The Foundations of Teaching English As A Foreign Language (Paul Lennon) (Z-Library)

The Foundations of Teaching English
as a Foreign Language
An introductory textbook that assumes no prior knowledge of linguistics

or second language acquisition, this book presents a comprehensive
overview of the theoretical foundations, methods and practices of Teach-
ing English as a Foreign Language (TEFL) for pre-service teachers.
Lennon covers the theoretical bases for TEFL and addresses second
language-acquisition research, past and present EFL teaching methodol-
ogy, as well as psychological and social approaches to individual lan-
guage-learner variation. Further chapters provide extensive yet accessible
coverage on essential foundational topics, including chapters on pronun-
ciation, grammar, vocabulary, literature and testing. Offering a socio-
cultural approach in which the teacher is seen as a facilitator and
supporter of students’ self-directed learning, this text provides the pro-
spective teacher with the knowledge and skills to be an effective educator
in the EFL classroom.
The targeted EFL focus makes this book ideal for pre-service teachers
and for teacher training programmes around the world. Each chapter
includes a Food for Thought section with questions for reflection and
a Further Reading list.
Paul Lennon is Professor of TEFL, Bielefeld University, Germany

(retired).
The Foundations of
Teaching English as a
Foreign Language
Paul Lennon
First published 2021
by Routledge
52 Vanderbilt Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2021 Taylor & Francis
The right of Paul Lennon to be identified as author of this work has
been asserted by him in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical, or
other means, now known or hereafter invented, including photocopying
and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and
explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Names: Lennon, Paul, 1951- author. | Routledge (Firm)
Title: The foundations of teaching English as a foreign language /
Paul Lennon.
Description: New York : Routledge, 2020. | Includes bibliographical
references and index. |
Identifiers: LCCN 2020007929 (print) | LCCN 2020007930 (ebook) |
ISBN 9780367250959 (Hardback) | ISBN 9780367250942 (Paperback) |
ISBN 9780429285998 (eBook)
Subjects: LCSH: English language–Study and teaching–Foreigh speakers. |
Second language acquisition–Study and teaching. | School management
and organization.
Classification: LCC PE1128 .L4532 2020 (print) | LCC PE1128 (ebook) |
DDC 428.0071–dc23
LC record available at https://lccn.loc.gov/2020007929
LC ebook record available at https://lccn.loc.gov/2020007930
ISBN: 978-0-367-25095-9 (hbk)

ISBN: 978-0-367-25094-2 (pbk)
ISBN: 978-0-429-28599-8 (ebk)
Typeset in Goudy
by Swales & Willis, Exeter, Devon, UK
Contents
Detailed Table of Contents vii

List of Figures xv
List of Tables xvi
Preface xvii
Acknowledgement xxii
1 Language Acquisition and Language Learning 1
2 Language Teaching Methods 31
3 Individual Learner Differences 59
4 Facilitating Classroom Learning 90
5 Pronunciation 121
6 Grammar 150
7 Vocabulary 180
8 Listening and Speaking 209
9 Reading and Writing 236
10 Literature in the Language Classroom 265
11 Language Testing 293
Index 323
Detailed Table of Contents
1 Language Acquisition and Language Learning 1

Settings 1
First Language Acquisition Settings 1
Bilingual Settings 5
Classroom Settings 7
Language-Learning Error 8
Contrastive versus Non-Contrastive Error 8
Overt versus Covert Error 9
Global versus Local Error 9
Language-Learner Language 10
The Monitor Model of Classroom Language Learning 12
Automatisation 13
Restructuring 15
Connectionism 16
Language-Learning Strategies 17
The Good Language Learner 17
Strategies of Young Children in Naturalistic Environments 18
Strategies of Classroom Learners 18
Self-Regulation, Autonomy, Motivation and Identity 19
Communication Strategies and Negotiation of Meaning 20
Socio-Cultural Theory and Interactive Language Learning 21
Food for Thought 23
Further Reading 24
References 24
2 Language Teaching Methods 31

The Grammar-Translation Method 31
The Direct Method 32
The Audiolingual Method 34
Situational Language Teaching (SLT) 37
viii Detailed Table of Contents
Total Physical Response (TPR) 39
Community Language Learning (CLL) 41
Communicative Language Teaching (CLT) 42
Task-Based Language Teaching (TBLT) 46
Cooperative Language Learning (CLL) 47
Self-Directed Language Learning (SDLL) 48
Competence-Based Language Teaching (CBLT) 51
The Post-Methods Era 53
Food for Thought 55
Further Reading 56
References 56
3 Individual Learner Differences 59

Aptitude for Language Learning 59
The MLAT 59
The LAB 63
The CANAL-F Test 64
Aptitude for Language Learning and First Language Acquisition 65
Motivation 65
Goal-Driven Behaviour 65
Social-Psychological Approaches 66
Cognitive Approaches 68
Process Approaches 70
Demotivation and Amotivation 71
Learner Attributions for Success and Failure 73
Personality 74
The Four Humours 74
The Meyers-Briggs Type Indicator (MBTI) 74
Anxiety 76
Risk-Taking, Inhibition and Sensitivity to Rejection 80
Willingness to Communicate (WTC) 80
Self-Esteem and Self-Efficacy 80
Tolerance of Ambiguity 81
Empathy 81
Cognitive Style 81
Field-Independent versus Field-Dependent Styles 82
Deductive versus Inductive Reasoning Styles 82
Global versus Particular Reception Styles 83
Sharpening versus Levelling Memory Styles 83
Impulsive versus Reflective Response Styles 84
Random versus Sequential Processing Styles 84
Detailed Table of Contents ix
Synthesising versus Analytic Styles 84
Language-Learning Style 84
Style Preferences 84
Style Stretching 85
Style Mismatches 85
Food for Thought 86
Further Reading 86
References 87
4 Facilitating Classroom Learning 90

The Orchestration of Classroom Interaction 90
The Syllabus 90
What Is a Syllabus? 90
The Grammatical Syllabus 91
The Lexical-Grammatical Syllabus 92
The Situational Syllabus 93
The Topic-Based Syllabus 93
The Notional-Functional Syllabus 94
The Output-Based Syllabus 94
The Multi-Track Syllabus 95
The Cyclical Syllabus 95
Needs Analysis 96
Teaching Materials 98
Information and Communication Technology (ICT) 100
Online Resources 100
Presentation Software 102
Virtual Learning Environments (VLEs) 103
Online Tools 103
Internet Research 107
Hand-Held Devices 108
Classroom Interaction 109
Teacher and Learner Classroom Roles 109
Input Modification and Interactional Modification 110
Learner Cooperation and Self-Direction 112
Provision of Feedback on Error 112
Some Rules of Thumb for Classroom Interaction 116
Food for Thought 117
Further Reading 118
References 118
x Detailed Table of Contents
5 Pronunciation 121
Whose Pronunciation? 121
Describing the Sounds of English 123
The Phonemes of English 123
Allophones 125
Producing the Sounds of English 127
Speech and Breathing 127
Voicing 128
The Monophthongs 128
The Diphthongs 132
The Consonants 133
Stress, Rhythm and Intonation 138
Stress 138
Rhythm 141
Intonation 142
Classroom Pronunciation Practice 144
Practice of Sounds in Single Words 144
Practice of Sounds in Connected Speech 146
Further Reading 148
References 149
6 Grammar 150
Inflectional Morphology 150
What Is Inflectional Morphology? 150
Verb Inflection 150
Noun Inflection 153
Adjective Inflection 155
Pronoun Inflection 155
Syntax 156
What Is Syntax? 156
The Noun Phrase 157
The Verb Phrase 157
Adverbials 161
Verb Complementation 161
Negative, Interrogative and Imperative Sentences 164
Simple, Compound and Complex Sentences 166
Minor Sentences 166
Approaches to Teaching Grammar 167
Pros and Cons of Explicit Grammar Teaching 167
Acquiring Grammar in the Classroom 168
Detailed Table of Contents xi
The Deductive Approach 169
The Inductive Approach 170
Comparing Deductive and Inductive Approaches 171
The Genre Approach 172
The Lexical Approach 172
The Cyclical Approach 173
Awareness-Raising 174
Grammar Games 174
Songs for Grammar Practice 175
Further Reading 178
References 178
7 Vocabulary 180
Word Formation 180
Word Meaning 182
Lexical and Grammatical Words 182
Dictionary Meaning and Meaning in Context 182
Criterial Attribute Theory 183
Prototype Theory 183
Words with Fuzzy Edges 184
Polysemy 184
Referential and Associative Meaning 185
Sense Relations 185
Collocational Relations 187
The Mental Lexicon 188
The Container Model of Lexical Storage 188
The Spreading Activation Model 190
What Information Is Stored for Lexical Items? 191
Stored Lexicalised Sentence Stems 193
Teaching Vocabulary 193
Vocabulary Acquisition: an Incremental Process 193
Enhancing Informal Acquisition of Vocabulary in the Classroom 195
Explicit Teaching of New Vocabulary 195
Vocabulary Practice and Consolidation 196
Teaching Affixes (Prefixes and Suffixes) 198
Teaching Lexical Collocation 200
Teaching Multi-Word Lexical Units 201
Teaching Lexicalised Sentence Stems 202
xii Detailed Table of Contents
Further Reading 205
References 205
8 Listening and Speaking 209

The Spoken Mode 209
Listening 211
Why Listen? 211
Listening Difficulties 212
Getting Started 213
Moving On 215
Checklist for Selecting Listening Texts 216
A Three-Phase Lesson Structure 217
Songs as Texts for Detailed Listening 219
Speaking 221
The Limitations of Traditional Classroom Discourse 221
Spoken Classroom Activities 222
Post Hoc Evaluation of Spoken Activities 232
Further Reading 234
References 234
9 Reading and Writing 236

The Written Mode 236
Reading 238
Reading Purpose and Reading Technique 238
A Three-Phase Lesson Structure 241
Alternatives to Traditional Reading Comprehension Classes 248
Writing 249
Written Genres 249
Four Major Approaches to Teaching Written Genres 249
Beginning Writing 255
Writing Short Descriptions and Narratives 256
The Transition to Discursive and Expository Writing 258
Using Pictures as a Stimulus for Writing 260
Further Reading 262
References 263
10 Literature in the Language Classroom 265

A Wide Definition of Literature for the EFL Classroom 265
Three Sets of Reasons for Reading Literature in the Language
Classroom 266
Detailed Table of Contents xiii
Language-Learning Benefits 266
Motivational Benefits 267
Intercultural Benefits 267
Short Poems 268
Suitable Poetic Genres 268
A Three-Phase Teaching Approach 269
Humorous Verse for Children 271
Humorous Verse for Adults 272
Descriptive Poems 272
Love Poetry for Advanced Learners 273
Polemical and Reflective Poems for Advanced Learners 275
Short Novels 276
Selecting a Short Novel 276
What Makes a Short Novel Too Difficult? 278
Screening Short Novels for Language Difficulty 279
Plannning a Three-Phase Reading Course 281
Reading Logs 285
Project Work 287
Links between Reading Logs and Project Work 287
A Three-Phase Plan 288
Four Possible Problem Areas 289
Nine Guidelines for Supervising Project Work 290
Further Reading 291
References 291
11 Language Testing 293

Approaching Language Testing 293
Teachers and Testing 293
Testing Purposes 294
The Three Phases of Testing 296
Testing Techniques 297
Criterion-Referenced versus Norm-Referenced Testing 302
Test Selectivity 303
Test Takers and Tests 305
Evaluating Language Tests 305
Validity 305
Reliability 307
Scorability 310
Economy 310
Administrability 311
xiv Detailed Table of Contents
Discriminability 311
Backwash on Teaching 311
Written Language Tests 311
Piloting the Test 311
Advance Information for Test Takers 312
Invigilation 312
Marking (Scoring) Procedures 313
Oral Language Tests 315
Techniques 315
Marking (Scoring) Procedures 317
Six Rules of Thumb for Oral Testing 319
Some Alternatives/Supplements to Traditional Tests 319
Teacher Reports 319
Continuous Assessment 320
Portfolio Assessment 320
Peer Assessment 320
Self-Assessment 320
Further Reading 322
References 322
Index 323
Figures
5.1 The organs of speech 127

5.2 The GB monophthongs 130
5.3 The GB diphthongs 132
Tables
5.1 Transcriptions for the 20 GB vowels 124

5.2 Transcriptions for the 24 GB consonants 125
5.3 Categorisation of the GB consonants 134
Preface
This is an introductory book about Teaching English as a Foreign

Language (TEFL). It aims to place the practice of TEFL on firm theore-
tical foundations as provided by linguistics, psycholinguistics, sociolin-
guistics, social psychology and the history of language-teaching
methodology. It is primarily intended for use in TEFL training courses
in universities and colleges. No prior knowledge of linguistics, language-
acquisition research, social psychology or language-teaching methodol-
ogy is assumed. All that is assumed is knowledge of basic traditional
grammatical terminology. The book is suitable for use in both pre-
service and in-service courses, although teacher trainers will use it in
different ways according to whether they are working with trainee
teachers, novice teachers or experienced teachers. Inexperienced trainees
will clearly require more support from their trainers than will experi-
enced teachers. The book can also profitably be used for self-study.
At the end of each chapter there is a Food for Thought section with
questions for critical reflection. Some of these are suitable for self-study;
some are marked as group tasks. A small number of questions are
intended especially for practising teachers, who are asked to draw on
their teaching experience, but most questions are aimed at novice or
trainee teachers and do not presuppose teaching experience. In some
cases the reader is asked to draw on his or her experience as a language
learner. Group tasks and questions are generally so designed that the
group as a whole will benefit from the pooling of individual contribu-
tions from group members, whether experienced, practising teachers,
novice teachers or trainee teachers, since experience of both language
teaching and language learning is always diverse, and trainee teachers may
be inexperienced teachers but they are usually experienced language
learners.
Only those who have never stood before a class at all may well wonder
what all the fuss is about in this book, as I myself would once have done.
Surely you just go in there and teach them, don’t you? This is one of the
reasons for devoting a chapter (Chapter 2) to language-teaching methods
xviii Preface
of the past and present, namely to demonstrate how very different
language teaching can be and has been over the years. What most of us
regard as ‘the way’ to teach is, ironically enough, probably more influ-
enced by the way we ourselves were taught than by anything we learned
in our formal teacher training. At the very least, any student who is using
this book as a course book should make sure that he or she has at least
observed – and far better actually taught – a few classes before the course
starts or while the course is running. Similarly, teacher trainers using this
book will find that their trainees will benefit far more from it, and be
more motivated to read it, if they have done even just a few hours’
teaching before the course starts.
Each chapter in the book is reasonably self-contained and can be read
of and for itself, but the book’s overall approach only becomes clear
when it is read as a whole and in the order in which it was written. The
first three chapters provide the theoretical bases upon which good
foreign language-teaching practice is grounded, and the teaching
approach advocated in the book emerges from these chapters. It is
a socio-cultural approach in which the teacher is seen as a learning
facilitator, and it is these introductory chapters which enable the reader
to grasp the book’s underlying rationale. Generally speaking, then,
unless readers/course participants already have a firm grounding in the
subject matter of the first three chapters, it would be inadvisable to omit
them. One might be tempted to think that by plunging in at Chapter 4
one is getting into the nitty gritty of classroom teaching from the start.
However, it is the foundations established in Chapters 1 to 3 which
underpin the discussion of teaching approaches, procedures and techni-
ques in subsequent chapters, and reference is often made back to the
early chapters.
Chapter 1 reviews aspects of first and second language-acquisition
research most relevant for language teaching, especially the effects of age
and setting. It also explains the construct of language-learner language,
sketching out the developmental processes of automatisation and restruc-
turing, and outlining the individual learning and communication strate-
gies learners employ, including negotiation of meaning in social
interaction.
Chapter 2 critically and selectively reviews the chequered history of
language-teaching methodology. The idea is not to recommend out-
moded methods but to help readers perceive what the strengths and
weaknesses of individual methods were and to take stock of the present
state of affairs, sometimes referred to as the post-methods era. Brief
mention is made of Content and Language Integrated Learning (CLIL),
which relies to a greater or lesser extent on English being acquired
through the learning of another subject or subjects in the English
medium, with or without some focussed language teaching as well.
Preface xix
Chapter 3 presents a synopsis of some cognitive, social and psycholo-
gical differences among individual learners which may influence their
preferred learning style, their motivation and their language-learning
achievement. Part of the thesis of this book is that there is not a single
best way to teach English, but that the challenge is to find suitable
approaches, techniques and materials for specific groups of learners
and, ultimately, each individual learner. One size does not fit all.
Chapter 4 takes up the idea of the EFL teacher as a learning facilitator
and makes the transition from principles to practice. The teacher is seen
as having various tools available to exploit the classroom setting’s full
potential as a language-learning environment and offset its disadvantages
as a language-acquisition environment compared to ‘real-world’ settings
where English is used. These tools include the syllabus, needs analysis,
teaching materials, Information and Communication Technology (ICT)
and the orchestration of classroom interaction to facilitate learning. Each
is considered separately. Practical guidance on how to react to spoken
error in the classroom is provided within the context of a socio-cultural
approach to classroom learning, which emphasises the importance of an
environment of constructive support for the learner which is furnished
by the teacher and classmates. In subsequent chapters it is assumed
readers are familiar with the matters dealt with in Chapter 4.
Chapters 5 to 7 are devoted to the three traditional levels of descrip-
tion of language, namely, pronunciation, grammar and vocabulary. Each
is treated separately and is assigned its own chapter. In each case, basic
linguistic descriptive information is first provided and then teaching
approaches, including references to materials and teaching techniques,
are put forward.
Chapters 8 and 9 look at English teaching from the traditional didactic
perspective of the ‘four skills’, namely, listening and speaking (Chapter 8)
and reading and writing (Chapter 9). These two chapters are closely
connected. The differences between the written and spoken modes and
their implications for language teaching are made clear. Each individual
skill is first analysed and described with special focus on genre, style and
register differences within a functional-pragmatic approach to written and
spoken communication. Attention is given to the individual phases of
lessons which focus on listening, speaking, reading and writing, respec-
tively, with a three-phase approach to lesson planning being elaborated.
Practical advice on tasks, activities and techniques is given and sample
materials are referred to or presented and worked through.
In presenting and discussing teaching techniques and materials, care is
taken to differentiate whenever necessary according to age level (primary
school to adult) and proficiency level (beginner to advanced) of learners.
The default case, where no age or proficiency level is specified, is the
intermediate-level adolescent or young adult learner. The focus is on
general English courses rather than English for Specific Purposes (ESP),
xx Preface
but ESP is referred to at various points. Most of the materials, activities
and games described are well-established, well-tried ones and are part of
the stock-in-trade of many EFL teachers. In many cases published
sources for teaching ideas and materials are referred to. These references
are intended as suggestions only and do not pretend to be exhaustive.
The teaching ideas advocated are not intended to be provocative,
experimental or even new. Rather, the ideas are chosen because they are
based on firm theoretical foundations and have enjoyed widespread
success in many teaching contexts. I myself have used many of the ideas
in various teaching situations with different learner groups. It has also
been my privilege to observe many successful trial lessons given by
trainee EFL teachers and when I expressly recommend a certain class-
room technique, it is often one that I have seen work well in lessons
I have observed. Getting a certain procedural detail right or wrong can
make all the difference, and when I caution against adopting a certain
classroom procedure, this is sometimes because I once made this very
mistake in my own teaching.
To the book’s emphasis on interaction, cooperation and self-direction
may be added an emphasis on intercultural learning, which finds expres-
sion in a whole chapter (Chapter 10) on the use of literature in the
English Language classroom. The focus is on using short poems in class,
using short novels as class readers and on doing project work based on
literature. Suitable sample poems are suggested for children and adults at
various proficiency levels and clear guidelines on how to use them
provided. Detailed advice on how to select and use a class reader and
how to introduce learners to keeping a reading log is given. The chapter
concludes with practical advice on implementing project work.
Much of what is said in Chapter 10 applies to extensive reading in
general, not just literature. The section on project work and its implica-
tions for furthering learner self-direction and attaining the ultimate goal
of learner autonomy is central to the book’s thesis. Project work is, of
course, not limited to literary projects and ideas for non-literary, inter-
net-based project work are touched on in Chapter 4. What is said there
should be read in conjunction with what is said in Chapter 10. Thus,
Chapter 10 is central to the book’s thesis and not peripheral.
The final chapter is an introduction to the theory and practice of
language testing. It is deliberately pitched at trainee teachers who have no
experience whatsoever of language testing. It is hoped that it will awaken
an interest in language testing which can be further pursued as
a practising teacher. It explains some different purposes for testing and
different types of test, as well as giving examples of some well-known
testing techniques. Test evaluation criteria are provided, as is practical
advice on administering written and oral tests, respectively. Finally, some
alternatives and supplements to traditional testing are presented. The
chapter stresses that language testing is an integral part of language
Preface xxi
learning, is important for building (and not destroying!) learner motiva-
tion, delivers feedback on progress to learners and teachers and provides
input to future course planning. It would therefore be a pity if this
chapter were omitted. Testing has become an important part of language
teaching in many, probably most, contexts and trainee teachers need
a grounding in the basics.
In writing this book I have worked hard to steer a difficult course
between the Scylla of overloading trainee teachers with too much theory,
which they would be unable to transfer to their daily lives in the class-
room, and the Charybdis of giving them a ‘thirty-ready-to-teach-lessons-
that-always-work’ book, which would run the risk of encouraging them
to take over ready-made lesson plans wholesale without reflection, with-
out understanding their underlying principles and without considering
their suitability for a particular learner group. My aim has been, rather,
to provide trainees with a firm theoretical foundation on which to base
their teaching so that they can plan their own lessons with full cogni-
sance of the wider classroom learning environment in which they are
working, teach with a critical eye as to how learning is progressing in real
time, reflect on the impact of their teaching and modify it accordingly for
the future.
Language learners and their needs will continue to change in the future
and the trainee teachers of today will need to adapt their teaching to
different situational demands over many years to come. How those
demands will have changed in 20, 30 or 40 years cannot be predicted.
What can be done is to provide trainee teachers with professional back-
ground knowledge and to sensitise them to relevant aspects of the
environment so that they can adapt successfully to changing situations
and enjoy doing so. This will make for career satisfaction.
Acknowledgement
The passage from Bank Shot by Donald. E. Westlake on pages 244–245 is

reprinted by permission of mysteriouspress.com.
1 Language Acquisition and
Language Learning
Settings
First Language Acquisition Settings

Long ago the Danish linguist Otto Jespersen (1922: 40) noted the stark
contrast between the almost universal and rather uniform success of
children acquiring English as their mother tongue compared to the
‘defective and inexact command’ of the language often achieved by
foreign language learners, whose success is also subject to tremendous
individual variation. Children produce their first word at about the age
of 12 months. By the age of about 18 months most children have an
active vocabulary of perhaps 50 words, that is, words they can use, not
just understand (Crystal 1986: 93; Saxton 2017: 156). By 24 months
children have a command of several hundred words and by the age of
six they can at least understand perhaps 10,000 words (Bloom and
Markson 1998: 68) or even 14,000 words (Clark 1993: 13).
However, the key to acquiring a language is the acquisition of its
syntax; in other words knowing how to combine words to form utter-
ances. Otherwise all one can do is to name things. Although there is
variation in the rate (speed) of acquisition of syntax, all children seem to
follow more or less the same path (route) (Brown 1973: 272–275). The
route remains the same because it is the relative complexity of individual
structures which determines the order in which they are acquired regard-
less of individual differences in children and the exposure they receive to
language. Clark and Clark (1977: 295) gave examples of typical utterances
by children aged between 12 months and two years eight months. These
range from one-word utterances (‘More’ at age 15 months), to two-word
utterances (‘More read’ at age 20 months), to what is sometimes called
‘telegraphic speech’ (‘Where go car?’ at age two years one month) and
finally to the production of simple sentences (‘What he can ride in?’, or
even ‘I want to open it’ at age two years eight months). Some children
are much quicker than others at moving through these stages, but,
broadly speaking, one-year-olds start with single words and before they
2 Language Acquisition and Language Learning
are two years old have progressed to two-word utterances. Telegraphic
speech, marked by reduced syntax, is characteristic of two-year-olds and
by the time they are three, children will be producing simple sentences,
some of which may be fully acceptable in the adult language although
others such as ‘What he can ride in?’ will still be grammatically deviant
(Clark and Clark 1977; Crystal 1986; Saxton 2017).
Just being exposed to language, for example by overhearing the con-
versations of others, listening to the radio or watching TV, is not enough
to acquire a language (Saxton 2017: 95–96). Children can pick up some
vocabulary from overheard language or TV but they cannot acquire
syntax in this way without interpersonal interaction (Saxton 2017: 96).
Language acquisition is intimately linked to the child’s own social inter-
active needs from the very start (Bloom and Tinker 2001). Even before
babies utter their first words at about 12 months of age, people talk to
them as if they understood what is being said, and babies do understand
some words before they can say their first words. Between the ages of six
months and one year children develop the ability to engage in ‘joint
attention’, that is, ‘the simultaneous engagement of two or more indivi-
duals in mental focus on a single external object of attention’ (Pence
Turnbull and Justice 2017: 137). An example would be a parent holding
up a toy for a child to look at, or looking at pictures with a child. Joint
attention seems to be important for early word learning (Adamson and
Chance 1998: 28).
Once they start to speak, children are trying to produce meaningful
utterances rather than just reeling off isolated words (Clark and Clark
1977: 314–316; Elliot 1984: 57; Pence Turnbull and Justice 2017:
180–181). Thus, ‘more’ may mean ‘I want some more’ and ‘no’ may
mean ‘I don’t want to go to bed, thank you very much.’ Toddlers try
to communicate even though they cannot make all the sounds of the
language, do not know the right word and cannot produce full
sentences (Clark and Clark 1977: 295, 397–401, 492–496; Saxton
2017: 151–156; 240–244). Child language represents a systematically
simplified linguistic system at the phonological, lexical and syntactic
levels. At the phonological level, for example, because the voiced velar
stop /ɡ/ is more difficult to make than the voiced alveolar stop /d/, /d/
may be substituted for /ɡ/ by two-year-olds, so that ‘go’ becomes ‘doe’
and ‘garden’ becomes ‘darden’ (see Pence Turnbull and Justice 2017:
166–167 on phonological simplification). At the lexical level, ‘garden’
may be used for gardens, parks and all open countryside as well
(lexical overextension, see Pence Turnbull and Justice 2017: 179). At
the syntactic level, ‘grammatical words’ such as articles and preposi-
tions will be omitted so that ‘telegraphic speech’ will be produced (see
Pence Turnbull and Justice 2017: 173). A combination of phonological,
lexical and syntactic reduction might result in ‘doe darden’ for ‘I want
to go to the park.’
Language Acquisition and Language Learning 3
Since it is often not clear to adults what the child wants to say,
children are constantly being asked to restate. Adults go to great
efforts to try to understand what the child wants to express and
frequently ask the child questions to find out what it means. Such
questions are termed ‘clarification questions’ (Demetras et al. 1986).
They are also found in conversations with foreign language learners
(Lightbown and Spada 2013 140; Sheen and Ellis 2011). Brown and
Hanlon (1970) found that adults are more interested in understanding
what the child wants to say than in correcting its errors and tend to
correct only factual errors explicitly. However, factual corrections
from the adult’s perspective may in some cases be language corrections
from the child’s point of view, as in, ‘No, that’s a purple sweater, not
a blue one’ (example from Saxton 2017: 104). Saxton (2017: 110–111)
also suggests that clarification questions may sometimes be posed
when a child makes a grammatical error.
Adults modify their speech style for children so that ‘child-directed
speech’ forms a special variety or register of language (Saxton 2017: 88,
112–116; Pence Turnbull and Justice 2017: 48–49). One feature of this
register is repetition of the child’s preceding utterance before the adult
moves on with new information. This constitutes approval of what the
child has said both factually and linguistically and helps to keep the
conversation on track. If the child’s last utterance was deviant, the adult
tends to repeat it in part but modify it or, in the case of telegraphic
utterances, expand it so that it is an acceptable adult language utterance.
These repetitions with variation are termed ‘recasts’. In this way, adults
keep the conversation going while expanding the syntax of the child’s
telegraphic utterances, improving morphology and making lexical
improvements as they go along rather than always interrupting and
explicitly correcting the child. ‘Recasts’, like clarification questions,
function as implicit corrective feedback to the child (Demetras et al.
1986; Saxton 2017: 102–107).
Whether all this is sufficient to explain how children acquire the
language is unclear. After all, some parents just say, ‘Shut up and eat
your chips.’ Yet these children also acquire English. The nativist school
of thought, associated with the American linguist Noam Chomsky,
maintains that the progress in language acquisition which children make
in the first few years of their lives is so rapid that they cannot be starting
from scratch, cannot be a tabula rasa, but must have some sort of
biological predisposition to acquire grammar; a sort of genetically estab-
lished blueprint of the common or universal grammatical rules or
principles which all languages obey. This is what is meant by universal
grammar (Chomsky 1965: 6–9, 30–37 and see Saxton 2017, Chapters 8–
10; Aitchison 2008, Chapters 3–7 for critical discussion). The blueprint
would have to be for universal grammar rather than the grammar of any
specific language because children will acquire whatever language (or
languages) they are exposed to in infancy, as countless examples of
internationally adopted children show.
The fascinating thing about first language acquisition is that it proceeds
incidentally and in step with cognitive development (Saville-Troike and
Barto 2017: 19). This is why the term ‘acquisition’ is used rather than
‘learning’. Children ‘pick up’ their first language in the first few years of
life provided people talk to them and they talk back. They do not have
to learn their mother tongue deliberately, nor do they get any instruction
in it. They are not taught it by their mother, father, siblings and
playmates in the way that a teacher teaches a foreign language (Lenneberg
1967: 125; Aitchison 2008: 71; Saxton 2017: 214–215). Even the most
devoted parent does not say, ‘I am going to sit down with three-year-old
Mary for two hours this afternoon to teach her some new vocabulary,
revise the present simple and continuous tenses and practise the “th”
sounds with her.’ Yet by the age of five or six children can carry on
conversations with other children and even with adults, provided the
adults make some concessions to them (Pence Turnbull and Justice 2017:
61, 211–213).
Whereas children acquire or pick up the skills of listening and speak-
ing incidentally, they do have to be taught how to read and write. This
usually starts in primary school, and the primary school years are
marked by further massive vocabulary expansion, perhaps on average
six new words per day between the ages of six and eight years(!) (Bloom
and Markson 1998: 68; Saxton 2017: 160–161). However, in the early
school years children also develop a better understanding of words they
already know, for example ‘ask’ versus ‘tell’, and they start to use
various clause-linking words such as ‘anyway’, ‘otherwise’ and ‘actually’
(Crystal 1986: 180–182). In writing, they progress from at first just
stringing sentences together by ‘and’ to gradually producing complex
sentences with a main clause and a subordinate clause linked by subordi-
nating conjunctions (Hunt 1965, 1970). A further development is that they
begin to command a variety of different speech styles and adapt their
speech to situation, learning that you talk to teachers differently from the
way you talk to other children and that you write differently from the way
you talk (Crystal 1986: 188–189; Lightbown and Spada 2013: 14). Meta-
linguistic awareness (awareness of how language works) also starts to
develop. Dawning metalinguistic awareness is manifested in the ability to
play with language (Crystal 1986: 185–188, 1998). This is responsible for
those corny language jokes, which children find so hilarious (Lightbown
and Spada 2013: 13). When my nephew was about seven, he said, ‘I’d tell
you a joke about a broken pencil, only it has no point.’
The neurologist Eric Lenneberg (1967: 142–179) suggested there is
a ‘critical period’ for acquisition of the native language (or languages)
which closes off at puberty at the latest. Lenneberg noted that the
mentally retarded can make slow but steady progress in language
acquisition until puberty, but then make no further progress. They seem
to run out of time. He cited two further sources of evidence to support
the critical period hypothesis. The first source is documented cases of
various unfortunate children who were for whatever reason deprived of
people to talk to them from infancy, with the result that they did not
learn to speak. Such children find it difficult to acquire language in later
childhood, even with specialist instruction (see Saxton 2017: 62–73;
Lightbown and Spada 2013: 22–23; Saville-Troike and Barto 2017:
88–89). The second sort of evidence comes from clinical data on patients
who have suffered language loss because of brain injury. Young children
who suffer brain injury involving language loss can re-acquire language
relatively quickly, presumably by using other areas of the brain. How-
ever, due to progressive loss of brain plasticity, this process becomes
increasingly more difficult, slower and less complete the older the child is
when injury occurs, and after puberty complete recovery may apparently
often be virtually impossible.
Bilingual Settings
It is quite possible under appropriate conditions for children to grow up
simultaneously acquiring two or even three languages from birth as if
each were the native language (Romaine 1995; Baker and Wright 2017:
88–98; Brown 2014: 66; Paradis 2009: 123). This is simultaneous bilingu-
alism. The two languages develop synchronously from infancy in keeping
with cognitive development. For this to work, the children need to have
adequate exposure to each language and appropriate people to speak to
in each language. As in monolingual settings, personal interaction is
necessary. Patterson (2002) found that the vocabulary of bilingual tod-
dlers aged 21–27 months was significantly related to being read to in each
language but not to watching TV. It is also important that the two
languages are kept apart for the child and not mixed. This is ensured,
for example, by one parent speaking only one language to the child and
the other parent speaking only the other language (‘the one parent, one
language principle’), although there are other successful paths to bilingu-
alism (see Romaine 1995: 183–185; Baker and Wright 2017: 91–93). In
the early years the languages may be acquired rather more slowly than if
only one language were involved (Brown 2014: 67). For further discus-
sion of simultaneous bilingualism, see Serratrice (2013) and Montrul
(2008: 94–97).
If acquisition of a second language begins after the age of three years,
this is no longer referred to as simultaneous bilingualism but as ‘con-
secutive (or sequential) bilingualism’ (Lightbown and Spada 2013: 30;
Baker and Wright 2017: 3, 88, 109, 432). The older the child is when
the second language is first introduced, the more likely it is that the child
will retain a foreign accent in the second language. Oyama (1976), for
example, conducted a study of 60 Italian immigrants to the USA. Native
speakers were asked to rate the subjects’ accents. It was found that
number of years spent in the USA had no effect on accent, but the
younger subjects were on arrival, the less marked their accent was.
Paradis (2009:
According to (Paradis 2009: 110)
110), phonology (accent) may actually be
affected from the age of five or six years onwards.
Presence or absence of a foreign accent may not be a good guide to
completeness of language acquisition as a whole. There are indeed some
famous examples of people who were introduced to English very late,
retained a foreign accent but became consummate masters of the lan-
guage. The English novelist Joseph Conrad (originally Józef Teodor
Konrad Korzeniowski), for instance, a native speaker of Polish who
acquired his English at sea after he joined the French(!) navy at the age
of 16 and then the British merchant navy at the age of 19, retained
a distinct Polish accent in his English until the end of his life. The
German-born former American Secretary of State, Henry Kissinger,
who emigrated to the USA when he was 15, also retains a recognisable
German accent. It seems in fact that, for second language acquisition,
there are a series of sensitive periods for various aspects of language or,
to put it another way, different aspects of language are differentially age-
sensitive (Seliger 1978). Phonology (accent) is more age-sensitive than
grammar, and grammar more age-sensitive than vocabulary, which can
continue to be acquired throughout life (Taylor 1978; Paradis 2009).
This is not to say that a nativelike accent cannot be acquired by hard
work in later childhood and beyond, at least by some people. Bongaerts
et al. (1995) found that a panel of four native speakers was unable to
identify the pronunciation of a small group of specially chosen Dutch
university teachers of English with excellent pronunciation as sounding
non-native compared to a control group of native speakers. Yet the
Dutch subjects had all started learning English at age 12 at school and
none of them had visited an English-speaking country before the age
of 15.
In a pioneering study by Patkowski (1980) native speakers were asked to
judge transcriptions of conversations held with 67 highly educated immi-
grants to the USA, all of whom had been resident for at least five years.
Thus, the phonological element was removed. It was found that those who
were six years or younger on arrival in the USA were generally judged to
be indistinguishable or barely distinguishable from Americans who had
grown up in the USA speaking English from the start. For those subjects
who immigrated between the ages of six and sixteen, the older the child was
when it immigrated, the more likely it was that its English was not native-
like even after a number of years living in the USA. Those who immigrated
after the age of about 16 were generally recognisable as not being native
speakers of English, even after many years of residence in the USA. Other
studies of US immigrants of various language backgrounds broadly
support these findings (Johnson and Newport 1989; DeKeyser 2000; Chis-
wick and Miller 2008). DeKeyser (2012: 447–454) reviews various studies
of the relationship between immigrants’ age of arrival in their new country
and their later proficiency in the second language.
However, whereas immigration at an early age virtually guarantees
nativelike acquisition, late immigration does not exclude this as
a possibility, particularly if accent is left out of account. The Patkowski
(1980) study found a lot of individual variation in attainment, with
a small minority of the late immigrants indeed achieving nativelike or
near-nativelike mastery. Just what this means is debatable. One would
expect a truly biological barrier to operate across the board with no
exceptions, so it seems likely that the barriers to complete acquisition in
later childhood and adulthood are social and affective as well as biologi-
cal (Herschensohn 2007: 173–182; Montrul 2008). It is also possible that
at least some age-related biological handicaps can be compensated for by
using general cognitive abilities, at least by some people (Selinker 1972;
Krashen 1982a; Paradis 2009).
Singleton and Muñoz (2011) argue that a wide range of cognitive,
social and cultural variables interact, including to what extent immigrants
continue to be exposed to the first language or not. It may be that it is
actually the firm establishment of the native language which is the main
obstacle to nativelike acquisition of a second language (Paradis 2009:
133–136; Montrul 2008: 22, 262–268). If exposure to the first language
ceases when exposure to the second language begins, as in the case of
international adoptees, for example, then attrition (loss) of the first
language may be dramatic and the second language takes over as the
new first language. This is referred to as ‘sequential monolingualism’ (see
Higby and Obler 2015: 647).
Classroom Settings
Classroom settings are necessarily impoverished acquisition environ-
ments. To compensate for the limited quantitative and qualitative expo-
sure to the language available in the classroom, and to take advantage of
the developing cognitive abilities of the older child, language teaching
often involves three components in various weightings:
explaining how the language works (instruction)

providing the learner with opportunities for using the language
(practice)
providing feedback on learner performance (especially error correction)
The higher level of ultimate achievement which younger starters are

likely to attain relative to older starters in naturalistic environments
does not necessarily apply to classroom learners. Younger children are
better at implicit acquisition than older children, but acquisition requires
massive amounts of exposure not available in the classroom (Paradis
2009). Older children, on the other hand, are better (faster) than younger
children at explicit learning of languages (and other school subjects)
because of their greater cognitive maturity, but their ability to implicitly
acquire is declining in the school years. In the long term the age at which
the child started to learn the foreign language does not seem to be
particularly significant for attainment in the educational system (see
Singleton and Muñoz 2011; Pfenninger and Singleton 2017).
Vocabulary is the aspect of language which seems to be least age-
sensitive. Nature seems to have recognised that our language equip-
ment must include a facility for updating lexis throughout our lives
(Libben and Goral 2015: 641; Jarema and Libben 2007: 3). This means
that foreign language learners can over many years continue to acquire
vocabulary and in some cases even catch up with or surpass mono-
lingual native speakers, particularly in certain areas of specialist
vocabulary.
Language-Learning Error
Contrastive versus Non-Contrastive Error

Errors which learners produce when speaking or writing in a foreign
language have received much attention from language teaching theor-
ists. Lado (1957: vii) argued that by analysing the points of difference
between the native language and culture and the language and culture
to be learned (contrastive analysis) one would be able to predict the
difficulties learners would encounter. It seems reasonable to assume
that error would be likely to occur at these points of difference.
However, whereas contrastive analysis tends to predict phonological
(pronunciation) error well, for grammar, vocabulary and pragmatics
(utterance meaning) it is a less reliable predictor (Brown 2014: 257).
Beyond the phonological level, it seems that the influence of the first
language is by no means the only factor influencing error. There are
certain errors in English which speakers of many different first lan-
guages tend to make, for example in the tense system or in preposition
choice. Such errors are termed non-contrastive or developmental
errors in contradistinction to contrastive errors, which are attributable
to negative interference from the mother tongue (Richards 1974).
Some errors, however, may be influenced by both developmental and
contrastive factors (Selinker et al. 1975). There are other influences on
error too, including the unfortunate effects of imperfect teaching
(Selinker 1972; Brown 2014: 261). For a review of studies on contras-
tive and non-contrastive errors, see Falla Wood (2017).
Overt versus Covert Error
A quite different distinction is that between ‘overt’ error and ‘covert’
error (Corder 1971). Overt error involves formal incorrectness as in,
*‘She live in England.’ (A preceding asterisk (*) marks a formally
unacceptable word, phrase or sentence.) Covert error results when
learners say something which is formally acceptable in English but
does not mean what they meant to say, as in, ‘If this is your first visit
to our country, you are welcome to it’ (notice reportedly seen in
a hotel). Identifying covert error can be very difficult because you can
never be sure about what someone really meant to say. However, even
overt error is a very elusive phenomenon. This is partly because there
are two broad approaches to the identification of error. One is based
on codified rules as found in grammar books and dictionaries. The
other is based on native-speaker judgements. This is sometimes called
the difference between correctness criteria and acceptability criteria.
There are problems with both approaches and there seems to be
a borderline area of infelicitous utterances. These infelicities may be
of two sorts: either not correct but acceptable (‘not right, but you can
say it’) or, conversely, correct but not acceptable (‘right but you don’t
say it’) (Azevedo 1980).
Global versus Local Error

Burt and Kiparsky (1974) drew a helpful distinction for describing and
understanding grammatical error. They identified ‘local’ and ‘global’
grammatical errors. Local grammatical errors include errors of verb
and noun morphology, article use, preposition choice, concord (agree-
ment), tense choice and many more. Such errors are relatively easy to
pinpoint and easy to correct. They often violate a specific grammar
rule and occur within a single clause. By contrast, global errors are
difficult to pinpoint and more difficult to correct because they occur at
the clause or sentence level. It is global errors which tend to seriously
impede communication, involving, as they do, sentence connectors,
tense continuity across clauses, word order, sentence structure, co-
ordination and subordination. When global errors occur in writing, we
teachers just tend to put those wavy lines under the whole sentence
instead of correcting them, so that the learner is not helped at all.
Sometimes we may write something cryptic in angry red ink, such as
‘Expression!’ or ‘Syntax!’ or ‘Not clear!’ or ‘Meaning?’ There is evi-
dence that native speakers who are not teachers tend to regard global
errors that disturb communication as the serious errors, whereas non-
native-speaker teachers tend to focus on local formal accuracy criteria,
with native-speaker teachers occupying a middle ground (Hughes and
Lascararatou 1982; Davies 1983).
Language-Learner Language
In 1967, the Edinburgh applied linguist Pit Corder published perhaps the
seminal article for our modern understanding of foreign language learn-
ing. It was tellingly entitled ‘The significance of learners’ errors’ and
suggested that there is system in a learner’s errors. Corder (1967) viewed
the foreign language learner as being engaged on a voyage of linguistic
discovery involving testing out hypotheses about the way the language
works. He suggested learners may be following a sort of ‘inbuilt sylla-
bus’, the ordering of which may be quite distinct from that of the
external syllabus they are ostensibly following. From this perspective,
learners have to go through the process of making errors as part of their
hypothesis testing. They have to proceed by trial and error. If learners
do not make errors, they will not get any better. Learners learn from
their errors. Corder (1967) suggested that learners’ errors will define
what he termed their ‘transitional competence’, a term which alludes to
Chomsky’s famous distinction between ‘competence’ and ‘performance’
(Chomsky 1965: 4). ‘Competence’ refers to the native speaker’s inter-
nalised language system in contradistinction to ‘performance’, which is
the language a speaker actually produces with all its surface
imperfections.
In summary, according to Corder:
Errors are a necessary part of linguistic development and represent

the discrepancy between the learner’s transitional competence and
the target language (language to be learned).
The learner’s inbuilt syllabus may determine the order in which the
grammar is acquired, and studying learners’ errors may supply clues
to this order.
A distinction can be drawn between ‘errors’ (of competence) and
‘mistakes’ (that is, performance slips or lapses). The latter are unsys-
tematic and therefore do not provide data on transitional
competence.
An analysis of learners errors can function as a window on the
learner’s transitional competence and help the teacher to plan future
teaching (error analysis).
However, error does not tell the whole story about the learner’s pro-
blems with the language since language deficit does not necessarily show
up as error. The most obvious example is that learners may avoid
structures which they are unsure of. Schachter (1974) found that learners
who did not have relative clauses in their native language tended to avoid
them in written work and use alternative structures in English, whereas
learners who did have relative clauses in their mother tongue tended to
use them more and hence made more mistakes.
The idea of transitional competence was taken up and developed by
Selinker (1972), who coined the term ‘interlanguage’ for what is nowadays
termed ‘learner language’. He suggested that learners develop a simplified
linguistic system in order to communicate. This system is permeable (open
to outside influences), dynamic, (developing) and systematic (regular cor-
respondences between forms and meanings) (Selinker 1972; Selinker et al.
1975; Tarone et al. 1976). Schumann (1978) compared learner language to
pidgin forms of languages, which are also simplified systems for basic
communicative purposes. Most learners seem, however, to stop develop-
ing in the direction of the target language at some point, at least in certain
aspects of their language – pronunciation being the most obvious one.
Selinker (1972) coined the term ‘fossilisation’ for such a state. Fossilised
interlanguage has lost its permeability and dynamism but retains its
systematicity (see Brown 2014: 264–266; Han and Odlin 2006; Han 2012).
No two learners will have identical learner languages, for each learner forges
his or her own language development along a path unique to that learner. This
route is influenced by external factors (the learner’s individual language-
learning experience) as well as internal factors (age, motivation and person-
ality). Even within a single learner the learner language will vary according to
linguistic and non-linguistic context, although context will not explain all the
variation found, so that some variation would appear to be non-systematic.
This non-systematic variation will include performance slips or mistakes (with
reference to the learner’s usual language use), but also free variability among
some forms (Ellis 1985: 76). Thus, although learner language is ‘systematic’
and errors are not random, the system may be difficult to detect.
The idea of ‘variable rules’ was introduced by Elaine Tarone to
account for contextual variation in language-learner speech. She argued
that production in the foreign language will vary systematically according
to task and proposed her ‘continuum paradigm’, in which a language
learner’s speech styles may be ordered along a continuum of how much
attention is paid to speech, with what she called the ‘vernacular style’
(colloquial speech) showing the most regular patterns, while other styles
show more variation (Tarone 1988). This implies that if researchers,
teachers or testers want to get a true picture of the learner language,
then they should observe natural conversation. However, if the learners
know they are being observed, then the naturalness may be affected since
the presence of the observer will result in attention being paid to speech
and very likely the learner will focus on trying to be correct. This is the
‘observer’s paradox’ (Labov 1972/1978: 209–210). Writing produced in
highly formal situations, such as examinations, is also likely to be a very
poor reflection of the learner’s underlying competence, since the learner
has time to reflect, correct and edit. Homework is even less reliable as
a guide to the true state of the learner language since additional external
sources such as dictionaries can be used and these will influence the
language produced.
The Monitor Model of Classroom Language Learning
The idea of variation in language-learner language was taken up by
Krashen (1981, 1982b), who distinguished between ‘monitored’ and
‘unmonitored’ production. This distinction formed the basis of his
Monitor Model. Krashen (1981, 1982b) argues that language acquisition
is an unconscious process which will only occur if learners are exposed
to language input which is neither too easy nor too difficult. This ideal
input should be at a level which, although just beyond what learners
could produce themselves, is nevertheless understandable for them in
context. Such input is termed ‘comprehensible input’ at the ‘i+1 level’.
Anything which is consciously learned about the language, especially
language rules, does not contribute to the unconscious process of
language acquisition. The Monitor Model maintains, in fact, that class-
room learners are building up two separate knowledge systems: an
unconscious or acquired system and a conscious or learned system. It
hinges on the idea that there are no links between the two systems,
conscious learning or knowledge about the language being available only
as a monitor which can edit or make changes to what the ‘acquired
system’ produces. This can be done either just before production,
involving hesitation while the change in plan is made, or after produc-
tion, involving self-correction (Krashen 1982b: 15).
For learners to be able to monitor their speech effectively, two condi-
tions have to be satisfied: they have to be free from time pressure and
their focus of attention has to be on formal correctness rather than on
communication. The theory also holds that learners differ individually in
terms of how much they use the monitor, there being ‘over-users’ and
‘under-users’ (Krashen 1978). The ‘optimal monitor-user’ would monitor
appropriately according to the specific communicative situation and
whether priority is to be given to accuracy or to fluency. Furthermore,
older children and adults will be more likely to use the monitor than
young children so as to compensate for their diminishing ability to
acquire naturalistically with increasing age and to take advantage of
their greater cognitive abilities.
However, Krashen (1982a) argues that the main barriers to language
acquisition after puberty may be not so much biological as affective. The
minority of adult learners who do successfully acquire a second language
very effectively (Selinker 1972) may owe their success to having over-
come the inhibitions which tend to block naturalistic acquisition after
puberty. In this way they may somehow reactivate the original acquisi-
tion path. Krashen refers to this as ‘lowering the affective filter’. The
affective filter (or barrier) is assumed to be lowered in low-anxiety
situations and raised in face-threatening, high-anxiety situations. Indivi-
duals themselves differ in their anxiety thresholds.
The teaching implications of monitor theory for classroom language
teaching are that learners should be provided with appropriate exposure
to the language (comprehensible input at the i+1 level) in a low-anxiety
situation rather than being supplied with knowledge about the language.
Krashen and Terrell (1983/1998) suggested various classroom activities to
promote such exposure. For adults and adolescents rather than children,
however, monitor theory holds that such an approach might profitably
be supplemented with formal teaching since older learners can more
effectively use the monitor.
The Monitor Model has remained controversial, especially its conten-
tion that ‘learning’ cannot become ‘acquisition’. It has also been objected
that the distinction between ‘conscious’ and ‘unconscious’ as the defining
distinction between ‘learning’ and ‘acquisition’ is not helpful since con-
sciousness is in any case a continuum not a dichotomy (McLaughlin
1990a; Searle 1992). The distinction between ‘learning’ and ‘acquisition’
seems to correspond to the distinction between ‘declarative knowledge’
and ‘procedural knowledge’ in cognitive science, although Krashen him-
self does not use the terms. Declarative knowledge is factual knowledge
(‘knowing that … ’) and procedural knowledge is skill (‘knowing how
to … ’). The Monitor Model appears to adopt an extreme non-interface
position on the relationship between the two sorts of knowledge. Scho-
lars who hold an interface position would argue that explanations, for
example of grammar, vocabulary or how to make the ‘th’ sounds, are
generally helpful for foreign language learners and that what is learned
consciously in the classroom can by dint of practice become part of the
acquired language system and be available for spontaneous use. For
critical appraisal of the Monitor Model and discussion of the interface
versus the non-interface positions, see Ellis (1985: 215–247), McLaughlin
et al. (1983), McLaughlin (1987: 133–153), Brown (2014: 288–293),
Saville-Troike and Barto (2017: 47–48, 78–81), Lightbown and Spada
(2013: 106–107) and Paradis (2009).
Automatisation
A distinction is drawn in cognitive science between two modes of
cognitive processing: ‘controlled’ and ‘automatic’. Controlled processing
is involved in the early stages of mastering a skill but with practice
processing becomes progressively automatised. Complex skills such as
carrying on a conversation, driving a car or playing tennis involve
a hierarchical series of sub-tasks. Human beings are limited-capacity
information processors, so that in using language, as in driving a car or
playing tennis, they are able to focus attention on only a limited number
of things at a time, and this focussed attention may be at a greater or
lesser level of consciousness. The process of language acquisition may
thus be regarded as proceeding stepwise or cyclically, with focus
constantly switching to new aspects of the language over weeks or
months, with some items being more in focus than others at any
particular point in time. In the early stages of learning English, even
pronouncing the sounds may be a painstaking process requiring much
effort and concentration. Beginning learners who have problems with the
‘th’ sounds, for example, may need to concentrate on putting their
tongue in the correct position on their upper teeth and expelling just
enough air to make an acceptable sound.
At lower-intermediate levels of proficiency, although pronunciation may
now have become partly automatised, learners may have to focus attention
on making grammatical choices while speaking, and lexical choice may also
involve much slow searching for words. These are controlled processes. As
learners become more proficient, however, phonology, grammar and much
lexical choice may become largely automatic and production can proceed in
larger phrasal units rather than word by word (automatic processing). This
is because, with increased practice, individual sub-components of the task
are bundled together into modules and performance is speeded up because
each sub-component does not have to be focussed on separately any more.
The automatisation of lower-level processes means that learners have more
attentional capacity available while speaking to concentrate on higher-order
aspects of production, such as discourse planning, attention to using an
appropriate style, use of idiomatic language and more varied and complex
syntax. They can also focus attention on their interlocutor. Only if enough
attentional capacity is available will they be able to notice the discrepancies
between aspects of their learner-English and the English of their interlocu-
tors and then modify their own language accordingly.
At the neurological level, it is assumed that learning to perform a sub-task
such as pronouncing the ‘th’ sounds, changing gear or shifting grip on the
tennis racquet for a backhand stroke involves activating a particular set of
neural impulses in the brain to form a circuit across the synaptic nodes which
join nerve cells. The more often the same neural pattern is activated (prac-
tice), the more established the pathways become, rather like the difference
between treading a path through undergrowth for the first time and treading
a well-trodden path. As performers become better with practice at a complex
skill such as speaking a foreign language, driving a car or playing tennis, many
lower-level skills, such as pronunciation, changing gear or shifting one’s grip
on the racquet for a backhand stroke, become progressively automatised.
Schneider and Shiffrin (1977: 51) defined controlled processing as:
… a temporary activation of nodes in a sequence that is not yet

learned. It is relatively easy to set up, modify and utilise in new
situations. It requires attention, uses up short-term capacity, and is
often serial in nature.
(Schneider and Shiffrin 1977: 51)
Automatic processing, by contrast:
… is triggered by appropriate inputs and then operates indepen-

dently of the subject’s control. Automatic sequences do not require
attention, though they may attract it if training is appropriate, and
they do not use up short-term capacity.
(Ibid.)
Speech production necessarily involves an optimal blend of controlled

and automatic processes, correctly adapted to situation. Whereas exces-
sive reliance on controlled processes makes for hesitant, non-fluent,
effortful production, too much automatisation, sometimes called ‘false
fluency’, is not ideal either. It results in over-reliance on clichés, in lack
of originality and flexibility and what is produced may be perceived by
the listener as shallow and superficial patter.
Restructuring
Language acquisition involves not just automatisation but also restruc-
turing of the learner language to bring it closer to the target language
(McLaughlin 1990b; Gass and Selinker 2008: 230–238). Automatisation
involves improving fluency and restructuring involves improving lan-
guage correctness. There is a trade-off relationship between the two, for
the more highly automatised deviant language sub-systems become – for
example non-nativelike pronunciation of ‘th’, deviant past-tense forms or
overextension of word meanings – the more difficult it is to restructure
them. The initial stages of restructuring involve controlled processes,
which are effortful and slow down production. As well as sounding less
fluent, learners may also appear to be backsliding (Selinker 1972). That is
to say, new errors may be introduced as the system is being reorganised
and before restructuring is complete. This pattern is termed U-shaped
behaviour (McLaughlin 1990b; Altarriba and Basnight-Brown 2013: 126).
Hakuta (1976) showed in a now famous case-study that in restructuring
their internal grammars learners are driven by two conflicting tendencies.
One is to make their grammar internally consistent and the other is to
make it externally consistent, that is, to adapt it to the language they hear
around them. The subject of the study was a five-year-old Japanese girl
called Uguisu who was acquiring English naturalistically in the United
States. The study started five months after she had arrived in the USA
and lasted for 15 months. One of the structures the study focussed on
was embedded ‘how to’ clauses. By the third month of the study Uguisu
was producing embedded ‘how to’ clauses with near 100% accuracy.
Then they declined to zero accuracy. However, from the 11th to the 13th
month of the study 50% accuracy was achieved: a remarkable example of
U-shaped behaviour.
It was possible to identify a number of stages in Uguisu’s acquisition
of the structure. In the first stage, ‘I know how to’ + infinitive was
learned as a ‘chunk’ or memorised pattern for a small number of specific
verbs, for example ‘I know how to ride a bike’, ‘I know how to swim.’ In
the second stage, the pattern was extended to other verbs, such as ‘show’
and ‘tell’ (‘I show you how to …’, ‘I tell you how to …’). In the third
stage, the infinitive element in the pattern was erroneously replaced by
an interrogative structure introduced by an interrogative adverb and with
inversion of subject and verb as in, *‘I know how do you write this.’
Hakuta suggests this backsliding at stage three occurred because
Uguisu was (presumably unconsciously) motivated by the need to
achieve internal consistency with her other interrogative embedded
sentences, which at this stage consisted of sentence + interrogative
structure as in, ‘I know where do you live.’ He predicted that in time
she would have restructured her deviant ‘how’ embedded clauses appro-
priately, but unfortunately the study had to be concluded before this had
happened. However, the other interrogative embeddings were progres-
sing to the norms of English at the end of the study (that is, with no
inversion in the embedding).
Connectionism
A related but distinct theoretical approach to language acquisition within
cognitive theory is that of connectionism. Connectionism provides
a theory of how words, phrases and whole utterances may be linked in
the mind. It builds on the idea of neural pathways becoming progres-
sively more firmly established by repeated use, but adds a sociolinguistic
component to the psycholinguistic one by additionally stressing the
importance of the specific contexts in which ‘chunks’ of language are
experienced and used. In other words, words, phrases or even utterances
are stored with a tag on their remembered contexts. This includes both
the linguistic context and the extralinguistic or social context: the
remembered social context may include who the speaker was, who the
addressee was and in what setting the chunk was heard. This is why it is
easier to remember language experienced in a memorable context. The
same applies quid pro quo to language met in reading.
A connectionist approach to language acquisition thus sees multiple
or branching neural interconnections becoming established in the
mind. These link remembered ‘chunks’ of language not only intralin-
guistically (sound, meaning and grammatical links), but also extralin-
guistically (similar contexts of use). Intralinguistic links involve
linguistic memory, while extralinguistic links tap into episodic
memory, or memory of our own personal experience. Remembered
context then forges links between words, phrases and utterances in
memory so that one linguistic element activates others from shared
contexts in a chain-like fashion (‘spreading activation’). Connectionism
elegantly accounts for the fact that such links in the mind will be, on
the one hand, idiosyncratic to each individual’s linguistic and cultural
experience but also shared by members of the speech community. It
also accounts elegantly for the fact that language experienced in real-
life contexts is more easily retained than language learned out of
context. The clear implication is that languages are best acquired by
using them.
Language-Learning Strategies
The Good Language Learner

