
(C PH)-the hypo thesis that animals , including humans, are gene tically
pro grammed to acqui re certain kinds ofknowledge and skill ar specific times
in life. Beyond those 'critical periods', it is either difficult or impossible to
acquire those abilities. Wi th regard to lan guage, the CPH suggests that chil­
d ren who are not given access to lan guage in infan cy and early childhood
(because ofdeafness o r extreme isolation) will never acquire language if these
deprivation s go on for roo long.
It is difficul t to find evidence for or against the CPH, since nearly all ch ild ren
are exposed ro language at an early age. However, history has do cumented a
few 'natu ral experiments' where children have been deprived ofco nt act with
language. Two of the most famous cases are those of 'Victo r' and 'G en ie'.
In 1799, a boy who became known as Victor was found wandering naked
in the woods in France . His Story was dram atized in a 1970 film by Fran cois
Truffaut called L'enfantsauvage ( The Wild Child) . When Victor was cap tured,
he was about 12 years old and completely wild, apparently having had no
cont act with humans. Jean -Marc-G aspard hard, a you ng doctor accu stomed
to working with deaf children , devo ted five years to socializing Victor and
trying to teach him language. Although he succeeded to some extent in devel­
oping Victor's sociability, memory, and judgement, there was littl e progress
in his language abiliry.
Nearly 200 years later, Genie, a 13- year-old girl who had been isolated,
neglecte d, and abused, was discovered in California. Because of the irrational
demands of a disturbed father an d the su bmission and fear of an abused
m o ther, Ge nie. had spent more than 11 years tied to a chair o r a crib in a
small, darkened room. Her father had forbidden his wife and son to speak to
Genie and had himself only growled and barke d at her. She was beaten when
she made any kind of noise, and she had lon g since resorted to complete
silence. Ge nie was undeveloped ph ysically, emotionally, and intellectually.
She had no language.
Language learning in early childhood 23
After she was discovered . Ge n ie was cared for and ed ucated with the par ­
ticipation of many teache rs and therap ists, including Susan C urtiss (1977).
After a brief period in a rehabilitation centre, she lived in a foster home
and attended special schools. Genie made remarkable progress in becoming
socialized and cogni tively aware . She developed deep personal relatio nships
and strong individual tastes and traits, Neve rtheless , after five years of expo­
sure to language, Genie's language was no t like'th at ofa typic al five-year old.
There was a larger th an normal gap between com prehension and prod uc­
tion. She used grammatical forms inco nsistently and overused formulaic and
routine speech.
Although Victor and Ge nie appear to p rovide evidence in support of the
CPH, it is difficult to arg ue th at the hyp o thesis is confirmed on the basis
of evidence from such unusual cases. We can not know wha t other factor s
besides biological ma ruriry mig ht have contributed to their ina biliry to learn
lang uage. It is no t possible to de ter m ine whether eithe r of them suffered
from brain damage, developmental delays, or a specific language impair­
ment , even before they were separated from normal human int eractio n.
A more approp riate test of the CPH is the case of child ren who co me from
homes where they receive love and care from thei r parents, yet do not have
access to language at the usual time. This is the case for some profoundly
deaf children who have hearing parents. Only 5-~ 0 per cen t of the pro­
foun dly deaf are born to deaf parenrs, and only these children are likely to be
exposed to ASL from birth. Hearing parenrs may not realize that their ch ild
cannot hear because the child uses other senses to in teract in an apparently
nor mal way. Thus, the early childhood period may be normal in most ways
but devoid of lan guage that is accessible to the child . These children's later
exr.erience in learn ing sign language has been the.sub ject of some impo rtant
research related to the CPH.
Like oral and wri tten languages, American Sign pnguage (ASL) makes use
of grammatical markers to indicate such things-as time (for example, past
tense) and nu mber. Th ese markers are exp ressed thro ugh specific hand or
body move me nts . ~.:
Elissa Newport (1990) and her colleagues srudied the abiliry of deaf users
of ASL to produce and comprehend grammatical markers. They compared
Native signers (who were exposed to ASL from birth), Early signers (who
began using ASL between fo ur and six years of age), and Late signers (who
began lear ning ASL after age 12). They foun d no differe nce between the
groups in some aspecrs of their use ofASL, for example in vocabulary knowl­
edge. However, on rests focusing on grammatical markers, the Native grou p
used the markers more consistently th an the Early group who, in tu rn, used
them more consisten tly than the Late group. The researchers co nclud ed that
24 Language learning in early childhood
thei r study sup po rrs the hypothesis mat m ere is a critical period for first lan ­
guage acquisition, whether mat language is oral or gestural.
Anomer line of research mat has given new insight into me importance of
early language experience comes from stu d ies of ' intern atio nal ado prees.'
These are children who were adopted at an early age by Familieswho did
not speak me language me child had heard during infancy. In their review
o f studies of international adoptees, [ohanne Paradis, Fred Genesee, and
Martha Crago (20 11) concluded that cognitive and linguistic outcomes were
generally very positive. Some com parisons of their language with that of
children me same age who had always heard me same lan gu age showed that
su btle differences persist even after several years, but these are not me kinds
of differences that most people would notice. Here again, of course, one
can no t know whether so meth ing o th er man a late exposure to the language
spoken in me adoptive environment also contributed to differences between
th ese children and others who did not experience an abrupt change in their
language en vironment. Nevertheless, with continuing researchon child ren's
ling uistic behaviours an d intui tion s. as well as m e neu rological stud ies of
in Fanr s' speec h pe rcepti on that we saw above, it is becom ing clearer that
langu age acq uisition begi ns at bi rth, an d poss ibly even befo re, as me ch ild 's
b rain is shaped by expo sure to m e lan gu age(s) in the environment.
The in natist perspecti ve is thus partly based on evid ence that m ere is a criti­
cal period for language acquisition. It is also seen as an explanation for 'the
logical problem of language acquisition', that is, me question of how adult
speak ers come to know me complex structure of their first language on the
basis of the limited sam ples of language to which they are exposed.
Interactionistldevelopmentalperspectives
Developmental and cognitive psycho logists have focused on the interplay
betwe en me inn ate learning ability of children and the environment in which
th ey develop. The y argue that me innatists place too much emphasis on me
'final state' (me competence ofadult native speakers) and nor enough on me
developmental aspects oflanguage acquisition. In their view, language acquisi­
rion is but on e example of the human child's ability to learn from experience,
and they see no need to assume mat there are specific brain structures devoted
to language acquisition. They hypothesize that what children need to know is
essentiall y available in the language they are exposed to as they hear it used in
thousands o f hours of interactions with the people and objects around memo
Psychologists artribu re considerably more importance to the environment
than the innarists do even though they also recognize a powerful learning
mechanism in the human brain. They see language acquisition as similar
to and influenced by the acquisition of othe r kinds of skill and knowledge,
rather man as something that is different from and largely independent of
Language learn ing in early childhood 25
the child's experience an d cog nitive develop ment. Ind eed, researchers such
as Dan Siobin ( 197 3) have long emphasized me close relationship between
ch ildren's cognitive development and their acquisition of language.
Piaget and Vygotsky
One of th e earliest proponents of th e view mat children's language is built
on their cogn itive development was me Swiss psychologisr/eplsrernologist,
Jean Piager (19 5 1). In me early decades of me 20th century, Piaget observed
infants and child ren in their play and in th eir interaction with obj ects and
people, He was able to trace the development of the ir cognitive understand­
ing of such things as object permanence (kn owin g that things hidden from
sigh t are still there) , me stability of quantities regardless of changes in th eir
appearan ce (kno wing that 10 pennies spread Out to form a long line are
not more numerous man 10 pennies in a tightly sque ezed line), and logi cal
inferencing (figuring OUt which pr operties o f a set of rod s (their size, weigh t.
mat erial , erc.) cause so me rod s to sin k and Others to float o n water).
It is easy to see how chi ld ren's cog nitive develop me nt wo uld par tly deter­
mi ne how the y acq uire lan gu age. Fo r examp le, the use of certain ter ms suc h
as 'b igger ' o r 'more' depends o n the childre n's unde rsta nd ing of the co n­
cep rs rhey represe nt . The developi ng cog nitive u nde rstanding is bu ilt on the
interacti on between th e child and the th ing s that can be o bserved o r manipu­
lated. For Piager, language was one o f a number of.~ymbol systems that are
developed in childhood. Language can be used to represent knowledge that
child ren have acquired through ph ysical interaction 'with the environment.
Anothe r influential student of child development'was the psychologist Lev
Vygorsky ( 197 8). He observed interactions among children and also between
child ren and adults in schools in me Soviet U nio n in the 1920 s and 1930s.
He concluded that language develops primarilyfrorn social interaction. H e
argued that in a supportive interacti ve environm ent, children are able to
advance to higher levels of knowledge and perfJrmance. Vygotsky referred
to a metaphorical place in which children could do mor e than they would be
capable ofdoing independently as the zone ofprox.iQ;tJ.1 development (Z PD ).
Vygo tsky observed the importance of con versations that child ren have with
adults and wirh other children and saw in these conversations the origins of
both language and thought. The conversations provide the child with scaf­
folding, that is, a kind ofsupportive structure that helps them make the most
of me knowledge the y have and also to acquire new knowledge.
Vygotsky 's view differs from Piager 's. Piaget saw language as a symbol system
that could be used to express knowledge acquired through interaction with
the physical world. For Vygorsky, thought was essentially internalized speech,
and speech emerged in social interaction. Vygotsky's views have become
increasingly central in research on second language development, as we will
see in Chap ter 4 .
 Langu age learni ng in early childhood 27
26 Language learning in early childhood

Cross-cultural research Researchers working in a 'lan guage socialization' framework have found
that the kind of child-directed spe ech observed in m iddle-class American
Since the 1970s, researchers have stu died children's language learning envi ­
homes is by no means universal. In some societies, adults do nor engage in
ronments in a great many differem cultural communities. The research has
conversation or verbal play with very yo un g children. For example, Bambi
focused not only on the development oflanguage itself, but also on the ways
Schieffelin (1990) found that Kaluli mothers in Papua New Guinea did not
in which the env ironment provides what children need for language acquisi­
consider their children co be appro priate co nversatio nal partners. Martha
tion . Betwee n 1985 and 1'997, Dan Slobin edited five volumes devoted co
Crago (1992) observed that in tradi tio nal In uit society, chil dre n are expected
research on the acquisit io n of28 lan gu ages, provid ing exam ples an d an alyses
to watch and listen to adults. Th ey are no t expec ted or en couraged to par­
ofchild lang uage an d th e language-learning enviro nm ent fro m com m un ities
ticipate in conversations with adults until th ey are older and have more
around the worl d. One of the most remarkable resources for ch ild lan guag e
developed language skills.
researchers is the C hild Language D ata Excha nge System (CHILDES) ,
where researchers have contributed child language data in dozens of lan­ Other researchers have observed that in some societies, young children inter­
guages in recorded and transcribed forms that are available as electronic files act primarily with older siblings who serve as th eir caregive rs. Even wi thin
from the C H ILD ES website (M acWhin ney 2000) . the United States, Shi rley Brice Heath (1983) and oth ers have documented
subs tantial differences in the ways parents in differen t socioeconomic and
One feature of cross-cultural research is th e description of child-rearing pa t­
et hnic groups interact with their ch ildren. Neve rtheless, in every society,
terns. Ca the rine Snow (199 5) and others have studi ed th e apparem effects
child ren are in situations in which they hear language that is meaningful
on language acq uisition of the ways in which adults talk to and interact
to them in their enviro nment. And they acquire the community language.
with young children. In middle-class North American ho mes, researchers
Thus, it is diffic ult to judge the long-term effect of the modifications that
observed that adults often modify the way they speak when talking co lime
some adults make in speech add ressed to chil dren. .
children. This child-directed speech may be characterized by a slower rate of
delivery, higher pitch, m o re varied intonation, shorter, simpler sentence pat­
terns, stress on keywords, frequen t repetition, and paraph rase. Furthermore,
The importance of interaction
copies of conversation emphasize the child 's immediate environment, picture The role of interaction between a language-learnin&'child and an interlocu­
books, or experiences that the adult knows the child has ha d. Adults often tor who responds co the child is illuminated by caseswhere such int eractio n
repeat the content of a ch ild's utterance, but they expand or recast it into a is missing. Jacqueline Sachs and her co lleagues (19$"1) studied the lan guage
grammatically correct sentence. For example, when Peter says, ' D um p tr uck! developmenr o f a child they called J im. He was a heari ng child 0 f deafparen ts,
Dump truck! Fall! Fall!', Lois responds, 'Yes, th e dump truck fell down. ' an d his only con tact with oral lan guage was th ro ugh television, which he
watched frequen tly. The family was un usual in. that the parents did not use
sigh language wi th Jim. Thus , altho ugh in other respec ts he was well cared
(~
MJ 3
0
p~-b.Je · '-.Ja~ She not-tk1e ~
CMs,'ttere.ti
for, Jim did not begin his linguistic developmen t in a normal en vironment
in which a parem com m unicated with him in either oral or sign language. A
~ effect5 of

hr:
SuC""­
~-.--J/), ) Li""':f-eA. / /A.ltJm..'.t~. t lca.( language assessment at three years an d nine months indicated tha t he was well
below age level in all aspects oflanguage. Although he attempted to express
f,v ~ i"~ ? ~ ~ ideas ap propriate "co his age, he used unusual, ungrammatical word order.
~?()-
. ,~ ~
~
o
~, When Jim began conversatio nal sessions with an adult, his expres sive abili­
ties began to improve. By the age of four year s and two mo nths most of the
un usual speech patterns had disappeared, replaced by language mo re typical
of his age. Jim's younger brother Glenn did not display the same type of
language delay. Glenn's lingu istic environment was different from Jim's: he
had his older b roth er-s-no r only as a model, but, mo re importantly as a con­
versati onal parmer whose interaction allowed Gle nn to develop language in
a more typical way.
28 L.mguage learning in (arly childhood
Jim showed very rapid acqu isition of English once he began co interact wi th an
ad ult on a one-co-one basis. The faCt mat he had f.U1ed coacquire language nor­
mally prior co this experience suggesrs mat impersonal sources oflanguage such
as television o r radio alo ne are nor sufficienr. One-co-one interaction gives chil­
dren access co language mat is adjusted co their level of comp rehensio n. When
a child does not understand. me adult may repeat o r paraphrase. Th e respon se
of me adult may also allow children co find our when their own urrerances are
understood. Television , for obvious reasons, does not provide such interaction.
Even in children's programmes, where simpler language is wed and topics are
relevant co younger viewers, no immediate adjustment is made for me needs
of an individual child. Once children have acquired some language, however,
television can be a source oflanguage and cultural information.
Usage-based learning
As more and more research has documented the ways in which children
inreracr with the environment, developmental and cognitive psychologists
find furth er evide nce that langua ge acq uisition is 'usage-based' . In thi s view,
lang uage acq uisitio n is possible because ofchild ren's gene ral cog nitive cap ac­
ities and th e vast n um ber o f op po rtuni ties they hav e co make co n nec tions
between the lan guage they hea r and wh at th ey expe rience in the ir env iro n­
men r. Sop hist icate d electro nic recording devices have been used co track
and count words and phrases children hear in their daily lives. Deb Roy
do cumenred his son's acquisition of words, showing the frequency and the
co nt exts for the occurrence of language. Most remarkable, perhaps, is the
demonstration of the power of interaction between the child and the adults
and how adults focus on the language me child has begun co use (Roy 2009).
The usage -based perspe ctive on language acquisition differs from the behav­
io urist view in that the emphasis is more on the child's abiliry co create
networks of associations rather than on processes of imitation and habit for­
mati on .·Referred co by var ious names, including cognitive linguistics, thi s
view also differs sharp ly from the lnnatists' because language acquisition is
not seen as requiring a separate 'mod ule of the mind' but rather depends on
th e child's general learning abilities and the contributions of the environ­
ment. As Elena Lieven and Michael Tomasello (2008) puc it, 'C hild ren learn
language from their language experiences-mere is no other way' (p.168) .
According to this view, what children need co know is essentially available to
them in the language they are exposed roo
So me of the early research in this framework was done in the context of con­
nectionism and involved compuce r simulations in which language samples
were provided as input to a fairly simple program. The goal was co show mat
the com p uter could ' learn' certain things if exposed to enough examples.
The program was found to be able to SOrt out the paceerns from the input
and even generalize beyond what it was actually exposed rooIt even made the
Language learning in early childhood 29
sam e kinds of creative 'm istakes' that child ren make, such as p u tting a regular
-ed ending on an irregular verb , for example, eared.
In a usage-based model, language acquisition involves not only associating
words with elemenrs ofexternal reality. It is also a process ofassociating words
and phrases with the other words and phrases chat oc cur with them , or words
with grammatical morphemes rhar oc cur with them. For example, children
learning languages in which nouns have grammarical gender learn co associ­
are the appropriate article and adjective forms with nou ns. So if children are
learning French, they learn mar La and une go with chaise (chair) and le and
un go with li vre (book). Similarly, they learn co associate pronouns with the
verb forms that mark person and number-il aim e (he likes) and nous aimons
(we like). They also learn which temporal adverbs go with which verb tenses.
Of particular importance co chis hypothesis is me fact that children are
exposed co many thousands of opportunities co learn words and phrases.
Learning takes place gradually, as the number of links between language and
meaning and am on g language form s are builr op . For usage-b ased th eori sts,
acq uisition of lan guage. whil e im p ressive, is not th e o nly rem arkable fear
acco mplished by th e ch ild . They co m pa re ir co o the r cogni rive and percep ­
rual learning, in clud ing learn ing co 'see' . Th at is, th e visua l abilities tha t we
take for gra nre d, for exam ple, focusin g o n and inrerpreting o bjects in o ur
visual field , are actually learned through experience.
Language disorders and delays
Although most children progress through the stages of language develop­
rnenr without significanr difficulty or delay, there are some children for
whom thi s is not the case. A discus sion of the various rypes of disabilities
(includ ing deafness, articulatory problems, autism, dyslex ia, and so on) mat
sometimes affect language development is outside me scope of this book. It is
essen tial mat parents and teachers be encou raged. to seek profe ssional advice
if they feel that a child is not developing language'norrnally, keeping in mind
that the range for 'normal' is wide indeed.
While most children produce recognizable first w~~ds by 12 months, some
may not speak before me age ofthree years. In very young children, one way to
determine whether delayed language reflects a problem or simply an individual
difference within the normal range is ro determine whether me child responds
co language and appears to understand even ifhe or she is nor speaking.
For older children , delays in learning co read that seem out of keeping with
a child's overall cognitive functioning may suggest that mere is a specific
problem in that domain . Some children seem to begin reading almost by
magic, discovering the mysteries of print with little direct instruction.
For most children, instruction that includes some systematic artenrion to
30 Language learning in early chi ldhood
sound-Ierrer correspondences allows them to unlock me treasure chest of
reading. Both groups fall within a normal range . For some children, however,
reading presents such great chal lenges that they need expert help beyond
what is available in a typical classroom.
Childhood bilingualism
Th e language development of chi ldren who lear n multiple languages during
childhood is of enormous importance throughout the world. Indeed, the
majority of the world's children are exposed to more than one language.
Some children learn m ultiple languages from earlies t childhood; others
acquire additional lang uages when they go to school. The acquisitio n and
maintenanc e of more than one language can open doors to many pe rsonal,
social, and economic opportunities.
Unfortunately, as Jim Cummins (2000) an d others have poi n ted out, ch il­
dren who already know one or mo re languages and who arrive ar their first
day of school without an age-app ropriate knowledge of the language of the
school have often been misdiagnosed as having lang uage delays or disorders.
Th is includes immigrant and minority language children who do nor speak
the school language at home and children who speak a different variety of
the school lan gu age. These children's knowledge of a different language or
language variety is often incorrectly interpreted as a lack of normal language
development and a lack of background knowledge for school subjects. They
may be placed in remedial or special ed ucatio n classes because schools are not
equipped ro provide an adequate assessme nt of children 's ability to use their
ho me language o r of their general cogn itive abilities or their knowledge of
school su bjects, learned through another language. Researchers have recen tly
mad e important pro gress in p rovid ing gu idelines that can help educators
distinguish between disability and diversity (Paradis, Genesee, and Crago
2011), but much practical work remains to be done so that children can
make the most of their cognitive and linguistic abilities.
Children who lear n more than one language from earliest ch ildhood are
referred to as 'simultaneous bilinguals', whereas those who lear n another lan ­
guage later may be called 'seq uential bilinguals'. We sometimes hear people
express the opinion that it is roo difficul t for children to cope with two lan­
guages. They fear tha t the children will be confused or w ill not learn either
language well. However, th ere is littl e su p po rt for the myth that lear ning
more tha n o ne lan guage in ear ly childhood is a problem for ch ild ren who
have adequate op porrunities to use each one. Th ere is a co nsiderable body
of research on children's ability co learn more than one language in thei r
earliest years. Al though some studies show minor early delays in one or both
lan guages for simultaneous bilinguals, th ere is no evidence that learning two
Language learning in early childhood 31
languages substantially slows down their linguistic development or interferes
with cognitive development.
Ind eed , many ch ildren attain high levels of proficiency in both lan guages.
Ellen Bialystok (2001) and other cognitive and developmental psychologists
have found convincing evidence m at achieving bilingual pro ficiency can
have positive effects on abili ties mat are related to academic success, such as
mecallnguisric awareness. Lim itations mat may be o bserved in me language
of bilingual individuals are more likely to be related to me circumstances in
which each language is learned man to any limitation in me human capacity
to learn more than one language. For example, ifone language is heard much
more often than the other or is more highly valued in me community, that
language may eventual ly be used better man, or in preference to, the other.
One aspect of bilingual language use is referred to as code switching-the use
ofwords or phrases from mo re than one lan guage wi thin a conversation. Fo r
example, a child who speaks bo th French an d English m ight say, 'I'm playing
with le cnduau' . Such switching between lan guages may sometimes reflect
the abse nce of a particular vocab ulary word or expression, but it can just as
often be th e intentional use of a word from the otherlanguage for a variety
ofin reracrio nal purposes. Highly proficient adul t bilinguals also code switch
when they speak to others who also know both languages. The use of both
languages within a bilingual context is not evidence of a lack of p roficien cy. It
may have many di fferent mo tivations, fro m expressJng solidarity ro making
a joke. Psychologists have shown that speakers of mo re th an one language
are constantly making choices abou t how ro express themselves and that
code switching is patterned and often predictable. Indeed, this experience
in making choices has been ide ntified as contributi ng to cognitive flexibility
rhro ughour life (Bialystok 2009).
,
As 'ch ild ren learn a second language at school, rhey need ro learn both the
variety oflanguage that children use among themselves (an d in informal set­
tings wi th familiar adults) and the variety that i5[used in academic settings.
In his ear ly research on childhood bili ngualis m, Jim Cummins called these
two varie ties BlCS (basic interpersonal co mmuni2tion skills) and CALP
(cognitive academic language proficiency). Charaderistics o f the two varie ­
ties overlap ro a certain extent, but there are important differences, not JUSt
in the range ofvocabulary that each requires but also in the way informatio n
is expressed. Mary Schlep pegrell (2004) and others have sought ro discover
what exactly it is that characterizes th ese varieties oflanguage and the interac­
tio n pa tterns that tend to go with th em, and some aspectS of the distinction
remain controversial. It is widely agreed, however, that the lan guage need ed
for academic discourse is mo re difficul t for ch ildren to acq uire than me info r­
mal language of day-to-day interaction (Cummins 2000).
32 Language learning in ear(y childhood Language learn ing in early childhood 33

Children entering school with little or no knowledge of me language spoken yer mastered me school language. and they have not continued to develop the

there may acq ui re BICS wirhin a relatively sho rt time- as lime as a year or family lan gu age. During me transi tio n period, they may fall behind in their

two . They learn from wa tching and imitating interactions among their peers academic learning . Unfo rtuna tely, me 'solutio n' educators sometimes pro pose

and between teachers and students. They make connections between fre­ co parents is that they should scop speaki ng me family language at home and

que ntly heard wo rds and ph rases and th e routines and recurring even ts of m e co ncentra te instead on speaking me scho ol lan guage with thei r children.

classroom, cafete ria, and playgro und. For th is reason , students are so metimes
The research evidence suggests that a better approach is co strive for addi­
perceived as ' fluent' in thei r seco nd language. This can lead teachers to assume
tive bilingualism-th e maintenan ce of the home language while the seco nd
mat any difficulti es in acade m ic tasks are nor du e to limi ted lan guage skills but
language is being learned. This is especially true if me parents are also lear n­
to othe r causes- fro m lack of mo tivation to learning disabilities. More care ful
ers of the seco nd lan gu age. If pare nts continue to use th e language mat they
observa tio n shows mar th e students, while fluent in social setti ngs, do not have
know best with th eir ch ildr en" th ey are ab le to exp ress th eir kn owledge and
me CALP skills needed for academ ic tasks such as understanding a p robl em in
ideas in ways m at are rich er an d mor e elabo rate man th ey can manage in a
marhemarics, defini ng a wo rd , o r writing a science report.
. language they do not know as well. Using their own lan guage in fam ily set ­
Virg inia Collier (1989) found that, for most students, acquiring age-appro­ tings is also a way for parents ro maintain their own self-esteem, especially as
priate CALP takes seve ral yea rs. As the seco nd language learner rries to catch they may have th eir own struggles with the new language outside th e home,
up, th e children who came to school already sp eaki ng the school lan gu age at work, or in me community. Maintaining me family language also allows
are continui ng to learn hundreds of new words every yea r and to learn the children co rerain family co n nectio ns with gran dparents or relatives who do
concep ts char these words rep resen t , If second language learn ers have limited not speak the new language. They benefit from me o ppo rt un ity to continue
knowledge of the schoo l language and do not have opportunities to continue both cognitive and affecti ve development using a language th ey understand
learn ing academic content in a language they already know. it is not surpris­ easily while they are st ill learning the second language. .
ing that they fall behind in learning the academic subjecr matter th at their
Other positive effects of bilingual or multilingual development go beyond

peers have continued to develop.

those that acc rue to the children and their families. Knowledge of more man

Children need time to develo p th eir seco nd lan gu age skills. M any people one language can also increase opportunities for cross-cultural communica­

assume that th is mea ns th at th e best app roac h is to start learn in g as early tion and econo mic cooperatio n amo ng people. As W~ have seen, develo ping

as possible and to avoid the use of me ch ild's p reviously learned languages. a seco nd language takes years. Bu t teachers. parents, and sru de nrs need

Certain ly, it is im portant for ch ildren to begin learn in g an d using th e school to know that the many benefits of additive bilingual ism w ill reward thei r

language as earl y as poss ible, but co nside ra b le research suggests th at co ntin­ patience and effort.

ued development of th e chil d's home lan gu age act ual ly co nt ribu tes in me
long term to m ore suc cessful acq uisitio n of th e scho ol language. Resear ch ers
an d ed uca to rs have exp ressed concern abo ut situatio ns whe re ch ild re n are
Summary
cut off fro m the ir fam ily lan gu age when they are very yo ung, spending long In this chapter we have focused on some of the.research on 'Children's early

h ours away from th eir fam ilies in sett ings whe re the ho me language is absent language develo pmen r th a t has inf!uenced research 0 n seco nd langu age acq ui­

or even forbidden . Lily \'V'on g Fillmore (2000) obs erved that when children sitio n. We have described three b roa d th eo retical perspect ives for exp laini ng

are 'su b me rged' in a di fferent language for long periods in pre-s chool or first language acquisition. In Chap te r 2 we will loojeac some o f th e find ings

day care, th eir development of the famil y language may be slowed down or of research examin ing th e de veloping lan guage of seco nd lang uage learners.

stal led b efore they have developed an age-app rop riate p roficiency in m e new
language. Event ual ly th ey may sto p speaki ng the fam ily language altogether,
and this loss of a common language can lead to sign ificant social an d psycho­
Questions for reflection
logical probl em s. Some researc h has found that the best predictor of children's voc abu lary
growth is the amount of language addressed to them by their parents and
Wallace Lambert (1987) called m e loss of one lan guage on me way to lear n­
other caregive rs .What have you seen in this chapter that is compatible w ith
ing anomer subtractive bilingualism. It can have negati ve consequences for
that flnding !
child ren's self-esteem , an d their relationships w ith fam ily m embers are also
likely to be affected by such early loss of the family language. In these cases, 2 Go to the children 's section of a library o r bookstore and loo k at the
ch ildre n seem to continue to be caugh t betw een two lan guages: they have not vocabulary used in books that are pub lished for children between three
34 Language learning in early childhood

and six years o ld.Compare these to books for young readers. aged six to
eight.What does this suggest about the impo rtance of continuing to read
SECOND LANGUAGE

2
to children after they have begun to learn to read at school? Finally. look at
the language used in textbooks for children at age 10 or I I.What can you
conclude about the challenge faced by English language learners entering LEARNING
school at this age?
3 Ifyou are or may be te aching a second language to a gro up of school­
aged learn ers with different first language backgrounds. can you think of
pedagog ical casks/activities in which children can display and use the ir LI
knowledge to help them learn the second language?

Suggestions for further reading

Berko Gleason,]. and N . Bernstein-Ratner (eds.) . 2009. The D evelopment
0/Language 7 th edn. N ew York: Allyn and Bacon.
Many o f th e chap ters by leading expercs in child language introduce
readers to the best-known findi ngs of the past 50 years of resear ch on ch il­
dren's language development. In addition, there are chapter s based on new
resear ch , using the kinds of technology that have only recendy become
available. Thus, th e rich database creat ed by researchers wi th no tepads,
tape recorders, and tools such as the 'wug test' is complem ented by stud ies
of the neurological bases oflanguage learning and language use.
Paradis, j ., F. Genesee, and M . B. Crago. 20 11. D ual Language
Development and Disorders: A Handbo ok on Bilingualism and Second
Language Learning 2nd edn . Balti more: Paul H . Brookes.
Th e authors describe language acquisition by child ren who learn more th an
one language sim ultaneously or sequentiall y, drawing on research from ed u­
catio n, psychology, and linguistics. They make th e research accessible by
their writing style, the inclusion of a glossary ofterms, and above all by relat­
ing the research to profil es of children who are acq uiring their languages in
a variety of home, school, and commun ity situations. The authors pro vide
insights into both normal and atypical multilingual development.
Pearson, B. Z. 200 8. Raising a Bilingual Child: A Step-by-Step Guidefo r
Parents. New York: Living Language (Rand om H ouse).
Addressing herself mainly to parents, Barbara Zurer Pearson (2008)
reviews resear ch from many stu di es and shows how ch ild ren beco me bilin­
gual in man y differ ent environ ments. She also emphasizes the advantages
of growing up wi th a kn owledge of more th an o ne language-from th e
evidence for cognitive flexibility to the benefits of cultural knowledge.
\'Q'ri rren in an ap proachable and humorous style , the text is sup po rted
by Zurer Pearson's thorough knowledge o f the research lit erature tha t is
included in the bibliograp hy.
Preview
In this ch ap ter we focus o n seco nd language learn ers' developing kn owl edge
and useof thei r new language. We begin by looking at the d ifferent co nt exts
for first and seco nd lan guage learnin g as well as th e 4ifferent characteristic s o f
learners in thes e contexts. We exam in e some of the errors that learners make
and discuss wha t errors can tell us about their knowledge of the language and
their ab ility to use that knowledge. We loo k at stages an d sequences in th e
acq uisitio n of so me synt actic and morphological features in the second lan­
guage. We also review some aspects o f learners' de velopment of voca bulary,
pragmatics, and phonology. .:,
ACTIVITY Explore contexts for second language learning
.;
A second language learner is different in many ways' from a young child
acquiring a first language or an older child learn inga second language.This
is true in terms of both the learners' characteristics.>( and the environments
in which the language acquisition typically cccurs.Think about how the
characteristics and learning conditions of the following learners may differ:
• a young child learning a first language ~~-'
• a child learning a second language in day care or on the playground
• an adolescent stu dying a foreign language in their own country
• an adult immigrant with limited or disrupted education working in a second
language environment and having no opportunity to go to language classes.
36 Second language learning Second language learning r
N o w ask yourself the following questions about th ese diffe rent learners. Using th e chart in Table 2.1,give you r o pinio n abo ut the presence or absence
I Do they already know at least one language? of learner characteristics and learning conditions for the four different
learners mentioned above. Use the following notation:
2 Are they cognitively mature? Are they able to engage in problem solving .
deductio n. and com plex mem o ry casks! =
+ usually present =
- usually absent
? = sometimes present. sometimes absent. or you're not sure
3 How well developed is their metalinguistic awareness? Can they define a
word. say what sounds make up that word. or scate.a rule such as 'add an First Second
-s to form the plural'? language language
4 How extensive is their general knowledge of the world? Does this Young child Adolescent Adult
Young child
knowledge enable them to make good guesses about what a second (at home) (playground) (classroom) (on the job)
language interlocutor is probably sa)'ing?
Learner characteristics
5 An~ they likely to be anxious about making mistakes and concerned about
sounding 'silly' when speaking the language? Another
6 Does the learning environment allow them to be silent in the early stages language
of learning, or are they expected to speak from the beginning! Cognitive
7 Do they have plenty of time available for language learning and plenty of maturity
contact with proficient speakers of the language? Metalinguistic
8 Do they frequently receive corrective feedback when they make errors in aware ne ss
grammar or pronunciation. or do listeners usually overlook these errors
W o rld
and pay attention to the meaning? kno wledge
9 Do they receive corrective feedback when their meaning is not clear.
Anxiety about
when they use the wrong word. or when they say something that seems
speaking
inappropriate or impolite? ..
lOis modified input available! That is, do interlocutors adapt their speech so Learning conditions
that learners can understand (for example. in terms of speed of delivery. Freedom to
complexity of grammatical structure. or vocabulary)? be silent
Then compare your views with the discussion of learner characteristics and Ample time
learning conditions below.
Corrective
feedback ,
.~

Learner characteristics (grammar and

pronunciation)
By definition . all second language learners, regardless of age, have already , :~..J
acquired at least one lan guage. This prior knowledge may be an advantage in Corrective
the sense that they have an idea of how languages work. On [he other hand, feedback
knowledge of other languages can lead learners to make incorrect guesses (meaning,
about how the second language works, and this may result in errors that first word choice.
politeness)
language learners would not make.
Modified input
Very young language learners begin the task of first language acquisition
without the cognitive maturity or rnetalinguisric awareness that older Photocopiable © Oxford University Press
second language learners have . Although young second language learners
have begun to develop these characteristics, they will still have far to go in Table 2. I Contexts for language learning
these areas, as well as in the area of world knowledge, before they reach the
levels already attained by adults and adolescents.
38 Second language teaming
On th e one han d , cognirive maruriry and meralinguistic awareness allow
older learners to solve problems and engage in discuss ions abour lan guage.
This is particularly important for those who are lear ning language in a class­
roo m, with limited time in contact with the language. On the other hand,
some theorists have suggested that the use ofthese cognitive skills-so valua­
ble for many kinds of tasks- can actually interfere with language acquisition.
Th ey argue m at successful language acquisiti on draws o n different mental
abilities, abilities that are spe cific to lan guage lear ning. It has been suggest ed
that older learners draw on their problem-so lving and metalinguistic abiliti es
precisely because they can no lo nger access the innate language acquisition
abiliry they had as young children. We will have more to say about this in
Chapter 3, when we discuss the role o f age in second language acquisition.
In addition to possib le cognitive differen ces, th ere are also attitudinal and cul­
tu ral differences berwee n children an d ad ults . Most ch ild learners are willing
to try to use the language-even when th eir pro ficiency is quite lim ited .
. Many adults and adolescents find it stressful when they are unable to express
themselves clearly .and correctly. Nevertheless, even very young (pre-school)
chi ldren differ in their willingness to speak a language they do not know well.
Some chil dren happily chatter away in their new lan guage; others prefer to
liste n an d participate silently in social interactio n with thei r peers.
Learning conditions
Young second language lear ners are often allowe d to be silent u ntil they are
ready to speak . They may also practise th eir seco nd language in songs an d
games that allow th em to blend their voices with those of other ch ildren.
O lde r second language learners are often forced to speak from the earli­
est days of their learning, whether to meet the req uirem en ts of classroom
instruction or to carry out everyday tasks such as shopping, medical visits, or
job interviews.
Another way in which younger and older learners may differ is in the amou nt
of time they can actually spend learning a second language. We know that
first language learners spend thousands ofhou rs in contact with the language
or languages spoken around them . Young second language lear ners may
also be exposed to their second language for many hours every day-in the
classroom. on the playground. or in fronc of the television. Older learners,
especially students in foreign lan guage classrooms, receive far less expos ure­
perhaps only a: few hours a week. Indee d , a typical foreig n lan guage student
will have no mo re than a few hundred hours of exposure, spread om over a
number of years. Adu lt learners who are immigrants or minoriry language
speakers often continue to use the language they already know as they fulfil
th eir daily responsibilities for work and family, and they may use the second
language only in limited situations.
Second language teaming 39
C lassroo m learn ers not only spend less tim e in co nt act with th e new lan­
guag e, they also tend to be exposed to a far smaller ran ge of discourse rypes.
For example, classroom learners are oft en taught language that is somewhat
form al in comparison to the language as it is used in most social settings. In
many fa reign language classes, teachers may even switch to their students' first
lan guage fo r discipline or classroom management, thus depriving learners of
oppo rtu nities ro experience uses of the language in real communication.
As we saw in C hapter 1, parenrs tend to respond to th e meani ng rather th an
co the grammatical accuracy of thei r children's language. Similarly, in second
language lear ning outside classrooms, errors that do not interfere with
meaning are usually overlooked. Most people would feel they were being
impolite if they in terrupred and corrected someone who was trying to have
a conversation with them. Nevertheless, interlocutors may react to an error
if they cannot understand what the speaker is trying co say. Thus, errors of
grammar and pronunciation may nor be remarked on, bur th e wrong word
choice may receive com m en t from a puzzled interlocutor. In a situation whe re
a second language speaker appears to use inappropriate language, interlocu­
tor s may feel un co mfortable, not knowi ng whether..the speaker intends to be
rude or simply does not know the polite way to say ~h:i't is intended. In this
case coo, espec ially berween adults, it is unlikely that the second language
speaker would be told that something had gone wrong. The only place where
feedback on error is typically presenc with high frc;quency is the lan guage
classroo m . Even there, it is no t always provided consis ten tly. In C hapters
5 and 6. research on the role of feedback in the classroo m will be reviewed.
One condition that appears to be common to learners ofall ages-though not
in equal qualiry or quantiry-is expos ure to modified or adap ted input. This
adjusted speech style , called child-directed speech in first language acquisi­
tioh, has sometimes been called foreigner talk .or teacher talk depending
on the contexts of second language acquisition. Some people wh o interact
regula rly with lan guage learners seem to have an intuitive sense of what
adjustments they need to m ake to help learners tlnderstand. Of course, not
everyone knows what adjustments will be most helpful. We have all witnessed
those pa inful conversations in which people seenj-to th in k that they can
make learners understand better if they simply talk louder! Some Canadian
friends told us of an experience they had in China. They were visiting some
historic temples and wanted to ger more informatio n about them than they
could glean fro m a guidebook, so they asked their guide some questions.
Unfo rtu nately, their limi ted Chinese and his non-existent Englis h made it
difficult for them to exchange information. The guide kept speaking louder
and louder, bu r ou r friends understood very little. Fina lly, in frust ratio n , rhe
guide concluded that it would help if they could see the informatio n, so he
rook a stick and began wri ting in th e sand-in Chinese characrers!
40 Second IdnguugeLearning Second Language Learning 41

language can be q uite perplexing ifo ne does no t have an overall picture o f the
LVtf",.to.4/la.u~, -o.~ e"t.ire pl... ",b;"S ~~ t-e ..,
is. eu.rre n I:' ~ l..U1aer r.epce;.r . It ~ M. ros' ,1k
'tP ~ -tj.,e. ,.....bl.·c. ~"""",",,':~,es
Sita-H.,( .u>D....c f.·.e
1I.tocJ<,S -f.-o...,
iCA eve
I:c " ,.
hete .. ~:-~ ) ~
?:
/
f)
~~
steps learners go through in acquiring the seco nd language.
In p resenting some of the findings of second language research, we have
included a number ofexamples oflearner language as well as some additional
sam p les to give you an opportunity to practise analysing learner language. Of
~' ~ c'~ ~ e!~.~0' course, teachers analyse learner language all the time. They try co determine
~, ~ ,> whether students have learned what has been taugh t and how closely their
language matches the target language. But progress cannot always be meas ­

. ·;tE1~.a
ured in these terms . Sometimes language acquisition progress is reflected
in a decrease in the use of a correct form that was based on rote memoriza­

'"
\.~
J1~ r Fl-t6HI M~,v...
I N FoP. M,q,T, ON
{,..C; -­
..
This bri ef d iscu ssion places th e em p hasi s on how bo th th e cha racte ristics of
learners and the contexts in which they acquire a seco nd language m ay be
different. In the following pages , we will focus more on similarities in how
their knowledge of the new language develops over tim e.
Studying the language of second
language learners
We have seen that children's knowledge of the grammatical system of their
first language is built up in predictable seq uenc es. For example, grammat i­
cal morphemes such as the -ing of the present progressive or the -ed of the
sim ple past are not acquired at the same time, but in a sequence. Are there
developmental seq ue nces for second language acquisition? How does the
prior knowledge of the first language affect the acquisition of the second (or
third) language? How does instruction affect second language acquisition?
Are there differences in the development oflearners whose only contact with
the new language is in a classroom and those who use the language in daily
lite? These are some of the questions researchers have sought to answer, and
we will address them in this chapter as well as in Chapters 5 and 6 .
Knowing more about the development oflearner language helps teachers ro
assess teaching procedures in the light of wha t they can reasonably expect to
accomplish in the classroom. As we will see , some characteristics of learner
tion or chunk learning. New errors may be based on an emerging ability
to generalize a particular grammatical form beyond the specific items with
wh ich it was first learned . In thi s sen se, an increase in error may be an indica­
tion of progress. For example, like nrst language learners, second language
learners usually learn the irregular past tense forms of certain common verbs
before they learn to apply the regular simple past -ed marker. That means
t ha t a learner w ho says ' I bu yed a bus ticket' may know more abou t En glish
gram mar than o ne who savs ' I bo ug h t a bus tickct., Wit ho u t further infer­
ma rio n. we cannot co ncl ude that th e one who says ' bo ugh t' wo uld use th e
regu lar past -ed marker w here it is approp riate , bu t the lear ner w ho says
' buyed ' has p rovided ev id enc e o f develo pi ng kn owledg e of a syste ma tic
aspect of English.
,
Teachers and researchers cannor read learners' minds , so they must infer
what lear ners know by observing what they do. Like those who study first
language acquisition , we observe learners' spontaneous language use, but
we also design procedures that help to reveal more about the knowledge
underlying their ob servable use oflanguage. WitP.0ut these procedures, it is
oft~n difficult to determine whether a particular. behaviour is representative
of something systematic in a learner's current language knowledge or sim ply
an isolated item, lear ned as a ch u nk . 0(
1
Like first language learners, second language learners do not learn language
sim ply through im itation and practice. They prod~c5 sentences that a re not
exactly like those they have heard. These new sentences appear to be based
on internal cognitive processes and prior knowledge tha t interact with the
language they hear arou nd them . Both first and seco nd lan gu age acquisition
are best described as developing systems with their ow n evolving rules and
patterns, no t simply as imperfect versions of the target language.
Contrastive analysis, error analysis, and interlanguage
Until the late 1960s, people tended to see second language learn ers' speech
simply as an incorrect version of the target language. Acco rding to the con­
trastive analysis hypothesis (CAH) , errors were assumed to be the result
42 Second language learning
Second language learning 43
of transfer from learners' first language. Detailed analysis of learners' errors
discover and describe different kinds of errors in an effort to understand
revealed , however, that not all errors made by second language lear ners can
how learners process second language data. Error analysis was based o n the
be explained in terms of first lan guage transfer alone. A number of studies
hypothesis that, like child language, second language learner lan guage is a
show that many erro rs can be explained better in terms of learners' develop­
system in irs own right-one thar is rule-governed and predictable.
ing knowledge of the structure of the target language rath er than an atte mpt
to tr ansfer patterns of the ir first language (Richards 1974). Furthermore, Larry Selinker (1972) gave the name incerlanguage to learners ' develo ping
some of the errors are remarka bly similar to those made by you ng first lan­ second lan guage knowledge. Analysis of a learner's inte rlanguage shows that
guage learners, for example, the use of a regular -ed past tense ending on an it has.some characreristics influenced by previously learned languages , some
irregular verb . characteristics of the seco nd language, and some characteristics, such as the
omission of function words and grammatical morphemes, that seem to be
A simplified version of the CAH would predic t that, where differences exist,
general and to occur in all interlanguage systems. Interlanguages have been
erro rs would be bi-directional, that is, for examp le. French speakers lear ning
foun d to be systematic, but they are also dynamic, continually evolving as
Englis h and English speakers learning French woul d make errors on paral­
learners receive more input and revise th eir hypotheses about the second
lei linguis tic featu res. Helmut Zo bl (1980) observed that thi s is not always
language. Th e path through language acquisi tion is no t necessarily smooth
the case. For example, in simple English sentences, direct objects, whether
and even . Learners have bursts of progress , then reach a plateau for a while
nouns or pronouns, come after the verb ('The dog eats the cookie. Th e dog
befo re something stim ulates furth er progress. Selinker also coined th e term
eats it.') . In French. direct objects that are no uns follow the verb (Le chien
fossilization to refer to the fact that so me features in a learner's language seem
mange Ie biSCUit-literally, 'The dog eats the cookie'). However, direct object
to Stop changing. This may be especially true for learners whose exposure to
pronouns precede the verb (Le chien le mange-li terally, 'The dog it eats').
the second language does not include instruction or-the kind offeedback that
The CAH would predict tha t a nativ e speaker of English might make the
would help them to recog nize differences between their interlanguage and
erro r of saying: ' Le chien mange le' when learning French, and that a native
the target language.
speaker of French might say 'The do g it eats' when learn ing English. In fact,
English speakers learn ing French are more likely to make the predicted
error than French speakers learning English . This may be due to the fact ACTIVITY Analyse learner language :.;
that English speakers learning French hear many examples of sentences with
The follow ing t exts were written by two learn ers of English. one a French­
subject-verb-object word order (for example, Le chien mange le biscuit) and
speaking secondary school student, the other a Chinese-speakin g adult learner.
make the incorrect generalizatio n-based on both the word order of th eir
first language and evidence from the second lang uage- that all direc t objects
Both learn e rs were des cribing a cartoon film entitle~ The Great Toy Robbery
(National Film Board of Canada) . After viewing the.film. they were aske d to
come after the verb. French-speaking learners of Engli sh, o n the other hand,
retell the story in writing, as ifthey were telling it to someone w ho had not seen
hearing and seeing no evidence that English direct object pronouns precede ~fi~ '
verbs, do not tend to use this pattern from their first language.
Read the texts and answer the following questions:1'
The findi ng that many aspec rs of learners ' language could not be explained
by the CAH led a number of researchers to take a different approach to ana­ I Can you unde rstand what each learner is trying to say?
lysing learners' errors . This approach, wh ich developed during th e 1970s, 2 Examine the errors made by each learn e r. What kirtd's of errors interfere
became known as 'error analysis' and involved detailed descriptions of the most with yo ur ability to understand?
errors second language learners made. The goal of this research was to dis­ 3 Do both learners make the same kinds of errors?
cover what learners really knew about the language. As Pit Corder observed
in a famous arti cle published in 1967 , when learners produce correct sen­ 4 In what ways do t he two interlanguages differ?
tences. they may simply be repeating something they have already heard; Learner I: Fre nch first language , secondary school student
when they prod uce sen tences that differ from the target language, we may Du r ing a sun ny day, a cowboy go in t he desert with his ho rs e . he has a big hat.
assume that these sentences reflect the lear ners' current understanding of the His horse eat a flo ur. In the same time. Santa Clause go in a city to give some
rules and pattern s of tha t language. We saw this in the example of a learner surprises. He has a red costume and a red packet of surprises.You have three
who says 'buyed' instead of 'bought.' Erro r analysis differed from contras­ robbers in th e mountain who sees Santa C lause with a king of glaces that it
tive analysis in tha t it did not set out to predict errors. Rather, it sought to permitted us to see at a long distance. Every robbers have a horse.They go in
44 Second I.mguage learning
the way of Santa Clause. not Santa Clause but his pocket of sur pr ises. After
they willgo in a city and they go in a saloon. [. . .]
(Unpublished data from P. M. Lightbown and B. Barkman)
Learner 2: Chinese first language. adult
This year Christmas comes soon! Santa Claus ride a one horse open sleigh to
sent present for children. on the back of his body has big packet. it have alot of
toys . in the way he meet three robbers.They want to take his big packet. Santa
Claus no way and no body help. so only a way give them, then three robbers
r ide their horse dashing through the town.There have saloon, they go to dr ink
so me beer and open the big packent.They plays toys in the Bar.They meet a
cow boy in the saloon.
(Unpublishe d data provided by M. J, Martens)
Perha ps the most striking thing her e is that man y erro r rypes are co m mo n ro
bo th learners. Furthermore. both make err ors of spelling and punctuation
tha t we m ight find in th e writing ofa yo ung first lan gu age spea ker o f Eng lish.
Even th ough Fren ch uses grammatical mor phemes to indicat e person and
number on verbs and C hinese does not, both these learners make errors of
subject-verb agreemenr-s-borh leaving off the third person -s marker and
overus ing it when the su bject is plural Ca cowboy go' and 'th ree robbe rs in the
mo untain who sees' by Learne r 1 and 'Santa C laus ride' and 'th ey plays' by
Learne r 2) . Such errors reflect learners' understanding of th e seco nd lan gua ge
system itsel f rath er than an attem pt to transfer characteristics of their first
lan gua ge. They are some tim es referred to as 'd evelop ment al' erro rs because
th ey are sim ilar to those made by children acquiring English as their first lan ­
guage. Sometimes these are errors ofovergeneralizarion , th at is, errors caused
by tryin g to use a rule in a conte xt where it does not belong, for exam ple, th e
-s end ing on th e verb in 'they plays'. Sometimes the errors are better described
as sim plification. where elements of a sentence are left out or whe re all verbs
have th e sa me form regardless o f person , number, o r ten se.
O ne can also see, especially in learner 2's text , th e influ ence of classroom
experience. An exam ple is the use of formulaic expres sions such as 'o ne horse
ope n sleigh' which is taken verbatim from a well-known Christmas song that
had been taught and sung in his English as a Second Language (ESL) class.
The vivid 'dashing through the town' probably comes from th e same source,
with the substitution of 'rown' for 'snow'.
For those who are familiar with the English spo ken by native speake rs of
French, so me of the errors (for exam ple, preposition cho ice 'in the same
tim e') made by the first learner will be seen as probably based on French.
Sim ilar ly, tho se familiar with the English of Chinese speake rs may recog­
n ize some word order patterns (for example, 'o n the back of his body has
big packet') as based on C hinese patterns . These may be called transfer o r
Second I.mgufJge learning 45
'interference ' errors . Wha t is m ost clear. however, is thar ir is often difficult
to determine the so urce o f errors. Thus , while erro r anal ysis has the advan ­
tage of describing what learners actu ally do rather than what they might do ,
it does not always give us clear insights int o why they do ir. Furthermore,
as Jacquelyn Schachter pointed out in a 1974 article, learn ers so meti mes
avoid using some features of lan guage th at they perceive to be diffic ult for
them. This avoidance ma y lead to th e absence o f certain errors, leaving th e
anal yst without information about so me aspects o f th e learn ers' develop­
ing intedanguage. Th e absence of particular erro rs is difficult to interpret,
and the phenomen on of 'avoidance' may itself be a part of th e learner's sys­
tema tic second langu age performance.
Developmental sequences
Second language learners, like first lan gua ge learn ers, pass throu gh sequences
of development: what is learned early by one is learned early by others.
Amo ng first lang uage learne rs, the existence ofd evelopmen tal sequences may
not seem surp rising because their language learning is partly tied to the ir cog­
n itive de velo pm ent and to thei r expe riences in learn ing abo ut relation sh ips
betwee n peop le, events, and objects arou nd them. But the cognitive develop­
ment of adul t or ado lescent second language learne rs is much more stable.
and thei r experiences wi th the langu age are likely to- be q uit e different, not
o nly from the experiences of a sm all child, but also different from each other.
Furth ermo re, second language learners already knowanother language th at
has different partern s for creati ng sent ences and wo rd form s. In light of th is,
it is more remarkable that we find developmental sequences that are similar
in the developing interlanguage of learners from ,different language back­
grounds and also similar to those observed in first langua ge acquisition of th e
same language. Moreover, the features of th e language that are most frequent
are not always learned firsr. Fo r exam ple, virt uallYFery Eng lish senten ce has
o ne or more arti cles ('a' or 'the'), but even ad vanced learn ers have difficulty
using these forms correctly in all contex ts. Finall y, altho ugh th e learner's first
language does have an influ en ce, many aspects of thcr.o9 developmental stages
are sim ilar among learn ers from different first langu;ge backgrounds.
In Chapter 1 we saw some developmental sequences for English first lan­
guage acquisition of grammatical morphemes, negation, and questions.
Researchers in second language acquisition have also examined these, as well
as other features . They have found patterns in th e development ofsyntax and
morphology that are sim ilar among learners from different language back­
grounds. Evidence for th ese developmental patterns first came from studies
o f learners wh ose primary learning en vironment was o utsid e th e classroom.
For examp le, JUrgen Meisel, Harald C lahsen, and Manfred Pienemann
46 Secondlanguage Learning Secondlanguage Learning 47

(1981) identified developmental sequences in th e acquisition of German by

speak ers of several Romance languages who had little o r no instruction. -ing (progressive)

plural

Subsequent research has shown that learners who rece ive instruction exhibit copula ('to be ')

similar developmental sequences and error pa tte rns. In the interlanguage of

English speakers whose only e~posure to German was in university classes in
Australia, Pienemann ( 1988) found patterns that were similar to those of th e
un instructed learners . In C hapter 6 , we will discuss other studies that have
inves tigated the influence of instruction on develop m ental sequences.
•
auxiliary (progr essive
as in 'H e is going')
article
Granunatica1 morphemes
Researchers have examined the develo pment of grammatical morphemes by
learners of English as a second language in a variety of environments, at
diffe rent ages , an d from different first lan guag e backgrounds. In analysi ng
each learner's speech, researchers identify the obligatory contexts for each
•
irregu lar past

.m o rphem e, that is, the places in a sentence where the mo rpheme is necessary
to make the sentence grammatically correct. Fo r example, in the sentence
'Yesterday I play baseball for rwo hours', the adverb 'yesterd ay' creates an
obligatory context for a past ten se, and 'fo r twO hours' tells us that th e required
form is a simple past ('played') rather than a pas t progressive ('was playi ng') .
Similarly, 'two' creates an o bligatory context for a plural-s on 'hours'.
For the analysis, ob ligatory co ntexts fo r each grammatical morpheme are
counted separately, that is, one count for simple past, one fo r plu ral, one for
third person singular present tense , and so o n . Afte r co u nting th e n um ber of
obligato ry COntexts, the researcher counts the correctly supplied morphemes.
The next step is to d ivide the number of correctly supplied morphemes by
the total n um ber of obligatory co ntexts to answer the question 'w hat is the
percentage accuracy for each morpheme?' An accuracy score is created for
each morpheme, an d these can then be ranked from highest to lowest, giving
an accuracy order for the morphemes.
The overall results of the studies suggested an order that was similar but not
ide ntical to the developmental sequence found fo r first language learners.
H owever, the order th e researchers found was quite similar among second
lan guage lear ners from d ifferent first language backgro u nds . For exam ple ,
most studies showed a higher deg ree of accuracy for plural -s than for posses ­
s,
sive - and for -ing than for regular pas t (-etl). Stephen Kr ashen summarized
th e order as shown in Figure 2. 1. The diagram sho uld be interpreted as
showing that ' lear ners will produce the morphemes in h igher boxes with
higher accuracy than those in lower boxes, but that within boxes , there is no
clear pattern of difference.
•
reg ular past -ed
third person singular -s ""
possessive's
Figure 2.1 Krashen 's (/982) summary of second language grammatical morpheme
acquisition sequence J!.
The sim ilar ity among learners suggests that the accuracy o rder canno t be
described or explained in terms of transfer from the learners' first lan guage,
and so m e researchers saw this as strong evidence against th e CAH . However,
a thorough review of all the 'm o rp hem e acquisition' studies shows tha t the
leafners' first language does have an infl uence on; acquisition seq uences. For
example, learners whose first language has a possessi ve form that resembles
the Eng lish s (such as German and Dani sh ) se1m to acqui re the English
po ssessive earl ier than those whose first language 'h as a very different way of
forming the po ssessive (such as Frenc h or Spani sh). And even though art i­
cles appear early in the seq uence, lear ners fro m manylanguage background s
(in cluding Slavic languages, Chinese, and Japanese) continue to struggle
with this aspect of English, even at advanced levels. Learners may do well in
supplying articles in certain obligatory contexts but not others. If the lan­
guage sample that is analysed COntains only th e 'easier' obligatory co ntexts,
the lear ner may have a m islead in gly h igh accur acy score.
Another reason why something as di fficult as English articles appears to be
acquired early is that the order in the diagram is based on t he analysis of
correct use in o bligatory co ntexts only. It does not take in to acco unt uses of
grammatical m orp hem es in places where they do not belong, for example,
when a lear ner says, 'The France is in Europe' . These issues led researchers
48 Second langu.lge learning
to que stion th e adequacy of ob ligatory co ntext anal yses as th e so le basis to r
understanding developmental seq uences. Teresa Pica (1983) argued that
accuracy scores should take account of overuse and incorrect uses to deter­
mine a 'score for target-like use rather than reflect only use in obligatory
Con te-xts.
The morphem e acquisition literature raises other issues, not least of them the
question of why there should be an order of acquisition for these language
fearu res.So me of the similarities observed in different studies seemed to be
due to the use ofparticular tasks for collecting the dar a, and researchers found
that different tasks tended to yield different results . Nevertheless, a number
ofstud ies have revealed similarities that cannot be explained by the data col­
lection pro cedures alone. As with first language acquisition, researchers have
not found a single sim ple explanatio n for the order. Jennifer Goldschneider
and Robert DeKeyser (200 1) reviewed thi s research and identified a number
of variables that contribute to the order. Salience (how easy it is to notice the
. morpheme), linguistic co mplexity (fo r example, how man y elements yo u
have to keep track of), semant ic trans paren cy (how clear th e meaning is).
sirni lariry ro a first language form , and freque ncy in the input all seem to play
a role.
N egatio n
The acquisitio n o f negati ve sente nces by second lan guage learners follows
a path that looks nearly identical to the stages we saw in Chapter 1 for first
language acquisition. However, second language learners from different first
language backgrounds behave somewhat differently within those stages. This
was illustrated in John Schumann's ( ) 979 ) research with Spanish speakers
learning English and Henning \'{fode's (1978) work o n G erma n speake rs
learning Engli sh ,
Stage 1
The negative element (usually 'no' or 'no r') is typ ically placed before th e verb
o r the elem ent being negated. Often , ir occur s as th e first word in th e sen­
ten ce because the subject is not there.
No bicycle .
I no like it.
Not my friend .
'N o' is preferred by most learners in this early stage, perhaps because it is
the negative form that is easiest to hear and recognize in the speech th ey are
exposed to. Italian- and Spanish-speaking learners may prefer 'no' because it
co rrespo nds to the negative form in Italian and Spanish (No tienen mu chos
libros). They may conrinue to use Stage 1 negation longer than other learners
because of the similarity to a pattern from their first language. Even at more
ad van ced stages, they may also use Stage) negativ es in longer sentences or
Second I.mguuge learning 49
when the y are under pressu re. Th us, sirnilarirv to a learn er's first language
may slow down a learner's progress th rough a particular developmental stage.
Stage 2
At this stage, 'no' and 'no t' ma y alt ernate with 'do n't' . H owever, 'do n't' is not
mar ked for person , numbe r, o r tense an d it may even be used before mod als
like 'can' and 'sho uld' .
He don't like it.
I don 't can sing.
Stage 3
Learners begin to place the negati ve element after auxiliary verb s like 'are',
'is', and 'can' . But ar thi s stage , the 'do n't' form is still no t fully an alysed .
You can not go there.
He was not happy.
She d on 't like rice.
At this stage, German speakers, whose first language has a st ructu re that
places the nega tive after the verb may gene ralize th e auxiliary-negative
pattern to verb-negative and produce sentences such as:
They co me no t [to] home, (Sie kommen nicbt nach Hause.i
Stage 4
In this stage , 'do' is marked for ten se, person, and number, and most in terlan­
guage senten ces appear to be just like thos e of the target langu age.
It doesn't work. We didn't have su pper.
H owever, some learners conrinue to mark tense, person , and number o n
both the auxili ary and th e verb .
I didn't wen t th ere.
Questions
Manfred Pienemann, Malcolm Johnston, and CSeoff Brindley (1988)
described a seq uence in the acquisition of qu estions by learners of Engli sh
from a variety of first language backgrounds. An adapted version of the
sequence is shown in Stages 1-6 below. The examples (except those in
Stage 6) come from French speakers who we re playing a game in which they
had to ask questions in order to find out wh ich picture the orher player
(the researcher) was holding. As we saw for negation , the overall sequence is
similar to the one observed in first language acquis ition. And again , there are
some differences that are attributable to first language influence.
Stage 1
Single words, formulae , or sent ence fragm ents.
50 Second language learn ing Second langu age learning 51

Dog?
ungrammatical to use inversion with a noun subject (" Pou rq uoi aiment les
Four children?
enfanrs le chocolat /).
What's that?

Stage 6
Stage 2 Complex questions.
D eclarative word order, no inversion, no fronting. question tag: It's better, isn'r it?
It's a monster in th e right corner?
negative q uestio n: Why can' t you go?
The boys th row the shoes?
embedded ques tion: Can you tell me whar the date is today?
Declarative order with rising intonation is common in yes/no questions in
Pienemann's developmental sequence for questions has been che basis for
informal spoken Fren ch . French speakers may hypothesize that in English,
a number of studies, some of which will be discussed in Chapter 6. Alison
as in French, inve rsion is optional.
Mackey and her colleagues have done a number of these studies, and she pro­
vided the da ta in Table 2.2. These examples come from three"aduIr Japanese
Stage 3
learners of English as a second language who were interacting with a na tive
Fronting: do-fronting, wh-fronting without inversion, other fronting.
speaker in a 'Sp ot the differences' task. In rhis task, learners have similar but
Do you have a shoes on yo ur picture?
no t identical pictures and they have to ask q uestions un til they work OUt how
Where the children are playing?
the pictu re they can see is different from the one their interlocuror has. Note
Does in this picture the re is four astronauts?
char p rogress ro a higher srage does not always mean char learners produce
Is the picture has rwo planets on top?
fewer errors .

French has an invariant form est-ce que (literally ' is it that') that can be placed

before a declarative sentence to make a question . For example, Jean aime le

ACTIVITY Analyse learners' questions
cinema becomes Est-ce queJean aime le cinema? ('is it that) John likes movies?'
Using the information about the .developmental sequence'fo r questions. circle
l'
French speakers may thi nk that 'do' or 'd oes' is such an invariant form and
the stage of second language question development thatbest corresponds to
contin ue to produce Scage 3 questio ns tor some time .
each question.
Stage 4
(Hint: Read all of each learner's qu est io ns before you begln.)
Inversion in wh- + copula;yeslno questions with orher auxiliaries.
c ,

. . Stage
Where is the sun?
\
Learner'
Is there a fish in the water?

I Where is he going and what is he saying? .( I 2 3 4 5 6

At Stage 4, German speakers may infer that if English uses subject-auxiliary

inversion, it may also permit inversion with full verbs, as German does, leading
2 Is the room his room? I 2 3 4 5 6
them to pro duce questions such as 'Like you baseball?' (Magst du basebaip.)
3 Is he caking out his skate board? ,. I 2 3 4 5 6
,:'\J

Stage 5
4 What is he thinking? I 2 3 4 5 6
Inversion in wh- questions with both an auxiliary and a main verb .

S The girl. what do you. what does she do. what is I 2 3 4 5 6

How do you say 'p roche'?
she doing?
Whar's the.boy doing?

French-speaking learners may have diffic ulty using Stage 5 questions in Learner 2
which the subject is a noun rather than a pro noun. They may say (and 6 Are they buying some things? I 2 3 4 5 6
acce pt as grammatical) 'Why do you like chocolate?' but not 'Why do chil­
7 Is they bought present? I 2 3 4 5 6
d ren like cho cola te?' In th is, they are drawing on French, where it is often
-J Sa ond lungudge learn ing
) - Second language learning 53

8 Is they're retirement people! I 2 3 4 5 6 No te that when the object possessed is a body part, French rvp ically uses a
defin ite ar tic le rather than a possess ive de termine r.
9 Is this perfume o r .. , I don't know. I 2 3 4 5 6
Il s'esi casse le bras = He broke the [his] arm.
I 0 And it is necktie! I 2 3 4 5 6
Joanna Whi te ( 1998,2008) studied the acq uisitio n ofpossessive determiners
by Frenc h-speaki ng students, adapting a developme ntal seq uence that was
Le arne r 3
first p rop osed by H elmu t Zobl (1984). White found a to tal of eight stages in
I I Are there any shuttle! Space shuttle! I 2 3 4 5 6 the sequen ce, but th ey can be gro uped into three ma in stages. The examp les
12 Inside , is there any girl? sho wn below co me from Fren ch -sp eaking students learning English . They
I 2 3 4 5 6
are describi ng ca rtoo n draw in gs of famil y eve nts and in te rac tio ns.
13 You don't see? I 2 3 4 5 6
Stage 1: Pre-emergence
14 What are, what the people wearing! I 2 3 4 5 6
No use of 'hls' and ' her'. Defi n ite ar ticle or 'yo ur' used fo r all perso ns, genders,
I S And they are carrying pink box! I 2 3 4 5 6 an d num be rs.
The little boy play w ith th e bicycle.

Answer key
H e have band -aid on the ar m , th e leg, th e stomach,

Learner I : Questions 1,4, and 5 are Stage 5 questions . Question 5 is

This boy cry in the arm of your mother.

interest ing because it shows th e speaker self-correcting, suggesting th at

Stage 5 is st ill a level th at requ ires so me greate r effo r t. Questions 2 and 3 There is one girl talk with your dad .

are Stage 4 questions.