Language-learning processes operate largely outside the learner’s con-
scious control. Language-learning strategies, by contrast, are ‘activities
consciously chosen by learners for the purpose of regulating their own
language learning’ (Griffiths 2008: 87). Whereas language-acquisition
processes are universal, biologically determined and inevitable, language-
learning strategies are individual, part of cognitively directed problem-
solving behaviour and optional. The learner also has to be suitably
motivated in order to employ them. There seems to be a link between
success in language learning and the employment of appropriate lan-
guage-learning strategies (Rubin 1975; Stern 1975; Naiman et al. 1978;
Griffiths 2008; Chen et al. 2020). Studies of successful language learners
in mixed environments where both tuition and exposure to speakers of
English are available indicate that these individuals tend to employ the
following strategies:
They focus on language both as a formal system and a means of

communication.
Both inside and outside the classroom they actively involve them-
selves in speaking so that they can learn the language by using it.
They do not expect to understand everything but this does not
prevent them from participating.
They employ inferencing skills to follow the drift of what is being
said.
When they are speaking, they try to paraphrase their way through
lexical gaps they may experience rather than letting communication
fail.
They are sensitive to feedback and try to learn from their mistakes
but are not afraid of taking risks and losing face by making mistakes.
According to the communicative situation, they place more or less
emphasis on fluency versus formal correctness, at times experiment-
ing with their language, at other times playing safe.
They are not only responsive to corrective feedback but are also
sensitive to how their interlocutors react emotionally to what they
say, that is they pay attention to socio-affective feedback.
They are flexible in choosing an appropriate strategy for the appro-
priate task and situation.
(Rubin and Thompson 1994; Griffiths 2008; Oxford 2011: 15)
Strategies of Young Children in Naturalistic Environments

In a now famous study of four Mexican children aged five to eight years
who were acquiring English naturalistically in mainstream school classes
in California, Wong Fillmore (1976) identified cognitive and social
strategies which the children adopted. Cognitive strategies ‘enable the
learner to figure out how the new language is structured, to interpret
meanings in it and to begin expressing themselves when using it’ (Wong
Fillmore 1976: 633). Social strategies ‘have to do with finding ways to
receive input on which to base the language learning and on making
efficient use of the social setting in which language is used as an aid in
that learning’ (ibid.). Five cognitive and three social strategies were
identified, which Wong Fillmore couched in the form of commands as
follows:
Cognitive strategies
Assume that what people are saying is directly relevant to the
situation at hand or what you are experiencing (metastrategy: guess).
Get some expressions you understand and start talking.
Look for recurring parts in the formulae you know.
Make the most of what you have (overextension of word meaning).
Work on big things and save the details for later.
Social strategies
Join a group and act as if you understand what is going on even if
you do not.
Give the impression – with a few well-chosen words – that you can
speak the language (formulae).
Count on your friends for help (interaction, negotiation of meaning).
(Wong Fillmore 1976: 633)
Strategies of Classroom Learners

Pioneering work by O’Malley and Chamot (1990) found that classroom
foreign language learners also adopt a strategic approach. It was possible
to identify metacognitive, cognitive and socio-affective strategies. Meta-
cognitive strategies are global strategic approaches which help to organise
learning. They include advance planning, monitoring and reviewing, and
can be applied to a variety of tasks to impart structure to task perfor-
mance. Cognitive strategies, by contrast, are used for a particular learn-
ing task. Among them are some well-known study techniques such as
note-taking or using a dictionary. Socio-affective strategies involve var-
ious forms of cooperative learning as well as consulting the teacher for
help and indulging in self-talk to reduce task anxiety (O’Malley and
Chamot 1990 119–120, 126). Oxford (2011: 241–262) discusses specific
strategies for listening, reading and writing as well as for learning
vocabulary and grammar.
Self-Regulation, Autonomy, Motivation and Identity

Learning strategies and especially socio-affective strategies have been
linked to the self-regulation of learning, the development of learner
autonomy and the maintenance of long-term self-motivation (Dörnyei
2005: 188–195; Ushioda 2003, 2008; Johnson 2018: 144–145; Oxford
2011: 7–42; Chen et al. 2020). The idea of self-regulation, not just of
language learning but of all purposeful human activity, was developed
within social psychology. According to one definition, ‘Self-regulation
refers to self-generated thoughts, feelings and actions that are planned
and cyclically adapted to the attainment of personal goals’ (Zimmermann
2000: 14). The cycle consists of three phases: forethought, performance
and self-reflection (ibid.: 16).
By learner autonomy is meant ‘the ability to take charge of one’s own
learning’ (Holec 1980: 3). Initially, learners will require teacher support,
which can gradually be withdrawn as they become increasingly self-
directed (Holec 1980: 9, 29). Holec identified five areas in which learners
should gradually take charge:
learning goals
learning content and progression
learning methods and techniques
monitoring of learning progress
evaluation of learning achievement
(Holec 1980: 4)
Good learners differ from bad language learners less in the specific
strategies they use and more in their ability to apply strategies appro-
priately to the task at hand in a flexible and eclectic manner (Reiss 1983;
Norton and Toohey 2001; Griffiths 2008; Johnson 2018: 134–145; Chen
et al. 2020). Strategy use is highly individual and one should be cautious
about regarding some strategies as inherently better than others for all
learners. In particular, learners from different cultures may have different
strategy preferences (Press 1996; Johnson 2005). Good language learners
tend, in fact, to be not only good strategy users but also to be autono-
mous learners (Little 1999: 13).
Teachers can help learners develop autonomy, maintain motivation
and ‘self-regulate’ their learning by giving them choices to make in their
learning, varying classroom activities, setting them classroom tasks with
clearly defined goals and discussing appropriate strategies (Dörnyei 2001:
71–86). Learners can benefit from some instruction in how to employ
specific strategies for specific tasks, for example vocabulary learning, but
they also need to be given space to develop their own personal strategic
techniques (Moir and Nation 2008). To make this possible, it is impor-
tant to create a low-anxiety learning environment conducive to the
development of self-confidence (Dörnyei 2001: 86–116). In this way
learners can find out what works best for them and ultimately develop
a sense of their own learner identity, their ‘L2 self’ (Dörnyei and
Ushioda 2009) and of the imagined English language community to
which they wish to belong (Norton 2013: 8; Noels and Giles 2013;
Norton and Toohey 2011).
Communication Strategies and Negotiation of Meaning

When learners encounter communication difficulties while speaking,
they may employ communication strategies to express what they want
to say in a makeshift way. Gass and Selinker (2008: 285) define commu-
nication strategy as ‘a deliberate attempt to express meaning when faced
with difficulty in the second language’. The following taxonomy is based
on Faerch and Kasper (1983). For other classifications see, for example,
Poulisse (1987) and Brown (2014: 129):
1. Using the native language or another language. This includes:

switching to the native language or another foreign language
(code switching)
inserting a native language (or another foreign language) word
anglicising a native language (or another foreign language) word
to make it sound English
2. Using one’s available English language resources. This includes:

overgeneralising a grammatical pattern (‘I suggested him to go’
on the model of ‘ask’)
overextending a word’s meaning (‘house’ for ‘kennel’)
paraphrasing (‘false hair’ for ‘wig’)
coining a word (‘meat-man’ for ‘butcher’)
giving examples (‘tables, chairs and sofas’ for ‘furniture’)
restructuring syntax (for example, abandoning a passive sentence
and using an active sentence)
3. Cooperative strategies of appeal to the interlocutor. These include:
pausing and looking at the interlocutor in a mute appeal for help
pausing and asking the interlocutor to supply a missing word
(‘What’s the word?’)
pausing and explicitly discussing the problem with the interlo-
cutor (‘negotiation of meaning’)
4. Paralinguistic strategies of gesture, mime and facial expression (for

example, yawning to indicate the word ‘tired’)
5. Retrieval strategies for a word that is on the tip of one’s tongue.
These include:
pausing until the word comes
saying words which sound similar
saying words of related meaning
saying translations of the word from other languages
Some learners are better than others at choosing the best strategy for the
particular communication problem at hand. They possess ‘strategic
competence’, which may be distinct from formal proficiency (Canale
and Swain 1980). Learners who regularly code switch in the classroom
may find it difficult to use English-based strategies when they have to
communicate with English speakers outside the classroom (Haastrup and
Phillipson 1983). By contrast, learners who can employ communication
strategies effectively, especially cooperative strategies, will not only com-
municate better but will also learn more effectively through communica-
tion, particularly at points where communication threatens to break
down and ‘negotiation of meaning’ takes place. At these points a skilled
interlocutor may provide the word, phrase or structure a learner is
struggling to find, or may correct the learner. It is believed that learning
is particularly likely to occur in such situations of joint focus on
a problem (Lightbown and Spada 2013: 114–115; Mackey et al. 2012.).
It is when learners experience problems in trying to communicate that
they become aware of their own language deficits. In this way, even if
they do not receive helpful input from their interlocutor, they will at
least become aware of their problem and can learn from their own
output. This is what is meant by the ‘output hypothesis’ (Swain 1985;
Swain 1995, 2005; Swain and Lapkin 1995).
Socio-Cultural Theory and Interactive Language Learning

The idea of learning by means of cooperative communicative interaction
is at the heart of the socio-cultural approach to language learning (see
Lantolf 2000; Lantolf and Thorne 2006; Brown 2014: 12–15). The
approach is based on the work of the Russian educationalist Lev
Vygotsky (1896–1934), whose work was not available in translation in
the West until long after his death. Vygotsky was not concerned with
language teaching but with general primary school education, which he
saw as proceeding through cooperation between the novice (child) and
the expert (teacher). Vygotsky (1978: 86) coined the term ‘zone of
proximal development’ (ZPD) to refer to the child’s potential for achieve-
ment under conditions of optimum support. He defined the ZPD as:
the distance between the actual developmental level as determined

by independent problem solving and the level of potential develop-
ment as determined through problem solving under adult guidance
or in collaboration with more capable peers.
(Vygotsky 1978: 86)
It was found that children of the same proficiency level might well vary
in the extent of their ZPD. Vygotsky argued that it was this variation in
ZPD rather than variation in starting level of proficiency which deter-
mined potential development (Vygotsky 1986: 187). The progress the
child makes educationally will be determined by the extent of its ZPD, its
ability and readiness to engage in effective interaction and the quantity
and quality of interactional support available to it (see Lantolf and
Thorne 2006: 263–290; Lantolf and Poehner 2008: 14–17, 2013:
142–143).
The term ‘scaffolding’ (Wood et al. 1976: 96) is nowadays used to refer
to such support, although Vygotsky himself did not use the term. With
reference to language learning, one important form of verbal scaffolding
consists of the way proficient speakers (experts) in conversation with
learners (novices) may employ various rhetorical devices such as para-
phrase, explanation, repetition and comprehension checks to support the
learner (see Saville-Troike and Barto 2017: 115–116, 119–120). Scaffold-
ing can also be provided by peers supporting one another in collabora-
tive activities such as peer editing, peer correction and peer feedback on
written work. This constitutes a form of mutual or reciprocal scaffolding.
If learners pool their knowledge in the production of a piece of colla-
borative writing, for example, they are able to achieve more than they
could individually. They may all be at a similar level of proficiency but
they will have different strengths and weaknesses in the language and so
can help one another. The idea of scaffolding may be extended still
further to include, for example, visual aids (Gibbons 2015; Baker and
Wright 2017: 287–290).The common factor is that the emphasis is never
on reducing the goal to be achieved but on supporting the learner’s
efforts to achieve the goal.
Socio-cultural approaches to language learning also place emphasis on
the importance of introspection, self-talk and self-regulation (Lantolf and
Thorne 2006: 72–79; 179–207). This builds on Vygotsky’s conviction
that language helps us to regulate our thought processes and grapple with
the problems of the world (Lantolf and Poehner 2013: 138–142).
Vygotsky noted that children indulge in private or egocentric speech as
a means of self-regulation as they talk their way through tasks and he
regarded this as the precursor to adult inner speech. At a certain stage of
cognitive development, egocentric speech, according to Vygotsky (1986:
87), ‘turns inward’ and becomes inner speech (see also Saville-Troike and
Barto 2017: 121–123).
Food for Thought

Q. 1. Is there anything you have learned in this chapter about (a) first
language acquisition and (b) bilingual acquisition in naturalistic
settings which you think language teachers should know? If so, say
what and why.
Q. 2. What do think is the greatest advantage that young language learners
(before puberty) have over older language learners (after puberty)
and what assets do post-puberty language learners have that young
language learners lack?
Q. 3. What are the advantages and disadvantages of classroom language
learning compared to naturalistic acquisition?
Q. 4. Do you think foreign languages should be taught just like other school
subjects, or should attempts be made to make classroom foreign
language learning as much like naturalistic acquisition as possible?
Q. 5. Imagine you are a teacher of an intermediate class and your learners
still produce *‘he come’, *‘she play’, etc. How would second lan-
guage acquisition theory explain that?
Q. 6. (Group task for practising teachers) Can you think of some develop-
mental errors and some contrastive errors your learners make? Do
they make any errors that seem to combine elements of both types of
influence? Which of the three types do you think are most common
among your learners? Are there differences in the proportions accord-
ing to whether the first language is close to or far removed from
English? Are there differences between child and adult learners?
Q. 7. (Group task) The Monitor Model holds that in order for optimal
acquisition to occur, the language that the learner is exposed to
must be comprehensible and just beyond the learner’s current
productive language level (comprehensible input at the i+1 level).
Do you think anything else is necessary?
Q. 8. (Group task) What language-learning strategies have members of
the group themselves used, perhaps for vocabulary learning, gram-
mar learning, writing, listening, reading or speaking in a foreign
language? Try to group these into metacognitive, cognitive and
socio-affective categories.
Q. 9. Look back at the taxonomy of communication strategies (CS).
Which of the strategies listed do you find most effective and
which least effective, both for achieving communicative goals and
promoting further language acquisition? Can you think of any
classroom activities which could be done to encourage learners
to use CS and how would you get them to use the effective ones
rather than the ineffective ones?
Q. 10. (Group task) What are the similarities and differences between the
i+1 level of input in the input hypothesis of Steven Krashen and
the zone of proximal development (ZPD) of Lev Vygotsky?
Further Reading
Enever, J. and Lindgren, E. (eds.) (2017). Early Language Learning: Complexity and
Mixed Methods. Bristol: Multilingual Matters.
Oxford, R. L. and Amerstorfer, C. M. (eds.) (2018). Language Learning Strategies
and Individual Learner Characteristics: Situating Strategy Use in Diverse Contexts.
London: Bloomsbury.
References
Adamson, L. B. and Chance, S. E. (1998). Coordinating attention to people,
objects and language. In Wetherby, A. M., Warren, S. F. and Reichle, J. (eds.),
Transitions in Prelinguistic Communication. Baltimore, MD: Paul H. Brookes
Publishing Co., pp. 15–37.
Aitchison, J. (2008). The Articulate Mammal: An Introduction to Psycholinguistics.
Fifth edition. London and New York: Routledge.
Altarriba, J. and Basnight-Brown, D. M. (2013). An information-processing
approach to second language acquisition. In Ritchie, W. C. and Bhatia, T. K.
(eds.), The New Handbook of Second Language Acquisition. Second edition.
Leiden: Brill, pp. 115–136.
Azevedo, M. (1980). The interlanguage of advanced learners. International Review of
Applied Linguistics, 18/3: 217–227.
Baker, C. and Wright, W. E. (2017). Foundations of Bilingual Education and
Bilingualism. Sixth edition. Bristol: Multilingual Matters.
Bloom, L. and Tinker, E. (2001). The intentionality model and language acquisi-
tion: engagement, effort and the essential tension in development. Monographs of
the Society for Research in Child Development 66/4: 1–91.
Bloom, P. and Markson, L. (1998). Capacities underlying word learning. Trends in
Cognitive Sciences 2: 67–73.
Bongaerts, T., Planken, B. and Schils, E. (1995). Can late starters attain a native
accent in a foreign language? A test of the critical period hypothesis. In
Singleton, D. and Lengyel, Z. (eds.), The Age Factor in Second Language Acquisi-
tion: A Critical Look at the Critical Period Hypothesis. Clevedon: Multilingual
Matters, pp. 30–50.
Brown, H. D. (2014). Principles of Language Learning and Teaching. Sixth edition.
White Plains, NY: Pearson Education.
Brown, R. (1973). A First Language: The Early Stages. Cambridge, MA: Harvard
University Press.
Brown, R. and Hanlon, C. (1970). Derivational complexity and order of acquisi-
tion in child speech. In Hayes, J. R. (ed.), Cognition and the Development of
Language. New York: John Wiley, pp. 11–53.
Burt, M. and Kiparsky, C. (1974). Global and local mistakes. In Schumann, J. and
Stenson, N. (eds.), New Frontiers in Second Language Learning. Rowley, MA:
Newbury House, 71–80.
Canale, M. and Swain, M. (1980). Theoretical bases of communicative approaches
to second language teaching and testing. Applied Linguistics 1/1: 1–47.
Chen, X., Wang, C. and Kim, D.-H. (2020). Self-regulated learning strategy
profiles among English as a Foreign Language learners. TESOL Quarterly 54/1:
234–251.
Chiswick, B. and Miller, P. W. (2008). A test of the critical period hypothesis for
language. Journal of Multilingual and Multicultural Development 29/1: 16–29.
Chomsky, N. (1965). Aspects of the Theory of Syntax. Cambridge, MA: The M.I.T.
Press.
Clark, E. V. (1993). The Lexicon in Acquisition. Cambridge: Cambridge University
Press.
Clark, H. H. and Clark, E. V. (1977). Psychology and Language: An Introduction to
Psycholinguistics. San Diego, CA: Harcourt Brace Jovanovich.
Corder, S. P. (1967). The significance of learners’ errors. International Review of
Applied Linguistics 5/4: 161–170.
Corder, S. P. (1971). Idiosyncratic dialects and error analysis. International Review
of Applied Linguistics 9/2: 147–159.
Corder, S. P. (1974). Error analysis. In Allen, J. and Corder, S. P. (eds.), The
Edinburgh Course in Applied Linguistics Vol. 3: Techniques in Applied Linguistics.
Oxford: Oxford University Press, pp. 122–154.
Crystal, D. (1986). Listen to Your Child: A Parent’s Guide to Children’s Language.
Harmondsworth: Penguin.
Crystal, D. (1998). Language Play. London and New York: Penguin.
Davies, E. (1983). Error evaluation: the importance of viewpoint. English Language
Teaching Journal 37/4: 304–311.
DeKeyser, R. M. (2000). The robustness of critical period effects in second
language acquisition. Studies in Second Language Acquisition 22/4: 493–533.
DeKeyser, R. M. (2012). Age effects in second language learning. In. Gass, S. M.
and Mackey, A. (eds.), The Routledge Handbook of Second Language Acquiistion.
London and New York: Routledge, pp. 442–460.
Demetras, M., Nolan Post, K. and Snow, C. (1986). Feedback to first language
learners: the role of repetitions and clarification questions. Journal of Child
Language 13: 275–292.
Dörnyei, Z. (2001). Motivational Strategies in the Language Classroom. Cambridge:
Cambridge University Press.
Dörnyei, Z. (2005). The Psychology of the Language Learner: Individual Differences in
Second Language Acquisition. Mahwah, NJ and London: Lawrence Erlbaum
Associates.
Dörnyei, Z. and Ushioda, E. (eds.) (2009). Motivation, Language Identity and the L2
Self. Bristol: Multilingual Matters.
Elliot, A. J. (1984). Child Language. Cambridge: Cambridge University Press.
Ellis, R. (1985). Understanding Second Language Acquisition. Oxford: Oxford Uni-
versity Press.
Faerch, C. and Kasper, G. (1983). Plans and strategies in foreign language
communication. In Faerch, C. and Kasper, G. (eds.), Strategies in Interlanguage
Communication. London: Longman, pp. 20–60.
Falla Wood, J. (2017). Errors in second/foreign language learning and their
interpretations. Education and Linguistics Research 3/1: 1–14.
Gass, S. M. and Selinker, L. (2008). Second Language Acquisition: An Introductory
Course. Third edition. New York and London: Routledge.
Gibbons, P. (2015). Scaffolding Language, Scaffolding Learning: Teaching English
Language Learners in the Mainstream Classroom. Second edition. Portsmouth,
NH: Heinemann.
Griffiths, C. (2008). Strategies and good language learners. In Griffiths, C. (ed.), Lessons
from Good Language Learners. Cambridge: Cambridge University Press, pp. 83–98.
Haastrup, K. and Phillipson, R. (1983). Achievement strategies in learner/native
speaker interaction. In Faerch, C. and Kasper, G. (eds.), Strategies in Interlanguage
Communication. London: Longman, pp. 140–158.
Hakuta, K. (1976). A case study of a Japanese child learning English as a second
language. Language Learning 26/2: 321–351.
Han, Z. (2012). Fossilisation: a classic concern of SLA research. In Gass, S. M. and
Mackey, A. (eds.), The Routledge Handbook of Second Language Acquiistion.
Han, Z. and Odlin, T. (eds.) (2006). Studies of Fossilisation in Second Language
Acquisition. Clevedon: Multilingual Matters.
Herschensohn, J. (2007). Language Development and Age. New York: Cambridge
University Press.
Higby, E. and Obler, L. (2015). Losing an L1 to an L2. In Schwieter, J. W. (ed.), The
Cambridge Handbook of Bilingual Processing. Cambridge: Cambridge University
Press, pp. 645–664.
Holec, H. (1980). Autonomy and Foreign Language Learning. Strasbourg: Council for
Cultural Co-operation of the Council of Europe.
Hughes, A. and Lascararatou, C. (1982). Competing criteria for error gravity.
English Language Teaching Journal 36/3: 175–182.
Hunt, K. (1965). Grammatical structures written at three grade levels. Research
Report No. 3. IL: National Council of Teachers of English.
Hunt, K. (1970). Syntactic maturity in school children and adults. Monographs of
the Society for Research in Child Development 35 (Serial No. 134, No. 1).
Jarema, G. and Libben, G. (2007). Introduction: matters of definition and core
perspectives. In Jarema, G. and Libben, G. (eds.), The Mental Lexicon: Core
Perspectives. Amsterdam: Elsevier, pp. 1–6.
Jespersen, O. (1922). Language: Its Nature, Development and Origin. London: Allen
and Unwin.
Johnson, D. (2005). It’s not what we expected! A case study of adult views in ESL
pedagogy. TESL Reporter 38/2: 1–13.
Johnson, J. and Newport, E. (1989), Critical period effects in second language
learning: the influence of maturational state on the acquisition of English as
a Second Language. Cognitive Psychology 21/1: 60–99.
Johnson, K. (2018). An Introduction to Foreign Language Learning and Teaching.
London and New York: Routledge.
Krashen, S. D. (1978). Individual variation in the use of the monitor. In
Ritchie, W. (ed.), Second Language Acquisition Research: Issues and Implications.
New York: Academic Press, pp. 175–184.
Krashen, S. D. (1981). Second Language Acquisition and Second Language Learning.
Oxford: Pergamon Press.
Krashen, S. D. (1982a). Accounting for child-adult differences in second language
rate and attainment. In Krashen, S. D., Scarcella, R. and Long, M. (eds.), Child-
Adult Differences in Second Language Acquisition. Rowley, MA: Newbury House,
pp. 202–226.
Krashen, S. D. (1982b). Principles and Practice in Second Language Acquisition.
New York: Pergamon.
Krashen, S. D. and Terrell, T. D. (1983/1998). The Natural Approach: Language
Acquisition in the Classroom. Hemel Hempstead: Prentice Hall. (First published
1983 by Alemany Press.).
Labov, W. (1972/1978). Sociolinguistic Patterns. Oxford: Basil Blackwell. (First
published 1972 by University of Philadelphia Press, Philadelphia).
Lado, R. (1957). Linguistics across Cultures: Applied Linguistics for Language Tea-
chers. Ann Arbor, MI: University of Michigan Press.
Lantolf, J. P. (ed.) (2000). Sociocultural Theory and Second Language Learning.
Oxford: Oxford University Press.
Lantolf, J. P. and Poehner, M. E. (2008). Introduction to socio-cultural theory and
the teaching of second languages. In Lantolf, J. P. and Poehner, M. E. (eds.),
Sociocultural Theory and the Teaching of Second Languages. London and Oakville,
CT: Equinox, pp. 1–30.
Lantolf, J. P. and Poehner, M. E. (2013). The artificial development of second
language skills. In Ritchie, W. C. and Bhatia, T. K. (eds.), The New Handbook of
Second Language Acquisition. Second edition. Leiden: Brill, pp. 137–158.
Lantolf, J. P. and Thorne, S. L. (2006). Sociocultural Theory and the Genesis of Second
Language Development. Oxford: Oxford University Press.
Lenneberg, E. (1967). Biological Foundations of Language. New York: Wiley.
Libben, G. and Goral, M. (2015). How bilingualism shapes the mental lexicon. In
Schwieter, J. W. (ed.), The Cambridge Handbook of Bilingual Processing. Cam-
bridge: Cambridge University Press, pp. 631–644.
Lightbown, P. M. and Spada, N. (2013). How Languages are Learned. Fourth
edition. Oxford: Oxford University Press.
Little, D. (1999). Learner autonomy is more than a Western cultural construct. In
Cotterall, S. and Crabbe, D. (eds.), Learner Autonomy in Language Learning:
Defining the Field and Effecting Change. Frankfurt am Main: Peter Lang, pp.
11–18.
Mackey, A., Abbuhl, R. and Gass, S. M. (2012). Interactionist approach. In
Gass, S. M. and Mackey, A. (eds.), The Routledge Handbook of Second Language
Acquiistion. London and New York: Routledge, pp. 7–23.
McLaughlin, B. (1987). Theories of Second Language Learning. London: Arnold.
McLaughlin, B. (1990a). ‘Conscious’ versus ‘unconscious’ learning. TESOL Quar-
terly 24: 617–634.
McLaughlin, B. (1990b). Restructuring. Applied Linguistics 11/2: 113–128.
McLaughlin, B., Rossman, T. and McCleod, B. (1983). Second language learning:
an information processing perspective. Language Learning 33: 135–158.
Moir, J. and Nation, P. (2008). Vocabulary and good language learners. In
Griffiths, C. (ed.), Lessons from Good Language Learners. Cambridge: Cambridge
University Press, pp. 159–173.
Montrul, S. A. (2008). Incomplete Acquisition in Bilingualism: Re-examining the Age
Factor. Amsterdam: John Benjamins.
Naiman, N., Fröhlich, M., Stern, H. H. and Todesco, A. (1978). The Good
Language Learner. Toronto: Ontario Institute for Studies in Education.
Noels, K. and Giles, H. (2013). Social identity and language learning. In
Ritchie, W. C. and Bhatia, T. K. (eds.), The New Handbook of Second Language
Acquisition. Second edition. Leiden: Brill, pp. 647–670.
Norton, B. (2013). Identity and Language Learning: Extending the Conversation.
Second edition. Bristol, Buffalo, Toronto: Multilingual Matters.
Norton, B. and Toohey, K. (2001). Changing perspectives on good language
learners. TESOL Quarterly 35/2: 307–322.
Norton, B. and Toohey, K. (2011). Identity, language learning and social change.
Language Teaching 44: 412–446.
O’Malley, J. M. and Chamot, A. U. (1990). Learning Strategies in Second Language
Acquisition. Cambridge: Cambridge University Press.
Oxford, R. L. (2011). Teaching and Researching Language Learning Strategies.
Harlow: Pearson Education.
Oyama, S. (1976). A sensitive period for the acquisition of a non-native phonolo-
gical system. Journal of Psycholinguistic Research 5: 261–285.
Paradis, M. (2009). Declarative and Procedural Determinants of Second Languages.
Amsterdam: John Benjamins.
Patkowski, M. (1980). The sensitive period for the acquisition of syntax in a second
language. Language Learning 30/2: 449–472.
Patterson, J. L. (2002). Relationship of expressive vocabulary to frequency of
reading and television experience among bilingual toddlers. Applied Psycholin-
guistics 23/4: 493–508.
Pence Turnbull, K. L. and Justice, L. M. (2017). Language Development from Theory
to Practice. Third edition. Boston, MA: Pearson Education.
Pfenninger, S. E. and Singleton, D. (2017). Beyond Age Effects in Instructional L2
Learning: Revisiting the Age Factor. Bristol and Blue Ridge Summit: Multilingual
Matters.
Poulisse, N. (1987). Problems and solution in the classification of compensatory
strategies. Second Language Research 2/3: 141–153.
Press, M. C. (1996). Ethnicity and the autonomous language learner: different
beliefs and learning strategies. In Broady, E. and Kenning, M.-M. (eds.), Promot-
ing Learner Autonomy in University Language Teaching. London: AFLS/CILT, pp.
237–259.
Reiss, M. A. (1983). Helping the unsuccessful language learner. The Canadian
Modern Language Review 39: 257–266.
Richards, J. C. (1974). A non-contrastive approach to error analysis. In
Richards, J. C. (ed.), Error Analysis. London: Longman, pp. 172–178.
Romaine, S. (1995). Bilingualism. Second edition. Oxford: Blackwell.
Romaine, S. (2013). The bilingual and multilingual community. In Bhatia, T. K.
and Ritchie, W. C. (eds.), The Handbook of Bilingualism and Multilingualism.
Second edition. Chichester: Wiley Blackwell, pp. 446–465.
Rubin, J. (1975). What the ‘good language learner’ can teach us. TESOL Quarterly
9: 41–51.
Rubin, J. and Thompson, I. (1994). How to Be a More Successful Language Learner.
Second edition. Boston, MA: Heinle and Heinle.
Saville-Troike, M. and Barto, K. (2017). Introducing Second Language Acquisition.
Third edition. Cambridge: Cambridge University Press.
Saxton, M. (2017). Child Language: Acquisition and Development. Second edition.
London: Sage Publications.
Schachter, J. (1974). An error in error analysis. Language Learning 24/2: 205–214.
Schneider, W. and Shiffrin, R. (1977). Controlled and automatic human
information processing. I: detection, search and attention. Psychological
Review 84: 1–66.
Schumann, J. (1978). The Pidginisation Process: A Model for Second Language
Acquisition. Rowley, MA: Newbury House.
Searle, J. R. (1992). The Rediscovery of the Mind. Cambridge, MA: M.I.T. Press.
Seliger, H. (1978). Implications of a multiple critical periods hypothesis for second
language learning. In Ritchie, W. (ed.), Second Language Acquisition Research.
New York: Academic Press, pp. 11–19.
Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics 10/3:
209–231.
Selinker, L., Swain, M. and Dumas, G. (1975). The interlanguage hypothesis
extended to children. Language Learning 25/1: 139–152.
Serratrice, L. (2013). The bilingual child. In Bhatia, T. K. and Ritchie, W. C. (eds.),
The Handbook of Bilingualism and Multilingualism. Second edition. Chichester:
Wiley Blackwell, pp. 87–108.
Sheen, Y. and Ellis, N. (2011). Corrective feedback in language teaching. In
Hinkel, E. (ed.), Handbook of Research in Second Language Teaching and Learning.
Vol. 2. New York: Routledge, pp. 593–610.
Singleton, D. and Muñoz, C. (2011). Around and beyond the critical period
hypothesis. In Hinkel, E. (ed.), Handbook of Research in Second Language Teaching
and Learning. Vol. 2. New York: Routledge, pp. 407–420.
Stern, H. (1975). What can we learn from the good language learner? Canadian
Modern Language Review 34: 304–318.
Swain, M. (1985). Communicative competence: some roles of comprehensible
input and comprehensible output in its development. In. Gass, S. M. and
Madden, C. G. (eds.), Input in Second Language Acquisition. Rowley, MA: New-
bury House, pp. 235–253.
Swain, M. (1995). Three functions of output in second language learning. In
Cook, G. and Seidlhofer, B. (eds.), Principle and Practice in Applied Linguistics:
Studies in Honour of Henry G. Widdowson. Oxford: Oxford University Press, pp.
125–144.
Swain, M. (2005). The output hypothesis: theory and research. In Hinkel, E. (ed.),
Handbook of Research in Second Language Teaching and Learning. Mahwah, NJ
and London: Lawrence Erlbaum Associates, pp. 471–483.
Swain, M. and Lapkin, S. (1995). Problems in output and the cognitive processes
they generate: a step towards second language learning. Applied Linguistics 16/3:
371–391.
Tarone, E. (1988). Variation in Interlanguage. London: Edward Arnold.
Tarone, E., Frauenfelder, U. and Selinker, L. (1976). Systematicity/variability and
stability/instability in interlanguage systems. In Brown, H. (ed.), Papers in Second
Language Acquisition. Special Issue No. 4 of Language Learning. University of
Michigan: Wiley, pp. 93–134.
Taylor, I. (1978). Acquiring versus learning a second language. Canadian Modern
Language Review 34/3: 455–472.
Ushioda, E. (2003). Motivation as a socially mediated process. In Little, D.,
Ridley, J. and Ushioda, E. (eds.), Learner Autonomy in the Foreign Language
Classroom: Teachers, Learners, Curriculum and Assessment. Dublin: Authentick,
pp. 90–102.
Ushioda, E. (2008). Motivation and good language learners. In Griffiths, C. (ed.),
Lessons from Good Language Learners. Cambridge: Cambridge University Press,
pp. 19–34.
Vygotsky, L. (1978). In Cole, M., John-Steiner, V., Scribner, S. and Souberman, E.
(eds.), Mind in Society: The Development of Higher Psychological Processes. Cam-
bridge, MA: Harvard University Press.
Vygotsky, L. (1986). Thought and Language. Newly revised and edited by Alex
Kozulin. Cambridge, MA: M.I.T Press. (First published in Russian 1934).
Wong Fillmore, L. (1976).The second time around: cognitive and social strategies
in second language acquisition. Unpublished doctoral dissertation. Stanford
University, Palo Alto, CA.
Wood, D., Bruner, J. and Ross, G. (1976). The role of tutoring in problem solving.
Journal of Child Psychology and Psychiatry 17/2: 89–100.
Zimmermann, B. (2000). Attaining self-regulation: a social cognitive perspective. In
Boekaerts, M., Pintrich, P. R. and Zeidner, M. (eds.), Handbook of Self-
Regulation. San Diego, CA: Academic Press, pp. 13–39.
2 Language Teaching Methods
The Grammar-Translation Method

In Europe, throughout the Middle Ages and the Renascence, right up to
the seventeenth century, Latin was the language used for international
scholarship and cultural exchange. Renascence scholars such as Sir
Thomas More and Erasmus corresponded in Latin. Only from the
seventeenth century onwards did the use of Latin as a lingua franca
slowly decline as French progressively took over as the European lingua
franca. Even so, Latin remained the language of scholarship and was
a central component of the education of the European aristocracy and
the expanding middle class throughout the eighteenth century and into
the nineteenth century. The grammar schools in Britain were so called
because they taught Latin grammar, not English grammar.
It was not until the nineteenth century that the teaching of modern
foreign languages started to attract the interest of educators and philolo-
gists in Britain and other European countries, notably France and
Germany, on a large scale. As foreign languages became part of the
school curriculum, it was natural that the method which had long been
used for the teaching of Latin and Greek was applied to teaching modern
languages. This method is now known as the Grammar-Translation
Method. The first Grammar-Translation course books for teaching
French and English appeared in Germany at the very end of the eight-
eenth century (Howatt with Widdowson 2004: 152) and the method was
almost universally used in school teaching from the mid-nineteenth
century until the mid-twentieth century for both the classical languages
and modern languages. It is still in use in many places today, particularly,
but not only, for teaching Latin and Greek. Great emphasis is placed on
accurate and close translation. The focus is on literacy rather than on
oral/aural skills, which is justifiable for Latin and Greek, which are ‘dead’
languages in the sense that they do not possess a surviving community of
native speakers, namely Romans or Ancient Greeks, with whom one
might need to converse. The method was also unproblematic for modern
languages provided one did not have to converse with native speakers.
In short, in the Grammar-Translation Method:
The mother tongue was the language of instruction.

Great emphasis was placed on starting to read and translate authentic
texts as soon as possible.
Translation was done both from the native language into the foreign
language and vice versa.
At first contrived (invented) sentences were translated and at later
stages authentic texts were used.
The overall goal was not only to be able to read, write and translate,
but also to benefit intellectually from the mental discipline involved
in mastering the language.
Grammar was taught deductively, that is, by means of rules and
example sentences.
Vocabulary was encountered in authentic texts and rote-learned with
mother tongue translations.
See Richards and Rodgers (2014: 6–7), Howatt with Widdowson (2004:
151–165), Cook (2001: 201–205) and Larsen-Freeman and Anderson
(2011: 13–23).
The Direct Method

In many ways the Direct Method was the complete reversal of the Gram-
mar-Translation Method. It aimed at enabling adults, many of whom had
already had some experience of learning a foreign language at school, to
develop oral conversational skills useful for transactions with speakers of
the foreign language in everyday life. It was never used in the state secondary
school system and was not intended to be, but with some modifications it
remained the basis of language teaching in many private language schools in
Europe right up to the 1970s. It was born in the late nineteenth century as
international contacts for leisure and business purposes grew and the
inadequacies of the Grammar-Translation Method for those who needed
to acquire conversation skills became increasingly apparent. In particular,
those who had learned by the Grammar-Translation Method were often at
a loss for words when confronted with a native speaker, unused as they
were to listening to natural spontaneous speech spoken at normal pace, or
to speaking the language themselves.
The Direct Method is sometimes called the Berlitz Method in allusion
to the Berlitz private language schools founded by Maximilian Berlitz,
who as a young German immigrant to the United States opened his first
language school in Providence, Rhode Island in 1878. Although Berlitz
did not invent the method, he popularised it on both sides of the
Atlantic. The first Berlitz School in Germany opened in Berlin in 1888,
Language Teaching Methods 33
and by 1914 there were nearly 200 Berlitz schools altogether, including
27 in Britain (Howatt with Widdowson 2004: 222).
Tuition by the Direct Method took place in small groups or on a one-
to-one basis, was orally based and was steered by the teacher, who asked
a series of more or less scripted questions to which the learners
responded. In addition to answering questions, learners sometimes had
to repeat sentences said by the teacher and perform actions as a response
to teacher directives. The Direct Method teacher was a very different
animal from the foreign language teacher in the secondary school system.
He or she was usually a native speaker of the language to be learned (the
target language), not a trained language teacher, preferably not a speaker
of the learner’s native language and in any case not allowed to use the
learners’ native language in class.
Grammar was not taught explicitly but was to be acquired implicitly by
the carefully planned sequences of teacher questions which introduced
grammatical structures according to a pre-set plan. Learners responded to
these questions with the help of initial prompts or even model answers
provided by the teacher. However, a translation of the teacher’s question
was never provided. Vocabulary was also introduced implicitly and was
explained, demonstrated, exemplified or illustrated by the teacher but
never translated. The teacher used language which was at least partly
understandable from context. In particular, great use was made of everyday
objects and pictures as well as gesture, facial expression, demonstration and
mime to make the meaning clear. The emphasis was overwhelmingly on
oral/aural skills with very little reading or writing. There were course books
which summarised each session, but these were mainly intended to be
looked at after the material had first been introduced orally.
In short, in the Direct Method:
The target language was the language of instruction.

The aim was to be able to understand and converse with native
speakers.
The teacher was usually a native speaker and not a trained teacher.
The emphasis was on everyday language.
Grammar was not explicitly taught.
Vocabulary was not introduced by translation but by demonstration,
mime, pictures and realia.
Emphasis was placed on grammatical and phonological accuracy,
with grammar and pronunciation mistakes being corrected (but not
explained) by the teacher and the corrected version being repeated by
the learner.
See Richards and Rodgers (2014: 11–14), Howatt with Widdowson

(2004: 217–228), Jin and Cortazzi (2011: 563–567) and Larsen-Freeman
and Anderson (2011: 25–34).
The Audiolingual Method
The next major innovation in language teaching was the Audiolingual
Method, which was developed in the United States in the 1950s. Its origins
go back, however, to the need in the immediate aftermath of the Second
World War to provide military personnel with foreign language interview-
ing skills so as to deal with displaced persons, refugees and prisoners of war.
The Audiolingual Method, claimed to be the first scientific method of
language teaching, based as it was on a behaviourist view of learning and
a structuralist view of language. Structuralist linguistics, as expounded by
the great American structuralist linguist Leonard Bloomfield (Bloomfield
1933), viewed grammar not so much as a system of abstract rules but as sets
of recurring sentence patterns in language use. This was an ethnographic
approach to grammar, employed by linguists investigating languages which
had no written records, for example the Native American languages. In their
field work ethnographers enlisted the help of a number of native speakers of
the language and proceeded by observing recurring patterns in speech and
inferring the meaning of sentences by comparing a sentence whose meaning
was known to a sentence similar in form whose meaning was unknown.
This approach to grammar as recurrent sentence patterns was taken up for
language teaching by Charles Fries (Fries 1945), who emphasised that the
sounds of a language must be mastered first, then the syntactic structures,
with vocabulary being held down to a minimum. Finally, vocabulary could
be expanded by inserting relevant new words into structures which had
already been mastered (a ‘slot-and-filler’ view of language):
Often ‘substitute frames’ furnish a simple device for the introduc-

tion or the practice of new words. Such ‘frames’ are sentences of the
same pattern and general context, in one part of which it is possible
to substitute a variety of words of the same class all of them fitting
the context.
(Fries 1945: 55)
The syllabus for Audiolingual courses was therefore grammatically

based, but the traditional grammatical categories of Latin and Greek
grammars, which the Grammar-Translation Method used, were aban-
doned in favour of sentence-length structures or patterns. Selection and
grading of structures was based on contrastive analysis, that is to say the
comparison of the language to be learned (target language) and the native
language. Contrastive analysis focussed on the grammatical systems of
the two languages so as to identify points of difference between them, for
example the Saxon genitive in English (‘my aunt’s pen’) versus the
prepositional construction for possession in French (la plume de ma
tante). The target language structures to be concentrated on were those
which differed most from the structures used in the native language to
express the same idea. Contrastive analysis was also applied to the sound
systems of the two languages (phonology), to identify sounds which did
not occur in the native language, and to the vocabulary (lexis) so as to
identify ‘false friends’, that is, words which sound similar in the two
languages but have different meanings, for example ‘circulation’ (English)
and circulation (French), which means ‘traffic’. However the main focus
was on grammatical structures (syntax).
The scholar most closely associated with Audiolingual Language
Teaching is Robert Lado, who studied under Fries at Michigan. It was
he who combined structuralism with insights from behaviourist psychol-
ogy, which sees human learning as being largely explainable in terms of
stimulus and response. Lado (1964: 36–37) enunciated five theoretical
principles, which he called ‘laws of learning’ which underlie Audiolingual
Language Teaching, here reproduced verbatim:
‘the fundamental law of contiguity’: when two experiences have occurred

together, the return of one will recall or reinstate the other.
‘law of exercise’: other things being equal, the more frequently
a response is practised, the better it is learned and the longer it is
remembered. Contrariwise, when a response is not practised, it tends
to be forgotten.
‘law of intensity’: other things being equal, the more intensely
a response is practised, the better it is learned and the longer it will
be remembered.
‘law of assimilation’: each new stimulating condition tends to elicit
the response which has been connected with similar stimulating
conditions in the past.
‘law of effect’: other things being equal, when a response is accom-
panied or followed by a satisfying state of affairs, that response is
reinforced. When a response is accompanied or followed by an
annoying state of affairs, it is avoided.
(Lado 1964: 36–37)
In the Audiolingual Method, emphasis was on listening and speaking

but in a highly programmatic way. Typically, an Audiolingual lesson
might begin with the class listening to a model dialogue contrived to
introduce a particular structure (for example, the instrumental structure,
using ‘with’, as in ‘I stir my tea with a teaspoon’). The example sentence
or sentences would then be repeated several times by the class, following
the teacher’s model, first in chorus, then individually. Next, a whole set
of different examples of the same structure but with different lexis (for
example, ‘I clean my teeth with a toothbrush’) would be repeated
(pattern drills). After that, learners might be directed, on the basis of
a model example such as ‘Do I stir my tea with a teaspoon?’ to make the
sentences they had been repeating into questions (manipulation drills).
The final step might be chain drills, in which one student asks another,
‘Do you stir your tea with a teaspoon?’ This student might reply,
‘Yes, I stir my tea with a teaspoon.’ Then on the basis of a cue
(stimulus) such as ‘open, door, key’ this student might ask the next
student, ‘Do you open your door with a key?’, and so it would go on
right around the class. Further refinements might include a cued chain
drill for questions which expect a negative answer such as ‘stir your
tea/toothbrush’ so that one learner asks, ‘Do you stir your tea with
a toothbrush?’ and the next learner answers, ‘No, I stir my tea with
a teaspoon.’ New structures could then be introduced using the same
lexis, for example ‘What do you clean your teeth with?’ and new
chain drills created.
The main focus was on learning structures and learners generally had
to work within the narrow range of vocabulary already introduced rather
than choose new vocabulary not yet practised since mistakes could arise
in this way. Just enough vocabulary was introduced to practise the
structure in question. If learners did make mistakes, they were immedi-
ately corrected by the teacher and required to repeat the sentence
correctly. Sometimes the whole class was required to repeat the sentence
to make certain that they did not pick up the error which a single learner
had made. The entire lesson was highly teacher-centred with the learner’s
role being to respond correctly to teacher stimuli. The military origins of
the method have been mentioned and it is reminiscent of the parade
ground in some respects. With the advent of language laboratories in the
1960s, audiolingual material could be recorded and learners could work
through the exercises systematically, receiving feedback for each utter-
ance they made.
In stark contrast to the Grammar-Translation Method, learners did very
little reading or writing. However, like the Grammar-Translation Method,
the Audiolingual Method failed to meet the needs of learners when they
had to converse spontaneously with native speakers. Learners tended to
remember the sentences they had practised in class, which usually turned
out to be communicatively useless in the real world: ‘I stir my tea with
a teaspoon’ is an authentic example from a course book I had to use many
years ago. For real-world language use, students had usually never learned
enough vocabulary to be able to fill in the slots appropriately in the
underlying structures. Even if they did have the requisite vocabulary
stored, the task of retrieving the words and the relevant sentence frame
and then inserting the words into the frame under time pressure, for
example while buying a ticket at a busy major railway station, was
insurmountable.
Furthermore, learners rarely practised speaking more than one sen-
tence at a time so that they were unable to cope with speaking even a few
sentences fluently at a stretch. The insistence on correct pronunciation
and grammar meant that progress was slow and learners lacked
confidence in their own abilities. A further drawback of the Audiolingual
Method was that classes tended to be either terribly boring or terribly
stressful. They might be boring because of all the repetitions, corrections
and lack of freedom to be original plus the sheer inanity of repeating
sentences with no scrap of relevance to any meaningful reality. They
might be stressful because – to avoid boredom themselves and to keep
things moving – teachers often adopted a no-nonsense, slightly bullying
stance (‘hurry-up now’, ‘answer right away’). This could be most unset-
tling for anxious learners and quite stressful even for those learners who
started out well-motivated and self-confident.
However, although now largely discredited, the Audiolingual Method has
not died out completely and there has been a renewed interest in the role of
rote learning, imitation and repetition in language learning in recent years
(see Scrivener 2011: 169–174; Harmer 2015: 234–235; Ur 2012: 82–83).
In short, in the Audiolingual Method:
The goal was automatic, unconscious mastery of the language.

The native language was banned in the classroom because language
mastery was equated with habit formation and the habits established
in and by the native language were held to interfere with the learning
of the new language (negative interference).
Learners were expected to respond immediately to stimuli and were
deliberately not given time to think about their responses.
Error occurrence was to be prevented and, if errors occurred, they
were to be immediately corrected by the teacher and then the correct
form was to be repeated by the learner.
Learners first listened to a new structure and then practised saying it.
Grammar was not taught explicitly, but rather learners were expected
to grasp the grammar intuitively by means of examples.
Correct pronunciation was insisted on from the start.
Vocabulary was in the early stages to be kept to a minimum until the
main structures had been mastered.
Drilling, that is repetition and repetition with variation, of structures
characterised Audiolingual Language Teaching.
See Richards and Rodgers (2014: 58–80), Johnson (2018: 151–155), Jin
and Cortazzi (2011: 567–568), Cook (2001: 206–211), Larsen-Freeman
and Anderson (2011: 35–50).
Situational Language Teaching (SLT)

The Audiolingual Method was never as popular in Britain as it was in the
United States. For British EFL teaching from 1960 onwards, Situational
Language Teaching (SLT) was the dominant method employed. An SLT
lesson consisted of three phases: ‘Presentation’, ‘Practice’ and
‘Production’. Frisby (1957: 136) justified this PPP lesson structure by
arguing that language learning involved first reception of new material,
then fixing it in memory and finally using it in actual practice until its use
became habitual. SLT differed from the Audiolingual Method in aiming
to achieve balanced competence in all four skills: listening, speaking,
reading and writing. However, any new structure or vocabulary item
taught was first to be introduced orally and its pronunciation mastered
before it was introduced in written form. As Pittman put it:
The learner has to hear the new material. He has to practise saying
that material until he knows it; and then, and only then, he may read
it or write it.
(Pittman 1967: 147)
SLT also placed great weight on the rigorous selection and grading of
structures and vocabulary with the simple being introduced before the
complex, the frequent before the infrequent and the more useful before
the less useful.
Structures were to be introduced in situations (Pittman 1967:
148–157). One way of doing this was by demonstration, using the class-
room surroundings. The present progressive, for example, could be
taught by the teacher opening and closing the window and saying, ‘I am
opening the window’ and ‘I am closing the window.’ However, many
structures cannot be taught using just the immediate environment of the
classroom, and so the SLT textbook was an important feature of the
method, as well as various visual materials such as wall-charts, flashcards,
pictures and drawings. SLT textbooks were specially written with scenes,
stories and illustrations to convey each structure. The idea was that the
situation should make the meaning clear without explicit grammatical
explanation (inductive learning). Each new structure was introduced
within a particular situation in a short text which told a story in a clearly
defined setting. This could be a picnic excursion, a house on fire, a bank
robbery or a visit to the circus. The texts were often accompanied by
a picture or pictures to further contextualise the language. The texts were,
however, highly contrived and chosen not for their thematic content but
to serve as mere vehicles to transport structures and vocabulary. They
have been much criticised since for their artificiality. But to write these
texts was a daunting task. They had to display the new structures in
sufficient variety for inductive learning of grammar to be possible, and
they could not include structures deemed too advanced to be encountered
at a particular stage. They also had to use the limited range of vocabulary
already known and could only introduce a limited number of new words,
and these words had to be of general service to learners. It was also very
difficult to select and grade vocabulary and at the same time fulfil the
lexical demands of the chosen topic.
In SLT textbooks the text and pictures constituted the raw mate-
rial for the Presentation phase. This was followed by rather mechan-
ical questions, substitution drills and other exercises (Practice phase)
and then a final section which involved freer work, often written
work, based on the text (Production phase). The lesson was generally
highly teacher-centred and learners performed under close teacher
control. Only in the final Production phase was more freedom
allowed and this phase might involve interactive group work of
various sorts.
In short, the main features of SLT were:
All four skills were covered but priority was placed on oral language.
Teaching took place in the target language.
The syllabus was organised according to structures.
Structures and vocabulary were rigorously selected and graded.
New structures and vocabulary were introduced and practised
in situations.
Structures were taught inductively.
Vocabulary was introduced so as to illustrate the structures in the
specific situation.
Vocabulary was selected and graded to ensure a basic service voca-
bulary became established.
In practice there were often conflicts between structural and lexical
selection and grading.
Through repetition and rehearsal error was to be avoided (practice-
makes-perfect principle).
Specially written short texts were used to introduce structures diffi-
cult to demonstrate in the classroom.
Each text had its own special setting, but taken as a whole the texts
sometimes formed a fictional narrative with recurring characters.

(2004: 249–250) and Cook (2001: 224–228).
Total Physical Response (TPR)

By the late 1960s the idea became popular, on the analogy of the child
acquiring its first language, of not forcing beginners to speak from the
very first lesson but to give them a ‘silent period’ in which they could
first just listen. In this respect, an interesting innovation in language
teaching was Total Physical Response (TPR), developed by James Asher,
a Californian psychologist (Asher 1969, 2009). TPR set out to link
language to action, supposedly in the way children acquire their mother
tongue (Asher 2009: 2.17–19; 3.2). According to Asher, there are three
guiding principles of TPR, which are here reproduced verbatim:
Understanding the spoken language should be developed in advance
of speaking.
Understanding should be developed through movements of the
student’s body.
Do not attempt to force speaking from students. As students inter-
nalise a cognitive map of the target language through understanding
what is heard, there will be a point of readiness to speak. The
individual will spontaneously begin to produce utterances.
(Asher 2009: 2.4)
In TPR the teacher gives commands which are carried out by the
whole class (‘Put your hands on your knees’) or individuals (‘Peter,
give your pen to Mary’). The commands which shape TPR lessons are
planned in advance down to the last detail. They come thick and fast
and are executed promptly in a fun atmosphere. The idea is that
internalisation of language takes place more effectively in this way
because motor activity occurs not just in the language-specific areas of
the brain but also in those involved in performing the relevant action
so that a neural link between language and action is established. This
is what is meant by ‘kinaesthetic learning’ (Asher 2009: 3.19). Asher
contends that TPR involves both hemispheres of the brain (see Asher
2009: 2.19–29; 3.4–7).
Grammar is not explicitly taught. The transition from the imperative
to other grammatical structures is made by nesting these within impera-
tives. For example, the past tense might be introduced as follows:
Josephine, if Abner ran to the blackboard, run after him and hit him
with your book.
(Asher 2009: 3–43)
According to Asher (2009: 2.4), ‘most of the grammatical structures of

the target language and hundreds of vocabulary items can be learned
through the skilful use of the imperative by the instructor’ (see also
Asher 2009: 3.37). Asher (2009: 3–44) suggests 12–36 new words can be
introduced in a one-hour lesson. This seems a lot by modern standards,
but only receptive, not productive, knowledge is required initially.
Learners are expected to recognise and respond to novel combinations
of items which have already been introduced in isolation, for example,
‘Sit on the table’ when ‘Walk to the table’ and ‘Sit on the chair’ are
already known (Asher 2009: 3–46). After perhaps 10–20 hours of instruc-
tion in this manner, learners are encouraged to take over the teacher’s
role and speak themselves (Asher 2009: 3.45; 4.18). In TPR students are
thus not ‘pushed’ to speak before they feel ready to do so. Initially,
learners’ pronunciation mistakes are not corrected until students are
‘ready to accept corrections’ (Asher 2009: 3.35). After 10–12 hours of
TPR, reading may be introduced but ‘the reading content is confined to
utterances students have thoroughly internalised through the imperative’
(Asher 2009: 3.25).
TPR was a radical departure from conventional language-teaching
wisdom of the time in not recommending blanket error correction. It
was Asher’s contention that inaccuracies are not important in the early
stages of speaking, based on the analogy of the child’s first efforts to
speak its mother tongue. After about 120 hours of teacher-centred
instruction, conversational dialogues are introduced. Other classroom
activities include role-play and slide presentations (to introduce vocabu-
lary). There is no course book and initially hardly any materials. At later
stages classroom objects and then pictures, slides, realia, word charts and
student and teacher kits are introduced (Richards and Rodgers 2014:
283). Error correction is kept to a minimum even in the later stages, and
reading and writing are introduced late. TPR is not well suited to coping
with abstract language, indirect language, the hypothetical, the condi-
tional and modality. Asher’s suggestion is to use pictures and words
written on cards (Asher 2009: 3.22–24). He also suggests TPR can be
used as a supplement to a course rather than the guiding principle of
a course (op. cit.: 3.18–19).
In short, in TPR:
The emphasis is on making learning an enjoyable experience in

a relaxed atmosphere.
The lesson is organised around directives and action-responses rather
than repetition and drills.
In its early stages TPR is highly teacher-centred but becomes learner-
centred later as learners take over the role of giving directives and,
still later, engaging in dialogues.
Error correction is kept to a minimum.
See Richards and Rodgers (2014: 277–288), Asher (1969, 2009) and
Larsen-Freeman and Anderson (2011:103–114).
Community Language Learning (CLL)

Community Language Learning (CLL) was one of a number of alterna-
tive, humanistic methods of language teaching which grew out of dis-
content with the Grammar-Translation and Audiolingual Methods.
Humanistic education can trace its origins back to Rousseau’s educa-
tional treatise Émile (1762/2009), which inspired early twentieth-century
reforming educators in Europe such as Montessori, Pestalozzi and
Rudolf Steiner. Developed by Charles Curran (Curran 1972, 1976),
a counsellor and professor of psychology, CLL seeks to develop the
personality of the learner as a whole rather than merely imparting
knowledge and skills; it emphasises personalised learning and stresses the
importance of affective and motivational aspects (Curran 1972: 26–27;
1976: 2–3, 22–23, 35–41).
The idea behind CLL is the encounter group in counselling and
psychotherapy (Rogers 1970) and Curran was much influenced by Carl
Rogers. The language class becomes an encounter, and the teacher, as
‘knower’, strives to help learners to become knowers without violating
their individuality. ‘Community’ refers to the learner group. CLL places
emphasis on learning as a social process. Only if the learners develop
into a community in which mutual trust replaces initial anxiety and
hostility will learning take place (Curran 1972: 28–36; 1976: 5, 22–25).
The group sits in a circle, with the teacher being outside the circle. The
learners decide on the topic to talk about and one starts talking. In the
case of complete beginners, the speaker will say a sentence in the native
language. The teacher, who commands the learner’s native language or
has an assistant who does, will translate sotto voce in chunks into English
and the learner will repeat each chunk out loud to the class. Another
learner will then continue in the native language. What learners say in
the target language will be recorded. After about 15 minutes the class will
listen to the recording played back to them.
If learners are not complete beginners, a learner starts by talking in
English about the topic agreed on by the group as best he or she can. The
teacher renders any non-English parts of the narrative into English and the
learner may incorporate these into the narrative, but will not be forced to.
At a later stage of development the teacher may provide corrective feed-
back. At even later stages, as the group starts to feel more secure,
collaborative narratives may be created, with the whole group offering
corrections and suggestions for stylistic improvement. To round off each
session, the teacher may record a monologue of the final spoken text.
Learners may listen to the recorded monologue outside class and may
transcribe it. Such transcriptions replace a course book, since no course
materials are used (see Curran 1972: 145–157).
In practice, CLL seems to be used more with adults than children,
especially adults of low proficiency who all have the same native language
and are striving to achieve basic conversational fluency. Like TPR, it can
also be used as a supplementary technique rather than as a complete
method. For more on CLL, see Curran (1972, 1976), Richards and Rodgers
(2014: 303–316) and Larsen-Freeman and Anderson (2011: 85–101).
Communicative Language Teaching (CLT)

The advent of Communicative Language Teaching (CLT) in the 1970s
represented a major paradigm shift, the influence of which persists to
this day. Current approaches to language teaching are refinements and
further developments of the communicative approach rather than further
paradigm shifts. CLT grew out of dissatisfaction with traditional lan-
guage teaching, was not motivated by a single all-embracing theory of
language or language learning and was a macro-level approach to language
teaching rather than a detailed, clearly laid out method at the micro-level
of classroom procedures. It revolutionised language teaching by placing
the major focus on the meaning of utterances in a situational context
(pragmatic meaning).
Pragmatics was the level of language which up to that time had been
largely neglected in language teaching. By the 1970s, however, it had become
clear that many learners, especially adults, who had achieved a reasonable
pronunciation, commanded the basic structures and had a sizeable vocabu-
lary, were unable to engage successfully in conversational ‘small talk’ in
English-speaking environments. They often experienced an inability to
apply their formal knowledge of grammar and vocabulary effectively in real-
life situations. They had trouble understanding longer stretches of natural,
fluent colloquial speech delivered at normal conversational speed because
they had been used to a teacher who spoke to them in idealised, slowly
spoken, single sentences. They often did not understand the idiomatic turns
of phrase used by native speakers in real-world situations either.
The goal of CLT was the achievement of ‘communicative competence’,
a term used by Hymes (1972), who argued that linguistic competence in
the sense used by Chomsky did not do justice to the abilities of
a proficient speaker of a language. To communicate effectively, not only
command of the formal system but also of socio-linguistic skills (‘rules of
use’) was required. He emphasised that not only must a learner know
whether and to what extent utterances are formally ‘possible’ (linguistically
correct), but also whether they are ‘feasible’ (not stylistically awkward),
‘appropriate’ (suitable in context) and ‘performed’ (actually used).
Before the advent of CLT, language teaching had tended to focus
mainly on the descriptive and narrative functions of language to the
neglect of other functions, especially the phatic function (the use of
language for purposes of social bonding). Phatic language became an
important component of CLT, influenced by the work of John Firth
(Firth 1950, 1951) and his student Michael Halliday (Halliday 1973). CLT
also drew on work in pragmatics by the language philosophers John
Austin (Austin 1962) and John Searle (Searle 1969). Austin’s book was
tellingly entitled How to Do Things with Words. It introduced the idea of
the speech act, which was further developed by Searle. In this perspec-
tive, apart from the literal or propositional meaning of an utterance (its
‘locutionary force’), two other aspects were of interest. The first was the
effect the speaker intended the utterance to have on the listener (‘illocu-
tionary force’), and the second was the effect it actually did have on the
listener (‘perlocutionary force’). Of particular interest for CLT were
‘indirect speech acts’, in which there was a discrepancy between syntactic
form and utterance function. An example would be a conditional
sentence functioning as a threat, as in, ‘If you do that again, I’ll break
your neck.’ Another example would be an interrogative sentence func-
tioning as a request as in, ‘Could you pass me the butter.’ In many cases
the speaker’s functional intention can only be understood in context,
and is often indicated by tone of voice.
In CLT less emphasis was placed on error correction and explanations
and more emphasis on using the language to communicate. This was in
line with the ideas expressed in Corder’s (1967) paper, ‘The Significance of
Learners’ Errors’, which was discussed in Chapter 1. It will be recalled
that Corder saw the language learner as retracing the path of the child
acquiring its first language, perhaps even following a biologically predeter-
mined ‘inbuilt syllabus’, but also employing cognitive skills in an active
process of hypothesis-testing to explore the parameters of the target
language (see also Mystkowska-Wiertelak and Pawlak 2012: 32–34).
The focus in the early days of CLT was on the spoken rather than the
written language. A CLT lesson might, for example, involve a simulation
or role-play in which groups of learners choose a package holiday from
a number of possibilities available in brochures or, nowadays, on the
internet. Their task might be to agree on the best solution for them
personally, perhaps within a stated budget and with other constraints.
According to Johnson and Johnson (1998), the success of
a communicative activity can be assessed post hoc by asking five ques-
tions, as follows:
Was language used to successfully transmit meaning?

Was the language situationally and socio-linguistically appropriate?
Did the task successfully stimulate cognitive processing?
Did learners indulge in the right level of risk taking?
Did learners have the opportunity to practise using the language
freely?
The communicative revolution, as it is sometimes called, also influenced

syllabus design. In this respect one may distinguish between ‘strong’ and
‘weak’ forms of CLT (Howatt 1984: 279; Littlewood 2011; Richards and
Rodgers 2014: 84–87). In its most radical form CLT abandoned the idea
of a syllabus altogether. Less radical forms strove to replace the tradi-
tional grammatical syllabus with a ‘notional-functional syllabus’, based
on notions such as ‘time’, ‘place’ and ‘the hypothetical’ and functions
such as ‘describing’, ‘narrating’, ‘apologising’ and ‘complaining’ (Wilkins
1976). However, it proved difficult to base selection and grading of
language material on notions and functions and so the grammar-based
syllabus was often retained but was supplemented by a functional/
notional strand. Strong forms of CLT also abolished the traditional,
rigidly sequenced course book of earlier times and replaced it with
a course book which was little more than a resource pool of materials.
Weaker forms of CLT retained a sequenced course book, organised so
as to cover both grammatical and notional/functional factors, but
included in addition quasi-authentic material. De-contextualised gram-
mar and vocabulary exercises now gave way to, or at least were supple-
mented by, communicative activities such as role-play, simulation and
group problem-solving activities, for example information-gap tasks and
jigsaw tasks. The teaching of writing was also later incorporated into the
weak communicative approach. Emphasis was placed on the conven-
tional language of various ‘communicative’ written genres, such as
letters, reports, newspaper articles and, more recently, e-mails and
blogs; in other words, on register (see Halliday 1978: 31–35).
CLT was not without its problems for teachers and learners, however.
One of the problems encountered by teachers whose pupils all spoke the
same mother tongue was that there was no real reason to communicate
in the foreign language in the classroom. Even in classes of learners with
various first languages, it is difficult to create opportunities for authentic
communication so that CLT classes tend to take on an element of make-
believe and ‘pretending’. As Littlewood (2011: 553) remarks, ‘Students
are sometimes given a steady diet of “planning a party” that will never
take place or “giving directions to the station” on the map of a non-
existent town.’ Not all learners are prepared to suspend disbelief as
required by these techniques.
In short:
• CLT grew out of disillusionment with traditional grammar-based

teaching and grammar-based syllabuses.
• Its focus was on language in use in specific situations (the hitherto
neglected area of pragmatics).
• The emphasis was on communication and learner involvement.
• The idea was to create situations in the classroom in which language
would be used naturally, the classroom would become a microcosm
of the real world and naturalistic language acquisition would thus
occur.
• The learner was viewed as an explorer of the language, an experi-
menter with the language and a negotiator in the language.
• The teacher was a facilitator of learning who provided input and
managed classroom activities rather than explaining the language
system, drilling syntactic structures and correcting errors of pronun-
ciation, grammar and vocabulary.
• The important thing was for learners to get their intended meaning
across rather than to produce formally correct utterances.
• Strong forms of CLT banned error correction, grammatical explana-
tion and use of the mother tongue from the classroom .
• In weaker forms of CLT, errors were selectively corrected, often at
the end of communicative activities so as not to disturb fluency, and
grammatical explanations in the mother tongue or in the target
language were permitted.

(2004: 326–349), Brown (2014: 205–236), Johnson (2018: 155–166),
Cook (2001: 211–220), Littlewood (2011) and Larsen-Freeman and
Anderson (2011: 115–130).
Task-Based Language Teaching (TBLT)

Task-Based Language Teaching (TBLT) is a further development of CLT.
By the 1990s efforts were being made to organise language teaching
around tasks which were performed by learners working together in
small groups of two to six or eight persons. The emphasis was on
successfully completing the task. This usually involved arriving at
a decision and often involved producing a physical product of some
sort (a report, a written recommendation, a brochure, a set of instruc-
tions or guidelines, for example). Both ‘authentic’ or ‘real-world tasks’
and ‘pedagogical tasks’ were deemed to be suitable for TBLT (Ellis 2003:
339, 347, 351). According to Ellis (2003: 6), authentic or real-world tasks
‘correspond to some real-world activity’. Real-world tasks are those
which learners might have to perform outside the classroom, for example
asking for and giving directions, booking a hotel room, filling
filing in various
sorts of forms. Pedagogical tasks include games, competitions and pro-
ject work of various sorts (see Legutke and Thomas 1991). Howatt with
Widdowson (2004: 366–367) justified the inclusion of pedagogical tasks
by arguing that the important thing is that the tasks engage learners’
imagination so that learners are concentrating not so much on language
learning as on the activity itself. In this way language learning occurs
incidentally.
TBLT involves learner empowerment. This means that the teacher
needs to release control and let learners work in small groups or pairs
not just for a few minutes at a time, as in CLT, but for almost the whole
lesson after a short initial phase in which the task is set up and explained
by the teacher. Attention is also paid to social forms of working,
especially to seating patterns, and the traditional pattern of rows of
learners facing a teacher at the front is abandoned. Instead, learners
may sit around tables in groups of perhaps four to six, or sit in
a horseshoe pattern, with the open part of the horseshoe being left for
the teacher to sit in. In particular, however, movement becomes a feature
of the classroom, with learners moving from one group to another,
seeking information or seeking a class member who possesses specific
information needed to complete the task. Furthermore, the working
groups are likely to be fluid and to change in composition according to the
requirements of each individual task. This social dimension of TBLT
provided the stimulus for further developments in language-teaching meth-
odology, especially Cooperative Language Learning and Self-Directed Lan-
guage Learning (SDLL), both of which will be discussed below.
In short, in TBLT:
The teacher may initially introduce a task, provide materials and set
an overall plan or timeline for the lesson before learners divide into
groups and work more or less autonomously.
After this preparatory phase, learners may plan, then perform and
finally reflect on the activity. The three phases (plan, perform, reflect)
are often performed cooperatively in groups, under the general
guidance of the teacher who may circulate from group to group to
monitor performance and progress and respond to any learner ques-
tions or appeals for help.
The teacher is available to facilitate accomplishment of the task, which
learners nevertheless perform in as autonomous a fashion as possible.
The learner is expected to take on an active and cooperative role in
all three phases of the lesson: planning, performance and reflection.
Traditional forms of assessment, especially traditional written exam-
inations, are often replaced by peer assessment and self-assessment,
project work and portfolio work.
For more on TBLT see Ellis (2003), Nunan (2004), Willis and Willis
(2007), East (2012), Richards and Rodgers (2014: 174–199), Brown (2014:
237–238), Johnson (2018: 167–172) and Larsen-Freeman and Anderson
(2011: 149–163).
Cooperative Language Learning (CLL)

Cooperative Language Learning differs only in emphasis from TBLT in
that it places the focus firmly on cooperation rather than task completion.
Tasks need not necessarily end in the achievement of a goal. Cooperative
forms of learning are underpinned by socio-cultural theory, in particular
the work of the Russian educationalist Vygotsky (1978, 1986), discussed in
Chapter 1. It was at first thought that the insights of Vygotsky were of
little relevance for classroom language teaching because they assumed
a one-to-one adult–child or novice–expert situation, whereas in the lan-
guage classroom there is only one expert (the teacher) available for a whole
class of novices. However, modern applications of Vygotsky’s ideas
include the idea of peer support, in which learners mutually support one
another in pairs or groups. In this way the group achieves more than any
of the individual members could alone.
Like TBLT, CLL involves learners working in small groups. The
implementation of group work in CLL, as in TBLT, may pose problems,
at least initially, and there are certain pitfalls which can be avoided:
Firstly, the aims of any specific piece of group work should be made
explicit by the teacher (or in self-directed project work by the
learners themselves) (Dam 1994: 509).
Secondly, for classes who have no experience of group work the
teacher may have to set up a framework within which the group
work is performed. This may include assigning specific roles to
certain group members, for example a secretary to keep a written
record, a timekeeper, a referee and a ‘language watchdog’ to record
and/or prevent language switch.
Thirdly, inexperienced young learners will need fundamental gui-
dance about techniques of negotiation and cooperation within the
group.
Fourthly, the teacher should make sure that inflexible group struc-
tures based on cliques do not develop. It must be ensured that group
membership changes over time, and this may require teacher inter-
vention. Personality clashes within groups should also be avoided if
possible (Dam 1994: 511).
In short, in Cooperative Language Learning:
The emphasis is on social interaction, the exchange of information

and the social benefits of discussion.
Learners find out about one another’s opinions, points of view,
personal histories, likes and dislikes.
Where the teacher has some freedom to determine the syllabus or
course content, then learner interests and areas of subject compe-
tence can be built in to the teaching schedule.
Where factual and linguistic knowledge are involved, some learners
may be factually more knowledgeable and others better at the
language, so that interaction results in mutual benefits.
See Richards and Rodgers (2014: 244–258) and Larsen-Freeman and

Anderson (2011: 186–190).
Self-Directed Language Learning (SDLL)

Self-Directed Language Learning is a further development of CLL and is
not to be confused with self-study or working in isolation. The ultimate
aim is learner autonomy, but the autonomous learner is a social learner.
The emphasis in SDLL is therefore on social learning under the guidance
of the teacher, who provides necessary support, which is gradually with-
drawn over time. This shift in emphasis towards learner self-direction is
sometimes referred to as learner empowerment. The aim is to produce
learners who can continue to learn on their own after formal language
education has finished. For this to occur, individuals must, under the
guidance of the teacher, develop increasing awareness of how they learn
best, what their preferred learning styles are and how they can apply
particular learning strategies in certain situations to enhance their learn-
ing. They should, however, ‘learn how to learn’ English implicitly
through the learning process itself rather than through explicit instruc-
tion (see Self-regulation, Autonomy, Motivation and Identity in Chapter 1).
It is advisable to introduce a measure of self-direction in the early
stages of language learning since it is extremely difficult to wean advanced
learners away from dependence on their teacher after years of teacher-
directed learning (Robinson 1981). Much depends on giving learners the
right amount of choice at the right time. There is a scale of choice types
through which learners should progress:
At the lowest levels learners can be offered a restricted number of

choices about which task to do or how to organise a task, for
example whether to do a specific task as group work or in pairs or
individually (see Pinter 2017: 111–126).
At a slightly higher level learners can also decide how tasks should
be divided up among the class and whether certain tasks should be
done in a particular order, or done in class or as homework.
At still-higher levels of development learners will be able to make
informed choices about materials and techniques.
Very advanced and experienced learners may be able to make
choices about teaching and learning methods and approaches.
The very highest level of choice concerns the negotiation of the
course content with the class (the negotiated syllabus). This is
unlikely to happen in school courses, but is a possibility in English
for Specific Purposes (ESP) courses for specific professional groups
with special needs (see Savage and Storer 2001).
Part of learner autonomy consists of the ability to assess one’s own

progress. This can be developed through the introduction of peer assess-
ment and self-assessment. These are not intended to necessarily replace,
but rather to supplement teacher assessment. In assessing their own
spoken and written productions as well as those of their peers, learners
will gain insights which would not otherwise be available to them. Peer
assessment of written work can be introduced successfully in secondary
school (see Bärenfänger 2012).
There is also an important intercultural aspect to self-directed learning.
For this reason SDLL should never be imposed wholesale on learners
without considering how well the approach chimes in with their own
educational and cultural background. In cultures which place a high
value on teacher authority, rote learning, imitation and closely guided
practice, the facilitating teaching style of SDLL may have to be arrived at
in easy stages. Learners who view the teacher as a language guru, who
have been used to a hierarchical mode of teaching and/or have an
authority-oriented learning style may see their own role as essentially
passive, may not like working in groups and may prefer the more
traditional teacher-fronted style to which they are accustomed (see
Sonaiya 2002: 106; Smith 2003: 260). More generally, self-directed class-
rooms may at first sight seem rather chaotic, noise levels may be higher
and there will be much more movement of learners compared to the
traditional classroom. Some learners may prefer a more disciplined
classroom. Learners who prefer to work alone or to consult the teacher
can be offered these options within the framework of choices available,
at least on occasion.
One way of getting learners started on the road to self-direction and
autonomy is to gradually introduce an element of peer presentation.
This involves learners, either singly, in pairs or in small groups,
preparing short presentations which they deliver to the class on some
topic they are especially interested in and possess expertise or special
knowledge about – perhaps an unusual hobby, or a period spent living
abroad. Adult learners can teach the class a little bit about their
profession or job. Another possibility is to introduce an element of
peer teaching. To start with, one or more learners can take over the
teacher’s role for a small part of a lesson. Learners can also be
encouraged, where appropriate, to bring in their own materials
(Worgan 2010).
Finally, for the teacher, too, things may not be easy. Some teachers
may find it difficult to adapt to the role of facilitator in SDLL. One
problem is that, as part of learning to become autonomous, learners will
make mistakes. It can be very difficult for the teacher to accept ‘wrong
choices’ which learners may make concerning techniques or materials.
Learner choices must be accepted and should not be overruled by the
teacher. Particularly in the early stages, the teacher should ensure that
the options available are all acceptable. For example, if learners are to
choose one of four pop songs to work on, then these must all be
didactically suitable.
In short, in SDLL:
The ultimate goal is complete learner autonomy.

The emphasis is on developing the ability to self-direct learning in
social contexts.
The teacher supplies support which is progressively withdrawn as
learners become more autonomous.
Learners first make limited choices concerning task sequencing, then
select materials and methods and ultimately decide on learning goals
and course design.
Peer teaching, peer presentations, peer assessment and self-assessment
may be used.
Authentic materials are often used and selected afresh for each
learner group and task.
See Richards and Rodgers (2014: 331–334), Pinter (2017: 111–126),

Benson (2011) and Dam (1994).
Competence-Based Language Teaching (CBLT)

Competence-Based Language Teaching (CBLT) defines language profi-
ciency in terms of what the learner can do with the language; that is to
say, as skill mastery rather than formal mastery. CBLT assumes that
language learning can be regarded as the development of a set of inter-
related skills, which range from simple to complex, with simpler skills
nesting within other more complex skills. More complex skills will build
on simpler skills in certain structured ways. Skills will cluster together to
form modules, and skills within a particular module will be closely
linked, and there will also be links between modules which share certain
sub-skills. Language proficiency in this view constitutes mastery of
a network of interrelated elements which can be recombined afresh to
meet new communicative needs. For example, writing e-mails combines
various elements of letter-writing skills with elements of the spoken
mode as well as some elements specific to e-mails. This results in the
specific language or register of e-mail writing.
According to Halliday (1978: 33–35), register is determined by the
‘context of situation’, which in turn comprises three components,
namely ‘field’, ‘tenor’ and ‘mode’:
‘Field’ refers to the physical and psychological setting of the

communication.
‘Tenor’ refers to the participants and the role relations between or
among them.
‘Mode’ refers to whether speech or writing is involved and more
detailed choices within the spoken or written channels.
(adapted from Halliday 1978: 33)
Concerning mode, nowadays one must also consider various electro-

nic forms, such as e-mailing, chatting, text messaging and Skyping, which
straddle the traditional divide between speech and writing to form
a ‘mode continuum’ (Gibbons 2015). ‘Field’, ‘tenor’ and ‘mode’, sub-
sumed under the term ‘context of situation’, will define the appropriate
register for any particular act of language communication.
In CBLT the content of the course becomes largely identical with the
skill-based proficiency goals set. In other words, in the classroom learners
work directly on the skills which they are aiming to master. The relevant
skill goals might include reading the headlines of a newspaper,
understanding written assembly instructions, tactfully declining an invita-
tion, writing a letter of recommendation or giving an after-dinner speech,
depending on the course and the needs of its learners. In the case of
English for Specific Purposes (ESP) courses, for example for computer
technicians, librarians or police officers, it may be possible to define goals
quite precisely by surveying learners in advance about the language skills
they will require. This is referred to as ‘needs analysis’ (Munby 1978;
Huhta et al. 2013). More will be said about needs analysis in Chapter 4.
In many school classes, however, it is not possible to predict learners’
future English language needs precisely, so that goals can only be
formulated in general terms. This situation is sometimes termed
TENOR (Teaching English for No Obvious Reason). Even in ESP
courses, needs analysis can sometimes be tricky. It is generally easier to
define the required skills for lower-level technical staff than for higher-
status professionals such as engineers, managerial staff and administra-
tors, who may need to indulge in a wide variety of linguistic activities
outside the narrow confines of their specialist knowledge, including
encouraging personnel, disciplining staff, interacting socially with collea-
gues and entertaining customers or visitors.
CBLT is an output-based, rather than an input-based, approach to
language teaching. It is the skills learners will hopefully possess at the end
of the course (output) which are set out in advance in the syllabus, not the
language content to be taught (input). The focus on output skills in CBLT
means that the syllabus has to be so designed that progress is made in
modular fashion from less complex to more complex communicative tasks,
but also from core to peripheral tasks in terms of the needs of the specific
group of learners at hand. The focus on output skills also influences how
learners are assessed. Continuous or ongoing assessment is often employed
in CBLT, with both success in performing the task, as well as the language
used, being assessed. CBLT is closely associated with the Common Eur-
opean Framework of Reference (CEFR) developed by the Council of
Europe. The CEFR sets competence-based, benchmark standards of profi-
ciency in languages (see Richards and Rodgers 2014: 165–168).
CBLT has a good track record and has been very successful in many
ESP programmes all over the world. It has also been successfully
employed in language programmes for adult immigrants, for example
the Australian Migrant Education Programme (Richards and Rodgers
2014: 160). Its adherents claim that learners respond positively to learn-
ing practical communication skills, which they perceive as relevant for
their future lives. Further, learners know at each stage in the course
exactly what they have achieved and can chart their own progress. In this
regard the use of portfolios in which learners record in ongoing fashion
the language skills they have acquired fits in well with the modular
approach and with continuous assessment.
The implementation of CBLT is, however, not without its problems,
especially in TENOR situations. Firstly, there is no one-to-one match
between structures and functions in language. Rather, it is in the nature
of language that structures have multiple functions, and different struc-
tures can often be used to express the same speech act but with
differences in register, style or politeness. Thus, it is difficult to define
a bundle of structures which learners must learn in order to perform any
given task. A second, related, problem is specifying the level of complex-
ity required of learners in performing a given task. This problem
becomes less critical if a cyclical syllabus, rather than a traditional linear
syllabus, is adopted. In a cyclical syllabus learners return to functions
such as apologising, refusing and objecting at increasingly sophisticated
levels of elaboration during the course of their language-learning careers.
A third and more fundamental criticism of CBLT is that no definitive list
of functional competences exists and it is unlikely that anyone could
produce such a list. The most serious and most fundamental criticism of
all, however, is that it is doubtful whether complex linguistic skills can
legitimately be approached in such a reductionist fashion, there being no
research-based evidence that the sum of the constituent tasks actually
does equal the complexity of the whole task.
To summarise, in CBLT:
• The goal is skill mastery.

• The course goals are made explicit in advance and are identical with
the syllabus.
• Mastery of the course goals can be demonstrated by performance.
• The emphasis is on individualised, student-centred instruction,
• Teaching is task-oriented.
• There is a focus on life-skills for successful functioning in society.
• The course is divided into modules to reflect the tenet that simple
skills nest within more complex skills.
• Continuous assessment is often used.
(See Richards and Rodgers,2014: 150–173)
The Post-Methods Era

The search for the ideal, universal and complete language-teaching
method has now been largely abandoned. Many successful teachers
today practise a principled eclecticism in that they combine elements
from various methods, without committing themselves to any single
method in its entirety. They recognise that the idea of a single universal
method is a pipe dream and that all methods have to be adapted to the
specific group of learners in the teaching situation at hand. Furthermore,
not only learners but also their teachers have their own preferences,
talents and inclinations, which also need to be considered. In order to
practise a principled eclecticism, each teacher needs to develop a theory
of what language is, how languages work, how they are acquired natur-
alistically and how they are learned in classrooms. Only then can
principled decisions be made as to what broad teaching approaches
should be adopted. This is infinitely preferable to just mixing and
matching techniques and activities from various methods on a rather
unprincipled basis.
This brief historical review of methodology has shown that up to
the 1970s classroom language teaching tended to focus on grammar
and vocabulary and to ignore pragmatics, especially the pragmatics of
conversation, which were increasingly in demand by learners. As these
deficits became apparent, there was a paradigm shift in language-
teaching theory from the late 1970s onwards. The new approach
became known as Communicative Language Teaching (CLT). Further
developments in methodology within the new paradigm included
Task-Based Language Teaching (TBLT) and Cooperative Language
Learning (CLL). The attempt was made within these approaches to
model classroom language learning on naturalistic acquisition and to
move way from teaching languages like any other academic subject.
This in turn involved downplaying formal, cognitively based teaching,
especially explanations of grammar and blanket correction of errors.
The emphasis was placed on interaction, conversation and language
use rather than on explicitly informing learners about the system of
the language.
However, at the level of classroom practice, the changes in methodol-
ogy described in this chapter found their way into classrooms only
gradually so that there was always a discrepancy between the state of
the art in ELT methodology and what actually went on in classrooms.
New methods were not necessarily applied universally, immediately or in
undiluted form and old methods were not abandoned overnight. Today,
within one institution, different teachers ostensibly teaching to the same
syllabus and using the same materials often teach in quite different ways.
Even course books that espouse a particular approach may be used by
teachers, whether deliberately or unknowingly, quite differently from the
way they were intended to be used. This is not surprising since teachers
develop their own preferred teaching styles and find they are better
suited to teaching within some approaches than within others. Trainee
teachers remain in many ways surprisingly impervious to the methodo-
logical input they receive during their professional training; later, as
practising teachers their particular brand of teaching often draws on the
way they themselves were taught at school and on what they have picked
up informally from colleagues as much as on what they learned about
teaching methodology at university.
Methodology theorists were often over-enthusiastic about new methods
which subsequently failed to live up to initial expectations. Furthermore,
there was a tendency for ELT methodology to develop in zig-zag fashion,
with each new method representing an overcorrection of the previous one.
In this respect, the unwillingness of teachers to immediately jump on every
new methodological bandwagon may well have shielded language learners
from some of the worst excrescences of methodology theory. The grass-
roots language teacher’s tendency towards compromise and eclecticism has
actually proved to be in tune with the more differentiated picture of
language learning which has emerged more recently. The teaching method
employed may well be less important than interesting content, the presence
of intrinsic learner motivation and a good working relationship between
teacher and learners, as well as a mutual conviction of the efficacy of what
they are doing.
The history of language-teaching methodology suggests that foreign
languages are, ironically enough, often best learned incidentally, when the
focus is not on the language as something to be learned but as
a communicative tool. A special case of such incidental language learning
is Content and Language Integrated Learning (CLIL), in which learners
study an academic subject such as history, economics, geography or
biology through the medium of a foreign language such as English. CLIL
falls outside the purview of this survey of language-teaching methodology
since it requires teachers trained in both the foreign language and the
academic subject in question and is not a methodological option for most
mainstream language teachers. For more on CLIL, which takes various
forms according to the educational system in which it operates in different
parts of the world, see Coyle et al. (2010) and Baker and Wright (2017) in
the ‘Further Reading’ section.
Food for Thought

Q. 1. There is some controversy about whether languages can be taught
in much the same way as other academic school subjects or not.
What do you think?
Q. 2. Do you see the role of the language teacher mainly as an explainer
of the language, a demonstrator of the language, or as a facilitator
who provides opportunities for practice in using the language
(learning by doing)? Do you think there are other major or minor
roles for the language teacher to play?
Q. 3. Can you match some of the language-teaching methods mentioned
in this chapter to the role views of the teacher mentioned in
Question 2?
Q. 4. Some people hold that the personality of the teacher is more
important than the teaching method. What do you think and why?
Which of the methods presented in the chapter are most dependent
on the personality of the teacher and which are least dependent on
the teacher’s personality? Can the teacher’s personality transcend
the method adopted?
Q. 5. The search for the single best language-teaching method has now
been largely abandoned. One size does not fit everybody and there
is more interest nowadays in matching teaching to learners. Take
some of the methods discussed in the chapter and consider what
sort of learners might be most suited or least suited to them.
Consider factors such as age, socio-cultural and geographical back-
ground, motivation, goals and proficiency level.
Further Reading
Coyle, D., Hood, P. and Marsh, D. (2010). CLIL: Content and Language Integrated
Learning. Cambridge: Cambridge University Press.
Long, M. (2015). Second Language Acquisition and Task-Based Language Teaching.
Malden, MA: Wiley Blackwell.
References
Asher, J. J. (1969). The Total Physical Response approach to second language
learning. The Modern Language Journal 53/1: 3–17.
Asher, J. J. (2009). Learning Another Language through Actions. Expanded seventh
edition. Los Gatos, CA: Sky Oaks Productions. (First edition 1977).
Austin, J. L. (1962). How to Do Things with Words. Oxford: Clarendon Press.
Bärenfänger, M. (2012). Peer correction in the English Language writing classroom:
an empirical study in the ninth grade of a Realschule. In Lennon, P. (ed.), Learner
Autonomy in the Language Classroom: Empirical Studies and Ideas for Teachers.
Frankfurt am Main: Peter Lang, pp. 147–175.
Benson, P. (2011). Teaching and Researching Autonomy. Second edition. Harlow:
Pearson Education.
Bloomfield, L. (1933). Language. New York: Holt.
Brown, H. D. (2014). Principles of Language Learning and Teaching: A Course in
Second Language Acquisition. Sixth edition. White Plains, NY: Pearson
Education.
Cook, V. (2001). Second Language Learning and Language Teaching. Third edition.
London: Hodder Arnold.
Corder, S. P. (1967). The significance of learners’ errors. International Review of
Curran, C. A. (1972). Counselling-Learning: A Whole-Person Model for Education.
New York: Grune and Stratton.
Curran, C. A. (1976). Counselling-Learning in Second Languages. Apple River, IL:
Apple River Press.
Dam, L. (1994). How do we recognise an autonomous classroom? Die Neueren
Sprachen 93/5: 503–527.
East, M. (2012). Task-Based Language Teaching from the Teachers’ Perspective.
Amsterdam and Philadelphia: John Benjamins.
Ellis, R. (2003). Task-Based Language Learning and Teaching. Oxford: Oxford
University Press.
Firth, J. R. (1950). Personality and language in society. The Sociological Review xlii/
1. (Reprinted in Firth J. R. 1957, Papers in Linguistics 1934–1951. London:
Oxford University Press, pp. 37–52).
Firth, J. R. (1951). Modes of meaning. Essays and Studies of the English Association
new series 4: 123–149. (Reprinted in Firth, J. R. 1957, Papers in Linguistics
1934–1951. London: Oxford University Press, pp. 190–215).
Fries, C. C. (1945). Teaching and Learning English as a Foreign Language. Ann
Arbor, MI: University of Michigan Press.
Frisby, A. W. (1957). Teaching English: Notes and Comments on Teaching English
Overseas. London: Longmans, Green and Co.
NH: Heinemann.
Halliday, M. A. K. (1973). Explorations in the Functions of Language. London:
Edward Arnold.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of
Language and Meaning. London: Edward Arnold.
Harmer, J. (2015). The Practice of English Language Teaching. Fifth edition. Harlow:
Pearson Education.
Howatt, A. P. R. (1984). A History of English Language Teaching. Oxford: Oxford
University Press.
Howatt, A. P. R. with Widdowson, H. G. (2004). A History of English Language
Teaching. Second edition. Oxford: Oxford University Press.
Huhta, M., Vogt, K., Johnson, E. and Tulkki, H. (2013). Needs Analysis for
Language Course Design: A Holistic Approach to ESP. Cambridge: Cambridge
University Press.
Hymes, D. H. (1972). On communicative competence. In Pride, J. B. and
Holmes, J. (eds.), Sociolinguistics. Harmondsworth: Penguin, pp. 269–293.
Jin, L. and Cortazzi, M. (2011). Re-evaluating traditional approaches to second
language learning and teaching. In Hinkel, E. (ed.), Handbook of Research in Second
Language Teaching and Learning. Volume 2. New York and London: Routledge,
pp. 558–575.
Revised third edition. London and New York: Routledge.
Johnson, K. and Johnson, H. (1998). Communicative methodology. In Johnson, K.
and Johnson, H. (eds.), Encyclopaedic Dictionary of Applied Linguistics. Oxford:
Blackwell, pp. 68–73.
Lado, R. (1964). Language Teaching: A Scientific Approach. New York: McGraw-Hill.
Larsen-Freeman, D. and Anderson, M. (2011). Techniques and Principles in Lan-
guage Teaching. Third edition. Oxford and New York: Oxford University Press.
Legutke, M. and Thomas, H. (1991). Process and Experience in the Language Class-
room. London: Longman.
Littlewood, W. (2011). Communicative language teaching: an expanding concept
for a changing world. In Hinkel, E. (ed.), Handbook of Research in Second Language
Teaching and Learning. New York and London: Routledge, pp. 541–557.
Munby, J. (1978). Communicative Syllabus Design. Cambridge: Cambridge Univer-
sity Press.
Mystkowska-Wiertelak, A. and Pawlak, M. (2012). Production-Oriented and Com-
prehension-Based Grammar Teaching in the Foreign Language Classroom. Heidel-
berg: Springer.
Nunan, D. (2004). Task-Based Language Teaching. Cambridge: Cambridge Univer-
sity Press.
Pinter, A. (2017). Teaching Young Language Learners. Second edition. Oxford:
Oxford University Press.
Pittman, G. A. (1967). Teaching Structural English. Revised edition. London: Ginn
and Company Ltd. (First published 1963 by the Jacaranda Press, Brisbane).
Richards, J. C. and Rodgers, T. S. (2014). Approaches and Methods in Language
Teaching. Third edition. Cambridge: Cambridge University Press.
Robinson, P. (1981). Language learning at the advanced level. English Language
Teaching Journal 35/2: 101–105.
Rogers, C. R. (1970). Carl Rogers on Encounter Groups. New York: Harper and Row.
Rousseau, J.-J. (1762/2009). Émile Ou De L’education. Paris: Flammation.
Savage, W. and Storer, G. (2001). An emergent language programme framework:
actively involving learners in needs analysis. In Hall, D. and Hewings, A. (eds.),
Innovation in Language Teaching. New York: Routledge, pp. 137–148.
Scrivener, J. (2011). Learning Teaching: The Essential Guide to English Language
Teaching. London: Macmillan.
Searle, J. R. (1969). Speech Acts: An Essay in the Philosophy of Language. Cambridge:
Smith, R. C. (2003). Pedagogy for autonomy as (becoming-) appropriate methodol-
ogy. In Palfreyman, D. and Smith, R. C. (eds.), Learner Autonomy across Cultures:
Language Education Perspectives. Basingstoke: Palgrave Macmillan, pp. 129–146.
Sonaiya, R. (2002). Autonomous language learning in Africa: a mismatch of
cultural assumptions. Language, Culture and Curriculum 15: 106–116.
Ur, P. (2012). A Course in Language Teaching. Cambridge: Cambridge University
Press.
Vygotsky, L. (1978). Mind in Society: The Development of Higher Psychological
Processes. Cambridge, MA: Harvard University Press.
Vygotsky, L. (1986). Thought and Language. Newly revised and edited by Alex
Kozulin. Cambridge, MA: M.I.T. Press. (First published in Russian 1934).
Wilkins, D. A. (1976). Notional Syllabuses: A Taxonomy and Its Relevance to Foreign
Language Curriculum Development. Oxford: Oxford University Press.
Willis, D. and Willis, J. (2007). Doing Task-Based Teaching. Oxford: Oxford
University Press.
Worgan, M. (2010). Promoting learner autonomy. Modern English Teacher 19/3:
25–28.
3 Individual Learner Differences
Aptitude for Language Learning
The MLAT
Some people are better than others at learning foreign languages. They
learn more quickly and apparently more easily than their peers. Like
intelligence, language-learning aptitude is taken to be a fixed internal
factor which cannot be developed but remains constant throughout life
for an individual relative to others of the same age. Special tests have
been developed which measure how well-suited to classroom language
learning an individual is. The most famous one is the Modern Language
Aptitude Test (MLAT), developed at Harvard University during a five-
year research project with 5,000 participants. It seeks to measure ‘how
well, relative to other individuals, an individual can learn a foreign
language in a given amount of time and under given conditions’ (Carroll
and Sapon 1959). The MLAT assumes that language-learning aptitude
consists of four components:
1. phonetic coding ability: an ability to identify distinct sounds, to

form associations between those sounds and symbols representing
them, and to retain these associations;
2. grammatical sensitivity: the ability to recognise the grammatical
functions of words (or other linguistic entities) in sentence
structures;
3. rote-learning ability for foreign language materials: the ability to
learn associations between sounds and meanings rapidly and effi-
ciently, and to retain these associations; and
4. inductive language-learning ability: the ability to infer or induce the
rules governing a set of language materials, given samples of language
materials that permit such inferences.
(Carroll 1981: 105)
The MLAT is administered by the Second Language Testing Founda-
tion Inc. (SLTF), a non-profit-making organisation which charges institu-
tions for using the test. Example questions are supplied online by SLTF
(http://lltf.net/mlat-sample-items). Versions of the MLAT are also avail-
able in languages other than English. It is important to bear in mind that
the MLAT is not testing ability in English or whatever language it is set
in, but ability to learn other languages in general.
In its standard form the MLAT requires 60–70 minutes of test time
and consists of a battery of five tests, as follows:
Part One (43 items), ‘Number Learning’, aims to test inductive

ability. Sample numbers from an invented language are presented
orally to learners, who then have to recognise novel numbers.
Example: The test takers hear ‘ba’ is one, ‘baba’ is two, ‘dee’ is
three, ‘tu’ is 20, ‘ti’ is 30, ‘tu-ba’ is 21, and so on, and have to
recognise novel numbers such as ‘ti-ba’, ‘ti-dee’, ‘tu-dee’.
Part Two (30 items), ‘Phonetic Script’, aims to test phonetic coding
ability. Test takers see five sets of four syllables written down and
spelt ‘phonetically’, and then hear the syllables pronounced.
Example: bot but bok buk
bok buk bov bof
geet gut beet but
beek bev but buv
geeb geet buf but
The task involves then hearing just one of the four syllables from
each set and matching it to its written ‘phonetic’ representation.
Part Three (50 items), ‘Spelling Clues’, aims to test phonetic coding
ability and inductive ability. However, this test part is done under
time pressure, so success also depends on ability to work quickly
and on knowledge of English vocabulary (or the vocabulary of
whichever language is used for the test). Test takers see a word
which is inaccurately spelt and they have to match it to one of four
(correctly spelt) words, three of which are distractors and one of
which is a near synonym of the source word.
Example: PREZNS (stimulus word).
Instruction: Match with one of the following:
‘kings’ ‘explanations’ ‘dates’ ‘gifts’ ‘forecasts’
Part Four (45 items), ‘Words in Sentences’, aims to test grammatical
sensitivity – without using grammatical terminology – and inductive
ability. Test takers are shown a model sentence with a word capita-
lised and a test sentence with five words underlined. They have to
identify the word which performs the same grammatical function as
the capitalised word in the model sentence.
Individual Learner Differences 61
Example: MARY is happy (model sentence).
From the look on your face, I can tell that you must
have had a bad day.
Part Five (24 items), ‘Paired Associates’, aims to test rote memory for
vocabulary. An element of working under time pressure is also
involved. Learners are given a set of 24 words from an unknown or
invented language and their English translation equivalents.
Example: Maya English
cɁon gun
siɁ wood
kɁab hand
kab juice
bat axe
(19 more words and translation equivalents)
Test takers are given two minutes to memorise the list. The vocabu-
lary is then tested by means of multiple-choice. Test takers have to
match a word to its English translation, with one correct solution
and three distractors being offered for choice.
Example: bat A. animal
B. stick
C. jump
D. axe
E. stone
There is also an elementary version of the test available, the MLAT-E,
which is intended for young learners aged roughly 9 to 12 years. It differs
from the standard MLAT in that it has only four parts. Part Five, the
‘Paired Associates’ test for rote memory has been dropped for some
reason. Sample questions from the MLAT-E are available online from
SLTF (http://lltf.net/wp-content/uploads/2011/12/MLAT-E-Sample-
Items.pdf).
The MLAT is more effective in identifying those few individuals who
will find it very difficult to learn languages than in spotting those few
highly gifted individuals at the other end of the scale. It has been widely
used in the United States and Canada by employers in the private and
public sectors, including the armed services, the civil service and the
diplomatic service, to select suitable learners for language courses and to
weed out unsuitable candidates from the start. The MLAT is also used
by some educational institutions for admission to degree courses in
foreign languages. Thanks to its modular structure, the MLAT produces
differentiated aptitude profiles. For this reason it is sometimes used
diagnostically to tailor language courses to the specific strengths and
weaknesses of a particular group of learners.
There is a relationship between how people score on language
aptitude tests and how they score on intelligence tests. However, the
MLAT is a better predictor of language-learning aptitude than perfor-
mance on intelligence tests, although three of the four components
tested in the MLAT, namely grammatical sensitivity, inductive ability
and rote-memory ability seem to be quite closely related to general
intelligence. The fourth component, namely phonetic coding ability,
seems not to be closely related to intelligence. This in turn suggests
that phonological aptitude may be distinct from lexical and gramma-
tical aptitude. In other words, foreign accent is not necessarily a good
guide to a learner’s overall proficiency (as was mentioned in Chap-
ter 1).
Those who have a good musical ear may be at an advantage in
acquiring the phonology of a foreign language after the critical or
sensitive period for nativelike acquisition has closed off. A study by
Sleve and Miyake (2006) found that musical ability (perception and
production of note and pitch changes) predicted phonological ability in
English as a foreign language (production and perception of phonemes).
However, musical ability did not predict grammatical or lexical compe-
tence. The subjects were 50 Japanese native speakers who had all come to
the USA after the age of 11 years and lived there for at least six months.
Half of them were students and the others were either employed locally
or were the spouses of people who were working or studying in the
USA. Most of them had first learned their English in school before
coming to the USA.
The MLAT does not seek to predict ultimate attainment but rather
how quickly an individual will learn a language in the classroom,
particularly in the initial stages. Ultimate attainment will be influenced
by additional internal factors such as interest in learning the language,
attitudes to the language and its speakers and both intensity of
motivation and the specific nature of the learner’s motivation. A host
of external factors, including the quality of instruction received and
opportunities to interact with speakers of the language outside the
classroom will also be influential. Carroll (1973) expressed the relation-
ship between aptitude and other factors influencing success in language
learning as follows:
In brief … if aptitude is reflected in how much time an individual

needs to learn something under optimal conditions of motivation,
opportunity to learn and quality of instruction, the role of aptitude
can be either increased or decreased when these other conditions are
varied.
(Carroll 1973: 6)
In the teaching environments for which it was developed, before the

advent of CLT, the MLAT generally produced correlations of 0.4 and
0.6 between aptitude and subsequent achievement tests in a foreign
language (Carroll 1981: 93, 96). This means that between 16% and 36%
of the variation in language test scores in a sample of learners is
accounted for by variation among learners’ MLAT score (see Carroll
1981: 96–97 for further discussion of how to interpret the correlations).
However, Ehrman and Oxford (1995) still obtained correlations of 0.51
for the MLAT with end-of-course proficiency ratings in reading and in
speaking for learners taught according to communicative methodology.
The subjects were 352 highly educated adult learners of various lan-
guages at the Foreign Service Institute of the U.S. Department of State.
The authors conclude that ‘the MLAT may be addressing learning
abilities that are independent of methodology’ (Ehrman and Oxford
1995: 78). Aptitude was a better predictor of success than all the other
factors included in the study: learning strategies, learning styles, person-
ality, motivation and anxiety.
For further discussion of the MLAT, see Johnson (2018: 110–117).
The LAB
Another successful and widely used aptitude test is the Language Apti-
tude Battery (LAB), developed by Paul Pimsleur and designed for use in
schools from the end of grade 6 to grade 12 (Pimsleur 1966). The battery,
which takes 50–60 minutes to administer, consists of six parts, as
follows:
Part One, ‘Grade Point Average’ asks test takers to report their
scores in their other academic subjects.
Part Two, ‘Interest in Foreign Language Learning’, aims to measure
the extent of learners’ self-assessed interest in learning the foreign
language.
Part Three, ‘Vocabulary as an Indicator of Verbal Intelligence’ is
a multiple-choice English vocabulary test based on matching one of
four words, for example ‘weak’, ‘sickly’, ‘strong’, ‘vigilant’ to
a source word, for example ‘vigorous’.
Part Four, ‘Language Analysis’, seeks to test ‘ability to reason
logically in terms of foreign language materials’. It uses an unknown
or invented language, and on the basis of example sentences learners
have to work out the grammar and answer multiple-choice questions
in which they choose the correct translation of an English sentence
into the unknown language from four possibilities.
Part Five, ‘Sound Discrimination’, tests the ‘ability to learn phonetic
distinctions and to recognise them in different contexts’.
Part Six (‘Sound/Symbol Association’) tests the ‘ability to associate
a sound with its written symbol’.
The LAB thus differs most markedly from the MLAT in its first two
parts, which have no parallels in the MLAT. Concerning the remaining
four parts, the LAB focuses more on auditory factors and less on
memory factors (which the MLAT-E also downplays).
Concerning ‘Grade Point Average’, there is indeed some evidence that
a general academic intelligence factor as well as academic study skills may
strongly influence foreign-language learning success in traditional class-
rooms. In a study by Saffron (1988), for example, it was found that for
secondary school children learning French at intermediate level in the
English Midlands the best predictors of achievement on the end-of-year
written exam were scores on a verbal reasoning test (in English) and exam
results in other subjects, with history being the strongest predictor
among school subjects. It may indeed well be the case that in formal
classroom settings successful learners are just using the same cognitive
abilities they successfully employ for other school subjects.
Concerning ‘Interest in Foreign Language Learning’, it is not surprising
that this variable should improve the predictive power of the LAB.
However, there are good reasons for excluding this variable from mea-
sures of aptitude and including it rather as an attitudinal measure. Unlike
intelligence and aptitude, interest in learning the language can and does
change over time and can be influenced by the teacher, other persons and
external circumstances. It is clearly not a fixed, internal factor.
The CANAL-F Test