Learne r 2: Questions 6 an d 9 could be Stage 4 que stion s. However, the fact
t hat questions 7 and 8 are Stage 3 questions suggests that this speaker
has not actually progressed from 'fro nt ing' to 'inve rsio n', particularly since
question lOis a Stage 2 question.
Stage 2: Emergence
Emergence of ,his' and/or ' her, with a stro ng preference to use onl y o ne o f
the forms .
The mother is dr essing he r littl e boy. an d she p ur her clo thes , he r pam, her

coa c, and then she fini sh . ,."

Learner 3: Questions I I and 12 are Stage 4 questions. Questions 13 and The girl m aki ng hisself bea utiful. She p u c the ma ke- u p on h is hand , o n h is

I5 are Stage 2 questions. Question 14 shows the speaker apparently on head, and hi s fat he r is su rp rise.

the verge of a Stage 5 question, then retreating t o a Stage 3 question .

Photo cop iable © Oxford Un ive rsit y Pre ss

Stage 3: Post-emergence
Differentiated use of ' his' and ' her bur not when th e objec t possessed has
Table 2.2 Questions by Japanese-speaking learners o(Eng/ish natural gen de r.
.'
The girl Fell o n her bicycle. She look h is fathe r an d cry.

The dad put her little gi rl o n h is sho ulde r, and aft er, o n his back.

Possessive determiners
A developmental sequen ce for th e En glish possessive form s ' h is' an d 'h er' has
been o bse rved in th e inrerianguage o f French- and Spanish -speaking learn­
ers. In English , the cho ice o f''h is' or ' her' (or 'its') is determined by the na tural
ge nde r o f the possessor. In Fren ch and Span ish (and m any other languages),
At the end of th e po st- emergence stage, in what Whft't: (2008) calls Stage 8,
learners finall y ach ieve erro r-free use o f'his ' and ' her' in all co nt exts includ­
ing natural gender and body parts .
The little girl with her dad play together. And th e dad cake h is girl o n his

th e co rre ct form o f the possessive d eterminer matches the grammat ical shou ld er and he hur t his back.

ge nde r o f the o bjec t o r person tha t is possessed. Th is ca n be illustrated w ith
th e fo llowing t ran slati on eq u ivale n ts for Fren ch and English:
Sa mere = h is mother o r her mother
Son chien = his dog or her dog
Ses enfants = his ch ild ren o r her ch ild ren
\V he n Eng lish speake rs learn Fren ch , o r othe r lan guages that use grammati­
cal ge nder as the basis For choosing possessi ve determine rs, the y must also
learn a new wa y o f determining th e ge nde r of the pos sessive determiner. Th e
need to learn che g ram ma rical ge nde r o f eac h and every no un furthe r add
to the challenge.
54 Second language learn ing
Relative clauses
Second language learners first acquire relative clauses that refer to nouns in the
subject an d direct object positions, and only later (and in so me cases, never)
learn to use them to modify nouns in other sentence roles (for exam ple, indi­
rect object an d ob ject of preposition). A summary of the observed pacrern
of acquisition for relative clauses is shown in Table 2.3. It is referre d to as the
accessibility hierarchy, and it reflects the apparent ease with which learners
have access to certain st ructures in th e target lan gu age.
Part of speech Relative clause
Subject The girl who was sick went home.
Dire ct ob ject The story that I read was lo ng.
Indirect object The man who[m] Susan gave the present to
was happy.
Object of preposition l-found the book that John was talking about.
Possessive I know the woman whose father is visiting.
Object of comparison The person that Susan is taller than is Mary.
Photo co piable © Oxford University Press
Table 2.3 Accessibility hierarchy for relative clauses in English
(adapted from Doughty 1991)
Unlike the study of grammatical morphemes , negation, and questions, the
study of relative clau ses was no t inspired by research on child language.
Rather, it cam e from patterns that Edward Keenan and Bernard Comrie
(1977) observed in a large number oflanguages. They found that those lan­
guages that included the structures at the bottom of the list in Table 2.3
would also have those at the top, but the opposite was not necessarily true .
Subseque ntly, Susan Gass (1982) an d others found that if a second language
lear ner could use one of the str uctures at the bo ttom of th e list, he or she
would probably be ab le to use any that precede it. O n the ot her hand, a
learner who could pro duce sentences with relative clauses in the subject or
direct object positions (at the top of the list) would not necessar ily be able to
use them in any of the clause types further down the list.
Despite the similarity of the general pa ttern, several types of first lan guage
influence have also been observed in the acquisition of relative clauses. Firs t,
it has been observed that for learners whose first lan guage does nor have a
particular clause type (for examp le, object of comparison ), it is more dif­
ficult to learn to use that type in En glish . Second, where learners have a first
language with a substantially d ifferent way of for ming relative clau ses (for
example, Chinese and Japan ese, whe re the relative clause precedes the noun
Second language learning 55
ir modifies ), they ma y avoid using relative clauses even when their interlan ­
guage is fairly advanced. Third, first language influence is seen in the erro rs
learners make. For example, Arabic speakers often produce both the relati ve
marker and the pro noun it replaces (for example, 'The man who I saw him
was very angry') as the y wo uld in Arabic.
Reference to past
A n umber of researchers, including jurgen Meisel (1987), have observed th e
develop ing ability to use lan guage to locate events in time. The research has
shown that learners from different first language backgrounds and acquiring
a variety ofsecond languages, acquire the lan guage for referring to past events
in a similar pattern.
Like young chil dren, learners with limited language may simply refer to
events in the order in which they occurred or mention a time or place to
show that the event occurred in the pase.
Viet Nam. We work too hard.
My son come. He work in resta urant.
Later, learners start to attach a gra mmatical morpheme marking the verb for
past , although it may not be the one that the target language uses fo r that
meaning.
Me working long time. Now stop. j .-
Past tense forms of irregular verbs may be used before-the regular past is used
reliably. '
We went to school every day. We spoke Spanish/
After they begin marking past tense on regular ver bs, learners may overge n­
era lize the regular -ed ending or the use of the wrong past tense form (for
example, the present perfect rather than the simple past ).
.,'
My sister cat ched a big fish. i
She has lived here since fiftee n years.
J\.,1
Kathleen Bardovi -Harlig (2000) and others have found that learners are more
likely to mark past tense in sentences such as 'I broke the vase' and 'My sister
fixed it with glue' than in sentences such as 'She seemed happy last week' or
'My father swam in thar lake'. These differences appear to be d ue to the 'lexical
aspect' , tha t is, th e kinds of meanings expressed by the different verbs . Learners
seem to find it easier to mark past tense on verbs that refer to something whose
end poi n t can easily be determined. These are referred to as 'accomp lishme nts'
an d 'ach ievemen ts' ('I ran thre e m iles.' 'My brother took an aspirin and went to
bed '). For 'activities' that may continue fo r some period ('I swam all afternoon ')
or 'states' that may be perceived as constants ('He seemed ha ppy to sit by the
lake') , learners use simple past markers less frequently.
56 Second I.mguage learning
First language can have an influence here too. Laura C ollins (2 002) in vesti­
gated the diffe rent English verb fo rms used by French speakers. The past tense
tha t is most commonly used in spoken French and that is usual ly a transla­
tion of a simple past form in English is a form that resembles th e p resent
perfect in English. Thus, the equivalent of 'Yesterday he ate an apple' is H ier
il a ma ng~ un e pomme- literally..'Yesterday he has eaten an apple'. Teachers
often comment on French speakers' tendency to overuse the present perfect.
In Collins' study, learners completed passages by filling in blanks with the
appropriate form of a verb. As expected, in places where English speakers
would use the simple past, French speakers did so metim es use the perfect
(either present perfect or past perfect)" forms. Furthermore, they used them
more frequently than a comparison group of Japanese speakers. However,
the French speakers were more likely to use perfect forms for achievement
and accomplishment verbs than for the states and activities. Collins observes,
'The [first lan guage] influence does no t appear to override the effect of lexical
aspect; rather it occurs within it' (p. 85).
M ovem en t through developmental sequences
\Y/e have seen in thi s section that, as in first lan guage acq uisition , th ere are
systema tic and p red icta ble developmental seq ue nces in seco nd lan guage
acq uisitio n. H owe ver, it is important to em p hasize that developmental stages
are not like closed rooms. Learners do not leave one behind when they enter
another, In examining a language sample from an individual learner, one
sho uld not expect to find behaviours from only one stage. On the contrary,
at a given point in rime, learners may use sentences typical ofseveral different
stages. It is perhaps better to think of a stage as being characterized by the
emergence and increasing frequency of new forms rather than by the com­
plete disappearance of earlier ones . Even when a more advanced stage comes
to dominate in a learner's speech, conditions ofstress or complexity in a com­
municarive interaction can cause the learner to slip back to an earlier.stage.
In addition, as we have already noted, progress to a higher stage does not
always mean fewer errors. For example, a learner may produce correct qu es­
tions at Stage 1 o r Stage 3, but those correct forms are not necessarily based on
underlying knowledge of subject-verb inversion . That is, correct questions
at Stage 1 are formulaic chu nks, not sentences that have been constructed
from the wo rds that mak e the m up. At Stage 2, learners have advanced, in th e
sense that they are forming o riginal questions, bu t the word o rde r of those
questions is not grammatical in the ta rget language. At Stage 3, questions are
formed by placing a question form (most often a tab- word o r a form of the
verb 'do') at the beginn ing of a sen tence with declarative word order, This
may result in questio ns such as ' Do you want to go?' that conform to English
patterns. However, when the learner asks a question such as '0 0 you can help
Second language learning 5­
me?' we can see that the learner's ime rlang uage ru le really is so me thi ng like
'Put a question word at the beginning of the sentence.'
Another important observation about developmental seq uences is the
way they interact with first language influence. Learners do not appear to
assume that they can sim ply transfer the structures of their first language
in to the second. Rather, as Henning Wode ( 197 8) and Helmut Zobl ( 1980)
observed , when they reach a developmental stage at which they perceive a
'crucial similarity between their first lang uage and their interlanguage,
they ma y generalize their first language partern and end up making erro rs
that speakers of other languages are less likely to make. They may also have
difficulty moving beyond that stage if their errors do not interfere with
com m un ica tio n.
More aboutfirst language influence
One reason that so me researchers rejected the hypothesis that 'transfer'
o r ' int erference ' wo uld best exp lain a learn er's d ifficulties wi th the target
language was the fact char co nt rastive analysis was closely assoc iate d wit h
behavio urist views oflanguage acquisition. In rejecti ng beha viou rism, so me
researche rs also disca rde d co nt rastive analysis. In d oing so, they potential ly
lost an essent ial so urce o f information abo ut lan gua ge acquisition .
Researchers at the European Science Foundation carried out a study that
created some valuable opportunities to examine mc::, influence of the first
language on second language learning. Adult language learners , most of
w hom had little or no second language instrucrion; were followed as they
learned anothe r European language. For each target language, learners from
rwo different first language backgrounds were co m pared , Also, for each first
language background, the progress oflearners in their acquisition of the twO
target languages was studied. As \'YIolfgang Klein and Clive Perdue (1993)
repOlt, there were substantial simi larities in the interlanguage patterns of the
learners, in spite of the great variety in the first anH second language combi­
nations. The similarities were greatest in the earlies t stages ofsecond language
acquisi tion , when learners produced similar simple Spntences.
There is no doubt that learners draw on the patterns of other languages they
know as they try to discover the complexities of the new language they are
learning. The patterns of those ear lier languages are firm ly established, and as
learners have experience with the new language, there is an interplay between
the new and old patterns. As Nic k Ellis (2009 : 153) put it, 'Th e lan guage
calcu lator has no "clear" burton.' In learning so meth ing new, we build on
wha t we already know.
58 Second language Learning
We have seen some ways in which the firsr language interacts with develop­
mental sequences. When learners reach a certain stage and perceive a similarity
to their nrSt language, they may linger longer at that stage (for example, the
extended use of preverbal 'no' by Spanish speakers) or add a sub-stage (for
example, the German speaker's in version of subject and lexical verbs in ques ­
do ns) to the seq uence which, overall, is similar across learners, regardless of
their firsr language. They may learn a second lan guage rule but restr ict its
app lication (for example, the French speaker's rejection of subject-auxiliary
inversion with noun subjects that we saw in Stage 5 questions on page 50).
. The first language may influence learn ers' interlanguage in other ways as well.
As we saw earlier, the pheno menon of avoidance that Jac quelyn Schachter
(1974) described appeared ro be caused at least in pa rt by learners' perception
that a feature in the rarget language was so distant and different from their
first langua!?e that they preferred not co try it.
Other researchers have also found evidence of learners' sensitivity ro degrees
of distance .or difference and a reluctance co attemp t a transfer when they
perceive the lan guages as roo different. In one revealing study, Hikan
Ringbom ( 1986) found that the interference errors made in Engl ish by both
Finn ish-Swedish and Swedish-Finnish bilinguals were most often trace­
able to Swedish, not Finnis h. The fact tha t Swedish and English are closely
related languages that actually do share many characteristics seems to have
led learn ers to take a chance that a word or a sentence structure that worked
in Swedish would have an English equivalent. Finn ish, on the other hand,
belongs co a completely d ifferent language family, and whether their own
first language was Swedish or Finnish, learners appeared reluctant to draw on
Finnish in learning English .
The risk-taking associa ted with this percep tion of similarity has its lim its,
however. For example, Eric Kellerman ( 19 86) observed tha t learners often
believe that idiomatic or rneraphorical uses ofwords are un ique co a particu­
lar lan guage. Kellerman found that Dutch learners of English were reluctant
to accept that certain idiomatic expressions or unus ual uses of words were
also possible in Englis h. For example, rhey rejected 'Th e wave broke on the
shore' but accepted 'He broke the cup ' even though both are straightforward
translations ofsentences with the Dutch verb breken.
Another way in which learners' first languages can affect second language
acquisition is by making it difficult for them to notice that something they
are saying is not a featu re of the language as it is used by more proficient
speakers. Lydia White (199 I) gave the example of adverb placement in
French and English. Both languages allow adverbs in several positions in
simple sentences. However, as the exam ples in Table 2.4 show, there are some
differen ces. English, b ur not French, allows SAVO orde r; Frenc h, but not
Eng lish, allows SVAO.
Second language learn ing 59
S =Subject V =Verb 0 =Object A =Adverb
ASVO
Often, Mary drinks tea.
Souvent, Marie bait du the.
SVOA
Mary drinks tea often.
Marie bait du the souvent
SAVO
Mary often drinks tea.
*Marie souvent bait du the.
SVAO
*Mary drinks often tea.
Marie bait souvent du the.
Noce: The asterisk (*) means that the sentence is not grammatical.
Phococopiable © Oxford University Press
Table 2.4 Adverb placement in French and English
It seems fairly easy for French-speaking learners of English to add SAVO to
their repertoi re and for English-speaking learners of French to add SVAO ,
but both groups have difficulty getting rid o f a pattern that does nor occur
in the target language if it is similar co one in their first language. Englis h­
speaking learners of French continue to accept SA'!.O as grammarical, and
French-speaking learners of English accept SVAO.:.AsWhite points OUt, it is
difficul t to notice that something is not present in.the input, especially when
its translation equivalent sounds perfecdy all rightand communication is not
disrupted. This may be even more challenging when learners interact with
Others from the same first language background. Their own errors are not
likely to cause misunderstanding and, in fact, they may hear Others make the
same errors. We will return co this when we look ai the role of instruction and
feedback on errors in Chapters 5 and 6. .
Jo, J
Our understanding of the relationship between 'n rSt- and later-learned
languages has been refined in recent decades. The term cross-linguistic influ­
ence is now often used, in part co reflect the fact that the relationship is by no
means unidirectional . That is, as we acquire a second or third lang uage, the
patterns that we learn can also have an impact on the way we use and under­
stand the language(s) we learned earlier. Current views of second language
development emphasize the interaction between the firsr language (or other
previo usly learned languages), cog nitive processes, and the samples of the
targe t language tha t learners encoun ter in the input. As extensive reviews by
Terence O dl in (2003) and SCOttJarvis and Aneta Pavlenko (2008) show, the
complexity of this relationship has ins pired scores of investigations.
60 St>cond Lmgu<lgt> Learning
So far thi s chap te r has focu sed on th e acquisition o f morphology and syntax
in the second language. We now tu rn to the learning ofsome other irn po rran r
co m po ne n ts of communicative competence: voca bulary. pragmatics, and
pron unc ia tio n .
Vocabulary
In 1980, Paul Meara characterized vocabulary learning as a 'neglected aspect
oflanguage learning '. Researche rs in the 1970s and ear ly 1980s were drawn
ro syntax and morphology because of the way error patterns and develo p­
mental sequences of these features migh t reveal something about u n iversals
in languages and language acq uisi tio n. H ow diffe rent things are now! J USt as
M eara was commenting on th e state of neglect, an explosion of research on
vocabulary learning was beginning, and the acq uisitio n of vocabulary has
become one of the mos t active areas in seco nd lan gu age acquisition research.
For most people, the importance of vocabulary seems ver y clear. As it has
o fte n been remarked , we can communicate by using words that are n ot placed
in the prope r order, pr on ounced perfectly, or marked wi th th e pr op er gram ­
matical m orphemes, but communicati on o fte n brea ks down jf we d o not
use the correct word . Although ci rcu m loc u tio n a nd gestures can sometimes
com pe nsa te. the impo rtance of vocab ulary ca n ha rdly be overes timated.
J)oYou ~EA'f:::
words they are exposed to may be m ore diffi cult, refer rin g to meanings that
rRE:NC~ ?
are not easily guessed from context, Marcella H I.! and Paul Nation (2000)
<,
(~
(If;; WE 1-"', I )X)N'T
KNOW AL.l­
-me woRDS if:T· ··
.J
( The first step in know ing a wo rd is sim p ly to recogn ize that it is a word. Paul
f of th is fact. So me of th ese tests ta ke the fo rm of word lists, and learners are
Second Languuge ft>urning 61
The challenge o f acq ui ring a large eno ug h vocab ula ry for succ essful comm u­
nication in a variety of set tin gs has been the focu s of m uch recent resear ch.
Every lan gu age has an asronishi ng ly large n umber of wo rds . Eng lish, which
has built its vocabulary from a grea t var iety o f so urce languages, is variously
estimated ro have anywhere from 100 ,000 to one million words, depending
in part o n ho w wo rds are counted . For examp le, so me would treat 'teach .
teacher. teaching, and taught' as sepa rate words whil e others would count
all of them as pan of o ne 'wo rd '- a single root from which the o thers are
derived.
An educated adult speaker of English is believed to know at least 20 ,000
words: some estimates suggest a number that is more than twice that. But
most everyday conversation requires a far smaller member, something more
like 2,000 words. Sim ilarly. although Chinese and Japan ese have ten s o f th ou­
sands of characters. mos t are rare , and no n- tech nical material can usually be
read with a knowledge of about 2.000 characters. Even so, acquiring a basi c
vocabulary is a slgnifican r accomplishment for a seco nd language learn e r.
As we saw in C ha pter 1. children learn th ousan ds of wo rds in the ir firSt lan­
guage wit h little observable effort. TIle task of acqu iring a large vocabulary
is q uite d ifferent for second lan gu age learners. For one thing, they are likely
to be ex pose d to far smaller sam p les of th e language to be learned . Also, th e
co ntexts in which seco nd language learners enco u nte r new voc ab ulary may
not be as helpful as those in which ch ildre n learn die first one o r two thou­
san d words of their first language. If they are olderchlldren or adults , the
showed that, in order to unde rstand a text witho ur frequent stops to co nsult
a di cti onary, one needs to know more than 9 5 pe,r cenr of the words-a rare
case for second langu age learners at mos t srages of acq uisi tion. Altho ugh the
two or three th o usan d most freq ue nt words in English make up as much as
80- 90 per cenr of most non-technical texts, less f~equenr wo rds are crucial to
th e meaning of many th ings we hear and read. F6 r examp le, th e meaning o f
a newspaper article abo ut a court case m ay be lost without the know ledge of
words suc h as 'testim o ny', 'alleged', or 'acco m plice'e....
Meara and his co lleagues (2005) have develo ped tests th at tak e ad vanrage
instructed to check 'yes' o r 'no' accordi ng to whe the r or no t they know the
wo rd. Eac h list also includes so m e items that loo k like E ng lish wo rds but arc
nor. The number of real wo rds that the learn er identifies is ad justed fo r guess ­
ing by a facror tha t takes acco u nt of th e n umber of no n-words th at are also
chosen . Such a procedure is more effective th an it m igh t so u nd. A ca refully
co nstructed list can be used ro esti ma te th e vocabulary size of even advanced
learne rs. For example. if shown the fol lowing list : 'Fro lip, laggy, scr u le. an d
62 Secondlanguage learning Secondlanguage learning 63

albeit' . a proficient speaker of English would know that only one of these there is naming in me w ritte n form or me pronunciation of me wo rds th em ­

words is a real English word, albeit a rare and somewhat odd one. On the selves mat gives a clue to their meaning. If students are to learn them , they

oth er hand, even proficient speakers mighr recognize none of me following must see or hear me words in contexts mat reveal their meaning, and, as a

items: 'go n io m erer, micel le, laminitis, rhrostle'. Even our computer's spell­ rule. they must do this many times before the link between a word and irs

che cker rejected twO our of fo ur, but all are real English words, according to meani ng is well established.

me New Oxford Dictionary ofAmerican English. On me other han d , some stu dents who .have never studied. English m igh t

Among m e factors that make new vocabulary m ore easily lear nable by already know words in List 2, beca use they are par t of an international

second language learn ers is m e freque ncy wi th which me word is seen, hear d, vocab ulary. Wim increasing internationalizatio n of co mm unica tio ns,

and understood. Paul Nation (200 1) reviews a n umber ofstudies suggesting many languages have ' bo rrowed' and adapted words from other languages.

mat a learner needs to have many meaningful encounters with a new word Studenrs chroughour che world may be surprised ro lear n how many words

before it becom es firmly established in memory. The estim ates range as high they already ~ow in me language they are tryi ng to learn .

as 16 ti mes in some studies. Even more encounters may be nee ded befo re a The wor ds in List 3 loo k difficult. They are rat her lo ng, not easily illust rated,

learner can ret rieve the word in Huen r speech or au tom atically und erstand and most are fairly infrequent in the lan guage. However, many students

me meaning of the word when ir occurs in a new COntext. The ability to would either 'know' chem on sigh t or learn them afte r a. single exposure

understand the meaning of most words without focused attentio n is essential because they look like the ir tra nslation equivalent in other languages that

for fluent reading as well as for fluent speaking. they already know. Some, such as 'natio n' and 'dictionary', are cognates

Frequency is no t me only faeror that de termi nes how easily words are lear ned , (words that have come fro m the same original roo t); others, such as 'deno ue­

howeve r, as illustrated by the wo rds in me three lists shown in Table 2.5. ment' and 'en tent e' , are bo rrowed words (words that have been adopted from

other languages) . These words ma t look alike and have shared meaning can

List I List 2 List 3 help learners expand their vocabulary.

friend hamburger gove rnment Teachers should not assume mat studen rs will always recognize bo rrowed

more Coke responsibility words or cognates in the ir second lan guage. Some eognates are identical in

form and mean ing, while others may require some knowledge of how spell­

town T-shirt dictionary ing patterns are related in me two languages (fo r example, 'water' and Wasser

boo k Facebook elementary in English and German respectively or 'm usic' and musique in English and

French) . Even with differe nr spellings, words are likely to be easier to recog­

hunt taxi remarkable

nize in thei r written form than th ey are in me spoken language . Learners may

sing pizza description need guidance in recognizing rhern , as illus trated in me following question,

hotel expression asked by an eight-year-old in a Quebec hockey arena: 'H i coach, comment on .

box
dit 'coach'en anglais?' ('Hey, coa ch. How do yo u s~ 'coach' in English?') . And

smile dollar international afte r a moment's reflecti on , English speakers may realize mat they know both

eye Internet denouement speciality items in a Japanese restauran t mat calls i~c:lf 'Sushi and Bisuteki.'

night disco entente On th e other ha nd, students may have particular difficulty with words that

look similar in the twO languages bu t have different meanings. These 'false

Phococopiable © Oxford University Press cognates' may come fro m different origins or they may have evolved differ­

entl y from the sam e origin. For example, the En glish verb 'deman d ' has a

Table 2.5 English words that may be 'easy' or 'difficult' (or second language learners d ifferen t meani ng fro m its Fren ch cousin demander, which mea ns 'request'

All of me words in List I look easy because the y are simple one-syllable words or 'ask a q uestio n', even though they came from the same Lat in verb .

that refer to easily illustrated actions o r objects. They are also quire common Some th eo rists have argued that second language learners, like children

words in English, appearing amo ng the 1,000 most frequent words. And yet , learning their first language, can learn a great deal of vocabulary with little

they are not likely to be known ro stu dents who have not had p revio us ins tr uc­ inten rional effor t. Stephen Krashe n (1989) has asserted mat me best source

tion in English or exposure ro the language outside sch ool. Furt hermore, of vocabulary growth is readi ng for pleasure. There is no doubt mat reading

 Second language learning 65

64 Second language learning

is an irnportanr po tential source of vocab ulary development fo r second lan­

guage learners as it is for first lan gu age lear ners. H owever, the re are so me
problems w ith t he notion th at vocabulary growth th rough reading requires
little effort. As no ted above, it is difficult to infer the meaning of a new word
from reading unless o ne already kn ows 95 pe r cent o r more of the oth er
wo rds. and learners usually need to have many meaningful enco unters with
a word befo re they recognize it in new con texts or p roduce it in th eir own
speaking an d wri ting. As we saw in Chap ter 1. Dee Gard ner's (2004) research
de mo ns trates th at ce rtain types of words are rare in narratives. Thus, StU­
dents who read mainl y fictio n may have little chance o f lear ning wo rds th at
are essential fo r th eir aca demi c pursuits. Conversely. reading m ainl y scie nce
texts w ill not p rovide many o pport un it ies to learn the vocab ulary of soci al
i nteractio n .
Research on voca bul ary learnin g through readi ng withou t focu sed ins tr uction
con firms mat some vocabulary can be learned w ithout explici t inst ruc tio n
(see C ha p te r 6, Stu dy 17) -.On the o the r hand , [anHulsti jn and Bacia Laufe r .
(200 1) and Others provide evidence t hat voca bula ry development is more
successfu l when learners are fully e ngaged in ac tivities that require th em to
attend carefully to the new words and even to use them in productive tasks .
Izabella Kojic-Sabo and Patsy Lighrbown ( 1999) found that effort and the
use of good learn ing strategies, such as keep ing a no tebook, looking words
up in a d icti onary, an d reviewing what h as bee n learned were associated with
Pragm atics .
be tte r vocab ulary development. Cheryl Boyd Zimmerma n (2009) p rovides Pragmat ics is t he study of how lan guage is used iri co ntext to exp ress suc h
many p ractica l sugges tio ns fo r teach in g voca bulary an d also fo r help ing thi ngs as d irectness. po lite ness, and defe rence. Even if learners acquire a
learners to continue learni ng o uts ide the classroom. voca b u lary of 5.000 words and a good knowledge of the syntax and mor­
pho logy of m e target lan gu age, they can sti ll e nco unte r d ifficulty in using
Even with ins truc tio n an d good strateg ies, me task of acq uiring an adequate language. They also need to acquire ski lls for interp reti ng requests, resp o nd­
vocab ulary is daunting. \'V'hat does it mean to 'know' a wo rd : ing politely to complime nts o rapo logies, reco gn lz lng h umo ur, an d managi ng
• G ras p th e ge ne ral m eaning in a familiar context? co nve rsatio ns. They need to learn to reco gnize th e many mean ings th a t the
• Provide a definition o r a tr ansla tio n equivalent? same sentence can ha ve in d ifferent situarlons vIjunk of the many ways one
• P rov ide ap pro p ria te wo rd assoc iatio ns? migh t interp ret an apparently simple question suc h as 'I s mat YOUt dog? ' It
• Identify its co m po nent pa rts o r ety mology ? mig ht pr ecede an exp ressio n o f admi ratio n fo r an att ractive pet , or it might
• Use the word to co mplete a sentenc e o r to create a new sentence? be a n urgent req uest to get the do g o ur o f me speal}w-'s flowe rbed. Sim ilarly,
• Use it m etaphori cally? the same basic m eaning is alt ered whe n it is ex pressed in di fferent ways. For
• Underst and a jo ke th at uses h omonym s (wo rds mat so und alike bu t exam ple, we wo uld pr ob abl y assume m at the relationship between speake r
m ean different th ings, such as 'cents', 'sense' , 'scen ts')? an d listener is very different if we hea r 'G ive m e that bo ok' or ' I wo nde r if
you'd m ind letting me have th at book when yo u've fin ish ed with it'.
Seco nd lan gu age learne rs whose goal is ro use the language fo r both social The study of how seco nd language learners acq uire thi s aspect o f langu age
and acade mic purposes must learn to do all th ese thi ngs. is refe rred to as 'i nt erlanguage pr agm atiCS' (Ba rdov i-Harlig 1999). So me of
th is research has focu sed o n th e ways in w hich learners ex press speech acts
suc h as in viting an d apo logizing in relatio n to d ifferences in (heir profi cien cy
level o r th eir first language ba ckground. O the r stud ies have exa m ine d learn­
ers' abi lity to perceive and co m p rehend pragmatic features in th e seco nd
66 Second languagelearning Second language learning 67

language and to judge whether a particular request is appropriate or inap­ Stage 5: Fine tuning

propriate in a specific context. Fine tuning of requestive force to participants, goals, and contexts.

Since the early 1990s more research has directly investigated the acquisition You could put some Blu Tack down there.

of second language prag matic ability. This includes lon gitu dinal and cross­ Is there any more white?

sectional studies describi ng the acquis ition ofseveral different speec h acts. O ne
that has been the focus of consi derable atten tion is 'requesting'. Req uests are
an interesting pragmatic feature to exam ine because there are identifiable ways
in which requests are made within particular languages as well as differences in
how they are expressed across different languages and cultures.
In a review of longitudinal and cross-sectional studies on the acquisition
of requests in Englis h, Gabriele Kaspe r and Ken neth Rose (2002) outline
a series of five Stages of development. Stage 1 consists of mi ni m al language
that is often incomplete and highly context-dependent. Stage 2 includes
primarily memorized routines and freque nt use of imperatives.. Stage 3 is
marked by less use o f form ulas, more productive speech, and some mitiga­
tion of requests. Stage 4 involves more complex language and increased use
of mitigation, especially suppo rtive statem en ts. Stage 5 is marked by more
refinement of the force of requests. The five stages , thei r characte ristics and
examples are given below.
Stage 1: Pre-basic
course' , whereas the non-native speakers said 'I will rake that course'.
Highly context-dependent, no syntax, no relational goals .
Learn ing how to make and reject suggestions has also been extensively inves­
tigated. Kathleen Bardovi-Harlig an d Beverly Hartford (199 3) o bserved
differences between the way in which native and non-native speakers of
English communicated with their professors as they discussed their course
selections in academic advising sessions. These differences contributed
to their greater o r lesser success in negotiating their academic plans . For
example, the non-native speakers did not initiate suggestio ns whereas na tive
speakers initiated a grea t d eal. There was also a tendency on the part of th e
non-native speakers to reject suggestions made by the advisor in ways that the
advisors might find rude or inappropriate. For example, they would reject
an adv isor's suggestion to take a particular course by saying 'I think I am no t
interested in that course' , instead of saying ' My schedule conflicts with that
course', or 'I th ink this other co urse would better meet my needs', which was
more typical of na tive-speaker rejection responses. 'The-non-native speakers
were also much less adept than the native speakers at using mitigation. For
example, native speakers were observed to say ' I think I would like to take this

Over a period of four an d a half mon th s, the rese arc~irs observed progress in

Me no blue. some aspects of the non-native speakers' pragmatic ability. For example, th ey

Sir. learned to take a moreactive role in the advising interaction and to provide

reasons for rejecting suggestions that the advisors-were likely to perce ive as

Stage 2: Formulaic
more credible or acceptable. Even so, they continued to have some d ifficulty

Reliance on unanalysed formulas and imperatives.

in mitigating their suggestio ns an d rejections. . ~.

Let's play the game.

For a long time, it was assumed that second language classrooms could

Let 's eat breakfast.

not provide appropriate opportunities for stude¥'tS to lear n many different

Don't look.
speech acts. This was especially true in reacher-fronted classrooms where

the dominan t int eractio n pattern was 'teacher initiation-learner respo nse­

Stage 3: Unpacking
Formulas incorporated into productive language use, shi ft to conventional teacher feedback' and where the emphasis was alm'~~t always on producing

full sentences that were grammatical ly correct (see fur ther d iscussio n of

in directness .
this in Chapter 5). In communicative, content-based, and task-based lan­

Can you pass the pencil please? guage teaching, there are more opportunities not only fo r a greater variety

Can you do ano ther one for me? o f input but also fo r learners to en gage in different roles and participant

organization structures (for example, pair and gro up work). This enables

Stage 4: Pragmatic expansion lear ners to produce and respo nd to a wider range of communicative func­

Addi tion of new forms to reperto ire, increased use of mitigation, more
tio ns. Furthermore, research o n the teaching ofpragmatics has demonstrated

co mp lex syntax.
that pragmatic features can be successfully learned in classroom settings and

Could I have another choco late beca use my child ren- I have five chil d ren. tha t expl icit rather than implicit instruction is most effective (Kaspe r and

Can I see it so I can copy it? Rose 2002). This is good news for fo reign language learners who do not have
(,8 Second language learning
exte nsive exposure to conversational in te raction ou tside the classroom. Th e
qu estion is no longer whe ther second lan guage pragmatics should be taugh t
but ramer how it can be integrared into classroom instruction.
Phonology
As noted earlier, grammar has been the focus fo r second language teac hers
an d researchers for a lo ng time. Vocabulary and pragmatics have also received
more attentio n in recen t years. H owever, we know less about pronunciatio n
and how it is learned and taught. Pronunciation was a central component in
language teaching when me audi olin gu al ap proach was dominant. Several
techniques for teaching pro nu nciation were developed at mat time, and
mo st of th em focused on th e pronunciation of segmenrals , getting learners
ro perceive and ro produce distinctions between single sounds in minimal
pai r drills (fo r exam p le, 'ship' and 'sheep') .
When -rh e au d io lingu al approach was repla ced by other ways of teach­
ing, atre nrio n [0 pron un ciation was min imized if not tot ally d iscarded .
Furthe rmo re, evidence for the critical period hypo th esis, sugges ting that
native-like pronunciat ion was an unrealistic goal for older seco nd lan guage
learners (see C hapter 3), led to the argument that instructi on al time woul d
be be tter spe n r o n teaching so me thi ng th at lear ne rs co uld learn mote success­
fully. When communicative language teaching (CLT) was nrs t introduced in
the late 1970s, little attention was given to me teaching of pronunciation.
If it was taught, the emphasis was o n su p raseg m en tals (rhythm, stress, and
in rona rio nj-c-as pecrs of pronunciati on mat were considered m ore likely to
affect communication (C elce-M urcia, Brinton, and Goodwin 1996) .
Although research o n me teaching and learning o f pronunciation is not as
extensive as m at in other language domains, there is theoretical and em pirical
wo rk ro help us understand the processes invol ved in phonological develop­
m erit in a second language and the factor s tha t co n tribu te to it. For example,
co nt rastive analysis helps ro explain some aspecrs offirst lan guage infl uence on
second lan guage learners' pronunciation . We can all th in k o f examp les from
ou r own experiences or those of o ur st udents. Japanese and Korean lear n ers
of Englis h often have problems hearing and p roducing I and r because these
sounds are no t d isti nct in th eir language. Spa nish speakers will often say ' I
e-speak e-Spanish' because Spa nish words do not have consonant clusters
beg inning with s at the beginning ofa word. French speakers may p lace stress
on the last syllab le of a word beca use French usually stress es the last syllab le.
Few languages have the th sounds that are frequent in En glish, and learn­
ers may substitute sim ilar sounds from their first language (for examp le, t
a t d, s o r z). Some times, howeve r, learners overcompensate fo r so unds mat
they know are d ifficult. Thus, learners may pronounce a th (as in 'th in' or
'this') where a t or a d so und belongs (saying 'th in' w hen they mean 'tin' for
Second language It.'aming 69
exam ple). Such erro rs are sim ilar to the overge neral izario n erro rs tha t we saw
for grammatical morphemes. If they rep lace earlier correct pronu nciation o f
t or d sounds, this may represem progress in learners' abil iry to notice and
produce the th so un d .
The relationship between perception and production of so unds is co m plex.
Evelyn Alcenberg (2 005) developed a se ries of tasks ro exp lo re Span ish
speakers' perceptions and production of English consonant clusters at the
beginning of a word . In one task, th ey had ro say whether certain invented
words were possible 'new English words'. The Spanish speake rs were quite
good at recognizing w hat English words are supposed to sound like. They
accep ted pseudowords like 'sp us' and rejected those like 'zban' , even though
born words would be unacceptable as 'new Span ish words' . Sh e found th at
they co uld usually wri te (from dictation) pseudowords with initial cl usters
such as sp and sm . However, in their own production , the se same learners
might still in sert a vowel at the beginning o f words such as 'spoo n' and 'sm ile'.
As we have seen with regard to gram mar and voca bulary, it is hypothesized '
tha t a grea ter d ifference between the learn er's nat ive lang uage an d th e target
lang uage ca n lead to greate r difficul ry. The evide nce supporting [he hypoth­
esis comes partly fro m the observation that it takes learn ers longer to reach a
hig h level ofH ue ncy in a particul ar seco nd or fo reign lan guage if th at lang uage
is su bstant ially different from the languages they alrefldy know. Fo r exam ple,
a speaker of C hinese faces a greater challenge in learn ing English than does
a sp eaker of German or Dutch . Language distance. affects pronunciation
as well as other language system s. Theo Bongaerrs '.(1999) collected speech
sam p les from highly proficient speakers who had learned Dutch in their
adulthood and who came from a wide variery o ffi ~s t language backgrounds.
W ~e n native speakers of Dutch were asked to judge the sp eech sam p les, o nly
those learners who spoke a language mat was closely related to Dutch (fo r
exam ple, English or German) were judged to have nat ive-like accents. None
of the speakers whose first languages were mor1' distant from Dutch (for
exam ple, Vietnamese) were judged to h ave n at ive-like pronunciation .
1 he re has been little research to document (he dev;~wpmental seq uences of
ind ividual sounds in seco nd language ph onological a~quisi tion . Nonetheless,
there is evide nce for similariry in the acq uisition o f some features ofstress an d
rhythm and it also clea r tha t the learner's first lan guage pl ays an im po rtant
role. O rner factors suc h as the amo unt an d rype of exposure to the targe t
lang uage and th e degree of use of the first language have bee n identified
as in fluen tial co ntrib uto rs to pronunciation. Thorsteri Piske, Ian MacKay,
and James Flege (200 I ) have repor ted that lon ger periods of exposure to
the seco nd lan guage ca n lead to improved pronunciatio n. It has also been
observed that adults who continue ro make great er use of thei r first language
may have stro nge r accenrs in the seco nd lan guage (Piske 2007).
70 Second language learning
Learners' ethnicaffiliation and their sense ofide ntiry are also related [Q how they
pro duce the sounds and rhythms ofa second language. Elizabeth Garbo n ron,
Pavel Trofirnovich, and Michael Magi d (2005) found a complex relation ship
between feelings of ethnic affiliation and second language learners' acquisi­
tion ofpronunciation. Among other th ings, they found that learners who had
achieved a high degree of accuracy in prono uncing the seco nd language were
sometimes perceived as being less loyal [Q their ethnic gro up than those whose
second language speec h retained a strong 'foreign accent'. Such percep tions
can affect learners' desire [Q achieve high levels of proficiency in the second
language , especially in contexts wh ere th ere are co nflicts between groups or
where power relationships imply a threat to one group's identity.
Pavel Trofimovich (2005 ) has looked at learning pronunciation from a
somewhat unusual perspective. His research raises questions about how well
learners perceive th e specific sounds of the new language wh ile their focus is
on meaning. Second language learn ers ofSpanish were asked to listen to a list
of familiar Spanish wo rds. For the purpose of comparison , they also heard
a list of words in English, their na tive language. O ne group of participants
were told [Q 'just listen' to the words; the second group were asked to pay
attention to how good the recording quality was; the third group were asked
[Q rate the 'pleasantness' of the things the words referred to. Then they heard
another list, which included bo th the original words and some new words,
an d they were asked to repeat each word as they heard it. Trofimovich then
compared how quickly each learner started to prono unce the words they had
already heard and the new words . Th e d ifference in the time it took them
to react to 'old' and 'new' words is a m easu re of how easily words could be
retrieved fro m memory.
As expected, the participants were always faster at retrieving the old words
in their native language, and rwo groups oflearners also showed this pattern
for their seco nd language. But the third group. who had been told to focus
. o n th e 'pleasantness' of the meanings, did not retrieve the old words faster.
Trofimovich suggests that when learners focus p rimarily on meaning. they
may not be able [Q also pa y attention to the sounds that make up the words.
In Chapter 6. we will review other research showing that learners sometimes
fail [Q notice certain language forms-grammatical morphemes, vocab ulary
words, syntactic patterns, pragmatic features-when their focus is on under­
Stand ing meani ng .
Few studies have investigated the effectiveness of pronunciation instruction ,
but the results of recent research suggest that it can make a difference, par­
ticular ly if the instruction focuses on suprasegrnenral rather than segmental
aspec ts of pronunciation (H ahn 2004). Tracey Derwing and her colleagues
(2003) carried OUt a series of Stu d ies to de termine how intelligi ble learners
were judged to be. Th ey found that learners who were given pronunciation
lessons emphasizing stress and rhythm were judged to be easier to understand
Secondlanguage learning 71
than learn ers who received lesson s focused o n ind ivid ual sounds. Even though
the learne rs wh o were given instruction on individ ual sounds were more
accurate in their use of those sounds, this did no t seem to increase listeners'
perception of the intelligibility of their speech to others. Findings like these
support the current emphasis on su prasegmentals in pronunciation classes.
One of the con troversial issues in pronunciation is related to the question
of whether the goal of secon d language acqui sition is to sound like a 'na tive
speaker.' One obvious problem with the question is tha t it suggests that there
exists a single correct variety of English, and th is is far from true. Not on ly
are there many different pronunciations of English by American, Australian,
British, Canadian 'native speakers' , the re are also many other varieties ofEnglish
that have come to be used as a linguafranca arou nd the world. Jane Setter and
Jennifer Jenkins (200 5) and Barbara Seidlhofer (20 11) are among the many
scholars who stress the role of English as a linguafranca (ELF). Indeed, there
are now far more speakers of ELF than of Englis h as a first language:
A related question is whether intelligibility rather than native-like pro­
nu nc iation is the standard that learners should strive toward. Stu d ies of
relatio nsh ips between Engl ish nati ve speake rs' perceptions of fo reign accent,
their perceptions ofcomprehen sibility, and their actual abil ity [Q understand
what speakers are saying show that the th ree are related. However, research
by Murray Munro and Tracey Derwing (20 11) shows that the presence of a
strong foreign accent does not necessarily resul t in r.educed int elligibility o r
comprehensibility.
Unfortunately, research evidence does not changethe fact that some listen ­
ers respond negatively to second language speakers'<pronunciation. In some
situations, accent still serves as a marker of group membership an d is used as
the.basis for d iscrim ination . Thus, some secondlanguage learners , particu­
larly thos e who have achieved a hig h level of knowledge and performanc e
in other aspects of the target language, ma y be motivated [Q approximate a
mor e 'native-like' accent for personal and professi~nal reaso ns. O ther second
language learners view this as irrelevant [Q their goals and objectives as users
of the second language (D erwing and Munro 2009); .... .~
Research related [Q teaching pro nunciation is gaining more arte ntion . It is
already clear that decori rexrualized pronunciation instruction is not enough
and that a combination of ins tr uction, e.xposure, experience, and motiva­
tion is required if learners are [Q change their way ofspeaking. Ro bin Walker
(20 10) provides guidance for teaching pronunciation in a way that recognizes
the importance of preparing students fo r interacting with other speakers of
English as a lingua fran ca.
-1
St!cond Lmguage [earning
Sampling learners' language
One of the challenges of study ing learners' language is the difficulry of co l­
lecting samples of their speech or writing that are large enough to ensure
that an alyses an d their findings are based on more than JUSt a few learners o r
fro m just a few examp les fro m a larger number oflea rne rs. Research ers often
find it d ifficult to recruit learners, to obtain th eir co nsent to participate in a
stu dy. and to persuade them to remain available over the time peri ods that
are necessary to show development. It can also be challenging to sched ule
sessions fo r recording sp eech or collecting writing sam p les, and to transcribe
o r digitize the speech sam ples for analysis .
Co mp u te r-based tool s are making it possible for resear chers to ask and
answe r new ques tions an d to revisit so me o f th e tentative answers to q ues­
tions th at have been around for a long tim e. For example. corpus linguistics
has provid ed us with large coll ections of naturally occ urring data that can
be used to discover information about the frequency o f different language
featu res (wo rds. ph rases. gra mmatical patt ern s) in a variety o flang uage co n­
texts and regist ers. Some of th ese co rpo ra co nt ain langu age sam ples that have
been co llected from newsp apers and conversatio ns. whe reas Othe rs are more
specific to part icu lar rypes o fla nguage. In additio n. the re are learner co rpo ra
and pedagogic corpora. More and mo re research ers are making th eir learn er
languag e data available to o the rs so that each corpus of learner language dat a
can be used for a number ofdifferent studies. Some of th ese corpora are avail­
able in CD or D VD format or online (Tarone and Swierzbin , 2009). The
links between corpus research and second language teaching are also becom­
ing increasingly apparent (Bennet, 2010; Sinclair. 2004). For example, if a
co rp us o f classroom language reveals that certain features occur frequ ently
in classroom input, teachers might decide to focus on features that occ u r less
frequenrly. Similarly if corpus research reveals that specific grammatical fea­
tur es are rarely used by native speakers in conversati onal interaction , teachers
(and textbooks) may de vote less time to th e o ral pr acti ce of th ese features.
Summary
The lan guage that second language learners produce and understand changes
as they have more exposure to the language and as they use it in a greater
var iety of situations. Describing those changes has been the focus of this
chapter. \'(fe have seen that there are strong paceerns ofsimilariry across learn­
ers o f different ages, learn ing in different contexts, and start ing from differ ent
first languagc backgrounds. The focus of this chapter has been mainly on
these similarities. In Chapter 3. we will turn our attention to some of the
ways in which learners differ from each other and how those individual dif­
ferences affect how quickly and how well th ey succeed in second language
acq uisitio n.
Second [angu, zge learning -:- 3
Questions for reflection
What are the general rules or patterns of negative sentences in English?
Look ing at th e develop mental sequence that has been described for English
negation. think about what learners seem to notice first. Is it word order?
Special words?What features seem hardest for them to acquire?
2 How would you collect samples of learner language for a study of the
acquisition of grammatical morphemes?What kind of speaking or writing
task would be most effective in leading learners to create obligatory
contexts for each of the morphemes listed in Figure 2.1? Do you think
some morphemes would be relatively easy to create contexts for? Which
ones do you think would be difficult! Do you think the 'wug test' would be
a useful tool?
J What aspects of learners' interlanguage are most likely to affect their
ability to use language effectively outside the classroom? Word order?
Grammatical morphemes? Vocabulary? Phonology! Pragmatics? Do you
th ink pr iorities for class room int er action and instruction reflec t t he
importan ce of t he se different language features!
Suggestions for further reading
Ellis, R. and G. Barkhuizen, 2005 . Analysing Leamer La ngu age. Oxford:
Oxford Universiry Press. "
Analysing Learn er Language introduces read ers co different approache s co
investigating learner language. It also serves as an extensive review of pub­
lished research using a range of m ethods and techniques for gathering and
an alysing data. Each ch apter is devoted co a particular anal ytic approach ,
including error ana lysis, frequency anal ysis, and socio cultural anal ysis. as
well as a cha p ter by Michael Barlow o n how com pu ter cools can be used .
Th e book will be of special interest to students who are em barking o n
seco nd language acquisition research at th e post-graduate level.
Nation, I. S. P. 200 1. Learning Vocabulary in Anorhft,JLanguagt!.
C am bridge: Cambridge University Press.
This comprehensive book cove rs research and theory and their implica­
tions for teaching and testing vocabulary development in a second or
foreign language. Although many books on voca bulary teaching and
learning have been published since thi s on e. it remains an essential and
accessible text for reach ers and post-graduate stu dent s who wish co under­
stand both the chall enges of vocabulary learn ing and those involved in
reaching and assessing vocab ulary knowled ge.
74 Second language learning

Tacone, E. and B. Swierzbin. 2009. Exploring Leamer Language. Oxford:

Oxford University Press.

3
Th e au thors collecred speech an d wri rin g samples from a group ofEn glish
language learn ers fro m different L 1 background s par ticipating in the same
IND IVID UAL D IFFEREN CES
tasks. This created a data base show ing how each learner tried to achieve IN SECOND LANGUAGE
th e same com mun ica rion goals. Th e text issupplemen ted by a OVD o f the
learn ers engaged in th e oral tasks. Exercises focus on differem ap proaches LEARNING
to un ders ta nding the lear ners' emerging language syste ms , includi ng error '
analysis, developmemal sequences, learners' response to feedback, an d
commun icat io n strategies. Many of the ideas tha t are introduced in th is
chapter o f How Languages Are Learned are illust rat ed in th e examples of
learners' lang uage th at are presen ted in th is book. ­

Preview
As V-(e saw in C hap ter 1, child ren are almost always succe ssfu l in acquiring
the language or lan guages th at are spo ken (o r signed) to them in early child­
hood, provided that they have adequate opportunities..[ Q use th e language
over time. This co n trasts with our expe rience of second lan guage learners,
whose success var ies grea tly. Both educators an d researche rs have an interest
in u nderstanding how the characteristics of individ uals are related to the ir
ability to succeed in learning a second language. ;: .
Many of us believe that individual d ifferences that are. inherent in the learner
can predict succes s or failure in lan guage learni ng. Such beliefs may be based
on our own experience or that of people we have known . For examp le, many
teachers are convinced th at extroverted students ~ho interact witho u t inh i­
bit ion in the second language and seek opporrunities to practise language
skills wUI be the most success ful learners. In add ition to an outgoing per­
sonality, other characteristics often believed to predict succ ess in language
lear ning are intelligence , motivation , and the ag9at which learning begins.
To what exte nt can we predict differences in the su ccess of seco nd lan guage
acq uisitio n if we have in fo rmatio n about learn ers' Rf:.!t50 nalities, the ir general
an d specific intellectual abilities, thei r mo tivation, or th eir age? In th is
cha pter, we will review some of the studies that have so ugh t to understand
th e relationships berween in d ivid ual diffe rences and learn ing outcomes.

AC T IV ITY Reflect o n la nguage le arning experience

Befo re you read t his chapt er. use th e questionnaire in Tab le 3. 1 to reflect o n
your ow n expe rience as a language learne r. Using additional co pies of the
quest io nnaire. interview seve ral friends. colleag ues . o r family members abo ut
t he ir exper iences of learn ing a second or fore ign language . Keep the response s
76 Indi vid UdI differences in second fdnguage learnmg Ind ivid ual ditftrences i

to the questionnaire and refe r to them as you read this chapter about
individual differences in second language learn ing. How much time have I I
you spent living in a place
where the language is
I
I a Wha t language do yo u spea k best? Do you speak more t han o ne
language equally well? spoken? I
I b When did you begin to learn this language (these languages)? Have you used the I
2 Which second or foreign language(s) have you learned with the language to learn other
most success? subjects at school? At
wha t level (elementary.
3 Which second or foreign language(s) have you learned with the secondary. university)?
least succ ess?
Do you have personal or
4 For the languages you mentioned in response to questions 2 and 3. emotional attachments
answe r the following questions in the appropriate columns: to this language? For
Languages learned , Languages not example. do you have
successfully learned successfully peers or family members
who speak this language?
How old were you when . I Do/did you enjoy I
yo u first tried to learn
the language? studying the grammar of I
I I t his language ?
Did you have a choice I I I

about learning this Do /did you en joy

language or were you studying vocabulary in
required to learn it? I this language? I
Do you currently speak Are/were you a successful student in other school subjects?
this language regularly? Do you think of yourself as a person who likes to socialize?

"

Do you regularly Do you think of yourself as a person who learns anew language easily?

read this language

for information or Photo cop iable © O xfo rd Universi ty Press
enjoyment? ,
How much of your Table 3.1 Individual differences in language learning experience
learning experience
with this language was
in a foreign language
Research on learner characteristjcs
classroom?
I "
Perhaps the best way [Q begin o ur discussion is [Q de scribe ho w research o n
If you no longer use this
language o n a daily basis. the influence o f individual differences on second lan guage learning is usually
can you estimate how done. \"'Vhen researchers are interested in finding out how a variable such as
many years you spent motivation is related to second language learning o utcomes. they usually
learning or using it? select a group o flearn ers and give them a questionnaire to measure th e rype
and degree of their m otivation. Then some kind of test is used to assess th eir
Estimate how many
secon d language proficiency. Th e test and the questionnaire are both sco red ,
hou rs of classroom
instruction yo u had for and the researcher uses a statistical procedure called a correlation. The cor­
this language. relation is an indication of how likel y it is that learners with high sco res
o n the motivation questionnaire will also have high score s on the language
 Individual differences in second language learn ing 79
78 Individual differences in second language learn ing
proficiency tests used in different stud ies do not measure the same kind of
test. If the rwo variables (motivat ion and lang uage proficiency) are found to
knowledge. For example, IQ ma y be less closely co rrelated to measures of
be correlated. the researcher will try to d iscover just what the relationship
conversational fluency than to tests that measure rneral inguisrlc knowledge.
between th em is. No te that co rrelatio ns m ay be pos itive or nega tive. That is.
o ne may find a panern su ggesting th at learn ers with highe r motivation scores Research o n individual di fferences must also take into account the soci al and
have hig her lan guage pro ficiency sco res (a positive co rrelatio n), 0 r 0 ne migh t , educational settin gs in which lear ners find themselves. Bonny Norto n and
in some circ ums tan ces, find th at lear ners w ith lowe r motivat io n scores do Kelleen Toohey (200 1) argue that, even when in dividuals po ssess so me of the
berter o n pro ficien cy measu res (a negative cor relation). charact eristics th at have bee n associa ted with successful lan guage learning .
their lan gu age acq uisition m ay not be successful if they are no t able to gain
Alth ough rhe correlation procedure seems scraigh tforward, it req uires car eful
access to social relationships in situations where they are pe rceived as value d
interp reta tio n. One pro blem is thar , unlik e varia bles such as heig ht or age ,
partnets in communication. Members of some immigrant and m inority
it is not po ssible to .d irecd y observe and measure motivation, extroversion,
grou ps are roo often marginalized by social and educational practices that
or even intelligence. Th ese are JUS t labels for an ent ire rang e o f behavio urs
limit th eir oppo rtunitie s to engage in co m municatio n wi th peers , colleagu es,
and char acte ristics. Fu rt hermore, characteristics such as th ese are no t inde­
and even teachers . In th ese social cond itio ns, individ uals who ap p roac h a
pendent of each ot he r, an d research ers have so me times used the same lab el
new langu age wi th the cognitive and motivational characte rist ics ty pical of
to describ e different sets of behavioural traits. Fo r exam ple, in motivation
su ccessfu l langu age lear ners m ay not achieve th e profi ciency th ar these char­
q ues tionnaires, learners may be "asked how often th ey use rheir second lan­
guage outside a classroom context. The assumptio n behind the question is acte ristics would predict.

that those who repo rt th at th ey frequen tly do so are h igh ly mo tivat ed to U nde rstandi ng the rela tio nsh ip between individ ual characte ristics, social sit­

lear n. Thi s seems reaso nable, but it is not so simple. If a learn er responds that uatio ns, and success in second lan guage learning is a' chal lenge, Nevertheless.

he or she frequen rly in teracts with sp eakers of the second language, it may research in this area is of great importance to both researchers and educato rs.

not be because he or she is mo re mo tivated to learn than o ne who repo rts less Researchets seek to know how di fferent cognitive and personality variables

interaction. Ra ther, it m ight be th at th is indi vid ual lives where the re are mor e are related and how they interact with learne rs' experiences, so th at they can

o pporrunities-or a grea ter necessity- fo r language p ractice th an th ose who gain a better un derstand in g of h um an learning. Edl)c.ators ho pe to find ways

report a low freq ue ncy of inte ract ion. Beca use it is usual ly im possible to of hel ping lear ners with different ch aracteristi cs to achieve success in second

separate thes e two variables (i.e. motiva tio n or desire to interact and oppo r­ lan guage learn ing . The larger community is also cdncerned because of th e

tu nities or the need to interact) , we cannot co nclude whethe r it is motivation, enormous impact second language learning has on'shaping op porruni ties for

necessity, or opportunity th at is bein g measured by th is q uestio n. ed uca tio n, em ployment, mobility, and other societal benefits.