The CANAL-F test (Cognitive Ability for Novelty in Acquisition of Lan-
guage – Foreign) (Grigorenko et al. 2000) was constructed in the light of
developments in communicative language teaching and language-acquisition
theory (see Dörnyei 2005: 50–53; Skehan 2012: 389–390). It uses an artificially
constructed language called Ursula as the basis for testing ability to acquire
vocabulary, comprehend extended texts, infer grammatical rules from text
and also make semantic inferences at the clause, sentence and discourse
levels. The underlying idea is that it is the ability to cope with novelty and
ambiguity which is decisive in determining language-learning aptitude. The
test assumes that there are five knowledge acquisition processes:
selective encoding (focussing on relevant information for the task)

accidental encoding (understanding background contextual information)
selective comparison (relating already learned information to the new
task)
selective transfer (applying already learned rules or patterns to the
new task)
selective combination (synthesising disparate items of learned
information)
Aptitude for Language Learning and First Language Acquisition
Carroll (1973) speculated that individuals may differ in the rate and
extent to which the ability to acquire a first language in childhood
subsequently declines after puberty and that these differences in fact
constitute differences in language-learning aptitude. There is some
empirical support for this idea. Abrahamsson and Hyltenstam (2008),
for example, studied over 100 immigrants to Sweden. Their first language
was Spanish and they had achieved virtually native-speaker levels in
Swedish as judged by Swedes. They were given an aptitude test and it
was found that for those who arrived in Sweden before the age of 12,
aptitude scores varied across the spectrum, while those who had arrived
after the age of 12 all scored very highly on the aptitude test. These
findings fit in with the hypothesis that early immigration virtually
guarantees nativelike acquisition of the new language while for late
immigrants (after puberty) high language-learning aptitude is required to
achieve nativelike or near-native proficiency. (Compare this finding with
the results of Patkowski’s 1980 study of Cuban immigrants to the USA,
discussed in Chapter 1.)
However, an alternative view is that foreign-language learning aptitude
reflects differences in first language acquisition aptitude. For although
first language acquisition is uniformly successful, there are differences in
the rate (speed) at which children acquire their mother tongue (see
Chapter 1). Evidence for a link between first language acquisition and
foreign-language learning aptitude is forthcoming from a study of 53
children whose first language development was closely monitored in
infancy (aged 39–60 months) as part of the Bristol Language Project
(Wells 1985). When these children were aged about 15 years, they were
tested for foreign-language learning aptitude. Some significant correla-
tions between certain aspects of first language rate of development and
foreign-language aptitude scores were found. These correlations were
sometimes as high as 0.4 to 0.5, which means the variations in first
language rate of progress accounted for 16–25% of the variation in
aptitude scores (Skehan 1986: 101). (See also Skehan 1989: 31–34,
2012: 388.)
Motivation
Goal-Driven Behaviour
Traditional psychological approaches to human motivation held that
human beings have ‘drives’ which move them to achieve ‘goals’ in order
to satisfy ‘needs’. Maslow (1970: 35–58) suggested there is actually a five-
tier hierarchy of needs. Lower-level needs are physiological in nature and
higher-level needs psychological in nature. Lower-level needs generally
have to be satisfied first and higher-level needs tend only to emerge when
lower-level needs are already at least partly satisfied. At the lowest level
are basic physiological needs such as food and shelter. Then come safety
needs, such as freedom from fear and the need for structure and order.
Then there are social needs, which Maslow calls ‘belongingness and love
needs’ (Maslow op. cit.: 43). The penultimate level in the hierarchy is
occupied by the need for esteem, both in the sense of self-esteem and of
being held in esteem by others, and finally, at the apex of Maslow’s
hierarchy, stands the need for personal fulfilment, ‘doing one’s own
thing’, as we would say today. Maslow termed this ‘self-actualisation’.
Unlike lower-level motivation, which is ‘deficiency-driven’ (lack of food,
shelter, love, for example), self-actualising motivation is ‘growth-driven’
(idem: 162), the goal being to ‘grow towards full humanness’ (idem: 104).
Those who have self-actualising motivation tend to be autonomous,
‘dependent for their own development and continued growth on their
own potentialities and latent resources’ (idem: 162). It may even be
possible for individuals with well-established self-actualising motivation
to forego, at least to some extent, gratification of some lower-level needs
(idem: 103). An example might be burning the midnight oil and thus
foregoing sleep when one is engrossed in one’s work.
Motivation to learn English may operate at various levels in the needs
hierarchy. Survival English courses for immigrants may well address
lower-level needs. On the other hand, foreign-language learning motiva-
tion in schools would normally relate to esteem and self-actualisation
needs, which only emerge when lower-level needs are satisfied. Thus,
a child who is not getting a breakfast on cold winter mornings is unlikely
to be interested in the English lesson in the second period and more
likely to be longing for the bell for break to sound so that it can get
something to eat (a basic physiological need). Similarly, a child whose
parents are breaking up and violently quarrelling at home will probably
experience a lack of fulfilment of safety and social needs and be in no
state to concentrate on learning English. Again, an adult learner who has
just lost his or her job or who has just been evicted by the landlord will
also be more concerned with satisfying safety and shelter needs than in
learning English. However, for something as complex as language learn-
ing, motivation may be multifaceted. A child’s motivation for learning
English at school may derive partly from its wish to please its parents
(need for love), partly from its wish to be seen as successful (esteem) and
partly from enjoyment of achievement (self-actualisation).
Social-Psychological Approaches
Gardner (1985: 50) suggested that not only will motivated language
learners work hard towards achieving their goal but will also, if ques-
tioned, say that they have a desire to learn the language and have
favourable attitudes towards learning it. In questionnaire-based studies,
mainly of schoolchildren learning French in Canada and the USA,
Gardner and his associates found that positive attitudes to the foreign
language and its speakers as well as to foreign languages and foreigners in
general tend to go hand in hand with success in learning a foreign
language and can counterbalance deficits in language-learning aptitude
(see Gardner 2010 for a review). Two broad attitudinal orientations, an
‘integrative’ and an ‘instrumental’ orientation were identified. According
to Gardner and Lambert (1972: 3), an integrative orientation involves
‘wishing to learn more about the other cultural community … to the
point of eventually being accepted as a member of that other group’,
while an instrumental orientation is concerned with ‘the more utilitarian
value of linguistic achievement, such as getting ahead in one’s
occupation’.
These two orientations are neither mutually exclusive nor exhaustive
(Gardner 1985: 51–52). Nevertheless, it was found in a series of studies in
schools that an integrative orientation to the foreign-language commu-
nity, meaning an openness to the community, a desire to know more
about it, to emulate some of its characteristics and in some sense to
belong to it, did seem to be more conducive to high levels of motivation
than a purely instrumental orientation. This was true to varying degrees
for schoolchildren learning French in Montreal, in Louisiana, in Con-
necticut and in Maine. However, in a study conducted among school-
children learning English in the Philippines, a predominantly
instrumental orientation to the language coupled with strong parental
support could be identified and seemed to be effective (Gardner and
Lambert 1972). Gardner (1985: 133–134) draws attention to the element
of ‘emotional involvement with the other community’ in an integrative
orientation, which may be lacking in an instrumental orientation, and
adds that in the case of an integrative orientation, ‘with proficiency in the
language comes the possibility of changes in self-identity’ (idem: 134).
Nowadays, for many learners identification with a geographically
identifiable native-speaker target community has become less relevant,
and is often replaced by the international community of users of
English as a global language. It is, however, still possible to have
either an instrumental or an integrative orientation to this global
community. In other words, the important psychological element of
identification in integrative motivation can still be a motivating factor
for learners without direct access to a community of native speakers
of English. In this regard Dörnyei (2005: 93–108) suggests redefining
the idea of integrative orientation within the psychological framework
of ‘possible selves’, including the ‘ideal self’ (Markus and Nurius
1986). The learner’s ‘ideal L2 self’ may include being part of the
global community of English users (see also Ushioda 2006: 150;
Ushioda and Dörnyei 2012: 400–401, 404–405; Dörnyei and Ushioda
2009). Dewaele (2013: 629–631) cites an unpublished study by Young
(2007) in which open-mindedness and cultural empathy were appar-
ently found to be to be good predictors of language-learning
achievement.
In schools where there is little direct access to a community of native
speakers, language learners may combine instrumental and integrative
orientations. In their study of learners of English aged 16–17 years in
Hungarian schools Clément et al. (1994) isolated three factors which were
important for optimum motivation. These were:
an integrative orientation (which did not exclude recognition of

instrumental value)
self-confidence in using the language (associated with low anxiety
levels)
positive attitudes to the classroom learning environment (including
the teacher, the other learners, teaching materials and classroom
activities)
A similar picture emerges from a study by Chambers (1994) of children

aged 10–17 years learning German at a school in North Yorkshire.
Although motivation was low among the 10–11 year-olds, who were
just starting to learn German, those who had been learning the language
for a few years were much more enthusiastic. The reason they gave was
their positive experience with their teacher, whom they said was helpful,
made the subject fun and emphasised its usefulness. These learners
seemed to combine instrumental and integrative orientations. All age
groups defined a ‘good’ subject as being useful in the career market, as
being interesting, as having a good teacher and as yielding good marks in
tests. Unimportant in deciding whether a subject was ‘good’ or not, on
the other hand, was the amount of homework required and the potential
to ‘mess around’ in class. In contrast to the older learners, the
10–11 year-olds were not able to distinguish between subjects they
enjoyed and those they perceived as useful. For them the fun element
was very important. The older learners, however, were able to state that
they chose German for three reasons, namely enjoyment, career useful-
ness and the social possibilities it afforded them. The oldest group
(16–17 year-olds) were also starting to show interest in the German way
of life.
Cognitive Approaches
Social-psychological approaches to language-learning motivation may be
contrasted with cognitive approaches. Cognitive approaches are often
applied to the learning of academic subjects in general, not just foreign
languages, in educational settings. They maintain that individuals have
limited time and energy at their disposal and therefore have to make
choices about distributing time and energy across various goals. Learning
English is only one possible goal among many, including perhaps other
academic subjects, career goals and family and social commitments.
Learners’ levels of commitment to learning English will be influenced
by how important they think the goal is and their expectations of success
for the effort expended. This commitment may fluctuate over time as
initial expectations are either confirmed or disconfirmed by experience.
Perceived success relative to the learner’s own goals enhances motivation
levels and perceived failure lowers motivation levels. This results in the
virtuous circle of the good learner and the vicious circle of the poor
learner: good learners perceive themselves as succeeding, which boosts
their motivation; this helps them to go on succeeding, which boosts their
motivation still further. Poor learners perceive themselves as failing,
which lowers their motivation, which leads to greater failure, which
lowers their motivation still more (compare Nuttall 1996: 127; Johnson
2018: 120).
But what makes a goal more or less important? In this regard Deci and
Ryan (1985) distinguished between ‘intrinsic motivation’ and ‘extrinsic
motivation’. Intrinsic motivation is self-determined, while extrinsic moti-
vation is other-determined. Both may co-exist. For example, you may be
reading a novel both for enjoyment (intrinsic motivation) and also
because it is on the exam curriculum (extrinsic motivation). In this case
there is congruence of intrinsic and extrinsic motivation. But in a specific
learning situation there may be a mismatch between the intrinsic goals of
the learner and the extrinsic goals of the syllabus or the curriculum.
Thus, a learner’s intrinsic goal may be conversational ability in the
language, but the extrinsic goal of the course may be for him to be able
to translate simple texts and write business letters.
Intrinsic motivation may be positive (achieving personal satisfaction)
or negative (avoiding personal dissatisfaction) in orientation. Similarly,
extrinsic motivation may be positive (receiving rewards) or negative
(avoiding punishment) in orientation. Particularly in the case of young
learners it may not always be easy for the teacher to decide whether
motivation is extrinsic or intrinsic. Children may appear to be intrinsi-
cally motivated, in that they are enthusiastic in class and work hard at
home too. However, in reality they may be extrinsically motivated by
their parents. The extrinsic motivation may take the form of promises of
presents or monetary rewards if they get good marks (other-imposed
extrinsic motivation). More subtly, the children may simply want to
please their parents (self-imposed extrinsic motivation). This may work
well at primary school since up to a certain age children cannot necessa-
rily distinguish between their own goals and the goals which adults
whom they love or respect impose on them. However, unless they
come to enjoy learning English for its own sake (intrinsic motivation),
they will at some point in secondary school rebel against learning English
just because their parents want them to. Concerning material rewards
(other-imposed extrinsic motivation), the stakes need to be constantly
raised if the effect is not to wear off. Even teacher-imposed external
motivation in the form of incentives such as gold stars for good work,
which may seem to be effective in the short to medium-term, may be
ineffective in the long term, since they do not necessarily help to develop
intrinsic enjoyment of learning. Without this, at some time in the future
the incentives will no longer work.
Process Approaches
Motivation to learn a language is a dynamic process in a constant state of
flux (Ushioda 1996, 2001; Dörnyei 2001a, 2001b, 2005). It is helpful to
distinguish between initial or starting motivation and continuing motiva-
tion. The latter is needed for persistence and perseverance (Dörnyei
2005: 83). As Williams and Burden (1997: 121) put it:
It is important to emphasise here that motivation is more than

simply arousing interest. It also involves sustaining that interest and
investing time and energy into putting in the necessary effort to
achieve certain goals.
(Williams and Burden 1997: 121)
This distinction will apply at the micro-level to any given language-

learning task, but it also applies at the macro-level to an entire language
course and to the whole language-learning process itself, extending over
years, perhaps even a lifetime. It is initial motivation that drives indivi-
duals to overcome inertia and invest energy in a purposeful, goal-
oriented way, but the real problem for learners of English tends to be
maintaining motivation over time after the initial novelty has worn off.
Motivation is likely to change both in intensity (quantitative change)
and in nature (qualitative change) during a learner’s language-learning
career, as Chambers’ 1994 study, discussed above, indicated. Only if the
quality of motivation develops over the years, will levels of motivation
remain high or rise. In the early stages of learning a language children’s
motivation is most likely to be mainly intrinsic (enjoyment) but there
may be extrinsic elements such as wanting to please parents or the
teacher. Later, additional extrinsic motivators may operate, including
passing exams and getting on in one’s job. The balance between an
instrumental and integrative orientation may shift over time, too (see
Johnson 2018: 122).
The dynamic nature of motivation means that for classroom teach-
ing purposes it is helpful to distinguish among three types of motiva-
tion which nest one within the other. ‘Backdrop motivation’ is
a learner’s underlying type and intensity of motivation to learn the
language; ‘task motivation’ refers to how appealing a particular sort of
language-learning task or activity is to a particular learner; ‘state
motivation’ refers to the learner’s mental and emotional state at
a particular moment in time while performing a language-learning
task, and it may fluctuate during performance of the task. State
motivation is influenced by backdrop motivation, task motivation and
how the learner currently perceives the ongoing task or activity,
especially concerning its level of ‘perceived challenge’, an idea which
will be taken up later in discussion of anxiety.
Tasks which learners like doing for their own sake will be motivating
(intrinsic motivation), and feelings of enjoyment experienced while and
after doing the task will also have positive motivational effects for the
future. This means that state motivation feeds back into task and back-
drop motivation. Ongoing feedback from the learning environment,
including the reactions of fellow learners and the teacher, especially
teacher praise and censure, also influences state motivation and feeds
back into task and backdrop motivation. Task outcomes, whether
extrinsic (test results) or intrinsic (personal feelings of success or failure)
will likewise feed back into task and backdrop motivation, which will be
boosted or dampened accordingly.
Demotivation and Amotivation

Unfortunately, the teacher may inherit demotivated learners (Ushioda
2003; Dörnyei 2001b: 141–155). For example, some learners may have
initially been enthusiastic about learning English but their expectations
of progress may have been quite unrealistic. After some weeks or
months or years they may start to perceive learning English as
a daunting task and motivation may drop: they become demotivated.
The demotivated learner has often been adversely influenced by repeated
negative learning experiences in the classroom, often by negative com-
parisons with other learners’ performance. Traditional grading systems,
which produce ranking lists and marks based on comparisons across
learners, contribute greatly to this process of demotivating all but the
highest-scoring learners.
What can be done for such learners? One approach is to direct their
attention to the progress they have made and away from comparison
with other learners. They should be encouraged to keep a learning log to
record their own progress. It may help them to go back to listening or
reading comprehension tasks they did some weeks or months previously,
do them again and experience how they now understand more. They can
compare their two performances, reflect on the experience and record
their comments and scores in their learning log. For such learners, the
marking of written work should in general de-emphasise comparison
with others and emphasise comparison with the learner’s own previous
performance. Feedback to learners on written work and also on speaking
should not only involve correcting mistakes (negative feedback) but
should also involve praise for positive language features (positive feed-
back). Marking systems should be transparent and should reward posi-
tive features and not just penalise error. The provision of suitable
support (scaffolding) for both spoken and written tasks to prevent
learner breakdown is also important (Gibbons 2015).
Learners who are amotivated rather than demotivated pose a different
problem for the teacher. Unlike demotivated learners, amotivated learners
have not lost their motivation along the way. Rather they quite simply lack
any motivation to learn English at all (Deci and Ryan 1985; Noels et al.
2000). However, it is not always easy to distinguish the two groups of
learners at first sight. For this reason it is important with any new class to
sound out learners’ attitudes and discover their reasons for learning
English. The following features tend to characterise amotivated learners:
They are not looking forward to learning the language as an enjoy-

able experience (lack of intrinsic motivation).
They do not have any incentives to learn from their families or
employers (lack of extrinsic motivation).
They are not interested in social interaction with the speakers of the
target language or in the target culture (lack of integrative motivation).
They see no relevance of the language to their future lives (lack of
instrumental motivation).
What can be done for these learners? Young amotivated learners in

schools may benefit from focusing on authentic uses of English in thematic
areas which are close to their own interests, for example in sport, fashion,
music, dance, food and drink. Especially effective is the presentation of
their idols in sport, music or fashion speaking English on video clips. This
is better than pointing out to them the relevance of English for their future
life as an adult, something which seems a long way away when you are 12.
School-exchange programmes, native-speaker guests, viewing authentic
TV or films, reading English youth magazines and organising e-mail
exchanges or chat groups may also all help in the longer term. It is,
furthermore, often valuable to talk to learners directly about their lan-
guage-learning beliefs. Learners may believe, among other strange things:
that they should only speak when they are sure they will not make
a mistake
that they should never write a sentence if they think it might have
a mistake in it
that every single English word has a one-word translation equivalent
in their own language
Learner Attributions for Success and Failure
Learners may ascribe their success or failure in language-learning tasks,
especially tests and examinations, to various causes. Attribution theory
seeks to systematically describe these attribution patterns (see Weiner
1986: 44–51; 1992: 248–253). Distinctions are drawn between attribu-
tions to causes which are perceived as internal versus external, stable
versus unstable and controllable versus uncontrollable. Attributions
which are perceived as being within the learner’s control, such as effort
expended, are more likely to maintain or enhance motivation than those
which are perceived as being outside the learner’s control, such as
aptitude or luck. Most successful language learners have effort attribu-
tions for success and failure. Effort is an internal, unstable and control-
lable attribution. Learners should be encouraged to adopt effort
attributions. Unsuccessful and poorly motivated learners may attribute
success and failure to external, unstable, uncontrollable factors, such as
task difficulty, mood, help or hindrance from friends or family and
teacher bias (Weiner 1986: 46–47; 1992: 250). Attributions such as
teacher bias and help or hindrance from friends or family may be
perceived by learners as uncontrollable by the learner but controllable by
others so that in the case of failure these parties may be blamed by the
learner (Weiner 1992: 251).
Learners may also have different attributions for success and failure.
Some undesirable mixed patterns which occur include:
attributing success to effort and failure to lack of aptitude. As long as

they perceive themselves as making progress, such learners will
persist. However, they are too easily discouraged by setbacks such
as a poor exam result.
attributing success to good luck and failure to lack of aptitude. Such
learners will see no reason to put in effort to learning the language
and will almost inevitably give up (drop out) as soon as their early
success becomes marred by disappointment.
attributing success to aptitude and failure to bad luck. Such learners
perhaps were good at languages in the early stages without trying
very hard, and as demands become greater with requirements for
more effort, their results deteriorate. However, instead of realising
that effort is now required, they attribute their failure to external,
unstable and uncontrollable circumstances (bad luck). To an outsider
they always seem to be making excuses.
Feedback to the learner, such as examination marks and teacher com-

ments on performance, will affect learners’ motivation differently accord-
ing to their attribution patterns. Thus, if the learner attributes success to
external, unstable and uncontrollable factors such as the good mood of
the teacher/examiner, easy questions or generous marking, then a good
mark will not necessarily act as a motivator. Similarly, a poor mark will
not motivate a learner to try harder next time if the learner does not
attribute failure to lack of effort. The learner may regard the disappoint-
ing mark as bad luck (an external, unstable, uncontrollable attribution),
personal malice on the part of the teacher (an external, unstable, attribu-
tion, controllable only by the teacher, not by the learner) or even as
confirmation that he or she has no ability (an internal, stable, uncontrol-
lable attribution). All these attributions are likely to result in effort being
reduced for the future rather than being increased.
But where do learners’ attribution patterns come from? Different
patterns of attribution will reflect both differences among learners in
past experience and in personality. Repeated experiences of perceived
failure in the educational system may in extreme cases result in the
phenomenon of ‘learned helplessness’ (Abramson et al. 1978):
a fatalistic stance in which individuals see themselves as being at the
mercy of external forces and unable to influence either their own lives or
their environment (external, stable, uncontrollable attributions).
For more on learners’ attribution patterns see Ushioda (2001: 114),
Williams and Burden (1999), Williams et al. (2001), Peacock (2010),
Sahinkarakas (2011).
Personality
The Four Humours

There is no ideal language-learning personality, or at least no such
personality profile has been identified. It is, rather, the case that different
personalities may be suited to learning languages in different ways.
A good starting point for thinking about personality is the classical
Greek taxonomy of the four humours, namely phlegmatic, sanguine,
choleric and melancholic. The phlegmatic personality tends to be slow
to take action, calm, unemotional and resistant to change. The sanguine
personality tends to be optimistic, impulsive, cheerful, confident and
excitable. The choleric personality is short-tempered, impetuous, ambi-
tious and dominant. The melancholic personality is inclined to be
reflective, self-searching, depressive and demanding.
The Meyers-Briggs Type Indicator (MBTI)

Modern psychology has developed personality tests based on more
sophisticated taxonomies, the most popular of which is the Meyers-
Briggs Type Indicator (MBTI). This involves four macro-variables
expressed as scales or continua. The end points of each scale represent
personality extremes, with most people occupying intermediate positions
on the scale. Certain language-learning advantages and disadvantages can
be assumed for a tendency towards either end of a scale, and maybe
those persons in the middle can combine the best of both worlds. The
four scales are:
extroversion versus introversion

Extroverts derive satisfaction from the regard of others while intro-
verts derive satisfaction from personal fulfilment. At the risk of
stereotyping, it may be said that extroverts tend to be talkative,
outgoing and to value social interaction, while introverts tend to be
reserved, to be ‘loners’ and value observation and contemplation.
Extroverts will tend to be more willing to interact in conversation
and to take risks than will introverts. On the other hand, introverts
are well-suited to private study, cognitive problem-solving and indi-
vidual project work, whereas extroverts prefer group work, group
presentations and co-operative learning.
sensing versus intuition
The ‘sensing–intuition’ distinction reflects the contrast between those
who focus on detail, facts and the literal as opposed to those who
prefer larger conceptual thinking and interconnections at the figura-
tive level. The ‘sensing’ personality will focus on the details of the
language and observe closely differences between his or her own
output and that of more proficient speakers, which the ‘intuitive’
personality will tend not to do so, but will, on the other hand, be
more suited to inferring meaning from context, working with figura-
tive language and creating mental models of the language.
thinking versus feeling
The ‘thinking–feeling’ distinction is intended to encapsulate the
difference between those who rely on reasoning and those who rely
on emotions. The ‘thinking’ learner will favour an analytic approach
to language learning but may be too focussed on being completely in
control at all times and therefore possibly anxiety-prone, whereas the
‘feeling’ learner will bond well with teachers and fellow learners and
may have a strong integrative orientation in contrast to the ‘thinking’
learner, who may possibly be more instrumentally oriented. On the
other hand, the ‘feeling’ learner will be easily put off by poor
bonding among the class or with the teacher.
judging versus perceiving
The ‘judging-perceiving’ distinction concerns the difference
between those who like to work within firm guidelines, do not
like loose ends or unresolved problems, and those who prefer to
leave their options open, improvise and adapt spontaneously to
new situations. Those inclined to ‘judging’ will tend to work
systematically through exercises and complete clearly defined lan-
guage tasks to their own satisfaction, while the ‘perceivers’ will
prefer to work with open-ended tasks and will shine when it comes
to adapting to new language forms, unaccustomed registers, accents
and situations. On the other hand, they may be more inconsistent
language learners in the long term and be easily distracted.
For more on personality and language learning, see Brown (2014:

156–157), Dörnyei (2005: 18–20), Johnson (2018: 128–133), Dewaele
(2013: 626–631).
As mentioned at the outset of this section, there are no simple
correspondences between a specific personality profile and the good or
bad language learner. In fact, personality profiles based on the MBTI or
other well-known personality tests rarely account for more than 15% of
foreign-language success, and sometimes studies find no relationships or
even relationships opposite to those expected (see Dörnyei 2005: 24–30).
This applies even to dimensions such as introversion–extroversion,
where one might expect extroverts to be better language learners. In
fact, research results are mixed (see Dewaele 2013: 626–628).
Part of the problem is that macro-scales subsume a large number of
primary characteristics relevant to language learning, such as anxiety,
risk-taking, inhibition, sensitivity to rejection, self-esteem, self-efficacy,
willingness to communicate (WTC), tolerance of ambiguity and empa-
thy. They are thus not fine-grained enough to identify relevant micro-
features, especially as the effects of some primary personality factors may
affect language-learning success not directly but indirectly by influencing
motivation (compare Dörnyei 2005: 20–24). Another factor that tends to
obscure the effects of primary psychological variables in correlation
studies is that it is not always the case that the relationship between
a personality trait and language-learning success is linear and incremental.
Some personality factors, such as risk-taking, require an optimum rather
than a maximum level for successful language learning. Some of these
primary psycho-emotional characteristics will now be considered. Since
anxiety is by far the most important of these, it will be considered first
and in most detail.
Anxiety
Anxiety is an important primary personality characteristic. Anxious
personalities experience feelings of tension and apprehension which
tend to be vague and all-pervasive rather than having a clearly identified
source. However, there is a distinction between anxiety as a relatively
stable component of personality (‘trait anxiety’) and anxiety in
a particular situation (‘state anxiety’), which even normally non-anxious
individuals may experience under certain circumstances (MacIntyre and
Gardner 1991; Dörnyei 2005: 198; Brown 2014: 150–152). If learners
continually experience ‘state anxiety’ performing a certain task, such as
reading out loud, then ‘task anxiety’ results. That is to say, when the
same task or situation is encountered again, previously experienced
feelings of anxiety are triggered before one even starts to perform the
task. The term ‘language-learning anxiety’ refers to feelings of tension
and apprehension specifically associated with the language-learning class-
room. It affects language achievement, language performance and lan-
guage-learning motivation and is probably the single most serious
inhibitor of language learning in classrooms (MacIntyre and Gardner
1994: 284–285).
The anxious language learner experiences feelings of apprehension,
excessive self-evaluation, worry over potential failure, concern over the
opinion of others and physiological responses such as increased heart
rate and sweating (Gardner and MacIntyre 1993: 5). At the core of
language-learning anxiety is a form of severe cognitive interference,
since attention is divided between task-related cognition and distracting,
self-related, derogatory cognitions (‘I can’t do this,’ ‘I’m making a fool of
myself in front of everyone’) (MacIntyre and Gardner 1994: 285). Anxi-
ety differs from prototypical fear, for example of dogs, rats or someone
aiming a loaded sawn-off shotgun at you, in that the anxious person finds
it difficult to explain exactly what he or she is anxious about; anxiety is,
rather, a dread of undefined consequences, particularly of negative social
evaluation. This social evaluation element explains why anxiety is often
highest in a real or perceived test situation (test anxiety) as well as
in situations of speaking before the class (stage fright). Excessive self-
consciousness and fear of negative social evaluation is often a problem
associated with adolescence, and language-learning anxiety levels are
usually higher after puberty than before.
Anxiety affects the ability to take in information (input level), to
process and retain it (processing level) and the ability to produce lan-
guage (output level), so that the language-learning process itself is
impaired (MacIntyre and Gardner 1994). Output is affected because the
ability to access stored lexical information, to construct sentences and to
manage discourse planning are all negatively affected. Pronunciation may
also be negatively affected. In communicative situations, each level feeds
into the next, compounding the problem and in severe cases leading to
complete communicative breakdown, with oral test situations being
particularly likely to provoke anxiety. In severe cases learners may after-
wards report that they felt they were ‘standing beside themselves’, unable
to influence their own behaviour. The negative feelings experienced also
contribute to lowering motivation for the future.
In general, anxiety levels rise as arousal levels rise as a function of
perceived task challenge. If a task is perceived as too easy or too difficult
then the learner will be in a state of under-arousal or over-arousal,
respectively and motivation and task performance will be negatively
affected (Dörnyei 2005: 198–199). It is therefore important to keep anxiety
levels low in the classroom, while maintaining an atmosphere of task
challenge. Optimum perceived task challenge results in optimum arousal
levels and at these levels slight anxiety is experienced, for example before
a test one is well prepared for. This is ‘facilitating anxiety’ and it is likely to
have a positive effect on task performance (MacIntyre and Gardner 1994:
285; Johnson 2018: 145, note 1). Increasing levels of anxiety beyond this
threshold, however, quickly become ‘debilitating anxiety’, which has
massively negative effects on task performance.
What can be done to help anxious learners? It is of paramount impor-
tance to give them ample time to perform tasks. Under no circumstances
should they be rushed. They will generally find reading, writing and
listening tasks less stressful than speaking tasks. Speaking to the whole
class will be most stressful and speaking to the teacher will be more
stressful than speaking to a peer, especially a sympathetic, friendly peer.
They will be happier in situations where they are not being assessed or
evaluated (Horwitz et al. 1986: 127). Online correction while they are
speaking will be extremely anxiety-provoking and the anxious learner will
often not be able to take in the correction anyway, much less process it
and remember it. For some learners anxiety may be limited to certain sorts
of speaking task, such as speaking in front of the whole class. Speaking
tasks should therefore be very carefully chosen for anxiety-prone learners.
It is essential that they experience a sense of achievement. Particularly
beneficial will be allowing them to speak about something which chimes in
with their own personal experience, interests and hobbies or on some
matter about which they possess specialist knowledge.
For anxious, as for demotivated learners, comparisons among learners
should be avoided and competition among individual learners should be
replaced by cooperation among individuals within groups. The element
of competition should be restricted to competition among groups (team
competition). In this way the positive, motivating aspects of competition
may be retained but the negative anxiety-provoking aspects eliminated.
A tolerant attitude to error should be encouraged and it should be
explained to learners that only by making mistakes can they learn. For
this reason teachers should refrain from correcting mistakes as much as
possible with anxious learners, and when they do this it should be done
in a warm, friendly and helpful manner, never censoriously. On the
other hand, teachers should be wary of too obviously giving all the easy
questions to anxious learners and of being exaggeratedly accepting of
poor performance from anxious learners, as they may interpret this as
pity, which will only raise their anxiety levels still further. The class as
a whole may also notice this special treatment, which is not desirable.
Tests are a particular source of anxiety for many learners, and teachers
should try to de-mystify tests in various ways. One way is to cultivate
transparency about tests, so that learners know what the test’s purpose is
and what is expected of them as well as what knowledge and skills are to
be tested. Anxiety levels will also be lowered if learners are informed in
advance about the form the test will take, when it will take place and
what its consequences will be. Where possible, learners should be
involved in the negotiation of the final mark by incorporating some
measure of self-assessment or peer assessment. Rather than just being
given a mark, learners should be given a report which shows where their
strengths and weaknesses lie, perhaps including marks for different sub-
sections of the test. A talk with each learner individually and in private
may be beneficial. All this takes teacher time and effort, of course. But it
may well be worth it in the long term.
Not only are some individuals more anxious than others, but some
classes are more anxious than others. A teacher may normally expect
anxiety levels to be at their highest in the early stages of a course with
a new class, and to come down as the course progresses and the learners
and teacher get to know one another and bonding develops. In this
regard, special efforts should be made in the early stages of a course to
ensure that learners’ communication experiences in English are favour-
able. However, a teacher may take over a new class which has inherited
language-learning anxiety as a result of prior negative experiences. In
such circumstances it may be very difficult to decondition the learned
response. A teacher who takes over such a class must be prepared to
work slowly and painstakingly over many months.
The problem of anxiety is compounded for the teacher by the fact that
it is not always easy to recognise the anxious learner. Anxiety is not to be
confused with shyness. Anxious learners do not necessarily appear to be
shy, and not all learners who do appear to be shy are anxious learners.
Anxious learners may exhibit any of a number of rather different
behaviour profiles. Some anxious learners tend to write little and say
even less, to be slow to learn, to be hesitant speakers and be unwilling to
self-correct when a mistake is pointed out to them. Other anxious
learners tend to be too dependent on the teacher and tend to focus too
much on not making mistakes rather than on expressing ideas. Others
again may appear to be uncooperative, unwilling to talk, introverted, or
even aggressive. In some cases, anxious learners may become disruptive.
Occasionally, anxious learners may play the clown to deflect attention
away from their perceived deficits and change the agenda of the class-
room so that they can excel in a subversive role. A quite different form
of behaviour is exhibited by other anxious learners, who concentrate
intensely in class, hang on the teacher’s every word, carry out instruc-
tions to the letter and devote excessive attention to the minutiae of their
written work, while still tending to adopt low-risk-taking behaviour by
avoiding difficult topics and tasks, not volunteering to speak unless
absolutely certain of themselves, being linguistically uninventive or
imprecise in expression (reduction strategies) and thinking carefully and
for a long time before saying anything.
Risk-Taking, Inhibition and Sensitivity to Rejection
Johnson (2018: 232–233) suggests that the ability to take risks is part of
‘strategic competence’ (Canale and Swain 1980, mentioned in Chapter 1).
Anxiety tends to be negatively associated with risk-taking, or to put it
another way, self-confident people are more likely to take risks. However,
not all low-risk-takers are anxious individuals. So risk-taking is a distinct
personality trait. Individuals vary along a scale of willingness to take risks,
but an optimal rather than a maximum level is conducive to successful
language learning. Very-low-risk-takers will be handicapped in learning
a foreign language because they will be unwilling to experiment in produc-
tion and try out things they are not sure about. But very high risk-takers
will also not be ideal language learners since they tend not to plan what to
say nor to monitor their speech. They will tend to be insensitive to
feedback and will not seek out feedback nor will they be interested in
trying to learn from their mistakes (see Brown 2014: 149–150).
Inhibition involves not being prepared to risk losing face (‘ego protec-
tion’). It is related to anxiety and negatively related to risk-taking (Brown
2014: 147–149).
Sensitivity to rejection is related to anxiety and inhibition and nega-
tively related to risk-taking. Learners who are sensitive to rejection need
much positive feedback and encouragement and should be shielded from
situations where there is a risk of undue censure by others.
Willingness to Communicate (WTC)

Negatively related to anxiety and related to risk-taking, but distinct from
it, is ‘willingness to communicate’ (WTC) (MacIntyre 2007; Dörnyei
2005: 207–211; Brown 2014: 146–147). WTC refers to the extent to
which learners, either in the classroom or in the foreign-language com-
munity, are prepared to speak even though it is not absolutely necessary
to do so. It seems that high levels of WTC may be important for
continuing acquisition beyond ‘good enough to get by’ levels.
Self-Esteem and Self-Efficacy

Learners who are highly sensitive to rejection will tend to have low self-
esteem. Self-esteem means an attitude of approval towards oneself. High
levels of self-esteem make it possible for learners to cope with the
inevitable setbacks of language learning and to persist with learning and
not lose motivation (see Dörnyei 2005: 211–214; Brown 2014: 144–145).
Closely related to self-esteem is self-efficacy, that is, the belief in one’s
ability to perform a specific task (Brown 2014: 145–146). It is possible for
learners who have generally low levels of self-esteem (‘I am not much
good at anything’) to be nevertheless confident of their abilities in one
particular domain, for example pronunciation or rote memorisation (‘I’m
good at this’).
Tolerance of Ambiguity
Learner personalities differ in other ways which are relevant for language
learning. These include tolerance of ambiguity, that is, to what extent
one needs to have uncertainty resolved immediately (‘closure-oriented’
personalities) and to what extent one can proceed with the next stage of
a procedure, such as listening or reading, while keeping a number of
options open until more information becomes available to clear up the
uncertainty (‘open’ personalities). Discovery-based teaching approaches
will be suited to ‘open’ learners and highly structured teaching which
proceeds in steps will appeal to ‘closure-oriented’ learners. Closure-
oriented learners find it difficult to keep on reading after encountering
a word they do not know. They also are sometimes dissatisfied until they
have a definite translation of a new word into their own language. Open
learners may, however, misunderstand utterances or texts because they
settle for a vague meaning and may thus miss the point. As with risk-
taking, an optimum rather than a maximum level is best suited to
language learning (see Brown 2014: 117–118; Johnson 2018: 130–131).
Empathy
Empathy is another relevant personality variable. Empathy is the ability
to put oneself in someone else’s shoes, to experience a situation from
another perspective (see Brown 2014: 153–154). Language learning is
a social activity and empathetic learners will respond well to role-play
activities, to drama-based approaches and to pronunciation work in
specific situations.
Cognitive Style
Cognitive style refers to the way in which an individual typically tends to
approach problems of all sorts in life (Wyss 2002: 1; Dörnyei 2005:
121–122, 124–125; Brown 2014: 113–114). Various components of cog-
nitive style have been developed heuristically. These components are
usually expressed as two end points of a scale. Individuals can be placed
at various points along the scale. Various researchers have created their
own taxonomies of cognitive style components, and unfortunately there
is no definitive taxonomy and much overlap from catalogue to catalogue.
However, certain well-established components of cognitive style seem
relevant for language learning.
Field-Independent versus Field-Dependent Styles
The most well-known of these is the ‘field-independent’ versus ‘field-
dependent’ scale (Brown 2014: 114–115; Dörnyei 2005: 136–139; Wyss
2002; Johnson 2018: 131–132; Ehrman and Leaver 2003: 396–397). This
is the difference between those who are able to isolate the components of
a coherent whole and those who find it difficult to decompose the whole
into its constituent parts. Those who can find hidden objects in
a picture, or make out forms within an ink blot, or see faces in wallpaper
patterns are said to be field-independent. However, the field-independent
learner’s ability to focus on detail may well mean that he or she is
insensitive to the wider context, while the field-dependent learner’s
inability to focus on detail may also imply that he or she is highly
aware of the wider context. The price for field-independence may be
field-insensitivity and the payout for field-dependence may be ‘field-
sensitivity’.
Field-independent learners will respond well to teaching approaches
which isolate specific elements of the language, deal with pronunciation,
grammar and vocabulary separately and focus on individual sounds,
specific grammar points and the meaning of single words (a ‘bottom-up’
approach). However, in text-based work they may tend to shut out
context and so not be able to cope with polysemous words, pragmatic
meaning and secondary, indirect meaning, including irony, implication
and allusion (field-insensitivity). Field-dependent learners, on the other
hand, will respond well to teaching approaches which deal with language
holistically in texts and which focus on the pronunciation of whole
utterances, the meaning of multi-word chunks and on identifying larger
syntactic patterns rather than morphological details (a ‘top-down’
approach). In text-based work they are likely to be skilled at invoking
context to infer the meaning of a polysemous word, pragmatic meaning
and various forms of indirect language (field-sensitivity).
Most language learners will be at neither extreme of the scale so they
will combine elements of both poles with a tendency towards one or the
other. Very good language learners may somehow strike the perfect
balance and be able to combine field-sensitivity with field-independence
according to the demands of the specific language-learning task.
Deductive versus Inductive Reasoning Styles

Some people tend to reason deductively, that is, they tend to set up
general principles or rules and then see if the concrete examples
conform to those rules or principles. Others tend to reason induc-
tively, that is, they tend to assemble examples and then work out
general rules or principles which explain those examples (Ehrman and
Leaver 2003: 398–399). In other words, deductive reasoners like to
proceed from rules to examples, whereas inductive reasoners like to
infer rules from examples. This has obvious relevance for grammar
teaching. Deductive reasoners will prefer traditional grammar teaching,
which presents rules and then gives isolated examples to illustrate the
rule. A preference for deductive reasoning often goes hand in hand
with field-independence. Inductive reasoners, on the other hand, will
prefer text-based grammar teaching, which encourages learners to infer
underlying grammatical rules from examples in texts. A preference for
inductive reasoning often goes hand in hand with field-dependence
(see Brown 2014: 97–98).
Global versus Particular Reception Styles

‘Global’ language learners tend to react to the general impression of
language they hear or read, while ‘particular’ learners tend to focus on
detail (Ehrman and Leaver 2003: 398). Field-independent learners tend to
be ‘particular’ and field-dependent learners tend to be ‘global’. The
global learner prefers extensive reading, for example of novels, to inten-
sive reading and is not distracted by partial non-comprehension, caused
for example by unknown vocabulary. The ‘particular’ reader prefers
reading closely rather than reading for gist and is well-suited to close
reading, of lyrical poems, for example. He or she may like to stop at the
end of each paragraph and look up unknown words.
Sharpening versus Levelling Memory Styles

‘Sharpeners’ prefer to note distinctions between similar entities for exam-
ple, between near synonyms such ‘pretty’, ‘beautiful’ and ‘handsome’, or
between dark ‘l’ and light ‘l’, while ‘levellers’ tend to group similar
elements together under a single heading and blur distinctions, so as to
simplify the learning task. (See Chapter 5 for the difference between dark
and light ‘l’.) This helps them to get on in the language more quickly, but it
may make their language inaccurate, for example if they use ‘pretty’,
‘beautiful’ and ‘handsome’ interchangeably or do not distinguish between
dark and light ‘l’. Ehrman and Leaver (2003: 401) explain the difference
between levelling-sharpening and global-particular as follows:
Levelling lumps information and reduces distinctions, whereas global

learning seeks the underlying meaning and overall pattern of an
experience. Sharpening seeks significant distinctions, whereas parti-
cular learning focuses on specifics without necessarily making dis-
tinctions of category.
Ehrman and Leaver (2003: 401)
Impulsive versus Reflective Response Styles
In speaking or writing, impulsive learners tend to think that the first
word that spontaneously comes to mind is likely to be correct, whereas
reflective learners prefer to think twice or try to recall rules rather than
relying on their intuition (Ehrman and Leaver 2003:402). Impulsive
learners may therefore be more fluent than reflective learners, but may
make more ‘careless mistakes’. In other terms, impulsivity may be
associated with under-monitoring and reflectivity with over-monitoring.
Random versus Sequential Processing Styles

‘Random’ learners do not like to work to a pre-imposed plan, whereas
‘sequential’ learners do (Ehrman and Leaver 2003: 398). Thus, sequential
learners like to work within a clearly defined syllabus, have clear lesson
plans and like to do written work according to a set of guidelines.
Random learners, by contrast, like to have freedom to adapt in response
to changing circumstances in an ongoing fashion. They are better at
unscripted language tasks, such as conversation.
Synthesising versus Analytic Styles

Synthesising learners like putting parts together to form a whole,
whereas analytic learners prefer to dissect in order to isolate key compo-
nents (Ehrman and Leaver 2003: 399). Field-independence tends to be
associated with a propensity for analysis and field-dependence with
a propensity for synthesis.
Language-Learning Style
Style Preferences
By learning style is meant ‘an individual’s natural, habitual and preferred
way(s) of absorbing, processing and retaining new information and skills’
(Reid 1995: viii). An individual’s preferred language-learning style may
differ from his or her preferred way of learning other subjects. Differences
among individuals in language-learning style will be influenced by cogni-
tive style and personality differences, as well as training, experience,
culturally transmitted beliefs, age and cultural background.
Teachers quickly become aware of language-learning style prefer-
ences among their learners. They notice that:
some prefer to work alone, others in cooperation with their fellows

some prefer to work in the aural-oral mode, others in the written
mode
some prefer to work with a textbook, others with authentic materials
some prefer highly structured activities, others freer activities
some prefer teacher explanation, others prefer discovery techniques
some prefer to plan activities in advance, others like to improvise
some prefer teacher-led activities, others prefer to work
autonomously
some prefer detailed teacher feedback and correction, others do not
some prefer teacher assessment, others prefer peer assessment or
self-assessment
There is no single best language-learning style for all learners. However,

language-learning styles are dynamic and should develop in the course of
the years in keeping with a learner’s personality, inclinations, abilities
and learning opportunities, including the style of teaching encountered.
Style Stretching
If learners are offered a wide choice of learning activities to perform, this
will enable each individual learner to choose tasks appropriate to his or
her language-learning style. However, learners should also be encouraged
to operate on occasion outside the comfort zone of their own preferred,
and possibly restricted, learning style (‘style stretching’). Taking
a learning-style test can help learners to raise their awareness of the
wide variety of learning styles that exist and encourage them to explore
new styles previously unknown to them. Well-known learning-style tests
are Joy Reid’s ‘Perceptual Learning Style Preference Questionnaire’,
Rebecca Oxford’s ‘Style Analysis Survey’, D. A. Kolb’s ‘Learning Style
Inventory’ and Cohen, Oxford and Chi’s ‘Learning Style Survey’. These
are all available online.
Style Mismatches
According to Ehrman (1996: 50), ‘Learning style mismatches are at the
root of many learning difficulties.’ Problems may arise if there is a clash
between a learner’s preferred learning style and the teacher’s preferred
teaching style or with the stipulations of the syllabus (Dörnyei 2005:
155). Indeed this may sometimes be a delicate matter, particularly where
cultural differences are involved. Learners from cultures where education
is more traditional, where the teacher is a figure of authority and where
teaching practices are far more autocratic may be alienated by open
approaches to teaching with an emphasis on autonomy and discovery
procedures, for instance.
Sometimes learners may have picked up misguided beliefs about
language learning. They may try to apply these ideas uncritically even
though they do not suit their own personality, abilities and inclinations.
In such cases it may be part of the teacher’s job to wean the learner away
from false ideas about how to learn. A possible scenario is a learner who
has been taught that rote learning is the only way to acquire vocabulary,
believes this, and thus tries to apply vocabulary rote-learning strategies
even though he or she may have a weak memory for rote learning but
great ability to retain vocabulary encountered incidentally in reading.
Food for Thought

Q. 1. Are there any individual learner differences you regard as impor-
tant in determining success in English as a Foreign Language which
have not been mentioned in the chapter?
Q. 2. The distinctions between intrinsic versus extrinsic motivation, on
the one hand, and instrumental versus integrative motivational
orientations, on the other hand, are two different distinctions
made by different researchers in different contexts. How would
you explain the interplay between the two distinctions?
Q. 3. (Group task) From the group’s experience as language learners,
what are some likely attributions for success and failure of good
language learners and poor language learners, respectively?
Q. 4. Why do you think learners often attribute success and failure to
different causes?
Q. 5. Can you see any problems which may arise if there is a mismatch
between the learner’s attributions and the teacher’s assumptions of
what the learner’s attributions are?
Q. 6. Complete a sample Myers-Briggs Type Indicator (MBTI) and,
while you are doing it, consider to what extent it is functioning
well in your own case. After completion, reflect on your experi-
ence and critically evaluate the test. A version of the test is
available online at:
www.lusd.org/cms/lib6/CA01001399/Centricity/Domain/711/THE
%20MYERS-BRIGGS.pdf
Q. 7. (Group task) Can you think of steps other than those suggested in
the chapter which teachers can take to control anxiety levels in the
classroom?
Q. 8. (Group task) Look back at the description of extroverts versus
introverts. What sort of course design might suit introverts and
what sort of course design might suit extroverts?
Further Reading
Pfenninger, S. E. and Singleton, D. (2017). Beyond Age Effects in Instructional L2
Learning: Revisiting the Age Factor. Bristol and Blue Ridge Summit: Multilingual
Matters.
Roberts, L. and Meyer, A. (eds.) (2012). Individual Differences in Second Language
Learning. Language Learning 62, Supplement 2. September 2012. University of
Michigan, MI: Wiley Blackwell.
References
Abrahamsson, N. and Hyltenstam, K. (2008). The robustness of aptitude effects in
near-native second language acquisition. Studies in Second Language Acquisition
30/4: 481–509.
Abramson, L. Y., Seligman, M. E. P. and Teasdale, J. D. (1978). Learned help-
lessness in humans: critique and reformulation. Journal of Abnormal Psychology
87/1: 49–74.
Brown, H. D. (2014). Principles of Language Learning and Teaching. Sixth edition.
White Plains, NY: Pearson Education.
Carroll, J. B. (1973). Implications of aptitude test research and psycholinguistic
theory for foreign-language teaching. Linguistics 112: 5–14.
Carroll, J. B. (1981). Twenty-five years of research on foreign language aptitude. In
Diller, K. C. (ed.), Individual Differences and Universals in Language Learning
Aptitude. Rowley, MA: Newbury House, pp. 83–118.
Carroll, J. B. and Sapon, S. (1959). Modern Language Aptitude Test, Form A.
New York: The Psychological Corporation.
Chambers, G. (1994). A snapshot in motivation at 10+, 13+ and 16+. Language
Learning Journal 9 (March): 14–18.
Clément, R., Dörnyei, Z. and Noels, K. A. (1994). Motivation, self-confidence and
group cohesion in the foreign language classroom. Language Learning 44/3:
417–448.
Deci, E. L. and Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in
Human Behaviour. New York: Plenum.
Dewaele, J.-M. (2013). Individual differences in second language acquisition. In
Ritchie, W. C. and Bhatia, T. K. (eds.), The New Handbook of Second Language
Acquisition. Second edition. Leiden: Brill, pp. 623–646.
Dörnyei, Z. (2001a). Motivational Strategies in the Language Classroom. Cambridge:
Dörnyei, Z. (2001b). Teaching and Researching Motivation. Harlow: Longman.
Dörnyei, Z. (2005). The Psychology of the Language Learner: Individual Differences in
Second Language Acquisition. Mahwah, NJ and London: Lawrence Erlbaum
Associates.
Dörnyei, Z. and Ushioda, E. (eds.) (2009). Motivation, Language Identity and the L2
Self. Bristol: Multilingual Matters.
Ehrman, M. E. (1996). Understanding Second Language Learning Difficulties. Thou-
sand Oaks, CA: Sage Publications.
Ehrman, M. E. and Leaver, B. L. (2003). Cognitive styles in the service of language
learning. System 31/3: 393–415.
Ehrman, M. E. and Oxford, R. L. (1995). Cognition plus: correlates of language
learning success. Modern Language Journal 79/1: 67–89.
Gardner, R. C. (1985). Social Psychology and Second Language Learning: The Role of
Attitudes and Motivation. London: Edward Arnold.
Gardner, R. C. (2010). Motivation and Second Language Acquisition: The Socio-
educational Model. New York: Peter Lang.
Gardner, R. C. and Lambert, W. (1972). Attitudes and Motivation in Second
Language Learning. Rowley, MA: Newbury House.
Gardner, R. C. and MacIntyre, P. D. (1993). A student’s contribution to second
language learning. Part II: affective factors. Language Teaching 26: 1–11.
NH: Heinemann.
Grigorenko, E., Sternberg, R. and Ehrman, M. E. (2000). A theory-based approach
to the measurement of foreign language learning ability: the Canal-F theory and
test. Modern Language Journal 84/3: 390–405.
Horwitz, E. K., Horwitz, M. B. and Cope, J. (1986). Foreign language classroom
anxiety. The Modern Language Journal 70/2: 125–132.
MacIntyre, P. D. (2007). Willingness to communicate in the second language:
understanding the decision to speak as a volitional process. The Modern Lan-
guage Journal 91/4: 564–576.
MacIntyre, P. D. and Gardner, R. C. (1991). Language anxiety: its relationship to
other anxieties and to processing in native and second languages. Language
Learning 41: 513–534.
MacIntyre, P. D. and Gardner, R. C. (1994). The subtle effects of language anxiety
on cognitive processing in the second language. Language Learning 44/2:
283–305.
Markus, H. and Nurius, P. (1986). Possible selves. American Psychologist 41:
954–969.
Maslow, A. H. (1970). Motivation and Personality. Second edition. New York:
Harper and Row.
Noels, K. A., Pelletier, L. G., Clément, R. and Vallerand, R. J. (2000). Why are you
learning a second language? Motivational orientations and self-determination
theory. Language Learning 50/1: 57–85.
Nuttall, C. (1996). Teaching Reading Skills in a Foreign Language. New edition.
Oxford: Heinemann.
Peacock, M. (2010). Attribution and learning English as a foreign language. ELT
Journal 64/2: 184–193.
Pimsleur, P. (1966). Language Aptitude Battery. New York. Harcourt, Brace and
World.
Reid, J. M. (1995). Preface. In Reid, J. M. (ed.), Learning Styles in the ESL/EFL
Classroom. New York: Heinle and Heinle, pp. viii–xvii.
Saffron, R. (1988). An investigation of the characteristics of the good language
learner. Unpublished MA thesis, University of Birmingham, UK.
Sahinkarakas, S. (2011). Young students’ success and failure attributions in lan-
guage learning. Social Behaviour and Personality: an International Journal 39/7:
879–886.
Skehan, P. (1986). Where does language aptitude come from? In Meara, P. (ed.),
Spoken Language: Papers from the Annual Meeting of the British Association for
Applied Linguistics Held at the University of Edinburgh, September 1985. London:
Centre for Information on Language Teaching and Research (CILT), pp. 95–113.
Skehan, P. (1989). Individual Differences in Second Language Learning. London:
Edward Arnold.
Skehan, P. (2012). Language Aptitude. In Gass, S. M. and Mackey, A. (eds.), The
Routledge Handbook of Second Language Acquisition. New York: Routledge, pp.
381–395.
Sleve, L. R. and Miyake, A. (2006). Individual differences in second language
proficiency. Does Musical Ability Matter? Psychological Science 17/8: 675–681.
Ushioda, E. (1996). Developing a dynamic concept of L2 motivation. In Hickey, T.
and Williams, J. (eds.), Language, Education and Society in a Changing World.
Dublin: IRAAL/Multilingual Matters, pp. 239–245.
Ushioda, E. (2001). Language learning at university: exploring the role of motiva-
tional thinking. In Dörnyei, Z. and Schmidt, R. (eds.), Motivation and Second
Language Acquisition. Honolulu, HI: University of Hawai’i Press, pp. 93–125.
Ushioda, E. (2003). Motivation as a socially mediated process. In Little, D.,
Ridley, J. and Ushioda, E. (eds.), Learner Autonomy in the Foreign Language
Classroom: Teachers, Learners, Curriculum and Assessment. Dublin: Authentick,
pp. 90–102.
Ushioda, E. (2006). Language motivation in a reconfigured Europe: access, iden-
tity, autonomy. Journal of Multilingual and Multicultural Development 27/2:
148–161.
Ushioda, E. and Dörnyei, Z. (2012). Motivation. In Gass, S. M. and Mackey, A.
(eds.), The Routledge Handbook of Second Language Acquisition. New York: Rou-
tledge, pp. 396–409.
Weiner, B. (1986). An Attributional Theory of Motivation and Emotion. New York:
Springer.
Weiner, B. (1992). Human Motivation: Metaphors, Theories and Research. Newbury
Park, CA: Sage Publications.
Wells, C. G. (1985). Language Development in the Pre-School Years. Cambridge:
Williams, M. and Burden, R. L. (1997). Psychology for Language Teachers: A Social
Constructivist Approach. Cambridge: Cambridge University Press.
Williams, M. and Burden, R. L. (1999). Students’ developing conceptions of
themselves as language learners. The Modern Language Journal 83/2: 193–201.
Williams, M., Burden, R. L. and Al-Baharna, S. (2001). Making sense of success
and failure: the role of the individual in motivation theory. In Dörnyei, Z. and
Schmidt, R. (eds.), Motivation and Second Language Acquisition. Honolulu, HI:
University of Hawai’i Press, pp. 171–184.
Young, T. J. (2007). Intercultural Communicative Competence and the Teaching
and Learning of English as a Foreign Language. Unpublished Ph.D. dissertation.
University of London, Birkbeck College.
Online Reference
Wyss, R. (2002). Field independent/dependent learning styles and L2 acquisition.
www.eltnewsletter.com/back/June2002/art1022002.htm
4 Facilitating Classroom Learning
The Orchestration of Classroom Interaction

In Chapter 1, learning English in a classroom setting was contrasted with
acquiring English naturalistically in an English-speaking community. It
was said that the classroom was an acquisition-poor environment com-
pared to being surrounded by the language and its speakers all day,
every day. However, the great advantage of learning English in the
classroom compared to trying to pick it up naturalistically is the avail-
ability of a professionally trained teacher who orchestrates classroom
interaction so as to promote acquisition, provides guidance, feedback
and explanation where necessary as well as encouragement and perhaps
sometimes censure. This chapter will first focus on four resources
available to the English teacher, namely the syllabus, needs analysis,
teaching material and Information and Communication Technology
(ICT). Then the focus will switch to classroom interaction itself, and
particular attention will be paid to teacher and learner classroom roles,
to input modification and interactional modification, to learner coopera-
tion and self-direction as well as to provision of feedback to learners.
Finally, some rules of thumb for teachers will be offered.
The Syllabus
What Is a Syllabus?
Most classroom teaching takes place within the framework of a syllabus.
This may be set by the institution, by an external examining body or by
the education ministry. The syllabus writer traditionally asks three
questions:
What are the goals of the course? (goal-setting)

Which elements of the language must be covered in order to achieve
those goals? (selection)
In what order must these language elements be covered? (sequencing)
Facilitating Classroom Learning 91
Nowadays the syllabus is often available online, and is likely to be
scrutinised not only by teachers and learners but also by parents,
employers and members of other educational institutions. As a publicly
available document, it makes for transparency, provides supportive
guidelines for teachers and can protect teachers from unfounded criti-
cism. On the other hand, the syllabus is a document which is imposed
on the teacher from above and can, under certain circumstances, be
constrictive. Some syllabuses are more prescriptive than others. Some
may lay down the language content to be taught, others may merely
specify end-of-course goals and give teachers considerable freedom on
how they achieve those goals. Some syllabuses may specify a course
book or other materials that have to be used, whilst others allow
teachers freedom in the choice of materials. Some syllabuses may specify
points in the course at which tests have to be conducted, while others
may not. Some syllabuses may be aimed at preparation for a final
examination (an examination syllabus).
The more prescriptive a syllabus is, the more constrictive it will be but
the more support it will provide for teachers in terms of methods to be
adopted, materials to be used and the selection and sequencing of lesson
content over weeks, months and even years. The less prescriptive
a syllabus is, the more freedom teachers have to select their own
materials and to use methods they think fit. On the other hand, teachers
may feel that a syllabus which specifies only broad goals but leaves
options open at the level of selection and sequencing does not offer
them the support they need. An example might be a syllabus that
specifies ‘negotiation skills’ as a goal but neither specifies the level of
competence to be aimed at, nor indicates the vocabulary and structures
to be introduced, or the topics, settings and participator roles to be
practised.
A syllabus does not necessarily entail a particular teaching method or
approach, but in fact certain methods are closely associated with parti-
cular sorts of syllabus (Ur 2012: 185–190; Richards 2013: 11). The
contents page of the course book, if there is one, will reflect the syllabus
(Johnson 2018: 192). Some major syllabus types will now be briefly
described.
The Grammatical Syllabus

This is the traditional syllabus, closely associated with the Grammar-
Translation Method and, in the guise of the structural syllabus, with the
Audiolingual Method. These methods were discussed in Chapter 2.
Many syllabuses today are to a greater or lesser extent still based on
grammatical selection and sequencing, although goals are unlikely to be
formulated solely in grammatical terms any longer (see Richards 2001:
153–154; Ur 2012: 186–187; Johnson 2018: 194–199; Long 2015:
207–208). In a strict grammatical syllabus the course was divided into
units, each of which dealt with a selected grammatical structure or item
in a pre-determined sequence. Criteria for selection and sequencing
might include frequency, difficulty, usefulness, learnability and authenti-
city (Richards 2013: 6), the two main criteria being frequency and
difficulty. In other words, frequent structures were to be taught before
less frequent structures and simple structures before complex structures.
However, the twin criteria of frequency and complexity are not in
themselves straightforward. Concerning frequency, some words of low
general frequency may occur frequently in the classroom environment
(‘board-duster’, for example). Concerning complexity, some grammar
items, the modal verbs for example, may be formally simple but func-
tionally complex (see Johnson 2018: 196–197). Furthermore, progress
from the simple to the complex and from the most frequently occurring
to the least frequently occurring items is not always identical. A very
obvious example is irregular versus regular verbs. Clearly, regular verbs
are formally simpler and therefore easier to learn than irregular verbs,
but it is precisely the most frequently used verbs which are the irregular
ones. The two most frequently occurring verbs are ‘be’ and ‘have’, and
these are both irregular. Usually they are taught first, so that the
frequency criterion is rightly given priority over the complexity criter-
ion. However, just how many, if any, other irregular verbs should be
introduced before regular verbs are introduced is a decision each sylla-
bus designer has to make.
A related problem is how strictly the sequencing principles of frequency
and complexity should be adhered to anyway. Too strict an adherence
means that all the bits of the grammar not yet ‘covered’ at any point in
time have to be excluded from classroom language and course-book texts.
This means that authentic texts cannot be used until very late and the
language of the contrived texts in the course book will tend to be over-
simplified and unnatural. For instance, a syllabus I once had to work with
taught beginners to say ‘Have you a book?’ This was done because the much
more frequent forms ‘Do you have …’ and ‘Have you got …’ were more
complex and were therefore to be taught later. However, ‘Have you
a book?’ sounds unnatural to many speakers, who reserve subject-verb
inversion with ‘have’ for its use as an auxiliary verb, as in ‘Have you seen
Peter?’ The result was teaching-induced unnaturalness in the resultant
learner language (see discussion of Selinker 1972 in Chapter 1).
The Lexical-Grammatical Syllabus

The lexical-grammatical syllabus is a refinement of the grammatical
syllabus. It includes as an addendum a list of words to be covered at
each stage of the course or for each unit of the course book (see Richards
2001: 154). The Direct Method, for example, was often based on strict
selection and sequencing of both grammatical structures and lexis. The
usual criterion invoked for selection and sequencing of vocabulary is
frequency of occurrence. However, this is not always unproblematic.
The most frequently occurring colour words, for instance, are the
primary colours and these are usually taught to beginners first. However,
a non-primary colour, such as ‘pink’, may be easier for learners to
pronounce and write than the primary colour ‘yellow’, and might also
be relevant to the learners for affective reasons, for instance being
a favourite dress colour. Learners in fact often want to know from the
start what their favourite dress colours are in English, and these may be
mauve, pink and lilac, for some learners, or beige, fawn and grey for
others. Some learners may want to know the colours of their favourite
football team, which may not always be primary colours either. Such
learner preferences have to be taken into account because they will
influence intrinsic motivation, which was discussed in Chapter 3.
The Situational Syllabus

The situational syllabus is associated with Situational Language Teaching
(see Chapter 2). The course book plays a key role since this is often
organised around a series of lessons, each of which has a particular
setting, for example, ‘At the Bank’ (see Richards 2001: 156–157; Ur
2012: 188; Long 2015: 212–214). The central element of each lesson is
often a reading passage or a printed dialogue. The idea is that the
language to be taught in each situational unit is suggested by the
situation, for example, ‘I would like to open an account’ for an ‘At the
Bank’ unit. The situation also specifies certain word fields, for example
the ‘money’ word field for ‘At the Bank’. However, for general English
courses rather than ESP courses, principled criteria for which situations
to choose in the first place tend to remain vague.
The Topic-Based Syllabus

The topic-based syllabus is a variation on the situational syllabus (see
Richards 2001: 157–159). Topics for young learners might include ‘the
family’, ‘pets’, ‘the weather’ and ‘hobbies’, to name just a few. Topic-
based syllabuses are well-suited to ESP courses since topics can be
selected which are relevant for the learners’ current or future profes-
sional needs. This can be done by means of ‘needs analysis’ (see
below). It is in Content and Language Integrated Learning (CLIL),
where a school subject is taught in English by a qualified subject-
teacher, that the topic-based syllabus finds its fullest expression. Here
the syllabus for the content subject replaces the language syllabus (see
Long 2015: 214–215).
The Notional-Functional Syllabus
The notional-functional syllabus (Wilkins 1976), mentioned in Chapter
2, is associated with Communicative Language Teaching (CLT) and is an
attempt to rely on functions and notions to structure a syllabus rather
than grammar, lexis, situations or topics. Functions might include ‘pro-
mising’, ‘agreeing and disagreeing’ ‘suggesting’ and ‘objecting’. Notions
might include ‘place’, ‘direction’, ‘distance’, ‘duration’ and ‘comparison’
(see Richards 2001: 37–38; Johnson 2018: 199–207; Ur 2012: 188–189;
Long 2015: 208–210). Such a syllabus is sometimes described as ‘analytic’
in contrast to grammatical and lexical-grammatical syllabuses, which are
‘synthetic’. Ur (2012: 185) explains the difference as follows:
Synthetic syllabuses provide a set of isolated language items (gram-

matical structures or lexical items, for example). The learner is asked
to learn these separately and then combine them to create or under-
stand meaningful phrases or sentences in order to deal with language
in context. Analytic syllabuses work the other way round. They
describe communicative abilities, tasks or functions (conveying
simple information, for example, or understanding the main points
of a text). These can then be analysed, and the required words or
structures are taught as an outcome rather than a starting point.
However, as already mentioned in Chapter 2, it has proved difficult to

specify a definitive list of functions or notions, let alone select and
sequence them in a principled manner. Functions just seem to proliferate
the more you think about them. Van Ek and Trim (1998: 27–47) listed
132 functions in their Threshold Level Syllabus. These were divided into
six categories: ‘imparting and seeking factual information’ (5 functions),
‘expressing and finding out attitudes’ (57 functions), ‘deciding on courses
of action’ (14 functions), ‘socialising’ (12 functions), ‘structuring dis-
course’ (28 functions) and ‘communication repair’ (16 functions). There
has thus understandably been reluctance in syllabus design to abandon
grammatical principles of selection and sequencing.
The Output-Based Syllabus

The output-based syllabus is sometimes called ‘the competence-based
syllabus’ or ‘the standards-based syllabus’ (see Richards 2001: 159; Ur
2012: 189–190). It is closely associated with Competence-Based Language
Teaching (CBLT), discussed in Chapter 2, since it specifies task-related
proficiency goals to be achieved rather than language to be learned, for
example, ‘introducing oneself’, ‘asking for directions’, ‘going shopping’,
‘making a travel booking by telephone’, ‘understanding an academic
lecture’, ‘writing a business report’. These are often formulated as ‘I
can …’ statements and describe outcomes of learning rather than provid-
ing information on teaching input. Syllabuses based on proficiency
benchmarks such as the Common European Framework of Reference
(CEFR) (see Council of Europe 2001: 231–257) or the Michigan English
Language Proficiency Standards (MELPS) are output-based.
Task-Based Language Teaching (TBLT), discussed in Chapter 2, is also
linked to the output-based syllabus since often the tasks specified as
output goals by the syllabus become identical with the task-based activ-
ities done in the classroom. This works particularly well for ESP courses.
In other words, the output-based syllabus solves the problem of selection
and sequencing of tasks in TBLT (Richards 2013). The snag with output-
based syllabuses, however, is that they do not specify for teachers the
required grammar and vocabulary required for mastery of desired out-
comes (Ur 2012: 189–190). They may also be vague about the level of
proficiency to be aimed at or the level of task complexity required.
The Multi-Track Syllabus

Many modern syllabuses are multi-track or ‘integrated’ syllabuses (see
Richards 2001: 164–165; Ur 2012: 190; Johnson 2018: 207–2018; Long
2015: 222–223). They seek to take into account grammar, vocabulary,
communicative functions, topics and output goals. For example, the
global course goals may be expressed in output terms; the internal
structure of the course (and textbook) may be organised around topic-
based units; the required language input (grammar, vocabulary notions
and functions) is then packaged into the individual topic-based units.
This involves syllabus writers in intricate, demanding and creative work,
matching desired output behaviour to grammar, vocabulary, functions
and notions, which have to be packaged into each chosen topic. The key
to successful modern syllabus design is without doubt flexibility within
an underlying systematic approach.
Course books which are based on a multi-track syllabus often design
their Table of Contents with four or five columns, one for the topic of
the unit, one for its grammatical focus, one for the vocabulary focussed
on and one for the notions/functions dealt with. They may in addition
specify the desired output competences for each unit (see Ur 2012: 190;
Johnson 2018: 193–194; 207–208). However, usually one track will be
primary and the others subsidiary. Often the primary structuring princi-
ple is to be found in the grammar/structures column (Harmer 2015: 212;
Johnson 2018: 208).
The Cyclical Syllabus

Traditional grammatical and lexical-grammatical syllabuses tended to be
essentially linear in structure: once a grammatical item had been covered
in the syllabus, or an item of vocabulary introduced, it was assumed it
had been mastered. The only concession to the fact that learning is
neither linear nor instantaneous was the inclusion of revision units in
the syllabus and the course book at various points. In the cyclical
syllabus, sometimes called ‘the spiral syllabus’, by contrast, learners
return over and over again to elements they have already encountered
so as to build on them, refine them and develop them to higher levels of
complexity in new settings, registers, genres and styles (see Bruner 1977:
13, 52–54; Richards 2001: 13–14, 151). An example would be advanced
learners returning to the modal verbs to explore additional modal mean-
ings and functions.
This list of syllabus types is by no means exhaustive. For mention of
some further types of syllabus, including the lexical syllabus, the proce-
dural syllabus and the process syllabus, see Long (2015: 210–212,
216–221).
Needs Analysis
In English for Special Purposes (ESP) courses there is sometimes no
syllabus set out in advance and the content of the course has to be
decided on by the teacher or course designer and tailored to the specific
needs of the learner group in question (see Richards 2001: 51–89; Huhta
et al. 2013; Johnson 2018: 199–203). This means that a needs-analysis
survey must be conducted in advance of the course before goals can be
set, the content of the course planned and teaching materials decided on.
Needs analysis involves eliciting from learners information about the
purposes for which they use English, in which contexts of use they feel
secure and in which contexts they feel they have deficits.
Needs analysis may be carried out in a number of ways (see Richards
2001: 60–63; Huhta et al. 2013: 16–32). Which way is chosen will depend
in part on the time available for conducting the analysis and evaluating
its results, the technical and human resources available as well as the
number of course participants and their availability before the course
starts. In the absence of a course designer, the course teacher will be
responsible for conducting the analysis. There are a number of well-tried
techniques:
The prospective learners may be observed in their professional or

everyday lives.
They may be interviewed.
They may be asked to complete a needs analysis questionnaire.
A survey of teachers who have previously taught the learners may be
conducted.
A survey of colleagues, employers or customers of the learners may
be conducted.
In general, the best procedure is to start with a structured questionnaire,
which should have two sections. In the first section learners should say
for what purposes they use the language in their daily lives or work
environment according to a set of categories provided. For each category
learners should indicate how often they use the language on a four-point
or five-point scale from ‘never’ to ‘very frequently’. These categories will
include receptive versus productive uses, written versus spoken uses and
can be subdivided still further, for example into formal and informal uses.
Further information can be obtained about whether English is used for
conversation, for oral presentations, for telephoning, for skyping and for
e-mailing. Uses will vary greatly according to the specific group, and all the
possible uses cannot be covered in a single questionnaire. Thus, shrewd
guesses as to possible uses need to be made and these options included. For
example, library staff may need to explain the cataloguing system to new
readers, to handle complaints and to advise readers on how to find a book,
among other needs. The final question should be an open-ended question in
which respondents can state any other uses of English they have which have
not been mentioned.
In the second section of the questionnaire learners should be given the
opportunity to say where they experience difficulties or deficits when
they are using English. This section should also consist of structured
items with the final item being an open-ended item. The structured items
should cover elements which can be specifically targeted in class, for
instance fluency, lack of specialist vocabulary, lack of discourse-linking
vocabulary, grammatical uncertainty, spelling problems, small-talk pro-
blems and perhaps also affective elements such as anxiety. However, it is
usually not effective to ask direct questions about affective matters, but
more effective to ask the respondents to recall their feelings in specific
situations. The items are thus best presented as statements which have to
be marked as true or false or marked on a four-point or five-point
frequency scale. An example statement to probe anxiety in telephone
use might be:
When I am talking in English on the telephone, I always/often/

sometimes/rarely/never feel nervous.
If the results of the questionnaire survey are unclear or surprising, then

follow-up interviews of certain respondents can be conducted. These are
best done as individual interviews rather than group interviews, since
certain dominant individuals may influence others in the group. Inter-
views can be short (just a few minutes) and should be semi-structured.
That is, on the basis of the questionnaire results, certain pre-scripted
questions should be asked (structured part), but the interviewer should
be prepared to ask unscripted follow-up questions so as to probe certain
areas in more depth (unstructured part). In cases where the teacher has
no prior knowledge or experience of the special uses of English for the
group in question, and is therefore unable to specify categories for
a questionnaire, then it may be advisable to replace the questionnaire
and semi-structured interviews with a small number of in-depth, unstruc-
tured interviews of selected learners, who can be asked to talk freely
about the uses to which they put English in their daily lives.
A quite different teaching situation concerning learner needs occurs in
Survival English (SE) courses, where the learners are often recent adult
immigrants with little or no knowledge of English. They have to learn as
quickly as possible how to cope in an English-speaking country whose
culture may be fully alien to them. Such learners may have limited
education and lack literacy skills in their native language. Added to this,
various bureaucratic procedures in the new country may be quite unfa-
miliar to them and have no counterpart in their home country. Some-
times these learners may be poor and socially disadvantaged; they may
not know anyone who speaks English and who knows the culture, so
that they have no competent person to rely on for help. They may
experience culture shock, depression or feelings of aggression. In such
cases, goals need to be defined in highly specific, situation-bound terms
(for example how to apply for official documents, how to explain pains
one has in a specific part of the body at the doctor’s, how to go
shopping). However, rather than assuming that certain needs exist and
no others, the teacher should first spend a lot of time trying to find out
about the specific linguistic and social problems of the learners by getting
them to tell their stories. In cases of very low English proficiency and in
the absence of another language as lingua franca, then recourse should be
made to mediation in the manner of Community Language Learning by
any available individuals who speak both English and the learners’ native
language (see Chapter 2).
Teaching Materials
Sometimes teaching materials are specified by the institution or by the
syllabus, often in the form of a course book; sometimes teachers have
complete freedom to choose their own materials; sometimes they may
have partial freedom: for example, they may be able to supplement
a course book with materials of their own choice or materials they have
developed themselves. Teaching materials can be broadly divided into
a number of broad classes:
didactic (contrived) materials

Didactic materials are defined as those designed specifically for
language teaching. They include written materials such as course
books, graded readers specially written for learners, grammar books
for learners and learners’ dictionaries, lists of irregular verbs,
vocabulary lists and flash cards. They also include visual materials,
such as picture-story sequences and course-book illustrations. Didac-
tic listening materials include scripted audio recordings specially
made to accompany a course book as well as recordings of individual
sounds and words for pronunciation practice.
authentic materials
Authentic materials are defined as those designed for purposes uncon-
nected with language teaching. They include written genres such as
novels, poems and newspaper articles, as well as grammar books and
dictionaries not written for foreign learners. Authentic visual materials
might include the teacher’s or the learners’ holiday photos. Also included
are audio and video recordings of songs, recitations, speeches and TV
programmes, as well as feature films and documentaries, demonstrations
from YouTube and even recordings of animal sounds or of household
things like a kettle boiling. Comic strips, cartoons and advertisements
combine text and pictures, which makes them very productive.
pseudo-authentic materials
These are materials written especially for teaching purposes which
masquerade as authentic materials. They include everything from
pseudo nursery rhymes written for teaching English to dialogues of
fictitious conversations produced by a textbook author to illustrate,
for example, giving directions or inviting someone to a party. Model
business letters with fictitious English-sounding addresses and corre-
spondents also belong to this group. Also included are scripted
audio recordings and specially shot videos for a language course.
These written and spoken texts are supposed to sound real or at least
natural, but they never do, of course.
semi-authentic materials
Semi-authentic materials are professionally produced, ‘doctored’
authentic materials. They include, for example, edited, simplified
and abridged versions of authentic texts, sometimes with vocabulary
glosses, footnotes and explanations.
teacher-produced hybrid materials
These are typically teacher adaptations of didactic materials to fit
a particular learner group in a particular situation, for example
a course-book text supplemented with a new task thought up by the
teacher or with an authentic picture from the internet, perhaps to
replace unsuitable original material. Sometimes course-book material
designed for a global market may for cultural reasons have to be
modified to make it suitable for the specific local class being taught
(Thornbury 2013; Richards 2001: 260–261).
Richards (2001: 252–254) discusses the pros and cons of using authentic
versus contrived materials. With specific reference to text-based materi-
als, didactic texts are necessarily unnatural texts. They tend to select
vocabulary and grammar to match what learners have already ‘covered’,
to use short sentences without connectors so that their discourse struc-
ture is unnatural, and, in the case of spoken contrived texts, the pace is
often slowed down considerably, which inevitably disrupts natural stress
and intonation patterns. In the early stages of learning this is the price
that has to be paid so as to ensure comprehensibility. However, it is
advisable to introduce some selected authentic texts (written and oral) as
soon as possible from lower-intermediate level onwards. Apart from the
linguistic and didactic advantages of exposure to natural language, there
are motivational advantages. Learners tend to perceive authentic texts as
‘the real thing’, find them more motivating than contrived texts of the
same difficulty level and invest more effort in working with them
provided the materials are in themselves varied and interesting (see
Peacock 1997; Sample 2015). Complete comprehension does not have
to be aimed at, and the criterion should be whether learners can under-
stand enough to feel a sense of achievement.
Distinct from teaching materials are realia, which are often used as
props for a task to be performed in the classroom. They include artefacts
from the target culture such as grocery products, coins and banknotes
(for shopping tasks), documents such as a passport or driving licence (for
tasks dealing with administration), but also items of clothing which are
specific to the target culture, such as a policeman’s helmet, or sports
equipment which is culture specific, such as a baseball bat or cricket bat
(for tasks dealing with sport). They can help to make a role-play or
simulation come alive and raise motivation levels.
Information and Communication Technology (ICT)
Online Resources
Larsen-Freeman and Anderson (2011: 199–200) emphasise that ICT repre-
sents both additional provision of resources for language learning as well as
an enhancement of the learning experience by opening up new opportu-
nities for learner autonomy. If used appropriately, ICT can make learning
English more efficient, more enjoyable, more educationally rewarding and
can ultimately enhance motivation to learn English. An approach to lan-
guage teaching which incorporates ICT is particularly well-suited to TBLT
and Co-operative Language Learning (both discussed in Chapter 2).
As with traditional teaching materials, online materials can be broadly
divided into authentic and didactic resources. Authentic online material
comprises resources not specifically produced for language learning.
Didactic online resources include online exercises, tasks, tests and gra-
phic material produced for language learners, either as a supplement to
a course book or independent of any course. They also include online
dictionaries and grammar explanations. There are various English-
teaching websites which offer complete lesson plans and materials. In
some cases corrective feedback and marking may be provided automati-
cally and the teacher can access this feedback and track learners’ pro-
gress. A further refinement is ‘adaptive learning’, whereby the software
itself tracks ongoing learner progress and level of achievement so as to
provide feedback as to which task or exercise should be done next (see
Larsen-Freeman and Anderson 2011: 200; Harmer 2015: 195–196; Dude-
ney and Hockly 2007: 117–119). The challenge for teachers is how to
integrate online resources into their teaching to didactic advantage rather
than just using them because they are there. A sure recipe for disaster is
just to hunt the internet for a lesson plan and materials which someone
has uploaded, to download it and try to use it in class without any
planning or preparation. But of course, you would not do that.
The incorporation of ICT into classroom language teaching is referred
to as ‘blended learning’. Online learning can be combined with class-
room contact teaching in three ways:
Learners can prepare for a class by using online sources and the class
might then consist of learner presentations, discussion and evalua-
tion of the online sources.
Learners can do online work as a follow-up to a class in the form of
homework. This might involve accessing supplementary authentic
material on the internet or performing additional language tasks from
online didactic resources.
Learners may be asked to access either authentic online material or
didactic online material such as exercises, tasks, grammar informa-
tion or dictionary entries at certain points in a lesson, either as
a planned structural element of the lesson or as problems crop up
(see Harmer 2015: 204–206).
Some of the beneficial uses of ICT in the EFL classroom include the
following (adapted from Stanley 2013: 1):
A vast store of information of all sorts, including information about

English, can be accessed.
Almost all varieties imaginable of spoken and written English can be
accessed.
An almost unlimited choice of authentic English texts to read, view
or listen to is available.
A great range of graphic, picture and photographic material is
available.
A great variety of EFL teaching materials, tasks, exercises and tests
for all levels is available.
Opportunities for communication with people all over the world
exist.
Quick and efficient production, editing and publishing of written,
spoken and multimedia texts, with opportunities for collaborative
authorship, is possible.
A community of English language users/learners within and beyond
the classroom can be created.
Virtual Learning Environments (VLEs) help to manage and organise
learning.
Opportunities for practice in speaking, listening, reading and writing
with specific-task focus are available.
Presentation Software
Presentation software such as ‘PowerPoint’, ‘Keynote’, ‘Impress’, ‘Pre-
sentations’ and ‘Prezi’ enables teachers and learners to do multimedia
presentations which incorporate text, pictures, graphics, video and sound
(Harmer 2015: 201; Scrivener 2011: 337). Online software such as
Popplet (www.popplet.com) allows the creation and presentation of
mind maps and there are online materials pools such as ‘LearningApps’
(learningapps.org), which allow teachers and learners to present and
adapt materials easily and attractively, work with these online individu-
ally or collaboratively and share them with others (see Grimm et al.
2015: 257–260). Initially, creation and presentation will be done by the
teacher but the emphasis should be on encouraging learners to progres-
sively take over responsibility themselves and to collaborate on making
presentations, performing tasks and creating and adapting their own
learning materials. Scrivener (2011: 338–339) suggests various sorts of
topics on which learners may present. In this way learners become
increasingly autonomous and may eventually decide on the content of
the course in collaboration with and under the advice of the teacher so
that a ‘negotiated syllabus’ is developed (Clarke 1991).
Here are five ‘Dos’ and three ‘Don’ts’ for using presentation software
and training learners in how to use it (see also Scrivener 2011: 338–339):
DO’S
Do vary the format and style of presentations rather than always

working with a fixed format.
Do make your presentations interactive by having blanks in the
slides for the audience to complete at various points.
Do use options for concealing text and for reading words one at
a time and synchronise this with challenging learners to guess what
comes next.
Do use the animated text option for presenting grammar, so as to
show changes in word order and addition of inflections, for example
in question and negative forms and in transformations from active to
passive sentences
Do use carefully chosen pictures rather than lengthy explanations for
presenting new vocabulary.
DON’TS
Don’t allow the slides to be a complete written version of the

presentation but rather a form of scaffolding in word and image to
be fleshed out by talk.
Don’t read out a presentation from slides, as this will make it
difficult for you to be flexible and respond to questions.
Don’t overload slides with clipart images which do not add anything
to the message.
Virtual Learning Environments (VLEs)

A VLE, sometimes called a Learning Management System (LMS), is
provided by platforms such as ‘Blackboard’, ‘Canvas’ and ‘Moodle’,
which combine the functions of e-mail, wikis and blogs as well as other
features such as forums and chat rooms. Stanley (2013: 17) recommends
student blogs for written homework, a ‘latest news’ forum and also forums
for learners to ask questions and receive teacher feedback. Class blogs,
where all class members have posting rights, encourage a sense of learner
community but student blogs and tutor blogs may also be employed, with
restricted posting rights. A wiki can be used as the class website and texts,
tasks, exercises and links to websites can be uploaded to it (Ur 2012:
240–241). A VLE can be used to manage either a classroom contact
course or an online course. Learners can see course content, including
links to websites or files and can do exercises and tests online. Via tracking
facilities, the course teacher can see who has logged in when and what
activities learners have done. Some exercises and tests can be automatically
marked and the results made available to the learner, while the teacher can
also assess other work and provide the learner with feedback. This can all
be done so that each learner can view only his or her mark, thus ensuring
privacy (Dudeney and Hockly 2007: 137–138, 153–154; Ur 2012: 240–241).
Online Tools
Various tools may be used either within a VLE or without one.
E-mail and Blogs

E-mail or blog exchanges with other classes at home or abroad may be
organised (Dudeney and Hockly 2007: 67–70; Stanley 2013: 34; Scrivener
2011: 340–341). Harmer (2015: 202), however, cautions that learners tend
to lose interest in responding to one another’s blogs unless they are given
direction and new impetus from time to time. Stanley (2013: 33) suggests
introducing ‘mystery guests’ whose identity learners have to guess.
Dudeney and Hockly (2007: 86–93) offer detailed advice on setting up
audio and video blogs.
Wikis
Wikis can be used for collaborative writing, interactive editing of written
tasks, for online discussion and for posting videos (Larsen-Freeman and
Anderson 2011: 202; Ur 2012: 240). This encourages social interaction
via writing (Larsen-Freeman and Anderson 2011: 201). Dudeney and
Hockly (2007: 96–98) describe a simple collaborative writing activity
using a wiki in which pairs of learners write a brief description of
a famous person, but deliberately make humorous factual mistakes,
which other learners have to spot and correct.
Audio and Video Files

YouTube (www.youtube.com) and other sites offer a wide range of materi-
als for listening and viewing. These include both authentic materials not
produced for language learners and didactic materials specially produced
for language learning. Relevant sites mentioned by Kiliçkaya (2018: 1363)
include English Language Listening Lab Online (www.ello.org), ESL lab
(esl-lab.com), ESL bits (www.esl-bits.net), Teacher Tube (www.teachertube.
com), Breaking News English (www.breakingnewsenglish.com). Real Eng-
lish (www.real-english.com) offers a vast store of authentic audio and video
material in English for listening practice in or out of class. There are other
useful sites for authentic listening and viewing. Kiliçkaya (2018: 1363)
mentions Voscreen (www.voscreen.com), a website which enables learners
to view short video clips from diverse sources.
This great repertoire of audio and video material can function within
the framework of blended learning to individualise the traditional listen-
ing comprehension class. For example, working in small groups, learners
can listen to a podcast or view a video out of class and draft comprehen-
sion questions for it (Walker and White 2013: 34). They can then
present the video to the class, who have to answer the questions. For
pronunciation practice rather than listening practice a wide range of
native and non-native varieties of English can also be found. A relevant
website is www.audio-lingua.eu (Kiliçkaya 2018: 1363).
Learners can also produce their own podcasts using software such as
‘Audacity’ (Dudeney and Hockly 2007: 98–102; Walker and White 2013:
36). A number of sites are available which enable learners to listen to
messages from classmates and the teacher, but also to record themselves,
for example, giving a short talk to fulfil an assignment. Kiliçkaya (2018:
1364) suggests this facility is valuable for anxious or shy learners who can
prepare their talk in advance and do not have to perform before their
classmates. Relevant sites are Audioboo (http://audioboo.fm), Vocaroo
(www.vocaroo.com) and Voxopop (www.voxopop.com).
Children’s songs and rhymes available in abundance on YouTube can be
used in class for young learners to sing along with and/or for listening
comprehension. One can usually find the written text of the song,
a rendition of the song – often with subtitles – and even a visual track which
can be helpful for comprehension and for raising the appeal of the video. For
example, ‘Ten Green Bottles’ is available with accompanying animated gra-
phics so that a bottle falls off the wall each time at the appropriate point in the
song. This can be used to promote prediction skills while listening. Concern-
ing poetry rather than songs, various websites provide readings of poems,
often by professional actors, sometimes even by the poet in person. Supple-
mentary material about the poem and the poet is also often easily accessible.
Sites such as YouTube are also invaluable for language-teaching pur-
poses for the countless videos of instructions posted by ordinary people
on how to operate various devices. These constitute examples of speech
accompanied by action. Often, the instructions or descriptions are
supported by action or by diagrams or pictures. This means that learners
can infer the meaning of words or phrases from the actions and from
their background knowledge of the topic.
Online Poster Walls

Online poster walls such as ‘Padlet’ (padlet.com) (see Harmer 2015: 202)
and ‘Glogster’ (edu.glogster.com) (see Grimm et al. 2015: 259; Stanley
2013: 36, 47) enable learners to post their own comments as well as
pictures and videos. Scrivener (2011: 341) suggests online poster walls can
be used for collecting birthday greetings, brainstorming ideas or collecting
feedback on a suggestion. Word-cloud creators, such as ‘Tagxedo’ (www.
tagxedo.com), ‘Wordle’ (wordle.net) and ‘Worditout’ (worditout.com),
display words in a text as a graphic illustration of their relative frequencies
and can be used for brainstorming a short text in a pre-reading activity (see
Grimm et al. 2015: 258–259; Scrivener 2011: 345–346; Stanley 2013: 101).
Video Communication Platforms

‘Skype’, ‘FaceTime’ or ‘AdobeConnect’ can be used for learner–learner
communication but also for teacher–learner communication and for
individual tutoring. They can also be used for presentations by guest
speakers with the possibility for interaction between speaker and audi-
ence (see Harmer 2015: 195). Dudeney and Hockly (2007: 71–85) offer
detailed suggestions for setting up a chat session with another class.
Multimedia Text-creation Software
Multimedia storytelling is one of many activities which can be performed
collaboratively. Learners can write stories which they illustrate with photos
from the internet using the ‘Flickr’ website (www.flickr.com). The Microsoft
software programme ‘Photo Story 3’ enables learners to add pictures,
recorded voice and background music to their stories (see Cheung and Lee
2013). The site ‘Storybird’ (storybird.com) allows learners to use pictures to
make stories, ‘Toondoo’ (www.toondoo.com) enables the creation of comic
strips and ‘Voki’ (www.voki.com) makes it possible to create avatars (‘vokis’)
which speak the words that learners type in (see Harmer 2015: 202–203).
Webinars
Other useful resources for advanced classes are online seminars or
lectures (webinars). These can be viewed in real time, and viewers can
communicate with other viewers via a chat box or text box. They are
also often recorded so that they can be viewed at any time afterwards,
for example the TED Talks, for advanced learners (see Scrivener 2011:
343; Harmer 2015: 206–207). Stanley (2013: 106) makes suggestions for
how a class can organise their own reading-for-pleasure webinar using
virtual classroom software such as ‘Wiz-IQ’ (www.wiziq.com).
Social Media Platforms

Social media platforms offer possibilities for learners to cooperate outside
the classroom in activities such as discussing a topic, making suggestions
about a project or producing a piece of writing. Learners can post to
a public social network page such as the British Council ‘Learn English’
pages for adults, teenagers and children, respectively (see Hockly and
Dudeney 2014: 48). Teachers can also liaise with their learners outside
the classroom on homework tasks (see Harmer 2015: 202).
Virtual Worlds
Virtual worlds such as ‘Second Life’ provide simulated reality within which
task-based learning can be done. Assuming another identity may help
learners to overcome anxiety and express themselves more freely than they
would otherwise have done. (See Harmer 2015: 195; Scrivener 2011:
347–348; Stanley 2013: 158–159; Dudeney and Hockly 2007: 154–156).
Online Dictionaries, Corpora and Concordance Programmes

Online dictionaries and apps for mobile devices are a great advance over
traditional print dictionaries and first-generation electronic dictionaries.
They are supplemented by the availability of various online corpora of
spoken and written English and, when used with a concordance pro-
gramme, are particularly useful for researching the collocation possibili-
ties of a word. Stanley (2013: 53–59) has a number of suggestions for
using concordance programmes as a basis from which to teach syno-
nyms, antonyms, polysemy and collocation (see also Dudeney and
Hockly 2007: 103–112). Kiliçkaya (2018: 1362) provides the internet
addresses of some major English dictionaries available online and notes
two great advantages of electronic and online dictionaries over conven-
tional print dictionaries. The first is that learners can listen to the
pronunciation of words and the second is that examples of the word in
use are available. However, even without a concordance programme, just
typing an unknown word or phrase into a search engine is sometimes
good enough. This usually generates lots of examples which help make
clear what the word or phrase means (see Harmer 2015: 203).
Internet Research
Learners can be sent on internet searches in the form of a ‘treasure hunt’
in which they have to answer a series of information questions. Some-
times one question can lead to another so that the answer to the previous
question is necessary to proceed to the next. Learners may need training
in how to use keywords effectively (see Dudeney and Hockly 2007:32–34;
Scrivener 2011: 342–345).
Dudeney (2007) suggested various activities based on accessing infor-
mation. Many of these have been part of the stock-in-trade of CLT for
many years, but are enhanced by the possibility of quickly accessing
information online, editing documents, adding pictures and printing out
material. The activities suggested include accessing song texts and photos
of the singer from the internet, using the ‘Time’ website of 20th-century
celebrities. Using the same site, celebrities can be invited to a virtual
dinner party. Other useful websites for creating tasks include weather
websites and the Yahoo headlines site. ‘Crayon’ can be used to create
a newspaper (Dudeney 2007: 49–58, 94–95). Another activity involves
writing a film review based on reviews available online (Dudeney 2007:
58). Learners can also design a holiday tour for various profiles of tourist
or plan their own holiday (Dudeney 2007: 61–62, 82). They can decide to
see a particular film, do some virtual shopping, search for an apartment,
design a holiday poster or view luxury apartments or houses (Dudeney
2007: 83–94, 104–105). One just has to find appropriate internet sites.
However, Littlewood’s (2011: 553) criticism, mentioned in Chapter 2
when discussing CLT, of make-believe activities of the ‘planning a party
that will never take place’ variety remains valid, regardless of whether
they are internet supported or not.
Much more promising for more advanced learners working within
a topic-based syllabus are internet-based research projects (see Dudeney
and Hockly 2007: 44–61). For example, a textbook unit for advanced
learners on The Spanish Armada can be greatly enhanced in the spirit of
‘blended learning’ if, after having read the textbook passage, learners are
invited to work collaboratively outside class, find out why the invasion
attempt failed and then report back in the next class and present their
findings. They will find quite different reasons advanced by historians of
different persuasions on the internet. A conflict can be engineered by
asking some groups to focus on the British War Museum’s version,
others on history schoolbook versions in Britain, others on the Spanish
viewpoint. Such research can be combined with writing a timeline for the
event in question using timeline software such as ‘X Timeline’ (see
Stanley 2013: 141–142). Another possible topic for research within
a course about the American Wild West might be the shoot-out at the
OK Corral in Tombstone, Arizona, immortalised in several Westerns,
which often pay scant attention to the geographical and historical facts.
Learners can find out how it really was.
Electronic portfolios constitute a possibility for learners to collect and
present work they have done in various internet-based projects over
a whole course or school year, and they can also be used for course
assessment. They differ from traditional portfolios only in the respect
that they are stored electronically and can include video and audio
material, blogs and websites as well as documents (see Dudeney and
Hockly 2007: 119–122).
Hand-Held Devices
Hockly and Dudeney (2014: 29–30) list various sorts of apps for hand-
held devices which can be profitably used for teaching and learning
English. Some of the simpler activities are based mainly on the text or
note-taking function of mobile devices. These include various dictation
activities, online notice boards, word clouds, following celebrities on
Twitter, text messaging, emoticon conversations and mind mapping
(Hockly and Dudeney 2014: 41–55).
At a slightly greater level of complexity are activities which involve
taking photos as a basis for language use, for example close-ups taken
from unusual angles of everyday objects, treasure hunts, making
a picture dictionary with ‘Flickr’, using photo-manipulation data to
make an alternative life story, making a story based on a series of
random pictures, making a photo collage to put in a time capsule and
annotating photos, to name only some possibilities mentioned by
Hockly and Dudeney (2014: 58–68). Even the conversation class about
what learners did at the weekend can be enlivened by asking learners to
bring in pictures they have taken of what they did (Jarvis 2015).
At a still-higher level of complexity are activities suggested by Hockly
and Dudeney (2014: 69–81) which require learners to make audio record-
ings. One activity is to record sounds from everyday life, such a kettle
boiling and use these as the basis for various sorts of quiz (idem: 72).
Another activity involves using a voice search engine to find out infor-
mation about famous inventions as quickly as possible (idem: 73). Yet
another activity involves using a speech-to-text/dictation app to practise
pronunciation (idem: 74).
Even more complex are activities involving video-recording. Hockly
and Dudeney (2014: 83) suggest that learners can, for example, create
a silent video tour of their own home: They can then present this in
class, providing a spoken commentary. This can provide the basis for
various spoken activities. Afterwards, using an audio voiceover app, they
can record their commentaries and upload the video to a class blog or
wiki page so that the videos are available for the whole class. A more
ambitious variation would be for learners to produce a video guide to
their own town (idem: 88). Other activities include writing subtitles to
a film with a non-English soundtrack using Bombay TV (www.grapheine.
com/bombaytv) (idem: 85) and producing a multimedia trailer for a book
using a video-production app such as Animoto (idem: 92).
Classroom Interaction
Teacher and Learner Classroom Roles

As already mentioned at the outset of this chapter, the classroom is an
acquisition-poor environment in which opportunities for varied expo-
sure and communicative interaction are both quantitatively and qualita-
tively limited. Classroom teaching therefore aims to compensate for
these deficits with a series of pedagogical interventions. These include
the syllabus itself, needs analysis, teaching materials and ICT. However,
the major instrument of intervention is the teacher in the classroom,
who fulfils at least the following six roles:
a provider of language knowledge

a language model
a facilitator of ‘learning by doing’
a provider of feedback (positive and negative)
a link to the target language culture
a motivator
The emphasis given to each of these roles varies according to the

teaching method or approach adopted, the learner group, the age of the
learners, their proficiency level, their learning goals and their type and
intensity of motivation. The teacher’s role also has to be adjusted to the
learning environment represented by the school, the educational system
and the wider cultural background of the learners (see Scrivener 2011:
13–19; Ur 2012: 16–18). However, in all teaching environments learners
must be provided with three opportunities:
to receive explanations about the language

to practise listening to, speaking, reading and writing the language
to develop both accuracy and fluency in speech and writing
Learners themselves have to take advantage of these opportunities and also

seek out additional learning opportunities outside the classroom. They
cannot be spoon-fed into acquiring English. All learners therefore have to
develop self-direction and the ultimate goal must be learner autonomy. The
self-directed classroom learner is to a greater or lesser extent:
an explorer of the language

a gatherer of information about the language
a builder and tester of hypotheses about the language
a language experimenter in conversation with fellow learners
an initiator, manager and regulator of language practice activities
a solver of communication problems which arise in conversation
an evaluator of language which is encountered in speech or writing
an explainer of the language to fellow learners
Input Modification and Interactional Modification

When speaking to learners of limited proficiency, people tend, quite
naturally, to modify their language in certain predictable ways. Skilled
teachers are good at tuning the degree of language modification to the
particular group of learners. The goal is to modify speech sufficiently so
as to ease comprehension and thus promote learning while keeping the
language as natural as possible. Saville-Troike and Barto (2017: 112–114)
mention a number of techniques, including the following:
regular and repetitive syntax

simple morphological structure
simple vocabulary
clearly defined pauses at clause and phrase boundaries
speech rate slowed down slightly
careful articulation
a level of loudness just slightly above that of normal speech
slightly stronger than normal stress on key words
topicalisation (topic mentioned first, followed by a comment as in,
‘Vienna, it is a nice city.’)
full forms such as ‘I will’ rather than contractions such as ‘I’ll’.
Saville-Troike and Barto (2017: 115) also identify techniques of ‘interac-
tional modification’. These involve devices such as paraphrase, repeti-
tion, expansion and elaboration. Unlike input modification, interactional
modification does not make for unnatural language at the phonological,
grammatical and lexical levels. This makes it a more useful teaching tool.
For example, a teacher can retrace and modify an utterance he or she has
already produced if it seems that the learner has not understood, as in
the following (invented) exchange:
Teacher: It will be an adventure.