Perhaps the most serious erro r in interp reting co rrela tions is th e conclusion Let's look at some of the in div id ual characcer lstlcsfhar have been investigated

that one o f the var iables cau ses the other. The fact tha t rwo things tend to in the effor t to discover explanations for differences in learning outcomes.

occur together or increase and decrease in a similar pattern does not neces­
sari ly mean tha t o ne ca uses the other. While it may be that one variab le
Intelligence {
infl uences th e o ther, it ma y also be th at both are influenced by something else
enti rely. Researc h on mo tivation is perhap s the best co nt ext in which to illus­ Th e term ' intelligen ce' has trad itionally been used to refer to performance on

trate this. Learners who are successfu l may indeed be h ighl y motivated . But certain kinds of tests. These tests are oft en associatM-with success in school,

can we conclude th at they became successful because of the ir mo tivatio n ? It and a lin k between intelligence and seco nd language lear n ing has sometimes

is also pla usible that early success heigh tened th eir mo tivation, or that bo th bee n reported. O ver the years, some research has shown that IQscores were

su ccess an d motivation are due to their special aptitude for lan guage learn ing a goo d means of p redicting suc cess in second language learning. H owever,

o r th e favourable context in which th ey were lear n ing. as suggest ed above, IQ test s m ay be more stro ngly relate d to metali nguis tic

knowledge th an to communicative ab ility. For exam ple , in a study w ith stu ­

Another diffic ulty in assessing rh e relatio nship between in d ivid ual learner dents in French im m ers ion p rog ramm es in C anada, Fre d Genesee (197 6)

characteristics and second language learning is how lan guage proficiency is foun d th at, while intelligence was related to th e development of Fren ch

defined and measured . In the L2 learning liter atu re, so me studies repo rt th at seco nd language readi ng, gra m mar, and vo cab ulary, it was unrelated to oral

learn ers with a highe r intelligen ce q uotient (IQ) are mo re success ful language productio n skills. This sug gests that th e kind o f abi lity meas ured by tr adi ­

learners th an those with a lowe r IQ , w hile othe r studies report no suc h cor ­ tio nal IQ tests may be a str ong pre dicto r when it comes to learning that

relatio n. One explanatio n fo r the se conflicting find ings is tha t the language

80 Individual differences In second langudge learning
invo lves lan guage anal ysis an d rule learning b ur may play a less important
role outside me classroom o r in classrooms where rhe instruction focuses
more on communication and interaction. Indeed, many students whose
general academic performance is weak experience considerable success in
second language learning if they are given me right opportunities.
Many ed uca to rs have been influenced by Howard Gardner's (I 99 3) pro ­
posal mat individuals have 'm ulti ple intelligences' and that traditional IQ
rests have assessed only a limited range of abilities. Among me 'm ultiple
intelligences' Gardner includes abilities in the areas of music, interpersonal
relations, and athletics, as well as me verbal intelligence that is most often
associated with success in school.
Language learning aptitude
Specific abilities thought to predict succes s in language learning ha ve been
stu d ied under the titl e of language learning 'a ptitu de'. One of the pioneers
in th is area, John C ar roll (199 1), has cha racte rized aptit ude in te rm s of the
abi lity to learn q u ickly. Th us, we may hyp othesize rhar a learner wirh high
ap titude may learn with greater ease and speed bu t th at o rhe r learners may
also be successful if they perseve re.
Over several d ecades, me most widely used aptitude measures have been the
Modern Language Aptitude Test (M LAT) (Car roll and Sapon 1959) and the
Pimsleur Language Aptitude Battery (PLAB) (Pimsleu r 1966). All me rests
are based on me view mat aptitude has several components, for example. the
ability to identify and memorize new sounds , understand the function of
particular words in sentences. figure our grammatical rules from language
samp les, and remember new words . \'V'hileearly research revealed a su bstantial
relationship between performance on the MLAT or PLAB and performance
in foreign language learning, these studies were conducted at a time wh en lan­
guage teaching was based on grammar translation or audiolinguaJ methods.
With me adoption of a more communicarive approach to teaching, many
teachers and researchers came to believe th at the abilities rargeted by these
tests were irrelevant to the process oflanguage acquisition. However, others
suggest that some of the abilities measured by aptitude tests are predictive
of success even in settings where the emphasis is on communicative inrerac­
rion . For example, Leila Ranta (2002) found that children who were good at
analysing language (one componenc of aptitude that is targeted by the tests)
were me most su ccessful lear ne rs in an English second lan guage programme
in which activities almost never in volved direct attention [Q grammar.
Nick Ellis (200 1) and orhers have hypothesized rhat working memory
(~I) capacity may be the mosr important variable in predicting success for
learn ers in many language lear n ing situations. Working memory, also called
Individual dlffi7'ences in second Language learning 81
'sho rr- rerm mem o ry' refe rs to th e active processing o f info rma tio n. Althoug h
long-term memory ca paci ty is very large indeed, working memory capacity
is limited. Thar means tha t only a certain amount of information can be
processed at a given rime, and individuals d iffer in me amount of informa­
rion they can process in working memory. Pet er Skehan (1989) suggests mar
successfu l lan guage learners need not be stro ng in all of th e co m po ne nts of
aptitude. For example, some may have strong memories but onl y average
abilities in language analysis.
Lear ners' strengths and weaknesses in thes e different aptitude components
may account for their ability to succeed in different rypes of instr uctio nal
programmes. In a C anadian language programme for adult learners of
French, Marjorie Wesche ( 19 8 1) studied the progress of students who were
placed in instructional programmes that we re eith er compatible or inc om­
patible with their aptitude profile. In the compatible groupings, students
with high analytic ability but average memory were assigned ro teaching that
focused on grammatical structures: learners with good memory but average
analytic skills were placed in a class o rgan ized around th e Fu nctional use o f
rhe second lan guage. In the inco m patible gro up i ngs~ students were placed in
classes that d id no t co rrespo nd to the ir aptitude profiles. \'V'esche reported a
h igh level o f stu de nt and teac her sarisfaction whe n stude nts were mat ched
with co m patib le teach in g env iro nments . In ad d itio n, som e evide nce indi­
cated that matched students were able to attain sigpi fican tly high er levels
of achievement than those who were mi smatched. Wh ile few schools could
o ffer such choices to their students, teachers may be able to ensure that their
teaching activities are sufficiently var ied to accommodate learn ers with dif­
fere nt aptitude profiles.
Furthe r su pport for the claim mat a particular. type of instruction can no t
benefit all learners in the same way co mes from a. study with seco ndary Stu­
dents of French as a foreign language in New Zealand. Rosemary Erlam
(2005) explored whether there was a relarionship.berween aptitude and the
effectiveness of three different types of instruction, which she called deduc­
tive, inductive, and structured in puc. Students were assessed on three measures
oflanguage aptitude: language analytic ability, phonemic coding ability. and
working memory. They were then divided into three groups and given differ ­
ent rypes of instrucrion on direct object pronouns in French.
Learners in the deductive instruction group received explicit ru le-based
grammar instruction followed by the opportunity to practise me rules they
had learned. Learners in me inductive group received no grammar instruc­
tion; instead they partiCipated in activities that encouraged them to figure
out the different meanings conveyed by direct object pronouns and men to
produce them. Learners in the structured input instruction group received
explicit rule-based grammar instruction but did not produce me target
82 individual differences in second language learning
forms. Insread rhey participated in activities that exposed them to spoken
and wr itten examp les of direct ob ject pronouns.
Ed am fo und that all learners benefited from the deductive instruction
regard less ofd iffere nces in apti tude. This was interp rete d as suP POrt for Peter
Skehan's (1989) hy po thesis rha r more srruc tured teaching m ay even o ut ind i­
vidual differences compared wi rh less st ructu red teaching. Edam's findings
also showed that lear ners wirh grea ter lang uage an alyr ic ab ility and memory
capacity we re able to benefit more fro m the in ductive and structu red in p ut
instruction o n writren (bur nor oral) tests. This suPPOrtS rh e hyporh esis that
learn ers w ith greater aptitude can fig ure OUt rhe rules of language based on
in put, an d that they are able to consolidate th is knowledge without rhe need
to pro duce language- at least in terms of their written ability.
Before we leave the ropic o flanguage learn ing apti tude, it is perhaps appro­
priare ro look at rwo extremes of the aptitude co ntin uum . Some p eop le
whose academic performance is usually very goo d find the mselves frus tra ted
in th eir attemp ts to lear n a foreign language. Lenore Ganschow and Richard
Spa rks (2001) an d the ir colleag ues have studied many cases ofyoung ad ults
who find fore ign la nguage learning exceedingly diffi cult . They identified
several ways in which rhese students differ fro m su ccessful learners. Most
perfo rm poo rly on at least some o f the m easures th at make up ap titude tests.
Some have pro blems wi th certain kinds of verbal ski lls, even in th eir own lan­
guage. What is perhaps most important about this research is that, with great
effort and instru cti onal suppOrt, some of th ese students are able to su cceed
in spite of th eir diffi cu lties. Th e challenge is ro find instructional approaches
th at m eer th e needs o flearn ers w ith a variety ofapti tu de profiles.
individual differences in second language learning 83
At th e orher end o f rhe aptitude continuum we find individuals wh ose
achievements seem to defy every prediction about what is possible in seco nd
language learning. Lo rraine Obler ( [ 989) repo rted on the case o f one
American man who seemed ab le ro acquire oral fluency in a new language
in 'a matter ofweeks '. Neil Smirh and Ian thi-Maria Tsimpli (1995) followed
a polyglot savant who learned many lan guages wirh appare nt ease. This
achievement was part icular ly asto nishi ng in ligh t of rhe fact char h is overall
cognitive fu nct io ning and soci al skills we re q ui te lim ited .
Such exceptional learners suggest that an aptitude for language learning is
at least partly independent of cognitive, social, and personality characteris­
tics that are often assoc iated wirh su ccessful learning. N evertheless, M ichael
Erard's (2012) review of the case s ofsome of h isro ry's mo st success ful learners
ofmultiple lan guages shows th a t thei r un us ual talent was also associated with
a w illingness to work hard at tasks that m any wo uld consider roo .bori ng or
diffi cul t, su ch as using word cards to study vocab ulary.
Learning styles .'
Some resea rchers hav e investigated individual diffe~e~~'e s in terms of ' learn­
ing style', de fined as an ind ivid ual's 'natural , hab itu al, and p referred wayis)
of abso rbing, processing , and retai ning new informatio n and skills' (Re id
1995: viii) . We have all heard pe o ple say t ha t rhey cannot learn something
until th ey have seen it. They would fall into the gro\:ip called 'visual' learn ­
er s. O thers, who m ay be called 'aud ito ry' learners .jseern to learn best 'by
ear '. For others, referred to as ' kin aest hetic' lea rness.rphysical action su ch as
mimin g or role-play seems to help rhe learning process, These are refe rred to
as perceptually-based learning styles. Co nside rab\~ research h as also focused
on \d istinctions berween di fferent cog ni tive learning styles. Individuals
have been described as field independent or fiel~ dependent, according to
whether th ey rend to separate details from th e general background or to see
rhings m o re holis tical ly. A typical measure of this ¢ gni tive style is the embed­
ded figures test, in which participants are asked to find a simple geo m etr ic
shape embedded in a more co mplex one. For a num~~ ofyears , it was wid ely
reported th a t rhere was a strong relatio nshi p be rween' field independ ence an d
success in second lan guage learni ng. H owever, a review of the research led
Z oltan Dornyei and Peter Skehan (200 3) to conclude that more research will
be needed to iden tify the nature of the relatio nship.
84 Indiuidual difJerences in second L.mguage learning
There are many questions abou t how learning.styles interact with success in
language learning." For on e thing, it is difficult ro determine wh eth er th ey
reliecr immutable differen ces o r wh ether they develop (and thus ca n be
ch.mgcd) rhr ou gh expe rience . There is a need fo r co nside rably more research .
Neve rt heless. w he n learners exp ress a preferen ce for seeing so methi ng wri tte n
o r spe nd ing more tim e in a language labor at or y, we sho uld not assume th at
their ways of working are wrong, even if the y seem ro be in conflict with th e
pedagogical approach we have adopted. Instead, we should encourage learn ­
ers to use all means available to them. At a minimum, research on learning
sty les should make us sceptical of claims th at a single teaching method o~
textbook will suit the needs of all learners .
Personality
A number of personality char acteristics have been proposed as likel y to affect
second language learning, but it has not been easy to co nfirm in empiri­
cal stu d ies. As with other research investigating the effects o f individual
characterist ics on second language learning, studies of a similar personality
trait produce different results. For example, it is often argued that an extro­
ve rted person is well suited to lang uage learning but research does not always
support this conclus ion. Although some studies have found th at success in
language learning is co rrelated with learners' scores on questionnaires meas­
uring characteristics associated with extroversion such as assert iveness and
ad venturousness, others have found that many successful lang uage learn ers
would not get h igh sco res on measures of extroversion. Lily Wong Fillmore
(1979) observed that, in certain learning situations, the quiet observant
learne r may have greater success.
Another aspect of personality that has been studied is inhibition. It has
been sugges ted that inhibition discourages risk-taking, which is necessary
Indioiduul dijferenCl!s in second language learning 8';
for progress in language lear ning. This is ofte n co nsidered to be a particul ar
problem for adolescents. who are more self-consciou s than younger learn­
ers. In a series of studies in the 1970s, Alexander Guiora and his co lleagues
( 1972) found suPPOrt for the claim that inhibition is a negative force, at
least for second language pronunciation performance. One study involved
an analysis o f m e effects of small doses of alcoh ol, know n for its ab ility to
reduce inhibition, on pronunciation. Study participants who drank small
amoun ts o f alcohol did better on p ronunciat ion tests than those who did not
drink an y. While results su ch as th ese are interesting, they ma y have more
to do with performance man with learning. We may also note, in passing,
that when larger doses of alcohol were administered , pronunciation rapidly
deteriorated!
Learner anxiety- feelings of wor ry, ner vousn ess, and st ress th at man y Stu­
dents experience wh en learning a second lan guage -has been exte nsively
investigated . For a long time. resear chers thought of anxiety as a perma­
nent fea ture o f a learner's personality. In fact, th e majori ty of lan guage
anxiety scales like th e Foreign Langu age C lassroom Anxiety Scale (Horwitz.
Horwirz, and Cope 19SiJ) measure a nxiety in th is way. So, tor example. StU­
de nts are assu med to be 'anxious' if they 'stro ng ly ag ree' with statements suc h
as ' I become nervous when I have to speak in the second language classroom'.
However, suc h qu estionnaire responses do not take acco unt of the possibility
tha t anxiety can be tempor ary an d co ntex t-spec ific..
.
Other resear ch ers in vestigating learner anxie ty in seco nd language classro oms
see anxie ty as dynamic and dependent o n par ticul arsituations an d circ um­
stan ces. Thi s perm its distincti ons to be made between tor example, feeling
anx ious when giving an oral presentation in front 'of the w ho le class but not
when interacting with peers in gro up work. \'«hatever th e context, anxiety
can 'Inrerfere with th e learning process. Peter MacIntyre (199 5) argues that
' because nervous stu d ents are focu sed o n both the task at hand and their
reactions to it . .. [they ] will not learn as qui ckl y a~ relaxed stu de nts' (p, 96).
O f course, it has also been argued th at not all an xiety is bad and that a certain
am ount of tension can h ave a positive effect and ~yen facilitate learning.
Experiencing anxiety before a test or an oral presentation can provide the
right combination of motivation and focus to succeed. Because anxiety is
often considered to be a negative terrn, so me resear chers have chosen to usc
other terms they co nsider to be more neutral . In a study of yo un g adults
learning French in an int en sive summer programme, Guy Spielman n and
Mary Radnofsky (200 I) used the term 'tensio n'. The y found chat tension , as
exper ienced by the learners in thei r stu dy, was perceived as both beneficial
and det rimental and that it was also related to rhe learn ers' social interactions
inside and outside th e classroom.
86 Individual differences in second language learning
A learner's willingness to communicate (WTC) has also been related to
anxiety. We have all experienced occasions when we tried to avoid commu­
nicating in a second lan guage. wrc may chan ge with the number of people
p resent. the topic of conversation, the fo rmality of the circumstances, and
even with whether we feel tired or energetic at a given moment. A collea gue
in C anada, w ho works in the area of second lan guage learn ing an d speaks
sever al lan gua ges, recently co nfessed tha t he avoided the corner sto re in his
neigh bourhood because th e p ro p rieto r always spo ke Fren ch to him. He rec­
ognized the p rop rieto r's efforts to help him improve his skills in this new
language, and was grateful for it, but, as he told us with embarrassment, it
was just easier ro go ro the store where he coulduse Eng lish.
This is consistent with research car ried OUt by Richard Clement, Peter
Maclnryre, and their colleagues, who argue that lear ne rs who willingly
com m un ica te in a wide range of conversational interactions are able co do
so beca use o f th eir comm unicative confidence. In a series' of studies they
have shown th at communicative confidence is shaped by two variables: how
relaxed L2 1earners are and how com petent (or incompetent) th ey feel about
their L2 ability. These factors are di rectly in fluenced by prev ious contacts
with L2 speakers and are considered to be the mai n conrribu ro rs to co m mu­
nicarive co nfidence (Clement, Baker, an d MacIntyre 2003).
Several other personality characteristics such as self-esteem, em pathy, domi­
nance, talkativeness, and responsiveness have also been studied. The research
does no t show a clearly-defined relat ions hip betwe en o ne pe rson ality trait
and second lan guage acq uisition. And, as indica ted earlier, the majo r d if­
ficu lty in inve stigating perso nality characteristics is that of idenrlfication and
measurem en t. Ano ther explanatio n has bee n offered for the mixed findings .
Personality variables seem to be more consisten tly related to co nversational
skills than to the acquisition of grammatical accuracy or academic language.
Finally, most of the research on personality variables has been carried out
withi n a quantitative research paradigm, that is, an approach that relies
heavily on relating learners' scores on personality questionnaires to th eir
language test performance. Some research ers have argued that a more quali­
tative approach to understanding an d investigating personality variables is
needed to adequately cap ture th eir depth and complexity. especially as they
emerge and evolve over time.
Despite th e contradicto ry resul ts an d th e problems invo lved in carrying ou t
research in th e area of personal ity ch aracteristics . many researchers believe
that personality will be shown to have an important influence on success in
lang uage learning. This relationship is an intricate one, however, in that it is
pro bably not perso nali ty alone, but the way in which it combines with o ther
factors, tha t influences second language lear ni ng .
Individual differences in second language learning 87
Attitudes and motivation
Robert Gardner and his coll eagues have carried out a programme of research
on the relationship between a learner's attitudes toward the second or foreign
language and its community. and success in second lan guage learning
(Masgorer an d Gardne r 2003) . As is the case with other variables, it is not
easy to determine whether posi tive attitudes prod uce successfu l learn in g or
successfu l learning engenders positive attitudes, or whether both are affecte d
by other factors. Although the researc h canno t prove that positive attitudes
cause success in learning, there is ample evidence th at positive attitudes are
associa ted with a willingness to keep lear n in g.
Motivatio n in second language lear nin g is a co mplex phenomeno n. It has
been defined in terms of rwo factors: on th e one hand, learners' communi­
cative needs, and on me other. their atti tudes towards the second language
community. If learners need to speak the second language in a wide range
of social situations or to fulfil professio nal ambitions, they will perceive th e
comm unicative value of the second language and are therefore likely [Q be
motivated [Q acquire proficiency in it. Similarly, if lear ners have favourable
attitudes towards the speakers of the language, they ;""ilhiesire more contac t
with them. Robert Gardner and Wallace Lambert (1972) coined th e terms
instrumental motivation (language learning for im mediate or practical goals)
and integrative motivation (language learning for personal growth and cul­
tu ral enrichment through contact with speakers of th~ other lan guage). For a
long time int eg rat ive motivation was considered [0 bethe stronger predi ctor
of suc cessful learning. In some contexts, however, instrumental motivati o n
was found [0 be a bett er predictor. Thus, both types-of mo tivatio n have been
foun d [Q be related [0 success in second language learni ng, However, in some
learn ing environments, it is difficult [0 distinguish' between thes e twO types
of orientation [Q the target language and its community. Fu rthermore, early
research tended [0 conceptualize motivation as a 'st able characteristic of the
learner. More recent work emphasizes th e dy na mi 4 nature ofmotivation an d
tries [0 account for th e changes th at take place over time.
Zoltan Dornyei (2001 a) developed a pro cess-orientqd model of motivation
mat consists of three phases. The first phase, 'choice motivation' refers to
getting started and [0 setting goals . the second phase, 'executive motivation',
is about carrying out the necessary tasks [Q maintain motivation, an d the
third phase, 'motivation retrospection'. refers to students' appraisal of and
reactio n [0 th eir performance. An example of how one might cycle th rough
these phases would be: a seco ndary school learner in Poland is excited abo ut
an upco m ing tri p to Spain and decides to take a Spanish course (choice moti­
vation). After a few months of grammar lessons he beco mes frustrated with
the course, StOps going to classes (execu tive motivatio n) and fin ally decides
to dro p the co urse. A wee k later a friend tells hi m about a grea t Span ish

88 Indiuidual diffirem:esin second Lmgu.tge f.:llnz ing
co nver sation co u rse she is taki ng, and his 'cho ice motivation' is activat ed
again. He dec ides ro register in m e conversation course an d in JUS t a few
weeks he develops so m e basic Spanish co nversatio nal skills and a feeling of
acco m plishmen t. H is satisfactio n level is so positi ve (mo tivation ret rospec­
tion) that he decides (Q enrol in a more advanced Spanish co urse when he
returns from his trip (Q Spain.
In a book devoted (Q helping second language teachers generate and maintain
learners' mo tivation. Dornyei (200 I b) pr op oses and descri bes conc rete and
innovative rnethods and techniques that can help teachers motivate learners
throughout these three phases.
M otivation in the classroom
In a teac her's mind, motivated studenrs are usually those who participate
actively in class, express interest in the subject matter. and srudy a great deal.
Teachers also have more infl uence on these behavio urs and the motivation
they represent than o n students' reason s for study ing the second language
o r their arrirudes coward the lang uag e a nd its speake rs. Tea che rs ca n make a
pos itive co ntr ib ut io n co stude n ts' m ot ivati on co learn if classr ooms are pla ces
that st ude nts enjoy co m ing co becau se the co n tent is inte resting and relevan t
co th eir age and level o f ab iliry, the learning goal s a re cha lleng in g yet manage­
able and clear, and the atmosphere is supportive. Teachers must also keep in
mind that cultu ral and age differences wi ll de termine the mos t appropriate
ways for them (Q motivate students.
Little research has in vestigated how pedagogy interacts directly with rnori­
va tio n in second/fo reig n lan guage classrooms. One excep tion is a study
by Marie Guilloteaux and Zolran Dornyei (2008) who explored the links
between teachers' motivatio nal practice and sruden ts' motivation for L2
learni ng. It was a large-scale study with 27 teachers a nd over 1,300 learners
in English as a Foreign Language (E FL) classrooms in Korea, The teach­
ers' mo tivat ional strategies were described using a classroom observation
scheme-the Motivation Orientation of Language Teaching (MOLT) .
MOLT iden tified 25 m otivat io nal p ractices used by the teachers that we re
relatively easy co de fine and co o bserve. They were divided into four ca t­
ego ries mat are descri bed below alo ng with exam ples o f the mo tivational
behavio urs included w ithin each.
1 Teac her d isco urse: arousing curiosity or atte ntion, promoting autonomy,
stating co m m un icative purpose/ utility of activity
2 Participation structure: gro up wo rk/pai r wo rk
3 Activity design : individ ual competitio n, team competition , in rellecrual
challenge. ta ng ib le task product
4 Encou raging positive retrospective self-evalua tio n an d ac tivity design:
effective p raise, elici ta tio n of self/peer correction sessio n, class app lause.
Individual diffirences in second language learning 89
ln each lesson , th e learn ers' m otivation was m easured in terms of th eir level
ofengagemenr. The proportio n ofstudents w ho paid atte ntio n , who actively
participated, an d who eagerly voluntee red d uring activities was calculated.
A three-level scale was used to measure engagement in each observed lesso n:
very low (a few students) , low (o ne third to rwo thirds of me students) and
high (mo re man rwo thirds o f the stude n ts}. Learners also co m pleted a ques­
tionnaire about their motivation levels specifically related co their EFL class.
The researchers fo und significant positive correlations between the teachers'
motivational practices, the learners' engagement behaviours, and me learners'
self- repo rt s on the questio nnaire. The resea rche rs acknowledge that corre­
lation results do not indi cate cause-effect relationships , Nevertheless. the
findings are important because this is me first study (Q provide 'any empiri­
cal evidence concerning the concrete. classroom-specific impact oflanguage
teachers' motivational strategies' (C uillo reaux and Dornyei 2008: 72) .
Identity and ethnic group affiliation
Social Fa ct o rs in the wider comm u nity can also affe.:: t m ot ivatio n . alt itu des,
and lan gu age learn ing success, One suc h facto r is th e social dynamic or
powe r relarionship betw een lan gua ges. Fo r exa m ple, me m bers ofa m in ori ty
gro u p learning the language o f a maj oriry group may ha ve differen t attitudes
and mo tivation from those of majority group members learn ing a minority
language. Even though it is impossible to predict the exact effect of such
societal factors on second lan guage learn ing , the facc'that languages exist in
social COntexts cannot be overlooked when we seek to unde rstan d th e vari ­
ables that affect success in learn ing. Children as well as adults are sens itive to
social d ynamics and power relationships . .' .
A g60d example of how relations of power in the ~o cial world affect interac­
rion between second lang uage lear ners and ta rget language speakers comes
from the work of Bonny Norton Peirce, Drawing from data collected in a
lo ngitu d inal case study of the lan guage lear n ing ;experiences of immigrant
women in Canada. she arg ues that concepts such as instrumental and inte­
gra tive mo tivatio n do nor ad equately cap ture the co rfip lex relati ons ofpower,
identity, a nd language learning, Instead, she uses the te rm ' lnvestrnen c' to
'cap ture th e relati o nshi p of the language lea rn er [an d his/ her idenrity] co the
chan g ing social worl d .' (Norton Peirce 1995: 10). All the participants in her
study we re highly motivated to lear n English . H o wever, the re we re social
situations in which they were reiucranr to speak and these were typically o nes
in which t here was a power im balan ce. The ir experiences in th ose situations
limited the opportun ities they had to practise and to conti nue to develop the
seco nd language outside the classro om .
Worki ng w it h immigranr c hil dren in English-medi um ki ndergart en classes,
Kellee n Toohey (2000) observed th a t they were q u ickly assig ned iden tities


90 Individual differences in second language learning
such as successful/unsuccessful. big/small, talkative/quiet. etc. in their first
year of schoo!. Of course, they also had the identity of ,being ESL'. Because
learners' identities impact on what th ey can do and how they can participate
in classrooms. this naturally affects how m uch they can lear n. For example.
one of the 'ESL' children was consistently excluded from imaginative interac­
tive activities with her peers; another learner was per ceived as so meone who
never listened or did the ' right thi ng'. Toohey argues th at th ese identi ties
could eventually lead to their isolation and to restricted or less powerful par­
ticipation in their classroom community. While Toohey is careful to point
OUtthat identities are not static and can change over time. it is equally impor­
tant to keep in mind that 'classrooms are organized to provide occasions
upon which some children look more an d some less able, an d jud gements are
made which become social facts about individual children' (Toohey 2000:
77).
Th e two stu d ies above describe how issues of ide ntity and investm ent play
im portant roles for both children an d ad ults when learn ing a second lan­
guage. Research has also do cu ment ed how these factors contri bu te in
complex and sometimes contradictory ways when learning a fo reign lan­
guage. For example, it has been observed that Japanese students are often
reluctant to speak English in communicarive lessons despire hi gh levels of
mo tivation to learn the lang uage. Furthermore, w hen students with high
levels of English lan gu age pro ficiency do communicare they often speak with
a strong Ja panese acce nt and int entio nally produce grammatical errors for
fear th ar th ey mig hr be perceived as considering rhemselves to be superior
(Greer 2000).
In a stu dy wirh secondary schoolJapanese learners ofEnglish as a foreign lan­
guage . Yasuyo Tomira (20 11) observed that the students were more willing to
communicare during activities that combined a foc us on form and meaning
rh an in exclusively meaning-based activities, In observations and interviews
with the students, she concluded that learners were nor w illing to inves r in
English communication with each or her unless th ey were able to esrablish
rh eir identities as 'learners' by disc ussing language form and raisin g q ues­
tions about grammar. In exclusively communicative acrivities learners were
reluctant ro use English to communicate their ideas or opinions for fear rhar
they would be identified as 'show offs' and prerending to be someone orher
than Japanese.
Learner beliefs
Second language learn ers are nor always aware of rheir individual cogn irive or
perceprual learning styles , bur virt ual ly all learners, particularly older learn­
ers, have srrong beliefs and op inio ns abou r how rhei r instruction sh ould be
delivere d . Th ese beliefs are usually based on previo us learn ing exp eriences
Individual differences in second language learn ing 91
and the assumprion (right or wrong) that a particular eype of instruction is
the best way for them ro learn.
Research on learner beliefs about the role ofgrammar and corrective feed back
in second language learning confirms th at there is often a mismatch between
sruden ts ' and teachers' views. In rwo large-scale studies Renate Schulz (2001 )
foun d that virtu ally all students expressed a desire to have their errors cor­
recred wh ile very few teachers felt this was desirable. In addition, while mo sr
students believed that 'formal study of the language is essential to th e even­
tual mastery of the language', JUSt over hal f of rhe reachers shared rhis view.
Shawn Loewen and his colleagues (2009) asked 74 5 learners of di fferent
lang uages to express their beliefs abour grammar instruction an d cor rective
feedback. They found that some bur nor all lear ners valued grammar instruc­
rion. For example, learners of English as a second language did nor value ir as
much as did learners offoreign languages. This might have been relared to the
fact that the ESL learners had more years of previous. grammar insrruction
compared to the foreign language learners and rhus rhey were less enthusi­
astic about ir and ready fo r mo re focus o n co mm unication. Overall, learn ers
did nor value corrective feedback as much as grammar instruction except for
th ose learn in g Chinese and Arabic who valued borh. The researchers sug ­
gested that this mighr be because the two languages are non Indo-European
and th us mo re difficult for English speakers to learn than, for example,
Spanish. German. and Fre nch. j,
Nina Spada and her research team (2009) have examined not whether lear n­
ers perceive grammar instruction as useful but racherif they have pre ferences
for when it should be taught. In a series ofstudies with learners (and teac hers)
of English as a second language and English as a foreign language, questio n­
naires were administered to 450 learners, asking rhemwherher rhey preferred
to focus on grammar separarely from or embedded wirhin co mmunicative
practice. The overall results indicared that both grpups of learners preferred
integraring at tention to grammar within commun'(cative practice. However,
rhe ESL learners also reported that they val ued separaring gram m ar from
communicarive interaction m uch more than rhe E¥,,~ learners. This makes
intuitive sense given rhar in the ESL context where t here are more oppo r­
tunities for communication outside the classroom. learn ers appreciare the
opportunity to focus on grammar separately from communicario n when
they are in the classroom. In the EFL context, however. where few opporru­
nities for exposure to rhe L2 are available, th e classroom is seen as th e only
place for communication-rhus a p refe rence fo r integrating grammar wirh
communicarive practice. The results from the ESL and EFL reache rs w ho
also par ticipated in the studies showed that they rended to share their Stu­
dents' views (Spada and San tos Lima 20 10 (in preparation».
91 Indi vid ua Ldijjerences in second !<mguageLearning
Individual differences and classroom
instruction
There are many questions about how the existence of individual differences
sho uld influence instruction . On a simple practical level , it is not possible
to r a teacher wi rh 50 st ude n ts-c-o r even one with 10 stude nts-c-eo customize
instruction to suit the abilities or preferences ofeach one. Nevertheless, there
can be little doubt that an instructional approach that rigidly adheres to a
single way of teaching all studen ts and an expectatio n that all students can
learn in the same way will deprive some students of learning opportunities.
Zo lta n Dornyei (200 5) has reviewed the research on individual differences
and proposes a number ofways for educators to help learners make the mo st
o f th eir individual abilities and learning preferences.
Learners' instructi onal preferences, whether due to inherent differences in
their approach to learning or to their beliefs about how languages are learned,
will influence the kinds of strateg ies th ey use in try ing to learn new mar e­
rial. Teach ers ca n help learn ers expa nd th eir repertoi re of learni ng strategies
and rhu s devel op g reater flexibility in t heir ways of approaching langu age
learni ng.
Age and second language learning
We now turn to a learner characte ristic of a different type : th e age at whi ch
learning begins. Age is easier to define and measure than personality, apti­
tude, or moti vati on, but the relationship between age an d success in seco nd
language acquisition is hardly less complex or controversial.
It is frequently obs erved that most children from immigrant famili es even­
tuall y speak the language of th eir new community with native-like fluen cy,
while their parents often fall short ofsuch high levels ofproficiency. especially
in th e spo ke n language. To be sure. man y ad ulr second language learners
achieve excellent language skills . One often sees reference to Joseph Conrad,
a native speaker ofPolish who became a majo r writer in the English language,
and it is not uncommon to find adult second language learners with a rich
vocabulary, sophisticated syntax, and effective pragmatic skills, even though
there may be subtle differences between their language use and that of those
who began learning the language wh ile very young.
/\5 we saw in Chapter 1, the Critical Period Hypothesis is that there is a
time in human development when the brain is predisposed for success in
language learning. It has been hypothesized that there is a critical period
tor second language acquisition JUS t as there is for first language acquisi­
tion. Developmental changes in the brain , it is a rgued, affect the natu re of
language acquisition , and language learning that occurs after the end of the
Indiuidual diffire>lCt?s in second wnguage learni ng 93
crirical pe riod may no t be based o n th e innate biolog ical struc t ures believed
to co nt rib ute to first language acquisition o r seco nd language acquisition in
early childhood. Rather, older learners may depend on more generallearn­
ing abilities-rhe same ones they might use to acquire other kinds of skills
or information . It is argued that these general learning abilities are not as
effective for lan guage learnin g as th e.m or e speci fic, inna te capacities that are
available to the yo un g child. It is mo st often claimed that the critical period
ends so mew here aro un d puberty, but so me researchers suggest it co uld be
even earlier. Others find evidence that there m ay be multiple critical periods,
related to different aspecrs of language learning. For example, the ability to
acquire rhe pronunciation patterns of a new language may end earlier than
the abiliry.ro acquire vocabulary.
Of co urse, as we saw in C hap ter 2, ir is difficult to compare ch ild ren and
ad ults as second language learners. In addition to possible biological di f­
feren ces suggested by the C ritical Period H ypothesis, the conditions for
language learning are often very different. Youn ger learners in informallan­
guage learning enviro n m ents usually have m or e tim e to devo te to learning
lang uage. They o fte n have more opportunities to hear-and use the lan guage in
enviro n me nts where th ey do not experience stro ng pre ssure to speak fluently
an d accu rat ely from the very beginnin g. Fu rthe rmo re, [heir ea rly im pe rfect
effo rts are of[en prai sed, o r at least accepted. O lde r learn ers are mor e likely to
find th emselves in situations char demand more co m plex language and [he
exp ressio n of more co mp lica ted ideas . Adults are often embarrassed by thei r
lack of mastery of the language and they may develop'a sense of inadequacy
mer experiences offrusrration in trying to say exacrlywhac they mean. Such
negative feelings may affect their motivation and willingness to place them­
selves in situations wh ere they will need to use the new language.
- -r,
Research based o n the CPH in addition to personal experience or informal
observation of adult learners' difficulties has led some educators and policy
makers as well as many parents to conclude that second language instruction
is most likely to succeed if it begins when learners 'are very young. However,
some studies ofthe second language development ofolder and younger learn­
ers learning in similar circumstances have shown [ha$cHder learn ers are more
efficient than younger learners. By using their meralinguistic knowledge ,
memory strategies, and problem-solving skills, they make the most ofsecond
or foreign language instruction. In educational settings, learners who begin
learning a second language at primary school level do nor always achieve
grea ter proficiency in the long run than those who begin in adolescence.
Furthermore, there are countless anecdotes about older learners (adolescents
and adults) who achieve excellence in [he second lan guage. Does this mean
that there is no critical period for second language acquisition?
94 Indiuidua! differences in second Language Learning
The criticalperiod: More than justpronunciation?
Most studies of the relationship between age of acquisition and second
lan guage development have concluded that older lear ners typically have a
no ticeable 'fo reign accent' in the spoken language. But what abo ut other
linguistic features? Is syntax (word order, overall sentence st ructure) as
depend ent on age of acquisition as p ho nological develo pment? What about
mo rphology?
M ark Patkowski (1980) studied the relatio nship between age and the
acquisition of fearures of a second language other than pronunciation. He
hypothesized that, even if accent were ignored, only those who had begun
learn ing th eir second language befo re the age of 15 could achieve full,
na tive-like mastery of th at language. Parkowski srudied 67 highly educated
im migran ts to the Un ited Sta tes. They had started to learn English at vario us
ages, b ut all had lived in the United States for more th an five yea rs. H e
compared them to 15 native-born Americans with a similarly high level of
education, whose variety of English could be considered the second language
speakers' target language.
The main questio n in Parkowski's research was: 'W ill there be a diffe rence
between learners who began to learn English before puberty and those who
began learning English later?' However, he also compared learners o n th e
basis of o ther characteristics and experiences that some people have sug­
gested might be as good as age in predicting or explaining a person's success in
mastering a second language. For example, he looked at the total amount of
time a speaker had been in the United Sta tes as well as the amount of formal
ESL instruction each speaker had ha d.
A lengthy interview with each person was tape-reco rded . Because Patkowski
wanted to remove the possibility that the results would be affected by accent,
he transcribed five-minute samples from the interviews and asked trained
native-speaker judges to place each transcri pt on a scale from 0 (no knowl­
edge of English) to 5 (a level of Englis h expected from an educated native
speaker).
The findings were quite dramatic. The transcripts of all native speakers and
32 out of 33 second language speakers who had begun learning English
before the age of 15 were rated 4 + or 5. The homogeneity o f the pre-puberty
learners suggests that, for this gro up, success in learning a seco nd language
was almos t inevitable. In contrast, 27 of the 32 post-puberty learners were
rated between 3 and 4 , but a few learners were rated higher (4+ or 5) and one
was rated at 2+ . The performance of this group looked like the SOrt of range
one would expect if one were measuring success in learning almost any kind
ofskill or kn owledge: some people did extremely well; some did poorly; most
were in the middle.
Indiuidual differences in secondLanguage Learning 95
When Patkowski examined the other factors thar m ight be thought to affect
success in second language acquisition, the picture was much less clear. There
was, naturally, some relationship between those factors and learning success,
but it often rurn ed out that age was so closely related to the other factors that
it was not really possible to separate the m completely. For example, length of
residence in the United States so metimes seemed to be a fairly goo d predic­
tor, However, it was often th e case that those with longer residence had also
arrived at an .earlier age. Sim ilar ly, amo unt of instru ctio n, when separated
from age, did not pred ict success as well as age of immigration did. Thus,
Patkowski found that for learners who acquire a second language primarily in
the 'natural' environment, age of acquisition is an important factor in setting
limits on the development of native-like mastery of a second language and
tha t this limitation does not apply only to pro n unciation.
Intuitions ofgrammaticality
Jacqueline johnson and Elissa Newport (I989) conducted a study of 46
C h inese and Korean speakers who had beg un to learn English at different
ages. All were students o r faculty members at an American un iversity an d all
had been in the United States for at least th ree years. The sru dy also included
a comparison group o f 23 native speakers of English . The participants were
asked to make grammaticali ry judgements of a large number of sentences
that tested 12 rules of English morphology and syntax. They heard recorded
sentences an d had to indicate whe ther each sentencdwas correct. Half of th e
sentences were grammatical, half were not. .>'
Johnson and Newport found that age of arrival in the United States was
a significant predictor o f success on the test. Learners who began earliest
achieved the highest scores on the judgement task. Those who began later
were less likely to judge the sentences correctly arid their performance on the
test varied more widely, .~
of
Robert DeKeyser (2000) carried our a rep lication th eJohnso n and Newport
study, working with Hungarian immigrants to the United States. He also
found a strong relations hip between age of immig~ation and performance
on th e judgement task. In addition, he asked participants to take language
aptirude tests and fou nd that, for participants who began learning English as
adults, aptitude scores were correlated with success. However, there was no
such correlation for those who learned English in childhood. These findings
appear to confirm the hypothesis that ad ult learners may learn language in a
way that is different from the way young children learn,
96 Indiuidua] differ ences in second language learning
Rate oflearning
So me researc h suggests that olde r lear ners may have an ad vantage in term s of
the rate of learning. They appear to lear n faster in th e early stages of seco nd
lan guage devel opment. In 197 8, C atherine Sno w and Marian Hoefnagel­
H ohl e published a study on a group of English speakers who were learning
Dutch as a seco nd language while living in the Netherlands. The lear ners
included children as yo ung as three years o ld as well as older children. ado ­
lescents, and adults. On tests administered when learners had been in the
COUntryfor less than a year. adolescents were by far the mos t successful learn ­
ers. They were ahead of everyone on nearly all of the tests, Furthermore, it
was the ad ults, not the children. whose scores were second best . In ot her
words, adolescents and adults learned faster than children in th e first few
months of exposure to Dutch.
By the end of the year, the children were catching up, or hac! surpassed.
the adults on 'several measures. Nevertheless, the adolescents retai ned t he
hig hes t levels of performance ove rall. The fact that the young ch ildren were
catc h ing up, toge ther wi th evide nce fro m o ther stud ies, sugges ts [hat [hey
wo uld pr ob ably su rpass the old er learners if they co nt inued to have adeq uate
oppo rtunity to use th e language. However, thi s stu dy shows th at adults and
adolescents can make co ns ide ra ble and rapid progress in their proficiency in
a second language in cont exts where the y use the language in social, personal .
professional , or academic interaction ,
One view of critical pe riod research that has had an important impact on
th e way we look at studies of language acquisition has been exp ressed in the
work o f Vivian Cook (200 8). H e makes a strong case for the inappropriate­
ness of using th e crit er ion of Tndisringuishable from a native speak er' as the
basis for su ccess in seco nd language acquisition. Indeed, Cook argues that
a second language speaker or bilingual person should not be compared to
monolingual native speakers because the real goal is 'm ulrico rn petence' , that
is. knowledge of m ultip le languages th at in form and enrich one another.
Resear ch on the long-term outcomes of second language learni ng as well as
the rate oflearning at different ages brin gs us to a ques tio n that is pro bab ly
ofgrea test interest to most read ers of this book: What can we co nclude abour
the role of age whe n learning takes place p rimarily in an ed ucational setting?
Age and second language instruction
Many peopl e w ho have never heard of the critical p eriod h ypothesis believe
th at , in school p rog ramm es for second or fo reign lan gu age teach in g, 'yo unger
is bett er '. H owever. bo th expe rience an d research show th at sta rring early is
no gua rantee ofsuccess and tha t olde r learners can attain high levels of profi ­
ciency in their seco nd lan gu age. In considering the best age at w hich to begin
Individual differences I
second language instruc tio n, it is essent ial to think carefully about the goa ls
of an instructional programme and the co nt ext in which it occurs befor e we
jum p to conclusions about the necessiry-or even the desirabiliry-of the
earlies t poss ible start (Lighrbow n 2008a) .
As we have seen, there isstrong evidence that differences in learning outcomes
are associated with age oflearning. We have also seen that, especially for older
learners, reaching high levels of second language proficiency involves apti­
tude, motivation, and the appro pria te social conditions for learning. Thus ,
dec isio ns about the age at which instruction sho uld begin canno t be based
solely on research on the critical period hypothesis, which focuses only on
age and on the attainment of native-like proficiency.
In educational settings. it is particularly important to assess th e goals and
the resources avai lable for second lan guage development. In these sett ings,
researc h has shown that older children and adolescents progress more rapidly
than younger ch ildren particu larly in the early stages oflearning. The knowl­
edge and sk ills tha t o lde r learn ers are abl e to acqu ire in a relati vely sho rt
pe riod of tim e wi ll satisfy th e needs of many learners whose goal is to use
the lan guage for everyday com m u nication, to succeed on foreign language
exam inatio ns. or to read texts for an academic course rather than to speak
with na tive-like p ron unciation.
When the o bjective of seco nd language learning is nati ve-like proficiency
in the target language. it may indeed be desirable for the learner to be com­
ple tely surro unded by the lan guage as earl y as pos sib le. However, as we saw
in Chapter 1, earl y intensive exposure to the second lan gu age may entai l
the loss or incomplete development of th e child 's first language. When the
goal is basic communicative ab iliry for all stu dent s in an ed ucational system.
an d whe n ir is assumed that the ch ild's native language will continue to be
an important part of the ir lives. it may be more efficient to begin second or
foreign language teaching later.
In mos t second- and fore ign -language classrooms. learn ers receive only a few
hours of instruction per week. Those w ho start la ter (for example. at age 10 .
11, o r 12) often ca tch up with th ose who begin earl ier. I n Clare Burs rall's
(19 75) lan dmark study, stu dents who had made progress in early -starr pro ­
gtammes, sometimes found themselves placed in secondary school classes
with students who had had no previous ins tructio n. Teachers who had both
the mo re advanced early -start stu dents and th e students who had had fewer
total ho urs of instruction te nded to teach to a lower co mmo n de nominato r,
and d ifferences between th e rwo groups of stude nts essentially disappeared .
Th is situation is not at all u ncommon . Further mo re, in ma ny educatio nal
settings, srarti ng ins truc tio n ear lier may no t actual ly entai l many m o re total
ho urs of instructi on . For exam ple, in Quebec, resp ond ing to p ressure from
parents, the age at whic h instruction in Eng lish as a second language began
98 Individ ual differences in second language /earning
was lowered in recent years fro m abour age 9 or 10 to age 6, but m e to tal
n umber of hours of instruction was not increased. Ramer, me nwnbe r of
minutes of instruction per week was spread over more years (Lightbown,
20 12). Thus, after years of classes, learners who have had an early start may
feel frusrrared by the lack of progress, an d thei r motivation to continue may
be d im inished . C learly th e age at which insrruction begins is not th e on ly
varia ble that determi nes success in the second language classroom.
For many years, it was difficult to compare early-start an d later-star t learners
because of all the varia tions in their educational contexts. Since the 1990 s,
many more studies have allowed us to investigate this question more effec­
tively. Some large-scale research pro jects have been particularly useful in
separa tin g the effect of age and other faccors in school-based foreign lan­
guage learn ing. For example, in Spain, me Barcelona Age Faeror (BAP )
project st udied the effects of changing the age of beginning co teach English
[Q Catalan/Spanish bilingual student s.
When the starting age for teaching English was lowered , Carmen Munoz and
her colleagues took advant age of me opportuni ty [Q compare me learning
outcomes for stu den ts who had started learning at different ages. They were
able to loo k ar students' progress after 100 , 4 16 , and 726 hours of ins truc­
tion . Th ose who had begun to learn later (aged 11, 14, or 18+) performed
berrer o n nearly every meas ure than those who had begun earlier (aged 8).
This was particularly true of m easures based on metalinguistic awareness or
analytic ability. O n listening comprehension, younger sta rters showed some
advantages . Munoz suggests mat th is may be based on yo unger learners' use
of a m ore implicit approach co lear ning while older learners' advantages may
reflect their ability to use more explicit approaches, based on their greater
cog nitive maturity. She points out mat, in foreign language instruction,
where tim e is usually lim ited , 'younger lear ners may not have enough time
and exposure [Q benefit fully from me alleged advan tages ofi m plicit learning'
(M unoz 2006: 33 ). .
One of me advan tages of me BAF project is mat the researchers were able
to follow the same learners' language develop ment over several years. Th is
enabled th em to examine whether the early lear ners would eventually surpass
the older learn ers as has been observed in the 'natu ral' setting. This did not
happen-although me you nger learners caught up, me older lear ners main­
tained their advantage over time.
Decisions about when to starr second language ins truction in schools should
be based on realistic goals an d on realistic estimates of how long it takes [Q
achieve the m . O ne or two hours a week will not produce advanced second
language speakers, no ma tter how yo ung the y were when they began . Older
learn ers m ay be able [Q make berrer use of the lim ited time the y have for
second language instr uction.
Individual differences in second language learn ing 99
Age is only one of m e characte ristics tha t det ermine m e way in which an
indi vidual approach es second language learning and the event ual success of
mat lear ning. The opportunities for learning (both inside and outside the
classroom) , me motivation to learn, and individual differences in aptirude
for language learning are also important factors m at affect both rate oflearn­
ing and eventual success in learning. It is important to remind ourselves mat
some older learners do achieve me highest level of success and mat many
more are able co use their languages in a vari ety of personal, social, an d work­
place activities.
ACTIVITY Reflect on individual differences and
language learning success
Look back at the notes you made inTable 3. 1 about your language learning
experience and that of your colleagues and friends.
Which cases confirm your expectations about th e variables that are
associated w ith success--or the lack of it- i n second language learning?
2 Which ones seem to challenge those ex pectations?
3 To what extent do you think t hat the contexts in which the learning took
place are responsible for the outcomes?
4 To w hat extent do you t hink t he differe nces in o utco me are due to
differences inherent in t he individuals?
S Do you think t hat the individuals might have experienced different
outcomes in different situations?
Summary
In this chapter, we ha ve learned mat me results of research on indiv id ual
differences are not always easy to interpret. Thi s is partly due [Q m e dif­
ficulty of defining an d measuring individual characteristics and to m e fact
m at the characteristics are no t independent ofon e another. In addition, rela­
tionshi ps between individual characteris tics an d learning en viro nments are
complex, and different learners will react differently to the same learning
conditions . Indeed , me same learner will react differently co th e same co nd i­
tions at different times. Researchers are beginning [Q explore the nature of
these complex interact ions , but it remains difficult to predicr how a par ticu ­
lar individ ual's characteristics will in fluen ce his or her success as a language
learner. Nonetheless, in a classro om , me goal of the sensitive teacher is co
create a learn ing environment w ith a wid e variety of instructio nal activities
so that learners with different abilities an d learni ng p refere nces can be suc­
cessful in learning a second language .
100 Indiuidua! differences in second l.mguuge learning Individual differences in second language learning 101

Robinson, P. (ed) . 2002 . Individual Differences and [nstructed Language

Questions for reflection
Learning. Am sterdam: John Benjamins.
Think of an exa m ple of a member of a majori ty group learning the language The co n tribu to rs co this edited collection focus on interactions between
of a minority grou p and on e of a member o f a minority group learni ng a individual learner characteristics and learning contexts. The chapters in
majority group's language . How might the power relationships betw ee n the first sectio n focus o n theoretical wo rk related to ap titude, motiva­
groups of speakers affect the attitudes of language learners? How might the tio n, anx iety, and emo tion . Eac h cha p ter in the seco nd sectio n descri bes
status of the languages affect opportunities for learning? research investigating how individual learner variables in teracr with a par­
2 As a second/foreign language teacher or learner, what are your views about ticular learning context to affect L2 learning. Thi s includes classroom an d
teaching grammar? Do you have any specific preferences fo r how it should laboratory stud ies examining learner varia b les in relation to different rypes
be taught or when! Do you know what your students' preferences might of instruction and stu d ies of natural versus instructed L2 learning.
be for grammar teaching. If not. do you think it would be useful to find out!
3 Ifyou were teaching English as a foreign language in a country with limited
opportunities for secondary and post-secondary education in English. what
recommendations wou ld you make regarding the age at which English
instruction would begin! What research would you draw on in supporting
your recommendations?

Suggestions for further reading

Dornyei, Z. 2005. The Psychologyo/the Langu age Learner: Individual
Differences in Second Language Acquisition. Mahwah , NJ: Lawren ce
Erlbaum and Associates.
Dornyei reviews decades of research on how individual differences affect
seco nd language learning. The book covers personality variables, aptitude,
motivation, learning sryles, learning strategies, and other individual char­
acteristics such as anxiety and willingness to communicate that may vary
according to me learning environment. Both thorough and accessible, this
review concludes by emphasizing the evidence mat individual differences
are strongly affected by the situation in which learning takes place ramer
than being 'context-independent and absolute.'
Munoz, C. (ed.). 2006. Age and the Rate 0/Foreign Language Learning.
Cleve don: Multilingual Matters .
The Barcelona Age Factor study is the basis of this edited volume. Ten
chapters report on var ious aspects of students' learning of English, com ­
paring the outcomes for students whose foreign language instruction
began at different ages. In addition ro me specific research reports on, for
example. me students' oral fluency, vocabulary, and rate of/earning, there
is an overview chapter in which Carmen Munoz. the projecr director and
editor of the volume, discusses the project in terms of broader issues ofage
and language lear n in g at school.
iI

 EXPLAINING SECOND

LANGUAGE LEARNING

Preview
A general theory o f seco nd language acq uisition needs to account for lan ­
guage acq uisitio n by learners with a variety of characteristics in a variety of
co ntexts . In th is chap ter we examine some of th e theories that have been
offered ro account for second language developmental progress and final
learning outcomes. We will loo k at how the behaviourisr and in natist expla­
nations for first language acquisition that we saw in Chapter 1 have been
extended to account for second language acquisi tion. We will also look at
some theories fro m cogniti ve psychology that have increasingly in formed
second language research in recent years. These theories emphasize th e way
th e mind perceives, retains, organizes, an d retrieves information. Finally, we
will loo k at sociocultural theory, a perspective that places second language
acquisition in a larger social context.

The behaviourist perspective

As we saw in Chapter 1, behaviourist theory explained learning in terms of
im itation, p ractice, reinforcement (o r feedback on su ccess). an d habi t for ma ­
tion . Much of the early research within behaviourisr theory was done with
labo rato ry animals, b ut the learning process was hypothesized to be the same
for humans.

Second language applications: Mimicry and

memorization
Behaviourism had a powerful influence on second and foreign language
teach ing, especially in N orth America, from the 1940s to the 197 0s. Nelson
Brooks (I 96 0) and Robert Lado (1964) were rwo p roponen ts of this per­
spective. Th eir infl uence was felt d irectly in the develop ment of widely used
104 E"pltlining second 141zgu age learnzng
aud iolingual teaching ma te rials and in teache r trainin g. C lassroom act ivities
em p hasized mimicry and me mo rization, and students learned dialogues and
sente nce pat tern s by heart . Beca use language develop ment was viewed as the
formation of hab its, it was assumed that a person lear ni ng a seco nd lan guage
would start off with the hab its formed in the first language and that these
habits would interfere with th e new ones needed for the second language.
Thus, behaviourism was often linked to the contrastive analysis hypothesis.
However, as we saw in Chapter 2, researc hers found that many of the errors
learners make are no t predictable on the basis of thei r first lan guage, nor do
they always make the errors that would be predicted by a simple comparison
of their first an d second languages. This discovery led to the reject io n ofbotli.
the contrastive analysis hypoth esis and behaviourism , lead ing to a period
during which both the role of the first language and the role of practice
in learning a second language received lim ited attention in both research
and pedagogy.
In ChaprerZ , we saw ample evidence that second language lea rners draw o n '
whar th ey already know-including pr eviou sly learned lang uages: H owever,
we also saw th at they are so me times relu ctant to transfer certain first lan guage
pattern s, even w hen the translatio n equ ivalent wo uld be co rrect. And we saw
tha t first language influ ence may become more app arent as mo re is learned
abo ut the seco nd lan guage, leading learn ers to see simi larities that th ey had
not perceived at an earlier stage. All thi s suggests tha r th e influence of th e
learner's first language may not sim ply be a matter of habits, but a more
subtle and complex process of iden tifying points of similarity, weighi ng the
evidence in suPPOrt of some particular feature, and even reflecting (tho ugh
not necessarily consciously) about w hether a cert ain feature seem s to ' belong'
in the target lan guage.
By the 1970s, many resear chers were convinced that behaviourism and the
COntrastive analysis hypothesis were inadequate explanations fo r second lan ­
guage acquisition . As we shal l see, however, as research o n second language
acquisition has evo lved, the explanatio ns offered by behaviourism and the
contrastive analysis hypothesis have been revisited and understood in terms
of new learning theo ries.
The innatist perspective
As we saw in Chapter 1, th e rejection of behaviourism as an explanation
for first lang uage acquisitio n was partly triggered by C ho msky's critiq ue of
it. Chomsky arg ued tha t innate knowledge of the principles of Universal
Grammar per mits all ch ild ren to acq uire the lang uage of thei r enviro nme nt
du ring a critical per iod of th eir develop ment. While C homsky di d no t make
specific claims abou t the implications of his theory for second language
learning, Lydia Wh ite (2003) and other lingu ists have arg ued th at Universal
Explaining second language learning 105
Gram ma r offers the best pe rspec tive from which to understa nd second lan ­
guage acquisition. O th ers. fo r exam ple Robert BIey-Vroman (1 990) and
Jacquelyn Schachter ( 1990) have sugges ted th at, although UG may be an
ap prop riate framework for understanding first language acqu isitio n, it does
not o ffer a good explan atio n for th e acqu isitio n of a second language, espe­
cially by learners who have passed the critical peri od. In their view, this mean s
that second lan guage acquisit ion has to be explained by some other theory,
perh aps o ne of the more gen eral psych ological theories described below.
Vivian Cook (2003) and others point o ut that there is still 'th e log ical
problem' of second language acquisition. That is, we need an explanation
for the fact that learners eventually know more about the language than they
co uld reasonably have learned if the y had to depend entirely o n the input
they are exp osed to . The implication is that knowledge ofUG must be avail­
ab le to seco nd lan guage lear ners as well as to first language lear ners. Some of
the the ori sts wh o hold this view claim that the nature and availa bili ty of UG
are the same in first and seco nd language acquisit ion . Others argue that UG
may be present and availab le to seco nd lan gua ge learn ers, b ut th at its exac t
na ture has been altere d by th e acq uisitio n of other lan guages.
Research ers wo rking wirhi n the UG framework also d iffer in th eir hypo rh­
eses about how formal instr uctio n or the availabi lity of feedback will affect
learn ers' knowledge of the seco nd language. Bonnie Schwartz (1993) , for
exam ple, concludes that instruction and feedback ch ange on ly su perficial
aspects of language performance and do not affect the underlying syste m ­
atic knowledge of the new lan guage. She argues that language acquisition
is based on the availab ility of natural language in the learner's environment.
Interaction with speakers of that language is sufficie nt to trigger the acquisi­
tion of the underlying structure of the language. Lydia White ( 199 1) and
o th ers agree tha r acquisition of many grammatical features of the new lan ­
guage takes place naturally when learners are engaged in meaningful use o f
the language. H owever, they also suggest that, because the nature of UG is
altered by the acquisiti on of the first language, second language learn ers ma y
som eti mes need explicit information abo ut what is not grammatical in the
seco nd lan guage. Otherwise, they may assu me that some structures of th e
first lang uage have equivalents in th e seco nd language when, in fact, the y
do no r. In C ha p ter 2 , we saw a good example of th is in Whi te's study of the
p lacement of Engl ish adverbs in sentences prod uced by Frenc h speakers.
In Chap ter 6 (Stu d ies 18 and 33), we will see some research o n th e effect of
instruction and feedback on suc h lan guage feat ures .
Resear che rs who study seco nd langu age acquisition from a UG pers pective
are usu ally interested in the lan guage co m pet en ce of advanced learners­
their kn ow ledge of com plex gra mmar-rathe r th an in the simple lan gu age
of beg inning learn ers. They are interested in whe ther the competence that
underlies the use of the second lan guage resem bles th e co mpetence underlying
106 Expla ining second language learning
Explaining second language learning 107
the lan guage perfor mance o f native speakers. Thus, their investigations often
invo lve grammaticality judgement or Other methods to probe what learners
know about the language rath er than observations of nat ural language use.

Second language applications: K rashen's

...
'Monitor M odel'

Perhaps the best known model ofsecond language acquisition influen ced by
Chomsky's th eo ry of first language acq uisition is Stephen Krashen's (1982)
Monitor Model, first described in the early 197 0s, at a time when there was
growi ng dissatisfaction with language teaching methods based o n behavio ur ­
ism, Krashe n desc ribed his model in terms of five hypotheses.
In the acquisition/learning hy po thesis, Krashen suggests that we 'acq uire' lan ­
guage as we are exposed ro sam ples of language that we understand in much
the sam e way that children pic k up the ir first lan guage-with no conscious
attention ro lang uage form. We 'learn' on the other han d through co nscious
attention to form an d rule learning. In Krashen's view, far mor e lan gu age is
acquired than learned. The Moniror Model been challe nge d by o ther researchers and theoris ts,
some ofwho m have argued that it is not possi ble to test Krashen's hypo theses
N ext, according ro the monitor hypothesis, second language users draw on
in empi rical research (Mclaughlin 1987) or th at he has drawn the wrong
what they have acquired when they engage in spontaneous communica­
con clusions from resear ch (White 1987). Nevertheless, h is ideas about
tion . Th ey may use rules and patterns that have been learned as an editor
seco nd language develop ment were in flue neial d uring a period when second
or 'monitor' , allowing them ro make mino r changes and polish what the
lan gu age teach ing was in transition from structu re-based approaches that
acquired system has prod uced . Such m o niroring takes place only whe n the
emphasized lear ning rules or memorizing dialogues, ro approaches that
speake r/w riter has plen ty o f time, is concerned abo ut producing cor rect lan­
emphasized using lan guage w ith a focus on meaning. Since then, as we will see
guage, and has learned the relevant ru les.
in Chapter 6, co m m unicative language teaching, including immersion, con ­
Th e natural order hypo thesis was based on the find ing th at, as in first language tent-based, an d task-based lan guage teaching, has been widely implemented .
acquisition, second lan guage acquisition unfolds in p redictable seq uences, as Krashen's hypotheses, especially th e comprehe nsib le input hypothesis, have
we saw in Chapter 2. The language rules that are easiest ro state (and thus ro been a source ofideas for research in second language acquisition. Classroom
learn ) are no t necessarily the first ro be acquired. research has confirmed that students can make agreatdeal ofprogress through
expo sure to co mprehensible input without direct instruction. Studies have
Th e comprehensi ble input hypothesis is that acquisition occ ur s when one
also shown, however, that students may reach a point from which they fail to
is exposed ro language that is co mprehensible and contains i + 1. Th e "i' rep ­
mak e further progress on some featu res of the second language u nless they
resents the level oflanguage already acquired, and the ' + I ' is a metaphor for
also have access ro gui d ed instruction. Some insigh ts from lear n ing theo ries
language (words, grammatical form s, aspects of pron un ciation) that is JUSt a
step beyond th at level. developed in cog nitive psychology help ro exp lain why th is may be so.

Krashen's affictive jilter hypothesis is proposed to account for the fact that
some peo ple who are exposed to large quantities of co mprehensible in p ut do
no t necessarily acquire lang uage success fully. Th e 'affective filter' is a meta­
phorical barrier that prevents learners fro m acquiring lan guage even when
app ropriate inp u t is available. Affict refers ro feelings of anxiety or negative
attitudes tha t, as we saw in C ha pter 3, may be associated with poor learn ­
ing out comes. A learner who is ten se, anxi o us, or bore d m ay jilter out input,
m aking it unavailable for acquisitio n.
ACT IVITY Examine t he M o ni t o r M od el
A number of write rs have questio ne d the validity of Kra shen's Monitor Model,
partly o n t he basis that it is difficult t o test the five hypotheses in e mp irical
stu dies. Ne ve rthe less, Kras he n's views have re mained influen t ial in seco nd
language tea ching.
I C an yo u th ink of so me reasons wh y t his might be so !
108 Explaining second fangullge learning
2 Which of the hypotheses do you find intuitively convincing!
3 Wh ich ones leave you sceptical! Why!
The cognitive perspective
Since jhe 1990s, research and th eories from cognmve psychology have
become increasingly central to our understanding ofsecond language devel­
opmen t. Some of these the ories use the computer as a metaphor for the m ind,
co m par ing lan guage acqu isitio n to th e capacities of co m pute rs for sto ring,
integrating, and retrieving information. Some draw on neurobiology, seeking
to relat e o bserved behaviour as dire ctl y as possible to brain acti vity.
As in first lan gua ge acq uis itio n, cogn itive and de velopmental psychologists
argue that there is no need to hypothesize that humans have a language-spe ­
cific module in th e brain or that acquisition and learning are d istinct mental
pro cesses. In their view, general theo ries of learning can account for the
grad ual development of co m plex syntax an d for learners' inabili ty to sponta­
neo usly use everything they know abou t a language at a given tim e. As not ed
above, some ling uists have also concluded that, wh ile the in narist perspective
provides a plausib le exp lanation for first language acquisition, something
else is requ ired for second language acqu isitio n, since it so often falls short
of full success. Fro m the cog nirive psych ology persp ective, however, first and
seco nd language acq uisitio n are seen as dr awing o n the sam e processes o f
perceptio n, memo ry, catego rizatio n, and generalization. The difference lies
in th e circ umstan ces oflearn ing as well as in what th e learners already know
abo ut language and ho w that pri or knowledge shapes their perception of the
new language.
InfOrmation processing
Cogn itive psychologists working in an information-processing model o f
human learning an d performance see seco nd lan gu age acquisition as th e
build ing up of knowledge that can eventually be called on automatically
for speaki ng and understanding. Robert DeKeyser (1998 ), Richard Schmidt
CWO1) and others have suggested that learners must pay attention at first to
any aspect of the language that they are trying to learn or produce. 'Pay atten­
tion' in this con text is accepted to mean 'using cognitive resources to process
information' but there is a limi t to how much information a lear ner can pay
att ent ion to. Thus, lear ners at the earliest stages will ten d to use most of their
resources to understand th e main words in a message. In that situation, they
may not notice the grammatical morphemes attached to some of the wo rds,
especially those tha t do not su bstant ially affect meaning. Gradually, through
expe rience and pr act ice, info rmatio n th at was new becomes easier to process,
and learners become able to access it q u ickly and even automatically. This
Explaining second language learning 109
frees up cognitive process ing resou rces to notice othe r aspec ts of the langu age
that, in turn, gradually become automatic.
For proficient speakers, choosin g words, pronouncing them, and string­
ing them rogether with the appropriate grammatical markers is essent ially
auto matic. Fu rthermore, much ofwhat the se speakers say is drawn from pre­
dic tab le patterns o flan guage that are at least partly formulaic, That is, fluent
speakers do not create new sentences by choosingone word at a tim e but rath er
by using strings o f words that typically occur together, This use of patterns
applies not onl y to idiomatic expressio ns, but also to much co nve rsatio nal
language and written language in a specific genre (Ellis, Simpson-Vlach, and
Ma ynard 200 8) .
Ano the r aspect ofauro rnaticity in lan gu age process ing is th e retr ieval ofword
meanings. When proficient listeners hear a familiar word, even for a sp lit
seco nd , they can no t help but understand it. Such auto matic responses do
not use up the kind of resources needed for pr ocessing new informati on.
Thus, p roficient langu age users can give their full attention ' to th e overall
mea n ing ofa text or conversatio n. whereas less pro ficient learne rs use mo re
of thei r attentio n on processing th e meaning of indi vid ual words and the
relationsh ips between them. Th e lack of automatic access to meaning helps
to expla in why second language reade rs need more time to understa nd a text,
even if th ey event ually do fully co m p rehe nd it. Th e information p rocessin g
model suggests that there is a limit to the amount o Hocused mental activity
we ca n engage in at one time.
Information pro cessing approach es to seco nd lan gu age acq uisitio n have
been explo red by many researchers. Drawing on J. R. Anderson's (199 5)
wor k, Robert D eKe yser (1 99 8, 200 1, 2007) and oth ers have investi gated
seco nd language acquis ition as 'skill learning'. They suggest that most learn­
in g, including language learning, starts with declarative knowledge, th at is,
kn owledge that we are aware of havin g, for example, a grammar rul e. Th e
hypothesis is that, through pr acti ce, decl arati ve knowledge ma y become pro­
cedural knowledge, or th e ability to use th e kn owledg e. Wi th continued
p ractice, the procedural kn owledge can become automatized and the learn er
ma y for get havin g learned it first as declarative knowledge.
Acco rding to this perspective, once ski lls become automatized , thinking
abou t the declarative knowledge while trying to perform the ski ll acrually
disrupts the smooth performance of it. Think, for example, of trying to drive
a car o r skate whi le intentionally thinking abou t and preparing every move.
With enough pra ctice, procedural knowledge eclip ses the declarative knowl­
edge, which, in time, may be forgotten. Fo r this reason , fluent speakers may
no t even realize that the y once po ssessed the declarative knowledge that set
the process in motion .

110 Explain ing second Languagelearning
Sometimes changes in language beh avio ur do no t seem [Q be exp lainabl e
in terms of a gradual build-up of fluency through practice. These changes
have been descri bed in terms of restructuring (Mclaughlin 1990). They
seem [Q be based on some qualitative change in the learner's knowledge .
Restructu ring may account for what appear [Q be bursts of p rogress, when
lear ne rs suddenly seem to 'pu t it all together', even though they have no t had
any new instr uctio n or appare n tly relevan t exp osure to the language. It may
also explain apparent backsliding, whe n a systematic aspect of a learner's lan­
guage incorpo rat es too much o r incorpor ates th e wro ng things. For example,
as we saw in Chapter 2, when a learner finally masters the use of the regular
-ed ending [Q show past tense, irregular verbs th at had previo usly been used
correctly may be affected. Th us, after months o f saying 'I saw a film ', th e
lear ner may say 'I seed' or even 'I sawed'. Such overgene ralizatio n erro rs are
not based on practice of those specific items bu t rather on th eir integrati on
int o a gene ral pattern.
Another concept from psycho logy offers insig ht into how learners store and
retrieve langu age. Acco rdi ng [Q tr an sfer-approp riat e processing (TAP ),
information is best retrieved in situations that are similar to those in which it
was acquired (Lightbown 2008b) . Th is is because when we learn something
our me mories also reco rd aspects of th e co ntext in which it was learned and
even the cognitive p rocesses invo lved in th e way we lear ned it, for exam ple,
by read in g or heari ng it. To da te, mo st of the research on transfer-app rop ria te
processing has been done in laborato ry experiments, for exam ple, co mparing
the learn ing ofword lists under different co ndi tions. However, the hypothesis
seems [Q offer a pla usible way of explaining a widely observed ph eno m enon
in seco nd language lear ning: knowledge tha t is acquired mai nly in rul e learn ­
ing or drill activities may be easier [Q access on tests th at resemble the lear n ing
activities than in communicative situ ations. On the Other hand, if learne rs'
attention is drawn co grammatical forms during communicative activities
in wh ich their cognitive resources are occupied with a focus on mean ing,
the retrieval of those forms on a grammar test may be more d ifficult. In
C hapter 6, a classroo m investigatio n of L2 lear ning influen ced by transfer­
appropriate pro cessing is described in Stu dy 40 .
Usage-based lea rning
As seen in the discussion of first lang uage acquisition in C hapter 1, cognitive
psycho logists, unlike innatists, see no need to hyp othesize the existence of
a neurological mo dule ded icated exclusively to lan gu age acqu isition. They
argue tha t what is innate is simply the ab ility to learn , rather than any specific
lingu isti c p rincip les. Some usage- based theories also attri bute less impor­
tance to th e kind of de clarative kn owledg e th at ch aracter izes skill learn ing
and trad itio nal str ucture- based app roaches to second lan gua ge ins tru ctio n.
As Nic k Ellis (20 02) explains, th e em phasis is on th e freque ncy wit h whic h
Explaining second language learning 11 1
learners enco unter specific linguistic features in the in put and the frequency
with which lan gu age features occur together. According to th is view, learn­
ers develop a stronger an d stronger nerwork of associations or connections
between these features as well as between lan guage features an d the contexts
in wh ich th ey occ ur. Eventually, th e presence of one situational o r lingu istic
featu re will activate the otherfs) in the learner's mind. For exam ple, learners
migh t get su bject-verb agreeme nt co rrect, not because th ey know a rule but
because they have heard examples such as 'I say' an d 'he says' so ofte n that
each subj ect pronoun activates th e co rrect verb form .
Con nections may be strong because the lang uage featu res have occurred
together freq uen tly or th ey may be relatively weaker beca use th ere have been
fewer opportunities to experi ence them to gether. Some of the evidence fo r
usage- based views co mes from th e o bservation mentio ned above that much
of the language we use in o rdinary co nversa tion or in particular genres is
predictab le, and to a considerable exte nt based on formulaic un its o r ch unks.
As sugges ted by N ick Ellis (200 3 , 2005) and ot hers , language is at least partly
learned in uni ts larger than sing le words, and sentences or phrases are no t
usually put toge ther one wo rd at a tim e. As no ted in Chapter 1, usage- based
research has shown that a learn ing mec hanism , simulated by a computer
program, can not on ly 'learn' from in put but can also generalize, even maki ng
overgeneralization errors.
The competitio n mod el
Elizabe th Bates and Brian MacWhin ney (198 1) described the 'co m pet ition
m odel' as an explanation for both first and second language acquisition that
takes into account no t on ly lan guage form but also lan guage me an ing an d
language use. Through expos ure to th ou san ds of examples oflanguage asso­
ciated wi th particular mean ings, speakers of a particular language co me to
understand how [Q use the 'cues' that signal specifi c functions . For example,
the relationship between words in a sentence may be signalled by word
order, grammatical markers, and the an im acy of the nouns in th e sentence.
Mos t lan guages make use of multiple cues, but they di ffer in th e p rimacy of
each. This becomes clear in a situa tion where the mean ing of a sentence is
no t immediately obvio us. What hel ps yo u figu re out the mean ing? English
uses word order as the most common indicator of the relationships between
sentenc e componen ts. Most English sentences have the order Su bject-Verb­
Object (5VO) . That is, the typical English sentence ment io ns th e su bject
first , the n the verb , then the object.
Two - and th ree-year old English-speaking ch ildre n can usual ly use cues of
ani macy and their kn owledge of th e way th ings work in the wo rld to int er­
pret odd sent ences. Th us, if th ey hear a string of words such as 'Box push
boy' , the y will act it ou t by making a boy do ll push a tiny box , focusing o n
112 Explaining second Lmgu,tge learning
the fact that th e 'boy' is the natural agent ofaction in this situation. H owe ver,
the SVO patrern is so stro ng in English that, by the time the y are four years
old, ch ildre n heari ng th is sent ence will ignore the fact that boxes do n't nor­
mally move on their ow n , and carefully demo nstrate how th e box pus hes th e
boy. For English speakers, wo rd order patterns are str onger than animacy
cues at thi s pai n e. At this age, ch ild ren may attri bute t he SVO relation ship
to sentences in th e passive voice. That is, 'Th e box was pushed by the boy'
may be interpreted as 'Th e box pushed th e boy.' Only later do they learn to
pay attention to the grammatical markers th at distinguish the active voice
sentence from th e passive word order.
In contrast , Spanish and Italian have more flexib le word order, and speak­
ers of these languages rely more on grammatical markers (fo r example, th e
agreement of subj ect and ver b, the case marking of pronouns) o r o n th e
an irnacv of nouns to understand how sen rene e elem ents are related. When
English speakers are learn ing these languages, th ey may have difficulry sup­
pressing their tendency to rely on word order as the basis for interpret at ion .
For examp le, an English speaki ng learn er of Italian may find it co nfusing to
hear sent ences suc h as II giocattolo gUtlrda if bambino (the toy-is loo king
at-the boy). An Italian speaker, accustomed to more flexib le wo rd order.
focuses on the ani macy of t he rwo nouns and concludes tha t th e most rea­
sonab le interpreta tion is th at the boy is looking at the toy. Acco rd ing to th e
co m petition model, seco nd language acq uisitio n requires that learners learn
th e relar ive impo rtanc e o f the different cues app ropriate in the langu age th ey
are learning (M acWhin ney 199 7).
ACTIVITY Look at how different cues lead to sentence
interpretation
Consider the following sente nces:
I The boy eats the apple.
same neural areas that are activated for L I processing but also .ac tivatio n in
2 The apple eats the boy.
o the r areas of the brain. This is not the case with younger learners who show
3 The dog sees the ball.
acti vation o nly in the areas for LI processing (Berett a 2011 ). Other stu d ies
4 The ball chases the dog.
h ave measur ed the electrical activity in br ain waves ro exp lore differen ces in
S The ball is chased by the dog.
the processing oflanguage in p ut. Some of this research has shown that as an
I Do they all follow the patterns of English grammar?
2 How can you tellwhich noun refers to the agent (the one who performs
the actio n)?
3 In each se ntence, what cue tells yo u which noun is the agent?
4 Is there more than o ne cue?
5 How are sentences 4 and S above different from each other?
Explaining second language learning 113
6 Accord ing to the competition model. how might these sentences be
interpreted by speakers of a language with a more flexible word order than
English? Wh at wou ld those speakers focus on?
The cog rnuve perspe ct ive e m phasizes the role of general h um an ab ili­
ties ro process and learn info rma tio n-c--i nc luding language-on th e basis
of experience. In recent years , the term 'co gni tive lingui stic s' has eme rged
an d highlights the view that language is but one of th e complex knowledge
syste ms that humans acquire. Peter Rob inson and N ick Ellis (20 08) suggest
tha t cognitive linguistics draws from an d builds on a number o f different
app roaches that have in common th e hypothesis that language is learned
th rou gh our perceptual an d cog n itive experiences and th at like all other
aspects oflear ning, lan guage learning involves th e discovery, categorization,
and determination of patterns through the use oflang uage.
La nguage and the brain
Anot her area of work wit hi n but no t lim ited to the cog ni tive perspective
is concerned wi th language lear ning and the b rain. Some of the questions
investigated include whether first and second languages are acquired and
represented in th e same areas of the brain and whe ther th e b rain processes
seco nd lang uage input d ifferently fro m first language in pu t. For a long
ti me the assumption was that lan guage fun cti ons were located in th e left
hemisphere of the brain. Nonetheless, recent brain imaging stu d ies show
activa tio n in different locati ons in both hemispheres of th e brain during lan ­
guage processing. This is rru e for bo th first and second languages. However,
differences ha ve been observed . depending on th e learners' age and level of
p roficiency. For example, when learners w ho acquire a seco nd lan gua ge later
in life are given a granlmatical task ro co m p lete, th ey show ac tivatio n in th e
L2 learn er's pro ficiency increases, the brain act ivity looks more like that of
first language processing. There is also evidence that semantic pro cesses are
the first to loo k more like L 1 processin g patterns followed by syntactic p ro­
cesses as proficiency in th e L2 in creases (H ah ne 200 1).
While it is fascinating to th ink about con nectio ns between seco nd langu age
learnin g and the brain. it is important ro keep in mind that th is is a yo u ng
di scip line. Furthermore, the limited research that has been conducted has
pro d uced mixed find ings. Th erefore any impl icati ons o f lan gu age and br ain
research for second language teaching are premature.

114 Explaining second language learning
Second language applications: Interacting; noticing;
processing, andpractising .
A number of hypotheses, theories, and models for exp laining second lan­
guage acquisition have been inspired by me cognitive perspective.
The interaction hypothesis
Evelyn Hatch (1978), Michael Long (1983, 1996 ) ,Teresa Pica (1994), Susan
G ass (1997), and many othe rs have argued mar co nversa tional interaction
is an essential, if nor sufficient, condition for second lan gu age acquisition.
These researchers have studied me ways in whic h speakers modify th eir
speech and their interaction parrern s in order to help learners participate
in a conversation or u nd erstan d meaning in a new language. Long (1983)
agreed with Krashen chat comprehensible inp ut is necessary for language
acquisition. However, he focused On me question of how in put could be
.made comprehensible. He argued mar modified interaction is the ne~s­
sary mechanism for making language comp rehensible. That is, what learners
nee d is opportunities to interact with ocher speakers, working rogerher to
reach mutual comprehension through negotiation for meaning. Through
these interactions, interlocutors figure our what they need to do ro kee p the
conversation go ing and make the input comprehensible to me less profi­
cient speaker. According to Long, m ere are no cases of beginner-level learners
acquiring a second language fro m native-speaker talk mat has not been mod­
ified in some way.
Modified interaction does not always involve linguistic simplification. It may
also include elaboration, slower speech rare, gestu re, or me provision ofaddi ­
tional contextual cues . Some examples of conversational modificario ns are:
1 Comprehension checks---efforts by me native speaker to ensure mar the
learner has understood (for example, 'Th e bus leaves at 6:30. Do you
understand?') .
2 Clarification requests---efforts by the learner to get the native speaker to
clarify someth ing mar has not been understood (for example, 'Could you
repeat please?'). These requests from me learner lead to fu rth er modifica­
tions by me native speaker.
3 Self-repetition or paraphrase-rhe more proficient speaker repeats his or
her sentence eit her partially or in its entirety (fo r example, 'She gOt lost
o n her way home from school. She was walking home fro m school. She
got lost. ').
Long (1996) revised the interaction hypothesis, placi ng more emphasis on
cognitive facto rs such as 'noticing and corrective feedback during inter­
act ion. When communication is d ifficul t, int erlocu to rs m ust 'nego tiate
for meaning', and thls negotiation is seen as me opportunity for language
Explaining second language learning 115
development. Related to this is Merrill Swain's (1985) comprehenSible
output hypothesis. She argued chat when learners must produce language
mat their interlocutor can understand, they are most likely to see me limits of
their second lan guage abil ity and the need to find better ways to express their
meaning. The demands of pro ducing comprehensible Output, she hy poth­
esized, 'push' learners ahead in their development.
The noticing hypothesis
Richard Schmidt (1 990, 2001) proposed the noticing hypothesis, sug­
gesring that nothing is learned unless it has been 'noticed' . N otici ng does
nor itself result in acquisition, but it is me essential starring point. From
this perspective, co mprehensible inp ut does no t lead to growth in language
knowledge unless the learner becomes aware ofa particula r lan guage feature.
Schmidt's original proposal of me noticing hypothesis carne from his own
experience as a learner of Portuguese. After mo n ths of taki ng classes, living
in Brazil, and keeping a diary, he began to realize that certain features of
lan guage that had been present in the enviro n ment for the whole tim e began
to enter his own second language system only when he had noticed memo
This was because they were brought to his attention in class or some other
experience made them salient. Drawing on psychological lear ning theories,
Schmidt hypothesized char second language learners could no r begin to
acquire a language feature until they had become aware of it in the in put.
Susan Gass (19 88) also described a learning process that begi ns when learn ­
ers noti ce something in me second lan guage that is different fro m what they
expected or that fills a gap in their knowledge of m e lan guage.
The q uestio n of whether lear ners must be aware that they are 'no ticing'
something in the input is me object of considerable debate . Accord ing to
info rmation processing rheories, anything that uses up our mental 'process­
ing space ' , even if we ilre not aware of it or attending to ir int enti o nally, can
contribute to learning. From a usage-based perspective, the likelihood of
acquisition is best predicted by the freq uency wit h which something is avail­
ab le for processing, not by the learner's awareness ofsomething in me input.
These q uestions about rhe im portance of awareness and attention con­
tinue to be the object of research. Several researchers have found ways to
track learners' attention as they engage in second lan guage interactio n. For
example, Alison Mackey, Susan Gass, and Kim McDonough (2000) had
learne rs watch and listen co them selves in Videotaped interactio ns and asked
questions leading them to explore what they were thinking as they partici­
pated in those interactions. Ro n Leow (1997) developed crossword p uzzles
chat learn ers had to solve wh ile rh inking alo ud , th us provid ing some insight
in to wha t they not iced about lan guage as they worked. Merrill Swai n an d
Sharon Lapkin (199 8) recorded learners in pair work an d kept trac k of me
116 Explaining second l mguage learning
language teatures the y mentioned. These research designs cannor tell us if
learne rs no ticed rhings cheyd id nor m entio n . However. rhey do make it pos­
sib le to ide n tify so me th in gs char learners we re aware o f and to look at how
th is awarenes s is related [Q measures of the ir language knowledge. The ext ent
[Q whic h lea rn ers' no ricing o f lan guage featu res affects their seco nd language
development wi ll come up ;1gain in our discussion of research on seco nd
language acquisition in the classroom in C hapters 5 and 6 .
Input processing
In his research wi rh American unive rsity students learn in g forei gn lan gu ages,
Bill Van Parre n (2004) o bserved many cases of students m isinterpreting sen­
reaces. Fo r example, as predi cred by the co m pe tition m odel discussed earl ier
in th is chapter. when English speakers heard sentences in Span ish, chey used
wo rd o rde r to interpret the relarionships amo ng the no uns in the sentence.
Th us. they interp reted ' La sigu« el seizor' as 'She (subjecr pronoun) follows
the man'. The co rrec t inrerpretarion is 'Her (objec t p ro nou n) follows rhe
man' (subject of the sentence). In orher words . the co rrect English trans latio n
wo uld be 'Th e man follows her'. In order to u nd erstand that . stude n ts need
ro learn that in Spanish , a pronoun o bjec t o ften precedes the verb and that,
rather than rely o n th e word o rder alone, it is essential to pay attention to
wh ether the form of the pronoun indicates a subject or an objec t.
Van Parten argued that the problem arose in part from the fact that learn­
e rs have lim ired processing capacity an d ca n no r pay at te ntion [ 0 form and
meaning ar the same ti me . Nor surprisingly. they ren d to g ive prio rity [Q
meaning, ove rlooking some fearures of the lan guage form. Whe n the conrext
in w h ich rhey hear a sentence hel ps rhem make se nse o f ir, thar is a goo d
strategy for u nd ersran di ng th e gen e ral idea. bur ir may interfere w ith learners'
progress in acqu irin g th e language. In C hap rer 6 we wi ll see ho w VanPatten
developed ins tructio na l proced u res that require learners to focus o n the spe­
cific language fearures in order [Q interpret the meani ng , th us pushing them
ro acquire those features.
Pro cessability theory
jurgen Meisel. H arald C lahsen, and Manfred Pienemann (198 1) studied rhe
acq uis itio n of G erm an by a group ofad ult mi gram wo rke rs who had littl e o r
no seco nd lan gu age instruction . Th ey analysed large sam ples o f th eir speech
and described th e details of developmenral seque nces in rhe ir pr oducti on of
sim p le and co m plex sen te nces . Th ey co ncl ude d rhat the sequen ce ofdevelop­
me n t for fearu res of synra.x and morpho lo gy was affected by how easy th ese
were to process. Ease o f processing was fo un d [Q dep end to a lar ge extent o n
the positio n of th ose features in a senre nce . Features rhar typically occ ur red
ar the beginning o r end of a senrence were easier ro process (and learn ) rhan
those in che m iddle. All learners acq uired th e fea tu res in th e sa me seq ue nce .
b:plll1lmg second l wguage learning 11­
even though they progressed at dirferent rates . Th e researchers also found
thar some lan guage feat ures did nor seem to be affected by rhese co nstralnts
an d could be learned and used by lea rn ers wh o were ar d ifferenr develo p men­
tal stages. These were referred to as variational features .
Pienemann (1999, 2003) d evelo ped processabiliry th eory on rhe basis of
research with learners of different languages in a variety of settings, both
instructional and info rmal . One irn po rran r aspect of h is theory is th e inte­
gra tio n of de velo p me ntal seque nces with first lan guage infl ue nce . He argues
that his theo ry explains why lea rn ers do no t simply transfer features from
their first lang uage ar ear ly stages ofacquisirion. Insread, they have to develop
a certain level of processing capacity in the seco nd language before they can
use their knowledge of the features rhar already exist in their first language.
We saw examples of this in the acquisition of negatives and questions in
Chap rer 2.
The role of p ractice
O ne co m po nenr o f language learn ing that has seen a renewal of interest
with in th e cog nirive persp ecti ve is pract ice. As we saw in di scu ssion s of the
be ha vio u risr perspecti ve, a n approach to learning that is based o n drill and
m ar separates pr actice from meaningful lan guage use d oes not usu ally lead
ro co m m u nicative competence. This does nor mean , however, that practice
is nor an essenrial co mponenr oflanguage learning. Robert DeKeyser ( 1998)
asserts that some classroom interpretations ofbehavio urism missed rhe po int
rhar practice is o nly effec tive if one practises t he behaviour th a t one wishes to
learn. As we w ill see in Chap ter G, the dr ills that characterized a udiolingual
ins truction often failed to make rhe co nnec rion between th e lan guage par­
terns being d rilled and the meaning(s) associared wirh them .
Researchers are now loo king m o re closely at how pr actice co nverts declarative
knowledge to pro cedural knowledge and th e n to automat ic pe rfo rmance .
No re that from the cognirive perspective, the practice needed for language
develo pmenr is no t mechanical, and ir is nor limired to the production of
language. Listening and reading are also affected by opportunities for prac­
rice. Lo ur des O rtega (2007) has pro posed three principles fo r practice in
the foreign lan gu age classr oom rh ar she sees as compati ble w irh rhe researc h
carried o ur from wh ar she calls rhe 'cog nltive-In te ractio n ist' pers pec tive:
1 Practice sho uld be inre racr ive.
2 Practice sho uld be m ean in gful.
3 The re sho uld be a focu s o n tas k-esse n rial fo rms.
Elizabe rh G atbonto n and No r man Segalowirz ( 1988. 2005) have devel­
oped an app roach to lan gu age teac hi ng ca lled ACCESS (Auro rnatizario n
in Com mu n ica tive Co n rexrs of Esse nrial Speech Seg ments) . Ir d raws on rhe
cog nit ive persp ective and is based o n classr oo m activities w hic h, by rhei r
118 Explaining second language learn ing
nature, require learners to use meaningful units of language repet itively
in contexts where there are genuine exchanges of meaning. The goal is to
provide opportunities for using these units wi th sufficient frequency that
they will become automatic. Segalowitz (20 10) has emphasized the impor­
tance of increasing the amount of language that can be used automatically,
th us freeing more cognitive resources for learning new thi ngs. Paul Nation
(2007) has suggeste d tha t automaticity, which he, like Segalowitz, refers to as
'fluency ' may be the most neglected aspec t of language teachin g in contexts
whe re instruction focuses primarily on meaning.
The sociocultural perspective
As we saw in Chapter 1, Vygorsky's theory assumes that cognitive develop­
ment, including language development, arises as a result ofsocial interactions.
Unlike the psychological theories that view th inki ng and speaking as related
but independent processes, sociocultural th eory views speaking and th ink­
ing as tightly interwoven. Speaking (and writing) mediates thinking, which
means tha t people can gain control over their mental processes as a conse­
quence of int ernalizi ng what others say to them and what they say to others.
This internalizing is tho ught to occur when an individual interacts with an
interlocutor within his or her zone of proximal development (ZPD)-that
is, in a situation in which the learner can perform at a higher level because of
the support (scaffolding) offered by an interlocutor.
In some ways, this approach may appear to restate some of th e hypotheses
encountered elsewhere in thi s chapter. In fact , people sometimes wonder
whether the ZPD is the same as Krashen's i + 1. William Dunn and James
Lamolf (1998) addressed this question in a review art icle, argui ng tha t it
is not possible to compare the rwo concepts because they depend on very
different ideas abo ut how development occurs. The ZPD is a metaphorical
location or 'site' in which learners co-construct knowledge in collaboration
with an interlocutor. In Kras hen's i + 1, the input co mes fro m outside the
learner and the emphasis is on the comprehensibiliry of input that includes
lan guage structures that are JUSt beyond th e learner's current developmental
level. The emphasis in ZPD is on development an d how learners co-con­
srruct knowledge based on their interaction with their interlocutor or in
private speech.
Vygotskyan theory has also been compared to the interaction h ypothesis
because of the in rerlocutor's role in helping learners understand and be under­
stood . These rwo perspect ives d iffer primarily in the emphasis they place on
the internal cogn itive processes. In the interact ion hypothesis, the emphasis
is on the ind ivid ual cognitive processes in the mind of the learn er. Interacti on
facilitates those cognit ive processes by givin g learners access to the in put the y
need to activa te internal processes. In Vygotskyan theory, greater importance
Explain ing second language tearning 119
is attached to the conversations th ems elves, wi th learning occurring through
the social interacti on . Soc ioc ul tu ral th eo ry holds that peopl e gain co n trol of
and reorganize their cogni tive processes during mediation as knowledge is
internalized during soc ial activity.
Second language applications: Learning by talking
Extending Vygotskyan th eo ry to second language acquisition, Jim Lan to If
(2000), Richard Do nato (1994) , and others are interested in showing how
second lan guage learners acq uire lan guage when they collaborate and inter­
act with other speakers. Traditionally, the ZPD has been understood to
involve an expert and a no vice . H owever, recent work has broadened the
term to include novice -novice or learner-learner interactions. An example
of this is in Communication task B in Chapter 5 (p. 137 ). In th at excerp t ,
the learners are struggli ng wit h French reflexive verbs as they try to co nstruct
a storyline fro m pic tu res. The example is fro m the work o f Merrill Swain
and Sharon Lapkin (200 2), wh o have investi gat ed sociocultural explanations
for second language learning in C anad ian Fren ch immersion p rogrammes.
Th eir work has its origins in Swain's comprehen sible o utput hypothesis and
the notion that when learners have to produce language, they must pay mo re
attention to how meaning is expre ssed through language than they ordi­
narily do for the comprehension of language. Swain (1985) first pro posed
the comprehensible output hypothesis based on the observation th at French
immersion studen ts we re considerably weaker in their spo ken and writte n
production than in their reading and liste ning comprehensio n. She advo ­
cated more op portunities for learners to engage in verbal production (i.e,
output ) in French immersion classroo m s. Since then, she and her colleagues
have carri ed Out extensive research to investigate the effects of o utput on
second lan guage learning.
Swain's early work on th e o ut put hypothesis was influenced by cognitive
theory, but more recent work has been motivated by sociocultural theo ry.
Using the term collaborative dialogue, Swain and Lapkin and their col­
leagues have carried out a series of studies to determine how second language
learners co-co nstru ct linguistic knowledge while engaging in production
tasks (i.e. speaking and writing) that sim ultaneo usly draw their attention to
form and meaning. As shown in Communication task B in Chapter 5, learn­
ers were testing hypotheses about the correct forms to use, discussing them
together and deciding wha t forms were bes t to exp ress their mea ning . Swain
(2000) considers collabo rative dialogues such as these as the co nt ext where
'language use and lan guage learning can co-occur. It is language use mediat­
ing language learning. It is cogn itive act ivity and it is social activity' (p. 97).
Therefore th e di fference berween the socio cultu ral perspective and that of
o ther researchers who also view interaction as important in seco nd lan guage


[ 20 bp/dining second langudge learning
acq uisiti on is that soc ioc ultu ral th eorists assum e (hat the cogni tive pr o­
cesses begin as an external socially mediated activity and eventually become
internalized . O the r inreractionisr models assume that modified input and
interaction provide lear ne rs w ith (he raw material tha t is interpreted and
analysed through internal cognitive processes.
Summary
In the end, what all theories oflanguage acquisition are intended to account
for is the ability of human learners to acquire language within a variety of
soci al and instructional environments. All of the theo ries discussed in this
chapter and in C hap ter 1 use metaphors to represent something that cannot
be ob served directly
Linguists working from an in na tist perspect ive d raw m uch of their evidence
from stu d ies of the complexities of profi cient speakers' knowledge o f lan ­
guage and from a nalysis of their ow n intuitions ab out la nguage. Cognitive
and developmental psychol ogists a rgue th at it is not eno ug h to kn ow wha t
the fina l sta te of knowledge is and that mo re a tte nt ion sho uld be paid to co r­
pu s-b ased studies of the input. as well as to th e developmental ste ps leadi ng
up to the achieve ment of hi gh levels of p ro ficiency.
Recent co gn itive perspe ctives ha ve o ften in volved com p u te r sim ula tio ns or
co ntrolled laboratory experim en ts where peop le learn specific sets of care­
fully chosen linguistic features, often in an invented language. Many linguists
argue that this does not entitle psychologists to gen erali ze to the cornplexities
o f the linguistic knowledge that learn ers eventually have.
Inr era crionists emphasize the role of negotiation for meani ng in conversa­
tional intera ctions . This perspective and the soc iocu ltu ral perspective provide
insights into the ways in which learners can gain access to new k nowledge
abo ut (he language when the y have suPPOrt from an interlocu to r. Some
linguists challe nge the inrcracrionist position, argu ing that much of what
learners need to know is not availab le in the in p ut, a nd so they puc gr eater
em phasis o n in na te principles o f language th at learners can d raw on.
Bot h lingu ists and psychologists draw som e of the ir evidence from ne uro­
logica l research. At present, most of the research o n language representation
in the brain an d spec ific neurological activity during lan guage processing is
inconclusive. However, advances in tec hnology are rapid ly increasin g oppor­
tunities to observe brain ac tivity mo re d irecdy. Suc h researc h will eventually
co nt ribut e to reinterpretations of researc h that previously could examine
o n ly the o bservable behaviour of learners speaking or performing other lan­
guage tasks .
Educato rs w ho are hoping (hat language acquisition theories wi ll give them
insight into lan guage teach ing practice are often fru strated by the lack of
Explaining second I.mguage learning [21
agreeme n t amo ng (he 'experts'. The co m p lexit ies of seco nd lang uage acq uisi­
rion, like those of first language acq uisitio n, represent puzzles that scientists
will continue ro wo rk on fo r a lo ng tim e. Research tha t has theory develo p­
ment as its goal has importan t lo ng-ter m significance for langu age teac hing
and learning, but agreement o n a 'co m plete' th eory o f language acquisiti on
is probably, at best , a long way o ff. Even if suc h agreeme nt we re reached,
there would still be questions about how the theory sho uld be interpreted for
language teaching practice,
While some teachers watch th eo ry development with interest . the y must sti ll
continue co teach and plan lessons and assess stu dents' performance in the
absence of a comprehensive theo ry of second language learning. A growing
body of applied research draws on a wide range ofthe o retical o rientatio ns,
so me tim es explicicly stated, so m etim es merel y implied . This research ma y
pr ovide information that is more helpful in guiding teachers' reflections
abo ut pedagogy. In C hap ters 5 and 6 , we will examine language acquisition
research th at has focu sed o n learn ing in the classroom .
Q ue st io n s for reflection
Several theories for L2 learn ing have been proposed in this chapter . Is o ne
of them more consistent with your own understanding of how languages
are learn ed? Ifso. ho w have your ex periences as a teacher or learner
brought you co this view?
2 Schm idt 's noticing hypothesis-that all second language learni ng in adu lts
invo lves awa ren ess of what is be ing learned-is somewhat con troversial.
That is, it has bee n argued t hat it is also possible to learn incidentally,
withou t any awareness or even an inte ntion to learn. However, second
language lea rners certai nly do have 'aha ' moments when they sudde nly
unde rsta nd so mething abo ut how t he target language works. Do you have
any exam ples of noticingfro m your own language learning ex periences. or
from th os e of your st ude nt s?
3 From the perspective of the interaction hypothesis. mo dified interaction
is se en as an essential res o urce for second language learn ers.This is
distinguished from modi fied (o r simplified) input . Can you th ink of so me
exa mples of each?What are some of th e feat ures of mod ified interaction
t hat you think are especially helpful to learn ers ?Are th e re some featu res
that may not support learning ?Wh at are th e co ntexts in wh ich sec o nd
language learn ers are most likely to benefit from modified interaction? Do
you thin k that simplified input is (also) important?
122 Explaining secondlanguage learning
Suggestions for further reading
Dornyei, Z. 2009. The Psychology ofSecond LanguageAcquisition. Oxford:
Oxford University Press.
This overview of the theories that have been proposed to explain second
lan guage acq uisition is both comprehensive and easy to read. Dornyei pro­
vides detailed trea tment of the theories th at are discussed in this chapter,
focusing pa rticularly on those arising from the research in cognitive
psychology. In addition, the book introduces the work in neurobiology
that provides a new level of explanation for language acquisition and use.
Swain, M., P. Kinnear, and L. Steinman. 2010. Sociocultural Theoryand
Second Language Education: An Introdu ction through Narratives. Bristol:
Multilingual Matters .
In this book the authors cover the key concepts of sociocultural th eo ry
(for exam p le, mediation, zone of proximal development. private speech,
collaborarive dialogue) through the use of narratives. The narratives come
from the voices of language learn ers and teachers from different educa­
tional COntexts. The book is of particular interest to readers motivated to
understand how sociocultural theory relates to the teaching and learning
ofsecond languages.
VanPatten, B. and]. Williams (eds.). 2007. Theories in Second Language
Acquisition: An Introduction. Mahwah, N]: Lawrence Erlbaum
Associates.
VanParten and Williams set out a list ofobservations that have arisen from
research studies in second language acquisitio n. Then, well-known authors
discuss how th e theo retical framework in which they have done their own
research would explain these observations. For example, there are chapters
on Universal Grammar. sociocultural theory, skill acquisition th eo ry, pro­
cessabiliry, and input processing. The chapters are brief (about 20 pages,
including discussion questions and readi ngs) and written in a style that is
accessible to those with limit ed background in resear ch and theory. The
final chapter, by Lourdes Ortega, provides a concise overview ofthe d iffer­
ent theories and identifies some ongoing ch allenges for explaining second
language acquisition.
5
OBSERVING LEARNING
AND TEACHING IN THE
SECOND LANGUAGE
CLASSROOM
Preview
In this chapter we explore different ways in which researchers have observed
and descri bed what goes on in second language classroo ms: Befo re we do th is,
let us take a moment to reflect on the differences between classroom settings
for language learning and other settings where people learn a new language
without instruction.
As we saw in th e activity in Chapter 2, learning a second language in a non­
instructional setting is differen r from learning in the classroom. Many believe
that learning 'on the street' is more effective . This belief may be based on the
fact th at mos t successful learners have had experience using the language
outside the classroom. What is special about th is 'natural' language learn ing?
Can we create the same environment in the classroom? Should we? Or are
there essential contributions that only instruction and no t na tu ral exposure
can provide?
Natural and instructional settings
Natural acquisition contexts should be understood as those in which the
lear ner is exposed to the language at work or in social interactio n or, if the
learner is a ch ild , in a school situation where mos t of the other chil dren are
native speakers of the targe t language and where the instruction is directed
toward native speakers rather than toward learners of the language. In such a
classroom, much of a child's learning take places in interaction with peers as
well as through instruction from the teacher.
In structure-based instructional environments , the language is taugh t to a
group of second or foreign language learners. The focus is on the lan guage
itself, rather than on the messages carried by th e language. The teacher's
goal is to see to it that students learn the vocabulary and grammatical rules
o f the target language. Some students in structure-based classes may have
124 Observing leaming and teaching in the second Lmguage classroom Observing learning and teaching in the second language classroom 125

oppo rruru nes to co nti n ue learning th e target language o utside th e class­ ACTIVITY C o m pare learning contexts
room; for others, the classroom is me on ly contact with that language. In The cha rt inTable 5.1 is similar to the one in Table 2.1 in Chapter 2.ln that
some cases, the learners' goal may be to pass an examination rather man to chart, we compared the profi les of first and second language learn e r s. In this
use the language for daily communicative interaction beyond the classroom. one, we compare natural and instructional contexts for second language
Co m m un icative. content-based, and task-based instructional environmen ts learning.Think about the characteristics of the four contexts represented by
also in volve learners wh ose goal is learning the language itself, but the sty le of each column, For each context, decide whether the characteristics on the left
instruction places the emphasis on interaction. co nversatio n, and language are present or absent. Mark a plus (+) in the table if the characteristic is typical
use, rather man on learning about the language. The topics mat are discussed of that context. Mark a minus (-) if it is something you usually do not find
in co m m un icative and task-based instructional environments are o fte n of in that context.Write '?' if you are not sure . Note that the 'Co mmunicative
general interest to me learner, for example, how to obtain a driver's license. instruction' column has been subdivided into teacher-student and student­
In content-based language teaching (C BLT), the foc~ of a lesson is usually student interaction .What happens when learners talk to each other? Is that
on the subject matter, such as history or marhemarics, which students are different from what happens in teacher-student interaction?
learning through me medium of the secon d language. In thes e classes, the
Characteristics Natural Structure- Communicative
focus may occasionally be on the language itself, but the emphasis is on using
acquisition based instruction
the language rathe r than talking about it. The language mat teachers use for instruction
teach ing is not selected so lely for the purpose of teaching a specific feature
of the langu age, bu t also to make sure learn ers have the lan guage they need Teacher- Student­
to interact in a variety of contexts. Students' success in these courses is often
student student
meas ured in terms of thei r ab ility to 'get th ings do ne' in the seco nd lan guage, Learning one
rather than on their accuracy in using certain grammatical features. thing at a t ime
Frequent
In naturalacquisition settings feedback on
When people learn languages at work. in social interactions, or on the play­ errors
gro un d, their experiences are often quite differ ent from those of learners Ample time for
in classrooms. C o m p lete Table 5.1 on the nex t page. As you look at m e learni ng
pattern of + and - signs you have placed in th e cha rt, you will probably find
High ratio of
it matches the descr iption s below.
native speakers
o Language is not presented step by st ep . Th e learner is exposed to a wid e to learners
variety ofvocabulary and str uctu res. Variety of
• Learners' errors are rarely corrected. If th eir interlocutors can under­ language and
stand what they are saying, they do not remark on the correctness of the discourse types
learners' speech. They would probably feel it was rude to do so.
Pressure to
• The learner is surrounded by the language for many hours each day. speak
Sometimes the language is addressed to the learner; sometimes it is
simply overheard. Access to
Q The learner usually encounters a number of different people who use modified input
the target language proficiently. Photoccpiable © Oxford Univers icy Press
• Learners observe or participate in many different types of language
events: briefgreetings, co m mercial transactions, exchanges of informa­ Tab le 5. I Contexts (or language learning
tion, arguments, instruction at schoo l and in workplace interactions.
126 Ob serving lea rning and teaching in the second language classroom
• Older children and adulrs may also enco unre r rhe wrirren language in
the use of video and web -based materials .
• Learners must often use their limited second language ability to respond
to questions or to get information. In these situations, the emphasis is
on getting meaning across clearly, and more proficient speakers tend to
be tolerant of errors that do no t In terfere with meaning.
• Modified in put is available in many o ne-to-one conversations. In sit u­
ations where many na tive speakers are involved in th e conversation,
however, lear ners may have d ifficulty getting access to language the y
can understand.
In structure-based instructional settings
The events and act ivities that are typical of structure-based instruction differ
from th ose encounrered in na tural acq uisi tion set tings. In grammar tran sla­
rio n approaches, there is considerable use of readi ng and writing, as learn ers
translate text s from one language to an o th er, and grammar rul es are taugh t
explicitly. In audio lingual ap proaches there is lit tle use o f translatio n, and
learners are expected to learn mainly through repetition and habit formation,
although they may be asked to figu re out the grammar rules for the senrences
they have memorized .
• Linguistic items are presented and practised in isola tion, one item at
a time, in a sequence fro m what teachers or textbook wri ters bel ieve is
'sim ple' to that which is 'co mplex' .
• Errors are frequently corrected. Accuracy tends to be given priority over
meaningful interaction.
• Learning is often limited to a few hours a week.
• In situatio ns of foreign lan guage learning th e teac her is often the only
native or proficient speak er the student comes in contact with.
• Students experience a lim ited range of language discourse types . Th e
most typical of these is the Initiation/Response/Evaluation (IRE)
exchange where the reacher asks a question, a student answers, an d the
teacher evaluates the response. The written language studen ts encoun­
ter is selec ted primarily to provide practice with specific grammatical
feat ures rather than for irs content,
• Students often feel pressure to speak or write the second language and
to do so correctly from the very beginning.
• Teachers may use the lear ners' native language to give instructions or
for classroom management. When they use the target language, th ey
tend to modify their language in order to ensure comprehension and
compliance .
Observ ing learning and teaching in the second language classroom 127
f!1e;h- i~":!:1~f'< ~~'e-oJI _
The a...x..1.'t1iJ ver-b, ""n.:c.."
"Dr-.~ t:4pPetVS after -t~e
subjeL-t.., ..,...,st ~ove 'tP k~t:e
i'l c:na.-L po~t"on .. .
~~>'4--3
<, r\~
~J~
~~
,~?-
Language classrooms are not all alike. The conditions for lear ning diffe r
in terms of the physical environment, the age and mo tivation of the Stu­
dents, the amount o f time available for learning, and many oth er variables.
Classroo ms also differ in terms of the prin ciples that guide teachers in thei r
language teaching methods and techniques. D esigners of commun icative
language teac hing programmes have sought to replace so me of th e charac­
teristics of st ructure-based instruction with those more typical of natural
acquisition con texts.
In communicative instructional settings
In com m un icative and co ntent-based instructio n, the emphasis is on the
communication of meaning, both between teacher and students and among
the sru dents themselves in group- or pair- work. Grammatical forms are
foc used on only in order to clarify meaning. The assumption is that, in focus ­
ing on mean in g, learners will acquire the language in a way that is similar to
natural acquisition.
• Input is sim plified and made comprehensible by the use of contextual
cues, props, and gestu res, rather than th ro ugh structural grading.
Students provide each o ther with simplified an d sometimes erroneous
input.
• There is a limited amount of error co rrec tion on the parr of the teacher,
and meaning is emphasized over form . Srudents tend no t to over tly
correct each other's errors w hen they are engaged in co m munica­
tive practice. Because the focus is on meaning, however, req uests for
128 Observing teaming and teaching in the second ftmguage classroom
clarifica tio n ma y se rve as implicit feedback. N ego tia ting for meaning
may help students see the need co say something in a different way.
• Learners usually have only limited time for learning. In a typical teacher­
fronted classroom with 25-30 students, individual students get very
little o p po rtu nity CO produce language in a GO-minute class, and when
th ey do, it is usually in the form of a sho rt response co a teacher's ques­
tion. When stud ents work in pairs or groups, they have opportunities
co produce and respond CO a greater amount and variety of language.
Sometimes, however, subject-matter courses taught through the second
language can add time fo r language learning. A good example of this
is in immersion programmes where mos t or all the subject matter is
taught to a group of students who are all second language learners.
• As in structure-based instruction, it is usually only the teacher who is
a proficient speaker. Learners have considerable exposure to the inter­
language of other learners, particularly in student-student interaction.
This naturally contains errors that would not be heard in an environ­
m ent where the interlocu to rs are nati ve speake rs, but it provide s m an y
mo re opportu nities for stu de nts to use th e target la ng uage tha n is th e
case in most teach er-fro nt ed activities.
• A variety o f discou rse ty pes m ay be int rod uced t hrough stories , pee r­
a nd gro up-work. the use of 'au the ntic' m aterials such as newspapers and
television broadcasts . Text m aterials m ay include both those modified
for seco nd language learners and those intended for native speakers. In
the latter case, teachers use instructional st rategies to help learners get
th e meaning, even if they do not know all the words and structures. In
student-stu dent interaction, learners may practise a range of sociolin­
gu istic and functional features oflanguage through role-play.
• There is little pressure to perform at high levels of accuracy, and there is
o ften a greater emphas is on co m preh ensio n than on production , espe­
cially in the early stages oflearning.
• Modified input: is a defining feature of this approach to instruction. The
teacher makes every effort to speak to students in a level of language
they can understand. Ifstudents speak the same first language, they may
have little difficulty in understanding each other. If they come from dif­
ferent language backgrounds, they may modify their language as they
seek to communicate successfully.
General descriptions of classroom instruction such as those above cannot
capture the individual characteristics of particular classrooms . For this
reason , researchers have developed a number of ways to study classroom
learning and teaching. We will discuss two approaches to classroom resea rch
in this chapter. We will look first at observation schemes, in wh ich research­
ers anticipate the occurrence of particular events and behaviours and make
note o f them within pre-planned frameworks or checklists. Then we will
Obser ving teaming and teachi ng in the second lmgut.lge classroom 129
loo k a t classroom ethnography, an ap proac h that requires th e o bserver to
describe what happens in the classroom , looking for patterns and relation­
ships. but trying not to limit the observation to any predetermined cat egories
or expectations.
Observation schemes
Many different observation schemes ha ve been developed for use in seco nd
language classrooms. They differ in several respects, includ ing the number
of categories they contain, whether they focu s o n qualitative or quanti­
tative descriptions, and whethe r they are used throughout a lesson or on
selected samples of classroom interaction . The schemes also differ in rela­
tion to whether they are used by o bservers in ' real time' while they are in
the classroom, o r used later outside the classroom to analyse audio or video
recordings or transcripts of such recordings.
O ne example of a scheme developed speci fically fo r seco nd language class­
roo ms is th e Com m u nicative O rien ta tio n o f Language Teac hi ng (C O LT)
Observat ion Sc he me descri bed by Nina Spada an d M aria Fro hlich ( 199 5l.
CO LT is di vided into two parts. Parr A d esc rib es teaching practices in terms
of co nte nt. focu s. and o rganization of ac tivity types . 'X'he n using Parr A ,
th e o bserve r ca n record, for exa m ple, whethe r the pedagogical ac tivities
a re teach er- or learner-centred, whether the focus is o n language fo rm o r
meaning, and whether there at e o ppo rt un ities for studen ts co choose the
topics for discussion. Parr B describes specific aspects of the language pro­
duced by teachers an d stud en ts, for example, how much (o r how little)
language students produce, whether their language production is restricted
in any way, the kinds of questions teachers ask. and wh ether and how teach ­
ers respond to learners' errors.
The COLT scheme and others like it have been used primarily in classroom
research that is intended to look at how differences in teaching practices are
related co differences in second language learning. Observation schemes have
also been used in the training of new teachers and in the professional devel­
opment of experienced ones.
Below is an activity in which you are asked co use a set of pre-determined
catego ries similar to those used in the COLT scheme co characterize the
nature of interaction between teachers and students and between students
and students.
Classroom comparisons: Teacher-student interactions
Excerpts from four transcripts of second language classroom interaction are
given in th is and the following section. The first twO present teacher-student
interaction. The transcripts come from classrooms that differ in their approach
 Observing learn ing and teachi ng in the second language classroom O bserving learn ing and teaching in the second language classroom 131
130

co secon d lan guage teaching; one of them represents structure-based instruc­ Unit 4 . Don't take your bo oks yet, don't take yo ur books. In 1,2 , 3
tion ; the other, a communicative approach. Stru cture-based approaches we worked in what ten se? What tense did we work on? OK?
emphasize language form throu gh either meralinguistic instruction (for s Past.
example, grammar translation) or pa ttern practice (for example, audiolingual) . T In the past-What auxiliary in the past?
s Did.
Wi th each tran script , th ere is a char t where you can indicate whether certain T D id (writes on board ' 1- 2- 3 Past'). Unit 4, Unit 4, we're goi ng co
th ings are happening in the int eractio n, fro m the point of view of the teac her work in the prese nt, present p rogressive, p resent conti nuous- O K?
and that of the stu dents. Before yo u begin readi ng the tran scrip ts, study th e
You do n't kn ow what it is?
followi ng interpretatio ns of th e categorie s used in the charts:
s Yes
1 Errors:Are there errors in the lan guage of either the teacher or the students? T Yes? What is it?
2 Feedback on errors: When stu dents make errors, do they receive feedback? s Little bit.
From who m? T A lit tle bit.
3 Gen uine q uestio ns: Do teachers and students ask q uestio ns to which they s
don't know the answer in advance? T Eh?
4 D isplay questions: Do teachers ask questions that they kn ow the answers s Uh, presen t continuo us
to so that learners can disp lay the ir knowledge of the lan guage (or lack T Present co nti nuous? What's that ?
of it)? s e-n-g
5 Negotiation fo r meaning: Do the teachers and students work to u nder­ T i-n-g
stan d what th e other speakers are saying? What efforts are made by the s Yes.
reach er? By the students? T What does that mean, present contin uo us?You do n't know? OK,
6 Meralinguis tic com m en ts: D o th e teachers an d students talk abo u t lan­ fine. What are you doing, Paul?
guage, in addition to using it to transmit infor mation? s Rien [nothing].
T N othing?
In the following excerpts, T rep resents the teacher; 5 rep resents a student.
s Rien-nothing.
(The first rwo examples come fro m unp ub lished data co llected by P. M .
T You're not doing anything? You're doing something!
Lighrbown, N . Spada, and B. Barkman.)
s Not doing anything.
T You're doing something!
Classro o m A: A structure-based appro ach

s Not doing anything.

(Students in th is class are I5-year-old French speakers .)
T You're doing something-Are, are you listen ing to me? Are you
talking with Marc? What are you doing?
Te a che r St uden t
s No, no--uh-listen­ uh-
Erro rs T Eh?
Fee dback on errors s co you .
T You're listening to me.
Genu ine quest ion s
S Yes.
Display question s T Oh. (writes 'What are you doing? I'm listening to you' on the board) .
s Je-[I .. .] .
Nego tiat ion for mean ing
T What are yo u-? You're excited.
Meta linguist ic co mments s Yes.
T You're playi ng wi th your eraser (writes 'I'm playing with my eraser'
Photocopiable © O xford Unive rsity Pre ss
on the board). Would you close the door p lease, Bernard? Claude,
T O K, we finishe d the book- we finished in the book Unit 1,2, 3. wh at is he doing?
Finished. Wo rkbook 1, 2, 3. So today we're going to sta rt wi th s Close the doo r.
l32 Observing learning and teaching in the second Lmguage classroom Observing learning and teaching in the second language classroom 1.33

T H e is closing the do or (w rites ' H e's closing the door ' on the board). T Every day? Ah! Does n't your b ro-(ina ud ible) h is bicycle? Could his
What are yo u doing, Mario? brother lend his bicycle? Uh, yo ur bro ther doesn't have a bicycle ?

s Yeah! A new bicycle (inaud ible) bicycle.

Classroom B: A communicative approach T Ah, well. Talk to your mom and dad about it. Maybe negotiate a new

bicycle for your brother.

(Student s in thi s class are IO-year-old French speakers. In this activity, they
are telling their teacher and their classmates what 'b ugs' th em. They have
s (inaudi ble)

T He has a new bicycle. But his brother needs a new one too.

written 'what bugs them' on a card or paper that they hold whil e speaking.)
s Yes!

Teacher Student T Hey, whoa, JUSt a minute! Jean?

s Martin's brother has-

Errors
T Martin, who has a new bicycle? You o r your brother?

Feedback on errors s My brother.

Genuine questions T And you have an old one.

s (inaudible)

Display questions T And your brother tak es your old one?

Negotiation for meaning s (inaudible) bicycle .

T His bicycle! H ow old is your b rother ?

Metalinguistic comments
s March 23.

Photocopi able © O xford University Press T H is b irthd ay?

5 Yeah !

s It bugs me when a bee stri ng me . T And how o ld was he?

T O h . wh en a bee stings me. s Fourteen.

s Stings me. T Fourteen. Well, why don't yo u tell your br other th at when he takes

T Do you get stu ng often? Does that happen oft en ? Th e bee stinging
many times?
s Yeah.
T Often? (Teacher turns to students who aren't paying attention) OK.
Sand ra and Benoit, you may begin working on a resear ch project,
hey? (Teacher turns her attention back to ' \'\'hat bugs me ')
s It bugs me (inaudible) and my sister pUt on my clothes.
T Ab! She borrows your clothes? When yo u're older, you may
app reciate it becau se yo u can switch clothes, ma yb e. (Turn s to chec k
ano ther student's wr itten work) Melan ie, this is yo urs, I will check­
OK. It 's good.
s It bugs me when I'm sick and my brother doesn't help me-my-my
brother, 'cause he-me-.
T 0 K. You know-when (inaud ible) sick, you're sick at home in bed
and you say, oh , to your brother or your sister: 'W o uld you please get
me a drink of water?' -'Ah! D rop dead! ' you know, 'Go play in the
traffic!' You know, it's nor very nice. Martin!
s It bug me to have-
T It bugs me. It bugzz me.
s It bugs me when my brother takes my bicycle. Every day.
your bike you will take his bike? And he may have more scratches
than he figures for. OK?
Characteristics of input and interaction
C om pare the rwo charts yo u hav e complet ed so far. What kinds of seco nd
language input and opportunities for interaction are available to learners
in each of the en vironments th at thes e tr ans cripts exemplify? HQw are the y
d ifferent?
Classroom A
I Errors: Very few on the part of the teacher, However her speech does have
som e peculiar characteristics typical of this type of teaching, for example,
the questions in statement form-often asked with dramatic rising into­
nation (for example, 'Yo u don't know what it is?'). Students don't make coo
many errors because they say very little and what they say is usually limited
by the lesson.
2 Feedback on errors: Yes, w henever students do make errors , the teacher
reacts.
3 Genuine questions: Yes, a few, but the y are almost always related to class­
room management. No questions from the students.
134 Observing team ing and teaching in the second language classroom
4 Display questions: Yes , almost all of the teacher's questions are o f this type .
Interestingly, however, the students sometimes interpret display ques­
tions as gen uine questions (T : What are you doing, Paul? S: Nothing.).
The teacher wants stu den ts co produce a sentence-any sentence-in
the 'prese nt continuous ' but the student worries th at he's about co get in
tro uble and asserts that he is do ing 'nothing'. This is a good exam ple of
how the teache r's p ragmatic intent can be misinterpreted by the student,
and o f how stro ngly, even in this setting, students seek co find genuine
meaning in language.
S Nego tiation for meaning: Very little, learners have no need co paraphrase
or request clarifications, and no opportunity co determine th e direction
of th e d iscour se; the teacher is focused on ly on th e formal aspects o f the
learners' language. All the effort goes into getting students co p roduce a
sentence with the present continuous form of the verb.
6 Meralinguisric com me nts: Yes, this is how the teacher begins the lesson
and lets the stu dents kn0:W what really matters!
Classroom B
1 Errors: Yes, students make errors. And even the teacher says so me odd
thi ngs sometimes. H er speech also contains in complete sentences, sim pli­
fied ways of speaking, and an in formal speech style .
2 Feedback on errors: Yes, sometimes the teacher repeats wha t the student
has said with the correct form (fo r example , 'he bugzz me '--emphasizing
the third person sin gular ending). However, this correction is no t consist­
ent or intrusive as the focus is pri marily on leering students exp ress th eir
meanings.
3 Genuine questions: Yes, alm ost all of the teach er's q uesti ons are foc used on
getting information from the students. Th e students are no t asking ques­
tions in this exchange. However, they do sometimes intervene co change
the direction of the conversation.
4 Dis play questions: N o , because there is a focu s on meaning rather than on
accuracy in grammatical form .
S N ego tiatio n for meaning: Yes, from the teacher's side, especially in th e
long exchange abou t who has a bicycle!
6 Meralinguisrlc comments: No. Even though the te-acher clearly hopes co
get stu d en ts co use the third person ending, she does no t say so in these
words.
You no do ubt no ticed how strikingly different these two transcripts are ,
even though the activities in both are teacher-centred. In th e tra nscript fro m
Classroom A, th e focus is on for m (i.e. grammar) an d in C lassroo m B, it is on
meaning. In C lassroom A, th e only purpose of the interaction is to practise
the present co ntinuo us. Although the teacher uses real classroom events and
some humour co accomplish thi s, there is no real interest in what st udents
Observing team ing and teaching in the second language classroom 135
are do ing. Rather th e teacher is highlighti ng th eir ability co say what they
are do ing, using th e correct verb form. There is a primary focus on correct
gram mar, displ ay questions, and error co rrection in the transcrip t from
C lassroom A. In th e transcript from Classroom B, the focus is on mean ing,
conversational interaction, and gen uine questio ns, although there are som e
brief references co grammatical accuracy when th e teacher feels it is necessary.
Classroom compa risons: Student-student interactions
Th is sectio n presents so me student-st udent interactions. The tran script s are
based on the interactions between seco nd lan guage learners engaged in dif­
ferent communicative tasks.
As in th e p revious section, the re is a ch art with each transcrip t where you can
indicate whether certain things are happening in the interaction.
Communication task A: Picture description
The foll ow ing tran script is of twO girls aged 11- 12 years, both ESL learn ­
ers in the ir first year of learning English in Australia. The first learner (S 1)
is from Hong Kong; the seco nd (S2) is from Somalia. They are engaged in
a task wh ere S 1 is describing a picture for S2 co d raw. They are sitting at a
table, separated by a sm all bar rier, so that they can see eac h other's faces and
hands (when they gestu re), bue no t each other's pic ture. The picture S 1 is
drawing is a black o u tline containing stic k figures-a boy flying a ki te an d
a girl holding his hand. Th e stick figu res are standing on some grass near
a tree. Sq uar e brackets indicate non-target pronunciation. (This tran script
comes from unpublished data collected by Alison Mackey, Rho nda Oliver,
and Jennifer Leeman.)
Student I Student 2
Errors
Feedback on errors
Ge nuine questions
Display questions
Ne go tiat ion for meaning
Meta linguist ic co mments
Phococopiable © Oxford University Press
S I And o-on th e right, there is a [tree]. It's a- a, the ki- , the kite is up .
(Po ints up in the air) This is the kite. (Po in ts up again ) Thi s is the
kite . (Points yet again ) An d the [tree] is u p there.
sz Three bird?
136 Obst!Tving learning and teachmg in the second language classroom Observing learning and teaching in rhe second language classroom lr
51 Huh ?

52 Is a three bird ?
Dara N ina
51 Huh?
Errors
5 2 Up , up-up me kite?

Feedback on errors
s 1 Yeah , m e kite is u-, the kite is up and me [tree ] is down. (Po inrs

di rections) Genuine questions

S 2 The [bird ] down?
Display questions
51 The kite-, the [tree] is down .
52 What's me [tee]? Negotiation for mean ing
51 H uh?
Metalinguistic comments
5 2 What's th e [tree]? (Imitates Lear ner 1's pro duct ion)
S 1 Feel? Pho coco piable © O xfor d Univers ity Press
52 Fell?
Fell down? (Poin ts down ) D Elle. [She]
DElle se so uvient, non , Ell« souuient q u'e lle a un ' " une pratique de
SINo, it's not m e fell down. N o , it's JUSt at the bo tto m .
5 2 The bird ?
chorale, alo rs elle se [eve. [She remembers, no. She remembers th at
sI No, the tree. . she has a cho ir practice so she gets up]
52 The tree? (Em p ha tic stress)
o Tout it co u p [Sudde n ly]

S 1 Yes.
NElle ssse [She ssse]

S 2 Ir is left and righ t?

o Elle souu ient [She remembers]

5 I It's right. (Poinrs)

N se so uvien t au souuientt [Reme mbers or rern cmber sr]

52 It's long? It 's [little]?

DElle sou uient . . . ahh , d ie se so uvien r .. . Ell« souuient Elle se

5 1 It's-what?
so uvien r, no . [She remembers . . . ahh , she remembers She

5 2 It's lo ng an d [Iittle]?
remem bers . . . She remembers, no]

s lUm, a lirtle-. It-, urn , a middle size. N pas . . . [not]

52 M iddle size tree?

o Ell« souuient q u'elle do icaller au band . . . ? [She remembers mar she
51 Yes.
has ro go to band]

5 2 It's llrrle, (Says as drawing me tree )

N Cho rale. [C hoir]

C o mmunica t ion task B: Jigsaw
The follow ing transcript is of rwo students in a Gra de 7 F re ~ch im mersio n
classroom. They are engage d in a jigsaw act ivity based o n a series of eigh t
o C ho rale. [C hoir]
N Tout it co up ell« souu ient q u'elle . .. do it aller it la cho rale [All of a
sudden she rem em bers that she has ro go to choir]
o [very softly] elle se so uvie nt ... non . [She remembers . . . no]
o Alo rs, elle [So, she]

pictures telling the Story o f a yo u ng girl bein g awakene d by her alar m clock N non, wait. Tour a coup elle .. . se souvienr? [no , wait .. . all of a

early in the morning. One stude n r has p ictures 1, 3,5 . and 7, and the oth er sudden she . . . rem embers?]

srude nr has pictures 2, 4, 6 , and 8 . Th ey take rums relling m e Story portrayed D Je pen se pas qu e c'esr se souvienr. [I don't th ink it's rem ember]

by me picrures and men th ey display all the pi ctures in sequence and write N oh, souvienr . , . so uvienr. [oh rem em ber . . . rem em ber]

the story th ey have JUSt rold . Telling me srory requires th e use ora number of o Elie souuient qu 'elle a Ie cho rale. [She rem emb ers th at she has choir]

reflexive verbs in Fren ch . In m e third person , th e form se is placed between N Qu'elle doit se prep arer. [ma t she has to get read y]

the subject and the finite verb. Thus, elle se ieve (she gets up ) and ellesesouuient D O ui. [Yes]

(she rem embers) are correct. Th e learn ers are called Oara (D) and Nina (N ). N pour le cho rale i . . no n, to u t it co up elie souuient qu'il ya un e pr atique

In the transc rip t incorrect uses of the reflexive verbs are in itali cs; o the r errors de cho rale. [for ch o ir . . . no . all of a sudden she rem embers that
are not marked . (Th e data are from Swain and Lapkin 2002.) th ere's a cho ir p ractice]
138 O bserving lea rning and teaching in the second language classroom
Characteristics ofinp u t and interaction
Compare the two ch ar ts yo u have completed . As before, wha t kinds ofsecond
language input and oppo rt uni ties for interacti on are available co learners in
each of the en vironments that these tra nscripts exemplify? How are they d if­
ferent from eac h other and the teacher-student interactions you looked at
pr evio usly?
Com m unication task A
1 Errors: There are m any errors in the speech of both learners. This includes
grammatical and pro n un ciatio n errors. These erro rs are p resent in several
b reakdowns in th e learners' conversatio n.
2 Feedback on errors: There is no error correction in terms of form as t he
learners struggle co understand each other's meaning. The difficulty they
are hav ing in communication may serve as a kind of implicit feedback.
That is, the fact that the in rerlocuto r does not understand may signal tha t
there is something wrong w ith what they ha ve said .
3 G enu ine questions: Yes, the re are m any genuine questi ons . Naturally, S2
asks m ost of these questio ns beca use she nee ds co get the informatio n fro m
S 1 in or der to draw the p icture. S 1 also asks some gen ui ne questions and
the se are almost always to ask for clarification.
4 Display questions: No, there are no display questi ons because they
are engaged in a real communication-gap exchange. S2 cannot see th e
picture that S 1 is describ ing. Therefore all the questio ns asked are gen uine
questions.
5 Negotiation for meanin g: Yes, indeed! Both lear ners are trying hard to
unders tan d each other, even th ough the y often fail to do so . This involves
m any co m prehensio n questions and clarifica tion req uests , as we ll as reperl­
tions of each other's utterances, often with emphasis , trying co unde rstan d
what the o ther learner has JUSt said.
6 M eral inguisric co m m en ts: No ne.
Communication task B
1 Errors: Bo th lear ners make several grammatical errors, most no tably th e
repea ted failu re to produce the reflexive form of th e verb se souvenir.
2 Feedback on errors: There is no act ual error co rrecti on provided. Neither
learner is really sure whar the correct form is. Instead, there is rneralinguis­
tic reflection and discussion as they try to figure out whether the y are using
the correct form of the verb se souvenir.
3 Genuine questions: The questio ns tha t are asked are genuine. The content
is language form, but the students are genuinely sharing information
about how to complete the task.
4 D isp lay q uestio ns: There are no d isp lay q uestions. The students are actively
collaborating to reconstruct the Sto ry and are aski ng genuine q uestions of
each other.
Observing learning and teaching in the second language classroom 139
5 Negotiation for meaning : At this po int in the interactio n, th e stu dents
have agreed o n the content of the story. Thus, there is mo re negotiation of
form, that is, more discussion of whe ther the y are using the co rrect forms
to say what th ey've agreed th ey wa nt to say.
6 Meralinguisric co mments: Although the y ar e not using words such as
'verb' or 'pro noun', the Stude nts are talki n g about language as th ey focus
on try ing to find the right form .
These two transcripts of Student- stud ent intera ction are very d ifferent from
eac h other. In the first commun ica tion task, the children are focused exclu­
sivelyon meaning and on try ing co understand each othe r in order to complete
th e in fo rm atio n -gap activity. They are constantly using co m prehension and
clarification requests as they negotiate fo r meaning in this task. In the second
student-student transcript, however, the learners are focused on both form
and meani ng. While recons tru cting the sto ry, they make several explicit
sta tem ents abou t whether th ey are using th e co rrect form of th e reflexive verb
se souvenir an d co ntinually question the grammatical accuracy of their use of
rh is for m as [hey conrinue co disc uss the co n te nt of the sto ry,
In the activities in the preceding pages, we have described an d compared
teacher- studen t and student-student interaction in terms ofsix observation
catego ries. Some observation sche mes use many more categories, cover ing a
broad range of inst ructional practices and procedures. O thers foc us o n one
specific feature of classroom instruction and imeractio n. In .the followi ng
sectio ns, we review classroom resea rch in which one particular feature of
instr uctio n has been examined. Five studies examine co rrective feedback,
fo ur investigate teach ers' use of questions and o ne describes th e amount an d
d istr ibu tion of time for L2 learning.
Correctivefeedback in the classroom
Study 1: Recasts in content-based classrooms
Ro y Lyster and Leila Ranta' (1997) de veloped an observation schem e which
describes different types of corrective feedback teachers give on errors and
also examines student uptake-an indication that the srudent has not iced
the feedback. . This scheme was devel oped in French immersion classroo ms
w here seco nd language srud en rs learn the target language via subject-matter
ins tru ct ion (i.e. content-based language teach in g). It has also been used co
desc ribe feedback in other types of seco n d language instruction.
They develo ped th eir sch eme by observing th e different types of co rrective
feedback provided during interaction in four French im m ersio n classrooms
wi th 9-II -year-old stud ents. They began their observations by usin g a co m ­
bina tio n o f some catego ries fro m Pan B of the C O LT scheme an d other
categori es from models that had examined feedback in both first and seco nd
140 Obsn"vmg learning and teaching in the second language classroom Obseruing learning and tea.·hing in sbe second language classroom 14 1

language learning. They adjusted so me of the categories [0 fit their data, th e nature o f the erro r but attempt [0 elici t the information from the stu de nt
and they also develo ped additional categories. This resul ted in th e id en ri­ (fo r example, 'Is it femi nine? ') .
ficat ion of six corrective feedback types, defined below. The defini tions are
s We look at th e pe op le yesterday.
tak en fro m Lyster and Ranta (1997 : pp . 46-8) . The examples come fro m
T 'W'hat's the ending we put o n verbs when we talk about the past?
1O-ll -year -o ld students in ESL classes that we have observed.
s e-d
Explicit correction refers [0 th e explicit provision of the correct form . As
Elicitation refers ro at leas t three techniques that teachers use to directly e1icit
the teacher provides the correct form, he or she clearly ind icates that
th e correct form from the students . First, teachers elici t completion of their
what th e student had said was incorrect (fo r example, ' O h, you mean . .. ',
own utterance (fo r exam ple, 'It's a .. .') . Second, teachers use questions to
' Yo u sh o uld say .. . ') .
elicit correct forms (fo r example, ... ' H ow do we say x in English? ') . Third,
s The dog run fasdy. teachers occasionally ask st uden ts to reformulate their utterance. '
T ' Fasrly' doesn't exist. ' Fas t' does not take -Iy. That's why I picked
'q u ickly'. s My father cleans the plate.
T Excuse me, h e cleans the - -?
Recasts involve the teacher's reformulation of all or part of a student's utter­ s Plates?
ance, minus the error. Recasts are generally implicit in that they are not
Repetition refers to the teacher 's repetition of the student's erroneous utter­
introduced by ' Yo u mean' , 'Use thi s word', or 'You sho uld say.'
ance. I n mos t cases, teach ers ad just th eir into na tion so as to hi ghl ight th e
SI 'W'hy yo u do n't like Marc?
erro r.
T \XThy don 't yo u like Marc?

SZ I do n't know, I don't like him .

Jn th is exa m pie, th e rep et itio n is followed by a recast:

5 H e's in the ba th roo m.
Note th at in this exa m p le the teacher does nor seem [0 ex pec t u ptak e from
T Bathroom? Bedroom. He's in th e bedroom .
S 1. It seems she is merely refo rm ulating the question S 1 has asked 52 .
In the next exam ple, the repetition is followed by rneralinguiscic co m m ent
Clarification requests in d icate to students ei ther that their utterance h as been
and explicit correction:
misunderstood by the teacher or that the utterance is incorrect in so m e way
a nd that a repeti tio n or a reformulation is required. A clarification request s We is .. .

includes phrases such as 'Pardo n me ' It may also include a repetition of T We is? But it 's two people, righ t? You see yo ur mistake? You see th e

th e error as in ' \xrha t do you mean by ?' erro r? 'W'hen it 's plural it's 'we are '.

T How often do you wash the dishes?

Lyster and Ranta found that all teachers in the content-based French immer­
s Fourteen .
sio n classes they observed used recasts more than any other type of feedback.
T Excuse me . (C lari ficatio n request)
Indeed, recasts accounted for more than halfo f th e to tal feedback pr ovided in
s Fourteen.
the fou r classes. Repeti tio n of error was the least frequenr feedback type pro­
T Fourteen what? (C larifica tio n request)
vide d . Th e other types ofco rrec tive feedback fell in becween. Student uptake
s Fourteen for a week.
was least like ly to occur afte r recasts and more likely to occur after clarification
T Fo u rte en tim es a week? (Recast)
requests, rneralinguistic feed back, and repetitio ns. Furthermore, elicitations
s Yes. Lunch and d inner.
and meralinguisric feedback not only resulted in more uptake, they were also
more likely [0 lead [0 a corrected form of the or iginal ut terance.
l ltfetafing uisticJeedback contain s comments, information, o r q ues tions related
to the correctness of the student's utte rance, witho ut exp licitly p roviding the Lyste r (I998) has argued that students receiving content-based la nguage
correct form . Meral inguiscic comments generally indicate that there is an teach ing (where the emp hasi s is on meaning not form ) a re less likely ro notice
error so mewhere (fo r exam ple, 'Can you find yo ur error?') . Also, rnetali n­ recasts than o ther forms of corrective feedback, because they m ay assume
guistic informatio n ge ne ra lly p rovides either some gramma tical te rmi nology that the teac her is responding to the co nt ent rather than the form of their
that refers [0 the nature of the error (for example, ' It's masculine') or a word speech . Indeed, the double challenge of making the subject-ma tt er com­
definition in the case of lexical er ro rs. Metalinguistic q ues tio ns also po int [0 prehe nsible and enhanci ng knowledge of the seco nd lan gu age itself withi n
conten t- based lan guage teach ing has led Me rrill Swain (1988) and others to
142 O bserving learning and teaching in the second language classroom
conclude that 'no t all content teaching is necessarily good lan guage teaching'
(p. 68). The challenges of content-based language teaching will be discussed
further in Chapter 6.
Since Lyster and Ranta reported their findings, many more observation
Stu d ies of corrective feedback in second or foreign language classrooms have
been carried our. Some of them report similar resulrs-s-rhar recasts are the
most frequentl y occurring type of feedback and that they appear to go unno­
ticed by learn ers. However, others report that learners do notice recasts in the
classroom. Below, [WO studies are described in which learners were observed
to no tice and to respond to recasts provided by their teachers.
Study 2 : Recasts andprivate speech
In a study with adult foreign language learn ers of Japanese, Amy Ohra
(2000) exam ined the oral language that learn ers addressed co themselves
during classroo m activ ities. She was ab le ro obtain this private speec h by
attaching microphones to individual students during classroom interaction
that focused on grammar and rneralinguisric instr uction. In this co ntext,
Ohta d iscovered that learners noticed recasts when the y were provided by
the instructor. Furthermore, learners were more likely to reac t to a recast
with private speech when it was directed ro an o ther learner or to the whole
class rather th an when the recast was di rected to the ir own errors. On the
basis of these find ings, she concluded that recasts do get noticed in classroom
interaction even if they do not lead to uptake from the student who originally
produced the error.
Study 3: Recasts in different instructional settings
Roy Lyster and Hirohide Mori (2006) compared learners' immedi­
ate responses to corrective feedback in French and Japanese im m ersion
classrooms. They found that the teachers in both contexts used correc­
rive feedb ack in similar ways. However, the effects of recast s on learners'
uptake were differe n t. In the Japanese immersion classes, learners freque ntly
repaired their utterances after receiving recasts whereas learners in the French
immersion classes rarely did. Instead, the greatest proportion of repa ir came
after prompts, th at is, the feedback types that indicated to students that a
correction was needed an d tha t encouraged them to self-correct .
In [heir efforts to understand these differences Lyster and Mori carried om
a detailed analysis of th e instructional characteristics in these classes. Using
the COLT observation scheme described above to capture differences in the
pedagogical practices, th ey discovered that there was an analytic orientation
in the Japanese classrooms leading the teachers and learners to focus their
atte ntion on language form and accu racy. Th e orient atio n in the French
immersion classes was mo re experienti al with a greater focus on content
and communication o f messages. Lyster and M ori argued tha t because the
Japanese learners' attentio n was regularly drawn to form, they were primed
Observing learning and teaching in the second langu age classroom 143
to notice the co rrective function of recasts. In the more meaning-oriented
Frenc h immersion classes, howe ver, recasts were less likely to signal to the
learner that the teacher was responding to a language err or. Thus is likely that
learners assumed tha t the teac hers' recast was simply a con firmation of what
they had said.
These findi ngs and those from other research led Lyster and Mori to propose
the counterbalance hypothesis . Acco rdi ng to this hypo th esis, feedback is
mo re likely to be noticed ifl ear ners are oriented in a direction that is op pos ite
to what th ey have become accustomed to in their instructional environment.
One example of th is would be that lear ners who receive L2 instruction that
is foc used on m ean ing/ coment need feedback tha t directs their attention to
form more explicitly.
Study 4: Correctivefeedback in context
Rho nda O liver an d Aliso n Mackey (200 3) carried out a descriptive study
of an Australian pri mary ESL classroom with 6-12-year-olds. They inves­
tigated whether teachers' provision and lear ners' use o f corrective feedback
differed depending on varying contexts for interactio n in a lesson . The y iden­
rified fo u r contexts in which teachers and learners interacted:
1 content exchanges: the teac he r imparted knowledge or asked questions
about the conten t of the cu rriculum
2 man agement exchanges: the teacher talked about the organization of the
lesso n an d appropriate classroom behaviour
3 communication exchanges: the emphasis was on studen ts using Englis h in
meaningful ways, and
4 exp licit lang uage -focused exchanges: the em phasis was on grammar and
the use of rneralinguistic term inology.
Oliver and Mackey found that learners produced significantly more errors in
the com m unicatio n exchanges. Thus opportunities for feedback were grea t­
est in this co ntext, The researchers found that feedback was provided in all
instructional contexts but that it was most frequent in th e explicit language­
focused exchanges, followed by conten t, comm un icatio n, and manageme m .
When they exam ined how learners reacted to the corrective feedback, they
foun d th at learners mod ified their omput most often within exp licit language­
focuse d exchanges, only some of the time in content and communication
exchanges, and never in management exchanges. Interestingly, the types of
corrective feedback also varied across co ntexts: recasts were used at a consist­
ently high rate in managemem, com municatio n, and con ten r exchanges,
bur less so in explicit language-focused exchanges; explicit corrective feed­
back was rarely pr ovided d uring co nte nt, management, and comm un icatio n
exchanges, and freq uen tly d u ring expli cit language-focused contexts.