Learner: at venture? (request for clarification)
The teacher can respond in various ways, using discourse-modification

tools. If one does not work, others can be tried.
Repetition
An adventure
Paraphrase
It will be an exciting holiday.
Expansion and elaboration
It will be very exciting, a big challenge. There is a mountain to climb
and you can go canoeing and hang-gliding.
These basic strategies can also be used where discourse modification is

made prophylactically, as in the following (also invented) example:
Teacher: It will be an adventure, an adventure (repetition), an exciting

adventure (repetition with expansion), yes, a big challenge (para-
phrase), a really exciting holiday (re-paraphrase). There will be
lots of exciting things to do, like mountain climbing, canoeing
and hang-gliding (expansion and elaboration).
Assuming the learners are following the gist of the discourse, the teacher
may continue in the same fashion, anticipating the next problem and
adopting the same strategy:
Teacher: You can go hang-gliding if you dare, if you dare (repetition), if

you are a brave and daring person (repetition with variation).
Hang-gliding (repetition) is only for brave and daring people
(paraphrase/repetition).
As part of their development in the direction of autonomy, learners

should be encouraged to make requests for clarification when they do
not understand either the teacher or another learner. When communi-
cating with other learners, they should also be prepared, if necessary, to
modify their original utterance if this is not understood by their inter-
locutor. This is important training for future real-world communication,
including future communication with less proficient speakers of English
than themselves.
Learner Cooperation and Self-Direction

For young beginning learners a considerable measure of teacher direction
is required, but training in how to learn autonomously should be
introduced as soon as possible (see Pinter 2017: 111–126). Learner
cooperation, self-direction and ultimately autonomy can be encour-
aged by:
clearly structured lessons, for example the three-phase lesson struc-

ture of ‘plan’, ‘perform’ and ‘reflect’ (PPR)
building-in certain ritual practices, for example having a language
watchdog, timekeeper, co-ordinator and secretary in the ‘perform’
phase. Rituals, however, should not be overdone, lest they impart an
unwanted rigidity to classes and they are best set up in cooperation
with the learners.
including choices for learners at various points in the lesson, such as
with whom they work, which material they use, which method they
employ to perform a task, whether they do a certain task in class or
at home
encouraging learners to modify and re-use materials as a springboard
for further activities (see Ur 2012: 203–209). This may entail expan-
sion, reduction, transformation into another mode, change of writer
perspective or re-contextualisation. Materials may be pooled and
montages may be made from various source materials for a new
task or activity, initially under the close guidance of the teacher, but
later in self-directed fashion.
Provision of Feedback on Error

For language learners there is always a discrepancy between cognitive and
linguistic development. This means that learners frequently want to
express ideas which they can easily express in their native language but
for which they lack the linguistic means in the foreign language. This
may lead them into error of various sorts, and this is good. Learners
need to make errors in order to test out hypotheses about the language
and actually learn from their errors, as was said in Chapter 1. However,
this does not mean that all errors have to be corrected. Firstly, if one
attempted to do this, much time for learners to speak would be lost,
communication would be severely disrupted and learners would in any
case be unable to learn from the corrections because of cognitive
overload. In addition, continual error correction might well be demoti-
vating. When language learners talk to native speakers in natural speech
situations, not many learner errors are corrected, precisely because this
would mean continually interrupting learners. Native speakers tend
usually to correct factual error. They are also quite likely to correct
irrelevant utterances, for example if a learner gives an answer which is
not the answer to the question asked. Vocabulary error is much less
likely to be corrected and syntactic error is rarely corrected. Errors of
omission of grammatical morphemes are very rarely corrected in natur-
alistic language use, probably because they do not usually impede under-
standing (Chun et al. 1982).
Errors occur on a spectrum from errors of competence at one
extreme to performance slips at the other, as was said in Chapter 1.
Errors of competence represent gaps or inaccuracies in the learner’s
knowledge of English – wrong ideas about the language, as it were.
Performance slips are quite different. The learner knows what is right,
but the wrong thing ‘comes out’ by mistake, due to incomplete auto-
matisation. However, an error such as *‘childs’ (for ‘children’) may be
a competence error for a beginner but a performance slip for an
advanced learner. In the latter case, correction is unnecessary and may
serve only to distract and even embarrass the learner. At the other
extreme, correction of competence errors which involve language far
beyond the learner’s current level of proficiency is unlikely to be
effective either. However, many errors learners produce are strung out
at various points along the spectrum from competence error to perfor-
mance slip. This middle ground represents more or less unstable
elements of the language which are as yet not firmly established in
competence. It is these middle-ground errors on which error correction
should focus. (For a review of various ideas on which errors to correct,
some of them different from those put forward here, see Sheen and
Ellis 2011: 599–600.)
Learners can benefit most from correction of those middle-ground
errors about which they are already uncertain and suspect there may be
something wrong with what they have just said. As part of their devel-
opment towards autonomy, learners should themselves gradually
become able to actively seek out support from the teacher when they
encounter difficulty while speaking, so that the teacher becomes less an
error corrector and more a prompter and provider of scaffolding. In
cases where learners have been considering which of two forms to use,
for instance, they may show their uncertainty by tone of voice, for
example giving a rising intonation to a word, to imply, ‘Is that right?’ or
make a mute appeal to the teacher by hesitating and using an appropriate
facial expression. In such cases of mute appeal, what the learner has said
may actually be fully acceptable. In this case he or she will nevertheless
benefit from brief positive feedback, even if this is just in the form of an
approving nod. Positive teacher feedback (praise), for instance for an
idiomatic turn of phrase, is in general immensely important not only for
the speaker but also for the whole class, who are likely to pay attention
to and perhaps take in the positive language feature. However, such
praise can become inflationary and should not be overdone.
The more pressure learners are under, the more errors they will
make. The more cognitively demanding the speaking task as a whole
is, the less surplus attentional capacity will be available for issues of
correctness and the likelihood of error will be greater. In a given speech
situation the learner may be focussing simultaneously on getting the
past tense form of a verb correct, choosing the right preposition and
thinking about how to formulate the following clause, so that due to
cognitive overload, error occurs. Furthermore, factors internal to the
learner, including level of arousal, attitude to the interlocutor as well as
anxiety levels, will also influence available attentional capacity and the
likelihood of error occurring. Anxiety was discussed in Chapter 3, and
in situations where learners are in a state of high anxiety, error correc-
tion should in any case be abandoned altogether (MacIntyre and
Gardner 1994).
To summarise: online classroom correction of oral error is a delicate
art and wholesale error correction has the potential to do a lot of
damage, particularly if performed insensitively. Error correction may be
counterproductive for the following reasons:
Learners may not need the correction if it refers to a performance

slip.
The correction may be too far above the learner’s current compe-
tence so that he or she is not yet ready to integrate it into his or her
learner language.
In any case, the learner may be focussed on what he or she is trying
to say and may not take in, understand or remember the correction
due to lack of free attentional capacity.
Learners may become embarrassed by correction and their anxiety
levels may rise so that free attentional capacity is further reduced.
The learner may misunderstand the correction, for example if the
learner incorrectly pronounces the word ‘iron’ with an intervocalic
‘r’ and is corrected, then he or she may in future also pronounce
‘irony’ without an intervocalic ‘r’, thus introducing a new error.
Sometimes the correction itself may be mistaken, that is, does not
correspond with what the learner wanted to say, so that the learner
will be confused.
Learners may be distracted from what they are trying to say, their
planning interrupted, and they may not be able to proceed.
Interruptions reduce the chance of learners to talk at length and thus
gain essential practice in planning longer stretches of speech (Holley
and King 1974).
Learners may adopt risk-reducing behaviour and in future say only
what they are sure is correct.
However, not all errors will necessarily disappear by themselves so that

appropriate error correction in the classroom is valuable (Swain 2000: 99;
Swain and Lapkin 1989; Sheen and Ellis 2011: 606; Keck and Kim 2014:
152). In particular, efforts must be made to ensure that a classroom pidgin
does not develop, involving formal and functional simplification as well as
transfer from the first language. Therefore, each language course needs to
strike its own individual balance between accuracy and fluency work.
Since trade-offs operate between accuracy and fluency in any given task,
it may well be preferable to separate fluency-based work and accuracy-
based work in class, with the main emphasis being on the former.
Error correction may be explicit or implicit. Explicit correction
involves direct teacher intervention and may be accompanied by an
explanation and/or a request for the learner to repeat a correct version
of the erroneous part of the utterance. In fluency-based sessions implicit
correction is often preferable: the teacher waits until the learner’s turn to
speak is over and then takes up the conversation, apparently in natural
fashion, and integrates into his or her utterance a corrected ‘recast’ of the
erroneous portion of the learner’s utterance (Lyster and Ranta 1997;
Lightbown and Spada 2013: 139–143; Keck and Kim 2014: 152–1557).
Loewen (2012: 25) cites many studies which find that recasts are the most
common form of classroom correction. This is obviously a more eco-
nomical and less intrusive procedure than direct correction, and it is
additionally less face-threatening. For the procedure to be effective,
however, learners may need to be sensitised to recognising recasts as
indirect feedback on error since there is always the danger that learners
may either not notice the recast or interpret it as mere stylistic variation
(see also Saville-Troike and Barto 2017: 117).
For accuracy-based sessions, by contrast, explicit correction may well
be appropriate. There are a number of techniques available:
The teacher says the correct version either immediately or at the end
of the utterance.
The teacher says the correct version and asks the learner to repeat it.
The teacher prefaces the correction by a comment such as, ‘No, you
can’t say that, you must say …’
The teacher tries to elicit the correct version, possibly giving various
clues to the correct form (‘you must use the past tense’, ‘no, it’s an
irregular verb’).
The teacher appeals to the whole class to supply the correct version.
The teacher not only corrects the error but adds a brief explanation.
The teacher not only corrects the error but tries to elicit an explana-
tion either from the learner or from the whole class, perhaps
supplying clues.
The teacher records a selection of errors made during the activity
and then afterwards either corrects them explicitly or tries to elicit
the corrections from the class. Explanations may also be given by the
teacher or elicited from the class.
Note 1: The advantage of elicitation procedures is that the cognitive

activity involved in retrieving the correct version and/or explaining why
it is correct helps the learner to retain it better.
Note 2: Concerning direct correction of pronunciation in accuracy-
based sessions, it is important that learners repeat the correct form.
Merely hearing the correct pronunciation is no guarantee that the
correction will be incorporated into production since the neuro-
muscular articulatory skill may not be available and will need practising.
However, if a learner cannot say the word correctly at, say, the third
attempt, then one should move on, rather than persisting too long and
possibly embarrassing him or her.
Note 3: Concerning grammatical and lexical mistakes, a brief explana-
tion given immediately can be useful for the learner, but longer explana-
tions should be postponed to the end of the activity or even to a special
form-focussed teaching session.
Note 4: If the technique of eliciting an explanation is adopted, one
should acknowledge the partial validity of incomplete explanations given
by learners and then supplement them or build on them rather than just
dismissing or ignoring them, which can be demotivating.
For further discussion of error-correction techniques, see Sheen and Ellis
(2011: 594), Loewen (2012: 25–26) and Lightbown and Spada (2013: 139–145).
Some Rules of Thumb for Classroom Interaction

To conclude this section, here are some rules of thumb for managing
classroom interaction, which you might like to reflect on and discuss
with one another:
DO’S
Do give learners ample time to respond to a question.

Do rephrase the question if there is no response after a certain time.
Do always acknowledge the truth value of a learner’s utterance first
of all, even if it is error-ridden.
Do expand the learner’s utterance to make it acceptable if a student
is struggling.
Do prompt the student with cues.
Do give students practice in speaking at a stretch so that they can
practise discourse planning.
Do be flexible and tolerant of learner answers that are not exactly
your own model answer.
Do incorporate vocabulary and structures which students themselves
have just produced into your own discourse and into the language of
the lesson.
DON’TS
Don’t be afraid of silence while learners are thinking.

Don’t necessarily make the student repeat your correction of gram-
mar or lexis.
Don’t give lengthy explanations which disrupt the conversation.
Don’t interrupt students unless absolutely necessary.
Don’t correct surface morphological errors in cases when you will
only be telling the learner what he or she already knows anyway.
Don’t react to error by just shaking your head or saying, ‘No.’ (This
is potentially anxiety-provoking for the learner, who will realise he
or she has done something wrong, but will not know what.)
Don’t react to error by saying ‘Again.’ This is highly ambiguous and
could from the learner’s point of view mean either ‘It is wrong’ or
‘Speak more clearly’, for instance.
Food for Thought

Q. 1. Look at the Table of Contents of some course books and decide
which of the syllabus types mentioned in the chapter underlies the
course book. If it is a multi-track syllabus, look at some of the
units and see if you can decide which strand is the primary one.
Q. 2. Imagine you are to teach a class of ten-year-old children in a European
country who are just about to start learning English for two one-hour
lessons per week at school. What rough goals would you set for the
first month of teaching? (There is no official syllabus available yet.)
Q. 3. In her study of classroom interaction in Hong Kong, Amy Tsui
found teachers sometimes tended not to give learners enough time
to think before responding to a question, for example in a reading-
comprehension lesson. Teachers tended to be intolerant of silence.
One of the teachers in the study said:
to me silence is the result of [the] teacher’s inertia; when silence

occurs, it means [the] teacher is not making the lesson productive
enough for students to learn …
(Tsui 1995: 86)
What would you say to this teacher? Think back to Chapter 2.
From what you learned there, can you think of any language-
teaching method and theory of language learning which might
underlie the ideas expressed by the teacher?
Q. 4. ‘Like you baseball?’ is an example of an error in question forma-
tion that learners at a certain stage of development sometimes
make. What has the learner erroneously assumed about question
formation in English? How might you react to the error in:
a) a classroom conversation session (fluency-based)?

b) an accuracy-based grammar practice session?
Q. 5. (Group task) Read Cheung and Lee (2013), who offer guidance and
advice on using ‘Photo Story 3’ for getting learners to perform digital
storytelling. Access either ‘Photo Story 3’ and/or some of the other
sites mentioned for multimedia text-creation software (or any others
you know). Decide on a suitable learner group, specifying age and
proficiency level, set goals and plan a digital storytelling lesson. Try it
out by doing it yourselves and discuss any problems which arise.
Consider how well it fulfils the goals you have set.
Further Reading
Gray, J. (ed.) (2013). Critical Perspectives on Language Teaching Materials. London:
Palgrave Macmillan.
Pawlak, M. (2015). Error Correction in the Foreign Language Classroom: Reconsidering
the Issues. Heidelberg: Springer.
References
Bruner, J. (1977). The Process of Education. Second, revised edition. Cambridge,
MA: Harvard University Press.
Cheung, O. and Lee, I. (2013). From story writing to digital storytelling. Modern
English Teacher 22/1: 48–54.
Chun, A., Day, R., Chenoweth, N. and Luppescu, S. (1982). Errors, interaction
and correction: a study of native-non-native conversations. TESOL Quarterly 16/
4: 537–547.
Clarke, D. F. (1991). The negotiated syllabus: what is it and how is it likely to work?
Council of Europe. (2001). Common European Framework of Reference for Languages:
Learning, Teaching, Assessment. Cambridge: Cambridge University Press.
Dudeney, G. (2007). The Internet and the Language Classroom. Second edition.
Cambridge: Cambridge University Press.
Dudeney, G. and Hockly, N. (2007). How to Teach English with Technology. Harlow:
Pearson Education.
Grimm, N., Meyer, M. and Volkmann, L. (2015). Teaching English. Tübingen,
FRG: Narr Francke Attempto.
Pearson Education.
Hockly, N. and Dudeney, G. (2014). Going Mobile: Teaching with Handheld Devices.
Peaslake, Surrey: Delta Publishing.
Holley, F. and King, J. (1974). Imitation and correction in foreign language
learning. In Schumann, J. and Stenson, N. (eds.), New Frontiers in Second
Language Learning. Rowley, MA: Newbury House, pp. 81–89.
Huhta, M., Vogt, K., Johnson, E. and Tulkki, H. (2013). Needs Analysis for
Language Course Design: A Holistic Approach to ESP. Cambridge: Cambridge
University Press.
Jarvis, H. (2015). From PPP and CALL/MALL to a praxis of task-based teaching
and Mobile Assisted Language Use. TESL-EJ 19/1: 19.
Keck, C. and Kim, Y. (2014). Pedagogical Grammar. Amsterdam and Philadelphia:
John Benjamins Publishing Company.
Kiliçkaya, F. (2018). Information and Communications Technology (ICT) in
Listening Instruction. In Liontas, J. I. (ed.), The TESOL Encyclopaedia of English
Language Teaching. Volume III: Teaching Listening, Teaching Speaking and Pro-
nunciation. Hoboken, NJ: Wiley, pp. 1360–1366.
Larsen-Freeman, D. and Anderson, M. (2011). Techniques and Principles in Lan-
guage Teaching. Third edition. Oxford and New York: Oxford University Press.
Lightbown, P. M. and Spada, N. (2013). How Languages are Learned. Oxford:
Loewen, S. (2012). The role of feedback. In Gass, S. M. and Mackey, A. (eds.), The
Routledge Handbook of Second Language Acquisition. London and New York:
Routledge, pp. 24–40.
Long, M. (2015). Second Language Acquisition and Task-Based Language Teaching.
Malden, MA: Wiley Blackwell.
Lyster, R. and Ranta, L. (1997). Corrective feedback and learner uptake: negotia-
tion of form in communicative classrooms. Studies in Second Language Acquisi-
tion 20: 37–66.
MacIntyre, P. D. and Gardner, R. C. (1994). The subtle effects of language anxiety
on cognitive processing in the second language. Language Learning 44: 283–305.
Peacock, M. (1997). The effect of authentic materials on the motivation of EFL
learners. ELT Journal 51/2: 144–156.
Richards, J. C. (2001). Curriculum Development in Language Teaching. Cambridge:
Richards, J. C. (2013). Curriculum approaches in language teaching: forward,
central and backward design. RELC Journal 44/1: 5–33.
Sample, M. G. (2015). Authentic materials: a motivational tool? Education and
Linguistics Research 1/2: 100–119.
Saville-Troike, M. and Barto, K. (2017). Introducing Second Language Acquisition.
Teaching. Third edition. London: Macmillan.
Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics 10/3:
209–231.
Sheen, Y. and Ellis, R. (2011). Corrective feedback in language teaching. In
Hinkel, E. (ed.), Handbook of Research in Second Language Teaching and Learning.
Volume 2. New York and London: Routledge, pp. 593–610.
Stanley, G. (2013). Language Learning with Technology. Cambridge: Cambridge
University Press.
Swain, M. (2000). The output hypothesis and beyond: mediating acquisition
through collaborative dialogue. In Lantolf, J. P. (ed.), Sociocultural Theory and
Second Language Learning. Oxford: Oxford University Press, pp. 97–114.
Swain, M. and Lapkin, S. (1989). Canadian immersion and adult second language
teaching: what’s the connection? The Modern Language Journal 73: 150–159.
Thornbury, S. (2013). Resisting coursebooks. In Gray, J. (ed.), Critical Perspectives
on Language Teaching Materials. London: Palgrave Macmillan, pp. 204–223.
Tsui, A. B. M. (1995). Introducing Classroom Interaction. Harmondsworth: Penguin.
Ur, P. (2012). A Course in English Language Teaching. Second edition. Cambridge:
Van Ek, J. A. and Trim, J. L. M. (1998). Threshold 1990: Council of Europe/Conseil de
L’Europe. Revised and corrected edition. Cambridge: Cambridge University
Press.
Walker, A. and White, G. (2013). Technology Enhanced Language Learning: Con-
necting Theory and Practice. Oxford: Oxford University Press.
Wilkins, D. A. (1976). Notional Syllabuses. Oxford: Oxford University Press.
5 Pronunciation
Whose Pronunciation?
The sounds, especially the vowel sounds, of a language inevitably
change over time. Evidence of this is that the rhymes used by
Elizabethan poets sometimes do not work today. Even in more
recent times there have been noticeable changes. If you listen to BBC
radio news or sports broadcasts of the 1950s or earlier and compare
them with their modern counterparts, you will notice that pronuncia-
tion has changed. Traditionally, the pronunciation target for foreign
learners of British English was ‘Received Pronunciation’ (RP). The
term was used by the British phonetician Daniel Jones to describe
the target pronunciation of his English Pronouncing Dictionary (EPD),
which was actually his own pronunciation:
I call it Received Pronunciation (abbreviation RP), for want of

a better term. I wish it to be clearly understood, however, that RP
means ‘widely understood pronunciation’.
(Jones 1926: viii)
The term RP was not invented by Jones, and in the first edition (1917)
of the EPD he had called his pronunciation model ‘Public School
Pronunciation’ (PSP). Just why he abandoned the term PSP in favour
of RP for the second edition is unclear. What he meant by ‘widely
understood’ is also unclear. However, Jones’s target pronunciation of
RP/PSP was in fact the accent used by, but not laid down by, the BBC
(founded 1922). It was just that the broadcasters were recruited from
the public schools (Abercrombie 1991: 48–49). It is true that the RP
accent would have been more widely understood in Britain than any
regional accent would have been. However, it is worth remembering
what George Bernard Shaw, himself an Irishman, wrote about English
accents at this time:
122 Pronunciation
It is impossible for an Englishman to open his mouth without
making some other Englishman hate or despise him.
(Pygmalion, 1916, Preface)
Early Watch with Mother broadcasts of the 1950s and 1960s, available
on YouTube, show how RP used to sound. These early broadcasts
were initially only receivable in the Home Counties, the heartland of
RP, and only by those socially privileged enough to own a TV. RP
was never the way most British or even most English people spoke.
Even in 1926, RP/PSP was used by only a small, privileged minority
of British speakers, so that it was always socially marked and
associated with the affluent south-east of Britain, although there
were RP speakers all over England, and of course public schools all
over Britain where RP was spoken. Nor was RP ever something
monolithic (Abercrombie 1991: 48–49). There was social variation
within RP, the RP of the aristocracy being different from the RP of
the educated middle class. As time went on and class barriers
became less rigid, RP became more open to outside influences. The
aspiring lower middle class, especially those from the south-east of
England, found it easy to modify their regional speech in the direc-
tion of RP when interacting with RP speakers. Such speakers never-
theless retained some features of their primary regional
pronunciation (see Cruttenden 2014: 76–78).
Since the 1960s RP has become increasingly influenced by other
accents (see Cruttenden 2014: 83–85 for a detailed description) and
these developments have been accompanied by more tolerant atti-
tudes to regional varieties of pronunciation, although the broad
urban dialects still carry a social stigma. Tolerance of variation in
pronunciation has been closely linked to changes in the social class
structure of Britain. As the traditional class barriers to power and
influence have been increasingly eroded, although not erased, a new
privileged class has grown massively in influence. This class is drawn
from pop culture, the New Economy, fashion, entertainment and
sport and has produced new social role models in music, sport, film,
TV and the digital world who are not speakers of RP but of other
varieties. The term ‘Estuary English’ is sometimes used to refer to
‘some people in public life who … now find it acceptable to speak
with some characteristics of the London area … such as glottal stops,
which in earlier times would have caused comment or disapproval’
(Roach 2009: 4). American pronunciation has also been influential.
Cruttenden (2014: 86) notes that a ‘mid-Atlantic’ accent is common
in pop stars.
Dissatisfaction with the label RP as the target pronunciation for
foreign learners was expressed by Windsor-Lewis (1972). He suggested
the term ‘General British’ (GB) on the analogy of ‘General American’
Pronunciation 123
(GA) for a target pronunciation of British English relatively unmarked
both socially and regionally. Gradually the term GB has come to be
preferred to RP. It represents a target which is in keeping with
changes in English pronunciation since the 1960s. Cruttenden (2014:
80–81) and Carley et al. (2018: 6) use the term GB instead of RP as
the target pronunciation for foreign learners. The Oxford Advanced
Learner’s Dictionary (2011: R45) defines GB as being based on the
pronunciation of younger speakers of RP and ‘a range of similar
accents which are not strongly regional’.
Cruttenden (2014: 80–81) uses the term Conspicuous General British
(CGB) to denote the now rather old-fashioned RP spoken for example by
the older Royals, including Prince Charles, in contrast to the more
modern pronunciation of, for instance, Prince Harry. Younger speakers
of GB may incorporate various features of regional English, especially
south-east regional and London English into their GB speech. There is
thus a fluid boundary between GB and those speakers of a regional
variety of English who approximate to RP. Cruttenden (2014: 81–82)
calls this variety Regional General British (RGB). He further identifies
a number of other standards in Britain, including Standard Scottish
English (SSE), London English, Estuary English (EE), Multicultural
London English (MLE) and General Northern English (GNE) (Crutten-
den 2014: 88–92).
Describing the Sounds of English
The Phonemes of English

The set of contrasting sounds a language employs to express meaning
comprises its phonemes (see Cruttenden 2014: 43–49; Roach 2009:
2–3; Collins et al. 2019: 12–16). Each language uses a limited set of
phonemes and no two languages use the same set. The set of pho-
nemes each language possesses is a subset of all the sounds which the
languages of the world use, which in turn is a subset of all the vocal
noises human beings are capable of. Some very exotic sounds are used
as phonemes in some languages. Some languages use clicks, for
example, which are not language sounds at all in English. Thus, the
‘tut-tut’ sound (a dental click,) which in English is a paralinguistic, not
a linguistic, expression of annoyance or disapproval, is used as
a phoneme in some African languages. And the lateral click (a click
with one side of the tongue against the roof of the mouth), which is
an alternative to ‘gee-up’ by some English speakers when talking to
horses, is also used in some languages as a phoneme (Cruttenden
2014: 34).
The sounds of language do not of themselves mean anything. They
combine to form words or word-parts which have meaning. English,
124 Pronunciation
like all languages, has a fixed inventory of distinctive sound contrasts,
for example, the vowel contrasts in the words ‘bed’ and ‘bad’ or the
final consonant contrasts in ‘bad’ and ‘bat’. Such pairs of words which
differ only in respect of one sound are called ‘minimal pairs’ (Roach
2009: 51–53; Cruttenden 2014: 43; Collins et al. 2019: 12).
There are generally agreed to be 44 phonemes in GB English,
comprising 20 vowels and 24 consonants (Roach 2009: 34). Since
there are more phonemes than letters of the alphabet, some addi-
tional symbols have to be used for a phonemic transcription.
Furthermore, there is no consistent correspondence in English
between sounds and the letters of the alphabet used to represent
them in writing (see Collins et al. 2019: 98–110). This is particularly
true of the vowels. At school many of us learned that there are five
vowels: ‘a’, ‘e’, ‘i’, ‘o’ and ‘u’. In fact, this is not a list of the vowel
sounds at all. It is a list of the five letters of the alphabet which are
used to represent vowel sounds in writing and never used to repre-
sent consonant sounds. It does not, for example, include the letter
‘y’, which is sometimes used to represent the vowel /ɪ/ (as in ‘lynch’),
the diphthong /aı/ as in /dye/ and the second element of the
diphthong /eɪ/ as in ‘day’. The letter ‘y’ nicely illustrates how poor
a guide English spelling is to pronunciation because ‘y’ can also be
used to represent the sound /j/, as in ‘yes’.
Tables 5.1 and 5.2 show the transcriptions for the GB English
vowel and consonant phonemes, respectively, following Cruttenden
(2014).
Table 5.1 Transcriptions for the 20 GB vowels
Monophthongs Diphthongs
• /ɪ/ as in ‘bid’ • /eɪ/ as in ‘page’/

• /i:/ as in ‘bead’ • /aɪ/ as in ‘five’
• /e/ as in ‘bed’ • /ɔɪ/ as in ‘join’
• /ɛ:/ as in ‘hair’
• /a/ as in ‘bat’ • /əʊ/ as in ‘home’
• / :/ as in ‘bar’ • /aʊ/ as in ‘now’
• /ᴅ/ as in bottom (first vowel)
• /ɔ:/ as in ‘bore’ • /ɪə/ as in ‘near’
• /ʊ/ as in ‘put’ • /ʊə/ as in ‘dour’
• /u:/ as in ‘boot’
• /ᴧ/ as in ‘bud’
• /ɜ:/ as in ‘fur’
• /ə/ as in ‘bottom’ (second vowel)
Pronunciation 125
Note 1: As Table 5.1 shows, vowels may be subdivided into mono-
phthongs (simple or pure vowels), for example /e/ in ‘bed’ or /a/ in ‘bat’,
and diphthongs (glides between two simple vowels), for example /eɪ/ in
‘pay’. Diphthongs start with the first vowel shown in the transcription and
move in the direction of the second vowel shown (see Roach 2009: 17–18).
Monophthongs may be either short, for example /ɪ/ in ‘bid’ or long, for
example /i:/ in ‘bead’. The notation (:) means that a vowel is ‘long’ rather
than ‘short’ (see Roach 2009: 13–14, 16–17; Cruttenden 2014: 38, 100–102).
Note 2: Older speakers of GB and speakers of CGB (RP) may well have
a diphthong /e / instead of a monophthong /ɛ:/ in words like ‘hair’ (Crutten-
den 2014: 118; Collins et al.: 2019: 94; Lorenz 2012: 51; Carley et al.
2018: 175).
Note 3: The symbol /ᴂ/ was traditionally used for transcription of the
vowel sound in words like ‘bat’ in RP. Cruttenden (2014: 84) prefers the
symbol /a/ since this vowel has evolved in GB to a more open vowel with
the jaw lowered, more like the vowel used in the north of England. The
older RP vowel, which was closer to /e/, tends now to be restricted to older
speakers of GB and speakers of CGB (see also Collins et al. 2019: 92).
Allophones
The phoneme is an abstraction and is never said (realised) twice in
exactly the same way, even by the same speaker. There are many reasons
for this, including whether one is speaking with one’s mouth full or not
but also, for example, how carefully one is speaking or to whom one is
speaking. Apart from variations of this sort, however, the realisation of
a phoneme will tend to vary systematically and consistently according to
the sounds it immediately follows and precedes within words and at
Table 5.2 Transcriptions for the 24 GB consonants
/p/ as in ‘pen’ /s/ as in ‘so’

/b/ as in ‘bad’ /z/ as in ‘zoo’
/t/ as in ‘tea’ /ʃ/ as in ‘she’
/d/ as in ‘dig’ /ʒ/ as in/vision/(middle consonant)
/k/ as in ‘cat’ /h/ as in ‘how’
/ɡ/ as in ‘got’ /m/ as in ‘man’
/tʃ/ as in ‘chin’ /n/ as in ‘no’
/dʒ/ as in ‘June’ /ŋ/ as in ‘sing’ (final consonant)
/f/ as in ‘fall’ /l/ as in ‘leg’
/v/ as in ‘voice’ /r/ as in ‘red’
/ɸ/ as in ‘thin’ /j/ as in ‘yes’
/ð/ as in ‘then’ /w/ as in ‘wet’
126 Pronunciation
word boundaries, that is to say, in different phonetic environments.
Different realisations of a phoneme which are clearly describable and
can be categorised systematically are termed allophones of that phoneme
(see Cruttenden 2014: 46, 308–310; Roach 2009: 33; Carr 1999: 39;
Collins et al. 2019: 13–14, 112–114; Carley et al. 2018: 2–3).
A nice example of allophones is provided by the two different ‘l’ sounds
in the word ‘little’. If you say the word ‘little’ slowly and concentrate on
where your tongue is for each ‘l’, you will probably feel the difference. In
the first ‘l’ (light ‘l’) the tip of your tongue is probably pressed against your
alveolar ridge, the hard boney ridge behind your front teeth, and the front
part of the tongue is arched up high towards the front part of the roof of
your mouth (hard palate). Light ‘l’ is a palatised ‘l’. In the second ‘l’ (dark ‘l’)
the back part of your tongue arches up and back towards the back of the
roof of your mouth (soft palate or velum). For this reason dark ‘l’ is
sometimes called a velarised ‘l’. This is a systematic difference: dark ‘l’
occurs word-finally after a vowel (as in ‘doll’), syllabically (as in the second
‘l’ in ‘little’) and after a vowel and before a consonant (as in ‘salt’). Light ‘l’
occurs word-initially (as in the first ‘l’ in ‘little’), in word-initial clusters (as in
‘blow’) and word-medially (as in ‘silly’). See Cruttenden (2014: 217–222) and
Carley et al. (2018: 30).
The difference between the two ‘l’ sounds does not carry any meaning
distinction. It is not a phonemic distinction but an allophonic distinction.
Dark ‘l’ and light ‘l’ are allophones of the /l/ phoneme in English. This
means that if you substitute a dark ‘l’ for a light ‘l’ or vice versa in any
English word you will never produce a different word. The pronunciation
may sound unusual or it may sound as if you are speaking with a regional
accent: Yorkshire, if you put dark ‘l’s’ at the beginning of words (the way
people from Leeds say the name of their city) or Irish, if you put a light ‘l’ at
the end of the word (try to say ‘the little people’ with an Irish accent).
The test of whether two sound differences are phonemic or merely
allophonic is the ‘minimal pairs test’. This means, if one can find two
different words which differ only in respect of the two sounds in question,
then the difference is phonemic, as in the examples of ‘bed’ and ‘bad’ and
‘bad’ and ‘bat’ already given. Problems arise for learners when a phonemic
distinction in English does not exist in their native language. Thus, /r/ and /l/
are phonemes in English as shown by the minimal pair ‘red’ and ‘led’. In
some Asian languages, Japanese, for example, there is no phonemic distinc-
tion similar to the English /r/ and /l/ phonemic distinction (Cruttenden
2014: 228). It is therefore difficult for speakers of these languages to keep /r/
and /l/ distinct when they speak English, especially if they have to say
something like ‘red and yellow lorries’. Similarly, they may find it difficult
to distinguish between the two sounds when listening to speech.
In general, unless words are said very carefully, there is a tendency for
one sound to take on features of an adjacent sound (assimilation) or even to
be omitted (elision), so that ‘grandpa’ may become ‘granpa’ or ‘grampa’ (see
Pronunciation 127
Cruttenden 2014: 256–258, 308–318; Roach 2009: 110–115; Carley et al.
2018: 296–301). This also happens across word boundaries in fluent, rapid
speech, particularly in set phrases, which may be said as if they were one
word, so that ‘good morning’ becomes ‘goomorning’. Some elisions and
assimilations may be frowned upon as ‘careless speech’ in some circles, as in
‘gimme’ for ‘give me’ or ‘wanna’ for ‘want to’. However, it is in the nature
of things that in fluent, natural speech words run into one another. There
are not clear spaces between them as there are in writing. This leads to
various rather corny jokes such as, ‘Where’s your grammar?’ ‘She’s gone to
the pictures with my grandad.’ And (heard at the seaside), ‘A yacht, a yacht!’
‘No, just pleasantly warm, thank you.’
Producing the Sounds of English

The evolution of our speech organs has placed us at a great advantage over
other species, for example apes and chimps, for the transmission of vocal
language. The whole area from the larynx to the lips is involved in making
speech sounds. This area is called the vocal tract and is shown in Figure 5.1.
Speech and Breathing

Since we have to draw breath while speaking, speech is uttered in short
rhythmic bursts adapted to the breathing cycle. In breathing out, air is forcibly
expelled from the lungs through the windpipe across the larynx into the throat
(pharynx) and out of the mouth. Speech sounds are made by modifying the
airstream that we expel in breathing out and allowing our vocal chords to
nasal cavity
hard palate soft palate

oral cavity
alveolar ridge
upper lip uvula
teeth tongue
pharynx
lower lip
vocal chords
larynx
Figure 5.1 The organs of speech

128 Pronunciation
vibrate intermittently. This produces the vowels and the voiced consonants
such as /b/ and /d/ versus the unvoiced consonants such as /p/ and /t/. The
difference between a vowel and a consonant is that for a consonant the
airstream is obstructed at some point between the pharynx and the lips by
two speech organs coming together and being released (the articulators),
whereas for a vowel this obstruction followed by release is lacking. A further
modification occurs for nasal sounds: in English /m/, /n/ and /ŋ/, in that for
them air is additionally allowed to pass out through the nose (nasal cavity).
Voicing
The vocal chords, which produce voicing, are located in the larynx (see
Figure 5.1). From the outside, the larynx, sometimes called the voice box, is
visible as the Adam’s apple and when people are speaking, you can see it
moving around. Voicing is produced by opening or closing the vocal
chords, which are two bony cartilages, rather like curtains, situated on
either side of the larynx. They can be progressively tensed or relaxed.
When completely relaxed or open, the air passes noiselessly through a gap
between them (the glottis) but when tensed or partly closed, they will
vibrate in the stream of air and produce a vowel sound which will go on
until it is necessary to draw breath. If learners put their finger and thumb on
either side of their Adam’s apple, they can feel the vocal chords vibrate
when voicing is in progress and stop vibrating when voicing stops. If you get
laryngitis, your vocal chords start to seize up and will not vibrate, so that
first of all you become hoarse, then you can only whisper and finally you
can say nothing at all. The same thing happens if you talk too much and
start to lose your voice. (See also Cruttenden 2014: 9–12; Roach 2009:
22–25; Collins et al. 2019: 32–36; Carley et al. 2018: 10–11.)
The Monophthongs
Vowels are, by definition, always voiced (Cruttenden 2014: 28; Carr
1999: 20). The air is allowed to pass from the windpipe, through the
vocal tract and out of the mouth unimpeded while the partly closed
vocal chords are vibrating. Differences among vowels are determined by
a number of features, especially:
the raising or lowering of the tongue in the mouth, from low to mid
to high
the protrusion or retraction of the tongue in the mouth from front
to central to back
the length of the vowel (long or short)
degree of lip rounding or lip spreading (a secondary feature in
English)
whether the vowel is oral or nasal (but English has no nasal vowels)
Pronunciation 129
All English vowels can be described along these dimensions and any two
vowels in English will differ along one or more of these dimensions.
Note 1: For low vowels the lower jaw tends to be dropped so that
these vowels are sometimes also termed ‘lax’ or ‘open’. High vowels
involve greater muscular tension to get the tongue close to the roof of
the mouth and so they are sometimes called ‘tense’ or ‘close’ vowels.
Note 2: Lip rounding only occurs in the back vowels in GB,
namely, /ʊ, u, ɔ, ɒ/. It is a characteristic of GB English that it does not
have front rounded vowels, unlike French or German. Lip rounding is
only a secondary feature distinguishing pairs of phonemes in English. To
appreciate the difference between a vowel with and without lip rounding,
consider the words ‘look’, in which the /ʊ/ vowel is rounded, and ‘luck’,
in which the /ᴧ/ vowel is not rounded. If you say the two words quickly
one after another, you can feel not only your tongue moving backwards
and forwards but your lips rounding (pursing) for ‘look’ and spreading
for ‘luck’. However the main difference between /ʊ/ and /ᴧ/ is the
position of the tongue, which is higher and further back in ‘look’ than
in ‘luck’.
The positions of the 13 GB monophthongs are shown in Figure 5.2.
The trapeze diagram is a stylised representation of the mouth in lateral
cross-section, with the front of the mouth being on the left and the back
of the mouth being on the right. The position of the symbols represents
the highest point of the tongue for each vowel. The two dots after some
symbols indicate a long vowel. The trapeze reflects the fact that there is
more perceived room available in the front of the mouth than in the
back of the mouth. This is partly because we have better neuro-muscular
control of the tongue in the front of the mouth.
The vowels may be briefly described as follows:
The high-front vowels /ɪ/ and /i:/
In English we have a short high-front vowel, /ɪ/ as in ‘sit’, and a long
counterpart vowel /i:/ as in ‘seat’. If you say the long vowel in ‘seat’, you
can feel your tongue pushing upwards and forwards in your mouth; if
you then go into the short vowel in ‘sit’, you may feel your tongue
retracting slightly. This is because there is a slight positional difference
between the two vowels, with the short vowel being slightly lower and
not so far forward (less tense and close). Some non-native speakers
whose native language does not have a long and a short high-fronted
vowel so close together find it difficult to distinguish them. In particular,
they tend to make the short vowel /ɪ/ too long and too tense so that it
can be confused with /i:/. Many speakers of GB tend to slightly
diphthongise /i:/ and this is a tip which can sometimes help learners to
keep the two vowels apart.
The mid-front vowels /e/ and /ɛ:/
130 Pronunciation
i: u:
ı Ʊ
е ә / ɜ: ͻ:
ɛ:
v
a ɒ
ɑ:
Figure 5.2 The GB monophthongs
We have a short, mid-front vowel /e/ as in ‘set’. If you say, ‘seat’, ‘sit, ‘set’,
you can feel your tongue dropping and lower jaw probably falling a little as
well. The long, mid-front vowel /ɛ:/ as in ‘hair’ is just slightly lower than the /e/
vowel. It is also distinguished from it by length. Cruttenden (2014: 119) notes
that all the words with /ɛ:/ have a following ‘r’ in the spelling, which is not
pronounced. He suggests that learners aim to produce a lengthened version
of /e/, opening the mouth more as well. Older speakers of GB and speakers of
CGB English may diphthongise /ɛ:/ to /eə/.
The short, low-front vowel /a/
We have a short, low-front vowel /a/ as in ‘sat’. If you say ‘seat’, ‘sit’,
‘set’, ‘sat’, you should feel your tongue continuing to drop and your
mouth opening wider. Older speakers of GB and speakers of CGB will
have a vowel for /a/ which is somewhat higher and closer to /e/. Speakers
with different pronunciations may occasionally misunderstand each
other, and might ask, for example, ‘Did you say your name was Jan or
Jenn?’ Learners may have similar problems (see Cruttenden 2014:
119–120). Sometimes there may also be confusion with /ᴧ/ (see below).
The short, low-central vowel /ᴧ/
If you now, after saying ‘sat’, concentrate on slightly retracting your
tongue, bunching it backwards just a little, you may start to make a /ᴧ/
vowel as in ‘sup’. This is a short, low-central vowel, actually rather close
to the /a/ vowel. Although individual speakers keep their own realisa
tions apart, listeners unused to a particular accent, can sometimes have
trouble distinguishing the two sounds, and may ask, ‘Did you say cup or
cap?’ Foreign learners hearing a new word like ‘palpable’ or ‘culpable’ for
the first time may not be sure which vowel it has (see Cruttenden 2014:
121–122).
Pronunciation 131
The mid-central vowels / / and /ɜ:/
Things are actually very crowded in the centre of the mouth, because
we also have a long and a short mid-central vowel. Both vowels occur in
the word ‘certain’, long /ɜ:/ in the first syllable and short / / in the second
syllable. The short vowel / / always occurs in unstressed positions in
words. If you just do nothing with your tongue and simply vocalise then
you will hit these two vowels, which are mainly distinguished from each
other by length rather than position. They should not usually be proble-
matic for learners.
The high-back vowels /ʊ/ and /u:/
English also has a short and a long high-back vowel to match the
short /ɪ/ and long /i:/ high-front vowels. These are short /ʊ/ as in ‘soot’
and long /u:/ as in ‘suit’. The short vowel is slightly lower and more
central than the long vowel, analogous to /ɪ/ and /i:/. If you go back to
saying /i:/ and just pull your tongue back and try to get it even higher in
the back of your mouth, arching it up to do so, and at the same time
rounding your lips, then you should make the /u:/ sound. Repeat several
times ‘seat’, ‘suit’ to get the feel of this. If you then say ‘suit’, ‘soot’, you
will notice the difference in vowel length, and you may also notice that
you probably relax your tongue and lower it just a little for ‘soot’. Both
the /ʊ/ in ‘soot’ and the /u:/ in suit have lip rounding. Collins et al. (2019)
report a tendency for /ʊ/ and /u:/ to have moved further forward in the
mouth in recent years than I have shown them.
The long, mid-back vowel /ɔ:/
To partly match the short, mid-front vowel /e/, English has a long,
mid-back vowel, /ɔ:/ as in ‘saw’. You can get to it by going from /e/ in
‘bet’ to /ɜ:/ in ‘Bert’ to /ɔ:/ in ‘bought’. You should feel how your tongue
progressively retracts and bunches up as you do so. You will also notice
that the vowel in ‘bet’ is short while the vowels in ‘Bert’ and ‘bought’ are
long. The /ɔ:/ in ‘bought’ has ‘medium lip rounding’ (Cruttenden 2014:
128) unlike the /e/ in ‘bet’ and the /ɜ:/ in ‘Bert’.
The low-back vowels /ɒ/ and /ɑ:/
We have a match for the short, low-front vowel /a/, as in ‘pat’ with the
short, low-back vowel /ɒ/ as in ‘pot’. We also have a long, low-back
vowel /ɑ:/ as in ‘part’. The short /ɒ/ in ‘pot’ has ‘slight open lip rounding’
(Cruttenden 2014: 126) unlike the long /ɑ:/ in ‘part’. Older speakers of
GB and speakers of CGB will have a more retracted position for /ɑ:/ than
shown in Figure 5.2 (see Cruttenden 2014: 124).
For description of the GB simple vowels see also Cruttenden (2014:
110–139), Roach (2009: 13–17), Collins et al. (2019: 89–96) and Carley
et al. (2018: 146–177).
132 Pronunciation
The Diphthongs
In Figure 5.3 the seven diphthongs are shown. These are /eı/ as in
‘page’, /aɪ/ as in ‘five’, /əʊ/ as in ‘home’, /aʊ/ as in ‘now’, /ɔɪ/ as in
‘join’, /ɪə/ as in ‘near’ and /ʊə/ as in ‘cruel’ and ‘dour’. The two last-
mentioned diphthongs with the second element /ə/ are termed ‘centring
diphthongs’. In traditional descriptions of RP there was a third centring
diphthong, namely /eə/ in words like ‘hair’. This pronunciation is now
less common, with the vowel being increasingly monopthongised to /ɛ:/,
as already mentioned. The centring diphthong /ʊə/ also seems to be
disappearing in word-final position in frequent words such as ‘poor’,
‘moor’, ‘you’re’. In such words it is often replaced by /ᴐ:/. Do you
personally distinguish in pronunciation between ‘poor’ and ‘paw’,
‘moor’ and ‘more’, ‘you’re’ and ‘your’? However, in less frequent words
such as ‘dour’ and words where it is preceded by /j/, as in ‘lure’ and
‘cure’ or followed by a consonant, as in ‘cruel’ and ‘cruet’ the diphthong
is still obligatory (see Cruttenden 2014: 332; Carley et al. 2018: 145).
The overall length of the diphthongs is similar to that of the long
vowels. The first vowel indicates that the glide starts in the region of that
vowel and moves in the direction of the second vowel. Roach (2009: 17)
makes the point that the first element is generally much longer and
stronger than the second element. It is best for learners to think of
themselves as gliding towards the second element but not quite getting
there. They should not be told to make the first vowel and then
the second because this will mean that they do not tail off in the
appropriate way. Individuals vary in the way they make their
diphthongs, and there is great regional variation within Britain and in
the varieties of English in the world.
Many learners tend to monophthongise their diphthongs and can be
helped by being shown the direction of the glide on a diagram such as
ıә Ʊә
еı әƱ
ͻı
aı aƱ
Figure 5.3 The GB diphthongs

Pronunciation 133
Figure 5.3 while they listen to the teacher or a recorded voice saying the
sound. They can then try to produce it themselves. Exaggerating the glide
of diphthongs should be avoided, however, as this may sound like CGB.
In Figure 5.3 the arrows just show the starting point and direction of the
glide, not where it finishes. In most cases, however, this will be well
short of the second vowel shown in the transcription. Learners should
not think they have to finish up saying the second vowel at the end of
the glide.
For descriptions of the GB diphthongs see also Cruttenden (2014:
140–156), Roach (2009: 17–18) and Collins et al. (2019: 96–98).
The Consonants
The consonants of English can be categorised according to the following
features:
place of articulation (bilabial, alveolar, velar etc.). This refers to the

point in the vocal tract between the larynx and the lips where the air
flow is interrupted.
manner of articulation (stop, fricative, affricate etc.). This refers to
the way in which the air is impeded by the speech organs, in
particular with regard to how complete the closure is.
presence or absence of voicing (voiceless versus voiced). This refers
to whether or not the consonant is produced when the vocal chords
are in open position (voiceless) or are partly closed and thus vibrate
(voiced).
presence or absence of nasality (oral versus nasal). This refers to
whether or not a consonant is made while the velum is lowered so
that air can enter the nasal cavity and resonate (nasality).
Table 5.3 presents a categorisation of the GB English consonants along

these four dimensions.
The following descriptions may serve as support for learners in cases
where their pronunciation of a particular consonant phoneme could be
confused with another consonant phoneme. Description of how to make
the sound (articulatory description) is, however, always best accompa-
nied by providing an acoustic model of the target sound by saying it in
a word and contrasting it with another word which differs from it only in
terms of the sound in question (minimal pairs), for example ‘bat’ and
‘bad’ for /t/ and /d/.
The (oral) stops /p, b, t, d, k, ɡ/
The air is stopped completely for a moment and then suddenly allowed
to escape with a little explosion. In English the oral stops are /p, b, t, d, k,
ɡ/, as in ‘pat’, ‘bat’, ‘tap’, ‘dad’, ‘cat’ and ‘get’. These six consonants in fact
134 Pronunciation
Table 5.3 Categorisation of the GB consonants
voiceless voiced
Stops bilabial /p/ /b/

alveolar /t/ /d/
velar /k/ /ɡ/
Nasal Stops bilabial /m/
alveolar /n/
velar /ŋ/
Fricatives labio-dental /f/ /v/
dental /ɸ/ /ð/
alveolar /s/ /z/
post-alveolar /ʃ/ /ʒ/
glottal /h/
Affricates post-alveolar /tʃ/ /dʒ/
Lateral alveolar /l/
Approximants post-alveolar /r/
palatal /j/
labial-velar /w/
consist of a voiceless–voiced pair at three points of articulation: the lips,

the alveolar ridge and the velum.
The place of articulation is at the lips for /p/ and /b/, which are called
bilabial stops because they are made by pressing the lips together and
then releasing them suddenly.
With /t/ and /d/ it is the tip of the tongue pressing on the hard bony
alveolar ridge immediately behind the upper teeth which makes the stop
(alveolar stop).
For the velar consonants /k/ and /ɡ/ the place of articulation is much
further back and the sound is achieved by pressing the back part of the
tongue up against the back of the roof of the mouth (the soft palate or
velum) and then releasing the pressure suddenly.
At the beginning of a word the important difference in sound between
voiceless /p, t, k/, on the one hand, and voiced /b, d, ɡ/, on the other, is the
presence of aspiration for the voiceless sounds and its absence for the
voiced sounds. By aspiration is meant that there is a hesitation
between the release of the articulators for /p, t, k/ and the beginning
of voicing for the following vowel (voice onset time). During this time
air escapes through the vocal chords making a breathy /h/ sound
between the consonant and the onset of the following vowel. In the
case of /b, d, ɡ/ the release is weaker and voicing starts at about the
same time as the release or even just before it. If learners do not
Pronunciation 135
aspirate their initial voiceless plosives, listeners take the sound to be
the corresponding voiced sound (see Roach 2009: 27–28; Cruttenden
2014: 163–166; Carley et al. 2018: 16–18).
The nasal stops /m, n, ŋ/
In addition, English has three voiced nasal stops /m, n, ŋ/, as in ‘mat’,
‘not’, ‘song’, at the same three points of articulation as the oral stops.
This means there is a three-way contrast at each point of articulation:
voiceless stop, voiced stop and nasal stop.
The bilabial, nasal plosive /m/ corresponds to the bilabial, voiced,
(oral) plosive /b/, in that the lips are closed.
The alveolar nasal /n/ corresponds to the alveolar, voiced, (oral) plosive /d/.
The velar nasal /ŋ/ corresponds to the velar, voiced, (oral) plosive /ɡ/.
The nasals differ from the corresponding oral consonants in that when
oral closure is made with the lips for /m/, the tongue and alveolar ridge
for /n/, or tongue and velum for /ŋ/, instead of the back of the soft palate
(the velum) being raised to close off the nasal cavity, for these three nasal
consonants it is lowered so that the sound resonates not just in the oral
cavity but also in the nasal cavity. This gives these three sounds their
characteristic nasal quality.
If learners say /b/ and then /m/ alternately, they will see that they close
and release their lips in exactly the same way. And they should be able to
feel a tickling in their nose when they say /m/, which is not present when
they say /b/. If they actually place their thumb and forefinger on either
side of their nose, they should even be able to feel the vibration when
they say /m/. They can do the same exercise for /d/ and /n/, this time
noting that their tongue is in exactly the same position for both sounds.
The same is true of /ɡ/ and /ŋ/.
Learners may find it more difficult to make the /ŋ/ sound than /m/
and /n/. This is because /ŋ/ does not occur at the beginning of words in
English. For more on /ŋ/ see Roach (2009: 46–48) and Cruttenden (2014:
215–217).
Whereas the oral plosives cannot be prolonged, the nasals can be
prolonged until you run out of breath, just like vowels. This is because
the air coming up from the lungs is expelled gradually through the nose
rather than suddenly through the mouth. Learners can prove this to
themselves by making a long drawn out nasal consonant, for example /m/,
and suddenly gripping their nostrils hard between thumb and forefinger as
they would to stifle a sneeze. The sound will stop completely.
The fricatives /f, v, ɸ, ð, s, z, ʃ, ʒ, h/
If the air is merely impeded but not completely stopped at the point of
articulation and allowed to escape gradually under constriction so that
there is friction, a fricative sound is produced. The English fricatives
are /f, v, ɸ, ð, s, z, ʃ, ʒ, h/, as in ‘fat’, ‘vat’, ‘thigh’, ‘thy’, ‘sat’, ‘zoo’,
136 Pronunciation
‘shop’, ‘treasure’ and ‘hat’. The first eight of these nine fricatives repre-
sent voiceless–voiced pairs at four points of articulation, namely the lips/
teeth (labio-dental), the teeth (dental), the alveolar ridge (alveolar) and
just behind the alveolar ridge (post-alveolar). We do not have bilabial or
velar fricatives in English.
The labio-dental fricatives /f/ and /v/ are made by allowing the air to
escape gradually between the lower lip and the upper front teeth.
To make the dental fricatives /ɸ/ and /ð/, the tip of the tongue lightly
touches the bottom and back of the upper teeth so that air can be
pressed through the teeth and around the tongue.
For the alveolar fricatives /s/ and /z/ the tongue tip is brought a little
further back so that the air is forced between it and the alveolar ridge.
For the post-alveolar fricatives /ʃ/ and /ʒ/ the tongue is pulled even
further back just behind the alveolar ridge where the roof of the mouth
(hard palate) begins.
The ninth consonant, the voiceless glottal fricative /h/, is the one made
furthest back in the mouth. It occurs only at the beginning of syllables
before a vowel. It does not have a voiced equivalent and is something of
an odd man out among the fricatives. Air is expelled forcibly from the
lungs before voicing begins for the following vowel and thus friction
occurs as the air passes through the glottis. Cruttenden (2014: 207)
describes /h/ as ‘a strong, voiceless onset to the vowel in question’.
The /h/ sound does not occur in the broad form of most British regional
accents, and even speakers of GB and CGB, when talking informally and
rapidly, tend to ‘drop their aitches’.
If learners sigh or pant, they will make a sound very like /h/. This is
a good tip for French learners of English. But they should not overdo it.
The affricates /tʃ, dʒ/
Affricates are a third class of consonant, and are formed by making
a stop followed by a fricative. In English these comprise the voiceless–
voiced post-alveolar pair /tʃ/, as in ‘churn’ and /dʒ/, as in ‘journey’. The
unvoiced /tʃ/ occurs as the first and last sound in ‘church’ and its voiced
counterpart /dʒ/, as the first sound of ‘jerk’ and the last sound in ‘dirge’.
The /tʃ/ sound actually starts as an unvoiced alveolar stop /t/, but instead
of the tongue being quickly and cleanly withdrawn from the alveolar
ridge, it is just slightly withdrawn (lowered and retracted) to make /ʃ/.
Similarly the /dʒ/ sound starts as a /d/ and glides into a /ʒ/.
One could more economically exclude these two affricates from the
inventory of phonemes and regard them as phoneme combinations,
particularly as there are other consonant combinations in English
which are not given phoneme status. The reason for treating them as
phonemes in their own right has something to do with subjective
impressions of what is one sound and what are two sounds. Speakers
are not normally aware of the fact that /tʃ/ and /dʒ/ are decomposable
Pronunciation 137
into two sounds but they are aware of this for other consonant
clusters such as /tr/, /sl/ or /dr/.
The lateral /l/
Laterals are another class of consonant, of which we only have one in
English, namely /l/, as in ‘lap’. In laterals a partial closure is made
between two articulators and air is allowed to escape from the side. For
English /l/, closure is made between the tip of the tongue and the alveolar
ridge, and either the front of the tongue is raised towards the soft palate
(for light ‘l’) or the back of the tongue raised towards the velum (for dark
‘l’) The sides of the tongue are pressed up against the roof of the mouth
bur air is allowed to escape from one or both sides. Voicing occurs and
there is little friction involved.
The post-alveolar approximant /r/
The /r/ phoneme, as in ‘rat’ in GB pronunciation, is described by
Cruttenden (2014: 222) as a ‘post-alveolar approximant’. It is quite
different from the various ‘r’-sounds in American English, Scottish
English, West Country English and Indian English. It is sometimes
termed a ‘frictionless continuant’ because no contact is made between
the tongue and the inside of the mouth, but the raising of the tongue
changes the path of the airstream just sufficiently for us to distinguish the
sound and regard it as a consonant rather than a vowel, even though it
does not fulfil the articulatory definition of a consonant (contact between
two articulators).
To make /r/, the tip of the tongue lunges towards but does not quite
touch the alveolar ridge, while the sides of the back of the tongue are
held against the upper back teeth. According to Cruttenden (2014: 224),
‘[T]he central part of the tongue is lowered … so that the effect … is

one of hollowing and slight retroflexion of the tip … The air stream
is thus allowed to escape freely, without friction, over the central
part of the tongue.’
The unrounded palatal approximant /j/ and rounded labial-velar approx-

imant /w/
These two sounds may be described as semi-vowels. Cruttenden
(2014: 228) defines a semi–vowel as ‘a rapid vocalic glide on to
a syllabic sound of greater steady duration’. There is therefore no
contact between articulators in these sounds. In English there are two
semi-vowels: the unrounded palatal approximant /j/, as in ‘yes’ and the
rounded labial-velar approximant /w/ as in ‘win’. Although these
sounds are phonetically vowels, they are counted as consonants for
functional and perceptual reasons, namely because they alternate with
138 Pronunciation
vowels to form syllables in the way consonants do, and we thus
perceive them as consonants.
For /j/ the rapid vocalic glide starts from the vowel position /i:/. The
lips tend to be spread.
For /w/ the rapid vocalic glide starts from the vowel position /u:/. The
lips are rounded.
The glottal stop /Ɂ/
The glottal stop is not a phoneme of English but it occurs frequently
in GB as an allophonic variant of /t/ at the end of words, as in /gᴅɁ/ for /
gᴅt/ or between vowels, for example in words such as ‘forgotten’. This
can cause comprehension problems for learners.
The glottal stop is made by suddenly and forcibly closing the vocal
chords completely for an instant and then releasing them suddenly so as
to abruptly cut off a preceding vowel.
In various regional varieties of British English, /Ɂ/ is very frequent.
Popular London speech is closely associated with glottal stops, which are
used to replace not just /t/, but also /p/ and /k/ at the end of words so
that word pairs like ‘right’ and ‘ripe’ and ‘rake’ and ‘rate’ may not be
distinguishable. This can cause further problems of comprehension for
learners in phrases like a ‘cup of tea and a piece of cake’, which
becomes /ɘk Ɂti:nɘpi:sɘkeiɁ/.
Over the last 30 years the glottal stop has become more and more
frequent in high-status speakers of GB, including the ‘young Royals’.
This seems to be a socio-linguistic influence of south-east English and
London regional speech, as increasingly spoken by many influential
people in show business, fashion, the arts and commerce in the London
area (Estuary English).
For descriptions of the GB consonants see also Cruttenden (2014:
161–237), Roach (2009: 26–30, 39–55), Collins et al. (2019: 63–78) and
Carley et al. (2018: 34–94).
Stress, Rhythm and Intonation
Stress
Consideration has to be given both to word stress patterns and utterance
stress patterns.
Word Stress Patterns