144 Observing learning and reaching in the second language classroom
O liver and M ackey's st udy em phasizes how differ ences in the instruc­
tional context affect teachers' feedback and learners' response to it. It is also
important to keep in min d that different corrective feedback types can be
int erp reted differen tly dependin g on how they are del ivered. Recasts are a
case in poim. In a study of co rrective feedback in four different instruc­
tio nal co nt exts. Younghee Sheen (2006) observed many co n trasting type s of
recasts , including recasts that were declarative versus interrogative. reduced
versus non-reduced, single words or short ph rases versus long phrases or
clauses. Recasts can also differ according to whether they are delivered with
or without stress or emphasis. In a study of adult learners of English , Shawn
Loewen and Jenefer Philp (2006) foun d that recasts containing 'prosod ic '
stress were thirteen times more likely to result in successful uptake' (p, 547),
that is, uptake in whi ch the student produced the co rrected form . However,
these interactions were not associated with improved performance on a sub­
seq uen t test.
Othe r facto rs that may affect learners' reacti ons to different types offeedback
include age and learn in g goal s. For exa mp le, adults may be more likely to
int erpret recasts as feedbac k on langu age form. particularly if a high level of
acc uracy in th e: seco nd o r fore ign lang uage is o ne of th eir goa ls.
In this chap ter we have exam ined the role of cor rective feedbac k prim ar ily
in terms of learners' oral producti on. Most of th is research has been co n­
cerne d with the effects of correc tive feed back in relation to learners' linguistic
growth an d mo re spec ifically, learn ers' grammatical de velopment. There is
also an extensive body of research that has investigated the role of corrective
feed bac k on learners' written production. This research has been primarily
co ncern ed with whether corrective feedback can help learners improve their
writing pe rfo rmance. Less attention has been given to whether and h ow
written corrective feedback co nt ribu tes to learners' linguistic development.
O ne recent study which has done th at is Younghee Sheen's research described
in Study 5 below.
Study 5: Oral and written correctivefeedback
Younghee Sheen (20 10) compared the effects of two typ es oforal and written
co rrective feed back on ad ult ESL learners' accurate use of articles. The oral
co rrective feedback consisted of recasts or metal inguistic information; par­
allel written corrective type s were d irect co rrection or direct rneralinguis ric
feedback. Five gro ups participated in the study, of which one was a control
gro up. The other groups each received one of the fo llowing: 1) oral recasts,
2) oral m eralinguisric feedback, 3) written dire ct co rrectio n. and 4) written
d irect meralingu isric feedback.
All groups participated in two 30-minute activities in which they were asked
[Q read a Sto ry and then retell it either in the written or the oral mode. Learners
in the oral corrective feed back group received either recasts or metalinguistic
Observing IMming and teaching in the second language classroom 14"5
feedback every time they made an article error while they retold the story,
Learners wh o wro te the sto ry received the ir narr ati ves with co rrectio ns for
article errors twO days later. To determ ine whethe r learners improved thei r
ab ility to use ar ticles co rrectly, their knowledge of articles was tested before
the instructional treatment, immediatel y after th e treatment, and again 3-4
wee ks later. The tests incl ud ed a spee de d di ctati on , a wri tte n narr ati ve, and an
erro r-co rrectio n task . All groups, except fo r the one th at received o ral recasts,
significan tly outperformed the control gro up o n all im med iate and dela yed
post-testS. Sheen lnterprers these findings as evidence that the medium (o ral
versus written) in w h ich co rrective feedback is provided is less important
than the expli citness of the co rrective feedback (recasts versus meralinguistic
feedback).
Questions in the classroom
Teac hers' qu estioning behavio ur has been the focu s of a good deal of research
in second lan gu age classrooms. Q uest io ns are fund am ental in engaging st u­
dents in interaction and in exp lor ing how much they understand. Two types
or questions that have been extensively examined are referred ro as 'disp lay'
questio ns (to wh ich the teac her already knows the answe r) and 'ge nui ne'
or ' referent ial' questions (to whic h the teacher may not know the answer).
The role they play in classroom interactio n has been exam ined in a num ber
of stu dies.
Study 6: Teachers' questions in ESL classrooms
Michael Long and Charl en e Saro (1983) examined the forms an d fun ctio ns
o f questions asked by teachers in ESL classrooms and com pared th em with
ques tio ning behaviours observed o utside the classro om berween nati ve and
non-native speake rs. Th ey were part icul arly interest ed in differ en ces between
the quantity of 'd isp lay' and ' info rmatio n' ('gen u ine' or 'referent ial') ques­
tio ns. Audio. reco rdings made of the int eractions between reachers and
stu dent s in six adult ESL classes revealed th at teach ers asked more d ispla y
qu estions th an informatio n quest ions. In the native speaker/non-nati ve
speaker con versati ons o utside rhe classroo m, referen tial questions were mor e
frequent than di spla y questions . The researchers co ncluded that teacher­
learner interaction is a 'grea tly distorted version o f its equivalent in the real
wo rld' (p. 28 4), and they argued that the in reractional struc tu re ofclassroom
co nversatio n sho uld be ch anged .
Even tho ugh language teaching methods have chan ged since th e Lon g an d
Sa ro study, o the r classroom stud ies on teachers' questioning behaviour have
also reported disproportionately higher numbers o f display questions. In the
co nte xt o f co m m un icative language teac hing, teachers have bee n urged to
use fewer d isplay questions becau se they are th ough t to lead to short, sim ple
responses th at require linle cognitive effort on the pa rt of the learner. Instead,
146 Observing learn ing and teaching in the second language classroom Observing learning an d teaching in the second language classroom 147

they have been encouraged ro ask more referential (or genuine) questions McCormick and D on at o suggest th at q uestions should be examine d wi thi n
since th e latter are thought to require more cognitive processing and to gen­ th e fram ewo rk of scaffo lded interactio n and with reference to the teacher's
erate more complex answers. goals in a particular lesson o r interaction .
More recently, however, a re-evaluation of display questions has taken place.
This is based on the observation that there are different ways in which dis play
Study 8: Open and closed questions
Ano ther distinction similar to th e one between disp lay and gen uine ques­
questions can be asked in classrooms, One is for the ~eac her ro ask a series o f
tions is that between open and closed q uestions. Clo sed q uestio ns typ ically
q uestions in a drill-like format such as 'D o you have a bro ther?' , ' D oes he have
a brother?', 'Do you have a sister?' , 'Does she have a sister?' In th is context, have only one possible answer an d they usually lead to simple one-word
responses, making them quick and easy to respond to. Open questions have
display questions do no t have a meaningful or communicative purpose. In
other COntexts , howe ver, display questions can serve important pedagogic mo re than on e possible answer and invite elaboration, typically leading to
longer and more complex answers, including, for example, exp lanation and
and interaction functions . The st udy below describes teachers' use of display
q uestio ns in a more positive light. reason ing. In content and language-integrated learning (CLIL) classes in
Austria, Christiane D alto n-Puffer (2006) observed and audio recorded the
Study 7: Scaffolding and display and referential questions types ofquestions asked by English teachers, as well as the responses stud ents
In a case study of one teacher's adult ESL class, -Daw n M cCormick and gave to the m . Stude nts produced a greater q uantity and q uality of outpu t
Richard Donato (2000) explored how th e teacher's questions were lin ked to after open questions. In addition, open questions that asked lear ners not JUSt
her instructional goals. Working within sociocultural theory, the researchers fo r facts but also for reasons o r exp lan ations led to th e most complex lin­
chose the concept o f scaffolding to in vestigate teacher questions as 'med ia­ guistic outcomes. D alton-Puffer co ncluded that asking more complex open
tional tools within the dialogue between the teacher and students' (p. 184) . questions would benefit learners in these CLIL classrooms but that thi s level
As we saw in Chap ter 1 and Chapter 4 , scaffolding refers to a process in of question/response int eraction requires a h igh level of com petence in the
which, for example, a more knowledgeable (or expert) speaker helps a less foreign language on the parr of the teacher.
kn owledgeable (or novice) learner by providing an interactional framework
that the learner can build on . Study 9: wait time and teachers' questioningpractices
Another aspect ofteachers' questioning behaviour is 'wait rim e'- the amount
McCormick an d Donaro identified six functions ofscaffolding (for example, of rime the teacher pauses after having asked a q uestion to give the student
drawing the novice's attention to the task , and simp lifying or limiting the time to resp ond. Jo anna White and Patsy Lighrbown (1984) did a quan­
task demands). Th e researchers examined another function-the teacher's tita tive an alysis o f wai t tim e in au d iolingual ESL classes. They fo und that
use of questions du ring scaffolded interactions-and how it contributed to teachers typicall y gave students no more than a second o r two before th ey
class participation and lear ner comprehen sion. In the example belo w, th ey d irected the question to another student or answered th e q uestion them ­
arg ue th at th e teacher's use of the disp lay question 'Who usually lives in selves. The y also tend ed to repeat or paraphrase the qu estion several times
palaces?' serves an important pedagogic fu nction because it draws the lear n­ rather than silen tly wai t for the stu dent to form ulate a response. Al tho ugh
ers' attention to the word 'palace' through the display question and facilitates such rapid q uestion/answer patterns are typical of audiolingual classes, the y
the learners' comprehension of the word. also occur in communicative instruction. Finding a balance between plac ing
T Palace?
roo m uch pressure on students to respond q uickly and crea ting awkward
51 Like castle?
silenc es seems to be a real chal lenge.
5 2 Special place, very good.
Research has sho wn that when teachers are tra ined to give thei r students more
5 3 Very nice.
time to respond to questions, not only d o students produce more responses
T Castle, special place , very nice . Who usuall y lives in pala ces?
but their responses are also longer and more complex. N ot surprisingly, th is
ss Kings.
effect has been observed to be stronger with open/ referential ques tions com­
T Kings, and queens, princes and princesses.
pared with closed/display qu esti ons (Lo ng er al. 1985 ).
ss Yeah

5 4 Maybe beautiful hous e?

In classroo ms with stu dents at diffe rent age levels and in different kinds of
T Big, beautiful ho use, yeah, really big.
inst ruction, find ing the righ t bala nce can lead to stu dents provi di ng fuller
answers, expanding their ideas , and more successfully processing the material
to be learn ed .
148 Obst!rving learning and tCtlchmg in the second !..znguage classroom
Study 10: Time for learning languages in school
Earlier in thi s cha p ter we calked about the differences betwe en learning a
seco nd language in the natural setting compared with the classroom (see
Table 5.1). One of the major differences is the amount of time available for
learning. In the natural serring there is ample time to learn w hereas in th e
typical classroo m setting, learners have limited time. One o f th e ways to
pro vide more time for learning a second/ fo reign language isvia co nt ent- based
instruc tion . H owever, thi s is nor always feasible or desirable (Ligh rbow n
2 0 12). Other alt ernatives include increasing the total instructional time o r
distributing time more intensively over the school year.
We know that it takes a great deal of time to learn a seco nd language, but
littl e resear ch has been done ro investigate how the distribution of instruc­
tio nal time affects L2 learning. Exceptions to this include th e wo rk of some
C anad ian resear chers wh o have examined different amounts and distribu­
rions o f rim e in English and Fren ch as a seco nd language programmes. In one
stu dy in Quebec, learners receiving intensive ESL instruction for five hours
eve ry day fo r five monrhs ofo ne schoo l year (in Grade 5 or 6) we re co mpared
to learners at the end of secondarv schoo l who h ad received the same total
amou nt of instruct ion sp read ove r 7-8 years of schooli ng. O n a nu mber o f
meas ures, the st ude nts who received th e intensive inst ruc tio n pe rfo rmed as
well as o r better rhan rhose whose instruction was deli vered in wha t has been
called a 'd rip feed ' ap proac h (Ligh rbow n and Spada 1994).
In subseq ue nt research , co m pariso ns were made between groups o f Grade 5
and 6 stude nts who participated in int ensi ve English language instruction
during a single schoo l year, but with the rime distributed differently: some
stude n ts received five hours o f English a da y for five months; others received
th e same total number of hours, doing two and a half hours o f Engli sh eac h
day for 10 months. Th e researchers found that both groups benefited from
th e overall increase in hours o f instruction with some additional advantages
for learners receivin g the more int ensive instruction (Co llins et al. 1999 ;
Collins and White 2 0 1 1). The advantages were evident not only in su perio r
language abilities but also in attitudes toward the language and satisfaction
with language learning experiences. Similar findings have been reported
for different models of intensive and core French programmes (N erren and
Ge rmain 2004; Lapkin, Hart, and Harley 1998).
The classroom observation stu d ies we have described in this chapter focus
o n specific features of classroom interaction. In these studies, the feature of
int erest was determined in advance of the observation on the basis of some
hypothesis about what aspects of classroom instruction and interaction are
important for learning, thar is, whether a particular type of corrective feed­
bac k led ro more learner repair, whether a particular type of qu estion led
to more learner OUtpUt, and how the distribution of time affected learning
Observing learning and teaching in the second language classroom 1-*9
outco mes. We now tu rn to a d ifferent a pproach ro describing and interp ret­
ing instructional sett ings for seco nd language learning.
Ethnography
Eth nograp hy is a way ofobserv ing teac hi ng and learni ng in seco nd or foreign
language classrooms without a set of predetermined categories. Instead, the
observer takes ext ensive notes of th e acti vities, practices, and interactions and
looks for the patterns that emerge. This approach to classroom observation
is sim ilar to the way in which an anthropologist tak es field notes in stu dy ­
ing a group of people in their natural surroundings. In doing ethnographic
research, the obs erver can either be a participant in th e classroom activities,
for example, as a teach er's aid , o r as a non-particip ant, some o ne who sits
quietly and unobtrusively in the background, observing and recording.
Ethnographic approach es to understanding teaching and learning involve
qualitative Stu d ies that are mu ch broader in sco pe th an th e stu d ies usin g
ob serva tio n sche mes descr ibed above. Th at is. et hnog rap hies in seco nd or
foreign language classroo ms do not focus so lely on learni ng or on teac hi ng
bu t also on soci al, cultu ral, and po lit ical realities and their imp act on learn ers'
cog nitive, linguist ic. an d social develop men t. Fo r example, Ma rtha Crago's
(1992) language sociali zation resea rch with Inui t ch ild ren led her ro argue
that if children come from a culture in whi ch silence is a resp ectful and effec­
tive way to learn from an adult, th eir second language instructor needs to
know this so that th e ch ild ren's behaviour is not misinterpreted as refusal to
participate o r inability to co m p rehend.
Here are summaries of three ethnographies carr ied our in seco nd and foreign
language classro oms: one in th e Sourh Pacific, one in C anad a, and on e in
Europe.
Study 11: Language in the home and school
Karen \'V'atson-Gegeo ( 1992) ca rried Out a longitudinal study over several
years with nine fam ilies in the Solomon Islands. She explored language-use
practices in th e home and in the schoo!. Observations in the homes revealed
en vironments that were rich and sti m ulating for both linguistic and cogn itive
de veloprnent. Nevertheless, a large number of the children failed in schoo!.
A detailed analysis uncovered many differences in language use and values
between the home and school setting. There was no use of the children's first
language in sch oo!. Their first language was replaced with a restricted and
o ften incorrect version of English . Although th ese language issues were con ­
trib uting factors to the children's failure, a broader analysis of the social and
cultu ral context revealed other, more influential factors at p lay. Evidently,
parr of the children's lan guage socialization experie nce at home included
parents negatively portraying their own experiences at school, expressing
150 Observing !earning and teaching in the second Language classroom
fear s abour th eir children's ability to succeed and raisin g fundamental q ues­
tions about the value of school in th eir lives. The researcher concludes m at
these factors were cen tral in contributing to the chi ldren's lack of continued
cogn itive and linguistic development in school.
Study 12: Separation ofsecond language learners in primary schools
In a lo ngi tudinal study, Kelleen Toohey (2000) observed a group o f chil ­
dren aged 5-7 in kin dergart en , G rade 1, and G rade 2 in Vancouver, Canada.
The gro up included children who were native speakers of English, as well as
ch ildren w hose home language was Cantonese, Hindi, Polish, Punjabi, or
Tagalog. All the children were in me same class, and English was th e medium
ofinstruction. Toohey identified three classroom practices that led to the sep­
aration of th e ESL children. First, the ESL children's desks were placed close
to the teacher's desk, on me assumption mat they needed more d irect help
from the teacher. Some of them were also removed from the classroom twice
a week to obtain assistance from an ESL teacher. Second. instances in wh ich
the ESL learners interacted more with each other usually involved borrow­
ing or len ding materials bur th is had to be done surreptitiously because me
teacher did no t always tolerate it. Fin ally, there was a ' rule' in me classroom
that children should not cop y one another's oral or written productions.
This was particularly problematic for me ESL children because repeating
the words of others was often the only way in which they could participate
in conversational interaction. According to Toohey, these classroom pr ac­
tices led to the exclusion of ESL students from activities and associations in
school and also in th e broader community in which they were new mem bers.
Fu rthe rmore, such practices did not co n tribu te po sitively to the ch ild ren's
ESL development.
Study 13: Sociopoliticalchangeandforeign languageclassroom discourse
In an erhnographic study of English-medium content classes in Hungarian
secondary schools, Patricia Duff (1995) examined me im pact ofsociopolitical
chan ges on pedagogical practice. She compared me structure and participa­
tion pa nerns of two classroom activities. One is a traditional activity called
a jefe!iJ which is a heavily ritualized recitation format closely associated with
Soviet-oriented pol icies mat were rejected after me fall of communism in me
late 1980s. As a result, in many English-medium classes in Hungary, me je!elis
was replaced by a more open-ended act ivity called 'stu dent lecture' in which
students prepared and presented material to the class in a less ritualized way.
In an examination of the kind of language produced by students when par­
ticipati ng in student lectures, Duff ob served a large number of spontaneous
co mm ent s and questions produced in English ramer than Hungarian. She
also noted how students appeared to incorporate feedback provided by the
teacher (and other students) in their subsequent production, how the teacher
and students worked together to negotiate meaning and form , and how th ey
Observing learning and teaching in the second Language classroom 15 1
developed th eir flue ncy, accuracy, an d co mprehe nsion skills in me p rocess.
On me basis of th ese findin gs, Duff co ncluded mat sociopo litical transfo r­
mation affects classro om practi ce and ultimately second language learning.
Summary
In this chapter we have reviewed some of me ways in which different features
of second language instruction can be described and interpreted. We h ave
presented descriptions and examples of how classroo ms differ in ter ms of
their overall instructional focus and provided examples of different ways in
which classroom observation has been carri ed out. We have included sum ­
maries o f Stud ies examining specific pedagogical features (i.e. corrective
feedback and question types ) as well as those examining the broader social ,
cultural, and political context and its relationship to second o r foreign lan­
guage learning.
We have also provided examples of ch arts and taxonomies tha t can be used
ro observe and desc rib e di fferent aspects of th e int eractio ns th at take place
between teachers and students in the classro om setting. We en courage you
to try om one of these too ls to observe and describe the interactions in yo ur
own classroom or mat of a colle ague. The information you gain from engag­
ing in this small-scale research may help you make decisions about your own
pedagogical practic e.
In the next chapter, we will examine different views abo ut how languages
are best lear ned in classroom sertings and examine some research relevant to
these positions.
Questions for reflection
Do you think you may have a preference for using a parti cular type of
corrective feedback in your teaching? For example,would you consider
yourself to be a 'recaster' or a 'prompter' or do you think you use a range
of different feedback types? What could you do to find out?
2 Are you teaching in multilingual classrooms where many of your students
have a first language other than the language of inst ruct ion? Do you group
these students together to respond to their needs or do you integrate
them with the rest of the class! What do you think are the advantages/
disadvantages of either decision!
3 The charts that you used to analyse the teacher-student talk and student­
student talk in this chapter include six features (for example, errors,
genuine questions, negotiation for meaning). Do you th ink there are other
features that should be included in this chart?You can base your decision
on the transcripts provided in this chapter or on your own teaching
experience.
1)-1~ Observ ing learning and teaching in the second Lmguage classroom

Suggestions for further reading

Sp ad a, N. and M . Frohlich. 1995 . The Communicative Orientation SECON D LANGUAGE
0/Language Teaching Observation Scheme: Coding Conventions and
Applications. Syd ney : M acmillan. LEARNING IN THE
This book describes the origins and purposes of th e Commun icative
O rie nta tio n of Language Teaching (C O LT ) observation scheme. COLT
CLASSROOM
desc ri bes th e pedagogical pract ices and verbal interactions thar take place
between teachers and stude nts w ith th e ai m of d escribing the ins t ruc tio n
as being mo re or less communicatively o rie nre d . It also provides detailed
guidelines and illustrations for using COLT including how to collect data.
ho w to code t he data, and how to analyse ir. Also included are descrip­
tio ns of how COLT has bee n used in seco nd/fo reig n language classrooms
th ro ugho u t th e wo rld.
Preview
To o h ey, K. 20 00. Learni ng English at School: Identity, Social Relations and
In th is c ha p te r, we exam ine six proposals fo r seco nd and foreign lan guage
Classroom Practice. C levedo n: M ulriling ual M atters. .
reach ing, pr ovide examp les from classroom int eract io n to illu strate how the
Thi s book add resses a co m mon ed ucat io nal practice in m an y co unt ries proposals ger tra nslated into classroo m p ract ice. and d iscuss research fin d­
in the world in which child ren fro m minori ry langu age backgrounds are ings t hat help to assess their effectiven ess. Th e label s we have given these
ta ught in mainstream English-med iu m classes. Throug h a lo ng itu d inal
prop osals are:
desc riprio n o f a group of chi ld ren learning En glish from ki ndergarten ro
rhe end o f Grade 2 in a Canad ian school, the reader is introduced ro a 1 Ge t it righr from th e beg in n ing
ran ge of social and crirical perspecti ves on ed uca tio n a nd how th ey can be 2 Just listen . . . and read
applied to ch ild second language learning. The bo ok provides importan t 3 Lee's talk
ins ight s and useful guida nce ab out how teachers a nd schoo ls can suPPOrt 4 G e t rwo for o ne
mi no rity lan gu age child ren in their efforts to become educared through 5 Teach wh ar is reachabl e
the m edium of En glish. 6 Ger ir righ r in th e end
W ajnryb , R . 1992 . Classroom O bservation Tasks:A Resource Book j or Language
Teachers and Trainers. Cam brid ge: C ambridge Uni versity Press. Proposals for teaching
This book is m ainl y addressed to reach ers to illustrat e how rh ey can use
Many th eories have been proposed for the best way co learn a seco nd lan ­
o bserva rio n to learn ab our th eir own teaching. Readers are guid ed rh rough
guage in the classroom . Even m o re teach in g methods and materials have
a variety of d ifferent rasks an d shown how to observe, analyse. an d refiect
be en developed co implement these theories . But the o nly way co an swer rhe
o n vario us aspecrs ofi nsrruc rio n including learners. language. lessons, and
question 'What is the best way to promote language learning in classro oms?'
reaching stra teg ies.
is through research that specifically investigates relarionships between reach­
ing and learning.
To assess proposals for classroom practice, we need to use a range of research
approaches, from large -scale quantitative co in-depchqualirative stu d ies. As
we saw in Chapter 5. quan ritarive research may be essent ially descriptive.
However. ir may also be experimental, in volving careful co n tro l of th e vari­
ables that m ay influence learning. The goal of experimental studies is usually
co identify specific variables that may affe ct learning sim ilarly in different
env iro nm ents and find ways of measuring these effects. These studies often
154 Second language learningin the classroom
invo lve large numbers of learners in an effo rt to avoid th e possibility th at
the unusual behaviour of one or two individuals might lead to a mis leading
conclusion about learn ers in general.
Qualitative research , incl udi ng ethnogra ph ies and case sru dies , ofte n involves
small n umb ers, perhaps one class or on ly one or rwo lear ners in that class.
The emphasis is not o n wha t is most general but rather on a th o rough under­
stan d ing of what is particul ar about what is happen in g in this classroom. As
pointed o u t by Ann Burns (20 10) and others, while quantitative and qualita­
tive research are important in assessing theoretical proposals, action research
carried our by teachers in th eir own classrooms is also essential to answer spe­
cific local questions. In th is chapter we focus mainly on experimental studies .
Th ese are studies that were designed to test hyp otheses abo ur how teaching
affects second lan guage learn ing. Readers are enco uraged to follow up with
further reading but also to exp lore related q uestions through research activi­
ties wit hin thei r own teaching and learning en viro nments.
1 Get it rightfrom the beginning
'Get it right from the begin ning' is probably the proposal that character­
izes more second and foreign language instr uctio n than any o ther kind.
Although communicative lang uage teaching has come to do minate in some
environments, the srrucrure-based approaches discussed in Chap ter 5, espe­
cially grammar translation, rem ain Widespread.
Th e gra mmar translat ion approach has its origi n in the teac hing of classi­
cal languages (fo r example, Greek an d Latin) . Students were presented with
vocabulary lists, often acco m panied by translat ion eq uivale n ts, and grammar
rules . The original purpose of th is approach was to help stu dents read lit­
era ture rather than to develop fluency in the spo ken lan guage. It was also
tho ught that thi s approach provided students with good m ental exercise to
help develop their in tellect ual and academic abilities .
In a typical grammar translation activity, student s read a text toge ther line
by line and are asked to translate it from the targe t language in to their native
lan guage . Stu dent s may answer comprehension questio ns based on the
passage, often in their first lan gu age. The teacher draws attention to a specific
grammar rule that is illustrated by the text (fo r example, a cer tain verb form).
Following th is, the students are given an exercise in which they are asked
to practise th e grammar rule, for exam ple, by fillin g in the blanks with th e
appropriate verb form in a series of decontextualized sentences th at mayo r
may no t be related to the text they have read and translated .
Audiolingual instruction arose in part as a reactio n to the grammar transla­
tio n app roac h. The argument was that, un like grammar translat io n teachin g,
in wh ich stu dents learn ed abo ut the lan guage , aud iol in gual teach ing would
Second language learning in the classroom 155
lead students to actually speak the language (Brooks 196 0: Lado 1964).
In C ha p ter 4, we saw that the audiolingual approach was based on beh av­
iou rism and contrastive anal ysis. The examples below reflect au d iolingual
teaching. It is evide nt that, even though the emphasis is on the or al language,
students rarely use the language spo nt aneously. Teachers avoid letting begin­
ning lear ne rs speak freely o ut ofconcerns that this wo uld allow them to mak e
errors . The errors, it is said, co uld become habits. So it is bett er to prevent
th ese bad habits before they happen and 'Get it right from th e beginn ing.'
Example 1
(A group of 15-year -o ld students invo lved in an exercise based on the sim ple
p resent of Eng lish verbs.)
S1 And uh, in the afternoon, uh, I come home and uh, uh , I uh,
washing my dog.
T I wash.
SI My dog.
T Every day you wash yo ur dog?
SI No . [ben]
S2 II n'a pas de chien! (= H e doesn't have a dog!)
s INon, mais on peut le d ire! (= No , but we can say we dol)
(Unpublished data fro m P. M. Lightbown and B. Barkman)
C learly, in this case, the student's real experience with his do g (or even th e
fact that he did or did not have a dog) was irrelevant. What mattered was the
correct use of the simple present verb!
&ample 2
(A grou p o f 12-year-old learners of English as a forei gn language.)
T Repeat after me . Is there an y butter in th e refrigerator?
CLAS S Is there an y butter in the refrigerator ?
T There's very little, Mom.
CLASS There's very little, Mom.
T Are th ere any to matoes in the refrigerator?
CLAS S Are there any tomatoes in the refrige rator?
T There are very few, Mom.
CLASS There are very few, Mom.
(Unpublished data fro m P. M . Lightbown and B. Barkman)
Pure repetition: The students have no reaso n to get involved or to th ink
abo ut what they are saying. Indeed, some students who have no idea what
the sentences mean will successfully repea t them anyway, while th eir minds
wander off to other things.
156 Second fanguilge It!aming in the classroom
97~/J ~/I/(\\)
~OJ
./
."'E!a>I:'S
o ON Tlfe J>e'Sl;. II. (, I(
(R~
\)/~ ~
(~~?-
Research findings
Many adulr learners, especially those with good metalinguistic knowledge
of their own language, express a preference for structure-based approaches.
Learners whose previous language learning experience was in grammar
translation classes may also prefer such instruction. As we saw in Chapter 3,
learners' beliefs about the kind of instruction that is best can influence their
satisfaction and success. The grammar translation approach is useful for the
study of grammar and vocabulary and can be valuable for understanding
important cultural texts. The audiolingual approach with its emphasis on
speaki ng and listening was used successfully with highly motivated adult
learners in intensive training programmes for government personnel in the
United States . However, there is little classroom research ro suPPOrt such
approaches for students in ordinary school programmes that must serve the
needs of students who bring different levels of motivation and aptitude co
the classroom. In fact, it was the frequent failure of traditional grammar
translation and audiolingual methods to produce fluency and accuracy in
second language learners that led co the development of more communica­
tive approaches co teaching in the first place.
Supporters of communicative language teaching have argued that language
is not learned by the gradual accumulation of one grammatical feature after
another. They suggest that errors are a natural and valuable part of the lan­
guage learn ing process. Furthermore, they believe that the motivation of
learners is often stifled by an ins istence on correctness in the earliest stages
Second languagt! It!armng in the classroom 15­
of seco nd lan guage learning. Th ese oppo nents o f th e 'G et it right fro m the
beginning' proposal argue that it is better to encourage learners co de velop
' fluency' before 'accuracy'.
Some researchers and educators have reacted to the version of communica­
tive language teaching that advocates an exclusive focus on meaning. They
argue that allowing learners too much 'freedo m' without correction and
explicit instruction will lead co early fossilization of errors. Once again we
hear the call for making sure that learners 'G et it right from the beginning'.
Unfortunately, it is difficult co test whether an emphasis on form in the earl y
stages of second language learning will, in the long run, lead co better results
than those achieved when the emphasis is on meaning in the early stages. To
test that hypothesis, it would be necessary to compare groups that are similar
in all respects except for the rype of instruction they receive. However, it is
not easy for researchers co find proper comparison groups. This is because
there are many parts of the world where one finds predominantly structure­
based approaches co language teaching, and in the se settings there arc no (o r
very few) classrooms wh ere the teaching places an emphasis o n.meani ng. O n
th e o ther hand , with th e widespread ado ptio n of communicati ve lan gua ge
teac hi ng in o the r parts of th e wo rld , it is difficult CO find classrooms th at
are exclusively struc tu re-based . No ne theless, so m e findi ngs from second lan ­
guage classro om research do permit us co assess the effects of instructi on th at
is strongly oriented co the 'G et it right from the beginning' approach . These
include descriptive studies of the interlanguage development of second lan­
guage learners in audiolingual programmes (Study 14), and comparisons of
the development ofsecond language proficiency between groups ofstudents
receiving different combinations of form- and meaning-based instruction
(Stu dy 15).
Study 14: Audiolingualpattern drill
In the late 1970s , Patsy Lightbown ( 19 83a / b) car ried OUt a series of lon­
gitudinal and cross -sectional investigations into the effecr of audiolingual
instruction on interlanguage development. The investigations focused on
French-speaking learners aged 11-16 in Quebec, Canada. Students in these
programmes typically participated in the rypes of rote repetitio n and pattern
practice drill we saw in Examples 1 and 2.
The learners' acquisition of certain English grammatical morphemes (for
example, plural -r and the progressive-ing) was compared with the acquisition
of grammatical morphemes observed in the interlanguage of uninstructed
second language learners (see Chapter 2 , p. 46). The results showed differ­
ences between the developmental sequences we saw there and the relative
accuracy with which these classroom learners produced them. These findings
suggested that the rype of instruction students had experienced-isolated
158 Second language learn ing in the classroom
parrern practice drills-resulred in a developmental sequence that appeared
co be different from that oflearners in more natural learning env ironm en ts.
For a time after their instruction had focused on it, learners reliably pro­
d uced a particular grammatical morpheme in irs obligacory contexts. For
example, after weeks ofdrilling on present progressive, students usually sup­
plied both th e auxiliary be and the -ing ending (for example, 'He's playing
ball'). However, they also produced one or more of me morphemes in places
where they did not belong CHe's wan t a cookie') . The same forms were pro­
duced with considerably less accuracy in ob ligatory contexts when they were
no lo nger being practised in class and when th e third person singular simple
present -s was being drilled instead. At this point, many students appeared
co revert co what looked like a developmentally earlier stage, using no tense
marking at all (for example, 'He play ball') . These findings provided evi­
dence mat an almost exclusive focus on accuracy and practice of particular
grammatical forms does not mean th at learners will be able to use the forms
correctly outside the classroom drill setting, nor that they will continue to
use them correctly once other forms are introduced. Not surprisingly, this
instruction, mat depended o n repetition and drill of decontexrual ized sen­
tences, did not seem co favour the development of comprehension, fluency,
or communicative abilities either,
Study 15: Grammarplus communicativepractice
In one of the earliest experimental studies of communicative language teach­
ing, Sandra Savignon (1972) studied me linguistic and communicative skills
of 48 college students enrolled in French language courses at an American
university. The students were divided into three groups: a 'communicative'
group, a 'cu ltu re' group, and a control group. NI groups received about fo ur
hours per week of audiolingual instruction where me focus was on me prac­
tice an d manipulation of grammatical forms. In addition, each group had a
special. hour of different activities. The 'com m unica tive' group had one hour
per week devo ted to communicative tasks in an effort to encourage practice
in using French in meaningful, creative, and spontaneous ways. The 'culture'
group had an hour devoted to activities, conducted in English, designed to
'foster an awareness of me Fre nch language and culture through films, music,
and art' . The control group had an hour in the language laboratory doing
grammar and pronunciation drills similar co those they did in their regular
class periods.
Tests . to measure learners' linguistic and communicative abili ties were
administered before and after instruction. The rests of linguistic competence
included a variety of grammar rests, teachers' eval uations of speaking skills ,
and course grades. The tests of communicative competence included meas ­
ures of fluency and of th e ability to understand and transmit information in
a variety of tasks, which included: discussion with a native speaker of French,
Second language learn ing in the classroom 159
interviewing a na tive speak er of French, reporting facts about oneself or one's
recent activities, and describing ongoing acti vitie s in French.
At the end of me period of instruction, there were no significant differences
between groups on the linguistic competence measures. However, th e com­
m unicative group scored significantly higher than me other twO gro ups on
the four communicative tests developed for the study. Savig no n in terp reted
these results as support for the argument mar second language programmes
that focus only on accuracy and form do not give students sufficient oppor­
tunity to develop communication abilities in a second language. Even more
important in the context of me 'G et it right from the beginning' approach
was the evidence that opportunities for freer communication did not cause
learners to do less well on measures of linguistic accuracy.
Interpreting the research
The studies reviewed above provide evidence to support me intuitions of
teach ers and learners that instruction based on the ' G er it right from' me
begin n in g' proposal has important limitations. Learners receiving audiolin­
gual or grammar-translation instruction are often un able to com m unicate
their messages and intentions effectively in a second lang uage. Experience
has also shown mat pr imarily or exclusively structure-based approaches to
teaching do not guarantee that learners develop high levels of accuracy and
linguistic knowledge. In fact, it is often very d ifficult to determine what stu­
dents know about me target language. The classroom emphasis on accuracy
often leads learners to feel inhibited and reluctant CO take chances in using
their knowledge for communication. The results from these studies provide
evidence that lea rners benefit from opportunities for communicative practice
in contexts where the emphasis is on understanding and expressing meaning.
It is important to em p hasize that in the Savignon study, all students contin­
ued to receive their regular, grammar-focused instruction. They differed only
in terms of me ptesence or absence ofan additional communicative practice
component. In other words, this study offers sup po rt for me argument that
meaning-based instruction is advantageous, not that for m-based instruction
is no t. The contributions of communicative practice and grammar-focused
instruction will be discussed in more detail in relation to the 'G et it right in
the end' proposal.
2 Just listen . . . and read
'JUSt listen . . . and read ' is based on the hypothesis that language acq uisition
takes place when learners are exposed co comprehensible input through lis­
tening and/or reading. As noted in Chapter 4, the individual whose name is
m ost closely associated with th is proposal is Stephen Krashen (1985, 1989).
This is a co ntroversial proposal because it suggests th at second language

160 Second language lea rning in the classroom
learners do nor need to produce lan guage in order to learn it. except perhaps
to get other people to provide inp ut by speaking to th em . According to this
view, it is eno ugh to hear (o r read) and underst and th e target language.
Read Example 3 to get a feel for how thi s the o ry o f classroom seco nd lan gu age
learning can be implemented in a classroo m . Th is descri p tion shows that
one way to obtain comprehensible input is to provide learners with listening
an d reading co m prehe nsion activities w ith no (or very few) op po rt un ities to
speak or int eract with th e teache r or other learne rs in th e classroom.
Example 3
It is time for English class at a pri m ary school in a French-speaki ng co m­
munity in New Brunsw ick. Canada . Th e classroom looks like a m in iatu re
language lab, with about thirty small desks, on each of which there is a cas­
sette player and a set oflarge ear ph ones . Aro und the roo m, shelves and racks
display scores of books. Each book is packaged with an audiocassette that
contains a reco rd ing of its content . The materi als are no t str ictly graded, bu t
some sets of books are very simple. and othe r sets are grouped so that they
are gradually more challenging. There are pre-school child ren's books with
a picture and a word or rwo on each page; illustrated stories with a few sen­
tences per page; picture dictionaries; ESL textbooks for child ren; illustrated
science books about animals, weather, vehicles, etc . Students (aged 8-10)
enter the classroom , select the material they want. and take it to their indi­
vidual works pace. They inse rt the cassette , put on th eir ear pho nes. an d open
their books . They hear and read English for the next 3 0 mi n utes. For so me of
the time th e teac he r walks aro und th e classroom, checking th at the machines
are runn ing smoothly, bu t she d oes not interact with th e stu de nts co ncern­
ing wha t they are do ing. Some of the stude nts are liste n ing wi th closed eyes;
others read actively, mouthing the wo rds silen tly as they follow each line with
a finge r. The classroom is alm ost sile nt except for th e so und of tap es being
inserted and rem oved o r chairs scraping as students go to the shelves to select
new tapes and boo ks.
Research findings
Research relevant to th e 'J USt listen . .. an d read' pro posal includes stu d ies of
co mp rehensio n- based teach ing and extens ive reading (Day et al. 20 1 1). We
will also look at some comprehension-based instruction in wh ich th e input
is ma nip ulated in ways that are intended ro increase the likelih ood that stu­
dents will pay atte nti on to lan gu age for m as well as m eaning.
Study 16: Comprehension-based instruction for children
Exam ple 3 was a description of a real pro gramme implem ented in experi­
me nt al classes in a Frenc h-speaki ng region in Canada . From th e beginn ing
of the ir ESL instructi on at age eigh t, stu dents only listen ed and read during
thei r daily 30- m in ute ESL period . Th ere was no oral practi ce o r interaction
Second language learning in the classroom 161
in English at all. Teachers did not 'teach' but provided organizarional and
technical sup po rt . Thus. lear ners received nat ive-speaker input from tapes
and books but they had virtu ally no interactio n in English with the teach er or
o th er learn ers. The y guessed at meaning by usin g the pictures o r by recogniz­
ing cognate wo rds char are similar in French and English. Occasionally they
could refer to translation equivalents of a few wo rds, taped inside a book's
~kro~~ .
Patsy Light bown and her colleag ues inves tigated the second language devel­
opment of hundreds of chil dren in this comprehension-based programme
an d compare d th eir lear ning with th at o f stu de nts in me regular ESL p ro­
gramme, which was mai nly an audiolingual ap proach. All m e stu de nts in
both programmeS had classes tha t last ed 30 minutes per day. After rwo years,
learners in the comprehensio n-based programme knew as much English
as (an d in some cases mo re than) learners in the regular pr ogramm e. Th is
was true not only for comprehension but also for speaking, even though the
learners in rhe experimental programme had never practised spoken Englis h
in their classes (Ligh rbown er al. 2002; Trofimovich et al. 2009).
The students' Engli sh language abilities were reassessed three years later, when
th ey were in Grade 8. Some students had co nt in ued in the comprehension­
o nly programme throughout that time. On comprehension measures and on
some measures of oral production, they continued to perform as well as StU­
dents in the regular programme. On other meas ures, some gro ups ofstu de nts
in the regular programme had made greate r p rogress, especial ly in wri ting .
Those students were in classes where the regular programme included not
on ly aud iolingual ins tructio n but also othe r speaking and writi ng co mpo ­
nents. teacher feed bac k, and classroom inte ractio n .
Study 17: Readingfor words
Find ing reading material for p rim ary school students learning a seco nd lan­
guage is challenging. Find ing reading material for adults in early stages of
second language acquisition is chal leng ing too, but graded readers specially
des igned for ad ult ESL learn ers are increasingly availab le. Th ese sim plified
literary classics, biogr aphies, romances, and thrillers offer int eresting and
age-a ppro p riate content, whi le the vocab ulary and writi ng sryle rem ain
sim ple. Marlise Ho rst (200 5) used sim plified read ers in a study of voca bulary
development among ad ult immigrants who were enrolled in an ESL pro­
gram me in a co m m unity cent re in M ontreal, C anada. Th e 2 1 particip ants
rep resen ted several langu age backgrounds and proficien cy levels. In ad di tion
to th e activ ities of th eir regular ESL class, students chose simplified read ers
tha t were made availab le in a class libr ary. Over a six-week period . students
roo k books home and read [he m o n th eir ow n.
H orst developed individualized voca bulary measures so mat lear ning co uld
be assessed in terms o f the books eac h student actually read . She found that
162 Second language Learning in the classroom
th ere was vocab ulary growth att rib utable ro reading, even over this short
period, and that th e more students read, the more words th ey learned. She
concluded that substantial vocabulary growth through reading is possible,
but that students must read a great deal (mo re than JUSt o ne or rwo books
per semester) to realize th ose benefits. As we saw in Chap ter 2, when we
interact in ordinary conversations, we tend to use mainly the 1,000 or 2 ,000
mo st freq uent words. Th us, reading is a particularly valuable so urce of new
vocabulary. St udent s who have reached an intermediate level of proficiency
may have few opportunities to learn new words in everyday conversation. It
is in reading a variety of texts that students are most likely to encounter new
vocabulary. The benefit of simplified readers is that students encoun ter a
reasonable number of new words. This inc reases the likelihood that they ean
figure out th e meaning of new words (or perhaps be motivated to look th em
up). If the new words occur often enough, students may remember them
whe n they encounter th em in a new context.
Other research that explores the 'Just listen '. . . and read ' proposal includes
stu dies in wh ich efforts have been made to draw seco nd langu age learn ers'
attention ro language forms in th e in put, for exam ple, by providing high­
frequency exposure to specific language features through an in p ut flood,
highlighting the fearures through enhanced input, and/or providing pro­
cessing instruction. All of these are described in more detail below but the
emphasis in all cases is on getting learners ro notice language forms in the
input, not on getting them to practi se producing the forms . The next th ree
studies are examples of th is researc h.
Study 18: Inputflood
Martha Trahey and Lydia White (1993) ca rried out a srudy with young
Frenc h-speaking learners (aged 10-12) in intensive ESL classes in Q uebec
such as those described in Chapter 5, Study 10. Th e goal of the research was
to determine whether high -frequen cy exposure to a particular form in the
instructional input would lead to better knowledge and use of that form
by the students. The linguistic form investigated was adverb place me nt in
English (see Chap ter 2, p. 58) . For approximately 10 hours over a two-week
period, learners read a series ofshort texts in whi ch they were expose d to liter­
ally hundreds of instances ofadverbs in English sentences-so ma ny that the
investigators referred to th is study as an ' in put flood'. There was no teaching
of adverb placement, nor was an y error correction provided. Instead, stu­
dents simply read the passages and completed a variety of comprehension
activities based on them.
Alt ho ugh learners benefited from th is exposure to sente nces with adverbs
in all the correct posi tions, their learning was incomplete. Th ey im p roved
in their acceptance of sentences with word order th at is grammatical in
En glish bur nor in French ('Th e children quickly leave school'). However,
Second language Learning in the classroom 163
th ey conti nued ro accep t senten ces that are gram ma tical in Frenc h bu t no t in
English ('The ch ildren leave quickly schoo l') . The stu dents' inab ility to recog­
nize th at adverbs in this position are ungram m atical in English suggests that
th e input flood could help them add something new to their interlan guage,
bu t d id nor lead them to get rid o f an error based on their first language. As
noted in Chapter 2, Lydia White (l99 1) argued tha t although exposure ro
language in put provides lear ners with positive evidence (in for matio n abo ut
what is grammatical in the second language) , it fails to give them negative
evidence (info rm atio n about what is not grammatical). Positive evidence is
not enough to permit learners to notice the absence in the target language
of elements that are present in their interlanguage (and their first language).
Thus, more explicit information abo ut what is not grammatical in the second
language may be necessary for lear ners' continued develop ment. This is d is­
cussed in more detail in the sect ion ' G et it right in the end' .
Study 19: Enhanced input
M ichael Sharwood Smith (1993) coined the term ' in put enhancem ent' to
refer to a variety of th ings that m ight draw learners' attention to featu res in
the second language, thus increasing the chan ces th at they would be lea rn ed .
In a study invol ving en hanced input, Joan na White (1998) exam ined th e
acquisition of possessive determ iners (specifically 'his' and ' her'; see Chapter
2, p. 53) by French-speaking learners aged 11-12 in inte nsive ESL classes.
Stu d en ts received approximately 10 hours ofexposure to hundreds of posses­
sive det erminers through a package of readi ng materials and comprehension
activities provided over a two-week period. The major difference between
this study and Trahey and White's in put flood is that typographical enhance­
ment was added. That is, every time a possessive determiner appeared in the
texts, it was in bold type , underlined, italicized, or written in capital lett ers.
The hypothesis was that th is would lead the learn ers to notice the possessive
determiners as th ey read the texts.
White compared the performance of learners who had read th e typographi­
cally enhanced passages wit h tha t o flearners who read the sam e texts wi thou t
enhancement. She found that both groups im proved in their kn owled ge
and use of these forms and that there was little d ifference between them. In
interpreting these find ings, White questions whether the en hancement was
sufficien tly explicit to draw the learners' attention to possessive determin­
ers. That is, even tho ugh the forms were h ighligh red by th e use of bold type ,
capital letters, etc., students d id no t learn how to choose the possessive deter­
miner to match the gen der of the possessor. In subsequent research , White
found that learn ers made more progress when they were given a sim ple rule
and then wo rked together to find the correct possessive deter miners (Spada,
Ligh rbown, and White 2005 ).
164 Second ltnguage learning in the classroom
/ tio n shows th at learners can make co nsiderable progress if they have sustained
Enhancing the input
Study 20: Processing instruction
Bill Van f'arren (2 004) and his colleag ues have investigated the effects of pro­
cessing instruct io n, ano the r app roac h to co m prehensio n- based learning. In
processing instruction, learne rs are p Ut in situa tio ns where they ca n nor co m­
p rehe nd a sen ren ee by depending so lely o n co n text . pri or kn owled ge, or other
clues. Rather they m ust focu s on the language itself. In one ofthe first studies,
ad ult learn ers ofSpanish as a foreign language received instruction on diffe r­
ent lingu istic forms , for example, object prono uns (VanPanen and Cadierno
1993) . As noted in Chapter 4, VanPatten found that English-speaking lear n­
ers ofSpa nis h tend ed to trea t the objec t pronouns, which p recede the verb in
Spanish, as if they were subject pronouns. Thus, a sentence such as La sigue
1 senor (literally 'h er (object) follows the man (su bject)') was interpre ted as
'She follows th e man'.
Two groups were co mpared in th e study, o ne receiving processing instruc­
tion, the other following a mo re traditional approach. The processing
instruction gro up received exp licit exp lanatio ns abou t obj ect p ronouns and
did some activities that drew the ir attentio n to the importan ce of noticin g
that o bject p ron ou ns could occur before th e verb. Then, th ro ugh a varie ty of
foc used liste ning an d reading exercises, learners had to pay atte n tion to h ow
the target forms were used in o rde r to understand th e mean in g. For example,
they heard or read La sigue el senor and had to choose whic h p icrure-a man
follow ing a wo ma n or a wo m an following a m an-eorresponded to the sen­
tence. A seco nd gro up of lear ners also received explicit info rmation about
the targe t forms but instead of focus ing on comp rehens ion pr acti ce through
processing instruct ion , th ey engage d in p roduct ion practice, do in g exercises
to practise the forms being taught. After the instruction. learners who had
rece ived the comprehension-based processing instructio n no t o nly did better
Second ltnguage learn ing in th« classroom 165
o n the co m p rehe nsio n tasks than learners in the p roduct ion gro up. they also
performed as well on production tasks.
Interp retin g the research
Researc h on comprehension-based approaches to second lan guage acquisi­
exposure to language they understand. The evide nce also suggestS, ho wever,
tha t co m prehensio n- based activities may best be seen as an excellent way
to beg in learning and as a supplement to other kinds of learning for more
ad van ced learners.
Comp rehension ofmeani ngful lan guage is the fou ndatio n oflanguage acqui­
sitio n. Active listening and readi ng for meaning are valuable components
of classroom tea chers' pedagogical pra ctices. Nevertheless, considerable
research and experience challenge the hypo th esis that comprehensible input
is enough. Van f'atren's research showed that forc ing students to rely on spe ­
cific ling uis tic features in order to interpret meaning increased the chances
tha t the y would be able to use th ese features in th eir own seco nd lan gua ge
p roduc tio n.
Anot her response to co mp rehension-based approaches is Merrill Swain"
(198 5) com prehensib le OUtput hypothesis. She arg ues tha t it is whe n Stu­
dents have to p rod uce lang uage tha t th ey begin to see the lim itati on s of th ei r
interlanguage (see C hap ter 4) . H owever, as we will see in the disc ussion of
th e 'Let's talk' proposal , iflearners are in situation s where thei r teachers an d
classmates unders tand them without difficulty, they may need additional
help in overcoming those limitations.
3 Let's talk
Advoca tes of the ' Let's talk' proposal em phasize th e im po rtance of access to
both co m prehensible in put an d co nve rsat ional inte ractio ns with teac hers
and othe r students. They arg ue th at when learners are given the opportunity
to engage in interaction. they are compelled to negotiate for meaning, that is,
to expres s and clarify the ir int enti o ns, tho ugh ts, opinions, erc., in a way that
permits th em to arrive at m utual understan d ing. Th is is especi ally tru e w hen
the learners are wo rki ng together to accom plis h a pa rtic ular goal, for exam ple
in task-based lan guage teachi ng (T BLT). According to the interactio n
hypothesis, negotiati o n leads learners to acqu ire the lan gu age for ms-the
words and the grammatical str uc tu res-that carry th e mea ning they are
attend ing to . This is th e theo retical view u nderlyin g the teacher- stu dent
behav io ur in the transcrip t from Class room B and from the student-student
interactio n in Co mmunicatio n task A in Chapte r 5.
166 Second language learn ing in the classroom Second language learning in the classroom 167

Negotiation for meaning is accomplished through a variety of modifications
that naturally arise in interaction, such as requests for clarification or confir­
mation, repetition wit h a questioning intonation, etc.
Look for negotiation for meani ng in the examples below and compare this
with the ex~p les given for the 'Get it right from the beginning' proposal.
Example 4
(A group of 12 -year-old ESL students are discussi ng a questionnaire about
petS with their teacher.)
5 And what is ~feed ' ?
th ese studies have been car ried o ut in labo rat o ry settin gs and are m oti vated
by Mi chael Long's ( 1996) updated versio n of the interaction hypothesis
(see Chapter 4). Compared with the original version Stating th at conver­
sational interaction promotes second language develo pment (Long 1983) ,
the updated vers ion integrates learne r capacities that co ntribute [Q second
language learning (for example, attention) and features of interactio n that
are most likely to facili tate lear ni ng. C o rrective feed back has been identified
as one feature that is believed to playa crucial role in helping learners [Q make
connections between form and meaning. In fact, as we will see later in th is
chap ter, research relevant [Q the updated interaction hypo thesis is more in

line with the 'G et it right in the end' proposal.

T Feed? To feed the dog?

5 Yes, but when I don't have a . . .

Study 21: Learners talking to learners
T If you don't have a dog, you skip the question.
In one of the early descriptive studies on learner interaction, Michael Long
and Patricia Porter (1985) examined the language produced by adult lear n­
Example 5

ers performing a tas k in pairs. There were 18 participants: 12 no n- na tive

(Studen~ from Classroom B, as they set tle in at the beginning of the day.)

speakers of English whose first language was Spanish, and six native English
T H ow are you doing this morning?
speakers. The non-native speakers were interm ed iate or advanced learn ers
s i I'm mad!
of English .
52 Why?

Each individ ual lear ner participated in separate discussions with a speaker
T Oh boy. Yeah, why?

fro m each of the three levels. For example, an intermediate-level speaker

51 Because this morning, my father say no have job this mo rning.

had a conversation with another inter mediate-level speaker, another wi th an

T Your farher has no mo re job this m o rn ing? Or you have no job?

advanced-level speaker, and ano ther with a native speaker of English. Long
5I My father.

and Porter compared the speech of native and non-native speakers in co n­

versatio ns, analysing the differences across p roficiency levels in conversatio n
How different these examples are from the essentially meaningless interac­
pairs . They found that learners talked more with other learners than they did
tion often observed in classrooms where the em phasis is on 'getti ng it right
with native speakers. Also, learners produced more talk wi th advanced-level
from th e beginning'. Such genuine exchanges of information must surely
learners than w ith intermediate-level partners, partly because th e conversa­
enhance students' motivation to participate in language learning activities.
tions with advanced learners lasted lo nger.
But do they, as advocates of this position claim, lead to successful language
Long and Porter examined the number ofgr ammatical and vocabulary errors
acquisition? Note, for example, that, although the conversation proceeded in
and false starts and found that learner speech showed no differences across
a natural way, th e student in Example 4 ne ver did find out w hat 'feed ' meant.
contexts. That is, intermediate-level learn ers did not make any more errors
with another intermediate-level speaker than they d id with an advanced or
Research findings
native speaker. This was an interesting result because it called into question
Most of the ear ly research that examined the 'Let's talk' proposal was descrip­
the argument tha t learners need to be exposed to a native-speaking model
tive in nature, focusing on such issues as: How does negotiation in classrooms
(i.e, teacher) at all times to ensure that they produce fewer errors. Overall,
differ from that observed in natural settings? How do teacher-centred and
Long and Porter concluded that although learners cannot always provide
student-centred classrooms differ in term~ of conversational in terac tio n ?
each other with the accurate grammatical input, they can offer each other
Do task types co ntribu te to different kinds of interactional modifications?
genuine communicative practice th at includes negotiation for mea ning.
Several studies also examined relarionships between modifications in conver­
Supporters of the 'Le t's talk' proposal atgue that it is precisely th is negotia­
sational interaction and comprehensio n.
tion for m ean ing that is essential for language acquisi tion.

In th e m id - 1990s researchers began to d irectl y explore the effects of inte rac­

tion on second language production and development over time. Most of

168 Second language learn ing in tbe classroom
Study 22: Learner language andproficie1lCY level
George Yule and Doris Macdonald ( [ 990) investigated whether the role
that different proficiency-level learners play in a two-way communication
task led to differences in their interactive behaviour. They set up a task that
required rwo learners to communicate information about th e location ofdif­
ferent buildi ngs o n a ma p and th e ro ute to get th ere. One learner, referr ed to
as the 'sender' , had a map with a delivery route on it , and this speaker's job
was to describe the deli very route to the 'receiver ' so that he or she could draw
the delivery route on a sim ilar map. The task was made mo re challenging by
the fact tha t there were minor differences between the two maps.
To determine whether there would be any difference in the interactions
acco rding to th e relati ve proficiency of the 40 adult participants, differ­
ent type s of learners we re paired together. O ne group had high-proficiency
learners in th e 'sender' role and low-proficiency learners in the 'receiver' role;
the other gro up had low-proficiency 'senders' paired with high-profici en cy
.receivers' .
\X'hen low-proficiency learn ers were senders, interactions were co nsiderably
longer and more varied than when high -p roficiency learners were the sen de rs.
The explanatio n for th is was tha t high -proficiency senders tended to act as
if the lower-level receiver had little contributio n to make in the co mpletio n
of the task. As a result , th e lower -level receivers were almost forced to playa
passive role and said very lirtle. When lower-level learners were the senders
however, much more negotiation for meaning and a greater variety of inrerac­
do ns between the two speakers took place. Based on these findings , Yule and
Macdonald suggest that teachers should sometimes place more advanced Stu­
dents in less dominant roles in paired activities with lower-level learners.
Study 23: The dynamics ofpair toork
In a longitudinal study with adult ESL learners in an Australian university,
N eo rny Storch (2 002) observed the patterns of int eraction between 10 pair s
of students co m p letin g different rasks over one semester. Sh e identified four
distinct parrerns of interaction: 'co llaborative' interactio n consisted of two
learners fully engaged with each other's ideas; 'domi na nt- d o m inan t' inter­
action was characterized by an unwillingness on the part of either learner
to engage and/or agree with the other's contributions; 'do m in ant- p assive'
consisted of one lear ner who was authoritarian and another who was willing
to yield to the other speaker; and 'expert-novice' interaction consisted of
o ne learner who was stronger than the other bur actively encouraged and
sup po rted the other in carrying our the task. To investigate whether the four
type s of interaction led to differences in learning outcomes, Storch iden­
tified learn ing events that occurred during the interactions (fo r example,
learning that the definite article is used with the names of some countries) .
Then she looked at whether that lan gu age knowledge was maintained in
Second language learning in the classroom [69
a su bseq ue nt task. Sto rch fo und tha t learn ers who participated in the col­
laborative and expert-novice pair s maintained more of their L2 knowledge
over time. Learners who participated in the dominant-dominant and domi­
nant-passive pairs maintained the least. Storch interp rets this as support for
Vygo tsky's theory of cogni tive development (see Chapter 4) and the claim
that when pai r work fu nctio ns collabo rat ively and learn ers are in an expert­
novice relationship, they can suc cessfully engage in the co-construction
o f knowledge.
Study 24: Interaction and second language development
Alison Mackey (1999) asked adult learners of ESL to engage in different
co m m unicative tasks, for example, story completion and picture sequen cing
with nat ive speake rs of En glish . The tasks were designed to provid e co n­
texts for learners to produce questions. Prior to participating in the tasks, the
learners were assessed in terms of the stage the y had reached in learning que s­
tions, as described in Chapter 2. Th ey were then divided into five groups.
The learners in G roup 1, referr ed to as' ' Inreracto rs.' ca rried o ut the tasks
with native speakers, w ho modified their language as they sough t to clarify
meani ng fo r the learne rs. Learne rs in Group 2 received th e same modified
input as learn ers in Gro up [ but they were not as advanced in th eir acquisi­
tio n ofq ues tio n forms and th us were referred to as the ' Lnter act o r Unreadies .'
Learn ers in Group 3 , th e 'O bservers.' were asked to listen to the learn er and
the native speaker as they carried o u t th e task but they did not interact in an y
way. Learners in Group 4, the 'Scripteds' carried out: the sam e tasks with the
native sp eakers but th e nati ve speakers used lan guage th at had been sim p li­
fied and scripted in such a detailed manner that communication breakdowns
d id not occ u r and thus no negotiation for meaning took pla ce. There was also
a co ntrol gro up of learners wh o did not particip at e in an y of the tasks but
co mp leted all the rests.
Both Inreract ors an d In reracror Un readies demonstrated more sustained
pr ogress in their quesrion formati on devel opment th an learn ers wh o did not
engage in interaction (i.e. Observers, Scripteds, and Control). Th e Observers
and the Scripteds were sim ilar to th e C ontrol group learners, wh o changed
very li trle . Mackey also notes that the significant increase in development for
the Inreractors was maintained on the delayed post-tests administered one
month after the treatment tasks. These results are interpreted as sup porr for
the hypothesis that negotiated int eraction leads to L2 development.
Study 25: Learner-learner interaction in a Thai classroom
In a study relevant to the updated version of the interaction hypothesis, Kim
McDonough (2 004) investigated the use of pair- and small-group activities
in English as a foreign language classes in Thailand. Students engaged in
interactional activities in which they discussed environmental problems in
thei r country. The topic was chosen as one that would generate contexts fo r

170 Second la nguage Learning in the classroom
the use of conditional clauses such as 'If people didn't leave water running
while brushing their teeth, they would save an estimated 5-10 gallons each
time' (p, 213) . Learners were audio-recorded as they discussed the environ­
mental problems. .
The recorded conversations were examined to see the extent to which stu­
dents used interactional features that are believed to facilitate second language
learning, for example, negative feedback (i.e. clarification requests, explicit
co rrection, and recasts) an d mo dified output (i.e. a learner's more accurate/
complex reformulation ofhis or her previous utterance). Learners were tested
on their ability to produce conditional clauses in a pre -test, an immediate
post-test, and a delayed post-test.
McDonough found that lear ners who had used more negative feedback and
modified Output in their interactions sign ificantly improved in the accu­
racy of their conditional clauses. Those who made less use of these feat ures
did nor. McDonough also explored opinions about the usefulness of pair
work and small-group activities, asking whether such act ivities contributed
to learning. She found that the students did not perceive pair- and group­
activities as useful for learning English. This was true both for students who
seemed to have made effective use of the interaction for learning and those
who had nor. The fact that learners were sceptical of the benefits of group­
and pair-work activities suggestS a need to take account of learners' beliefs
about learning (see Chapter 3) and to share with them our reasons for using
these activi ties.
Interpreting the research
Research based on the interaction hypo th esis has investigated facrors that
contribute to the q uality and quantity of interactions between second
language learners. It has provided some useful information for teach ing.
Certainly, the studies by Long and Porter, Yule and Macdonald, and Storch
contribute to a better understanding ofhow to organize group and pair work
more effectively in the classroom. The Mackey and McDonough studies are
two examples of research that have measured second language development
in relation to different aspects of conversational interaction. Mackey's study
used one-on-one pair-work activities between trained native speakers and
non-native speakers focusing on a single grammatical feature in a laboratory
context. Thus it is difficult to relate the findings to the kind of interactions
that take place in classrooms. The McDonough study helps to fill this gap
because it is a classroom study that also demonstrates the benefits of inrerac­
tion on second language learning over time.
Recently, a num ber of labo rato ry studies have also examined the effects of
different interactio nal featu res on speci fic aspects o f second language lear n­
ing over time. Several studies have shown that implicit corrective feedback
Second language Learning in th e classroom 171
(for example, recasts ) in pair-work situations is beneficial. A recent review of
this research confirms that the positive effects for recasts are strongesr in the
laboratory setting (M ackey and Goo 2007) . This may be because recasts are
more salient in pair work, par ticularly if only one form is recast consistently
(Nicholas, Lighrbown, and Spada 200 1).
In McDonough's classroom study, recasts (and other forms of corrective
feedback) were more likely to have been no ticed beca use the Thai learners
were accustomed ro traditional grammar instruction and a focus on accuracy.
This is not always the case, however. As we learned in Chapter 5, when the
instructional focus is on expressing meaning through subject-matter instruc­
tion, the teachers' recasts may not be perceived by the learners as an attempt
to correct their language form but rather as JUSt another way of saying the
same thing. Later in this chapter we will look at classroom studies related to
the 'G et it right in the end' posi tion that have investigated the effects of more
explicit corrective fc;edback on second language learn ing. .
4 Get two for one
In content-based language teaching, learners acq uire a second or foreign
language as they study subject matter taught in that language. It is imple­
mented in a great variety of instructional settings, for example, immersion
programmes and the content and language-integrated learning (C LIL) pro­
grammes in Europe described in Study 8 in Chapter 5. Other educational
programmes such as the European Schools extend this further by offering
instruction in two or more languages in addition ro students' home language.
The expectation of this approach is that students can get 'two for one', learn­
ing the subject ma tter content and the language at the sam e time.
In immersion and CLIL programmes, studen ts choose (or their parents
choose for them) ro receive content-based instruction in a second language.
In many educational situations, however, no other option is available. For
example, in some countries, the only language ofschoo ling is the language of
a previous colonial power. In others, educational ma terials are not available
in all local languages, so one language is chosen as the language ofeducation.
In count ries of immigration, students often have access to schooling o nly
through the majority language. Other students may have access to bilingual
education programmes that allow some use ofa language they already know,
but the tran sitio n ro the majority language is usually made with in a year
o r two.
Research findings
In many contexts fo r content-based instruc tion, it is simply assumed that
students will develop bo th th eir academic skills an d second language ability.

172 Second Lmguage learning in the classroom
In recent years. researchers ha ve so ugh t to exam ine this assum p tio n mo re
critically.
Study 26: French immersion programmes in Canada
Research in Canadian French immersion programmes is often cited in
suPPOrt of th e 'Ge t two for one' prop osal. M ost immersion pr ogrammes are
offered in primary and seco ndary schools, but some universities also offer
co nt ent -based instr uctio n th at expands opportunities for students to use
their second language in cognitively challenging an d informative courses.
What have the studies shown?
In terms of popular ity and longevity, French im mersion has been a great
success. Thousands of English-speaking Canadian families have chosen thi s
option since its first implementation in the 1960 s (Lambert and Tucker
1972), both in areas where French is spoken in the wider community an d in
those where French is rarely heard outside the classroom. Numerous stu d ies
have show n that Fren ch immersion stu den ts develop fluen cy, high levels of
listening com p rehe ns ion, and co nfide nce in usin g th eir seco nd langu age.
Th ey also ma intain a level o f success in the ir academic sub jects that is com­
pa rable to th at of th eir peer s whose ed uca tio n has been in English (Genesee
1987). Over th e years, how ever, ed ucato rs and research ers bega n to exp ress
co nce rn about students' failu re to achieve h igh levels of performan ce in so me
aspec ts o f French grammar, even after several years a Hull-day expos ure to the
second language in these programmes.
So me researchers argue that the difficulty French immersion learners experi­
en ce in their L2 production shows that comprehensible input is not enough
(H arley and Swa in 1984). They claim that the learners engage in roo little
language production becau se the classes are largel y teacher-cen tr ed. Students
are observed to speak relatively little and rarely req u ired to give extende d
answers. This permits them to operate successfully with the ir incom plet e
knowledge of th e language becau se they are rarely pushed to be more preci se
or more accurate . \~en stu de nts do speak, com m un icatio n is usually sat ­
isfactory in spi te of nu mero us errors in their speec h becau se the learners'
interlanguages are infl uenced by the same first langu age, the same learn ing
env iro nm ent, and the same limi ted contact wit h the target language ours ide
the classroom. Teachers also tend to understand students ' in rerlanguage, so
there is rarely a need to negoriare for m eani ng. Such successful comm unica­
tion makes it diffic ult for an individ ual learner to work our how his or her use
of the language d iffers from rhe target language.
An other expla nation for students' lack of p rogress on ce rtain lan guage fea­
tures is their rarity in French im mersion instructio n. Fo r example, Merrill
Swain ( 1988) observed that even history lesso ns. where past ten se verbs
might be expected to occ ur, were often delive red in the ' histo rical present'
(fa t example, 'Th e ships go down to the C aribbean ; th ey pick up sugar and
Second langu,tge learning in the classroom 173
they take it back to Eng land . . . '). Roy Lyster ( 1994) foun d that th e use of
the seco nd perso n prono un uous to pol itely address an individual was used so
rarely in classes that even afte r years of immersion instr uction , stu dents did
nor use it appropriately. Elaine Tarone and Merrill Swain (1995) no ted th at
learners with only classroom exposure to the language did not have access to
th e s pe ech styles that wo uld be ryp ical of int eraction am ong nati ve speake rs
of the same age. Increasingly, it was suggested that subject matter instruction
needed to be complem ented by instruction that focused o n language form,
includ ing pragmatic features o f th e language. In so me experimen tal studies,
learners did benefit from fo rm-focused instruction on particular language
featu res (see th e ' G et it right in th e end' proposal).
Study 2 7: Late immersion under stress in Hong Kong
In the 1960s the educational system in Hong Kong moved from one in
whi ch students studied eith er exclusively in English or in Canto nese to one
in which rhe majority of students stud ied in Cantonese in primary scho o l
. (G rades 1-6) an d in Eng lish at seco ndary school (G rades 7- 13). These
late Eng lish im mersion p rogram mes were pop ular wit h C hinese parents
who wa nte d their ch ildren to succeed p ro fession ally and academically in
the inte rnational co mmunity. They were also seen as be ing consis tent with
the Ho ng Kong govern ment 's goa l of maintai ni ng a hig h level of Chinese­
English bilingualism.
In review ing some o f the resear ch o n teaching and learn in g beh aviours in
lat e English immersion classes in H ong Kong seco ndary sch ool s, Keith
Johnson (1997) raised co nce rns abo ut the ability o f the edu catio nal system
to meet the demands for such programmes. He noted that students lacked
the English proficien cy needed to follow the secon da ry level cu rriculum suc­
cessfully. He also observed teach ers' difficul ties in effecti vely deli vering th e
co n tent beca use of limitations in their own En glish profi ciency. He argued
th at several pedagogic behaviours consrlbured to th e inability o f learners to
make adeq uate linguistic pro gress in thes e En glish immersion programmes.
One of them was teach er talk that con sisted o f English, Chinese, and 'Mix'
(a combination o f English and C h inese) .
Observational classroom studies revealed that Chinese and MLx pred omi­
nat ed in th e speech of teachers and that students interacted with the teacher
and with each other in English only in min imal ways . Many students came
to the first year of secondary school withour any lite racy skills in Englis h.
To compensate for th is, teachers employed a variety of st rategies to help
stu dent s co m p rehend texts . They red uced the vocabulary load . sim plified
the grammar, encouraged the use of bilingual dictionaries , and provided
stu d en ts wit h su pplementary no res and charts in Chinese to assist their com­
prehension. Johnson observed tha t, while 'the texts are no t translated, they
are essentially pr e-tau gh t so that by the time students come to read the texts
174 Second kmguage learn ing in the classroom
for themselves the more able students at least are sufficien tly familiar with
the content to be able ro deal with them' ( p.I 77) . Although these strategies
helped students un derstand the content, they may not have helped them
learn ro use th e syntactic and discourse structures in the second language to
establish form-meaning relationshi ps. Therefor e it is no t surprising that th e
st andards of reading in English at age 15 were repo rted to 'be significan tly
lower than those for C hinese. At the sam e time, howeve r, the ed uca tio nal
o ut comes for Hong Kon g student s in content su bjects contin ued to be high,
com parable to , and in so me areas superior to , achievements in other devel ­
oped countries. In addition, th e levels of Chinese Ll reading proficiency
remained high.
In spi te of p ro fession al development efforts ro help teachers achieve the d ual
goals of language and content instruction , Phil ip H oare an d Stella Kong
(2008) find that many teachers in th e H ong Kong immersion p rogrammes
conti nue to have difficulty implementing immersion pedagogy. They attrib­
ute this in part ro the pressure teachers feel in a sociery.where performance
on examinations is paramount. To ensure that their students do well on the
content exams, teachers o ften feel that they must teach in Chinese or in a
simplified English that does not give st udents access ro the lang uage th at is
appropriate fo r high-level academic work.
Study 2 8: Dual immersion
In recent years, legislation has lim ited the availability of bilingual education
for most minoriry language st ude nts in th e Un ited States. In most states,
English language learners' education must take place entirely in English , or
with on ly minimal suPPOrt for learning through their first lan guage. As we
saw in Chapter I, the result ofthis ap proach is often su bt ractive bilingualism.
Children grad ually lose their first lan guage or fail ro develop it for academic
purposes. In addition, they often fall behind in their academic work because
they do not yet have the English language skills needed for dealing with th e
grade-level subject matter.
Some jurisdictions allow 'dual immersio n' as an exception to the Strict
enforcement ofins truction thro ugh English on ly. In dual immersion, minor­
iry language studen ts learn English in classro oms whe re English-speaking
children also learn the minoriry language students' home lan guage. Patsy
Lighrbown (2007) observed classroom interaction and learning outcomes
in a school where an equ al number of native English- and native Spanish­
speaki ng students shared the classrooms. Starting in kindergarten, half th eir
instruction was delivered in English by an English native speaker and half in
Spanish by a na tive speaker of that lang uage. Teachers coo rdinated closely
to ensure tha t the su bject rna rre r inst ructio n in th e two languages was co m­
plement ar y rath er than red und anr. Stude nts' performance on a variery of
meas ures administered through Grade 3 showed that the programme was
Second language learn ing in the classroom 17 5
be neficial for their d evelop ment of En glish lan guage skills, and Spani sh ­
speaking students made especially rapid p rog ress in readi ng. O f particul ar
importance was the fact that studen ts also continued ro develop their Spanish
language skills in ways that were not available ro students whose instruction
was either in English only or in transitional bilingual classes where there was
very lit tle suPPOrt fo r their home language.
In other d ual immersio n programmes, the number of students an d the
amount o f time are distri bu ted differe n tly, often with more time devoted to
the teaching and lear ning of English, but sometimes with an early emphasis
on the m inority language th at resembles the Canadian French immersion
programmes, that is, where the English-speaking students receive nearly all
th eir early instruction in their second language. Th is ap proach is preferred
in settings where it is not possible ro have a substantial number of students
from the minoriry language. Indeed, a number of different models of d ual
imm ersio n have been developed, but they all are based on the p rinciple that
the co n tin ued develop ment ofa ch ild 's home language is a strong foundation
on which ro bu ild second language abil it ies (H oward er al. 2007).
In recent years, several research reviews ha ve exam ined th e evidence fo r
di fferent app roaches to educating English language lear ners in the United
States (August and Shanahan 2008; Genesee er al. 2006). The researc h con­
firms better outcomes, in bo th En glish language learning and subject ma tter
knowledge, for minority lan guage stu d en ts in programmes that support the
students' home lan guages than fo r those in English only o r 'early exit' bilin­
gual program mes where they receive only to ken opportunities ro co ntinue
learning through their home language. Kathryn Lindholm-Leary (200 1)
also found benefits for the majori ty language studen ts in dual immersio n
programmes, where they share the challenges and achievemen ts of second
language learn ing with mi noriry lan guage students. She suggests that this
app roach 'has th e potential to eradicate the negative status of bilingualism
in the U S' (p.1).
Study 29: Inuit children in content-basedprogrammes
In an aboriginal com m un iry in Q uebec, Canada, N ina Spada and Patsy
Lightbown (2002) observed the teaching and learning of school su bjects and
language with Inui t ch ildren. The children had been educate d in their first
language, Inukritut, from kindergarten ro Grade 2 (aged 5-7). Then, except
for occasional lessons in Inuit culture, their education was in one of Canada's
official langu ages, Fren ch or Eng lish . Nearly all students had so me d ifficulry
coping with sub ject matter instruction in thei r second language.
In a case study o f one Frenc h secondary-level class, the y observed instr uc­
tional activi ties, analysed instr uctional materials, and assessed stu dent s'
ab iliry ro un derstand and to pro d uce wri tten French. In the observation
data from a social studies lesso n, it was evident that the teacher had to work


176 Second language Learning in the classroom
very hard co help st ude nt s understand a text o n beluga whales. H e did this
in many ways-by paraphrasing, repeating, sim plifying, checking for com­
prehension, gestures, etc . Despite his efforts it was clea r that most students
understood very little of the te-xt. In their French language classes, these same
st uden ts also lacked the terminology they needed co talk about grammatical
gende r in relati on to adje cti ve agreement.
\'V'hen the students' performance on a wide range of measures was examined
to assess their knowledge of French (fo r example, vocabulary recognition ,
reading comprehension, writing) . it was evident that the students did not
have the French language skills they needed to cope with the demands of
typical secondary-level instruction. Furthermore. even though many of the
stude nts were able to speak French informally outside class, their oral abili­
ties we re limited when th ey had to di scu ss more co m plex academic subject
matter. As we saw in Chapter 1. teachers are sometimes misled by students'
ab ility to use th e language in informal settings, concluding that their aca ­
demic difficulties co uld not be due co lan gu age problem s.
The stu de nts' lackor age-a p prop riate acad em ic Frenc h is a serio us p roblem.
Solving it wi ll involve complex educat ional. social . an d cultu ral quest ions.
One pedagogical element that mi ght co nt rib u te to a so lu tion is a bett er
balance be tween lang uage and sub ject matter instr uc tio n. focu sin g o n th e
language that the students need to suc cee d in schoo l. An other possibiliry is
that furthe r development of the learners' L 1 literacy would better prepare
them for seco nd lan guage and subject matter learning. There is another good
reason to support students' development of Inuktirur. There are increasing
co ncern s that Inuktirut will be lost as future generations shift to English or
French as their preferred language. An educational system that encourages
the de velopment of both firSt and seco nd languages may ensure the su rvival
or this heritage language (Taylo r. Caron, and McAlpine 2000).
Interpreting the research
Cont ent-based language teaching has many advantages. In general . it
in creases the amount of time for learners co be exposed to the new language.
It creates a genuine need to communicate. motivating students to acquire
language in order to understand the content. For older students , there is the
or
advantage content that is cog nitively challenging and interesting in a way
that is often missing in foreign language instruction, especially where lessons
are designed around particular grammatical forms.
N everth eless. there are also some problems with content-based instruction.
Our research with Inuit children adds further evidence co Jim C ummins'
( 19 84) claim that students need 5-7 years before their ability co use the
la nguage for cognitively challenging academic material has reached an age­
appropriate level. For students From disadvantaged minority groups, this
Second language learnmg in the classroom 177
delay in co mi ng co g rips with sc hooling can have lasti ng effects, as we saw in
th e discussion or su btractive bilingualism in C hapter I . Majority language
students in immersion programmes-in Canada and in Hong Kong-seem
to do well in learning subject matter, an d it is also no teworthy that they
recei ve a substantial amount of subject matter instruction through their
first language over the full co urse of their acade m ic careers. Similarly. dual
immersion programmes allow students from each language group to con­
tinue development of the home language and to co ntin ue learning subject
matter content in that language. However. although students in content­
based language instruction are able to communicate with so me fluen cy in th e
seco nd language. they often fall short of the high levels of linguiscic accuracy
that their years of schooling in the language might predict.
In recent years, propo ne nts of co nren t-based in srructi on have stressed the
need co recall that content-based language teaching is still language teaching.
For exam ple. jana Echevarria. Mary Ellen Vogt, and Deborah Short (2004)
have done research and developed teacher education pro grammes th at sh ow
the effect iveness of lesso ns t har have bo t h corue n r o bjectives an d lan guage
objectives.
5 Teach what is teachable
Manfred Pienemann ( 1988) and his co lleagues have tried to explain why it
often seems that some things can be taught successfull y whereas other th ings .
eve n after extensive or intensive teaching. seem to remain unacquired. Their
research provides evidence that some linguistic str uct u res, for exam ple, basic
word order in sentences (bo th sim ple and complex) develop along a predict­
able developmental path . These are labelled developmental features. The
developmental stages of English questions that we saw in C hapter 2 are based
on this research. According to Pienernann (1988) an y attempt to teach a
Stage 4 word-order pattern to learners at Stage 1 will not work because learn­
ers have to pass through Stage 2 and ger to Stage 3 before they are read y to
acquire what is at Stage 4. As we saw in 'G et it right from the beginning',
students may produce certain struc tures after they have been taught them
in class, but cease to use them later because they are not fully integrated into
their interlanguage systems . The underlying cause of the stages has no t been
fully explained, but processabiliry theory (see Chapter 4) suggests that they
may be based at least in part on learners' developing ability to notice and
remember elements in the st ream of speech they hear.
Resear chers supporting this view also cla im that certain other aspects of
lan guage -for example, individual vocabulary items-can be taught at any
time. Learners' acquisition of these variational features appears to depend on
facto rs su ch as m otivation . th e learners' sense o f identity. language aptitude ,
178 Second language learning in the classroom Second language learning in the classroom 179

and th e quality of insrrucrion, including how learners' identi ties and culrures Example 8

are acknowledged in the classroom. (Interviewing each other abo ut house pr eferen ces)

In Example 6 below, we see a teacher trying to help students with the word 5 I Is yo ur favourite house is a split-level?
order of questio ns. The st udents seem to know what the teacher means, but 52 Yes.
th e level oflanguage the teacher is offering them is beyond their current stage T You're saying 'is' twO times dear. 'Is your favo urite house a spl it-le vel?'
of develop men t. Students are asking Stage 3 q uestio ns, whic h th e teach er 5I A sp lit-level.
recasts as Stage 5 questions . Th e students react by simply answering th e ques­ T O K.
tion or acce pting th e teacher's fo rm ulati on.
In Example 9 the student asks a Stage 3 question, and th e teacher provides a

Example 6 Stage 4 correction that the student imitates. The interaction suggestS that the

Students in an intensive ESL class (l 1-12-year-old French speakers) inter­ student is almost ready to begin producing Stage 4 q uestions. Note, however,

viewing a stude nt who had been in the same class in a previous year (see that the student does not imitate the possessive 's, som eth in g tha t French.

Classroom B in C hapter 5). speakers find very d ifficul t.

s I Mylene, where you pur your 'Kid of the Week' pos ter?
Example 9

T Where did you put your poster when you go t it?

(' Hide an d seek' game)

52 In m y room.

5 Do the boy is beside the teacher desk?

(Two minu tes later)
T Is th e boy beside the teacher's desk?
53 Beatrice, where you pur your 'Kid of the Week' poster?
5 Is th e boy beside the teacher desk?
T Where did yo u pu t your poster?
Research findings
54 M y poster was on my wall and it fell down.
The 'Teach what is teachable' view suggesrs that while variational features
In Example 7, the student is using the 'fron ting' strategy that is typical of
of the language can be taught successfully at various points in the learners'
Stage 3 q uestions. The teacher's corrective feedback leads the stu dent (0
development, develo pmental features are best taught acco rdi ng to the learners '
imitate a Stage 4 question.
internal sche d ule. Fu rthermore, although learners may be able (0 produce more
advanced forms on tests or in very restricted pedagogical exercises, instruction
Example 7
cannot change the 'natural' developmental course. The recommendation is to
(Th e same gro up of studen ts engaged in ' Famo us person' interviews.)
assess th e learners' developmental level and teach what would naturally come
next. Let us examine some srudies th at have tested th is hypothesis.
5I Is your mother play piano?
T 'Is your mother play piano? ' OK. Well, can you say 'Is yo ur mother Study 30: Ready to learn
'play piano?' or 'Is yo ur mother a p iano player?'
M anfred Pienemann (1988 ) investigated whe ther instruction permitted
5I ' Is your mo th er a piano player?'
learners to 'skip' a stage in th e natu ral sequence of de velo pment. Two groups
52 N o.
of Australian un iversity students of German who were at Stage 2 in their
In Example 8, the teac her draws the student's attention to th e error and also acquisitio n of German word order were taught the rules associa ted with
provides the correct Stage 4 question. This time, however, the feedback is no t Stage 3 and Stage 4 respectively. The instr uction took place over twO weeks
followed by an imitation or a refo rmulatio n of the question, bur simply by and during this tim e learners were provided wi th expli cit grammatical rules
an answer. and exercises for Stage 3 an d 4 constructions. Th e learners who received
ins truc tion on Stage 3 ru les moved easily into thi s stage from Stage 2.
However, those lear ners who received instruction on Stage 4 rules either con­
tinued (0 use Stage 2 rules or moved only into Stage 3 . That is, they were no t
able (0 'skip' a stage in the develo pmental sequence. Pienemann in terp rets
his results as suPPOrt for the hyp o th esis that for some lin gu istic st ructu res,
learners can not be tau ght w ha t they are not developmentally ready (0 learn .
180 Second language learning In the classroom Second language learning in th e classroom 181

Study 31: Readies, un readies, and recasts

doing?'). This parrern in the students' performance ap pe-ars to be related to a
Aliso n Mac key and Je nefer Philp (1998) in vestigat ed whethe r ad ult ESL
quest io n rule in their hrst lan guage that we saw in Chap ter 2. That is, in French,
learne rs who were at different stages in th eir acq uisitio n o f qu esti ons co uld
questio ns with no un s in su bject position are not inverted (for exam ple, "Peut­
advance in their production of th ese forms if they received implicit nega­
Jean uenir chez moi? ='Can John come co my house?'). In French qu estions wi th
tive feedback (i.e. recasts) in co nversational interaction. As described in
pronoun subjects, however, inversion is permitted (fo r example, Peut-il uenir
Chapter 5, recasts are paraphrases of a learner's incorrect urrerance tha t
chez moi? = 'Can he come to my house?') .
invo lve repl acing one or mo re of me incorrect components w ith a cor rect

form wh ile mai ntaining a foc us o n meaning. Th e researchers were interested

These results in dicate mat instruction timed co match lear ners' develop­
in discovering w he the r ad ult lear ners who received modified interaction with
me ntal 'readi ness' may move the m into mo re advanced stages, bur their
recasts we re ab le co advan ce in th eir production of q uestion forms more than
performance may still be affected by other factors. In this study first language
learne rs who received m odi fied inte ractio n without recasts. Furt hermo re,
influen ce seems co be respo nsible for th e learners' inability co generalize their
they wan ted CO explo re whether learners wh o we re at more ad vanced stages
knowledge of inve.rsion to all questions.
of q uestion development ('readies') would benefit more from interaction

with recasts than lear ners at less advanced stages of question dev elopment
Interpreting the research
('unr ead ies'). The results revealed mat me ' readies' in the interaction plus
The resul ts of these studies sugges t that targeting instructional or interac­
recasts group improved mo re tha n me ' read ies' in th e interaction without
tio nal in pu t to lear ners wh en they are deve lopmen tally ready to progress
recasts group. However. the 'un read ies' who were exposed co recasts did not
further in the second language can be . beneficial. H oweve r, other factor s
sho w more rapid improvement than those who were not exposed co recasts.
such as rype of inpur andfirst language influence can interact with learners'
develo pment al readiness in co m plex ways. If we compare the rypes o f instrue­
Study 32: Developmental stage andfirst language influence
rional/ in teracrio nal in pu t acro ss the three studies , Pienemann provided th e
N ina Spada and Patsy Lighcbown ( 1999) have also in vestigated the acqu i­
most exp licit instruction to learners who were both ' ready' and 'u nready' .
sitio n of questions in relation to learners' developmental ' read iness'.
The results showed tha t learners who were 'ready' moved into the next stage
French-speaking st udents (aged 11-12) in int ensive ESL classes received
of develo pment, whereas learners who were not ' ready' did not. The results
high-frequency exposure to q uestion forms that were one or rwo stages
of th e Mackey and Philp study also offer some sup port for m e teachabiliry
beyo nd thei r develop mental level. Learners who were judged on oral pre­
hypothesis but reveal th at developmental readiness is not the only predictor
tests (Q be at Stage 2 or 3 were given h igh freque ncy exposure co Stage 4 an d
of success. Th e fact th at m e 'readies' be nefited mo re from recasts tha n the
5 questio ns in th e inst ruc tio nal in put.
'un read ies' sugges ts th at the rype of instru ctio nal/ interactio nal in put is also
The materials that contained the more advanced question forms were
im poreant. The Spada and Ligh tbown study shows how me learn ers' first lan ­
designed (Q engage th e learn ers mainly in com prehension practice. There was
guage may interact with developmental read iness in Contributing to learni ng
no student producti on an d th us no co rrective feed back, nor was th ere any
outcomes. Furthermore, in th at stu dy there was no explicit instructio n on
exp licit instruction o n q uestion format ion. The researchers wanted co know
questions. Learners we re sim p ly exposed to a high freque ncy of correctly
whethe r Stage 3 learn ers (i.e, those co nside red co be developmentally ' ready ')
formed h igh er-stage questions in the in put. Thus, th ey received increased
would be nefi t mo re from the hig h-frequency exposure (Q Stage 4 and 5 q ues­
'exp os ure' bu t no ' instr uction' , and, in the en d, they did not show as much
tio ns than the Stage 2 learn ers, who we re not yet devel opmentally ' ready'.
developme ntal change as learners wh o received foc used instruc tion .

Learners' perform an ce on an oral pOSt-test measure indicated no advantage for

So me research ap pears to offer co unter-evide nce to the claim tha t it is ben­
me Stage 3 lear ners. In fact, the re was little progress for eith er group. H owever,
eficial to teach what is d evelopmentally nex t. Th is includes several s tudies
on a task m at requi red learn ers co judge th e grammaricaliry of written qu es­
that have used m e accessibili ty hierarchy (see C hap te r 2) to descri be seco nd
tio ns there was evidence th at all stu de nts had som e kn owledge o f Stage 4 and
lan guage learn ers' pr ogress in th eir acqu isition o f relative clauses. Resul ts of
5 ques tions. A mo re de tailed exam ina tion of the learn ers' perfor mance on th is
these stu dies suggest that wh en low-level learners (fo r exam ple, th ose who
task showed tha t stude nts tended to acce pt Stage 4 an d 5 q uestions whe n the
use relat ive cla uses only in subject positi on ) are ta ugh r relative clauses th at are
subject o f the sente nce was a pron oun (for exam ple, 'Are you a good student?' o r
several stages beyond the ir curre nt level, th ey not o nly learn wha t is taugh t,
'When are you going to eat breakfast?') . \'7hen the subject of the sentence was
they also acquire the relative cla use posi tion(s) berween me one tau gh t and
a noun , however, there was a ten den cy for students co reject high er stage qu es­
the o ne(s) they alread y kn ew. In so me ins tances th ey even learn how to use
tio ns (for exam ple, 'Are the students watchi ng TV?' or 'What is yo ur brother
relative clauses beyo nd the level they were tau ght (Am mar an d Lighr bown
200 5; Eckman, Bell. and N elso n 1988; Ham ilton 1994) .
182 Second Language learning in the classroom
At first glance, thi s research seems ro con tradict Pien ernan n's claim th at
learners should be taught what is 'next'. However, it is also possible that
the de velopmental parhs of differenr linguist ic features are based on diffe r­
en t sorts of processing ab ilities. For example. Carherine Doughty (1 99 1)
suggested that once learners have learned ro use relative clauses in one po si­
tio n (usually rhe subject posi tion), rhere is no constraint on their abili ty ro
learn rhe ot hers. What aU rhe Studies of relative clause teaching and learning
have in common is that learners acquire th e relative clauses in an order very
similar to rhe accessibili ty hierarchy. That is, whe ther or not they learn what
is taught, th ey make progress by lear n ing subject, rhen direct object, then
indirect object, and so on.
Th e 'Teach what is teachable' position is of great porenrial interest ro syl­
lab us p lan n ers as well as teachers. However, it must be emphasized that a
description of a learner's develop m en tal path is not in itsel f a tem plate for a
syllabus. Th ere are n umerous practical reaso ns for th is, not least rhe fact th at
o nly a small n umber of language features hav e been described in term s of a
developmen tal seq uence. While Pienernann's wo rk o n processability rheory
(see Chapter 4) provides insights into the pri nciples that ma y make so m e
features more difficult than others, those pri nciples are not easily tran slated
into instructional sequences.
As Patsy Lighrbown ( 1998 ) has suggested, the 'Teach what is teachable'
researc h is irnporranr primarily for helping teachers understand why Stu­
de nts don't always learn wha t they are taugh t- at least not immediately.
The research also shows tha t instruction on language th at is 'too ad vanced'
ma y still be helpful by providing learners with samples oflanguage that they
will be able ro inco rporate into their inrerlanguage when the time is right.
H owever, many other factors need ro be taken into consideration in choosing
language feat ures ro focus on. We will return ro rhis point after we discuss rhe
final proposal for language teaching, 'Get it right in rhe end .'
6 Get it right in the end
Proponents of rhe 'G et it righ t in rhe end' proposal recognize an important
role for form-focused instruction, bur rhey do no t assume that everyth ing
has to be ta ught. Like advocates of rhe 'Let's talk', 'Two for one', and rh e 'JUSt
listen ... and read ' positions, the y h ave co ncl ude d tha t ma ny lang uag e fea­
rures-from pro n un ciation ro vocabulary and grammar-will be acquired
narurally if learners have adequate exposure ro rhe language and a motivation
to learn. Thus, while they view co m p rehensio n-based, conrenr-based , task ­
based, or other types of essentially meaning -focused instructio n as crucial
for language learn ing, they hypothesize th at learners will do better if they
also have access ro some form- focus ed ins tructio n. They argue that learners
will benefit in terms ofb orh efficien cy of rhei r learni ng an d rhe level of p ro­
ficien cy rhey will eventually reach.
Second Language learn ing in the classroom 183
Advocates of this p ro posal also agree wirh advocates of the 'Teach wha t is
teachable' view that so me rhings can no t be taught if rhe teach ing fails to
take the student's readiness (stage of development) in to account. This pro­
posal differs from rhe 'Teach what is teachable' proposal, ho wever, in th at it
em phasizes rhe idea that some aspects of language must be taught and may
need ':0 be taught qu ite explicitly, There are a number of siruatio ns in wh ich
guidance- form-focused ins truction or corrective feedback-is expected to
be especially important. For example, when lear ners in a class share the same
first language, the y will make errors that are partly rhe result of transfer from
thar shared Ianguage. Because rhe errors are not likely ro lead to any kind of
communication breakdown, it will be virt ually impossib le for learners to
discover the er rors on their own .
Examples 10 , II , and 12 are taken from a classroom where a group o f 12-y ear­
old French speakers are .Iearning Engl ish . In Example 10, they are engaged
in an activity where the words in senrences are reo rdered ro form new sen ­
tences. Th e following sentence has been placed o n the board : 'So metim es my
mother makes good cakes '.
Example 10
T Anorher p lace ro put our adv erb ?
5I After makes?
T After makes.
52 Before good?
T My mother make s sometimes good cakes.
53 No .
T No , we can't do th at . It sounds yucky.
53 Yucky!
T Disgusting. Horrible. Right?
54 Horrible!
This is hardly a typ ical grammar lesson! "An d yet the students' at ten tion is
being drawn to an error that virtually all of them make in English.
Proponents of 'Get it righ t in rhe end' argue that what learners focus on
can eventually lead to changes in the ir interlanguage systems, not JUSt ro
an appearance of change. However, rhe supporters o f this p roposal do not
claim that focusing on particular language points will prevent learners from
maki ng errors or that they will begin using a fo rm as soon as it is taught.
Rather, they sugges t that th e focused instruction will allow learners to notice
the target fearu res in subsequent input and interaction.
Form -focused instruction as it is understood in th is position does not always
involve rnecaiinguisric explana tions, nor are learners expected to be able to
exp lain wh y so me thing is righ t or wrong. They claim simply that rhe learners
need to no tice how rhei r language use differs from that of a more proficient
speaker. As we will see in the examples below, teachers who work in rhis

184 Second Language learn ing In the classroom
approach loo k for the right moment to crea te increased awareness on the
parr of the learner-ideally, at a time whe n the learner is m otivat ed to say
somethi ng an d wan ts to say it as d earl y and correctly as possi ble.
Examp le 11
(The students are practising ~ollowing instructions; one student instructs,
the others colour.) .
51 Make her shoes brown .
months in eithe r G rade 5 or Grade 6, st udents (aged 10-12) spent mos t of
T Now, he r shoes. Are those Mom's shoes or Dad's shoes?
every schoo l day learning Englis h th rough a variety of communicative inter­
52 M om's.
active activ ities .
T Mo m's. H ow do you kn ow it's Morn's?
51 Because it's he r shoes ,
As we saw in Chapter 4, Fren ch-s peaking learn ers of English ha ve di fficu lty
with 'his' and 'h er' beca use French possessives use the gram matical gt nder of
t he objec t possessed rather th an th e na tural ge nde r o f the possessor in select­
ing the appropriate possessive form. The teacher is aware of th is and-b riefly,
without interrupting the activity-helps the learners no tice the correct form .
Example 12
lTh e stu dents are playing 'h ide and seek ' with a doll in a doll 's house, as king
questions until they find ou t where 'G eo rge' is hiding. Although a model tor
correct q uestions has been written o n the board, th e game becomes quite
lively a nd stu de nts spo ntaneously ask q uestio ns that reflect the ir inrerlan­
guage srage .)
s I Is Geo rge is in the living room?
T You said 'is' two ti mes, dear. Listen to you-you said 'Is George is
in?' Look o n the board. 'Is George in the ' and then yo u say th e nam e
of the ro o m.
51 Is G eor ge in th e living roo m?
T Yeah.
SI I w in!
I
I
Note tha t th e teach er's brief interventio n does not distrac t the student fro m
his pleasure in th e game, demonst rating th at foc us o n form does not have to
interfere wi th genu ine inte ractio n .
Proponents o f 'G et it right in the en d' argue that it is so meti mes necessary
to draw learners' attentio n to th eir errors and to focu s o n certai n linguistic
(voca bulary or g ram mar) poin ts. H owever, it is d iffere nt from th e 'Get it
right fro m the begi nning' prop osal in acknowledg ing that it is appropriat e
for learners to engage in mean ingful lan guage use fro m the very beginn ing of
thei r expos ure to the seco nd language. Th ey assum e th at m uc h o f lan gu age
acq uisition will develop naturally our of suc h lan gu age use, with o ut formal
ins tr uc tio n th at focuses on the language itself.
St?condLanguage learning in the classroom 185
Research findings
A grea t deal of research has examined issues relat ed to this pro posal. This
includes bo th descriptive and expe rime ntal studies .
Study 33: Form -focus experiments in in tensive ESL
Since the 1980s. researchers have investigated the effects of form-focused
instructio n and corrective feedback on the developing English of French­
speakin g Students participati ng in intens ive ESL classes in Quebec. Fo r five
In descrip tive studies invo lv ing almost 1,000 students in 33 classes, Patsy
Lighcbown and Nina Spada 0990 , 1994) observed that teachers rarel y
focused on language form. The emphasis of the teaching was o n act ivities
that focu sed o n meaning rather than form, opportunities for spontaneous
interactio n, a nd the provision o f rich an d varie d co mp rehens ible input. In
the se classes , learners developed good listening co m p rehe nsio n and co m m u­
nicative co nfide nce in En glish . H owever. th ey co nt inued to have probl em s
with linguistic accuracy and complexity.
In expe riment al stu d ies w ith a smaller number o f classes, th e effect s of form­
focu sed instructio n and corrective feedback were exam ined with respect to
adverb placeme nt and question formation . In the first study, Lydia \'V'hite
selec ted adve rb placeme nt for in vestigatio n because of the diffe rences
between English and French that have already been discussed (see Chapter 2
and Study 17 in 'Just listen . . . and read') . The hypothes is was that learn ers
wo uld pe rsist in usin g adverb placeme nt rules co nsiste nt with Frenc h (th eir
first lan guage) if they were no t exp licitly told how rules for adverb place­
ment differ in English and French. Questio ns we re selec ted fo r th e second
study beca use th ey h ave been extensively inves tigated in the literat ure and
co nsiderable co mparison data were availa ble, particularly with regard to
dev elo pmental stages (see C h apte r 2) .
Bot h experimental and comparison grou ps were tested before and afte r the
pe riod of special inst ructio n. Throughout th e per iod of th e expe rime nts, all
st ude nts co nt in ued to parricip are in th e regul ar co m m u nica tive acti vities
that w ere typical of their in structi on . Th e researchers gave each teac he r a set
of pedagogical macerials to be used for th e special fo rm -focused instruct io n .
The experime ntal groups received app roximately eigh t ho urs of instructio n
on adverbs or q uesti ons ove r a two-week period . This include d so me exp licit
teachi ng of th e rul es associa ted wi th each str ucture as well as co rrective feed­
bac k d ur ing the practice activ ities .
Learners who received explicit instruction on adverb placeme nt dramatical ly
o utpe rforme d the learner s who did not. This was foun d o n all the pos t-tests
186 Second language learning in the classroom
(im m edi ately following ins tru ctio n and six weeks later). In me follow-up
tests a year later, however, me gains made by m e learners who had rece ived
me adverb instructio n had disappeared and their perfo rmance on this Struc­
tu re was like mat of un in structed learn ers (White 199 1).
In th e question study me instr uct ed gro up also mad e signiflcan dy greater
gains than m e uninstructed gro up on th e written tasks im mediately follow ­
in g instr uctio n. Fur thermore, they maintained their level of kn owled ge on
later testi ng (six weeks and six months after instruction). Th e instruction also
contrib uted to improvement in or al performance that was sustained over
time (White, Spada, Ligh rbown , and Ranta 199 1).
Th e difference in long- ter m efiecrs of the two studies m ay be d ue to a differ­
ence in the availability of me target forms in me classroom input to which
learners were exposed. Analysis of classroom language showed mat adverbs
were extremely rare in classroo m spee ch, giving learne rs little opportuni ty
to mai ntain their: newly-acquired knowledg e th ro ugh continued expo sure
and use. In contrast, mere w.e re hu ndreds of op porrunlries to hear and use
ques tio ns every day in m e classroo m . Once lear ners ha d been given some
foc used instruction , it seems they were able to continue to advance in their
knowledge and use of questions (Spada and Llghrbown 1993 ).
In several of the studies carried out in intensive ESL p rogramm es, there is
evidence of me strong influence of me learner's first language on their second
language dev elopment. In Stu dy 32 , we described the ten den cy of intensive
ESL learners to reject in versio n in qu estions when the su bject is a no un but to
accept inversion wh en the subject is a pronoun, consistent with th eir first lan ­
guage. The influence of the learners' first lan guage in thei r acquisitio n of the
possessive determ in ers ' his' and 'her' was obse rved wi th this group oflearners
(see Chapter 2 an d Study 18). This led to me ques tion of whether form­
focu sed instructi on mat includes expli cit contrastive information about how
the first and second language diffe r wo uld help in th eir developme nt ofques­
tion formation and possessive determ ine rs. In a study to expl ore th is, learners
who received instruction on possessive de termi ners improved mo re in their
knowledge and use of this fea tu re than did learners who received instruction
on question forms . This finding appeared to be related to differences between
the fo rm- mean ing connections of these two features. That is, a misused pos­
sessive determiner ('He's going ho me with her mother') is more likely to lead
to a co mmunication breakdown than an ill-forme d question (for example,
'Where he 's going?') . Res ults like th ese po in t to the importance of cons ide r­
ing how instruction ma y affect language features in different ways (Spad a,
Lighrbown, and White 2005 ; White 2008).
As we saw in th e disc ussion of the ' G et two for one' proposal, there is
growin g evide nce that learners in COntent-based programmes such as French
immersio n ne ed m ore opp ortu niti es to focus on form an d receive corrective
Second language Learning in the classroom 187
feed back . A n um ber of studies have exp lo red the quest ion of how this can
best be accomplished .
Study 34: Focusing on gender in French immersion
Birgit Harley (1998) exam ined the effects of instruction with young children
in French immersio n programmes. Six classes of Grade 2 children (7-8 years
old) were given focused instruction on a language feature chat is kn own to be
a persistent problem for French immersion studenrs-i-grammarical gen der.
For 20 minutes a day over a five-week period these children carried our many
activities based on children's games (for example, 'I spy ') m at were modified to
draw their attention to gender distinctions and which required them to choose
between femin ine and masculi ne ar ticles (une or un, la o r Ie) . Srudenrs were
also taught how certain noun endings provide clues about gender (for example,
-ette in la bicyclette for feminine, and -eau in le bateau for masculine).
The students were pr e-tested on th eir knowledge of gram matical gender via
listening and speaking tests before th e ins truction began and the same rests
were administered im m ediately after instr uction and then again five months
later. Learners w ho received instruction were much better at recognizing an d
producing accurate gen der distinctions for familiar nouns than those who
did no r receive instruction. However, the instruction did not enable learn ers
to generalize their learning co new nouns . Harley's interpretation of th is is
th at coo much new vocabulary was introduced in the lat er teaching acti vities
and this meant chat teach ers spent more tim e teach ing the meaning ofwords
than the noun en dings and their relationship co gender. Therefore, 'th e in put
on noun endings was simply no t available in su fficient quantity and intensity
for th e majority o f students to establish me predictive relevan ce of the noun
en dings in questio n' (p. 169).
Study 35: Focusing on sociolinguisticfonns in French immersion
Roy Lyster (994) exam in ed me effects of form-focused instruction on th e
knowledge and use of sociolinguistic style variations in three classes of Grade
8 French immersion stu den ts (about 13 years old ). One of the mai n features
examined in his study was me d istinctio n between the use of second person
pronouns tu and V0 11S. In addressing an individual, tu is used co indicate infor­
mality and familiarity while uous is used as a marker of respectful po liteness , or
social distance between speakers. Prior co instruction, immediately after, and
again one month later, the learners were tested on their ability co produce and
recognize these forms (in addition co ochers) in appropriate con texts.
The instruction cook place for abo ut 12 hours over a five-week period .
During th is time, students in rhe experimental classes were given explicit
instruction and engaged in guided p ractice activities chat included role-plays
in a variety of form al an d informal co ntexts and corrective feedback fro m
teachers and peers . Students in the two comparis on classes co ntin ued with
their regu lar instruction witho ut any focused teac hing or guided practice
188 Second language learning in the classroom
in using soci olin gui sticalJ y appropriate forms. On the immediate post -test ,
learners in the experimental classes performed significantly better than lear n­
ers in the comparison classes on both written and oral production tasks and
the multiple-choice test, and these benefits were maintained when learners
were tested a month later.
Study 36: Focusing on verbfonns in content-based science classrooms
Catherine Doughry and Elizabeth Varela ( 1998) carried OUt a study with a
group ofESL learners in their science classes. One class ofmiddle-school stu­
den ts (11-14 years old) from a variety offirst language backgrounds received
corrective feedback on past tense and conditional verb forms in English.
For several weeks, while studenrs were engaged in oral and written work
related co a series of science reports, me teacher provided corrective feed­
back on th eir errors in past tense and conditional forms-both explicitly and
im plicirly. Students' abiliry CO use chese forms was assessed before and after
the experimental period and again rwo months later. Their performance was
co m pared co that of a gro up o f Stu dents who were in another scie nce class
doing [he sam e scie nce repo rts but w ho did not receive co rrective feedback
on the verb forms .
Studen ts who received the co rrective feed back mad e mo re prog ress in using
past an d co ndi tio nal forms th an the co m par ison gro up both im me diat ely
aft er the period of focused feedback and two monrhs later. Their progress
was assessed in terms o f both increased accuracy and the presence of inter­
language forms chat showed students were doing more than repeating forms
th ey had heard.
Study 37: Recasts andprompts in French immersion classrooms
In C hapter 5, we saw som e o f Roy Lyster's descriptive research on the dif­
ferent types of co rrective feedback provided by teachers in Canadian French
immersion programmes and learners' immediate responses (uptake) co mat
feedback. More recently, Lyster (2 004) explo red the effecrs ofform-focused
instruction and feedback rype on second language learning in an experimen­
tal study with Grade 5 srudenrs in French immersion classes. There were
three experimental groups and a comparison group. The experimenral groups
received approximately nine hours of explici t instruction over a five-week
period, d uring which their attention was drawn to grammatical gender and
the fact that word endings can give a clue co grammatical gender in French
(see Study 34). Students in rwo of me experimental groups also received cor­
rective feedbac k in the form ofeither recasts or prompts when they produced
erro rs in grammatical gender. These feedback types differ in that recasts
provide learners wi th the correct model , whereas promprs signal the need
for a correcrion and requi re the student co produce the target form through
clarification requests, elicitation. and metalinguisric clues (see Chapter 5
for definitions and examples of these different types of feedback). The chird
Second language learning in the classroom 189
expe rime nt al group received the ins tr uction bu t no co rrec tive feed back, and
th e co m p ar ison group received neither instruction nor corrective feedback.
All groups continued their regular French immersion p rogram me of con­
tent-based instruction chrougho ut the srudy and they were all tested before
the instructional treatment, immediately after, and again three monchs later.
O n the post-tests all three experimental groups (i.e. chose who received
instruction) were significanrly more accurate chan the comparison group
in assigning grammatical gender. In addition, the instr uctio n + prompts
gro up did sign ifican rly better than the instruction + recasts group on the
written measures. However, there were no significant differences between
the three experimental groups in terms of/earners' performance on the oral
tasks. Lyster interprets th is finding as a task effect. That is', because of the
time-co nsumi ng nature of oral tasks, o n ly a randomly selected su b-sam p le
of students participated in chis part of the Study. These students met with
the researcher in three intensive on e-on-one sessions. In order to ensure the
accuracy of the d ata , the researcher en couraged students co speak as clearl y as
possi ble because previou s research had shown th at learners so metimes used
a 'hybrid ar ticle' tha t cou ld be inter pret ed as eit he r masculin e or fem inine.
Th is em phasis o n th e clear articulation of articles provided all learne rs with
individ ualize d arte n tion o n the target featu re an d th us may be th e reaso n
why all three gro ups perfo rmed simi larly on th e oral measure.
Study 38: Focus on form through collaborative dialogue
Motivated by sociocultural theory and the idea chat language learning occurs
in dialogue, Merrill Swain and Sharon Lapkin (2002) observed the language
development of two Grade 7 French immersion students as the y wrote a
Story collaboratively. Later, in a 'no ticing' activity; the students compared
wh at they had written with a reformulated version of the Sto ry. Th e students
also took part in a stimulated recal l of their noticing activity. Swain and
Lapkin wanted to find out what students noticed about differences berween
thei r original versio n and the reformulated one and whecher they made
revisions to their original stories based on their collaborative talk about the
reformulated version ,
The talk that learners produced in all phases of the researc h was recorded,
transcribed and coded for language-related episodes (LREs) , 'any part of the
dialogue where learners talk about the lang uage they produced, and reflect
on their language use' (p. 292). An excerpt of the learners' collaborative talk
from this study is presented in Chapter 5, Communication Task B. The LREs
were coded in terms of whether they focused on lexical, grammatical, or dis­
course features . The researchers used the original story that the rwo learners
created together as a pre -test and the stories chat each learner constructed as
a post-tese. Both learners were much mo re accurate on the POSt-test version
of the Story. The researchers conclude that the multiple opportunities for
190 Second language learning in the classroom
learners to engage in collaborative talk on the language features in question
led them to a greater understanding of their correct use.
Study 39: Focus on form in task-based instruction
In a descri ptive study investigating the importance of th e teacher's role in
task- based instruction, Virginia Sarnuda (2001) explored ways of guiding
ad ult E5L lear ners' attention to form-meaning relationships by focusing on
expressions of possibility and probability (fo r example, 'migh t', 'could', 'it's
possible'). In a task design that took learners through a 'meaning to form to
meaning progression', learners were first asked to work in groups to speculate
on the identity ofan unknown person (for example, age, gender, occupation)
by looking at a set ofobje cts thought to have come from that person's pocket.
In carrying out th is task, learners were observed to produce expressions of
p robability and possibility such as 'It's possible that he smokes' and 'maybe
it's a girl' , but few instances of modal auxiliaries (for example , 'must', 'may')
were used .
In the second phase of the task, the students were asked to come together as a
who le group to tell each o ther what they had decided. During this phase, the
teacher acted as a co-communicator and maintained the focus on meaning
but gtadually shifted to form by using the language that the lear ners had
produced on their ow n and ptoviding them with alternative ways ofexpress­
ing uncertainty. Initially, this was done implicidy. For example if a lear ner
said something like 'We th ink uh 50 per cent he smokes', the teacher said
'So you're not certain that he smokes?' After each group had presented, the
teacher provided a more explicit focus. She drew the learners' attention to
other ways of expressing possibility an d probability by overtly talking about
language form as shown in the excerp t below (p. 13 1).
ST Businessman
T Businessman ninety? OK So you're 90 per cent certain he's a
businessman, right? Here's another way to say this. You think it's 90
per cem certain. so you chink he must be a businessman. He must be
a businessman (writes it on the board) . So this (points to 'must be' on
board) is showing how certain how sure you are. Not 100 per cent,
bu t almost 100 per cent. 90 per cent.
In the final stage of the task, the students prepared and presented a posrer
based on their conclusions about the identity of the unknown person to
the whole class. During this time, the teacher responded to the content and
not the form of (heir work. When the researcher examined the differences
between expressions of probability and possibility that the students used in
the firSt stage of this task and compared it with the final stage , there was evi­
dence of improvement in that many more instances of modal auxil iaries were
present in the learners' speech.
Second language learn ing in the classroom 191
Study 40: The timing ofform-focused instruction
Nina Spada and her research team carried out a study to exam ine whether
there may be a better time in the instructional sequence to draw learners'
attention to form (Spada et al. 2012). Two classes of intermediate-level
adult ESL learners were provided with 12 hours of instruction that differed
in terms of whether attention to form was em bedd ed in communicative
activities or separated from communicative prac tice. They are referred to as
integrated and isolated form-focused instruction. The target feature was the
passive construction and learners were tested on their knowledge of it before
instruction, immediately after instruction. and again three weeks later.
A second question motivating the research was whether the rwo types of
instruction might lead to different kinds of L2 knowledge. This question
was informed by transfer-appropriate processing theory and the idea that
we are more likely to remember something we have learned if the cogni­
tive processes that are activated during the learning process are the same as
those activated during retrieval (see Chapter 4): Thus , the researchers were
interested in whether learners who obtained their knowledge of the passive
while participating in com m un icative interactio n (i.e. integrared FFI) were
better at retrieving that knowledge on an oral communication task than the
learners who received isolated FFI. Similarly, they wanted to explore whether
learners who obtained thei r knowledge of the passive structure in grammar
activities that were separated fro m communicative practice (i.e. isolate d FFI)
were better at retr ieving their knowledge on a written grammar test than the
learn ers who received integrated FFI.
Learners in both the integrated and isolated FFI classes improved signifi­
cantly on both language measures over time. The find ings also revealed some
support for TAP in that learners who received integrated FFI outper formed
the isolated FFI learners on the oral communication task and the learn ­
ers who received isolated FFI outperformed the integrated learners on the
written grammar resr.
The overall results of this study point to the complementarity of the two
types of ins truction, likely due to the fact that they both provide a focus on
form and meaning, albeit at different times.
ACTIVITY Match pedagogical activities with teaching
proposals
Below are brief descriptions of 12 pedagogical activities. Match each activity
with the teaching proposal it represents and explain how you reached that
conclusion. For example, an activity such as 'Fill in the blanks with the correct
form of the verb' represents the 'Get it right from the beginning' proposal
because such grammatical exercises are typical of the grammar translation
approach with its emphasis on rule learning and accuracy. Keep in mind
192 Second language Learn ing in th e classroom Second language Learning in the classroom 193

tha t in some cases. an activity is compatible with more than one teaching possessive determi ne rs than q uestio ns becau se the re is a stro nge r fo rm­
prop o sal.For exa mple, if the se ntences in the 'Fill in the blanks' activity came meaning connection with possessive determiners man with questions.
from an earlier draft of a letter w ritten for a co mmunicative activity.it might
Other language featu res for which form-focused instruction may playa crucial
be co nsist ent with the 'Get it right in the end ' proposal because it int egra tes
role are those mat are influenced by the learn ers' first language, particularly
attention to language form in a meaning-based activity.
wh en there are misleading similarities between L1 and U. Th e difficulty may
I Role-playa conversation between a travel agent and a tourist.
be inc reased in second language classrooms w here learn ers share me same first
2 Memorize a dialogue abo ut buying airline tickets.
language and reinforce each other's first language-based errors.
3 Underline the past tense ver bs while rea ding a sto ry.
Finally: the rules associated with some language features are more complex
than others. For example, the article system in English is both complex and
4 Arrange illustrations in the co rrect sequence after listening to a st o r y.

abst ract and notoriously difficult to teach and learn. Thus, learners may be
S W ork with a partner to write a story based on a cartoon strip.
better off learning abo u t articles via exposure in the input. On the o th er
6 Rearrange a set of scrambled words to form correct questions.
hand a simple 'rule of thumb' suc h as 'put an -$ at the end of a noun to make
it plural' may be a better target fo r instruction. In a recent meta-analysis of
7 Deba te or discuss a topic that was featured in a newspape r ar ticle.

th e effecss of type of in struction on 'co m plex' and 'sim ple' language features,
S Watch an episode of Sesame Street.
however, Spada and Tom ita (20 10) report that explicit instruction promoted
9 Demonstrate and describe the steps in a science experiment.
learningJo r both types o flang uage features.
10 Interview a mystery guest and try to discover his or he r occupation.
Research o n integrated and iso lat ed FFI is a rern inde r that the tim in g o f

II Playa game of 'Simo n Says'.

fo rm- focuse d instructi on may also make a di fferen ce in L2 1earning. Sa rnu da's

stu dy w ith adult ESL learners is a good exam ple of integ rated FFI, illust rat­

12 Work in small groups to choose the ideal candidate for a job.

ing how teac he rs can effec tively di rect students' attention to form w ith in

task-bas ed in struction. The finding th at iso lated and integrated FF I lead to

different kinds ofL2 knowledge is intrigu ing and resonates with th e experi­

Interpreting the research

ence of many teachers . That is, reachers of second/foreign languages know

The overall results of the studies described above provide support for me
that explicit rule -based grammar teaching without co m m un icative practice

hypothesis mat form-focused instruction and co rrective feedback can help

is likely to lead to a fairly good knowledge of the rules of grammar but nor

learners improve the ir knowledge and use of particular grammadcal features.

th e ability to use the rules in meaningful and spontaneous language produc­

There is also co m pellin g evidence that more explicit attention to form is par­
tion. These differences in L2 kn owledge have been vario usly referred to as

ticularly usefu l within communicative and content-based second and foreign

d eclarative versus procedural and learning ve rsus acquisition , as discussed

language programmes. This has been confirmed in reviews and meta-analy­

in Chapter 4, or as explicit versus im plicit kn owledge. Expli cit knowledge is

ses of many studies that ha ve investigated the contribution of form-focused

typ ical ly des cribed as co nscio us and analysed, whereas im plicit knowledge is

instruction to U learning (N o rris and Ortega 2000; Spada 2011 ). Some

co nsidered to be lnruitive and unanalysed.

results also show, however, that the effects of instruction are not always

Some theo rists and researchers claim rha t L2 instruction can lead to exp licit

long-lasting. This may be related to whether there is continued exposure to

knowledge only. Furthermore, they argue tha t the results from meta-analyses

a linguis tic feature in the regular classroom input after the experimental treat­

show ing positive effects for L2 instructio n are due to the fact that the rests used

merit ends .

to assess learn e rs' progress in me majority of studies have measured expl icit

We have also seen that form-focused instruction may be more effective with
knowledge using, for example, disc rete-poim grammar tests. Fo rt unat ely,

some language features than with others. For example, the successful lea rn ­
recent research has included a grea ter variety o f language measures to rap

ing of the tul uous distinction in Lyster's (1994 ) study could be d ue to the
in to learn ers' intuitive L2 knowledge such as oral communication tasks and

fact th at learning tu and uous is essentially a matter of learni ng rwo im po r­

time-p ressured tasks that require learners to retrieve thei r knowledge quickly

tant vocabulary items and thus may have been less d ifficul t to learn th an
without having tim e to 'th ink abou t it'. Nonetheless, we won't have a clear

syntactic features that affect meaning in less obvious ways. In the intensive
answer to th e question ofwhat type ofknowledge resul ts from L2 instruction

ESL research, learn ers may have been more successful after instruction on

194 Second language lea rn ing in the classroom
Second language learning in the classroom
until valid and reliable tests of both im plicit and explicit knowledge are used
in a larger num be r of studies (Ellis er al. 20 09).
Similar issues have been raised about research on corrective feedb ack but the
central focus of th is wo rk has been on investigating whe the r certain types of
cor rective feedb ack are mo re effective than oth ers. The results fro m Lyster's
stu dy in French im me rsion pro grammes suggests that learners benefit more
from feedback th at pushes th em to self-correct (i.e. p ro m pts ) th an from feed­
bac k th at p rovides the correct form (Le. recasts). Research in o ther conte xts,
however, has p roduced diffe rent results . Fo r example, the majo rity ofla bora­
tory st udi es of co rrective feed back report benefits for recasts over o the r types
o f corrective feed back includi ng prompts (Mackey and Goo 20 07). Th ese
conflict ing find ings are likely related to differen ces in context-the labora­
to ry is a mote controlled environment than the classroo m , whe re there are
competing dem ands on lear ners' attention. As a resul t, learners may notice
certain types of feedback in th e one-on-one laborato ry int eractions more
than they do in the classroo m in communicative or COntent-based classes,
where the primary focus is on mean ing.
As d iscussed in Chapter 5, the specifi c pedagog ical activity in which correc­
tive feed back is provided also plays an im po rtan t role in terms o f whe ther
learn ers recognize it as corrective feedback. Th e tim ing o f corrective feed back
may also be impo rtan t in L2 learning. To date little research has explore d
whether it is pre ferable, fo r example, to provide feed back during or afrer
com m un icative pr act ice. O ne study of this issue was carried OUt by James
Hunter (20 12). He investigated the effectiveness of feed back that the teac her
provided after stu dent s ha d par ticipated in st udent-led conversations. His
findings show that such an ap p roach can result in a h igher proportion of
repa ir than feedback provi ded in whole-class teacher-led activities.
Recently there have been a number of meta-analyses of studies investigat­
ing the effectiven ess of L2 corrective feedb ack o n L2 oral prod uction. The
results are m ixed , wi th some reporting benefits for recasts over other types of
feed bac k (Li 2010) and others reporring advantages for p rompts over recasts
(Lyster and Saito 20 10). Until there is greater con sens us o n the co nt rib utions
of different types of corrective feedback on L2 [ear n ing, a prudent approach
would be to provide learners with a var iety of d ifferent types of corrective
feed back and to keep in mi nd the co unterbalanc e hypothesis presente d in
Chapter 5, whic h suggests that more explicit corrective feedback may be
effective in contexts where th e learners' attention is focused on m ean ing /
content while imp licit feed back may be sufficient to at tract learners' atten­
tion in contexts where the focus of ins truction is typically o n language fo rm .
1
195
Assessing the proposals
Although there is still much work to do , it seem s evident that proposals rep­
resenting an almost exclusive focus on form or those representing an almost
exclusive focus on meaning alone can not be recommended. Approaches that
provid e atten tion to form wi thi n communicative and con tent -based int erac ­
tio n receive the most support from classroo m research.
We kn ow that some exceptionally gifi:ed learne rs will succeed in second
lang uage learn ing regardless of the teaching meth od . In the schools of the
world, grammar translation is no dou bt the most widely applied method
an d mo st of us have met individuals whose advanced p roficiency in a fore ign
language develo ped Out o f th eir experience in such classes. Simila rly, audio­
lingual instruction has produced highly p roficient second language speakers.
However , we also know-from personal experience and research findings­
that th ese methods leave many learners frustrated an d unable to participate
in ordi nary con versations, even mer years o f classes. G rammar tra nslation
and audiolingual approaches will continue to be used, but th e evide nce sug­
gests that 'Get it righ t from the beginning' does no t correspo nd to th e way
the majority of successful second language learners have acq uired their pro ­
ficiency. On the other hand, in throwing out co ntrast ive analysis , feedback
on error, and rnetalinguistic exp lanations and gu idance, the 'comm unicative
revoluti on' may have gone toO far.
There is increasing evide nce that learners continue to have difficulty with
basic structures of the language in programmes that offer littl e or no fo rm ­
focused inst ru ctio n. This calls into question extre m e versions of the 'Just
liste n . .. and read ' and ' Get two fo r o ne' pro posals. While there is good
evidence that lear ners make considerable progress in both com prehen­
sion and production in comprehensio n-based programmes, we do not find
suPPOrt for the hypothesis that lan gu age acquisition wi ll take care of itself if
second language learners sim ply focus on meaning in comprehensi ble in put.
Comprehension-based ap proaches are most successfu l wh en they include
guided attention to language features as a co mpo nent of instructio n.
Th e 'Le t's talk' proposal raises sim ila r co ncerns. Opportunities for learners to
engage in conversati onal interactions in group and paired activ ities can lead
to increased comm unicative competence an d the ability to manage conversa­
tio ns in a second lan guage. However, the research also shows that learners m ay
make slow prog ress on acquiring more accurate and sophi sti cated language
if there is no focus on form. This is espe cially tru e in classes where students'
shared langua ge an d learning backgrounds allow the m to communicate
successfully in sp ite of their errors. Because 'Let's talk' em phasizes mean ing
an d attem pt s to simulate 'natu ral' comm un icati o n in co nversational inter­
action, the students' focus is naturally on what they say, not how to say it.
Fu rt he rmore, when feed back on error takes th e form o f recasts , lear ners may

Second language learning in the classroom
196 Second language learning in the classroom
interpret it as a co nt in uatio n o f the conversati on ramer than focus on form.
Th us, programmes based on the 'Le t's talk' approach are incomplete on their
own, and learners' gains in confidence and conversational skills may not be
marched by their development of more accurate and complex language.
Id s important to emphasize that the evidence to su ppo rt a role for for m­
focused instruction and corrective feedback does not suggest a return to the
' G et it right from the beginning' approach. Research has shown that learn­
ers do benefit considerably from communicative interaction and instructi o n
that is meaning-based. The results of research in French immersion , other
co nt ent- based language teac hing, and communicative ESL are strong
indicators that learners develop higher levels of fluency through pr imarily
meaning-based instruction than through rigidl y grammar-based instructi o n.
Th e problem is that certain aspectS oflinguistic knowledge and performance
are not fully developed in such programmes.
Research investigating the 'Teach what is teachab le' proposal is no t yet at a
poi nt where it is possible ro say to teach ers: 'H ere is a list oflin gui stic features
and the o rde r in wh ich th ey will be acq ui red . You shou ld teach them in th is
order'. Th e number of featu res that researche rs have in vestig at ed in experi­
me ntal studies wi thi n th is framewo rk is far to o sm all. O n the othe r hand,
the re has been no stro ng evid ence tha t teaching acco rd ing to th e develop ­
mental sequences is necessary or even desirable or that it will improve the
long-term results in lan guage learning. What is mos t valuable about this pro­
posal is that it serves to help teachers set realistic expectations about th e ways
in which learners' interlanguage may change in response to instruction . The
im p lica tions of 'Teach what is teachable' may be seen primarily in me fact
th at genuine progress in second language development must be measured
in ways that include, but are not limited to , increased accuracy in language
production .
According to th e 'G et it right in tire end' proposal, classroom activities
sho uld be built primarily on creating opportunities for students to exp ress
and understand meaningful language. However, this proposal is based on the
hypothesis that form-focused instruction and corrective feedback are also
essent ial for learners' continued growth and development. The challenge is
to find m e balance between meaning-based and form-focused activities. The
righr balanc e is likely to be different acco rdi ng to th e ch aract er istics of me
learners. The learn ers' age , rnetaling uisric sophistication, prior ed ucational
experiences , rnoriva tion , and goals, as well as the similarity of the target lan ­
guage co a language already known need to be taken into account when
decisions are made about the amount and type of form focus to offer.
Class room data from a n umber of studies offer support fo r the view that
form-foc used instruc tion and cor rec t ive feedback provided within me
co nt ext o f comm uni ca tive and content-base d programmes are more effective
1
19~
in promo ting second lan gu age learn ing tha n programmes that are lim ited
to a virtu ally exclusive emphasis on comprehension , Huen cy, or accuracy
alone. Thus, we would argue mat second lang uage teachers can (and should)
provide guided , form-foc used instruction an d corrective feedback in certain
circu m stances. For example, teachers sh o uld not hesitate to correct persistent
erro rs th at learners seem not to notice wi tho ut focused arren tion. Teach ers
sho uld also be especially aware of errors mat m e majority of learners in a
class are making when they share me same first language background . They
should not hesitate to provide co ntr astive information about how a partic­
ular structure in a learner's first language differs from me target language.
Teachers might also try to become more aware oflanguage features that are
JUS t beginning to emerge in the seco nd language development of their Stu­
dents and provide so me guided instruction in me use of th ese forms . It can
also be useful to encourage learners to take part in the process by creating
activities that draw their attention to th e forms they use in communicarive
activities, by developing contexts in which th ey can provide each other with
feedbac k, and by encou raging th em to ask quesrions about langu age.
De cisio ns about wh en and how to provide form focus must take in to account
d ifferences in learner characte ristics, of co urse . Q u ite d ifferent approaches
would be appropriate for, say, trai ned lingu ists learn ing a fo urth o r fifth
language, yo ung ch ild ren beginnin g their schooling in a seco nd langu age
enviro n ment , both yo un ger and olde r irnrnigrants w ho cannot read and
w rire th eir own language, and adolescents stu dy ing a foreign language for a
few hours a week at schoo l.
Summary
M any teache rs are aware of the need to balance form focus and m eaning
focus , and they may feel that recommendations based on research simply
co nfirm th eir current classroom practice. Although this ma y be true to so me
extent , it is hardly the case that all teachers hav e a clear sense of how best to
acco m plish their goal. It is not always easy to step back from familiar prac­
tices and say, 'I wonder if this is really th e most effective way to go abo u t
th is?' Furthermo re, it can be difficult to try OUt classroom practices that go
against the prevailing trends in their educational contexts. Many teachers still
work in enviro nments where mere is an emphasis on accuracy that virtually
excludes spo ntaneous language use in the classroom. At me same rime, the
introd uction of communicarive language teaching methods has sometimes
resulted in a co m plete rejection of attention to form and error-correction in
second lan guage teaching. But it is not necessary to choose between form­
based and meaning-based instruction. Rather, the challenge is co find the
best balance between these two orientatio ns.
198 Second language Learning in the classroom
Second language Learning in the classroom
Classroom-based research on second language learning and teaching has
given us parcial answers co many questions. Through continuing research
and experience, resear chers and teachers will fill in more details, always rec ­
ognizin g m at no Single answe r wi ll be adequate for all lear n ing environmen ts.
Among the questions we will continue to ask are these:
• How can classroom instructio n provide the righ t balan ce of meaning­
based and form- focused instruction?
• Which features of lan guage w ill respond best co for m-focused instruc­
tion, and which will be acquired without explicit focus if/earners have
adequate access co m e language?
• Which learners will respo nd we ll co rne ral in gui sric informatio n
an d which w ill require so m e other way of focusi ng attentio n on
language form?
• When is it best co draw learners' attention co fo rm-before, aft er, or
during communicative practice?
" How should corrective feedback on lan guage form be offered?
e When sho uld lear ners be allowed [Q focus their attention on th e content
of thei r utt eran ces?
Continued classroom-centred researc h, including me action research by
teachers in their ow n classro o ms , will provide further insights into th ese and
o ther important issues in second langua ge teaching and lear ning.
Que sti ons for reflection
I Keeping in mind that individual lea rner differe nces play an important role in
seco nd language learn ing, do you t hink a particu lar learn er pr ofile might be
more compatible with on e of the teaching proposals than anothe r ?
2 If you were go ing to experiment with a new app roac h to teac hing in your
classroom. wh ich of th e six pr o posals des cribed in this chapt e r woul d yo u
choose?Why?
3 This chapte r co ncludes with the sugges tion that 'Get it right in the e nd' is
the best ap pro ach. Is t his con sist ent wit h your own views ?Why/w hy not?
1
199
Suggestions for further reading
Ellis, R. 2012 . Language Teaching Research and Language Pedagogy. Malden ,
MA: Wiley-Blackwell.
In thi s volume, Ellis focuses on research tha t is designed specifically co
inves tigate the role of teach ing in sec o nd language learning. He review s
bo th the m ethods an d m e find ings from a vast n umber of studies . Ellis
app roaches the task from his d ual perspectives as a resear cher and as edi tor
of me journal Language Teaching Research, which publishes me wo rk of
scholars an d educators who are investigating language teaching around
the world. Throughout the book, links are made be tween me research and
its implications for language pedagogy.
Hedge, T. 2000 . Teaching and Learning in the Language Classroom. Oxford :
Oxford Universiry Press.
Th is is a useful reference book fo r the classroom teacher. It covers a wide
ran ge of topics releva nt to the 'teach in g an d learn ing of second/foreign
lan guages. It is divided into fo ur sections: a framework fo r teaching and
learning, teaching the language system, de veloping the language skills,
planning, and assessing learning. Eac h chapter moves from th eo retical to
p ractical co nsideratio ns an d there is extensive use of tasks , activi ties, an d
teach ing materials to motivate reade rs co reflect o n th e ideas p resented in
relation to their own practice.
Lyster, R. 2007. Learning and Teaching Language through Content:
A Counterbalanced Approach. Amsterd am: John Benjamins.
This bo ok provides a comprehensive description and analysis oflanguage
teaching and lear n ing in co ntent- based classrooms. With a pr imary bur
no t exclusive focus on research in French immersion programmes in
C anad a, Roy Lyster synthesizes decades of empirical work m at has sought
solutions co the challenges of teaching language and content simultane­
ously. The au thor's experience as both teacher and researcher is evident in
the useful connections m ad e be tween theory and practice.
Nation, I. S. P. and]. M a calist er. 20 10. Language Cu rriculu m Design.
N ew York: Routledge.
Paul Nation has proposed four 'strands' mat are seen as essential-and of
equal importance-in a language teaching programme. The four strands
are: m eaning-focused input, m ea n in g-focused o utput, lan guage-focused
learning, an d fluency development. Together they tep resent a balanced
app roa ch ro language teach ing mat is compatible with research o n class­
room learning. O rigin ally base d on Nation's research in vocabulary
learning, me four strands may also be see n as the elements ofa lesson , a syl­
labus, o r even a curriculu m. N atio n (20 07 ) introduces m e ideas, an d th is
book elaborates on the fu ll range of issues related [Q cur ric ul um desi gn.
 POPULAR IDEAS ABOUT

LANGUAGE LEARNING
REVISITED '

Preview
In thi s chapte r, we ret urn to th e 18 statements tha t yo u respo nded to in the
Introduction and summarize some of the related researc h and theory that
we have discussed in thi s book, shari ng so me of our ow n views about these
popular opinions.