The relevant unit for stress is the syllable. Stress refers to the degree of
respiratory and muscular energy with which a speaker produces
a syllable in relation to neighbouring syllables. Words of two syllables
have one stressed and one unstressed syllable. In words of more than
Pronunciation 139
two syllables there will be one syllable which receives primary stress,
while other syllables may receive secondary (weaker) stress or be
unstressed. In transcription a stressed syllable is shown by a stress mark
(ᶦ) placed before the syllable. A stressed syllable is perceived by the
listener as being louder and at a different pitch, often at a higher pitch,
than unstressed syllables or perceived as shifting in pitch, as in
the second, stressed syllable in /aᶦround/ (Roach 2009: 75). There also
tend to be differences in vowel length and vowel quality between stressed
and unstressed syllables, with vowels in unstressed syllables tending to
be shortened and often to collapse towards the short central ‘schwa’
vowel / /, which occurs only in unstressed syllables, as in the first vowel
of /poᶦtato/ and /aᶦround/. However, other vowels are found in unstressed
syllables, for example /ɪ/, as in /ᶦdevil/, /ʊ/, as in /ᶦinput, / ʊ/, as in /ᶦphoto/
and /ᴧ/, as in /ᶦhiccough/ (see Roach 2009: 73–74; Collins et al. 2019:
130–132; Cruttenden 2014: 242–244).
Generally, the first syllable of a two-syllable or three-syllable word is
the stressed syllable, as in /ᶦhappy/, /ᶦhappiness’/. However, there are many
exceptions. In particular, borrowed words may maintain the stress pattern
of their source language. Thus, /hoᶦtel/, /deᶦsire/, /reᶦport/, /deᶦssert/, which
are all of French origin, retain their French stress pattern with the stress on
the second syllable. The word ‘potato’, transcribed /p ᶦteit ʊ/, maintains
its Spanish stress patterning, with primary stress on the middle syllable. If
you say ‘potato’, however, you will notice that there is secondary stress on
the final syllable, while the first syllable is unstressed. For words of four
syllables or more there is a tendency for the primary stress to be placed on
the last-but-two syllable, as in /serenᶦdipity/ and /comᶦmunicate/ (Collins
et al. 2019: 132–133; Cruttenden 2014: 248; Roach 2009: 74–79).
When a monosyllabic word is said in isolation, as in reading out a list, its
syllable receives stress, whereas in connected speech it may or may not be
stressed, according to its importance in the uttterance. The stress pattern of
words of more than one syllable, on the other hand, is generally fixed in the
sense that it does not usually change in connected speech (Collins et al.
2019: 132). But never say never, and for some exceptions, where stress-shift
may occur in connected speech, see Roach (2009: 86).
In some cases a word has two different stress patterns to mark the
difference between its different functions as two parts of speech. When
a word can be used as both a noun and a verb, as in the case of ‘present’,
then the noun, /ᶦprez nt/, usually has stress on the first syllable and the
verb, /pr ᶦzent/, has the stress on the second syllable. When a word can
be used as an adjective and a verb, as in the case of ‘perfect’, then the
adjective, /ᶦpɜ:f kt/, has the stress on the first syllable and the verb,
/p ᶦfekt/, has the stress on the second syllable.
Compound nouns usually have stress on the first element, for exam-
ple /ᶦblackbird/ versus /ᶦblack ᶦbird/ and /ᶦgreenhouse/ versus /ᶦgreen
ᶦhouse/. However, some compounds are not written as one word but
140 Pronunciation
nevertheless are treated as one word for stress placement, for exam-
ple, /ᶦrunning shoes/, where ‘running’ is a gerund (verbal noun).
Running shoes are shoes for running in, not shoes which run, which
would be /ᶦrunning ᶦshoes/ stressed like /ᶦrunning ᶦwater/ (water which
runs). In this case ‘running’ is a present participle used as an adjective
(example from Wells 2006: 101).
Unfortunately for learners, however, some compound nouns do not
conform to the principle of stress on the first element, but are ‘end-
stressed’, that is they have their primary stress on the second element
with a secondary stress (not shown in the following examples) on the first
element. These include compounds where the first element names
a material or ingredient (paper ᶦbag, apple ᶦpie etc. in contradistinction
to ᶦpaper clip, ᶦapple tree). Other compound nouns which receive stress
on a syllable in the second word include the names of people (Jim
ᶦPeters), names of institutions such as schools, hotels etc. (The King’s
ᶦHead, Eton ᶦCollege), compounds in which the first element is a time
location (afternoon ᶦtea, April ᶦshowers, summer ᶦholiday, Easter vaᶦca-
tion) and compounds which specify a place, typically road names etc.
(Oxford ᶦRoad, Trafalgar ᶦSquare, New ᶦYork). However, if the place is
a ‘street’ rather than a ‘road’, then the first element is stressed (ᶦOxford
Street). For more examples and discussion see Collins et al. (2019:
133–135); Lorenz (2012: 102–103) and Wells (2006: 100–102).
Adding a suffix to a word may well affect its stress pattern, for
example /ᶦlemon/ versus /lemonᶦade/, /ᶦengine/ versus /enginᶦeer/. See
Collins et al. (2019: 133), Lorenz (2012: 104–107), Gut (2009: 89–93)
and Cruttenden (2014: 248–255).
Utterance stress patterns
As already mentioned, in connected speech words of two or more
syllables generally maintain their stress pattern while monosyllabic
words may be stressed or unstressed. In connected speech stress or
prominence is given to those monosyllabic words which are important.
Generally speaking, these are lexical words (nouns, verbs, adjectives and
adverbs) rather than grammatical words, as illustrated by the following
example from Collins et al. (2019: 136):
I’ve ᶦheard that ᶦJack and ᶦJane ᶦspent their ᶦholidays in Jaᶦmaica.
If you just say the words that receive stress, you get the message. If you
just say the words which are unstressed, you do not get the message. There
are in fact about 40 monosyllabic grammatical words which have two
possible pronunciations. Firstly, they have a ‘strong form’, sometimes
called the citation form, used when the word is said in isolation, is said in
a list, is spoken in a declamatory formal manner or is stressed for special
emphasis in connected speech. Secondly, they have a ‘weak form’, in which
their vowel is reduced, often to / / or /ɪ/. This is the form normally used in
Pronunciation 141
connected speech unless the speech style is slow, formal and declamatory
or the word is to be specially emphasised. Examples, with the strong form
given first, include ‘than’ /ðan/ and /ð n/; ‘at’ /at/ and / t/; ‘were’ /wɜ:/
and /w /; ‘can’ /kan/ and /k n/. Carley et al. (2018: 235) note, however, that
a small number of grammatical words do not take a weak form, for
example, ‘in’, ‘if’, ‘it’, ‘on’ and ‘they’.
The more rapid and informal speech is, the more complete the collapse of
the strong-form vowel towards /ɘ/ or /ɪ/ is likely to be, and in very rapid,
informal speech the vowel virtually disappears. This gradation is illustrated
by the following three ways of saying ‘can’ in ‘I can come’:
(a) ᶦai ᶦkᴂn ᶦkᴧm (b) ᶦai kɘn ᶦkᴧm (c) ᶦaikɘnkɘm
In the first case, which suggests slow and very deliberate speech, each
syllable is given even stress. In the second case, which is more natural in
conversation, ‘can’ is unstressed and in the third case, which suggests more
rapid speech, the whole utterance is given the stress pattern of a three-syllable
word, with primary stress on the first syllable, the second syllable unstressed
and secondary stress on the final syllable (not shown in the example).
Within the pattern of stressed syllables contrasting with unstressed
syllables speakers can impart special emphasis within an utterance by
giving special stress to a certain syllable. For example, in the above case,
if ‘I’ is given very strong stress, this implies, ‘I, unlike certain others, can
come.’ If ‘can’ is given very strong stress, this implies ‘I can come but
I don’t want to’ (or ‘don’t have to’). If ‘come’ is given very strong stress,
then the implication may be, ‘I am willing to turn up but not willing to
do anything.’ Note also that whichever word is stressed, the stressed
vowel will also most likely be lengthened for added effect. Learners may
need practice in doing this.
For lists of weak forms and discussion see Lorenz (2012: 54–56),
Collins et al. (2019: 21–25), Gut (2009: 85–88), Cruttenden (2014:
273–2777) and Carley et al. (2018: 235–237).
Rhythm
The patterning of stressed and unstressed syllables produces rhythm
because, although the number of unstressed syllables between stressed
syllables varies, the stressed syllables are, or at least seem to be, uttered at
regular intervals of time. If there are many unstressed syllables between the
stressed syllables, then speakers appear to skip over them by shortening
their vowels (Collins et al. 2019: 136–138). Stressed syllables thus tend to
be longer than unstressed syllables. In this respect English is a ‘stress-timed’
language. If you were to speak by saying each word just as you would if
you said it in isolation, your speech would sound very unnatural, the way
aliens from outer space were supposed to speak in old films, or the way
142 Pronunciation
early computer-generated speech sounded. Moreover, there would be no
variation in pitch, a feature not mentioned so far.
On rhythm see also Lorenz (2012: 113–115), Roach (2009: 107–110)
and Cruttenden (2014: 271–273).
Intonation
Pitch refers to how high or low on the scale a stressed syllable is said,
relative to the pitch range of the particular speaker, and whether
a movement of pitch occurs on the syllable. In fact there are recognisable
and systematic movements of pitch in connected speech. This is what is
meant by intonation. Intonation patterns differ not only in different
languages but also in different English accents. This makes the underlying
melody of, for instance, Welsh, Scottish, Liverpool or Birmingham
English recognisable (Collins et al. 2019: 141–142). The following brief
description refers to GB pronunciation.
Speech is produced in short units variously called intonation units,
tone units, tone groups or intonation phrases. These units corre-
spond roughly to information units and vary in length from a single
word to a short phrase, clause or short utterance. Speakers tend to
pause at intonation unit boundaries to draw breath and perhaps to
think ahead and plan the next unit. Each intonation unit has its own
intonation pattern or tune, sometimes called its intonation contour.
Whether reading a text, making a speech, or participating in
a conversation, speakers repeatedly have three decisions to make:
how to divide what they say into intonation units (tonality)

which of the stressed (accented) syllables in each intonation unit to
give pitch prominence to (tonicity)
which pitch movement to choose for each intonation unit (tone)
The choices made tend to influence meaning in various ways.

Concerning tonality, the sentence ‘We don’t know who she is’ could
be said as one tone unit, but could also be said as two tone units, either
with the boundary after ‘we’, after ‘don’t’ or after ‘know’. It could also
be said as three tone units, with boundaries after ‘we’ and ‘know’ to
impart different emphases.
Concerning tonicity, intonation units are structured as follows:
PRE-HEAD (optional) + HEAD (optional) + NUCLEUS + TAIL (optional)
Only the nucleus is obligatory. It consists of a stressed syllable and

usually occurs towards the end of the unit. The nucleus, or ‘tonic
syllable’, as it is sometimes called, is given a noticeable pitch movement
so as to make it particularly prominent. It is also normally perceived as
Pronunciation 143
being louder and longer than any other stressed syllables surrounding it.
For the listener, the word in which the nucleus occurs is thus highlighted
as the most important word in the intonation unit. The nucleus may be
preceded by a head containing one or more stressed syllables. The first
of these (or the only one) is termed the ‘onset’ of the head. The head
continues up to and including the syllable before the nucleus (tonic
syllable). Any unstressed syllables before the onset of the head are
termed the pre-head. Any syllables following the tonic syllable constitute
the tail. They tend to decrease in loudness and to follow the direction of
the pitch movement on the tonic syllable. They are by definition
unstressed (Wells 2006: 6–10; Collins et al. 2019: 142–154).
The following utterance could be said in more than one way, but would
probably be said as two intonation units. The break between intonation
units is shown by the vertical line (│). The onset syllable of the head is
shown by (ᶦ). The nucleus or tonic syllable is shown in bold type.
ᶦSend reinforcements │we’re ᶦgoing to advance
The first intonation unit consists of head (‘send rein-‘), nucleus (‘-force-‘)
and tail (-ments’). The second intonation unit consists of pre-head (‘we’re’),
head (‘going to ad-‘) and nucleus (‘-vance’). The onset of the head in the first
intonation unit is ‘send’ and in the second intonation unit the onset is ‘go-’.
Concerning tone, often the onset is at a high pitch (as it most likely would
be for both intonation units in the above example). The head is often said at
a slightly falling pitch, with a rise just before the tonic syllable (nucleus).
Often the pitch movement on the tonic syllable (the nuclear tone) is a falling
one. Falls tends to signal finality (Collins et al. 2019: 156). Rising tones
signify open-endedness (Wells 2006: 9–10). They may be used in an
introductory clause to imply, ‘Wait-for-it, there’s more to come.’ In the
above ‘send reinforcements’ example, both intonation units could be said
with falling intonation on the nucleus, but it would be possible instead to
have a rise in the first intonation unit. This would make it clear that ‘we’re
going to advance’ is the reason for the request for reinforcements.
The most basic distinction is between falling and rising tones on the tonic
syllable, but falls may be subdivided into those which fall from high- to low-
pitch levels (high falls) and those which fall from mid- to low-pitch levels
(low falls). Rises, which are much less common, may be subdivided into low
rises (from low to mid-pitch level) and the less common high rise (from mid-
to high-pitch level). Also important, although much less common again, are
two complex tones: the fall-rise and rise-fall. The fall-rise moves from high,
to low, to mid-pitch levels. The fall-rise is the tone which is often used to
imply something which is not explicitly stated (Wells 2006: 10). The rise-fall
moves from mid-, to high-, to low-pitch levels. It is the least common
nuclear tone and like the rise indicates lack of completion (Wells 2006: 10;
Collins et al. 2019: 146).
144 Pronunciation
Intonation patterns may be said to have at least the following possible
functions, according to Wells (2006: 11–12):
a focussing function
They highlight the most important information.
an attitudinal function
They may indicate the speaker’s attitude to what is being said, for
example boredom, interest, seriousness, disbelief, doubt, irony and
sarcasm (Wells 2006: 11).
a grammatical function
They can show whether an utterance is to be taken as a question,
statement, request or command, for example. Attitudinal and gramma-
tical meanings are often intertwined. Thus, the usual intonation pattern
for statements, commands and ‘wh-’ questions is falling intonation,
which implies finality, while rising intonation, which implies open-
endedness, is often found in ‘yes’/‘no’ questions. However, statements
said with a rising intonation may imply disbelief or incredulity. Collins
et al. (2019: 158–159) note that commands said with rising intonation are
typically softened into requests and ‘wh-’ questions said with rising
intonation are more friendly and obliging (as in ᶦWhat can I do to
↗help you?)
a discourse function
They can show the relationship between clauses, as in the ‘send
reinforcements, we’re going to advance’ example, where the intona-
tion pattern chosen can show that there is an implicit causal relation-
ship between the two clauses.
a psychological function
They make speech easier to understand, to produce and to
remember.
an indexical function
They may help to mark the social or professional status of the
speaker in terms of dominance relations.
On intonation see also Lorenz (2012: 111–117), Gut (2009: 104–135),

Cruttenden (2014: 277–301) and Roach (2009: 119–160).
Classroom Pronunciation Practice
Practice of Sounds in Single Words

If there are phoneme contrasts which are problematic for the learner,
then these must be focussed on as a first priority. Usually only
a relatively small number of phoneme contrasts will be problematic for
a particular learner group. Which phonemes are problematic will depend
largely on the learner’s first language. It is therefore helpful to have an
Pronunciation 145
understanding of the phoneme inventory of the learner’s mother tongue
and how this contrasts with English. For the major world languages these
contrasts are well described and easily accessible. Collins et al. (2019:
182–207) provide outline descriptions for Spanish, French, Italian,
German, Polish and Japanese.
As a general rule, one should first establish whether learners can
perceive the difference in sound between two phonemes by playing
recordings of minimal pair words to them before working on improving
pronunciation. If perception is the problem, then it will be helpful to do
minimal pair exercises of the ‘ship’ or ‘sheep’ sort in which learners have
to make a discriminatory decision. For various exercises see Baker
(2006a, 2006b), Hewings (2004: 51–58), Marks (2007: 12–55), Hancock
(2012: 12–49) and Carley et al. (2018: 95–131, 197–230).
Once learners can perceive the difference, they will sometimes be able
to produce the sounds correctly. If not, then one can adopt an articu-
latory rather than a perceptual approach and explain to them how the
sound is made in terms of place and manner of articulation (see Under-
hill 2005: 2–47 for detailed explanations and activities). However, some
sounds are better suited to being explained in terms of articulation than
others. Generally, articulatory explanation:
will work better for consonants than for vowels

will work better for plosives, fricatives and affricates than for /r/, /w/
and /j/
will work better for sounds made in the front of the mouth than for
those made in the back of the mouth
will work better for sounds made with the lips, teeth, alveolar ridge
and tip of the tongue than with the rest of the tongue
will work well for the voiced–voiceless distinction because learners
can feel when their vocal chords are vibrating by placing finger and
thumb on their Adam’s apple
will work well for nasal sounds because learners can do the finger
tests described earlier in this chapter
Vowel sounds are very difficult to explain by describing the position of

the tongue verbally. It is far better to provide learners with a vowel
diagram so that they can see the position of the problem vowel relative
to the vowel sounds they can already make.
Minimal pair exercises can be used by learners to practise producing
sounds. Working in groups, each learner can in turn say one word of
a suitable minimal pair (for example ‘cap’ and ‘cup’) with the other
learners in the group trying to identify which word was intended. This
can be done as a game with learners being dealt a card from a pack, with
one word on each card.
146 Pronunciation
Learners who have trouble making the GB /r/ may not be helped at all by
the articulatory explanation given earlier in this chapter because the front
part of the tongue appears to make no contact with the alveolar ridge at all
for this sound and most of us do not know where it is when we make the
sound. By contrast, learners who find it difficult to make sounds with
tongue-tip contact will be much helped by an articulatory approach.
Once learners can perceive and produce the phonemic contrasts
acceptably, the next step is to discover if they have problems perceiving
and producing the allophones of the English phonemes, for example
dark and light /l/. A further problem is that learners may carry over
allophones from their own language which do not apply to the English
phoneme. For example, for German-speaking learners the English voiced
plosives /b, d, ɡ/ are generally unproblematic since German has a similar
set. However, in word-final position German devoices its voiced plo-
sives, so that they are pronounced as /p, t, k/. German speakers tend to
carry this over into English, so that /rip/ and /rib/, /bit/ and /bid/, /rick/
and /rig/ may be indistinguishable in their pronunciation. Sometimes
they may not even perceive the sound difference between such words
when they hear them, which can lead to comprehension problems.
After individual phonemes and their allophones can be perceived and
produced acceptably, the next step is to consider sound combinations,
especially consonant clusters, for example /ɸ s/ as in ‘cloths’ or /ðz/ in
‘clothes’. Consonant clusters which occur across word boundaries also
have to be considered. For exercises on consonant clusters, see Hewings
(2004: 73–78; 2007: 20–25), Marks (2007: 56–63), Hancock (2012: 50–59)
and Carley et al. (2018: 276–301).
Most learners can benefit from practice on word stress in words of
more than one syllable. For exercises, see Underhill (2005: 51–57,
151–159) and Hewings (2004: 106–141). Hewings (2004: 113–117) also
presents some useful activities on the stress pattern contrast between
numerals such as /thirᶦteen/ and /ᶦthirty/. It is the stress difference rather
than the suffix itself which mainly distinguishes these pairs for listeners.
Practice of Sounds in Connected Speech

Pronunciation practice in the classroom should not be limited to isolated
words, and most pronunciation practice should focus on connected
speech. Stress, rhythm and intonation as well as weak forms deserve
particular attention. Learners will also benefit from paying attention to
the elisions and assimilations that take place in connected speech. By
learning how to produce rapid speech themselves, learners will be more
easily able to understand rapid speech they are exposed to. For activities
and exercises, see Dalton and Seidlhofer (1994: 27–31), Carley et al.
(2018: 296–301), Hancock (2012: 72–109), Hewings (2004: 94–98,), Hew-
ings (2007: 48–49, 58–69, 84–113) and Marks (2007: 82–89, 98–109).
Pronunciation 147
Ur (2012: 248–50) suggests reciting jazz chants and ‘tongue twisters’. An
example tongue twister would be ‘She sells sea shells on the sea shore and
the shells that she sells are sea shells to be sure’ in order to provide
practice in /s/ versus /ʃ/. Tongue twisters can be recited over and over
again at ever-increasing pace for each recitation, first with the entire text
displayed for the learners, then with individual words being obliterated
one by one until the learners can recite the piece at speed by heart. The
teacher can tap out a beat and progressively raise the pace. Later a member
of the class can take over this task. Stressed syllables and intonation marks
can also be indicated in the text displayed, and then progressively
poems, jokes and
removed. For activities using tongue twisters, limericks, poems
short authentic texts, see also Hewings (2004: 219–230).
For more advanced learners, carefully chosen snippets of, for example,
TV soap operas can be chosen as a basis for showing learners how over-
tones such as incredulity, sarcasm, uncertainty or conviction are imparted
by phonological features, especially intonation. Learners should be encour-
aged to imitate the model and then incorporate these features in meaningful
contexts into their own speech, for example in role-play. Examples would
be ‘What can I do?’ or ‘You’re not going to marry her.’ ‘What can I do’ can
be said in various ways to impart a range of emotions from complete
indifference to extreme self-commitment. ‘You’re not going to marry her’
can be said variously to imply, for instance, ‘Surely you are not going to
marry her’, ‘I forbid you to marry her’ or ‘It has been settled that you are
not going to marry her.’ Learners can then be asked to act out in pairs
a scene culminating in such a sentence used with a specific intonation. As
a follow-up, when learners have mastered the different intonation patterns,
a quiz can be done in which the sentence is said by a learner out of context
and the class has to guess which meaning is intended.
Food for Thought

Q. 1. Look back at Tables 5.1. and 5.2 and transcribe the words below
using the symbols given.
tough bough cough dough hiccough thorough

laugh through heard beard bird dead
bead bed deed meat great threat
suite straight bat debt moth mother
both bother broth brother here there
dear fear bear pear does rose
lose goose choose cork front word
ward font front worth sword do
go thwart cart lice rhythm swat
148 Pronunciation
You probably found the vowels sounds more difficult to
transcribe than the consonant sounds because there is a more
consistent correspondence between sound and writing for con-
sonants than for vowels. Which do you think are the vowel
sounds where there is a reasonably consistent correspondence
between spelling and pronunciation, and which are the highly
problematic ones? Are monophthongs more problematic than
diphthongs or vice versa? Are there any patterns? Could you
offer learners any help or advice about letter-to-sound
correspondences?
Q. 2. Imagine you have learners who pronounce ‘chess’ and ‘jazz’ in
such a way that the two words sound almost the same. First
transcribe the two words and describe for yourself the pronun-
ciation problem for each of the three phonemes. Then decide
what remedial action you would take with the learners. Act out
a brief remedial pronunciation session for ‘chess’ and ‘jazz’,
with some colleagues playing the part of the learners, and after-
wards reflect with your colleagues on how successful the ses-
sion was.
Q. 3. (Group task) Consider this utterance (example taken from Wells
2006: 9):
We’re planning to fly to Italy.
Say this as one intonation unit and then identify the component
parts of pre-head (if present), head (if present), nucleus and tail (if
present). If there is a head, mark the onset syllable. Decide what
tone you gave to the nucleus and mark this too. Experiment now
with other ways of saying the utterance, involving shifting the
nucleus and/or shifting the tone placed on the nucleus. See if you
can match these to specific contexts. (If you find this difficult, do
the exercise the other way around; that is, imagine some situations,
say the utterance as you would naturally say it in context, and then
decide what intonation patterns you have used.)
Finally, see if you can write two brief dialogues whose con-
texts demand two different intonation patterns on a particular
sentence common to both dialogues. If you have access to
learners, try out your dialogues with them. See how well they
do before and after you offer any help on how to say the
sentence in question.
Further Reading
Odden, D. (2008). Introducing Phonology. Cambridge: Cambridge University Press.
Pronunciation 149
References
Abercrombie, D. (1991). Fifty Years in Phonetics. Edinburgh: Edinburgh University
Press.
Baker, A. (2006a). Tree or Three? An Elementary Pronunciation Course. Second
edition. Cambridge: Cambridge University Press.
Baker, A. (2006b). Ship or Sheep? An Intermediate Pronunciation Course. Third
Carley, P., Mees, I. M. and Collins, B. (2018). English Phonetics and Pronunciation
Practice. London and New York: Routledge.
Carr, P. (1999). English Phonetics and Phonology: An Introduction. Oxford: Blackwell.
Collins, B., Mees, I. M. and Carley, P. (2019). Practical English Phonetics and
Phonology: A Resource Book for Students. Fourth edition. London and New York:
Routledge.
Cruttenden, A. (2014). Gimson’s Pronunciation of English. Eighth edition. London
and New York: Routledge.
Dalton, C. and Seidlhofer, B. (1994). Pronunciation. Oxford: Oxford University
Press.
Gut, U. (2009). Introduction to English Phonetics and Phonology. Frankfurt am Main:
Peter Lang.
Hancock, M. (2012). English Pronunciation in Use: Intermediate. Second edition.
Hewings, M. (2004). Pronunciation Practice Activities. Cambridge: Cambridge Uni-
versity Press.
Hewings, M. (2007). English Pronunciation in Use: Advanced. Cambridge: Cam-
bridge University Press.
Jones, D. (1926). An English Pronouncing Dictionary. Second edition. London: Dent.
(First edition 1917).
Lorenz, F. (2012). Basics of Phonetics and English Phonology. Berlin: Logos.
Marks, J. (2007). English Pronunciation in Use: Elementary. Cambridge: Cambridge
University Press.
Roach, P. (2009). English Phonetics and Phonology: A Practical Course. Fourth
Shaw, G. B. (1916). Pygmalion: A Romance in Five Acts. New York: Brentano.
Underhill, A. (2005). Sound Foundations: Learning and Teaching Pronunciation.
Revised edition. Oxford: Macmillan.
Ur, P. (2012). Vocabulary Activities. Cambridge: Cambridge University Press.
Wells, J. C. (2006). English Intonation: An Introduction. Cambridge: Cambridge
University Press.
Windsor-Lewis, J. (1972). A Concise Pronouncing Dictionary of British and American
English. London: Oxford University Press.
Dictionary
(2011). Oxford Advanced Learner’s Dictionary of Current English. Compiled by
A. S. Hornby. Managing editor: Joanna Turnbull. Eighth edition. Oxford:
6 Grammar
Inflectional Morphology
What Is Inflectional Morphology?

Inflectional morphology refers to how the form of a word may change to
express grammatical meaning. In English inflection usually involves
adding a suffix (word ending) to the base form, as in ‘give’ (base form),
‘gives’, ‘giving’, ‘given’. But sometimes it involves a vowel change, as in
‘give’ – ‘gave’. Technically, such a vowel change is ‘vowel gradation’
(Crystal 1996: 72) rather than inflection, but I will include it under
inflection in this chapter. Inflection never changes the word class of the
original word: ‘gives’, ‘gave’, ‘giving’ and ‘given’ are all forms of the verb
‘give’. For more on inflectional morphology see Yule (2017: 75–77).
In comparison to some other languages: Latin, Russian and German,
for example, English has a slimmed-down system of inflections. It was
not always like this. Old English (c. 700–1100 AD) was much richer than
Modern English in inflection. As English developed over the centuries,
however, it gradually lost much of its inflectional morphology and came
to rely increasingly on word combinations and word order to express
grammatical meaning. (See Quirk and Wren 1957: 19–58 for more on
Old English inflections.)
In English today, inflection occurs in verbs, nouns and adjectives as
well as in personal, possessive, reflexive, relative and interrogative pro-
nouns. There is also the modification of the indefinite article ‘a’ to ‘an’,
before nouns beginning with a vowel sound. This is really just
a pronunciation modification which is shown in the spelling.
Verb Inflection
In English, verb inflectional morphology marks person and number as
well as tense and aspect. The ‘base form’ of the verb is the uninflected
form listed in the dictionary (‘play’, ‘give’, ‘be’, ‘have’ etc.). It is always
identical with the infinitive (without ‘to’). With the exception of the verb
Grammar 151
‘to be’, it is also the form used for the simple present tense in the first
person singular, as in ‘I play’, ‘I give’, ‘I have’ etc.
Third Person Present Singular

In the present tense only the third person singular of the present
tense is marked by being assigned the suffix ‘–s’ (as in ‘he/she/it plays’)
or ‘-es’ (as in ‘washes’). There is one great exception and one slight
exception to this pattern. The verb ‘to be’ is the great exception and
does not follow the pattern at all: ‘I am’, ‘you are’ ‘he/she/it is’, ‘we/
you/they are’. The verb ‘to have’ is the slight exception to the pattern
in that instead of adding the suffix ‘-s’ to form ‘he/she/it *haves’,
reduction to ‘has’ occurs.
The pronunciation of ‘-s’ varies according to whether it follows a voiced
or voiceless sound. The suffix is pronounced /z/ following a vowel sound, as in
‘plays’ /pleiz/, or a voiced consonant, as in ‘reads’ /ri:dz/, or when it is syllabic,
as in ‘washes’ /wɒʃɪz/, shown in writing as ‘-es’. It is pronounced /s/ following
a voiceless consonant as in ‘sits’ /sɪts/. Most learners can make the necessary
distinction in pronunciation with a little practice after it has been pointed out
to them.
First and Third Person Singular Past of ‘To Be’

In the past tense, verbs have a single form for all persons, singular and
plural. This applies even to the verb ‘have’, which is always ‘had’ in
the past. The verb ‘to be’ is the great exception: it takes the form ‘was’
for first and third person singular (‘I/he/she/it/was’) and ‘were’
for second person singular (‘you were’) plus all plural persons (‘we,
you, they were’).
Tense (Present versus Past)

The basic tense distinction in English is between present (‘give’, ‘play’)
and past (‘gave’, ‘played’). It is marked by inflection of the base form of
the verb. The simple present tense form is identical with the base form
of the verb apart from the third person singular (see above). The way the
past tense is formed varies according to whether the verb is ‘regular’ or
‘irregular’. Regular verbs form their past tense by adding ‘-ed’ to the base
form (‘play’ – ‘played’). Irregular verbs depart more or less radically
from this scheme and there are a number of different patterns that can be
discerned. There are about 170 irregular verbs. These include the two
most frequently used verbs, ‘to be’ and ‘to have’, and many other verbs
of frequent occurrence such as ‘bring’, ‘drink’, ‘eat’, ‘give’, ‘lie’, ‘put’,
‘rise’, ‘send’ ‘sit’, ‘sleep’, ‘stand’, ‘take’ and ‘wake’.
152 Grammar
Aspect (Progressive versus Perfect)
Most verbs are ‘action verbs’ and English has two verb participles, the
progressive participle and the perfect participle, which show whether the
action is to be regarded as in progress (progressive aspect) or completed
(perfective aspect). (In Latin perfectum means ‘completed’.) The progres-
sive participle is formed by adding ‘-ing’ to the base form, regardless of
whether the verb is regular or irregular (‘play’ – ‘playing’; ‘give’ –
‘giving’). The perfect participle of regular verbs is identical with the past
tense form (‘played’). For irregular verbs there are a number of patterns:
for ‘give’ the past participle is ‘given’.
Finite and Non-Finite Verb Forms

The perfect and progressive participles together with the infinitive are
non-finite parts of the verb. Unlike the present and past tense forms
(‘play’ – ‘played’; ‘give’ – ‘gave’), which are finite forms of the verb, they
cannot take a subject (I give but not *I giving’ or *I given). There is one
other non-finite part of the verb, namely the gerund. A gerund is
a ‘verbal noun’ formed from the base form of the verb by adding ‘-ing’,
as in ‘Seeing is believing’. The form of the gerund is always identical in
form with the present participle (see Crystal 1996: 155; Greenbaum and
Quirk 1990: 312–313, 378; Huddleston and Pullum 2005: 32–33; Carter
and McCarthy 2006: 905).
Irregular Verb Patterns

To be able to use any irregular (or regular) verb, one needs to know only
its base form, past tense form and perfect participle. There are partial
patterns of systematicity among irregular verbs. The following seven
verb ‘families’ do not pretend to cover all possibilities but may help
learners:
1. verbs which form their past tense regularly only to depart from
regular inflection in forming the perfect participle (‘swell’, ‘swelled’,
‘swollen’).
2. verbs which repeat the base form for both past tense and perfect
participle (‘cut’, ‘cut’, ‘cut’)
3. verbs which change the medial vowel for the past tense and revert to
the base form for the perfect participle (‘run’, ‘ran’, ‘run’)
4. verbs which have three different vowels for the three forms (‘sink’,
‘sank’, ‘sunk’)
5. verbs which change both the medial vowel and the final consonant
sound to form the past tense and then repeat this form for the
perfect participle (‘bring’, ‘brought’, ‘brought’)
Grammar 153
6. verbs which change the medial vowel for the past tense and then add
‘(e)n’ to the base form for the perfect participle (‘give’, ‘gave’, ‘given’)
7. verbs which have two alternative forms for both the past tense and
perfect participle, one regular and one irregular (‘smell’, ‘smelled’ (or
‘smelt’), smelled (or ‘smelt’)
See Crystal (1996: 73) for a comprehensive classification.
Modal Auxiliary Verbs

All that has been said about verb inflectional morphology so far applies to
lexical verbs and does not apply to the nine modal auxiliary verbs, ‘can’,
‘could’, ‘may’, ‘might’, ‘will’, ‘would’, ‘shall’, ‘should’ and ‘must’. The
modal verbs are always either followed by the ‘bare’ infinitive (base form)
of a lexical verb, that is without ‘to’, as in ‘I can help’, or, if they stand
alone, a following ‘bare’ infinitive is ‘understood’ as in, ‘Can you help?’
‘Yes, I can.’ Modal verbs are never inflected. They always use the base
form even in the third person singular (‘he can’, ‘she could’, ‘it may’ etc.),
nor do they have any non-finite forms (infinitives, participles and ger-
unds). For example, one cannot say *‘to must’, *‘musting’ or *‘musted’.
To express the past meaning of ‘must’ the past tense of ‘have to’ (‘had
to’) is used. Concerning the other eight modals, in reported speech
‘could’ functions as the past tense of ‘can’, ‘might’ as the past of ‘may’,
‘would’ as the past tense of ‘will’, and ‘should’ as the past tense of ‘shall’.
However, in other uses, particularly requests such as, ‘Can/could you
lend me your bike?’ the meaning difference between ‘can’ and ‘could’,
‘may’ and ‘might’, ‘will’ and ‘would’ is often one of tentativeness rather
than of ‘past’ versus ‘present’ (Crystal 1996: 103).
In addition to the modal verbs there are so-called ‘catenative verbs’,
such as ‘appear to’, ‘seem to’, ‘have to’, ‘want to’, ‘need to’, which
express modal meaning but generally behave morphologically like lexical
verbs except for the fact that, like modal verbs, they have no passive
forms (see Carter and McCarthy 2006: 401–403).
On verb inflection generally, see Crystal (1996: 70–77), Greenbaum
and Quirk (1990: 24–40), Huddleston and Pullum (2005: 29–39, 273–277)
and Carter and McCarthy (2006: 405–429).
Noun Inflection
Noun inflection in English is limited to number (singular versus plural)
and also to distinguishing the genitive case from the unmarked case. See
Crystal (1996: 108–121), Greenbaum and Quirk (1990: 93–99), Huddle-
ston and Pullum (2002: 340–349) and Huddleston and Pullum (2005: 82,
277–279).
154 Grammar
Noun Plural Inflection
The usual way of forming the plural of nouns in writing is by adding the
suffix ‘-s’. The way the ‘-s’ is pronounced is influenced by the preceding
sound. If the singular ends in a voiceless consonant, it is pronounced /s/,
as in ‘cats’; and if it ends in a voiced consonant or a vowel it is
pronounced /z/, as in ‘dogs’ and ‘zebras’. If the singular already ends in
the sound /s/, /z/, /ʃ/, /tʃ/ or /dʒ/, the suffix is made syllabic because it
would otherwise be impossible to pronounce the consonants together. In
writing, this is shown as ‘-es’, pronounced /iz/, as in ‘races’, ‘mazes’,
‘crashes’, ‘churches’, ‘badges’ (see Greenbaum and Quirk 1990: 93–94).
A minor complication is that some very few nouns which end in /f/
change the consonant to its voiced equivalent for the plural and add /z/,
as in ‘leaf’, ‘leaves’, rather than *‘leafs’. However, the change is shown in
the spelling (Crystal 1996: 114; Greenbaum and Quirk 1990: 95).
Seven nouns change their vowel to form the plural: ‘man’ – ‘men’,
‘woman’ – ‘women’, ‘foot’ – ‘feet’, ‘tooth’ – ‘teeth’, ‘mouse’ – ‘mice’,
‘louse’ – ‘lice’ (Crystal 1996: 114; Greenbaum and Quirk 1990: 95). It is
worth spending a little time on the pronunciation of ‘woman’ – ‘women’
when introducing it to beginners before they ever see the written form. It
is a very difficult plural because both vowels change: the first, stressed
vowel changes from /ʊ/ to /ɪ/ and the second, unstressed vowel changes
from / / to /ɪ/: /wʊm n/ – /wɪmɪn/. The spelling is particularly unhelpful
for learners.
There are a few difficult words of foreign origin such as ‘appendix’ –
‘appendices’ which maintain the plural of the original language, Latin in
this case (Crystal 1996: 115; Greenbaum and Quirk 1990: 96–97). There
are also a few exceptional plurals such as ‘oxen’, ‘children’, ‘cattle’, which
maintain earlier Germanic plural endings. Some very few nouns, such as
‘sheep’, have the same form for both singular and plural (‘one sheep’, ‘two
sheep’). There are also some nouns, sometimes called ‘non-count nouns’,
which have no plural, such as ‘money’, ‘music’, ‘milk’ and ‘butter’ (Crystal
1996: 112–113). There are also a few problematic nouns from Latin which
hover uneasily between the Latinate plural and the regular English plural
(Crystal 1996: 115). Do you personally say ‘croci’ or ‘crocuses`, ‘syllabi’ or
‘syllabuses’? (See also Greenbaum and Quirk 1990: 97–99).
Noun Genitive Inflection

The genitive case is distinguished from the unmarked case in the singular in
both speech and writing (boy – boy’s). In the plural genitive the distinction
is made only in writing (boys – boys’). As an alternative to the genitive, the
‘of’ construction’ can be used, as in ‘the father of the boy’ and ‘the father of
the boys’. Thus, English has both an inflectional and a syntactic means of
expressing possession. The genitive construction tends to be preferred for
Grammar 155
persons and animate nouns and the ‘of’ construction for inanimate nouns
(compare ‘John’s foot’ versus ‘the foot of the mountain’) but much depends
on the linguistic context of the sentence in which the phrase occurs.
On the genitive, see Crystal (1996: 124–127), Greenbaum and Quirk
(1990: 102–107), Huddleston and Pullum (2005: 279–280) and Carter and
McCarthy (2006: 361).
Adjective Inflection
To form the comparative and superlative forms of one-syllable adjectives,
the inflectional suffixes ‘er’ and ‘est’, as in ‘grand’, ‘grander’, ‘grandest’ are
added. For adjectives of three or more syllables, ‘more’ and ‘most’ are
combined with the adjective, as in ‘beautiful’, ‘more beautiful’, ‘most
beautiful’. This latter method is syntactic not inflectional because words
are added. Adjectives of two syllables can usually take both forms, but
some second syllables, such as ‘-id’ prefer the syntactic form and some, such
as ‘-y’ prefer the inflectional form: ‘stupider’ sounds rather odd compared
to ‘more stupid’, and ‘prettier’ sounds slightly more natural than ‘more
pretty’. There are very few complete exceptions to the system, the most
obvious ones being the irregular ‘good’, ‘better’, ‘best’ and ‘bad’, ‘worse’,
‘worst’. However, the adjective ‘old’ has in addition to the regular compara-
tive and superlative forms ‘older’, ‘oldest’ the forms ‘elder’, ‘eldest’, which
are used only for family members, as in ‘my elder brother’, ‘my eldest
sister’, and are optional rather than obligatory.
On adjective inflection see Crystal (1996: 164–165), Greenbaum and
Quirk (1990: 152–156), Huddleston and Pullum (2005: 280–281) and
Carter and McCarthy (2006: 759–773).
Pronoun Inflection
Personal Pronouns
The personal pronouns ‘I’, ‘we’, ‘he’ ‘she’ and ‘they’ have an objective
case, namely, ‘me’, ‘us’, him, ‘her’, and ‘them’. However, ‘you’ and ‘it’
do not have separate inflected forms for the objective case (see Carter
and McCarthy 2006: 376–382). The possessive pronouns, ‘mine’, ‘yours’,
‘his’, ‘hers’, ‘its’ (actually rarely used as a possessive pronoun), ‘ours’ and
‘theirs’ may be seen as genitive forms of the corresponding personal
pronouns (see Carter and McCarthy 2006: 382–383). The reflexive
pronouns (‘myself’, ‘yourself’, ‘himself’, ‘herself’, ‘itself’, ‘ourselves’,
‘yourselves’, ‘themselves’) represent further inflectional variation on the
personal pronouns (see Carter and McCarthy 2006: 384–386 and also
Huddleston and Pullum 2005: 105 for an overview in tabular form).
Idiomatic expressions like ‘a friend of mine/yours/his/hers/ours/theirs’
etc. (rather than the unidiomatic and maybe ungrammatical ‘a friend of
156 Grammar
me/you/him/her/us/them’ etc.) are unusual in that they combine ‘of’
followed by the genitive case rather than the objective case (Carter and
McCarthy 2006: 360). With nouns rather than pronouns, ‘of’ can be
followed by either the objective or the genitive case, as in ‘a friend of my
uncle’ versus ‘a friend of my uncle’s’ (Crystal 1996: 126). Many people
perceive a slight meaning difference, however, between the two expres-
sions, the latter meaning ‘a friend from among my uncle’s friends’.
Relative and Interrogative Pronouns

The relative and interrogative pronoun ‘who’ (singular and plural), used
only to refer to persons, is inflected for the genitive (‘whose’). In the
objective case ‘whom’ is an alternative to ‘who’ as relative pronoun in
formal, mostly written contexts, as in ‘the lady/ladies who(m)
I mentioned/to whom I referred’. However, especially in less formal
contexts, the objective relative pronoun may be, and often is, omitted
altogether, as in ‘the lady/ladies I mentioned’, ‘the lady/ladies I referred
to’ (see Crystal 1996: 140–141; Carter and McCarthy 2006: 386–389).
‘Whom’ as an alternative to ‘who’ as an interrogative rather than
a relative pronoun in the objective case is very formal indeed, as in
‘Whom did you mention?’ and ‘To whom did you refer?’
On pronoun inflection generally, see also Crystal (1996: 146–155),
Greenbaum and Quirk (1990: 111, 114) and Huddleston and Pullum
(2005: 102–110).
Syntax
What Is Syntax?
Syntax refers to the way words are combined to form larger units such as
phrases, clauses and sentences. Word order is very important for sen-
tence meaning in English because English has lost much of its inflectional
noun morphology. There are no inflectional morphemes to show
whether a noun is a subject, direct object or an indirect object in
a sentence. English is an SVO language, that is to say, in simple
declarative sentences (statements consisting of one clause) in English the
usual order of elements is subject + verb + object. Thus, in the sentence
‘The man bit the dog’ the word order makes it clear that the man does
the biting and the dog gets bitten even though this is a rather unlikely
event. In SVO sentences, grammatical subjects are ‘actors’ and gramma-
tical objects are ‘patients’ or ‘recipients’, that is to say they are acted
upon. For more on SVO word order see Crystal (1996: 36–38), Green-
baum and Quirk (1990: 205), Huddleston and Pullum (2005: 65–71) and
Carter and McCarthy (2006: 778–779).
Grammar 157
The Noun Phrase
The subject slot in an SVO sentence has to be filled by a noun phrase
(NP). A NP consists minimally of a ‘head’, which must be a noun or
pronoun. Word order within the NP is rather fixed. There is a closed set
of ‘determiners’ which may precede the NP head:
a/the/my/your, etc./some/this/that dog
Then there is a slot for an open set of ‘premodifiers’, mainly adjec-
tives, the order of which is partly fixed, with colour words, for example,
coming close to the following noun:
the famous, old, brown dog
After the head of the NP there is a slot for post-modification, often by
a participial phrase (‘lying peacefully’), prepositional phrase (‘in the
yard’) or both:
the famous, old, brown dog lying peacefully in the yard
Alternatively, a relative clause may post-modify a NP head:
the famous, old, brown dog which was lying peacefully in the yard
On noun-phrase structure see Crystal (1996: 104–107), Greenbaum
and Quirk (1990: 363–364), Huddleston and Pullum (2005: 82–84) and
Carter and McCarthy (2006: 318–328). On adjective order before nouns
see Scrivener (2010: 64–65).
The Verb Phrase

The verb phrase (VP) consists minimally of just a verb, as in ‘slept’, but can
also include up to four accompanying auxiliary verbs, all of which proceed
the lexical verb (the ‘head’), as in ‘She may have been being blackmailed’
(Crystal 1996: 68; Greenbaum and Quirk 1990: 41–43; Carter and
McCarthy 2006: 395–403). Some grammarians, especially those working
within the tradition of transformational generative grammar (TG), extend
the term VP to also include any other obligatory elements that follow the
verb in the sentence, for example in ‘The dog bit the man’ the VP would
then be regarded as ‘bit the man’, which would in turn be subdivided into
verb (V) plus NP (see Huddleston and Pullum 2005: 64–65). This definition
of the VP corresponds to what was called the ‘predicate’ in traditional
grammar. It is not the one adopted here. In this section, three ways in
which the form of the VP expresses grammatical meaning will be consid-
ered. These are tense, aspect and voice.
Tense
In English there are only two tenses formed by inflection of the verb,
namely the simple present tense (Mary sings in the choir) and the simple
past tense (Mary sang in the choir) (Crystal 1996: 92; Greenbaum and
Quirk 1990: 47–48; Carter and McCarthy 2006: 926). The so-called
158 Grammar
‘simple future’ and ‘simple conditional’ tenses in English are actually
composite forms made up of the modal auxiliaries ‘will’ and ‘would’,
respectively, followed by the base form of the lexical verb (Mary will/
would sing). For the sake of convenience I will refer to such composite
tense forms as ‘tenses’ too. These four tense forms are all unmarked for
aspect.
Aspect
‘Aspect’ refers to how verbs show (are marked for) completion (perfec-
tive aspect) or lack of completion (progressive aspect) of the action.
A further aspect option available is the ‘perfective progressive’, arrived
at by combining perfective and progressive aspect marking. There are
thus four aspects to the verb: unmarked, perfective, progressive and
perfective progressive. The term ‘simple tenses’ is used for tenses which
are not marked for progressive aspect (Crystal 1996: 98). Each of the
four aspects has four tenses (present, past, future and conditional). This
makes 16 tenses in all.
UNMARKED ASPECT
The four unmarked tenses are simple present (Mary sings), past
(sang), future (will sing) and conditional (would sing).
PERFECTIVE ASPECT
The four unmarked tenses can each be marked for perfective aspect
by combining the appropriate tense of the ‘have’ with the perfect
participle of the lexical verb (sung). This produces the simple present
perfect (Mary has sung), past perfect (had sung), future perfect will
have sung and conditional perfect tenses (would have sung).
PROGRESSIVE ASPECT
The four unmarked tenses can each be marked for progressive aspect by
combining the appropriate tense of ‘be’ with the progressive participle
of the lexical verb (singing). This produces the present progressive
(Mary is singing), past progressive (was singing), future progressive
(will be singing) and conditional progressive (would be singing).
PERFECTIVE PROGRESSIVE ASPECT
The four unmarked tenses can each be marked for both perfective
and progressive aspect by combining the appropriate tense of ‘have’
with the past participle of ‘be’ (been) followed by the present
Grammar 159
participle of the lexical verb (singing). This produces the present
perfect progressive (Mary has been singing), past perfect progres-
sive (had been singing), future perfect progressive (will have been
singing) and conditional perfect progressive (would have been
’).
singing).
Voice
The above 16 tenses are all in the active voice. ‘Voice’ (active or passive)
refers to whether the grammatical subject of the sentence is the doer
(active voice) or the recipient of the action (passive voice), as in ‘she sang
a sea shanty’ (active) versus ‘a sea shanty was sung’ (passive). Each of the
16 active voice tense forms has a corresponding passive voice tense form.
In the passive as in the active voice there are four aspects: unmarked,
perfect, progressive and perfective progressive. Within each aspect there
is a present, past, future and conditional tense:
UNMARKED ASPECT IN THE PASSIVE
• The four unmarked passive tenses are formed by combining the

appropriate simple tense of ‘be’ with the perfect participle of the
lexical verb (sung). This produces the simple present (the sea shanty
is sung), past (was sung), future (will be sung) and conditional tenses
(would be sung) in the passive voice.
PERFECTIVE ASPECT IN THE PASSIVE
• The four perfective passive tenses are formed by combining the appro-
priate tense of ‘have’ with the perfect participle of ‘be’ (been) followed
by the perfect participle of the lexical verb (sung). This produces the
present perfect (the sea shanty has been sung), past perfect (had been
sung), future perfect (will have been sung) and conditional perfect
tenses (would have been sung) in the passive voice.
PROGRESSIVE ASPECT IN THE PASSIVE
• The four progressive tenses are formed by combining the appropri-

ate tense of the auxiliary ‘be’ with the present participle of ‘be’
(being), followed by the perfect participle of the lexical verb (sung).
This produces the present progressive (the sea shanty is being sung),
past progressive (was being sung), future progressive (will be being
sung) and conditional tenses (would be being sung) in the passive
voice.
160 Grammar
PERFECTIVE PROGRESSIVE ASPECT IN THE PASSIVE
The four perfective progressive passive tenses are formed by combining

the appropriate tense of the auxiliary ‘have’ with the perfect participle
of ‘be’ (been), followed by the progressive participle of the auxiliary ‘be’
(being), followed by the perfect participle of the lexical verb (sung).
This produces the present perfect progressive (the sea shanty has been
being sung), past perfect progressive (had been being sung), future
perfect progressive (will have been being sung) and conditional perfect
(would have been being sung) tenses in the passive voice.
For more on the active and passive voice see Crystal (1996: 88–91),
Greenbaum and Quirk (1990: 44–46), Huddleston and Pullum (2005:
240–246) and Carter and McCarthy (2006: 793–802, 929).
Note 1: Order of auxiliary verbs in composite tense forms