ACTIVITY Review your opinions

In the Introduction. we asked you to indicate how strongly you agreed with some
popular ideas about language learning.Before you continue reading this chapter.go
back and comp lete the questionnaire again.Compare the responses you gave then
and thos e you would give now. Have your views about second language acquisition
been changed or confirmed by what you've read in the preceding chapters?

Reflecting on the popular ideas: Learning

from research
1 Languages are learned mainly through imitation
It is difficult to find support for the argument that lan guages are learned
mainly through imi tation, because nrst and second language learners produce
many novel sentences that th ey could no t hav e heard before. These sentences
are based on thei r developi ng under stan ding of how the language system
works. This is evident in ch ild ren's sentences such as 'I'm hiccing up and I
can't stop', and 'It was upside down bur I turned it upside right' , and with
second language learners who say 'Th e cowboy tided into town', or 'Th e ma n
th at I spo ke to him is angry'. These exam p les and many others provide evi­
dence that language learners do more than internalize a large list of imitated
202 Popular ideas about language !earning revisited
and memorized sentences. They also identify patterns in the language and
extend them to new co ntexts.
If we use a narr ow definit ion of imi ta tion (the im mediate repetition of all or
part ofan o th er speaker's utterance) we find that so me ch ild ren im ita te a grea t
deal as they acquire their first language. Even these childre n, however, do not
im itate everything th ey hear. Instead, they selectively imi tate certain wo rds
or structures tha t they are in th e p rocess of lear ning. Fu rtherm ore, child ren
who do little overt imitation lear n language as quickly and as well as tho se
who imi tate more. Thus, this type of imitation may be an individual learning
strategy but it is not a universal ch arac teris tic oflanguage learners.
Some secon d language learne rs also find it useful ro imitate sam ples of th e new
language. Classroom researchers have observe d students who repeat what they
hear others say, and some advan ced lear ners who are determined to improve
their p ronu nciation find it helpful to spend time carefully listening to and imi­
tating language in a language labo ratory or tu torial. However, for beg inni ng
learners, the imitation and rote memorizatio n tha t characterizes audiolingual
approaches to language teaching is not effective iflearners do not also use the
sentences and p hrases they are prac ticing in meaningful interaction. Learners
need ro do more than recite bits ofaccurate language in d rills and dialogues.
Nevertheless, recent findings from corpus linguistics have p rovided a new
appreciation for form ulaic lan guage use. We know from the d iscussion of
usage-based theories d iscussed in C hap ter 4 that a great deal of na tural Ian­
guage use is p redictable on the basis of the frequency with whic h wo rds o r
phrases occur rogerher, Learners create strong associations between lan guage
features that tend to occ ur together. Thus, lang uage is parrly learn ed in chunks
larger than single words. However, this internalization of the input does no t
depend on the learner's imitation of all or part of another person's utt erance
in a rote-repetition fashion. It is the co m bined exposure ro language fearures
in the inp ut and their use in meaningful exchanges th at leads to learning.
2 Parents usually correct young children w hen they m ake
gram.matical errors
There is considerable variation in the extent to which parents correct thei r
children's speech . The variation is based partly on the ch ildren's age and
partly on th e paren ts' social, ling uistic, and educational background. When
chil dre n are very young, pare nts rarely comment on grammatical errors,
although they may correct lapses in poli teness or the choice of a word that
doesn't make sense. As child ren reach school age, parents may co rrect the
kinds of non-standard speech that they hope their children will OUtgrow, fo r
exam ple, 'M e and Fred are go ing outside now'.
Extensive observatio ns of parents and ch ildren show that, as a rule, pare nts
tend to focus on meaning rather than form when they correct chil dren's
Pop ular ideas about language learning revisited 203
spee ch . Thus, th ey may co rrect an incorrect wo rd choic e, an incorrect state­
ment of the facts, or a rude remark, but the y do no t often react ro errors th at
do not interfere with communication. What this tells us is that child ren
cannot depend on consistent corrective feedback in order to learn the basic
structure (th e word order, the grammatical morphemes, the intonation pat­
terns) of th eir language . Fortunately, they ap pear to be ab le to acqui re th e
ad ult fo rm o f the lan guage with little or no explicit feedback.
The case fo r second language learners is more complex. On the one han d ,
both children and adults can acquire a great deal of lang uage without any
formal instr uction or feedback on error. On the other hand, the evidence
suggests that, wi th o ut corrective feedback an d gu idan ce, secon d lang uage
learners m ay persist in using cer tain ungrammatical for ms for years.
3 Highly intelligent people ar e good language learners
Th e kind of intelligence that is measured by IQ tests is often a good pred ictor
ofsuccess in classrooms where the emp hasis is o n learning abou t th e language
(fo r example. grammar ru les). Peop le who do well o n IQ tests ma y do well
o n other kinds of tests as well. However, in narurallanguage learning settings
an d in classrooms where interactive language use is emphasized, research
has shown that lear ners with a wide variety of intelle ctual ab ilities can be
successful language learn ers. This is especially true if the emphasis is on oral
co mmunication skills rather than meralinguistlc knowledge .
Most important , it m ust be recalled that language learning involves a great
ma ny different skills and ab ilities that are no t m easured by IQ tests. Stu dents
sho uld not be excluded from opportunities to lear n another language on
the gro unds tha t the y do not have the academic ab ility to succeed . In man y
educational contexts, students from immigran t or mino rity groups have no
cho ice ab out learning a second lan guage. What is essential is to find ways to
engage the different abilities th at students bring to the learn ing enviro nment.
4 The best predictor ofsuccess in secon d language acquisition
is motivation
Everyone agrees th at learn ers who Want to learn ten d to d o better than those
who don't. But we must not interpret this too rigidly. Sometimes, even highly
moti vated learn ers encounter grea t challenges in language lear ning. We kn ow,
for example, tha t lear ners who begin lear ni ng a second language as ad ults
rarely achieve th e fluency and accuracy that childre n do in first lan guage
acq uisition . Th is sho uld not be taken as evidence that ad ult second lan guage
learners are not motiva ted to learn . It may be a reflectio n o f chan ges that
come with age or of other ind iv id ual d ifferen ces such as language lear ning
ap tit ud e, how the instr uction inte racts with ind ivid ual learn ers' styles an d
pre ferences for learnin g, how m uch tim e the learner can devo te to learn ing
204 Papular ideas ab out language learning revisited
the new language, and what o ppo rt uni ties the learner has to use th e language
outside the classroom.
Teachers have no influence over learners' intrinsic motivation for learning a
seco nd language. Students come [0 classrooms from different backgrounds
a nd life experiences, all of which ha ve contributed [0 their motivation [0
learn and their arrirudes toward the target language and thecomrnuniry with
which it is associated. The principal way that teachers can influence learners'
motivation is by making the classroom a supportive environment in which
stu de nts are stimulated, engaged in activities that are appropriate [0 their age ,
interests, and cultural backgrounds, and, most importantly, where students
can experience success. This in turn can contribute [0 positive motivation .
leading to still greater success. .
5 The earlier a second language is introduced in school
programmes, the gre~ter the likelihood of success in learning
The decision ab out w he n to introduce seco nd o r fore ign la ng uag e in st ru cti on
mus t dep e nd o n th e o b ject ives of th e lan gu age program me in th e parti cu lar
soc ial co ntex t of th e school, W he n the o bjective is nat ive-lik e pe rfor man ce
in the second lan gu age, th en it may be des ira ble to begin expos ure ro the
language as early as possible. as long as lear ners have exte nsive exposure to
a nd opportunities to use th e second lan guage in a var iety of contexts . The
research evid en ce is fairly strong th at those who begin second language learn­
ing at an early age a re most likely to eventually be indistinguishable from
na tive speakers .
However, even in cases where native-like proficiency is targeted, it is impor­
tant to recognize certain disadvantages of an early star t for seco nd language
learning. \Vhen an earl y star t means that children have little opporruniry to
contin ue to develop their first language, the resulting subtractive bilingual­
ism may ha ve lasting negative consequences .
For ch ild ren from minori ry-Ianguage backgrounds, programmes promoting
the development of the first language both at home and at school may be
more important for long-term success in the second language than an early
stare in the seco nd language itself. Research shows that a good foundation in
th e child's first language, including the development of literacy. is a sound
base to build on. Children who can begin their schooling in a language they
already know will have more self-confidence, will be able [0 learn more effec­
tively in the early school years, and will not lose valuable time in a period of
limbo during which the y struggle JUSt to understand what is happening in
the classroom .
For many children, there is no opportunity to have their early schooling in
their first language . They are members of smal l minoriry groups where it is
not practical for schools to offer them an educational p rogramme in thei r
Papular ideas about language learning revisited 205
first lan guage . o r th ey live in jurisdi cti ons whe re legislation has mandat ed a
single language of education tor all children, regardless of their background.
For the se children, it is crucial [0 have sensitive educators wh o respect the
children's difficulty, who encourage parents to maintain the home language,
an d w ho un derstan d that second language learning takes time and effo rt .
Fo r foreign lan guage instruction or for second language instruction where
th e level of proficiency that is targeted is not native-like performance by all
students, the situation is quite different. When the goal of the educational
p ro gram m e is basic co m m un ica tive skill for all students, and where there is a
strong commitment [0 maintaining and developing the child's first language,
it can be more efficient to begin second language teaching later. Older chil­
dren (fo r example, 10-year-olds) are able [0 catch up quickly with those who
began earlier (fo r example, at 6-7 years old) in programmes offering only a
few hours a week ofinstruction . This is especially true if the for eign language
course include s a period of more intensive exposure to the new language.
A.ll scho o l p rog rammes sho uld be base d o n realistic esti mates of how lon g
it takes to lea rn a seco nd lan gu age. O ne o r two hou rs a week-even for
seve n o r eigh t years-will not produce adv anced seco nd lan gua ge speakers.
Th is 'drip-feed' approach ofte n leads to fru stratio n, as learn ers feel that th ey
have bee n studying 'for years' withou t maki ng m uc h progress. Sad ly, th ey are
ofte n righc about thi s.
6 Most of the mistakes that second language learners make are
due to interference from their first language
First, we should recognize that knowledge ofone or more languages can con­
tribute positively to many aspects of second or to reign language learning. If
the languages are relatively close co usins (for example, English and German,
Span ish and French), there is much that learners already ' kn ow'- includ ing
-t he alphabet, cognate words. as well as some basic principles of syntax,
On the other hand , the transfer of panerns from the native language is one
of the major sources of errors in learner language. When errors are caused by
learners' perception of some partial similariry between the first and second
languages, they may be difficult to overcome, especially when learners are
frequently in contact with other learners who make the same errors .
Aspects of the second language that are different from the first language will
not neces sarily be acquired later or with more difficulry than th ose aspects
that are sim ilar. Second language learning is not simply a process of putting
seco nd- language words into first-language sentences. In fact, learners may
not always be able to take advantage of similarities unless they are pointed
o u r to rh ern , Learners can be overl y discriminating, failing to take advantage
ofsim ilarities because they assume, sometimes incorrectly, that the languages
must be different.
206 Popula r ideas about langu ag( l(arning reuisite.
However, me first language is nor me only influence on second language
learning. Learners fro m different backgrounds often make me same ki nds of
errors, an d some of thes e errors are remarkably similar to those made by first
lan guage lear ners. In such cases, second-language erro rs are evidence of me
learners' efforrs to discover m e structu re of the targe t language itself rather
man attem pts to tran sfer patterns from fheir firsr lan guage.
7 The best w ay to learn new vocabulary is through reading
This statement is true but it does not tell me whole Story. Children expand
their vocab ulary dramatically during their school years, and reading is me
major source o f th is growth . Second language learners can also increase their
vocab ulary knowledge th ro ugh reading, bur few seco nd language learners
will read th e amount of target langu age text m ar a ch ild reads in m e co urse of
more m an a d ecad e of schooling.
Researc h evidence suggesrs tha t second language learners benefit from oppor­
tunities to read material that is interesting and important to them. However,
thos e who also receive guidance from instr uc tio n and develop goo d strate­
gies for lear n ing and remembering words will be nefit more m an those who
simply focus on getting the m ain ideas fro m a text. What is per ha ps most
striki ng in me research is the evide nce m at in order to successfully guess me
m eani ngs of new words in a text , a reade r usuall y needs to know more than
90 per cent of me words in mat text.
8 It is essential for learners to be ab le to pronounce all the
individual sounds in the second language
Research o n pronunciation has shown tha t second language speakers' ab ility
to m ake themselves understood depends more on their ab ility to rep roduce
the phrasing and stress patterns-the 'melody' of the language-than on
their ability to ar ticulate each individual sound. Ano ther important em p ha­
sis in current research is the undeniable fact that most languages o f m e wo rld
are spoken in many diffe rent varieties. Thus, it no longer seems appropriate
to insist that learners be tau gh t only one language variety or that only native
speakers of a particular variety are the best teac hers. Rather, learners need to
learn to understand and pro du ce language varieties that will permit them to
engage in communicative inceraction with th e interlocutors they are most
likely to encounter.
9 Once learners know 1,000 words and th e basic struct ure ofa
second language, they can easily participate in conversations
with native speakers
It is true th at most conversational language involves only a relatively limited
nu mber of wo rds and sentence typ es. H owever, learners will find it easier to
Popular ideas about language learning revisited 207
understand and to make themselves und erstood if th ey also have an under­
standing ofso me of the pragmatic feature s ofthe new language. For exam ple,
it is useful for them to focu s their attention on suc h things as how speakers
show respect. apologize, or make requests . Th e cul tural d ifferences in these
types of interactions so metimes lead to comm unication breakdown or mis­
understan d ings, even when the words and the sentence structu res are correct.
10 Teachers should p resen t grammati cal rules one at a ti me, and
learners sho uld practise each one before go ing on to another
Second lan guage learn ing is not simply linear in irs development. Learners
may use a par ticular form accu rately at Stage x (suggesting mat they have
lear ned that fo rm), fail to produce the fo rm (o r make errors whe n they
attempt it) at Stage y, and produce it accurately again at Srage z. The decline
in accuracy at stage y may show that learners are incorporating new informa­
tio n about the language into their incerlanguage. We saw, for example, how
learn ers may ask co rrect fo rm ulaic questions such as 'Wh ar's that? ' or 'How
do you say proche in English? ' an d th en produce questions like 'What you're
doing wit h that?' at a larer rime . Language development is no r just adding
one ru le afte r ano ther, Rather. ir involves processes of incegrating new lan ­
guage for ms an d patterns inco an existing inrerlanguage , readjusting and
resrructuring u nt il all the pieces fir,
Some structure-based approaches to teaching are based on the false assump­
tio n that seco nd language development is an accumulation o f ru les. This can
be seen in the organization of textbooks that introduce a parti cular language
feature in the first unit and reinforce it in several subsequent units , and then
move on th e nex t featu re, with only rare opportunities for lear ners to p ractise
the ones previously raught. This isolated presenta rion an d pr actice of on e
structure at a time does not provide learn ers with an opportunity to discover
how different language features compare and contrast in normal language
use. It is also likely mat, without opporrunities to continue hearin g, seeing,
and using them , the language features learned in the firsc unit will have been
forgotten long befo re the lase.
11 Teachers should teach simple language structures before
complex ones
Research has shown mat no ma tter how language is prese nced to learn ers,
certain structures are acquired before others. Th is suggesrs that it is nei ther
necessary nor desirable to restrict learners' exposure to structures ma t are per ­
ceived in linguistic terms to be 'sim ple'- particularly when this invo lves the
isolated presentation, ordering, and p ractice o f''simple' to 'complex' features .
At the same tim e, there is no do u b t mar second lang uage learners benefit from
th e effo rts of na tive speakers an d fluent bilinguals to modify the ir speech

208 Popular ideas about la ngu age learn ing revisited
[Q help them understand. Th e language used in modified interaction may
contain a variety of lin gu isric structures, some 'sim p le' and some 'com plex'.
However, it also includes a range ofadjustments that enable second language
learners to engage in interactions with native and more advanced speakers
of the second language more easily-mo re repetition. slower rate o f delivery,
paraphrasing, etc .
Teachers must also be aware, however, that some linguistic forms are so rare
in classroom language that learners have little opportunity [Q hea r, use, and
learn them if th e teacher does no t make a point of providing them. These
are not necessarily difficult o r complex forms . As we saw in Chapter 6 . some
'sim ple' language forms turn our to be extremely rare in classroom language.
even in content-based instr uctio n.
12 Learners' errors should be corrected as soon as they are made
in order to prevent the formation of bad habits
Errors are a natural part o fl anguage learnin g. Thi s is true of the development
of a chi ld's first langu age as well as of seco nd langua ge learn in g by ch ild ren
and ad ults . Errors reflect the parterns of learners' develo pi ng interlan gu age
systems- showing ga ps in their knowledge, overgeneralizario n of a second
lang uag e rul e, o r an inap p ro p riate tra nsfer of a first lan gu age pattern to the
second language.
Teachers have a responsibility to help learn ers do their best , and this includes
th e provision o f explicit, form-focused instruction and feedback on error.
When errors are persistent, especially wh en they are shared by almost all stu­
dents in a class, it is important to bring the problem to their attention . This
does nor mean that learn ers should be expe cted to adop t the correct form
or pattern immediately or consistently. If the error reflects a developmental
stage, the instruction or feedback may be useful only when the learn er is ready
for it. It rnay be nece ssary to repeat feedback on the same error many times .
Of cour se, excessive feedback on error can have a negative effect on motiva­
tion; teachers need to be sensi tive to their students' reactions to correction .
The amount and type of co rrec tion that is offered will also vary according to
the specific char acteristics of the students , as well as thei r relationship wi th
the teacher and with each other. Children and adults wi th little ed ucation
in their first language will not benefi t greatly from sophisticated metalin­
guis tic explanations , but university students w ho are advanced learners of
the lan guage may find such explanations of great value. Immediate reaction
to erro rs in an oral communicatio n setting may embarrass some students
and discourage them from speaking wh ile others welcome such correction
as exactly what is needed ro help them notice a persistent error at JUSt the
moment when it occu rs.
Popula r ideas about language {earning revisited 209
13 Teachers should use materials that expose students only to
language structures they have already been ta ught
Such a p roced ure can provide comprehensible inp ut of course, bur-given a
meaningful context -learners can comprehend the general meaning of oral
or written texts that contain vocabulary and structures they ha ve not 'mas­
tered'. Thus, restricting classroom second language materials to those th at
contain little or nothing that is new may have several negative consequences.
There will undoubtedly be a loss of motivation if students are nor sufficiently
challe nged. Students also need to develop str ategies for dealing with 'real ' o r
'auth entic' material if they are eventually going to be prepared for lan guage
use outside the classroom. They do this first with the teacher's guidance and
then independently. Restric ting students to ste p-by-ste p exposu re to the lan­
guage extends their dependen cy.
When a particular language feature is introduced for the first time, or wh en
the teacher feels there is a need for correction of a persistent problem , it
is ap p ro p riare to use narrow-focu s mat erials th at isolate o ne element in a
co nt ext w he re othe r th in gs seem easy. But it wo uld be a disservice to stu ­
de nt s [Q use such mater ials exclu sively or even predo m inantl y. \'{'e shou ld
reme m be r tha r learners who successfully acq uire a seco nd langu age outs ide
classrooms certainly are exposed to a grea t variety of forms and structu res
they have not mastered .
14 When learners are allowed to interact freely (for example, in
group or pair activities) , they copy each other's mistakes
Th e language that learners hear and read serves as input to their language
developm ent . Th e cog nitive processes that allow th em to learn from input
are not 'sh ut down' wh en they are interacting with other learners. Thus, when
learners interact with each other. they ma y provide some incorrect input.
Fur the rmo re, wh en learners come from the same first language background
and are at roughly the same level of pr oficiency, the y are likely to understand
each other very well , eliminating the need for negotiation for meaning that
might lead th em to replace their inrerlanguage pattern s with more target­
like ones . Nevertheless, the benefits of pair and group work far outweigh the
disadvantages, especially if the tasks are properly designed.
If the activities are well designed and learn ers are appropriately matched,
pair and group work provides far more practice in speaking and participating
in conversations than a teacher-centred class ever could. Somewhat surp ris­
ingly, resear ch has s how n that learners do not produce any more errors in
their speech when talking to learn ers at similar levels of proficiency than th ey
do when speaking to learners at mo re advanced levels or [Q native speakers.
The resea rch also shows, however, that learners at similar levels cannot o rd i­
narily provide each other with information that would help to correct those
210 Popula r ideas about language lea rning revisited
errors. No neth eless, tasks can be devised in such a way mar learners working
together can discover nor only how to express or interpret meaning bur also
how to discover the correct panerns in me seco nd language. In order for this
to happen, che casks m ust be carefully planned ro give lear ners access ro m e
new language they need.
Group and pair work is a valuable addition co the variety o f activities that
encourage and pro mote secon d lan guage development. Used in combina­
tion w ith individual work and reacher-centred activities, it plays an essential
role in language teach ing and learning .
15 Students learn what they are taught
Teachers know from experience ma r students don't lear n everything they are
taugh t! Forrunately, learners also learn a great deal that no one ever reaches
them. They are able ro use the ir own internal learn ing abilities to discover many
of the parrerns and assoclations th at underlie the language they are learning. In
this sense, studen ts learn much more man they are explicitly raughr .
Some reaching methods rypically give learners the opportunity to learn only a'
restricted n umber ofwords and senrence rypes. Even when the language reach­
ing method provides much richer lan guage in pur, me fact that someth ing
is taught or made available in the in put does no t mean learners will acquire
it right away. For example, some aspects of me second language emerge and
evolve according ro developmental sequences , and learners may be more likely
to lear n certain lan guage fearures when they are developmentally ready. Thus,
arremprs ro teach aspeCts of language that are roo far away from the learner's
current stage of development can be frustrating. Other language features , for
example, vocabulary, can be taught ar any rime, as long as the learners are in ter­
ested in the opportunity ro learn and the teaching methods are app ropriate co
the learn er's age, interests , needs , experiences, and learning styles.
16 Teachers should respond to students' errors by correctly
rephrasing what they have said rather than by explicitly
pointing out the error
Th is kind of feedback, referred to as ' recasts' has been fou nd to be by far the
most common rype offeedback in second language classrooms. This has been
shown to be true for learners ar different ages and in different instructional
senings- from au diolingual to communicative and co n ten r- based instruc­
tion. A recast has -the advantage o f not imerrupdng the flow of in teraction .
It is seen as Indirect and police, a way ofgivin g students the information th ey
need without em ba rrassing them.
Resear ch in classes with a general focus on grammar an d accurate language
use shows that learners are responsive to th is kin d of feedback. Resear ch in
which learn ers interact individually with interlocuto rs has also shown tha t
Popular ideas about language learning revisited 2 11
recasts are perc eived as corrective feedback, even though learners may not
always know exactly which language features the feedback is focused on.
However, in co ntent-based instruction (fo r example, immersion classes) and
in communicative instruction with younger learners, recasts often appear to
be misinterpreted. Learners seem ro hear them as confirmation of meaning
rather than as correction o fform. In these situations, recasts have been found
to be more effective if the teacher has a method ofsign alling to me stu dent­
rone ofvoice, gesture, or facial exp ression-mat says to me student, 'I think
I understand what you are saying, and I'm showing you how you can say
it better'.
17 Students can learn both language and academic content (for
example, science and history) simultaneously in classes where
the subject matter is taught in their second language
The advantages of conren t-based instruction are num erous. Motivation is
increased when the material that is used for language teachi ng has an inher­
ent value to th e students: it creates a genuine, immediate need to learn the
language . Content-based instruction is also often associa ted with the oppor­
tunity to spend more time in contact with che language, without losing out
on instruction in other subject matter. In addition, the range of vocabulary
and language fearures mar students encounrer in learning academic subjects
is more varied than that which is typically available in second and foreign
lan guage classes.
Research has confirmed m at studen ts in content-based and im mersion
classes develop com p rehensio n skills, vocabulary, and general communica­
tive competence in me new language. Teachers and researchers have also
found, however, that rhe ability to understand me content and to functio n in
classroom interaction does nor ensure mat students w ill continue to improve
in certain aspects of their second language, especially in areas of accuracy
on lan guage fearu res mar do not usually interfere with meaning. Th us, for
example, srudents can spen d years in Frenc h im mersion without achiev­
ing accuracy in marking nouns for gender or verbs for tense. Experimenral
studies in which an element of form-focused instruction was added to the
content-base d instruction have shown mat, with guidance, students can
improve in these areas as well. Born students and teac hers need to keep in
mind mat co nrent-based instruction is also language teac hing.
18 Classrooms are good places to learn about language but not
for learning how to use language
Some structure-based approaches ro lan guage teach ing have ten de d to treat
lang uage as a set o f grammar rules or as separate bits ofinformarion m ar need
to be learn ed before learners can use m e language as a comm u nicative too l.
Other approaches such as communicative language teac hing, co nt en t-based,
212 Popular ideas about language Learning reuisited
and task-based instruction start from the principle that we learn language
by using it to achieve a goal. for example, understanding a story, making
a medical appointment, writing a science report, or joining the fun on the
playground. With this in min d . classroom activities are designed to prepare
students to continue learning outside the classroom. by giving them experi­
ence in language UJeS that are like those they will encounter there.
As we saw in Chapter 6, some theorists argue that second language instruction
can only lead to knowledge about language (explicit knowledge) and question
whether instruction can lead to the ability to use the language spontaneously
and fluently (implicit knowledge) in a wide range ofcommunicative COntexts.
Contemporary approaches to L2 instruction target the development of both
types of L2 ability by ensuring that students use the language in meaningfi.i.l
interaction inside the classroom and that they learn effective strategies for using
the language outside the classroom when they have opportunities to do so.
ACTIVITY Support your opinion
Choose two or three of the 'popular ideas ' that you find especially important.
For each of these. identify and discuss how one or more of the research
studies you have read about in this book has strengthened your agreement!
disagreement with the statement or has led you to change your views.
Conclusion
Knowing more about second language acquisition research will not tell you
what to do in your classroom tomorrow morning. We hope, however, that
this book has provided you with information that encourages you to reflect
on your experience in teaching. We hope, in addition, that this reflection will
co nt ribute to a better understanding ofyour responsibilities as a teacher and
your students' abilities and responsibilities as language learners .
.As we have seen, language learning is affected by many factors. Among these
are the personal characteristics and experiences of the learner. the social and
cultural environment both inside and outside the classroom. the structure of
the native and target languages, opportunities for interaction wi th speakers
of the target language, and access to corrective feedback and form-focused
instruction. It is clear that teachers do not have control over all these factors,
Nevertheless. a better understanding of them will permit teachers and learn­
ers to make the most of the time they spend together in the rwin processes of
teaching and learning a second language.
GLOSSARY
The glossary contains items that have a special or technical meaning in second
language acquisition research and second language teaching. The definitions
are intended to reflect the terms as we use them in this book.
accessibility 'hierarchy: A ranking of relative clauses developed by Keenan
and Comrie (1977). Differem languages use relative clauses to modify
nouns in different grammatical roles . According to the accessibility
hierarchy. for example, most languages allow relative clauses for sentence
subjects, while fewer languages allow them for the object of comparison.
accuracy order: The relative accuracy of grammatical form s in learner
lan gu age. For exam ple, learners are oft en mor e accurat e in usin g plur al -J
s.
than in usin g possessive So me researc he rs have inferred that an accuracy
orde r is eq uivalent to a developmental seq ue nce .
action research: Research carried out by teach ers. o ften in their own
classroo ms or in collaboration with other teachers. The research goals and
questions are local and specific to their own teaching environment.
active lisreningi A teaching technique in which students not only listen
but also show their comprehension by thei r responses.
additive bilingualism: Learning a second language without losing the first.
American Sign Language (ASL): Th e gestural language used by many
North Americans who are deaf or who interact with others who are deaf. It
is a true language, with complex rules of structure and a rich vocabulary, all
expressed through motions of the hands and body.
audiolingual approach: An approach to second or foreign language
teaching that is based on the behaviourist theory of learning and on
structural linguistics, especially the contrastive analysis hypothesis. This
instructional approach emphasizes the formation ofhabits through the
repeti tion, practice, and memorization ofsentence patterns in isolation
from each other and from COntexts of meaningful use.
auditory discrimination: The ability to distinguish language sounds, for
example minimal pairs suc h as ship! sheep.
behaviourism: A psychological theory that all learning, whether verbal or
non-verbal, takes place through the establishment of habits. According w
this view, when learners imitate and repeat the language they hear in their
 1
214 Glossary
surrounding environment and are positively reinforced for doing so, habit
formation (or learning) occurs.
bilingual education:· Schooling in which students receive instruction in
two (or more) languages, usually the ir home lan guage and a second
language.
bilingualism: The ability to use more than one language. The word itself
does not specify the degree of proficiency in either language.
brain imaging: A variety of techniques that allow researchers to observe
and track activity in the brain.
child-d.iceeted speech: The language that caretakers address to children.
In some cases, this language is Simpler than char which is addressed to
ad ults. In some cultures, it is also slower, higher pitched, more repetitive,
and includes a large number of questions.
chunk: A unit oflanguage chat is often perceived or used as a single'unir.
Chunks include formulaic expressions such as Thank you or What's that? but
also bits of language tha t frequently occur together, for example, ice cream
cone or significant difference.
classroom observation scheme: A tool (often in the form of a grid) that
consists ofa set of predetermined categories used to record and describe
teaching and learning behaviours.
cognate: A word in o ne language that comes from the same origin as a
word in another language and has the same meaning, for example, 'nation'
in English and nation in French or uaca and uache (cow) in Spanish and
French. The term false cognate is used to refer to words that may come from
the same origin but have evolved to have different meanings, for example,
librairie (bookstore) in French does not have the same meaning as library in
English.
cognitive: Relating to how the human mind receives, processes, stores, and
retrieves information. The focus is on internal learn ing mechanisms that are
believed to be used for learning in general, not JUSt language learning alone.
cognitive maturity: The ability to engage in problem-solving, deduction,
and complex memory tasks.
collaborative dialogue: A conversation between learners in which they
wo rk together to solve a problem, for example, reconstructing a story they
have heard. While the focus is on the task, learners may also focus on the
elements oflanguage that rhey need to complere the rask.
communicative competence: The ability to use language in a variety of
settings, taking into account relationships between speakers and differences
Glossary 215
in situations. The term has sometimes been interpreted as the abili ty to
convey messages in spire of a lack of grammatical accuracy.
communicative language teaching (CLT): CLT is based on the premise
that successful language learning involves not only a knowledge of the
structures and forms of a language, but also the functions and purposes tha t
a language serves in different communicative settings. This approach to
teaching emphasizes the communication of meaning in interaction rather
than the prac tice and manipulation of grammatical forms in isolatio n.
competence: Linguist Noam Chomsky used this term to refer to
knowledge oflanguage. This is contrasted with performance, which is the
way a person actually uses language- whether for speaking, listening,
reading, or writing. Because we cannot observe competence directly, we
have to infer its nature from performance.
comprehensible input: A term introduced by Stephen Krashen to refer to
language that a learner can understand. It may be comprehensible in part
because ofgestures, contextual information, or prior knowledge/experience.
comprehensible output hypothesis: The hypothesis that successful
second language acquisition depends on learners producing language (oral
or written). Swain (1985) proposed this hypothesis in response to Krashen's
( 1985) comprehensible input hypothesis.
comprehension-based instruction: A general term to describe a variety of
second language programmes in which the focus of instr uction is on
comprehension rather than production.
connectionism: A theory of knowledge (including language) as a complex
system of units that become interconnected in the mind as they are
encountered together. The more often units are heard or seen together, the
more likely it is that the presence of one will lead to the activation of
the other.
content and language-integrated learning (CLIL): An approach to
content-based language teach ing that has been developed primarily in
secondary schools in Europe.
content-based language teaching (CBLT): Second language instruction
in which lessons are organized around subject matter rather than language
points. For example, in immersion p rogrammes, students study science ,
history, mathematics, etc. in their second language.
contrastive analysis hypothesis (CAH): The expectation that lear ners will
have less difficulty acquiring target language patterns that are similar to
those of the first language than th ose that are differe nt.
control group: In experimental studies , a group of/earners th at differs
fro m the experimental group on ly in terms of the single variable that the
216 Glossary
researcher is investigating. Performance of me control gro up is used to show
that the variable in question is the best (o r only) explanation for changes in
the experimental group. Also so meti m es referred to as 'comparison group' .
corpus (plural: co rpo ra): A principled collection of oral or w rit ten
language samples mat can usually be accessed and explored with computer­
based tools. Some of me rno sr fam o us corpora contain millions of wo rds
from, for exam ple, newspapers. Samples of the language produced by
learners have also been collected for second language acquisition research.
corpus linguistics: An approach to the srudy oflanguage that is based on
the analysis of language corpora. See corpus .
corrective feedback: An indication to a learner m at his or her use of th e
tar get language is in correct.
co r rela ti o n s A sta tistical procedure that compares the relative frequen cy o r
size of different variables in order to determine whether there is a
relatio nsh ip between them .
counterbalance hyp o thesis: Th e hypothesis that lea rn er s' attention wi ll
be drawn to classro om even ts tha t are d ifferent from those they are
accustomed to .
critical period hypothesis (CPH) : The p ro posal th at th ere is a limited
pe riod during w hic h language acq u isition can oc cur.
cross-linguistic influence: The effect on kn owledge of o ne language by
the kn owl edge o f ano th er. This term is pr eferred over previous terms suc h as
interference to indicate mat knowledge of one language can be beneficial to
learning another. The term also reflects the fact mat me influence can go
fro m a known language to th e o ne being learned bu t also from the new
language to one already known .
cross-sectional study: A study in which parti cipan ts at di fferent ages and/or
stages of development are studied . This co n trasts wi th longirudinal studies,
declarative knowledge: Informati on th at we h ave and kn ow we hav e. An
exam p le would be a rule such as 'th e verb must agre e with me subject ro
form a correct sentence'. In some skill learning theories, it has been
hypothesized that all learning begins with declarative knowledge. This
contrasts with procedural knowledge.
descriptive study: Research mat does not involveany manipulation,
cha nge, o r intervention in me phenomenon being stud ied . The researcher's
goal is ro observe and record what is happening. This co nt rasts with
experimental sru d y.
Glossary 2 17
developmental features: Th ose aspects o f a language which , acco rding ro
Pienemann and his colleagues, develop in a particular sequence, regardless
of input variation , learner motivation , or instructional intervention .
developmental sequence: The order in which cert ain fearures of a
lan guage (fo r example, negation) are acquired in language learning. Also
called developmental stages o r order o f acquisition.
display question: A question to which the asker already knows me an swe r.
Teachers often ask these questions (fo r example, What colou r is your shirt?)
to get the learner to display his or her knowledge of the language.
enhanced input: Input mat is altered in an effort to make some language
featu res more salient to learners. It can be more or less explicit, ranging
fro m explicit rneralinguistic co m m ents to ty pograph ical enh anc eme nt
(bold type or underlining) or exaggerated stress in speaking.
ethnography: D escriptive research in which the ob server seeks to
understan d a group or co m m u nity fro m wi thi n its own perspect ive. The
research requi res extensive per iods of observatio n as well as co nsu ltation
with group m em bers to validate the observer's desc riptions .
exp e rimental study: Research des igned to test a hypo thes is abo ut the
im pact of one o r mo re spec ific var iables o n ano the r var iab le. A stric tly
expe rime nt al scudywo uld have 'experim ental' and 'co ntro l' gro ups th at
di ffer from each o the r only in th e pres ence or absence of th e variable(s) of
inte rest. In ed ucatio nal research, it is often difficult to create all of me
co ndi tio ns m at permit a study to be termed as a 'gen uine' expe rime nt al
study. In this book, me term is used in a non-technical sense to refer to
research in which an anempt has been made to investigate a sin gle varia ble
in an educational setting.
field independent/field dependent: This distinction has been used to
des cribe people who differ in their tenden cy to see th e for est o r th e trees.
That is, so me people (calle d field independent) are very quick to pick o ur
the hidden figure s in a complicated drawing. Others (called field
dependent) are more inclined to see the whole drawing and have difficulty
separating it in to parts.
first language (Ll , mother tongue, native language): The language first
learned. Many children learn more man one language from birth and may
be said to have more than one 'first' language.
foreigner talk: The modified or simplified language that some native
speak ers address to seco nd language learners. A special category of foreigner
talk is teacher talk .
foreign language learning: This refers to the learning of a language,
usually in a classroom setting, in a context where the target language is not
218 Glossary
widely used in the community (for example, lear ning French in China).
This is sometimes contrasted with 'second language learning', where the
language being learned is used in the community (for example, learning
Italian in Florence).
form-focused instruction: Instruction that draws attention to the forms
and structures of the lan guage within the context of communicative
interaction. This may be done by giving metalinguistic information, simply
highligh ting the form in question, or by providing corrective feedback.
formulaic: Expressions or phrases that are often perceived and learned as
unanalysed wholes . For example, a child or second language learner may
first hear 'What's that?' as a single uni t oflanguage rather than as three
units.
fossilization: This term is used to describe a persistent lack of change in
interlanguage patterns, even after extended exposure to or instruction in
the target language.
function words: Words that are used mainly as linking or supporting
words for nouns, verbs, adj ectives, and adverbs. For example, prepositions
('to' , 'for', 'by') and articles ('a', 'the') are two types of function words. They
have litt le or no meaning when they occur alone, bu t they have an
important effect on the meanings of the words they accompany.
generalization: Extending a pattern learned in one context to another
one. See also overgeneralization.
genuine question: A question to which the asker does not know the
answer in advance (for example, What didyou do fast weekend?). Also called
'referential' or 'informatio n' questions. Cont rasts with display question .
grammar translation: An approach to second language teaching
char acterized 9Ythe explicit teaching of grammar rules and the use of
translation exercises.
grammatical morphemes: Morphemes are the smallest units oflanguage
that carry meaning. A simple word is a morpheme (for example, 'book'),
but when we talk about 'gram matical morphemes' we are usually referring
to smaller units that are added to words to alter their meaning (for example,
the -s in 'books' indicates plural) or function words (fo r example, the) which
are ordinarily attached to another word.
grammaticality judgement: A test or task in which participants are asked
to make a decision about whether a sentence is grammatically correct or not.
hypothesis (plural: hypo theses): A sta teme nt of a possible fact tha t can be
tested th rough research . Most empirical research starts from one or mo re
hypotheses and involves the design of a stu dy that can either show support
for the hypo thesis or disprove it.
Glossary 2 19
immersion programme: An ed uca tio nal programme in which a second
language is taught via co nte nt- based instruction. That is, students stu dy
subjects such as mathematics and social stu dies in their second language.
Typicall y, students in imm ersio n programmes shar e the same first lan guage
and teachers adjust their instructional language and materials to meet the
needs of second language learners.
information processing: A psychological theory based on the idea that
learners' cog nitive resources are limited and they can't pay attention to
everything at the sam e time. But with repeated experience and practice,
things which at first required attention become auto matic, leaving more
atte ntion available fo r focus on something else.
innatist: In language acquisition research, this is the theoretical perspective
based on the hypothesis that human beings are born with mental structures
that are designed specifically for the acquisition oflanguage.
input: The language that th e learne r is exposed to.(either written or
spoken) in the environment:
input flood: A technique for pro viding a large number of examples of a
particular language feature in the input to learners. It has been used in
research projects to explo re q uest ions related to whether comprehensible
input is sufficient for lan guage acquisition.
instrumental motivation: Motivation that is essentially prac tical, such as
the need to learn the language in order to get a better job.
integrative motivation: Motivation for second language learning that is
based on a desire to know more about the cul ture and co mm unity of the
target language group and even a desire to be more like members of that
group.
intensive ESL: In thi s book, 'i n tensive' ESL is used to refer to an
instructional approach in Quebec where lO-12-year-old French-speaking
students learn English as a second language. Most Quebec Students in this
age group have onl y an hour or two ofESL instruction each week.
'Intensive ESr.: classes provide much more time. Most of the classes
observed in intensive ESL research set aside one five-month block of time
in one school year and devot e full days to ESL instruction during th at
period. The pedagogical approach observed in these classes is
p redominan tly communicative language teaching. In contras t to
immersion programmes, intensive ESL classes do no t usually include
content-based langu age teaching.
interaction hypothesis: The hypothesis that language acquisitio n is based
both on learners' innate abilities an d on opportunities to engage in
conversations, often those in which other speakers modify their speech and
220 Glossary
their interaction parcerns {Q match the learners' com m u n ication
requirements. The innate abilities are not seen as being specific to language
or language acquisition.
interlanguage: A learner's developing second language knowledge. It may
have characteristics of the learner's first language, characteristics of the
seco nd language, and some characteristics that seem ro be very general and
tend ro occur in all or most interlanguage systems. Inrerlanguages are
systematic, but they are also dynamic. They change as learners receive more
input and revise their hypotheses about the second language.
interlocutor: A participan t in a conversation.
language acquisition/language learning: In this book, these two terms are
most often used interchangeably. However, for some researchers, most
notably Stephen Krashen acquisition represents 'unconscious' internalization
oflanguagt knowledgt, which takes place when attention is focused on
meaning rather than language form, and IMming is described as a 'conscious'
process th at occ u rs wh en th e learner's objective is {Q learn abou t the langu age
itself rathe r tha n ro un de rsta nd messages co nveyed thro ug h the language.'
langu ag e-related episodes (LREs): Parts of co nve rsatio nal interactions in
wh ich lang uage learners talk abo ut the lan gu age forms th ey are usin g and
engage in self- and peer -correction .
longitudinal study: A study in which the sam e learners are studied over a
period of time. This contrasts with a cross-sectional study.
meaning-based instruction: See communicative language teach ing.
meta-analysis (plural: meta-analyses): A statistical procedure that allows
researchers ro combine the findings from a large number of quantitative
studies in order ro assess the overal l parcerns of findings on a similar ropic .
metalinguistic awareness: The ability {Q trea t language as an object, for
exam ple, being able to define a word, or {Q say what sounds make up that
word .
mitigation: In pragmatics, a phrase or tone of voice used {Q reduce or
so ften the possible negative impact of what is said.
modified input: Adapted speech that adults use to address children and
nati ve speakers use ro address language learners so that they will be able ro
understand. Examples of modified input include shorter, simpler sentences,
and basic vocabulary.
modified interaction: Adapted conversation patterns that proficient
speakers use in addressing language learners so that the learner will be able
{Q understand. Examples of interactional modifications include
co mprehensio n checks, clarification requests, and self-repetitions.
Glossary 221
morpheme: See gram ma tical morphem es.
native-like: The ability to comprehend and produce a second language at a
level of performance that is indistinguishable from that of a native speaker.
native speaker: A person who has learned a language from an early age and
who is deemed ro be fully proficient in that lan guage. Na tive speake rs differ
in terms ofvocabulary and stylistic aspects oflanguage use, but they tend ro
agree on the basic grammar of the language. The notion 'native speaker'
must always be undersrood within a specific geographic region or
socioeco no mic group because there is wide variation among 'native
speakers' of most languages. .
natural order: See developmental sequence.
negotiation of form: An interaction in which language learners work
toward the correct form in a context where meaning is undersrood. If a
teacher is involved in the interaction, he or she seeks ro guide students {Q
Rnd the right form inst ead o f pro viding it for them.
negotiati o n fo r meaning: Int eractio n betwee n spea kers who make
adj ustments {Q th eir speec h and use other tech niques ro repair a breakdown
in co mm unication. See also mo dified interaction.
noticing hypothesis: Th e hypo thes is, pr op osed by Rich ard Schmi dt. th at
language learners learn only that which th ey have first 'no ticed' or become
aware of in the input.
obligatory contexts: Places in a senten ce where a particular grammatical
form is required if the sentence is to be correct. For example, in the sentence
' Last week, my brother rent a car', the speaker has created an obligatory
COntext for the past tense by the use of 'Last week', but has not used the
required form of the verb in that context.
order of acquisition: See developmental seq ue nce.
overgeneralization: This rype of error is th e result of trying to use a rule or
patt ern in a context where it doe s not belong, for exam ple. putting a regular
-ed ending on an irregular verb , as in 'buyed ' instead of ,bought' .
pattern practice drill: A teaching technique in which learners are asked to
practise sentences chosen to represent particular linguistic forms. Typical of
the audiolingual approach.
performance: The way we use language in listening, speaking, reading,
writing. Performance is usually contrasted with co m petence, which is the
knowledge that underlies our ability to use language. Performance is subject
ro variations due to inattention or fatigue whereas competence, at [east for
the mature native speaker, is more stable.
222 Glossary
phonemic: Small differences in language sounds that can change meaning
within a particular language. For exam ple, the consonants p and b are
phonemic in English, but not in Arabic.
pragmatics: Aspects oflanguage use that go beyond vocabulary and
grammar to includ e rules of how to use lan guage appropriately in different
contexts and with different speake rs. It also includes an understanding of
the implied as well as th e explici t meaning oflanguage.
private speech: Th e language we use when we are talki ng to ou rselves, no t
expecting anyone to hear or respond.
procedural knowledge: Knowledge that underlies fluent or automatic
performance. It is contrasted with d eclarative kn owledge.
processahility theory: M anfred Pienernann's theory oflearners' developing
ability to process linguistic elements in different sent ence posi tions.
processing instruction: An approach to instruction in which learners are
given explicit in formation abo ut the language feature co be learned and
th eir pr actice activities involve the co mprehension (no t production) of
sentences or texts that cannot be understood without a focus on th e
language itself. The app roach was developed by Bill VanPatten.
qualitative research: An approach tha t uses detailed descriptions of the
phenomena being studied rather than counting or measuring the exact
amount of some specific variable or variables. Q ualitative research requires
extensive o bservatio n and ins ightful interpretation.
quantitative research: An approach that requ ires precise counts or
numeric measurements of variables. In a quantitative study, both the
variable that is believed to affect learning and the learning itself are
measured or 'q uan tified' . Quantitative research requires careful selection of
the measurem ents that will be used to repre sent the variables being studied.
rate oflearning: The speed with which learners progress in th eir language
development.
recast: To repea t a learner's incorrect utte rance , makin g ch anges th at convert
it to a correct phrase or sentence. ' Recast' is also used as a noun, that is, a
recast is the interlocutor's modified!corre cted form of the learner's utterance.
register: A style or way of using language that is typ ical o f or approp riate
for a particular setti ng. For example, speaki ng and writing usually require
d ifferent registers; the register used in wri tin g a research report is differe nt
fro m tha t used in writing a letter co a friend .
restructuring: Cognitive activity that is seen as causing changes in the way
information is organized in the brai n, even though no new inform atio n has
bee n learned.
Glossary 223
scaffolding: Th e language th at an interlocu to r uses co suppo rt the
communicative success of another speaker. It ma y includ e the provision o f
missing vocabulary or the expansion o f the speak er's incomplete sentence.
second language (L2): In this boo k, the term refers to any language other
than th e first language learned. Th us, it may actually refer to th e third or
fourth language.
segmental: The individual sounds of a language. Contrasted wi th
'sup rasegrnen rals', which are patte rn s of intonation.
significant difference: Th..is is a tech nical term that refers co differences
between groups which, according co a variety ofstatisti cal tests, are unlikely
to have happened by chance. Such diffe ren ces can be sm all or large. Their
'significan ce' is due co th e consistency of the di fferences as well as their size.
simplification: Leaving out elements of a sentence, for example, using the
same form of a verb regardless of person, number, tense ('I go today. H e go
yesterd ay') .
sociocultural theory: An explana tion for knowledge and learn ing th at is
based on the assumption th at all learning is first social then individ ual.
Learning is viewed as a process that is socially mediated, th at is, it is
depen dent on dialogue in face-co-face interaction . The claim is that during
communication, learners join tly construct kn owledge which is intern alized
by the individual.
standard variety: The variety o f a given lan gu age that is typically used in
formal writing and formal public speaking (including broadcasting) . Th e
standard variety ofwidely spoken languages may be different in d ifferent
places. For example, American English, British English, Can ad ian English ,
an d Indian English each has its own standard variety, as well as numerous
eth nic, regional, and socioeco nom ic varieties .
structural grading: A technique for organizing or sequencing material in a
textbook or lessons . The basis for the organization is a gradual increase in
co mplexity of grammatical features.
subtractive bilingualism: Partially or complete ly losing the first language
as a second lan gu age is acquired .
suprasegmentals: The sounds of a language that involve the melody an d
rhythm of the language (e.g, stress an d intonation), rath er than the
pronunciation ofindividual sounds.
target language: The lan guage being learned, whether it is the first
language or a second (or th ird o r fourth) language.
task-based language teaching (TBLT): Instr uct ion in which classroom
activities are 'tasks' sim ilar to th ose lear ners mi gh t engage in outside the
 GIoJSary 225
224 Glossary

second or foreign language classroom. Tasks may be complex, for example,
creating a school newspaper. or more limited, for example, making a phone
call to reserve a train ticker.
teacher talk: See modified input and foreigner talk.
transfer: The influence ora learner's first language knowledge in the
second language. Also called 'interference' . The term 'cross-linguistic
influence' is now preferred by many researchers. It better reflects the
complex ways in which knowledge of the first language may affecr learners'
knowledge and use of another.
facto rs co nt ribute to wrc including social, individ ual, situatio nal, and
mo rivarionai.
working memory (WNf) : The cognitive 'space' in which we actively
process new information or information that is currencly in focus. Also
called 'short-term memory' .
zone of proximal development (ZPD): The metaphorical 'place' in which
a learner is capable of a higher level of performance because there is suPPOrt
from interaction with an interlocutor. In Vygotsky's theory, learning takes
place through and during interaction in the learner's ZPD.

transfer-appropriate processing (TAP): Cognitive psychologists have

observed that when we learn something new, we also internalize some aspecrs
of the circumstances and thinking processes that were present when we
learned it. The TAP hypothesis is that knowledge is easier to retrieve ifwe are
returned to or can recreate those circumstances and thinking processes.
universal grammar (VG): Innate linguistic knowledge which, it is
hy po thesized, consists of a set of principles common co all languages. Thi s
ter m is associated with C homsky's theory o f language acq uisitio n.
uptake: This term is sometimes used in a general sense to refer to what a
learner notices and /or retain s in second language input or instruction.
Lyster and Ranta's ( 1997) definition refers co a learner's observable
imm ed iate response to corrective feedback on his/her utterances .
variable: An element or characteristic that can be measured or defined.
Variables can differ in different groups or change over time within a group
or individual. Some examples ofvariables that are commonly examined in
language acquisition research include the amount of time a person has been
learning the language, scores on aptitude rests, and performance on
measures oflanguage knowledge.
variational features: In contrast to the developmental features in the
framework developed by Pienemann and his colleagues, variational features
(for example, vocabulary, some grammatical morphemes) can be learned at
any point in the learner's development.
variety: A way of speaking and using language that is typical of a particular
regional, socioeconomic, or ethnic group. The term 'dialect' is sometimes
used. Some language varieties are st igmatized as 'uneducated' but each
language variety has its own rules and patterns that are as complex and
systematic as those of the so-ca lled 'standard ' language. Among the most
studied non-standard varieties of English are British Cockney and African­
American Vernacular English .
wiUingness to communicate (wrC): The predisposition oflearners
toward or away from communicating in a second/foreign language. Several
BIBLIOGRAPHY

Th is list b rings togeth er all the wo rks cited in th e text as well as th ose that
have been suggested for further read ing.
Altenberg, E. P. 2005 . 'The judgment, perception , and production of
consonant clusters in a second language.' International Review 0/.Applied
Linguistics 43/ 1: 53-80.
Ammar,A. and P. M. Light bown. 2005 . 'Teach ing marked linguistic
strucru res-s-rnore about the acquisition of relative cla uses by Arab lear ners
of English' in A. Housen and M . Pierrard (eds.): Investigations in Instru cted
Second Language A cquisition. Amsterdam: Mouto n de Gruyte r. pp . 167- 98.
Anderson, J. R. 1995. Learning and Memory: An Integrated Approach.
New York: W iley.
August, D. and T. Shanahan (eds.) . 2008 . Developing Reading and Writing
in Second-Language Learn ers. N ew York: Rou tled ge.
Bardovi-Harlig, K. 1999 . 'Exp [0 ri ng the i nte rlan guage 0 f interlanguage
pragmatics: A research agenda for acquisitional pragmatics.' Language
Learn ing49/4: 677- 7 13.
Bardovi-Harlig, K. 2000. TenseandAspect in Second Language Acquisition :
Form, Meaning, and Use. O xfo rd : Blackwell.
Bardovi-Harlig, K. and B. S. Hartford. 1993 . 'Learn ing the ru les of
academic talk: A longitudinal study of pragmatic development. Studies in
Second Language Acquisition 15/3 : 279-304.
Barlow, M. 2005. 'Computer-based analyses oflearner lang uage' in R. Ellis
and G . Barkhuizen (eds.). Analysing Learner Language. O xford : O xford
University Press. pp . 335-57.
Bates, E. A. an d B. MacWhinney. 1981. 'Second-language acquisition
from a fu nctio nalist per spective: Pragmatic, semantic, and perceptual
strategies ' in H. W initz (ed.): Native Language and Foreign Language
Acquisition. Annals o/ the New York Academy of Science. N ew York: N ew York
Academy o f Science, Vol. 379: 190- 2 14.
Bennet, G. R. 20 10. Using Corpora in the Language Learning
Classroom: Corpus Linguisticsfor Teachers. Ann Arbor, MI: U nive rsity
of Mi chigan Press.
228 Bibliography Bibliography 229

Beretta, A. 20 11. 'Th e language learning brain' in M . Long and Celce-Murcia, M., D. M., Brinton, and J. M. Goodwin. 1996 . Teaching
C. Doughty (eds.): The Handbook o/Language Teaching, Oxford: Wiley­ Pronun ciation: A Reference[or Teachers 0/English to Speakers 0/Other
Blackwell. pp . 65-80 ,
Languages. Cambridge: Cambridge University Press.
Berko Gleason,]. 'The child's learning of English morphology.' m"rd 14,
Chomsky, N. 1959. ' Review of"Verbal Behavior" by B. F. Skinner.'
150-77.
Language35 /1: 26-58.
Berko Gleason,]. and N. Bernstein-Ratner (eds .). 2009. The Development Clement, R., S. C. Baker, and P. D. MacIntyre. 2003. ' W illingness to
ofLanguage 7 th edn. New York: Allyn and Bacon.
communicate in a second language: The effects of context, norms, and
vitality.' Journal o/Language and Social Psychology 22/2 : 190-209.
Bialystok, E. 2001. Bilingualism in Development: Language, Literacy, and

Cogn ition. Cambridge: Cambridge University Press.

Collier, V. P. 1989. ' H ow long? A synthesis of research on academic
achievement in a second language.' TESOL Quarterly 23/3: 509-31 .
BialystokE. 2009. 'Effects of bilingualism on cognitive and linguistic
pe rformance across th e lifespan' in 1. Gogolin and U. Neumann (eds. ): Collins Cobuild Concordance and Collocation Sampler
Streitfal! Zioeisprachigkeit--« The Bilingualism Controversy. Wiesbaden, http://www.collins.co. uk! corpusICorpus.Search.aspx
G ermany: VS Verlag fur Sozialwissenschaft:en. pp. 53-68.
Collins, L. 20 02. 'The roles ofLl influence and lexical aspect in the
Bley-Vroman, R. 1990 . 'Th e logical problem offoreign language learning.' acquisition of temporal morphology.' Languag« Learn ing 52 /1: 43 - 94.
linguistic Analysis 2011- 2: 3-47.
C o ll in s, L. and J. White. 20 11. 'An int ensive look at in rensiry an d
Blo o m , L. 1991 . Language Development from Two to Three. Cambridge:
language learn ing.' TESOL Quarterly45/ 1: 106-33.
Camb ridge Univers ity Press.
Co llins, L., R. H. Halter, P. M. Lightbown, and N . Sp ad a. 1999.
Bloom, L. and M. Lahey. 1978. Language Development and Languag«
'T ime and the distribution of tim e in L2 instructi on .' TES OL Quarterly
Di sorders. N ew York: John Wiley. 33/4: 655 -80 .
Bongaerts, T. 1999 . ' U ltim ate attainment in L2 pronunciation: The case Conboy, B. D. and P. K. Kuhl. 2011 . ' Im pact of second-language
of very advanced late L2learners' in D. Birdsong (ed .): Second Language experience in infancy: brain measures of first- and second-language speech
A cquisition and the Critical Period Hypothesis. Mahwah , NJ : Lawrence perception.' Developmental Science, 14/2: 24 2-48 .
Erlbaum and Associates . pp . 133-59. Cook, V. 200 3. 'The poverty-of-the-stimulus argum ent and structu re­
Brooks, N. 1960 . Languag« and Language Learning. New York: dependency in L2 users of English .' International Review 0/ Applied
Harcourt Brace .
Linguistics 41/3: 201 -21.
Brown, R. 1973 . A First Language: The Early Stages. C amb ridge, MA:
Cook, V. 2008. Second Language Learn ing an d Language Teaching 4 th edn .
Harvard University Press.
London: Hodder Education.
Burns, A. 20 10 . Doing Action Research in English Language Teaching: A
Corder, S. P. 196 7. 'Th e significance of learners' errors.' Intern ation al
Guide [or Practitioners. New York: Routledge.
Review o/Applied Linguistics 5/ 2-3 : 161-69 .
Burstall, C. 1975 . ' French in the primary school: The British experiment:
Crago, M . 1992. 'Co m m u nica tive interaction and second language
Canadian Modern Languag« Reoieio 31/5: 388--402 . acquisition: An Inuit example.' TESOL Quarterly 26/3 : 487-505 .
Carroll, J. 1991. 'Cognitive abilities in foreign language aptitude: Cummins, J. 1984. Bilingualism and Special Education: Issues in Assessment
Then and now' in T. Parry and C. Stansfield (eds.) : Language Aptitude and Pedagogy. Clevedon: M ultilingual Matters.
Reconsidered. Englewood Cliffs, NJ : Prentice hall.
Cummins, J. 2000. Language, Power and Pedagogy: Bilingual Children in
Carroll, J. and S. Sapon. 1959 . The Modern LanguagesAptitude Test. the Crossfire. Clevedon: Multilingual Matters.
San Antonio, TX: Psychological Corporation .
Curtiss, S. J97 7. Genie: A Psycbolinguistic Study ofa Modern-Day
Wild Child. New York: Academic Press.
230 Bibliography Bibliography 23 1

Dalton-Puffer, C . 2006. 'Q uestio ns in CLIL classrooms: Strategic Dornyei, Z. 2009 . The PsychologyofSecond Language Acquisition. Oxford :
questioning [Q encourage speaking' in A. Martinez-Flor and E. Uso (eds.): Oxford Un iversity Press.
New Perspectiveson Teaching the Language ProcessingSkills. Ams terdam:
D ornyei, Z. and P. Sk eh an . 2003. 'Individ ual differences in second
Mo uto n de Gruyrer, pp. 187-213.
lan guage learn ing' in C. ] . Doughty an d M . H. Long (eds.): The Handbook
Day, R,J. Bassett, B. Bowler, S. Parminter, N . Bullard, N. Prentice, ofSecond Language A cquisition. Oxford: Blackwell. pp. 589 -630.
M. Furr, M. Mahmood, D. Stewart, an d T. Robb. 20 11. Bringi ng
Doughty, C. 199 1. 'Second language instructio n does mak e a difference:
Extensive Read ing into the Classroom. O xford : Oxford Un iversity Press.
Evide nce fro m an em pirical study ofS L relativizarion.' Studies in Second
DeKeyser, R M. 1998 . ' Beyo nd focus on form: Cognitive perspectives on Language Acquisition 13/4: 43 1-69.
learning and practicing second language grammar' in C. ]. Doughty and
D oughty, C. and E. Varela. 1998. 'Communicative focus on form' in
]. W illiams (eds.): Focus on Form in Classroom Second Language Acquisition.
C. Dough ry and] . Will iams (eds.): Focus on Form in Classroom SLA .
Camb ridge: Cam bridge U niversity Press. pp. 42-63.
Cambridge: C ambridge University Press. pp. 114-38.
DeKeyser, R. M . 2000. 'The robustness of critical period effects in
Duff, P. 1995. 'An ethnography of comm unica tion in immers ion
secon d lan guage acquisitio n.' Studies in Second Language Acquisition 22/4 :
classrooms in Hungary.' TESOL Quarterly 29 /3: 505-37 .
493-5 33.
Dunn, W. E. and ]. P. Lantolf. 1998. 'Vygo rsky's Zone of Proximal
DeKeyser, R M. 2001 . 'Aurornariciry and au tomatization' in P. Rob inson
Developmen t and Krashen's i+1: Incom m ensur able const ructs; .
(ed.): Cogn ition and Second Language Instruction. Cambridge: Cambridge
incom mens urable theo ries.' Language Learning 48/3 : 4 11-42.
University Press. pp. 12 5- 5 1.
Echevarria, J ., M. Vogt, and D. J . Short. 2004. Making Content
DeKeyse r, R M . 2007. 'Introducti on: Situ ating the concep t of prac tice' in
Comprehensiblefor English Language Learners: The SlOP Model. Bosron :
R. DeKeyser (ed.): Practice in a Second Language: Perspectives from Applied
Pearson .
Lingu isticsand Cogn iti ve Psychology. Camb ridge: Cambridge University
Press. pp. 1-18. Eckman, E, L. BeU, and D. Nelson. 1988. ' O n the generalizatio n
of relative clause instruction in the acquisi tion of English as a second
de Villiers, J . G. and P. A. de Villiers. 1973. 'A cross-sectio nal study of the
lan guage.' Applied Linguistics 9/ 1: 1-20.
acqu isition of grammatical mo rphemes.' Journal 0/Psycholinguistic Research
2/3: 267-78. Ellis, N . C. 200 1. 'Memory for langu age' in P. Robinson (ed.): Cognition
and Second La nguage Instruction. C am bridge: Camb ridge U niversity Press.
Derwing, T. M. and M.J. Munro. 20 09. 'Purring accent in its place:
pp.33-68.
Rethi nking o bstacles to communicat ion.' Language Teaching42/4: 476-90 .
EUis, N. C. 2002. ' Freq uency effects in language acq uisitio n: A review with
Derwing, T. M. and M . Rossit er. 2003. 'The effects of pronunciation
impli cations for theories of implicit and explicit lang uage acquisition.'
instruction on th e accu racy, fluency, and complexity ofL2 accented
Studies in Second Language Acquisition 24/2: 143-188.
speech .' Applied Language Learning 13/ 1: 1- 17.
Ellis, N . C . 2003. ' Co nstr uctio ns, chu n king, and connectionism: The
Donato, R 1994 . 'Collective scaffold ing in second language learn ing' in
emergence of second language structure' in C.]. Do ughty and M . H . Long
] . Lantolf and G . Appel (eds.): Vygotskian Approaches to Second Language
(eds.): The Handbook ofSecond Language Acquisition. O xford: Blackwell.
Research. Norwood, N] : Ablex. pp. 33-56.
p p. 63-103.
Dornyei , Z. 200 1a. Motivational Strategies in the Language Classroom.
Ellis , N. C. 2005 . 'At the int erface: Dynam ic interactions of explicit and
C ambridge: Cambridge U niversity Press.
im plicit language knowledge.' Studies in Second Language Acquisition
Dornyei, Z. 2001 b. Teaching and Researching Motivation. H arlow: 27/2: 305-52.
Longman. Ell is, N. C. 2009. ' O ptimizing the input : Freq uency and samp ling
Dornyei, Z. 2005. The Psychology ofrhe Language Learner: Indi vidual in usage-based and fo rm-focus ed learn ing' in C.] . Dou ghty and
Differences in Second LanguageAcquisition. Mahwah, N]: Lawrenc e M. H . Long (eds.) : The Handbook o/Language Teaching. Malden, MA:
Erlbaum and Associates. W iley-Blackwell. pp. 139-1 58 .
232 Bibliography
Ellis, N. c., R Simpson-Vlach, and C. Maynard. 2008 . ' Form ulaic
language in native and seco nd language speakers: Psycholinguisrics, corpus
linguistics, and TESOL.' TESOL Quarterly 42/3 : 37 5-96 .
Ellis, R 2012. Langu age 'Teaching Research and Langllage Pedagogy. Malden,
MA: Wiley-Blackwell.
Ellis, R. and G. Barkhuizen, 200 5. Analysing Learner Language. Oxford:
Oxford Uni versity Press.
Ellis, R, S. Loewen, C. Elder, R. Edam,]. Philp, and H . Reinders. 2009.
Implicit and Exp licit Knowledge in Second Language Learning: Testingand
Teaching. Bristol, UK: Multilingual Matters.
Erard, M. 201 2. Babel No More: The Searchfor the World's Most
Extraordi nary Language Learn ers. New York: Free Press.
Erlam, R 2005. ' Language aptitude and its relationship to instructional
effectiveness in second language acquisition.' Language Teaching Research
9/2: 147-1 71. . behavior in th e com municative lan guage classroom. ' JALTJournal
Ganschow, L. and R. L. Sparks . 200 1. 'Learn ing difficulties and foreign
lang uage learning: A review of research and inst ruction .' Language Teaching
34 / 2:79- 98.
Gardner, D. 2004 . 'Vocabulary input thro ugh extensive reading:
A comparison of words found in children's narrative and exposito ry
reading materials .' Applied Lingu istics 25/1: 1-37.
Gardner, H . 1993. Multiple Intelligences: The Theory in Practice. New York:
Basic Books.
Gardner, R. C. and vv. E. Lambert. 1972. Attitudes and Motivation in
Second Langllage Learning. Rowley, MA: Newbury House.
Gass, S. M. 1982. 'Fro m theory to practice' in M . Hin es and
\Y/. Rutherford (eds.): On TESOL '8 1. Washington, D C: T ESOL.
pp .129-139.
Gass, S. M. 1988. ' Inrcgrarlng research areas: A framework for second
language studies. ' Applied Linguistics 9/2: 198-2 17.
Gass, S. M . 199 7. Input, Interaction, and the Second Language Learner.
Mahwah, N] : Lawrence Erlbaum and Associa tes.
Garbo nt on, E. and N. Segalowitt. 1988. 'Creative automatization:
Classroom Second Language Acq uisition . Camb ridge: Cambridge Univers ity
Princ iples for promoting fluency within a communicative framework .'
Press. pp. 156-74 .
TESOL Quarterly 22/3: 47 3- 92 .
Gatbonton, E. and N. Segalowitz, 200 5. 'Rethinking communicative
language teaching: A focus on access to fluency.' Canadian Modern
Language Review 61/3 : 325-53.
Bibliography 23.3
Gatbonton, E., P.Trofimovich, and M. Magid. 2005. ' Learn ers' ethnic
group affiliation and L2 pronunciation accuracy: A sociolinguistic
investigation.' TESOL Qua rterly 39 /3 : 4 89- 5 11.
Genesee, E 1976 . 'Th e role of intelligence in second language learning.'
Language Learni ng 26/ 2: 267-80.
Genesee, F. 1987 . Learning through Two Languages. New York:
Newbury House.
Genesee, E , K. Lindholm-Leary, vv. M . Saunders, and D. Christian
(eds.). 2006. Educating English Language Learners:A Synthesis ofResearch
Evidence. Camb ridge: Cambridge University Press.
Goldschneider, J. M. and R M . DeKeyser. 200 1. 'Ex plaining th e narural
order ofL2 morpheme acq uisition in English: A meta-analysis of multiple
determinants.' Language Learning 51/ 1: 1-50.
Greer, D . L. 2000. 'The eyes of hiro: A Japanese cultura l mo nito r of
22/ l: 183-95 .
G uillo tea ux , M. J. and Z. Dornyei. 2008 . ' Mo tivating language learn ers:
A classroo m-o riented investi gati on of th e effects o r"mo tivatio nal st rategies
o n student motivation.' TESO L Quarterly 42 /1 : 55-77 .
Guiora, A., B. Beit-Hallahami, R Brannon, C. Dull, and T. Scovel.
1972 . 'Th e effects ofexperimentally induced changes in ego states on
p ro nunciation abiliry in a seco nd language: An exploratory stu dy.'
Comp rehensive Psychiatry 13/5: 421-8.
Hahn, L. D . 2004 . 'Primary stress and inrelligibiliry: Research ro motivate
the teaching ofsuprasegmenrals .' TESOL Quarterly 38/2 : 20 1-23 .
Hahne, A. 2.00 1. ' What's different in seco nd-language processing?
Evidence from event-related brain porenrials.' Journal ofPsycholinguistic
Research30/3: 251 -66 .
H amil to n, R. 1994 . ' Is irn plicar io nal generalization unidirectional and
maximal? Evidence from relativizarion instructio n in a second languagc.'
Language Learning 44/ 1: 123-57.
H arley, B. 1998. 'The role of focus-on -form tasks in promoting child
L2 acq uisition' in C. Doughty and ]. W illiams (eds.): Focuson Form in
Harley, B. and M. Swain. 1984. 'Th e inrerlanguage of immersion students
and irs implications for second language teaching' in A. Davies, C. C riper,
and A. Howatt (eds.): Interlangl/age. Edinburgh: Edinburgh University
Press. pp . 291 -3 11.
234 Bibliography
Hatch, E. 1978. 'D isco urse analysis and second language acquisition' in
E. Hatch (ed .): Second Language Acquisition: A Book ofReadings. Rowley,
MA: Newbu ry. pp . 401-35.
Heath, S. B. 1983. ways with WO rds'. Cambridge: Cambridge University
Press.
Hedge, T. 2000. Teaching and Learning in the Language Classroom . O xford:
O xford U niversiry Press.
Hoare, P. and S. Kong. 2008. ' La te immersion in Hong Kong: Still stressed
but making progress?' in T. W Fort une and D . J. Tedick (eds.): Pathways to
Klein, w: and C. Perdue. 1993. Utteran ce Stru cture. Amsterdam:
Multilingualism: Em erging Perspectives on Immersion Education. Cleve don:
John Benjamins. http://www.mpLnllworldlcgllapp/esf/es[html
Multilingual Maners. pp . 242-66.
Kojic-Sabo, I. and P. M . Lightbown. 1999. 'Stu dents' ap proaches to
Horst, M. 2005 . 'Learning L2 vocabulary through extensive reading:
vocabulary learn ing and their relationship to success.' Modern Language
A measurement study.' Canadian Modern Language Review 6 1/3: 355-82.
journal 83/2: 176-92 .
Horwitz, E. K. , M. B. Horwitz, andJ. Cope. 1986 . 'Foreign language
classroom anx iety,' Modern Language journal 7 0/ 2: 125-32 .
Howard, E. R., J. Sugarman, D. Christian, K. Lindholm-Leary, and
D. Ro gers. 200 7. Guiding Principles for Dual Language Education 2nd edn.
Washington, DC: Cent er fo r Applied Linguistics. hcc p:llwww.cal.org/rwi/
guid ingprinciples.h tm
Hu, M. and I. S. P. Nati o n. 2000. 'U nkn own vocabulary density and
reading comprehension.' Reading in a Foreign Language 13/ 1: 403-30 .
Hulstijn,J. and B. Laufer. 2001. 'So me empirical evidence for the
involvement load hypothesis in vocabulary acq uisition.' Language Learning
5 1/3 : 539-58 .
Hunter, J. 20 12. 'Small talk : developing fluency, accuracy, and com plexity
in speaking.' ELTjournal 661 1: 30-41.
Itard, J .-M .-G . 1962. The Wild Boy ofAveyron (L'enfant sauvage). New York:
Meredith.
Jarvis, S. and A. Pavleoko. 2008. Crosslinguistic Influence in Language and
Cognition. New York: Ro utledge.
Johnson, J. and E. Newpo rt. 1989 . 'Critical period effects in second
language learning: The in fluence of maturational state on the acquisition of
Englis h as a Seco nd Language.' Cognitive Psychology 21 11: 60-99.
Johnson, R. K. 1997. 'Th e Hong Kong education system: Late immersion
under stress' in R. K. Johnson an d M . Swain (eds.): Immersion Education:
International Perspectives. C amb rid ge: Cambridge Universiry Press.
pp . 171 -89.
BibLiography 23 5
Kasper, G. and K. R. Rose. 2002. ' Pragmat ic Development in a Second
Language.' Language Learning 52/Supplement 1.
Keenan, E. and B. Comrie. 1977. 'Noun phrase accessibiliry and Un iversal
Grammar.' Li ngu isti c Inquiry 8/1: 63-99.
Kellennan, E. 1986. 'An eye for an eye: Crosslinguistic cons trai nts o n th e
develo pment of me L2 lexicon' in E. Kellerman and M. Sharwood Sm ith
(eds.): Crosslinguistic Influence in Second Language Acquisition. New York:
Pergamon. pp. 35-48.
Krashen, S. 1982. Principles and Practice in Second Language Acquisition.
Oxford: Pergamon.
Krashen, S. 1985 . The Input Hypothesis: Issues and Implications. London:
Longman.
Krashen, S. 1989 . 'We acquire vocabulary an d spelling by readi ng:
Additio nal evidence for the inpu t hypothesis.' Modern Language journal
7 3/4 : 440-64.
Lado, R. 1964. Language Teaching: A Scientific Approach. New York:
McGraw-Hill.
Lambert, w: E. 1987 . 'Th e effects of bilingual and bicu ltu ral experiences
on children's attitudes and social perspectives' in P. Homel, M. Pali], an d
D . Aaronson (eds.): Childhood Bilingualism: Aspects ofLinguistic, Cognitive,
and Social Development. Mahwah, NJ : Lawrence Erlbaum an d Associates .
pp .197-221.
Lambert, w: E. and G. R. Tucker. 1972. Bilingual Education ofChildren:
The St. Lam bert Experiment. Rowley, MA: Newbury.
Lantolf, J . P. (ed.), 2000. Sociocultural Theory and Second Language
Learning. Oxford: Oxford Universiry Press.
Lapkin, S., D. Hart, and B. Harley. 1998 . 'Case study of co m pact Core
French models: Attitudes and achievement' in S. Lapkin (ed .): French
Second Language Education in Canada: Empirical Studies . Toronto, Canada :
Universiry ofToromo Press. pp. 3-30 .
Leow, R. P. 199 7 . 'Attention, awareness, and foreign language beha vior.'
Language Learning 47/ 3: 46 7-505 .
236 Bibliography Bibliograph} 237

Li, S. 20 10. 'Th e effecti veness of co rrect ive feedback in SL\: A meta­ Loewen, S . and]. Philp. 200 6. 'Re casts in th e ad ult Eng lish Ll class:
analysis.' Language Learning 60 /2 : 309-365. characteristics, explicitness, and effectiveness. ' Modern Language[o urna
90/4: 536-56.
Lieven , E. and M. Tomasello. 2008 . 'C hildr en's first lan gu age from a usage
perspective' in P. Ro binson an d N . C. Ellis (eds .): Handbook of Cogn iti ue Loewen, S., S. Li, F. Fei, A. Thompson, K. Nakatsukasa, S. Ahn, and
Linguisti cs and Second Language Acquisition. New York: Roucledge. X. Chen. 2009. 'Second language learners' beliefs about grammar
pp ..168-96. instructio n and error co rrectio n .' Modern Langu ageJo u m al 93/ 1: 91 - 104 .
Lightbown, P. M. 1983a. 'Acquiring English L2 in Quebec classrooms' Long, M . H . 1983. 'N ative speaker/non-native speaker conversation and
'in S. Felix and H. Wode (eds.): Language Development at the Crossroads. the negotiation of comprehensible input.' Applied L inguistics 4/2: 126-41.
T iibingen : Gunter Nar r, pp . 15 1- 7 5.
Long, M . 1996 . 'The role of the linguistic environment in second language
Ligh t bown, P. M. 1983b. 'Exploring relarionships between developmental acquisition' in W. Rit chie and T. Bhatia (eds .): Handbook ofSecond
and insrru ctio nal sequences in L2 acquisition' in H. W. Seliger and Language Acquisition. New York: Academic Press. pp . 413-68 .
M . H . Long (eds.) : Classroom-Orienud Research in Second Language
Long, M . H. and P. Porter. 1985 . 'Group work, interlanguage talk , and
Acquisition. Row ley, MA : Newbury House. pp. 21 7-45 .
seco nd language acquisition .' TESOL Quarterly 19/2: 20 7-28.
Ughtbown, P. M . 1998. 'The importance of riming in focus on form '
Long, M. and C. Sa to. 1983. 'C lassroo m foreigner talk discourse: Forms
in C. D oughcyandJ. \Xfilliams (eds.): Focus on Form in Classroom
an d fun cti on s of teachers' q uestions' in H. W. Seliger and' M . H . Lon g
Second La nguage Acquisition. Cam bridge: Ca m b ridge Un iversity Press,
(eds.) : Classroom -Oriented Research in Second Language Acquisition. Rowley,
pp. I77-%.
J'v1A: Newbury House. pp. 268-85 .
Ligh t bown, P. M. 2007. 'Fair tra de: Two-wa y bilin gual ed uca rio n.'
Long, ] ., C. Brock, G. Crookes, C. Deicke, L. Po tt er, and S. Zhang.
Estudios de Lingidstica Inglesa Aplicada 7: 9-34.
1985. 'The effect of reach ers' qu est ioning partern s and wai t tim e on pu pil
Ligbtbown, P. M . 2008a. 'Easy as pie? Children learning languages.'
parti cipation in public h igh school classes in H awaii for stu den ts of limited
Concord ia WO rking Papers in Applied L inguistics 1:1- 25.
English proficiency.' TESOL Quamrly 19/3 : 605-07.
hrrp :/ /doe. concordia. cal copal/ index.p hp/cop al/volumes.hrml

Lyster, R. 199 4. 'Th e effect of functional-analytic reaching on aspects of

Lightbown, P. M. 2008b. 'Transfer approp riare processing as a model fo r
French immers ion students' sociolinguistic co mpetence.' Applied Linguistics
classroom seco nd language acquisition' in Z. Han (00.) : Understanding
15/ 3: 263-87.
Second La nguage Process. C levedon , UK: Mulrilingual Maners. pp . 27-44.

Lyster, R. 199 8. ' Recasts, repetition and ambiguity in L2 classroom

Light bown, P. M. 2012 . 'Intensive L2 inst ruction in Canada: Why nor
discourse.' Studies in Second La nguage Acquisition 20 /1 : 51-81 .
immersion?' in C. Munoz (ed .): Intensive Exposure Experiences in Second

Lyster, R. 2004. ' D ifferential effects of prompts and recasts in form ­

Language Learning. Bristo l, UK: Mulrilingual Matters. pp. 25-44.

focused instruction.' Studies in Second Language A cquisition 26/3: 399-432.

Light bown, P. M. and N . Spad a. 1990. ' Focus on form and corrective

Lyster, R. 2007. Learning and Teaching Language through Content:

feedback in cornmunicative lan gu age reach ing: Effects o n seco nd language

A Counterbalanced Approach. Amsterdam: John Benjarnins.

learning.' Studies in Second Language Acquisition 12/ 4: 429-48.

Lyster, R. an d H . M ori. 2006. ' Interactional feedback and ins tructional

Ligh t bown, P. M. and N. Spada. 1994. 'An innovative program for

cou nterbalance.' Studies in Second Language Acquisition 28/2: 269-300.

p rimary ESL in Quebec: TESOL Quarterly 28/3: 563-79.

Lyst er, R. and L. Ranta. 1997. 'C orrect ive feedback and learner uprake:
Ligh t bown, P. M ., R. Halter,]. L. White, and M. H o rst. 2002.

Negotiation of form in communicative classrooms.' Studies in Second

'Com p rehensio n-based learn ing: The limits of "do it yourself".' Canadian

LanguageAcquisition 19/ 1: 37-61.

Modem Language Rev iew 58/3: 42 7-64.

Lyst er, R. an d K. Saito. 2010. 'O ral feedback in classroom SLA:

Lindholm-Leary, K. 2001. D ual Language Education. Clevedon, UK:

A m eta-analysis.' Studies in Second Language A cquisition 32/2: 265 - 302.

Mulrilingual Maners.

238 Bibliography
Macintyre, P. 1995. ' How does anxiety affect second language learning?
A reply to Sparks and Ganschow. ' Modern Language journal 791 1: 90-9.
Mackey, A. 1999. 'Input, interaction and second language development:
An empirical study ofquestion formation in ESL.' Studies in Second
Language Acquisition 21 14: 557-87.
Mackey, A. and]. Goo. 2007. ' Int eraction research in SLA: A meta­
analysis and research syn thesis .' in A. Mackey (ed.): Con versational
Interaction in Second Language Acquisition: A Series of Empirical Studies.
Oxford: Oxford Universiry Press. pp. 40 7-452.
Mackey, A. and]. Philp. 1998. 'C onversatio nal interaction and second
language development: Recasts, responses and red herrings.' Modern
Language journal 82 /3: 338-56.
Mackey, A., S. M. Gass, and K. McDonough. 2000. 'How do learners
per ceive interactional feed back?' Studies in Second Language Acquisition
22/4: 47 1-497.
MacWhinney, B. 1997. 'Seco nd lang uage acq uisition and the competition
mode l' in A. M . B. de G root and J. F. Kroll (eds.): Tutorials in Bilingualism:
Psycbolinguistic Perspectives. Mahwah, NJ: Lawrence Erlbaum and
Associates. pp . 113-44.
MacWhinney, B. 2000. The CHILDES Project: TOols for Analyzing
Talk (2 volu mes) . Mahwah , NJ : Lawrence Erlba um and Associates.
http:/ /childes.psy. cmu.edu/
Masgoret, A. M. and R C. Gardner. 2003. 'Atti tudes, motivation, and
second language learning: A me ta-analysis ofstudies conducted by Gardner
and associates.' Language Learning 53/1 : 123- 63 .
McCormick, D. E. and R. Donato. 2000. 'Teacher questions as scaffolded
assistance in an ESL classroom' in J. K. Hall and L. Verplaetse (eds.): Second
and Foreign Language Learning through Classroom Interaction. Mahwah. NJ:
Lawrence Erlbaum and Associates. pp. 183-201 .
McDonough, K. 2004. ' Lear ner-learner interaction during pair and small
group activities in a Thai EFL context.' System 32: 207-24.
McLaughlin, B. 1987 . Theories ofSecond Language Learning. London:
Edward Arnold .
McLaughlin, B. 190/0. 'Restructuri ng.' Applied Linguistics I II 1: 113-28.
Meara, P. M. 1980. 'Vocabulary acquisition: A neglected aspect oflanguage
lear ning.' Language Teaching and Linguistics Abstracts 13: 221-46.
Meara, P. M. 2005. 'D esigning vocabulary rests for Englis h, Spanish
and other languages' in C. Buder, M . dos Angeles Cornez-Gonzales, and
Bibliography 239
S. D oval-Suarez (eds .): The Dyn amics ofLanguage Use: Functional and
Dyn amic Perspectives. Amsterdam: John Benjamins.
http ://www.lognostics.co.ukltools/ pp . 269-83.
Meisel,]. M. 1987. 'Reference to past events and actions in the development
of natural second language acquisition' in C. Pfaff (ed.): Firstand Second
Language A cquisition Processes. Cam bridge, IvfA: Newbury House.
Meisel, J. M., H . Clahsen, and M. Pienemann. 1981. 'O n determining
develo pment al stages in natural second language acq uisitio n.' Studies in
Second Language Acquisition 3 /2 : 109- 3 5.
Mwioz, C. 2006. 'Th e effects of age on foreign language learni ng: The BAF
project' in C. M unoz (ed. ): Age and the Rate ofForeign Language Learning.
Clevedon , UK: Multilingual Matters. pp. 1-40.
Munro, M. J. and T. M. Derwing. 2011. 'Th e fou ndations of accent and
intelligibiliry in pron unciation research.' ~anguage uaching44 /3: 3 16-27.
Nagy, W. E., p"A. Herman, and R.Anderson. 1985. 'Learning words from
context.' Reading Research Quarterly 20/2: 233-5 3.
Nation, I. S. P. 2001. Learning Vocabulary in Another Language.
Cambridge: C am bridge Universiry Press.
Nation, I. S. P. 2007. 'The four strands.' Innovation in Language Learning
and uaching I l l : 1-12.
Nation, 1. S. P. and]. Macalister. 2010. Language Curri culum Design.
New York: Routledge.
Netten,]. and C. Germain (eds.). 2004. 'Int ensive French in Ca nada .'
Canadian Modern Language Review 60 /3 : pp . 249-435 .
Newpo rt, E. 1990. 'Maturational constraints on language learning.'
Cognitive Science 14/1 : 11-28.
Nicholas, H., P. M. Lightbown, and N . Spada. 2001 . ' Recasts as feedback
to language learners.' Language Learning 51/4: 719-58.
No rris, ] . M. and L. O rtega. 2000. 'Effectiveness ofL2 instruction:
A research synthesis and quantitative meta-an alysis.' Language Learning
50 /3: 417-528 .
Norton Peirce, B. 1995. 'Social identity, invest ment, and language
learn ing. TESOL Quarterly 29 11: 9-31 .
Norton, B. and K. Toohey. 200 1. 'C han ging perspe ctives on good
language learners.' TESOL Quarterly 35/2: 307-22.
Ohler, L. 1989. 'Exceptional secon d lan guage lear ners' in S. Gass,
C. Madden, D . Preston, and L. Selinker (eds.): Variation in Second
240 Bibliography
Language A cquisition, Vol. II: Psycbolinguistic Issues. C levedo n: Multilingual
Marrers . pp. 141-59.
O dlin, T. 2003. 'Cross-Iingulsric influence' in C. J. Doughry and
M. H. Long (eds.) : TheH andbook ofSecond Language Acquisition. Oxford:
Blackwell. pp. 436-86.
Ohta, A. 2000. ' Rethinking recasts: A learner-centered examination
of corrective feedback in the Japanese classroom' in J. K. Hall and
L. Verplaetse (eds.) : Second and Foreign Language Learning ,hrough
Classroom Interaction. Mahwah, NJ: Lawrence Erlbaum and Associates.
pp .47-71.
Oliver, R. and A. Mackey. 2003 . ' In teractio nal context and feedback in
child ESL classrooms.' Modern LanguageJournal 87/4: 519-33.
Ortega , L. 2007. 'Meaningful L2 practice in foreign lang uage classrooms:
A cognitive-Interacrionist SLA perspective' in R. DeKeyser (ed.): Practice
in a Second Language: Perspectives from Applied Linguistics and -Cognitive
Psychology. C ambridge: Cam bridge Uni.versity Press. pp . 180-207 .
Paradis, J., F. Genesee, and M. B. Crago. 201 1. D ual La nguage
Development and Disorders:A Handbook on Bilingualism and Second
Langu age Learn i ng 2nd edn. Baltim ore: Paul H . Brookes.
Patkowski, M. 1980 . 'The sensitive period for the acqu isition of syntax in a
second language.' Language Learning 30/2: 449-72.
Pearson, B. Z. 2008. Raising a Bilingual Child: A Step-by-Step GuidefOr
Parents. New York: Living Language (Random H ouse).
Piaget, J. 1951. Play Dreams, and Imitation in Childhood. New York:
Norton. (Originally published as La fOrma tion du symbole chez len[ant.
Neucharel: Delchaux er Niestle.)
Pica, T. 1983. 'M ethods of morpheme q uantification: Their effecr on
the interpretation ofsecond language data.' Studies in Second Language
Acquisition 6/ 1: 69 - 78 .
P ica, T. 1994. ' Research o n negotiation : What does it reveal abo ut second
language acquis itio n? Conditions, processes, and outcomes' Language
Learning44/3: 493-527.
P ien em ann, M. 1988. ' De term in ing the influ ence of instruction on L2
speec h processing.' AlIA Review 5/ 1: 40-72.
Pi en em ann, M. 1999. Language Processing and Second Language
Development: Processability Theory. Amsterdam: Joh n Benjamins.
Bibliography 24 1
Pienemann, M. 2003 . ' Lan guage processing capacity' in C. J. Doughty
and M. H. Long (eds .): The Handbook ofSecond Language Acquisition.
Oxfo rd: Blackwell. pp . 679-714 .
Pien em ann, M., M . Jo hnston , and G. Brindley. 1988. 'Constructing an
acquisition-based procedure for second language assessment.' Studies in
Second Language Acquisition 10/2: 21 7-4,.3 .
Pimsleur, P. 1966. The Pimsleur Language Aptitude Battery. New York:
Harcourt, Brace, Jovanovic. .
Piper, T. 2006. Language and Learning: The Home and School Years 4th edn,
U ppet Saddle River, NJ: Merrill/Prentice-Hall.
Piske, T. 2007. ' Im plications ofJames E. Flege's research for the foreign
language classroom' in O . Bohn and M . Munro (eds.): Language Experience
in Second Language Speech Learning. Amsterdam: John Benjarnins.
pp.301-314 .
P iske, T., I. R. A; MacKay, and J. E. Flege. 2001 . :Factors affectin g degr ee
of foreign accent in an L2: A review.' [ournal ofPhonetics 29 /3 : 191 -215 .
Ranta, L. 2002. 'The role of learn ers' language an alytic ab iliry in the
co mm unicative classroom' in P Robinson (ed .): Individual D ifferencesand
Instructed Language Learning. Amsterdam: John Benjarnins. pp . 159-80.
Reid ,]. (ed.). 1995. Learning Styles in the ESUEFL Classroom. New York:
Hei nle and Heinle.
Richards,J. 1974. 'A non-contrastive approach to error analysis' in
J. Richards (ed.) . Error analysis. Londo n: Longman. pp. 172-88.
Ringbom, H . 1986. ' C rosslingu istic influence and the foreign language
learning process' in E. Kellerman and M. Sharwood Smith (eds.):
Crosslinguistic Influence in Second Language Acquisition. New York:
Pergamo n Press. pp . 150-72. .
Rob in so n , P. (ed.). 2002. Individual Differences and Instructed Language
Learning. Ams terdam: John Benjamins.
Robinson, P. and N. C. Ellis. 2008. 'Conclusio n: Cognitive linguis tics,
seco nd language acq uisition, and L2 instruction-issues fo r research'
in P Robinson and N. C. Ellis (eds.): H and book of Cognitive Lingu istics
and Second Language Acquisition . New York: Routledge. pp. 489-545.
Roy, D. 20 09. 'New horizon s in the stu dy of child language acquisitio n.'
INTERSPEECH-2009, Brighton , UK.
Sachs, J., B. Bard, an d M. Johnson. 1981. 'Language learn ing with
restric ted inp ut: Case studies of two hearing children of deaf parent s.'
Applied Psycholinguistics 21 1: 33-54.

242 Bibliography Bibliography 243

Samuda, V. 2001 . 'G uiding relationships between form and meaning Sharwood Smith, M. 1993 . ' Inp ut enhancement in instructed SLA:

during task performance: The role of the teacher' in M. Bygate, P. Skehan, Theoretical bases.' Studies in Second Language Acquisition 15/2: 165-79.

and M . Swain (eds.): Researching Pedagogic Tasks: Second Language Learning,

Sheen, Y. 2006. 'Corrective feedback and learner uptake in communicative

Teaching, and Testing. London: Longman. pp . 119-40.

classrooms across instructional settings.' Second Language Teaching Research

Savignon, S. 1972. Communicative Comp etence: An Experiment in Foreign­ 8/3 : 263-300 .

language Teaching. Philadelphia, PA: Center for Curriculum Development.

Sheen, Y. 2010. ' D ifferent ial effects o f oral and written corrective feedback

Schachter, J. 1974. 'An error in error analysis.' Language Learning in the ESL classroom.' Language Teaching Research 32 /2: 203-34.

24/2: 205-14.
Simpson, R. C, S. L. Briggs,]. Ovens, an d]. M. Swales. 1999 . The

Schachter,]. 1990. 'On rhe issue of completeness in second language Michigan CorpusofA cadem ic Spoken English (MICASE). Ann Arbor, MI:

acquisition.' Second Language Research 6/1: 93-124. The Regents of the University of Michigan. http://micase.e1icorpora.info.

Schieffelin, B. 1990 . The Give and lake o/Everyday Life: Language Sinclair,]. M. (ed.). 2004. How to Use Corpora in Language Teaching.

Socialization o/Kaluli Children . Cambridge: Cambridge University Press. Philadelphia, PA: John Benjamins.

Schleppegrell, M. 2004. The Language o/Schooling. Mahwah, N]:
Lawrence Erlbaum Associates.
Schmidt, R. 1990. 'The role of consciousness in second language learning.'
Applied Lingu istics 11/1: 17-46.
. Skehan, P. 1989 . Individual Differences in Second Language Learning.
London: Edward Arno ld.
Skinner, B. F. 1957 . Verbal Behavior. New York: Appleton-Century-Crofts.
Slobin, D. I. 1973. 'C ognitive prerequisites for the development of

Schmidt, R. 2001. Attention' in P. Robinson (ed.): Cognition and Second grammar' in C. A. Ferguson and D. 1. Siobin (eds.): Studies of Child

Language Instruction. Cambridge: Cambridge University Press. pp. 3-32. Language Development. New York: Holt, Rinehart, and Winston.

Schulz, R. A. 2001. ' C ultural differences in student and teacher pp . 175-208.

perception s conceming'rhe role of grammar teaching and cor rective Siobin, D. I. 1985-1997. The Crosslinguistic Study 0/Language . Vols 1-5 .

feedback: USA-Colombia.' Modern LanguageJournal 85/2 : 244-58 . Mahwah, N]: Lawrence Erlbaum and Associates.

Schumann,]. 1979 . 'Th e acquisition of Engli sh negation by speakers of Smith, N. and I. M. Tsimpli. 1995. The Mind ofa Savant: Language

Span ish: a review of the liter ature ' in R. W Andersen (ed.) : TheAcquisition Learning and Modularity. Oxford: Blackwell.

and Use ofSpanish and English as First and Second Languages. Washington,
Snow, C 1995. 'Issues in the study of input: Fine- tuning, universality,

DC:TESOL.
individual and developmental differences, and necessary causes' in

Schwartz, B. 1993. 'O n explicit and negative data effecting and affecting P. Fletcher and B. MacWhinney (eds.): The Handbook ofChild Language .

competence and linguistic behav ior.' Studies in Second Language Acquisition Oxford: Blackwell. pp. 180-93.

15/2: 147-163.
Snow, C and M. Hoefnagel-Hohle, 1978. 'The critical period for

Segalowitt, N. 2010. Cognitive Bases o/Second Language Fluency. lan guage acquisition: Evidence from second language learning.'

New York: Routledge. Child Development 4914: 1114-28.

Seidlhofer, B. 2011. Understanding English as a Lingua Franca. Oxford: Spada, N. 20 11. ' Beyond form-focused instruction: Reflections on past,

Oxford University Press. present and fut ure research.' Language Teaching 44/2: 225-36.

Selinker, L. 1972. 'Inr erlan guage.' International Review 0/Applied Spada, N. and M. Frohlich. 1995. The Communicative Orientation

Linguistics 10/2 : 209-31 . 0/Language Teaching Observation Scheme: Coding Conventions and

Applications. Sydney: Macmillan .

Setter,]. and].]enkins. 2005. 'Pronunciation.' Language Teaching

38/1: 1-17. Spada, N. and P. M. Lightbown. 1993. 'Instruction and the development

of questions in L2 classroo ms.' Studies in Second Language Acquisition

15/2: 205-24.

244 BibLiography
Spada, N. and P. M. Lightbown. 1999. ' Insrructio n , L1 influence and
developmental readiness in second language acquisition.' Modern Language
JournaL8311: 1-22.
Spad a, N. and P. M . Light bown. 2002. 'L 1 and L2 in the education of
Inuit children in northern Quebec: Abilities and perceptions.' Language
and Education 16/3 : 212-40.
Sp ada, N . and Y. Tomita. 2010. ' Interact io ns between type of instruction
and type oflanguage feature: A me ta-analysis .' Language Learning
60/2: 1-46.
Sp ada, N., P. M. Lightbown, and ]. L Whit e. 2005 . 'The importance
oHorm/mean ing mappings in explici t form -focussed instruction' in
A. Housen and M. Pierrard (eds .): Investigations in Instructed Second
Language Acquisition. Amsterdam: Mouton de Gruyter, pp. 199-234.
Spada, N . and M . Santos Lima. 2010. ' EFL and ESi. teacher and learner
views about integrated and isolated form-focused instructi on.' Proceedings
of th e IX Inrern ati o nal Conference o f the Brazilian Assoc iatio n of C ana d ian
Stud ies: Po rro Alegre, Brazil.
Spad a , N., K. Barkaoui, C. Peters, M. So, and A. Valeo. 2009.
.De veloping a questionnaire to m easure learners' preferen ces for isolated
and integrated form-focused instruction.' System, 37/ 1: 7 0- 8 1.
Spada, N., L. Jessop, W; Suzuki, Y. Tomita, and A. Vale o . 2009. The
Conrributions of Isolated an d Inregrated Fo rm-Focused Instruction
on Different Aspects of L2 Knowledge and Use. Paper p resented at the
American Association for App lied Linguistics: Denver, CO.
Spielmann, G. and M. J . Radnofsky. 2001 . ' Lear ning language under
tension: New directions from a qual itative study.' Modern Language journal
85/2 : 259-78 .
Storc h, N. 2001 . ' Patterns of interaction in ESL pair work.' Language
Learning 52/1: 119-58.
Swain, M. 1985. 'Comm unicative competence : Some roles o f
comprehens ible input and comprehensib le o ut put in its develo pm ent '
in S. Gass an d C. Madde n (eds.): Input in Second Language Acquisition.
Rowley, MA: N ewb ury H o use. pp. 235 -53 .
Swain, M. 1988. 'M anipulating and complementing content teaching to
maximize second language learning. ' TESL Cana da journal Gt s: 68-83.
Swain, M. 2000. 'The ou tput hypo th esis and beyond: Me diating
acquisit io n th rou gh collabo rat ive d ialogue' in] . P. Lanro lf{ed. ):
Sociocultural Theory and Second Language Learning. Oxford: Oxfo rd
University Press. pp . 94-114.
Bibliography 145
Swain, M. and S. Lapkin. 1998. ' Int eractio n and seco nd language
lear ning: Two ado lescent Fre nch immersion stude nts working toge ther.'
Modern Language[ournal 82/3: 33 8-56.
Swain, M . an d S. Lapkin. 2002. 'Talking it through: Two French
immersion learners' respo nse to reformulation.' Internationaljournal of
Edu cational Research 3 7/3-4: 285-304.
Swain, M., P. Kinnear, and L Ste inman. 2010 . Sociocultural Theory and
Second Language Education: An Introduction through Narratives. Bristo l:
Multilingual Maners .
Taro ne, E. and M. Sw ain. 1995 . 'A sociolinguistic perspective on
seco nd-language use in immersion classrooms.' Modern Language [ournal
79/2: 166-178.
Tarone, E. and B. Swierzbin. 2009 . Exploring Learner Language. Oxfo rd:
Oxfo rd University Press.
Taylor, D ., M. Caron, and L. McAlpine. 2000. 'Patterns o f perceived
lang uage abili ty and use in Arctic Q ue bec.' Bilingual Education and
BilinguaLism 3/4: 28 3-96.
Tomita, Y. 20 11. The Role of Form-Foc used Instruction : Learn er
Investment in L2 C o mm un ication. Unpubl ished thesis . University of
Toronto. h ttps:!!rspace.library.utoronw.caJbitstream/1807/29892/ 1/
Tornira, Yasuyo_20 1106_PhD_thesis.pdf
Toohey, K. 2000. Learning EngLish at School: Identity. Social Relations and
Classroom Practice. C levedon: Multilingual Maners .
Trahey, M . and L White. 1993. ' Positive evidence and preemption in
the second lang uage classroom.' Studies in Second Language Acquisition
15/2: 18 1-204.
Tro6m ov ich, P. 2005. 'Spoken-wo rd process ing in native and second
lang uages: An investigatio n o f auditory word priming.' Applied
PsychoLinguistics 26/4: 479-504.
Tr ofim ovich , P., P. M. Lightbown, R. H. Halter, an d H. Song. 2009.
'Co m prehension- based practi ce: The develop m ent o f L2 pro n unciation
in a listen ing and read ing p rogram .' Studies in Second Language Acquisition
31/4: 609-39.
VanPatten, B. 2004 . ' Inp ut pr ocessing in SLA' in B. Van Patte n (ed .):
Processing Instruction: Theory, Research, and Commentary. Mahwah, N] :
Lawrence Erlba um and Associa tes. pp. 5-3 1.
VanPatten , B. and T. Cadiemo. 1993. ' Explici t instructio n and input
processing .' Studies in Second Language A cquisition 15/2: 225-41 .
246 Bibliography Bibliography 247

VanPatten, B. and]. Williams (eds.). 200 7 . Theories in Second Language White, L. 2003 . Second Language Acquisition and Universal Grammar.
Acquisition: An Introduction. Mahwah. NJ: Lawrence Erlbaum and Cambridge: Cambridge University Press.
Associates.
White, L., N . Spada, P. M. Lightbown, and L. Ranta. 1991. 'In put
Vygotsky, L. S. 1978 . Mind in Society. Cambridge, MA: Harvard enhancement and syntactic accu racy in L2 acquis ition.' Applied Li ngu istics
University Press. 12/4: 416-32.
Wajnryb, R 1992 . Classroom Observation Tasks: A Resource Bookfor Wode, H. 1978 . 'D evelopment al sequences in naturalistic L2 acquisition'
Language Teachers and Trainers. Ca m bridge: Camb ridge University Press. in E. M. H atch (ed.): Second Language Acquisition: A Book o/Readings.
Walker, R 20 10. Teaching the Pronunciation 0/English as a Lingua Franca. Rowley, MA: Newbury House. pp . 10 1-1 7.
O~ford: Oxford University Press. Wode, H. 1981. ' Language acqui sitional universals: A unified view of
Watson-Gegeo , K. A. 1992 . 'Thick explana tion in the ethnograph ic study language acqu isition' in H . Winirz (ed.): Native Language and Foreign
Langu age Acquisition. N ew York: Annals 0/the New YorkAcademy ofSciences
of child socialization: A longitudinal study of the p roblem of school ing
for Kwara'ae (Solomon Islands) child ren' in W A. Corsaro and P. J. M iller 379. pp. 218-34.
(eds .): Interpretive Approaches to Children's Socializa tion. Special issue ofNeu: Wong Fillmore, L. 1979 . 'Individ ual differences in second language
D irectionsfor Child Development 58: 5 1-66. acquisition' in C. J. Fillmo re, W -S. Y. Wang , and D . Kemp ler (eds.):
Individual Differences in Language Ability and Language Behavior.'
Werker,]. E, w: M . Weikum, and K. A. Yoshi da. 2006. 'Bilingual
speech processing in infants and adu lts' in P. McCardle and E. Hoff (eds.): New York: Acade mic P ress, pp. 203 -28. .
Childhood Bilingualism: Researchon Infancy through School Age. C levedon: Wong Fillmore, L. 2000. 'Loss of family languages: Should educarors be
Mu ltilingual Matters. pp.1 -18. concerned ?' Theory into Practice 39/4: 203 -10.
Wesche, M . B. 198 1. 'Language aptitude measure s in scream ing, matchin g Yule, G. and D . Macdonald. 1990. 'Resolving referential conflicts in
stu dents with metho ds, and diagnosis oflearni ng pro blems' in K. Diller L2 interaction: The effect of pr oficiency and interactive role.' Language
(ed. ): Individual Differences and Universals in Language Learning Aptitude. Learning 40/4: 539-56.
Rowley, MA: Newbury H ouse. pp. 119-39.
Zimmerman, C. B. 2009. WOrdKnowledge: A Voca bula ry Teacher's
White, J. 1998. 'Getting the learners' attention: A typographical input Ha ndbook. Oxford: Oxfo rd University Press.
enhancement stu dy' in C. Dou gh ty and J. Williams (eds.): Focus on Form in
Zobl, H. 1980. 'The formal and development al selectivity of Ll influence
Classroom SLA . Cambridge: Cambridge University Press. pp . 85- 113.
on L2 acquisition.' Language Learning 30 / 1: 43-5 7.
White,]. 2008. 'Speeding up acquisition of his/her. Explicit Ll/L2
Zobl, H. 1984. 'Th e wave mo del of linguistic change and interlanguage
contrasts help' in J . Philp , R. Oliver, and A. Mackey (eds.): Second Language
systems .' Studies in Second Language A cquisition 6/2: 160-85.
Acquisition and the Younger Learner: Child's Play? Amsterdam : John
Benjamins. pp. 193- 22 8.
White, J. and P. M. Lightbown. 1984 . 'Asking and answering in ESL
classes.' Canadian Modern Language Review 40/2: 228-44.
White, L. 1987. 'Againsr comprehensible input: the inp u t hypothesis an d
the development ofL2 competence.' Applied L ingu istics 8: 95-11 0.
White L. 1989. Universal Grammar and Second Language Acquisition.
Amsterdam: Joh n Benjamins.
White, L 1991. 'Adverb placeme nt in second language acquisition: Some
effects of positive and nega tive evidence in the classroom .' Second Language
Research 7/2: 133-61.
INDEX

Glossary entries are shown by 's' after the page number.

accent, in second lan guage 69-70, 7 [, 90. 94

alcohol. effect on pronunciation 85

ACCESS (A uto matization in American Sign Language (ASL) :1.3. :1.13g

Communicative Contexts of Essential anxiety 85, 86. 100, 101, 106

Speech Seg me nts ) II7 aptitude, lan guage learning 78. 80-3

accessibility hierarchy 54-5, 181, 213g

aspect (of verbs) )5-6

accuracy o rder 46--8, 113g

arre ntio n

acquisition/learning hypothesis [06, 193, 113g

attentional resou rces 108-9

actio n research IH. 2-13g

noticing hypothesis II5. 177 . 183-4,

active listen ing 16 5, 213g

189, lUg

activi ties
arrirudes, lear n ers' 87-8 .

analyse learn er langua ge .B - 4

audi o lingual approach 68. [0 4, 154-9 ,

analyse learners' questions 51-2­

195. 113g
co m pare learning co ntexts [24-5
aud ito ry discrirnination 6, 213g

exam ine the M oni tor M odel 107

auditory learners 83

exp lo re co ntexts tor seco nd lan guage

authe n tic materials 20 9

learn ing 35
auro ma ticiry 108-9

loo k at how different cues lead to sentence

avoi da nce (o f diffi cult featu res) 45, 58

incerpretarion [[2
Barcelona Age Faeror proj ect 98

look for and compare negotiation for

behaviour ism 113g U~ also audiolingual

meaning [66
approach

match pedagogical activi ties with teaching

in first language acquis ition 14-19

p roposals 191- 2­ in second language learn ing 57. I03-oj.

opinions on statements about language

bel iefs of learners 90-1. 170

learnlng 3-4
BleS (basic in te rpersonal communication

reflect o n individ ual differences an d

ski lls) 31-1

language learnin g success 99

bili ngual ed ucation programmes t71, lqg

reflect on language learnlng exp erience

bilingualism 114g

75-7
childhood bilingualism 30-3

review you r op inions 2-0[

child lan guage acquisition

supporr yo ur o pinion 2!I-ll

code switching 31

reache r-srudenr interaction transcripts

dual immersion [/4-5

130-1
'bo rrowed ' wo rds 63

try o ur the 'wu g' test 8-9

brain
additive bilingualism 33, 213g
fu nctioning and language II3
ado lescen r learners 85. 93-6 S~~ also pub erty
imagi ng II3. :1.4g
adult-child interaction 26-7
CALP (cog niti ve academic language

adve rb placem en t 58-9, 163. 185

pr oficiency) JI-1

affect 10 6
care take r talk se« child-directed speech;

affective filter hypothesis 10 6

modified input

age
characteristics of learn ers see lear ner

ad ult learners 93, 96 .156

characteristics

chi ld lan guage acq uisition 5-34

chi ld- d irect ed spee ch 16-7, 2I4g

an d instruction in seco nd lan guages 96-9

chi ld language acq uisitio n 5-34

and in rer pretario n of recasts 144

Child Lan guage Data Exchange System

of learner 9:1.-6
(C H IL DES) :1.6
250 Index Index 25 1

children
com p rehe nsion-based instruction 159- [65. an d develo pme n tal stages 180
early start for lan guage learning 93. 96-9 ,

first language 5-34

215g diffe rent first lan guages affectin g
20 4- 5

second language 92-9

com prehens io n checks 1I4 develo pm ent of negation, possessives.
EFL vs ESL CO ntexts 91, [50

cho ice motivation 87- 8

co nditional mood 170, [88 questions, 'referen ce to pas t', relati ve
elicitation I4! . 188

chunks 10 . 1II. 102. 1I4g IU alto formulaic

connectionism 28-9. 215g em bedded figur es test 83

clauses 48-56

lan gu age
content-based language reachi ng (CBLD firSt language transfer causing 'errors' 42,
Engl ish as a ling ua fran ca (ELF ) 71

clar ification req uests 114, 140. 188

215g IU also immersion programmes 44-;. 47, 163. 181. 186 , 20 5- 6
enh an ced input / in put enhancemen t 162- 3.

classroom s co nt ent an d langua ge-int egrated lea rnin g an d proc essing capab ility II6 •
217g
classroom instr uction and indi vid ual (CLIL) 147, 171, 215g
and pro nun ciatio n 69
env ironment. language 1(( also classroo ms;
d ifferences 92-6
focus o n usage [2+ [27-9. 2II- 12
cross-sect io nal stu d ies 8, 66 , 'l16g in put
classroom lear nin g 39
'get twO for one' 171-7
deaf ch ildren
amount of expos ure (tim e) im po rtant 38.
classroo m observation schemes [29-49 ,
and increasing th e tim e available for
and th e Cri tical Period Hypothesis 23
69, !l8, [48-9

2I4g
language study [48
and innatlsrn'zr
and th e behaviourist perspective 14-19

and learning pragmatics 67-8

learning verb fo rms in science class 188
and the need for interaction 27- 8
and the innatist perspective 20

methods of reachin g and lear ni ng [53-99

and pragm atics 67
deaf parents 23. 27
an d the in reractio nisr pe rspe ctive 24-5

motivation in th e classroom 88-9

an d recasts (nor notici ng) [41
declarative know ledge 109, [17. 193, 1I6g
natural acq uisitio n setti ngs [25-6

closed ques tions [47

Contrastive An alysis Hypothesis (CAH) deductive instruction 82
errors J(( alto cor rective feedback
co-construction [19
41-5, 47, 57, 68. 104 , 215g
delayed lan guage develop rnenc. zo-jo
absen ce o f erro rs may be due to
code switchi ng 31
con trol groups 157, 2[5g
descriptive stu d ies 14 3, 116g
avoidance, nor proficiency 45. 58

cognates 63, u 4g
cor p us linguistics 72, 202, 216g
developmental seq uences 217g
can be a sign of progress 4[ , 56, IIO

cognitive devel opment, an d [he

corr ective feedb ack 216g IU also recasts
accessibil ity hierarchy 54-5. 2t3g
erro r anal ysis approach 42- 3

ln reractionist pers pectiv e 2.4-5

clarificatio n requests 140
cross-lan guage sim ilar ities 45
erro rs ma de by second lan gua ge learners

co gn itive linguistics 28-9 , 110-11, 1I3-[8

contex tual im po rtan ce 143
de velo pmental erro rs 44
similar to chi ld err ors 42

cog ni tive maturity 37-8 , 214g

effect on oral prod ucti o n 194
devel o pmental fea tures 177. 217g
teacher feedback on [30

cog n itive psychology 108-[8

elicitation 141
di fferent acco rd ing to ins tr uct io n type
eth n ic group affiliation 70, 71, 89-90

co llaboration
error co rrectio n in communicative

157- 8
eth nography [49-51. 217g

collabo rative dialogues II9 . 189, 214g instructional settings 127-8

diso rders an d dela ys 29-30
executive mo tiva tio n 87-8

collabo rative interaction 168

exp licit correction 140, 208, 2Io-II
first language acquisition generally 6-12
experim ent al studies [53-4, 217g

I U also learner- learner interac tio ns;

an d first language acq uisitio n 202-3
first lan guage infl uence o n seco nd
exp licit knowledge 193. 212

pai r work
and 'get it righ r in th e en d' 182-95
lan guage 57, [80--[
extroversion 84--6

com m unicative confidence 86

an d the innatist pe rspective 105
grammatical mo rphemes 7-9, 46-8. [57-8
false cog nates 63

CLIL (content and language-integrated

and the interaction hypothesis 167
movement through (seco nd language
feedback lee co rrective feedback

lear ning) 147, [7[. 215g

lear ner bel iefs abou t 91
de velopment) 56-7
feleles rso

CLT (com m unicative lan guage teaching) 215g

metalinguistic feed back 140-[, 143-4, 183- 4
negation 10, 48-9
field in dependent/field dependent learning

and apti rude restin g 80

nega tive feedback in learner- learner
possessive det erminers 52- 3
styles 83. 2[7g
compared [0 structure-based instruction
in teractions [70 , [71
pragmatics 66-7
first language (Lr , mother to ngu e. native
156-7
pos itive reinforcement 14
pronunciation 69
,language) 217g

and phonology 68
repetitio n 14!
q uestion formation [0-[2
acquis itio n o f, gene rally 5- 34

and pragmatics 67
self-correctio n vs other correctio n 194
of relative clauses 54-5
im po rtance of cont in ued dev elopment of

teacher-student interaction uanscripts

stu dying corrective feed back in the
secon d language acquis ition generally 45-57
L1 in im m igran r situatio ns 32-3 , 97,

132- 3
classroom 139-45
and 'teach what is teach able' 177-82, [96
174, 175, 176. 20 4

use of d isplay q uestio ns 145--6

tim ing of co rrective feed back 194
dialects I U varieties of languages infl uenced by second language (cross ­

Communicative Orientation of Language

Uptake 139. 141. 142 , 188, 224g
disorders and delays
linguis tic influe nce) 59--60

Teaching (C O LD 129
wri tten corrective feedback 144 - 5
in first lan guage acq uisition 29-30
infl uence o n second language lear n ing

com perence. lingu istic vs co mmunicative

co rrelation 77-8, 216g in learning second languages 82
57- 60 I(e cross- linguistic influence

[58-9. 215g
counterbalance hypo th esis 1+3, 1I6 g disp lay questions [30, 145-9, 21n fluency
com pe ti tio n model Ill- 13
C rit ical Period H ypothesis (C PH ) 22-4. d rills
as aurornaticity II8
co m prehensible inp ut 106, 165-71
92- 6, 216g
audiolingu al pattern drills 157- 8
learn ing befo re accu racy in
and authenti c materials 209. 215g
cross-cul tu ral research . into first language
an d behaviourist approaches II7
com m unicative ap proaches 157

'just liste n . .. and read ' 159-65

acq uisition 26-9
d rill-like dis play quesri ons 146
m istaken int erprerarion of 32
an d m odified interactio n II4
cross-lingu istic infl uen ce 59-60. 216g I U pa tte rn practice dr ills [57- 8
focus on fo rm IU fo rm -focus ed instructio n
and th e Mo nitor M odel 106-7
also Con trastive Analysis H ypoth esis 'd rip feed' approach to instr uctio n [48 , 10 5
focus on m eaning I U alto comm un ica tive
co m pr ehensible outpur hypothesis II4, 1I9. (CAH) du al im m ersio n 174- 5
lan gu age teaching (C LD ; co n ten t­

165. 215g bi-directional 59-60

based lan guage teaching (C BLT )

 and the beh aviouri st perspective [8-1 9

immersion p rograsnmes 79-80, 128. 142. II!,
inte lligibility. as goal of pronun ciatio n 7t learn er-Iearne r interactions 119, 128, 135-9.

and the competition model 111

172-5. t87-9 , 219g
intensive ESl 148. 162. 219g 167-7° .209

'get it right in th e end' 182-95

immigrants
intensive vs 'dr ip feed' instruction models learning condi tio ns 38-4 0

and lack of accuracy 195

im po rtance of contin ued development of
148, 2°5 lear ning vs acquisition 106, 193

negotiation for meaning 14, 128, 130,

LI 32-3 . 97. 1/4 , t75. 176 , 204
interacrio n
'let's talk' 165- 71

165--6. :l.2lg second language acquisition 79 . 89-90.

collabo rative d ialogues II9. 189
lingua franca , English as I!

and pronunciation 70
94,95. 171
co nversational interaction 14, 165- 71
longitudinal studies 7-8. 9. 66. 89, 98, 149.

foreigner calk 39. 217g

implicit knowledge 193 •
importance of interaction in first language
nog
Foreign Language Oassroom Anxiety Scale 35
individual differences 75- I0 1
acquisition 27-8
meaning. focus on I~~ focus on meaning
foreign language learning 217g
inductive instruction 81-2
interaction hypothesis 1I4. 1I8. 165. 167,
memory
form-focused instruction 213g U:~ also
information-processing model 108-10 , 119g
[69-70. 119g
and the information processin g m odel
corrective feedback
information questions (gen uine questions)
int eractionist/ developmental perspective,
1°9-10

beneficial in content-based instruction 173

t30 • 145-9
child language acquisition 24- 5
memorization 103-4

and 'get it right in the end' 182-95

inhi bitio n 84-5
lack of in ' just listen ... and read'
retrieving new words 70

useful in some circumsrances 196-7

Initiation/Response/Evaluation (IRE)
approaches 159--65
working memory capaci ty 80-1

formulaic language 218g IU also chunks

exchan ges 126
lear ner-learner interactions 1I9, 128,
meta-anal ysis 193, !log
at early stages of learning 56
innatisr perspeccive on child language
135-9. 169-70. 209
meralinguistic awareness n og

and information-processing model 109

acquisition 20-4,.104- 7. 219g
modified interaction iU modified input
and bilingualism 31

much language learned as III

input l l9g I~~ also comprehensible input;
need to rake social setting in to account
rnetalinguisric feedback 140-1, 143, 183-4

in seco nd language errors +1­ modified input when researching 79

and older learners 38

and wh- questions I~

flood 162-3. 219g inte rfere nce u< c ross-linguistic influence
in pre-sc hoo lers 13

fossilization 43. t 57. 218g

amo unt o f expos ure (time) im po rtan t 38, inrerlanguage 220g
mimicry 103-4 se« also i rn i rario n

front ing, qu estion formation u , 49-51, 17S

69 , 128, t48- 9
ge nera lly 43
mi tigation 66, 6 7. l 20g

func rion wo rds 7. 43, 218g

enhanced in purli n pu t enhancement 162- 3
and 'get it righ t from the beginning' 157-8
Modern Language Ap titud e Test (MLo\T)

gende r, grammatical 29. 52-3, 187

freque ncy of encounte rs with new
in terlan gu age pragm atics 65-6
8o
'Genie' 22- 3
ma terial is im po rtant 62
Interlocu to rs iU interaction
modi fied input H Og

generalisatio n 42. 4+ 218g

in the in n arisr perspective on ch ild
' int ern atio nal adoprees' 24
benefits o f 207-8

genu ine questions 130, 1+5-9 , 218g

language acquisition 21
international vocabulary 63
chi ld-directed speech 26-7

gestural (sign) language 23

in put flood 162-3. 219g
investm ent (of learner in learn ing) 89-90
in communicative langu age teaching

'get it right fro m the beginning' 154-9.

input hypothesis 106
IQ (int elligence quotient) 78-80 . l 0 3
(C l T) 127-8

166.195
input processing n6
isolated form-focused instruction 191
in co nte nt-based language teaching

'get it right in the end' 182-95

the ' log ical problem' o f seco nd language
'j USt listen ... and read' 159--65
(C fsl T) 173-4. 176

'get twO for one' 171- 7

acqu isition 1°5
kinaesthetic learners 83
enh ance d input 162- 3. 2l7g

gram mar
struct ured input 81- 2
language acqui sition order see developmental
foreign er talk 39. 2l7g

in babies' first utterances 7

instruction (teach ing)
seq uences in learner-learner interaction s t70

grammatical gender 29, 52- 3. 187

age and second langu age instruction 96-9
language disorders and dela ys 29-30 modified interaction II4, zzog IU also
gram m aticality judgements 95. w 6. 218g
amount and distribution of time available
language distance 69 I U also cross-linguistic corrective feedback; modified input

grammatical morphemes. child

1.j.8
influence
in natural acquisition sett ings 1].6

acquisition of 7-9, 218g

'd rip feed' vs int ensive [48, 205
language learning aptitude 80-3
and comprehensible input II4

grammatical morphemes. second

explicit instruction and pragmatics 67-8
language-related ep isodes (LREs) 189, H Og
teacher talk 39

language acquisition of 46-8

explicit instruction and pronunciation
' language socialization' 27
modified interaction II5, nog I« also
learner beliefs about grammar instruction 9t
70-1
lea rner characteristics
co rrective feedback; modified input

obligatory contexts 46-8, 158, l21g

explicit instruction an d vocabulary
generally 37-8
modified outpUt 1I5. 119.165

grammar trans lation methods 80, 126,

learning 64
age 92--6
Monitor Model (i +1) 106-7. 118

154-9. 195, 218g

and the innatist perspective 105
a~itudes and motivation 70 , 87-9
morphemes, grammatical 7-9, 46-8 , nlg

grammaticality judgement 95. 106. 218g

instructed vs uninstructed learners 46-7
and classroom instruction 92
rnorher tongue see first language

group work 170. 209 U:~ also pair work

instructio nal programmes and aptitude
identity and group affiliation 70 . 89-90
motivatio n

habits 14, 104

p rofiles 8t
Intelligence 79-80
difficulties in research ing 78

human in p ut , need for 6. 28

instructional settings 124-9 I~< also
lan guage learning aptitude 80-3
and excessive co rrectio n 208

ident ity 70. 89-90

classrooms
learner beliefs 90-t. 170
motivation retrospection 87-8

imitation
teachabiliry hypothesis 177-82
learning styles 83-4
not always a predicto r of success 203-4

and first language acquisition 15-t9

instrumental motivation 87. 219g
perso nality 84-6
teachers' motivational practices 88-9

and the behaviourisr perspective 14-1 9

integrated form-focused insrructio n 191
readiness to learn 180-1
,IS variab le in second language learning

and second languages 41. 201-2

int egrative motivation 87. 2t9g
researc h methods 75- 7
87-8

inrelligence 7 9-80, 203

254 Index
Motivation Orientation of Language
Teaching (M O Ln 88
rnulticornperence (as better goal th an native
abiliry) 96
multip le int elligences 80
native lan guage see first lang uage
native-like ability 68, 71, 96, azrg
socio linguisrics
nat ive speak er 3, 24, 4-2 , 5r; 67 , 69 , 7r, 94,
pre-schoo lers lear n ing social forms of
96, 97, 114 , 126, 145, 22lg
natural acq uisition setti ngs 123- 9
narural order hypothesis 106, 221g IU also
developm ental seq uences
negatio n '
chi ld acq uisitio n of 9-10
secon d lan guag e learning 48-9
nega tive feedback I U clar ificati o n req uests;
correcti ve feedb ack
nego riatio n for rnean ing. tra , 128, 130, 165-{).
221g
nego tiation of fo rm 139 , zz rg II!I! also
form-focused ins tr uctio n
neurologi cal resear ch I13
no ticing hyp othesis II5, 177 . r83-4, 189, 2llg
object pr onou ns 164
learn er in teractions
o bligatory con texts 46-8, 158. az rg
su b mersi o n m ethods 32
obs ervatio n
su btra ctive bili ngu alism 32-3. t74, 177 . 223g
ethno graphy r49-51
sup rasegm enrals 68. 70 . 223g
ob servatio n schemes 129-48
op en q uesrio ns 147
or der of acquisition ul g I I!I! developmen tal
seq uences
cvergene ralizar io n errors 221g
th e st udy ) 189
an d the behavio ur ist perspect ive 18- 19
researchi ng teache r- st ude nt int eractions
in first language acq uis irio n 12
and the info rma t io n processing model 110
in pronunciation 68-9
in second language lear ning gene rally 44
pair wo rk 168-9 , 170 . 209
paraphras ing 1I4 . 147 II!I! also recasts
parent-child interacti on 26-7
pasl tense 55-{). 173
patt ern p ractice drills 157-8. 22rg
patt erns in lan guage. lear nin g 18
pee r group, lear n ing fro m I19 II!I! also
lear ner-lear ner im eracrions
perceptio n of so un ds 69 , 70
perceptually-b ased lear ning sryles 83
perfo rm an ce 12, 45, 71. 79. 85. 94 , 96, 107 ,
117, 186, u lg
personal characterist ics of lear ners II!I! lear ner
characteristics
perso nal ity 84-6
pho nemi c d isti nctio ns 222g
infants' ab ility 6
seco nd language learners 69
phonology 68- 71, 206
po lyglor savam 83
pos itive reinforcement 14
possessive de terminers 46-8. 52-3 . 163
power relatio nships 70. 89
practi ce
an d the behaviourls t perspect ive 14- 19
and cogni tive perspectives II7
pragma rics 222g
im portance of learning 207
imerlanguage p ragmarics 65- 6
po lite prono uns 173. 187
registers (differem lan g uag e fo r d ifferem
situati o ns) 14. 173, r87. 222g
an d seco nd lan gu age lear ning 65-8
pre-school years. an d lan gu age acq uisirion
12-13
pri vat e speech II8. 142. 222g
procedural knowled ge 109. 117. 193, 222g
processabiliry theory 1I6-17. 177, 182, 222g
p rocessing capac ity II6
processing instr uc tion 16 2, 164-5. 122g
profi cien cy
an d bil ingual ism 31- 2
diffic ulties in resear ch ing 78-9
learner language an d proficiency level
r67-8
pro m pts
clar ificatio n requests I14. 140, 188
elicita tion 141. 188
enco uragi ng self-co rrection 188-9. 194
p ron un cia tion 68-71
ro le in mo tivation 204
psych o logical theories II!/! behaviourism:
teacher-lear ner in teractio ns 129-35. [45
info rmario n proc essing mod el
pu be rty 94-5
q uest io ns
child acquisirio n of ro-ta, 19
display q uestions r30
seco nd lan guage lear nin g 49-5r. 180
teacher q uestioning in th e classroo m 145-9
rate oflearnin g 96 , 97 . 222g
readi ness ro learn [8c>-1
reading
effect of literacy o n firsr language
deve lopmen t 13
'just listen" . an d read' 159-65
and practice II7
p ro blems in lear ning to read 29-30
read ing as a sourc e of voca bulary growth
63-.f, 206
readi ng m aterial fo r learn ers 161-2
recasts 2Ug
comparison o f feedback th at co rrects vs
p ro m p ts 194
in co nten t-base d classroo ms 139-45
in immersion classrooms 188- 9
in learner-learn er inreractions 170 , 171
learners may not notice r95
mos t common kind of feed back 21C>-1I
for 'readi es' and 'un read ies' 180
stress (p rosodic) r44
refere nc e ro past 55-6
referen tial questions (gen uin e q uestio ns) 130,
145-9
reflexive pronouns 2C>-1
registers (di fferem lan guage to r differen r
situations) 14, 173, 187. 222g
rela tive cla uses 54-5, 181-2
repe ti tion 141 I I!I! also imi ta tio n; prompts
req uests . as p ragm atic feature 66
research m ethods
acti on resear ch r54. 213g
(classroo m) o bservatio n sche m es 129-49
co rrelation 78
corpus Iingu isrics 72, 202, 216g
cross-secrional stud ies 86. 216g
descriptive stud ies 143, 216g
difficulry in findi ng co m parison groups
157
ethnography 149-51
experime ntal stud ies r53-4
lon gitu d inal st udies 86
observa rio n schemes 129-49
q ualita tive researc h 86. 129. 149, 154. 222g
q uan ti tative resear ch 86. 129, 153. 222g
researcher paradox (researcher affecti ng
129-35
sam pling issues 72
for st udy ing learner characte ristics 77-8
rest rucruring 1°9-10. 222g
routines (form ulai c lan guage) II!/! fo rm ulaic
langu age
scaffolding 25, u 8, 146-7, 223g
seco nd lan gua ge 159, 223g
school years , an d first lan guage acq uisitio n
13- 14
segrne n tals 68. 70. 123g
self-co rrec tion vs o ther cor rect ion 194
self-re pe ti tio n U4
sensitive pe riod (Critical Period H ypothesis)
22-4,9 2- 6
seq uential bilingual s 30
sho rr-re rrn memory 8C>-1
sign lan guage 23
Sign ifican t di fference [59, 189 , 223g
silence. use of in classroom I.p . 149
sim pli fica tio n 44 . 114, 223g
Index 255
simplified readers [61
sim ultaneous bilinguals 30
social int eractio n
and th e interactionist perspective 24- 5
soc ializa tio n, lan guage 149-50
soc ioc ultural th eo ry U8-19, 146, 169. 223g
language /2- 13
socio lin guistic fo rms 18i - 8
sociopo litical change r5C>-1
speech an d langu age diso rders 29-30
stages II!/! deve lop mental seq uences
Standard var iery r4. 223g
stim ulat ed recalls 189
str ess (an xiety) 85, 106
stress (prosodic)
im portance in making yo urself
unde rstood 206
as part of ph on o logy 68. 69 , 70 , 7t
in recasts 144
str uctural gradi ng 223g
srudenr-srude nc in teractio ns II!/! learn er­
large t lan guage 223g
task-based langu age teaching (TBLn 67. 165,
190, u3g
teach ers
teacher questio ning in th e classroom
45-9
teachers' mo tivati on al practices 88-9
teach er talk 39. 224g
teach in g, classroo m I I!I! audio lin gual
app roach ; co mmun icative language
teaching (CLn ; co ntent -based
lan gu age teach ing (CBLn ; focus on
meaning; form -focused instruction;
grammar translati on me thods; task­
based language teachin g (TB Ln
'teach w hat is teach able' 177-82. 210
technology
co m puter -base d tools fo r sam pling 72
in put for ch ild lang uage acq uisitio n needs
to be h umans not electronic 6, 28
'teleg raph ic' sen te nces, ba bies' 6-7
timing
amount of exposure (tim e) in new
language impo rtant 38, 69, u S, 148- 9
256 Index

of beginning instruction in second

language 93, 9~, 204-5
of corrective feedback [94
of form-focused instruction [9[
transfer 224g see cross-linguistic influence
transfer-appropriate processing (TAP) 110,
• 191, 2~
rwo for one 171-7
typographical enhancements of input [63
UG (U niversal Grammar) 20, 104-5 , 224g
uptake [39. 141. 42, 188, 224g
usage-based learn ing 1I0-I1
usage-based perspectives on child lan guage
acquisition 28-9
variab les, personal 224g f~~ learner
characteristics
variational features 116, 177--9, 224g
varieties oflanguages 14, 31, 71. 206 . 224g
'Victor' 22
visual learner s 83
vocabulary
amou nt needed for conversation 6[,
162,206
can be taught any time 178
first language development of 14
growth through reading 162
learning straregies 64­
second language learning 60-4
wait time 147
wh-words to-l2, +9-51
willingness co communicare (\vro 86, 224g
word identification 61-2
retrieval of word meanings 109
word order
adve rb placement 58-9
basic word orde r and predictable
development paths 177. 178, 179
word order and meaning 111-12
working memory capaciry 80-1, 225g
'wug rest' 8-9
younger the benet (for starting second
language instruction) 93, 96-9,20+-5
zone of proximal development (Z PD) 25,
118, 225g