When a verb is used in combination with a lexical verb, it is called an
‘auxiliary verb’. A distinction is usually made between the nine modal
auxiliary verbs, already discussed, and the three ‘primary auxiliary verbs’,
‘be’, ‘have’ and ‘do’ (Crystal 1996: 69). ‘Do’ has not yet been mentioned, but
it is combined with lexical verbs to form questions and negatives (‘I don’t
dance’; ‘does she drive?’) and will be dealt with in due course. Unlike the
modal auxiliaries, the three primary auxiliaries can also be used as lexical
verbs themselves (‘they are happy’; ‘she has a car’; ‘he is doing the washing’).
In the active voice up to three, and in the passive voice up to four
auxiliaries can occur before the lexical verb, which in one of its forms
always occupies the final slot. ‘Will’ and ‘would’, if present, always
occupy the first slot in the composite tense form. Furthermore, ‘will’ and
‘would’ can always be replaced by another modal auxiliary with
a corresponding change of meaning. No modal auxiliary can co-occur
with another, nor can ‘have’ and ‘had’ occur together as auxiliaries. The
auxiliary combinations ‘am/is/are being’ and ‘been being’ occur only in the
passive and only immediately before the perfect participle of the lexical
verb (‘he may be being followed’; ‘she may have been being blackmailed’).
Note 2: The distinction between tense and time

The real problem for learners is how to use the tenses appropriately.
There is no close correspondence between the tense labels ‘past’, ‘present’
and ‘future’ and the time expressed. Consider ‘If I had a hammer …’, ‘It is
her birthday next Thursday’, ‘Oil floats on water’, ‘She is arriving on
Wednesday’. In particular, in addition to expressing future meaning ‘will’
has a variety of modal meanings. In ‘boys will be boys’ it expresses
inevitability; in ‘will you please be quiet now’ it is an insistent request; in
‘she’ll be landing in San Francisco at this very moment’ it is an assump-
tion. These are all modal rather than temporal meanings of ‘will’.
Grammar 161
For more on the modal auxiliary verbs, including modal meanings of ‘will’,
see Crystal (1996: 102–103), Greenbaum and Quirk (1990: 60–68), Huddle-
ston and Pullum (2005: 54–58) and Carter and McCarthy (2006: 638–657).
For more on tense, aspect and time see Crystal (1996: 92–99), Green-
baum and Quirk (1990: 47–60), Huddleston and Pullum (2005: 42–53)
For more on auxiliary verbs see Crystal (1996: 76–81), Greenbaum and
Quirk (1990: 34–40), Huddleston and Pullum (2005: 37–42) and Carter
and McCarthy (2006: 598–636).
Adverbials
Grammarians use the term ‘adverbial’ to cover not only single-word
adverbs, as in ‘She handled the package carefully’, but also adverb
phrases, as in, ‘She handled the package very carefully indeed’,
a prepositional phrases (PP), as in, ‘She handled the package with extreme
care’ or even NPs functioning as adverbs, as in ‘She handled packages
carefully that summer’. The term adverbial thus refers to a functional
rather than a structural category. Some verbs demand an adverbial to
complete their sense (that is as complement), for example the verb ‘lie’,
which requires a place adverbial. Thus, *‘The package lay’ is not acceptable
but ‘The package lay on the shelf’ is. Any sentence, however, can be
expanded by the addition of one or more optional adverbials to provide
additional information about the verb, especially concerning time,
manner, place and frequency. Single-word adverbs of frequency, such as
‘never’, ‘always’, ‘seldom’, ‘rarely’, ‘often’, ‘sometimes’, ‘occasionally’, are
usually placed immediately before the verb and after the subject. Other
adverbials tend to be placed either at the beginning of the sentence or
following the verb. The usual order is manner, place, time.
The following example shows how a sentence can be expanded by
adding optional adverbials:
The dog often (frequency adverbial) slept happily (manner adverbial)

in the yard (place adverbial) that summer (time adverbial).
For more on adverbials see Crystal (1996: 166–179), Greenbaum and

Quirk (1990: 158–187), Huddleston and Pullum (2005: 78–80) and Carter
and McCarthy (2006: 578–596).
Verb Complementation
Verb complementation refers to what elements, if any, are required to
complete the meaning of a verb in a simple declarative sentence, that is,
a sentence of one clause in the form of a statement. The possibilities are
object (O) and complement (C) and various combinations thereof. The term
162 Grammar
‘verb complementation’ is, perhaps rather confusingly, used to cover both
objects and complements as elements following the verb (see Huddleston
and Pullum 2005: 70, 73). Objects may either be direct (accusative case) or
indirect (dative case). If there is only one object, then it is direct. Each verb
has its own set of complementation possibilities. There are five basic
complementation patterns in terms of compulsory elements:
SV The dog (S) slept (V)

SVO The dog (S) bit (V) the man (O)
The dog (S) barked at (V) the man (O)
SVC The dog (S) was (V) happy (C)
The dog (S) was (V) an Alsatian (C).
The dog (S) was (V) in the yard (C)
SVOO The dog (S) gave (V) the man (O) a lick (O)
SVOC The dog (S) made (V) the man (O) happy (C)
The dog (S) made (V) the man (O) a famous figure (C)
The dog (S) got (V) the man (O) into trouble (C)
See also Crystal (1996: 38), Greenbaum and Quirk (1990: 204–207),
Carter and McCarthy (2006: 527–528) and Huddleston and Pullum
(2005: 77–78). Classifications vary slightly from scholar to scholar.
The verb is the pivotal element in a sentence, both semantically and
grammatically, and it is the ‘transitivity status’ of the verb which determines
which, if any, elements must follow it to complete its meaning and thus
complete the sentence (Crystal 1996: 38, 48–51; Greenbaum and Quirk
1990: 204–205; Huddleston and Pullum 2005: 63–78; Carter and McCarthy
2006: 504–506).
The following verb-transitivity possibilities exist, and some verbs can
be used in more than one way:
Intransitive verbs (such as ‘sleep’, ‘die’, ‘smile’ and ‘laugh’) require

neither an object nor a complement, nor do they permit one. Any-
thing that follows them is an optional adverbial. The sentence
pattern is SV (‘the dog slept’).
Transitive verbs (such as ‘bite’, ‘give’, ‘make’ and ‘find’) can be sub-
divided into monotransitive, ditransitive and complex transitive verbs:
1. Monotransitive verbs
Verbs such as ‘bite’ and ‘find’ take one (direct) object. The sentence
pattern is SVO (‘the dog bit the man’). This category includes
prepositional verbs, which take a prepositional object (‘the dog
barked at the man’). (See Greenbaum and Quirk 1990: 338–339.)
2. Ditransitive verbs
Verbs such as ‘give’ take a direct and an indirect object. The
indirect object comes before the direct object. The sentence
pattern is SVOO (‘the man gave the dog a bone’). (The distinction
Grammar 163
between direct and indirect objects corresponds to the difference
between the accusative and dative cases in Latin grammar.)
With some, but not all, ditransitive verbs an alternative is to
replace the indirect object by a PP introduced by ‘to’ or ‘for’ and
to place this after the direct object (‘the man gave a bone to the
dog’). Huddleston and Pullum (2005: 72) analyse this pattern
variation as SVOC (see below). With ‘give’ this option is not
available in some non-literal or idiomatic senses (*‘the dog gave
a lick to him’, *‘she gave a dressing-down to me’).
3. Complex transitive verbs
Verbs such as ‘make’, ‘get’ and ‘put’ take a direct object plus
a complement. The sentence pattern is SVOC (‘the dog made the
man a millionaire’; ‘the dog made the man happy’; ‘the man put the
dog on the table’). The object complement provides additional
information about the object, which is necessary to complete the
sense. As the above examples show, C can be an NP, an adjective, an
adjective phrase (AdjP), or an adverbial – usually a place adverbial.
Other possibilities include an infinitive clause with or without ‘to’
(‘the man put the dog to work’; ‘the man heard the dog bark’) and
a participial clause (‘the man heard the dog barking’).
4. Copular verbs
A small group which includes ‘be’, ‘seem’ ‘resemble’ and
‘become’ cannot take an object and require a subject comple-
ment. The sentence pattern is SVC (‘the dog was an Alsatian’;
‘the dog seemed happy’; ‘the dog lay in the yard’). The distinction
between a direct object and a subject complement is that whereas
the transitive verb affects or has an effect on the following object,
the subject complement refers back to and provides additional
information about the subject, which is necessary to complete the
sense. As the above examples show, complements may be a NP,
an adjective/AdjP or an adverbial – usually a place adverbial.
Note 1: In all sentence patterns, optional adverbials (adverbs and adverb

phrases) can be added (Crystal 1996: 61; Greenbaum and Quirk 1990: 204;
Huddleston and Pullum 2005: 78–80; Carter and McCarthy (2006: 504–529).
Note 2: Many verbs can be used both transitively and intransitively,
for example ‘approach’. One can say ‘the dog (S) approached (V)’ and
‘the dog (S) approached (V) the man (O)’ (see Crystal 1996: 51; Carter
and McCarthy 2006: 505–506 for relevant verb lists). For some verbs,
such as ‘eat’, ‘drink’ and ‘smoke’, but also ‘ring’, the difference between
transitive (SVO) and intransitive (SV) usage is that in the SV sentence
pattern the object is not explicitly stated but is taken for granted (under-
stood), as in, ‘You rang (the bell), madam.’ Sometimes, however, the
intransitive use of a verb is similar in meaning to the passive of the
164 Grammar
transitive use, for example, ‘the bell was rung’/‘the bell rang’; ‘the door
opened’/‘the door was opened’; ‘dinner was cooking’/‘dinner was being
cooked’. See also Carter and McCarthy (2006: 506–507).
Note 3: The copular verbs ‘look’ and ‘appear’ are roughly synon-
ymous as copulas (‘the dog looked/appeared tired’). However, ‘appear’
can also be used intransitively in the sense of ‘become visible’ as in ‘the
stars appeared’, and ‘look’ can be used transitively with a prepositional
object, as in ‘she looked at me’ or as a complex transitive verb in ‘she (S)
looked (V) me (O) in the eye (C)’.
Note 4: Some transitive verbs can be followed by a finite clause, others
by an infinitive clause and others by a gerundial clause functioning as
direct object of the verb. Sometimes a single verb can take more than
one option, as in, ‘She promised (that) she would visit him’ and ‘She
promised to visit him’, but not *‘She promised visiting him’. It is difficult
for learners to get these choices right and they tend to overextend use of
the infinitive option, particularly if their own language does not have
a gerund, producing, for example, *‘She considered to visit him’ instead
of ‘She considered visiting him’.
For more on verb transitivity see Crystal (1996: 48–59), Greenbaum
and Quirk (1990: 204–206), Huddleston and Pullum (2005: 78), Carter
and McCarthy (2006: 508–509).
Negative, Interrogative and Imperative Sentences

All the sentences so far mentioned have been positive declarative sen-
tences. The way a simple declarative sentence is transformed into its
negative counterpart depends on whether an auxiliary verb is present or
not. If an auxiliary verb is already part of the tense, then the word ‘not’
is added immediately after it, as in ‘The game is not being played.’ If
there is no auxiliary in the declarative sentence (simple present and
simple past tenses), then the ‘dummy’ auxiliary ‘do’/‘does’ plus ‘not’
(present tense) and ‘did’ plus ‘not’ (past tense) is added and placed
between the subject and the base form of the lexical verb, as in ‘she
does not play’, ‘she did not play’. For more on negation see Greenbaum
and Quirk (1990: 223–230), Huddleston and Pullum (2005: 149–157) and
Carter and McCarthy (2006:729–731).
To make declarative sentences interrogative, if an auxiliary is already
present, then this is brought to initial position in the sentence (‘fronted’), as
in ‘Is she playing?’ or ‘Can she play?’ Another way of expressing this is to
say that the subject and the auxiliary verb are ‘inverted’. In the case of the
simple present and simple past tenses, in the absence of an auxiliary, the
‘dummy’ auxiliary ‘do’/‘does’ (present tense) or ‘did’ (past tense) is added
and placed in sentence-initial position, as in ‘Does she play?’, ‘Did she play?’
‘Do’ used as an auxiliary to form questions and negatives is called
a ‘dummy’ auxiliary because it is a place-filler when an auxiliary verb is not
Grammar 165
available. ‘Do’ and ‘did’ cannot co-occur with an auxiliary in questions
(*Does she is playing? or *Does she can play?). Nor can the third person
singular ‘-s’ be marked on the lexical verb because it is already marked on
the auxiliary. Such ‘double marking’ is not permitted by the grammar
(*‘Does she plays?’). However, in the early stages of mastering questions
learners tend to do both these things, especially in writing.
The answer to interrogative sentences (‘yes/no’ questions) may be ellip-
tical, ending after the first auxiliary or a subsequent auxiliary (so-called short
answers). However, it is not acceptable to break off after ‘being’. Thus, as an
answer to the question ‘Is he being good?’ *‘Yes he is being’ or *‘No, he isn’t
being’ is not possible. As an answer to ‘May he have been being black-
mailed?’ ‘Yes, he may’, Yes he may have’, ‘Yes, he may have been’ are all
possible, but not *‘Yes, he may have been being’. Negative interrogative
sentences can be arrived at by two routes: either by making the interrogative
sentence negative or by making the negative sentence interrogative to
produce, for example, ‘Is she not playing?’ ‘Does she not play?’ ‘Did she
not play?’ If contracted forms are used, the fronted auxiliary attracts the
negative particle in all persons and for all auxiliaries, as in, ‘Isn’t the game
being played?’ ‘Doesn’t she play?’ and ‘Didn’t she play?’
The negative interrogative is not a neutral question but a loaded
question, implying expectation of the positive answer. ‘Isn’t New York
the biggest American city?’ is tantamount to saying, ‘Surely New York is
the biggest American city’ and expects the answer, ‘Yes, it is.’ The same
implication is expressed by the ‘negative tag question’ following
a positive statement, as in ‘New York is the biggest American city, isn’t
it?’ This also expects the positive answer, ‘Yes, it is.’ By the same token,
a ‘positive tag question’ following a negative statement implies a negative
answer. ‘San Francisco isn’t the biggest American city, is it?’ expects the
answer, ‘No, it isn’t.’ For more on interrogative sentences see Green-
baum and Quirk (1990: 231–237), Huddleston and Pullum (2005:
161–168) and Carter and McCarthy (2006: 534–537).
In English there are first person, second person and third person impera-
tives. In second person imperative sentences the verb form used is identical
with the base form of the verb as in, ‘Listen.’ The subject ‘you’ is ‘under-
stood’ or taken for granted and can be added for emphasis, insistence or
belligerence, as in, ‘You listen to me, mate.’ Imperatives have a wide range
of functional uses, including commands (‘Quick march’), instructions
(‘Light the blue touch paper …’) and warnings (‘… and stand well clear’),
but also exhortations (‘Walk on with hope in your hearts’). In
conversation, second person imperatives can sometimes sound brusque or
imperious (compare ‘Would you like to sit down’ with ‘Sit down’). There is
an emphatic form of the imperative with ‘do’, which can be inviting (‘Do
come in’; ‘do have another biscuit’) or insistent (‘Do be quiet’). The second
person imperative is negated with ‘do not/don’t’, as in, ‘Don’t go.’
166 Grammar
The first person imperative is formed with ‘Let me’ (singular)/‘Let us/let’s’
(plural) followed by the base form of the lexical verb. In the singular it can
express deliberation as in, ‘Let me see.’ However, it is also used in appeals as
in, ‘Let me take you by the hand’. In the plural it can also function as an
exhortation, as in ‘Let us pray’, ‘Let us go then you and I’ (T. S. Eliot), or
a suggestion as in ‘Let’s have a party’, ‘Let’s dance’, ‘Let’s spend the night
together’ (The Rolling Stones). Third person imperatives are also formed
with ‘Let’, as in ‘Let them eat cake’, but are rare and elevated in tone
(Greenbaum and Quirk 1990: 243). The first person negative imperative can
be formed either by negating the verb as in, ‘Let me/us not be too hasty’ or
by using the dummy auxiliary ‘do’ and negating that, as in, ‘Don’t let me/us
be too hasty.’ Negative third person imperatives are rare. See Crystal (1996:
44–45), Greenbaum and Quirk (1990: 241–243), Huddleston and Pullum
(2005: 170–172) and Carter and McCarthy (2006: 541–543).
Simple, Compound and Complex Sentences

Simple sentences consist of just one main clause, as in, ‘The dog slept.’
Sentences may, however, be simple, compound or complex:
Compound sentences consist of at least two clauses joined by one of

the three co-ordinating conjunctions ‘and’, ‘or’ and ‘but’ as in, ‘The
dog slept but the man worked.’ If the subject remains the same from
one clause to the next, it may be omitted after the first mention
(ellipsis) as in, ‘The dog slept and snored.’
-Complex sentences consist of at least one main clause and at least one
subordinate clause. A wide range of subordinating conjunctions such
as ‘because’, ‘when’, ‘where’, ‘while’, ‘since’, ‘as soon as’, ‘although’
and ‘despite the fact that’ may introduce subordinate clauses.
Since a clause always contains one, and only one, finite verb, the number
of clauses in a sentence is identical with the number of finite verbs.
Compound sentences in which main clauses are chained together by
‘and’ are not necessarily much more difficult to understand than simple
sentences. Complex sentences are more difficult, however, and the more
subordinate clauses a sentence has, the harder it becomes to understand.
Particularly difficult are sentences with ‘embedded’ subordinate clauses.
This occurs when a subordinate clause is inserted within another clause,
rather than following or preceding it. Extremely difficult texts often have
multiple embedding, in which a subordinate clause is inserted into another
subordinate clause, which is already embedded in the main clause.
See Crystal (1996: 192–205), Greenbaum and Quirk (1990: 262–291)
Grammar 167
Minor Sentences
Particularly in speech, informal writing and notices and messages, commu-
nication sometimes occurs in a sort of syntactic shorthand, for example:
‘No milk today’, ‘Gone to lunch’, ‘Maybe’, ‘Difficult to say.’ These syntac-
tically incomplete or ‘minor’ sentences depend on the listener or reader
being able to recover the missing elements from context. Thus, the milkman
knows that ‘No milk today’ means, ‘Please do not leave any milk today’, or
words to that effect. For more on minor sentences see Crystal (1996: 34–35).
Approaches to Teaching Grammar
Pros and Cons of Explicit Grammar Teaching

There is a controversy as to whether grammar should be explicitly taught
at all (see Thornbury 1999: 14–28; Ur 2011, 2012: 78–79; Mystkowska-
Wiertelak and Pawlak 2012: 29–59; Keck and Kim 2014: 146–149). Argu-
ments against the explicit teaching of grammar include the following:
It is impossible to teach explicitly the grammar of English (or any

other language) in its entirety because the grammar has not been fully
described.
Even native speakers are usually quite unable to explain explicitly the
grammar which they command implicitly.
Grammatical rules formulated by pedagogic grammarians are often
necessarily inadequate, incomplete and sometimes misleading (see
Harmer 2007: 225–227).
The conscious learning of oversimplified grammar rules may actually
blind the learner to noticing the more subtle grammatical regularities
of language in input and can lead to fossilisation (see ‘Language-
Learner Language’ in Chapter 1).
Classroom time is thus better spent on enhancing the quantity and
quality of exposure to English and getting learners to interact in the
classroom so as to implicitly acquire grammar (see Krashen 1999).
Learners do not like grammar lessons.
However, the advocates of some measure of explicit (formal) grammar

teaching make the following points:
The massive exposure to input and opportunities for interaction

which the child acquiring its native language experiences cannot be
even remotely approached in the classroom, which is an acquisition-
poor environment relative to that for native-language acquisition (see
‘Classroom Settings’ in Chapter 1 and further mention at the start of
Chapter 4).
168 Grammar
Since naturalistic grammar acquisition is age sensitive, adult naturalis-
tic acquirers tend not to acquire the grammar completely (see ‘Bilin-
gual Settings’ in Chapter 1). Even children in bilingual education
programmes without explicit grammar instruction may retain slight
grammatical deficits at the end of their schooling (Thornbury 1999: 50;
Baker and Wright 2017: 260). Thus, some measure of explicit gram-
mar teaching may be beneficial, even necessary, to counteract these
internal (biological) and external (environmental) disadvantages.
If learners are left to themselves, their learner grammar may fossilise
at a level at which they can understand and communicate more or
less effectively for their current needs (Swain and Lapkin 1995:
371–372; Thornbury 2005: 31–32).
Learners do not like bad grammar lessons but they do like good
grammar lessons.
For these reasons, with the exception of very young learners, explicit
grammar teaching of some sort can be beneficial. The older the learners
are and the more academic or intellectual they are, the more they will
benefit from grammar teaching of some sort. The teacher’s role is to
support learners in their discovery of the grammar, to give learners a leg-
up on the grammar ladder in accordance with Vygotskyan ideas of
scaffolding discussed in Chapter 1, and thus enhance the grammar-
acquisition process. However, neither we as teachers nor our learners
should have too high expectations of explicit grammar teaching. One
cannot expect newly encountered items of grammar to be immediately
incorporated into the learner’s internalised system and be available for
use. It may require a long and gradual process of experimentation before
the item passes from declarative knowledge into procedural knowledge
and finally becomes automatised because processes of restructuring will
be involved (see Chapter 1).
Acquiring Grammar in the Classroom

Learners need to be provided with more than just focussed grammar
lessons, on the one hand, and more than just communicative tasks which
ignore necessary grammatical structures, on the other. Keck and Kim (2014:
171–195) provide suggestions for setting up communicative tasks to practise
grammar, and suggest that such tasks can be designed along a scale accord-
ing to the degree of explicit focus on grammatical form and teacher feed-
back on error. Scrivener (2011: 157–158) suggests learners need:
wide exposure to English through extensive reading and listening

activities
practice in spotting examples of a particular grammatical feature in
specially written texts packed with many examples of that feature
Grammar 169
focus not only on grammatical form, but also on meanings and
functions (uses)
focussed practice of the item with the availability of teacher feedback
and correction
to use grammatical items in natural communicative contexts
to reflect on grammar items, to record them in written form and to
return to them repeatedly over time
But what is the best way to teach grammar? The following approaches are
worth considering:
the deductive approach

the inductive approach
the genre approach
the lexical approach
the cyclical approach
The Deductive Approach

The deductive approach, which Harmer (2015: 231–235) calls ‘Explain
and Practise’ and Scrivener (2011: 159) calls ‘Present–Practise’, is the
traditional one. It was used in the Grammar-Translation Method, dis-
cussed in Chapter 2, and today is still probably the most widely used
approach and is closely associated with ‘The three P’s’ lesson plan:
Presentation, Practice, Production (see Harmer 2015: 65–68). The new
item of grammar is first introduced and explained with examples (the
presentation phase). Next, the learners are given exercises in which the
new structure is practised (the practice phase). This phase also serves to
provide the teacher with feedback about how well the learners have
understood the explanation and whether they are ready to proceed to
the final phase or whether further explanation is necessary (repetition of
the presentation phase). The final phase is then a production phase, with
the emphasis on free production in contrast to the focussed practice of
the second phase. In the production phase, learners might, for example,
be asked to perform an oral communicative task, perhaps as role-play, or
a free communicative writing task, so that their attention is focussed on
communication rather than solely on the grammatical feature in
question.
In the Grammar-Translation Method the language of instruction was
the learners’ native language so that grammar explanations were formu-
lated in the native language. Some writers, for example Ur (2012: 80),
advise giving grammar explanations in the mother tongue for students of
lower proficiency levels. However, this only works in classes where the
learners all have the same mother tongue and the teacher speaks it too.
In any case, the real problem is that explanations will often tend to be
170 Grammar
abstract and difficult to understand anyway. Whether you tell learners
that the third conditional form is ‘an unreal condition in the past’ in
English or in any other language, they will still be baffled. It seems
therefore to be advisable to introduce some limited grammatical meta-
language in English early because otherwise it will be difficult to suddenly
switch to explaining grammar in English as learners become more
advanced.
In fact, many modern lessons for introducing new grammar adopt
a modified deductive approach in that they maintain the PPP structure
but avoid long explanations and rely on learners’ being able to infer the
patterns from carefully chosen examples through which they are guided
by the teacher in the presentation phase. A good example is Harmer’s
(2015: 239–240) lesson to introduce the present simple progressive, third
person singular. In such an approach the explanation, as formulated in
the course book, can then be presented at the end of the presentation
phase, or even at the end of the whole session.
For more on deductive approaches see Scrivener (2010) and Thorn-
bury (1999: 33–47).
The Inductive Approach

The alternative approach to deductive grammar teaching is the inductive
approach. This is sometimes called a discovery approach. Harmer (2015:
235–237) calls it ‘Discover and Practise’. Scrivener (2011: 166–169)
distinguishes between teacher-directed ‘guided discovery’ and ‘self-
directed discovery’. Rather than being first presented with explanations
illustrated by examples, learners are just shown example sentences and
invited to induce the grammatical rules which underlie them. To this
end, the examples must be carefully chosen so as to highlight gramma-
tical differences among otherwise similar sentences.
This approach is associated with a different three-phase lesson plan:
‘the three I’s lesson plan’ (Illustration, Identification, Inferencing)
(McCarthy and Carter 1995). In the first phase (Illustration), learners are
presented with example sentences of the structure to be introduced.
Often contrasting sentences will be shown which differ only in terms of
the new grammatical structure (for example, progressive versus simple
tense forms). Pictures may be used to indicate the meaning difference
between contrasting sentences. In the second phase (Identification),
examples of the new structure are identified in texts or with context
provided in other ways, for example by pictures, and their function
investigated, with support from the teacher. In this phase learners may
also do exercises which manipulate the sentences in various ways, for
example expanding, reducing or adapting sentences to exploit the gram-
matical options available. In the final phase (Inferencing), the class
reflects on what they have been doing and tries to identify grammatical
Grammar 171
patterns and principles illustrated by the sentences and finally to for-
mulate an explanation, if necessary with supportive scaffolding provided
by the teacher. If learners can work with a concordance, then it is
possible to combine an inductive approach with self-study and self-
directed learning (see Thornbury 1999: 54).
Total Physical Response (TPR) (see Chapter 2) is a long-established
technique which relies on inductive learning from the very start. Since
the imperatives used in TPR are paired with actions, learners are encour-
aged to induce the relationships. This can work for prepositions, for
example (Put your hands behind your head, in front of your face, on the
table, under the table etc.). After this has been done, learners can be
asked to recall the prepositions and their meaning. The same technique
can be used to teach positive and negative imperatives, and learners can
be asked to then take over the teacher’s role to see if they are then able
to produce the forms (see Thornbury 1999: 55–57).
Comparing Deductive and Inductive Approaches

Some aspects of grammar are better approached deductively and some
inductively. In this regard it is important to distinguish those where the
main problem is the mastery of form from those where the difficulty lies
in function rather than form. Often grammatical form is better taught
deductively and grammatical function inductively. For example, the form
of the conditional sentences lends itself to a deductive teaching
approach, since the system of tense shifting for the three conditional
sentence types is relatively easy to explain. However, the functional
differences among the three conditional forms are better taught by
following an inductive approach. An inductive approach is also well-
suited to exploring the functional differences among various ways of
expressing futurity in English (simple future tense, future continuous
tense, present continuous, ‘going to’ and simple present tense).
Advocates of the inductive approach stress that the approach is
learner-centred rather than teacher-centred and is process-oriented
rather than product-oriented. Learners are encouraged to self-direct
their learning and actively involve themselves. It is argued that learners
will retain rules that they themselves have formulated (or at least been
involved in formulating) better than rules which have been supplied to
them by the teacher or textbook. The opponents of inductive grammar
teaching complain that in practice the inductive approach is extremely
time-consuming and that learners are often unable to formulate the
explanation themselves and may even go off on a wrong track so that in
the end the teacher has to supply the explanation anyway. Additionally,
learners may become frustrated and demotivated by their own fruitless
search for the explanation. Rather than setting deductive approaches off
against inductive approaches, however, it is perhaps better to regard
172 Grammar
them as complementary weapons in the teacher’s armoury. They may be
used to vary teaching style and also to appeal to different learner styles.
For further sample lessons within the inductive approach, see Thorn-
bury (1999: 58–67) and Harmer (2015: 246–247).
The Genre Approach

The genre approach is concerned with teaching the grammatical features
which characterise specific text varieties or genres. It can be handled
more or less deductively or inductively. In the first phase an example
text of the specific genre in question is read, and learners are either asked
to identify idiosyncratic grammatical-stylistic features, or these are
pointed out to them, for example the agentless passive in scientific
texts. In the second phase learners may be given a second text in the
same genre with phrases or even whole sentences omitted and be asked
to complete the text appropriately. This can be done on a multiple-
choice basis, if desired. Alternatively, learners may be asked to supply
the missing introduction to the text, or to complete the text, or to match
text segments to their appropriate original texts (gap filling). In the final
phase learners may be supplied with a specific communicative situation
and asked to write an appropriate text in the same genre on their own
initiative.
The genre approach often tends to focus on syntax rather than
morphology and to be concerned with stylistic aspects of grammar
rather than with ungrammaticality versus grammaticality. Lexical and
pragmatic aspects may also be included if the features which characterise
the genre straddle the boundaries between grammar, pragmatics and
lexis. As well as being suitable for teaching ESP writing, it is also well-
suited to working with features of conversational grammar, especially
with advanced learners, using recordings of natural conversation or
transcriptions, if available. Features to focus on might include ‘left-
dislocated structures’ such as ‘That house on the corner, is that where
you live?’ and ‘right-dislocated structures’, such as ‘It’s very nice, that
road up through Skipton to the Dales’, as well as positive ‘tags’ with
positive sentences, such as, ‘She’s lovely, she is’ (examples from
McCarthy and Carter 1995: 211–213).
For a sample lesson based on the genre approach see Thornbury (1999:
85–89).
The Lexical Approach

The fourth approach to grammar teaching is the lexical approach, which
recognises that grammar teaching overlaps with vocabulary teaching. It is
a bottom-up approach based on ‘word grammar’. Word grammar oper-
ates at the interface between grammar and lexis and focuses on the
Grammar 173
grammatical information assumed to be attached to lexical entries in the
mental lexicon. This includes information on morphology, for example
noun plurals, adjective comparison and verb complementation. The
approach starts with the grammar of individual words and from this
basis sentences are built up. Words are grouped into word classes in
advance and introduced within specific word fields within certain the-
matic units as may be laid down by the syllabus.
To illustrate this with an example for noun plural morphology: for
young children in primary school at an early stage of learning there might
be a unit about a farm within which the semantic field of farm animals
might be introduced. Starting with the sentence frame, ‘Farmer Brown
has three cows,’ various regular plurals (‘cows’, ‘horses’, ‘goats’, ‘hens’)
as well as irregular plurals (‘sheep’ and ‘geese’) could be introduced as
learners adapt the sentence frame on a slot-and-filler basis to construct
their own sentences. In this way the morphology of noun pluralisation is
taught.
Another focus, suitable for older and more advanced learners, might
be verb complementation. This could be embedded in a thematic teach-
ing unit on giving a party. Verb complementation possibilities might
then be introduced in sentences such as ‘John suggested inviting Celia’,
‘Bill offered to buy the bread’ and ‘Anne suggested that Henry should
write the invitations.’ These sentences can be used as models or skeletal
structures to be filled appropriately with other lexical items. In this way
the grammar of gerund, infinitive or ‘that’-clause complementation is
being taught.
The Cyclical Approach

The cyclical approach is less an alternative to the other approaches
than a supplement to them. The learning of grammar is not a linear
but a cyclical process which continues throughout a learner’s career.
The cyclical approach goes beyond the idea of just practising forms
which have already been learned, as in classic grammar revision
sessions, and involves consolidation and extension of already known
grammar. As their proficiency develops, learners repeatedly return to
earlier, simplified rules and patterns they have learned so as to revise
and expand on them. This includes adding new meanings or functions
for grammatical forms already encountered. For example, the mor-
phology of the English modal verb system is relatively simple, but the
multiple meanings or functions of modal verbs are extremely complex
and cannot be mastered all at once. When modal verbs are first taught,
only a limited number of their meanings can be introduced. Within
a cyclical approach to language teaching, however, even advanced
learners will return again and again to modal verbs so as to refine
and extend their functional command.
174 Grammar
An example lesson would be a writing task for advanced learners in
which they are asked to write a guarded personal reference for a job
applicant about whose suitability for the position in question they have
reservations. This is a classic case of indirectness in language which is
expressed in part by verb modality. Students are given some model
references and work in groups (collaborative writing). Each group is
given a specific candidate profile with a particular weakness and asked
to write the reference. Afterwards, groups can exchange references so
that each group can take on the role of a prospective employer, read the
reference and see if they can successfully ‘read between the lines’.
Finally, the whole class can reflect on and discuss the uses of the modal
verb system for hedging and expressing reservations indirectly. Lessons
can also be designed for other areas of grammar, including, for instance,
non-conditional uses of ‘if’ (‘If you wouldn’t mind …’, ‘If I were
you …’), indirect speech acts involving interrogative and imperative
sentences and modal uses of ‘will’.
Awareness-Raising
In addition to focussed grammar lessons, attention should also be paid
to awareness-raising of grammar items encountered incidentally in the
classroom, perhaps during a reading session. Awareness-raising
involves directing the learners’ attention to a grammar item of interest
which might otherwise pass unnoticed. What is a grammar item of
interest? The answer is: items that are just slightly beyond the learners’
current proficiency level, or items which have recently been formally
introduced in a focussed grammar session, or items which are known
but which are in this case being used in a different sense or function or
in a variant form. Awareness-raising should not involve greatly dis-
rupting the lesson. The aim is to sensitise learners to the grammar
point in question so that they are more likely to notice similar
patterns for themselves in the future. It is not to be expected that
learners will necessarily master the item or pattern in question imme-
diately. (See Thornbury 1999: 24–25.)
For more on grammar teaching, see Thornbury (1999), Thornbury
(2005), Scrivener (2011: 156–184), Ur (2012: 76–87), Thaler (2012) and
Harmer (2015: 228–257).
Grammar Games
Grammar games provide additional practice and are helpful for proce-
duralisation and automatisation because in a game learners do not focus
attention on grammatical forms but rather on meaning. Within a PPP
lesson plan, the third phase, the production phase, can take the form of
a grammar game aimed specifically at the grammar point in question.
Grammar 175
A simple game is ‘Split-pairs sentences’. This can be used, for example, to
practise ‘unlikely conditions’ (type 2 conditionals). Each student receives
two slips of paper. On the first slip of paper they write the first half of an
improbable condition in the first person, which they are free to compose
themselves, for example, ‘If I won a million dollars’. On the second piece of
paper they complete the condition, for example, ‘I would take a trip around
the world.’ They fold each slip of paper in two so that the writing is not
visible. The first and second slips of paper are collected separately by two
students and each pile is put into a separate box or bag. Students then take it
in turns to come out and draw a paper from the first box, read it out and
then draw a paper from the second box and immediately read this out so as
to complete the sentence. The results may be humorous, silly, not mean-
ingful, or surprisingly witty or apt by chance. The papers are not returned
to the boxes so that each sentence produced is a new one. At the end of the
game the boxes are empty. The game can be made less challenging by giving
each learner the first half of a sentence on a slip and asking learners just to
write a completion on a second slip of paper. After the game has been
played once, subsequent rounds can include an assessment by the learners
at the end as to which was the best sentence that was produced by chance.
See Scrivener (2011: 177) for a variation on this game.
For further grammar games, see Harmer (2015: 253–256) and Scrivener
(2011: 176–179). For a board game which involves creating sentences by
combining lexical words and inflectional morphemes see Thornbury
(2005: 84–87). See Thaler (2012: 60–69) for a variety of games, including
grammar quizzes, panel games and ‘Memory’. The following game for
practising word order within the sentence is a slight adaptation of
a suggestion by Thaler (2012: 69).
Learners form groups of six. Each person first writes down
a grammatical subject (number and person must be specified in advance),
folds their paper over and passes it on to the next person who writes
a monotransitive verb (agreeing with the subject). The paper is passed on
to each member of the group with the following additions being made in
this order: object, adverbial of manner, adverbial of place, adverbial of
time. At the end the group decides which is their funniest sentence, and
these can be presented in plenum to the whole class and a vote cast on
which is the funniest sentence of all.
Songs for Grammar Practice

The thrust of this chapter has been that learners probably learn grammar
better from examples than from rules. This is exemplar-based learning (see
Ur 2011: 519). Songs or snatches of songs known by heart are well-suited
as exemplars. Some songs repeat, with lexical variation, over and over
again specific grammatical structures and are therefore particularly bene-
ficial. Songs are probably best used as a supplement to the grammar
176 Grammar
lesson, although in some cases they could be used to introduce a structure
and would lend themselves to a text-based inductive approach. Songs may
be sung or listened to by the class, as desired. They do not necessarily
have to be sung by the learners to be remembered. More important is that
they appeal to the learners. Songs include everything from nursery rhymes
through traditional ballads to pop music.
One of many children’s songs suitable for singing is ‘Ten green
bottles’. Anyone who can sing ‘There are ten green bottles standing on
the wall/And if one green bottle should accidentally fall/There’ll be nine
green bottles standing on the wall’ already knows a lot of English
grammar implicitly. They know how to use the existential construction
(‘there is/are’) and how to use a participial phrase consisting of the
present participle (‘standing’) plus a PP (‘on the wall’) to post-modify
a NP (‘ten green bottles’). They also know something about the order of
elements before nouns, namely that numbers precede colours. They
implicitly know how to form a ‘type 2’ conditional sentence using
‘should’, which is a variant form (‘should fall’ for ‘fell’) not often taught
explicitly in the classroom. This is presumably the way native-speaker
and bilingual children acquire and practise the grammar of English,
namely implicitly. Young learners will enjoy singing this song and are
unwittingly learning syntactic structures which will easily become estab-
lished in verbatim memory and serve as a template for other sentences
they may wish to construct on the same pattern. It is not necessary to
make the grammar explicit to learners, although this can be done.
For adolescent and adult learners, songs can be chosen to listen to
which hinge around a particular grammatical structure. Here are some
suggestions for focus on various tenses:
For type 2 conditional sentences: ‘If I Had a Hammer’; ‘If I Had

a Million Dollars’; ‘If I Loved You’
For the present progressive tense: ‘Oh What a Beautiful Morning’;
‘Sailing’
For the simple past tense: ‘Yesterday’; ‘The Banks of the Ohio’
For the past progressive tense: ‘The Tennessee Waltz’
For the present perfect tense and questions: ‘Where Have All the
Flowers Gone?’
For the future tense: ‘We Shall Overcome’
For the future tense and questions ‘What Shall We Do with the
Drunken Sailor?’
For the modal verbs and questions: ‘Blowing in the Wind’
Clearly, it is best to choose songs which appeal to the learner group, and
some of the above songs might be best suited to a class of senior citizens.
Learners themselves can be asked to find songs which best illustrate
a particular grammatical structure. This helps raise metalinguistic awareness.
Grammar 177
Food for Thought
Q. 1. (Group task) Look at a list of irregular verbs, for example the list in
Greenbaum and Quirk (1990: 29–34). You will notice that the list
includes many of the most commonly used verbs in the language,
but also a few rather rare verbs (‘wed’, ‘weep’, ‘withhold’ and
‘withstand’, for example). Why do you think some verbs are
regular and some are irregular, especially as all new verbs which
enter the language are initially regular?
Q. 2. Select what you consider the 20 most important irregular verbs for
learners and see how well they fit into the seven groups mentioned
in the chapter. How might you set about teaching these and how
many would you teach at a time to your elementary level learners?
Q, 3. (Group task) In the chapter, examples were given of various ways
of expressing future time and examples were also given in which
the ‘will’ future tense is used without reference to future time. Can
you think of example sentences in which present time and past time,
respectively, are expressed without using the corresponding tense
form? Conversely, can you think of examples in which a present
tense or past tense form is used without expressing the correspond-
ing time?
Q. 4. (Group task) Consider the ‘Leroy Brown’ song below. In this song
the irregular comparative and superlative forms of ‘bad’ are ‘reg-
ularised’ for facetious rhetorical effect:
And it’s bad, bad Leroy Brown

The baddest man in the whole damned town
Badder than old King Kong
And meaner than a junkyard dog …
(Jim Croce ‘Bad, Bad Leroy Brown’ from the album Life and
Times, 1973)
www.youtube.com/watch?v=QvwDohEEQ1E
Can you imagine using this song in a grammar lesson? If so, how
would you use it and for what purpose? Can you think of any other
examples of non-standard grammar which frequently occur in pop-
ular songs? Should authentic song texts with non-standard grammar
be excluded from the classroom or should they be deliberately
included, and if so, why, how and at what level?
Q. 5. Consider the grammar ‘of’ ‘much’, ‘many’ and ‘a lot of’ with
reference to the categories ‘number’ (singular and plural) and
sentence form (statement, question and negation). Can you work
out when you would be likely to use ‘much’ and ‘many’ and ‘a lot
of’? How would you introduce these forms to beginners? Have
a look at some course books, if you wish, and see what they do.
178 Grammar
Further Reading
Nassaji, H. and Fotos, S. (2011). Teaching Grammar in Second Language Classrooms:
Integrating Form-focussed Instruction in Communicative Context. New York:
Routledge.
Ringe, D. (2018). An Introduction to Grammar for Language Learners. Cambridge:
Rinvolucri, M. (1984). Grammar Games. Cambridge: Cambridge University Press.
References
Carter, R. and McCarthy, M. (2006). Cambridge Grammar of English:
A Comprehensive Guide to Spoken and Written English Usage. Cambridge: Cam-
Crystal, D. (1996). Rediscover Grammar with David Crystal. Revised edition.
Harlow: Longman.
Greenbaum, S. and Quirk, R. (1990). A Student’s Grammar of the English Language.
Harlow: Longman.
Harmer, J. (2007). The Practice of English Language Teaching. Fourth edition.
Harlow: Pearson Education.
Pearson Education.
Huddleston, R. and Pullum, G. K. (2002). The Cambridge Grammar of the English
Language. Cambridge: Cambridge University Press.
Huddleston, R. and Pullum, G. K. (2005). A Student’s Introduction to English
Grammar. Cambridge: Cambridge University Press.
Keck, C. M. and Kim, Y. (2014). Pedagogical Grammar. Amsterdam: John
Benjamins.
Krashen, S. D. (1999). Seeking a role for grammar: a review of some recent studies.
Foreign Language Annals 32/2: 245–257.
McCarthy, M. and Carter, R. (1995). Spoken grammar. What Is It and How Can We
Teach It? English Language Teaching Journal 49/3: 207–218.
Mystkowska-Wiertelak, A. and Pawlak, M. (2012). Production-Oriented and Com-
prehension-Based Grammar Teaching in the Foreign Language Classroom. Berlin:
Springer.
Quirk, R. and Wren, C. L. (1957). An Old English Grammar. Second edition.
London: Methuen and Co. Ltd.
Scrivener, J. (2010). Teaching English Grammar: What to Teach and How to Teach It.
Oxford: Macmillan.
Swain, M. and Lapkin, S. (1995). Problems in output and the cognitive processes
they generate: a step towards second language learning. Applied Linguistics 16/3:
371–391.
Thaler, E. (2012). Ten Modern Approaches to Teaching Grammar. Paderborn, FRG:
Schöningh.
Thornbury, S. (1999). How to Teach Grammar. Harlow: Pearson Education.
Grammar 179
Thornbury, S. (2005). Uncovering Grammar. Oxford: Macmillan. (First published
2001).
Ur, P. (2011). Grammar Teaching. In Hinkel, E. (ed.), Handbook of Research in
Second Language Teaching and Learning. Volume 2. New York: Routledge, pp.
507–522.
Yule, G. (2017). The Study of Language. Sixth edition. Cambridge: Cambridge
University Press.
7 Vocabulary
Word Formation
The minimal meaningful unit of language is the morpheme. Morphemes
are of two sorts: ‘free-standing’ morphemes and ‘bound’ morphemes.
A free-standing morpheme is a morpheme that can stand alone as
a word, for example ‘respect’ or ‘ox’. A bound morpheme, on the other
hand, cannot stand alone as a word, but is always part of a word. There
are two sorts of bound morphemes: ‘inflectional’ and ‘lexical’. Inflectional
morphemes were dealt with at the very beginning of Chapter 6. Examples
of inflectional morphemes would be, for example, the ‘-s’ in ‘respects’,
which marks the verb as third person singular in the present tense, or the
‘-en’ in ‘oxen’, which marks the noun as plural. You will recall from
Chapter 6 that inflectional morphemes are always suffixes (word endings)
in English, although sometimes inflection can be shown by vowel change
rather than adding a suffix, as in ‘give’ – ‘gave’, ‘foot’ – ‘feet’. In this
section the focus will be on lexical morphemes (sometimes also called
‘derivational’ morphemes).
Lexical morphemes carry lexical rather than grammatical meaning and
in English can be either prefixes or suffixes. The process of combining
prefixes and/or suffixes with free-standing morphemes such as ‘respect’ is
by far the most common means of word formation in English (Yule
2017: 62). For example, if the prefix ‘dis-’ is added to ‘respect’, we have
the new word disrespect. Adding the suffix ‘-able’ produces respectable’,
adding the suffix ‘-ful’ produces ‘respectful’. Adding both a prefix and
a suffix to ‘respect’ gives us ‘disrespectable’ or ‘disrespectful’. These two
words can be analysed into prefix plus ‘stem’ plus suffix.
The meaning change caused by adding lexical morphemes, whether
prefixes or suffixes, to a stem is generally systematic, consistent and
predictable. For example, the prefix ‘dis-’ carries the meaning ‘lack of’
in nouns such as ‘disrespect’, ‘disbelief’ and ‘disregard’. The prefix ‘un-’
is used to produce the antonym of an adjective, as in ‘unhappy’, and is
extremely productive. Just think of all the adjectives it can be combined
with! But of course it cannot be combined with all adjectives. For
Vocabulary 181
example, the opposite of ‘opportune’ is ‘inopportune’. But there are
usually patterns to be discerned. Consider ‘unwanted’, ‘unloved’,
‘unheeded’ versus ‘misguided’, ‘misplaced’, ‘misdirected’, where the
meaning of the verb seems important for the prefix which goes with it.
Some prefixes are available in antonym pairs: ‘super-’ and ‘sub-’, ‘intra-’
and ‘extra-’, ‘pre-’ and ‘post-’. There are also some treacherous cases,
however, where a prefix does not mean what one might expect: for
example, ‘flammable’ and ‘inflammable’ turn out to be synonyms rather
than antonyms: ‘If only my English teacher had told me that,’ as the oil-
rig worker said to his surgeon.
Prefixes usually change the meaning of the original stem word without
changing the word class, for example ‘terminate’ (verb) – ‘exterminate’
(verb). There are a few exceptions, however, for example, ‘witch’
(noun) – ‘bewitch’ (verb) (see Todd 1987: 43–44 for more examples).
Lexical suffixes, however, differ from prefixes (and from inflectional
suffixes, which never change the word class of the stem) in that they
always change the word class of the original word, as in ‘like’ – ‘likeable’,
‘eat’ – ‘eatable’, ‘measure’ – ‘measurable’. Here are some very produc-
tive, systematic, suffix patterns:
‘-ly’ forms adverbs from adjectives (‘possibly’, ‘sensibly’)

‘-ness’ forms abstract nouns from adjectives (‘happiness’, ‘sadness’)
‘-able’ forms adjectives from verbs (‘eatable’, ‘likeable’)
‘-ful’ forms adjectives from nouns (‘wonderful’, ‘hateful’)
‘-ise/-ize’ forms verbs from nouns (‘hospitalise’, ‘hospitalize’)
‘-er’ forms ‘agent nouns’ from verbs (‘painter’, ‘writer’)
‘-ish’ forms adjectives from nouns, sometimes with negative over-
tones (‘childish’, ‘oafish’)
‘-like’ forms adjectives from nouns, sometimes with positive over-
tones (‘childlike’, ‘godlike’)
See Crystal (1996: 109) for a list of suffixes that form abstract and
concrete nouns, respectively, from verbs, nouns and adjectives.
Although adding prefixes and suffixes to a stem is the most common
means of word formation in English, it is by no means the only one.
Another very common way is compounding, or putting two words
together to form a new word, as in ‘bookcase’, ‘fingerprint’ and ‘sun-
burn’. There are other, rather less common processes of word forma-
tion. These include ‘blending’, in which the beginning of one word is
joined onto the end of another as, in ‘brunch’ ‘(breakfast’ + ‘lunch’) or
‘smog’ (‘smoke’ + ‘fog’). Another means of word formation is ‘clipping’
or the reduction of a word of more than one syllable to a single syllable,
as in ‘flu’ and ‘porn’. Sometimes two words may be clipped and put
together to form a single word, as in ‘typo’ for ‘typographical error’.
Words may also be ‘borrowed’ from another language and for this
182 Vocabulary
reason may or may not be written in italics (zeitgeist, ‘blancmange’). For
other word formation possibilities see Yule (2017: 55–71). For an activity
for advanced learners on word formation see Ur (2012: 193–195). For
further discussion of lexical morphology see Carter (1998/2012: 24–26),
Todd (1987: 43–45), Yule (2017: 62, 73–88), Saxton (2017: 190–199) and
Huddleston and Pullum (2005: 264–265, 281–288).
Word Meaning
Lexical and Grammatical Words

The distinction between lexical and inflectional morphemes finds find its
parallel in the distinction between lexical and grammatical words. Lexical
words comprise nouns, verbs, adjectives and adverbs. Grammatical
words comprise all other word classes and include determiners (such as
the, a, each, my, some, this), pronouns (such as she, anyone, something),
conjunctions (such as but, and, because, since, although) and prepositions
(such as on, in, during). There is thus a distinction between lexical and
grammatical word meaning. The meaning of lexical words involves
reference to the outside world (extra-linguistic reference). The meaning
of grammatical words involves reference to other words (intra-linguistic
reference). Thus, a pronoun stands for a noun, a conjunction joins two
clauses and the possessive determiner ‘my’ establishes a relationship
between a following noun and the speaker or writer.
Grammatical words also differ from lexical words in that there is
a fixed, limited number of them in each class so that one can name all
the words in the class. Moreover, the stock of grammatical words
changes only very slowly over time, for example the slow demise of
‘thou’, ‘thy’, ‘thine’ in the nineteenth century. Lexical words, by con-
trast, are not only much more numerous, but new lexical words are
being coined all the time and others are falling into disuse. It is impos-
sible to list all the lexical words in a language not only because there are
so many, but because the stock is constantly changing. Grammatical
words are thus said to be ‘closed class’ words. Lexical words are ‘open
class’ words. Discussion in this chapter will focus on lexical words. The
processes of word formation discussed in the previous section, you will
have noticed, involved only lexical words. For more on lexical versus
grammatical words, see Fries (1952), Carter (1998/2012: 23–24) and
Aitchison (2012: 119–130).
Dictionary Meaning and Meaning in Context

There is a fundamental distinction to be drawn between a word’s ‘dic-
tionary meaning’ and its ‘meaning in context’. The meanings of a word as
cited in a dictionary represent its potential meaning range. Meaning in
Vocabulary 183
context refers to the meaning a word takes on as used on a specific
occasion in a specific context by a specific writer or speaker. Words such
as ‘young’ and ‘old’, for instance, can mean quite different things in
different contexts. Thus, when John F. Kennedy was elected President of
the USA at the age of 43, he was a very young president but he would
have been a very old footballer. ‘Old friends’ can be young people under
30 if they have known each other since their schooldays; an old girl-
friend or old boyfriend is an ex-boyfriend or ex-girlfriend, regardless of
age. ‘Old chap’ is just a term of endearment and has little or nothing to
do with age. Part of knowing the meaning of ‘young’ and ‘old’ is
knowing their possible meanings in a variety of contexts (see Aitchison
2012: 77–79).
Criterial Attribute Theory

One approach to dictionary word meaning is to define what distinctive
semantic features (criterial attributes) distinguish one word from another.
For instance, a mare is a female horse, a stallion a male horse and
a gelding is a castrated horse; a foal is a young horse, a colt is a young,
male horse and a filly is a young, female horse. There are other features
of horses which are not criterial for any of the above examples: colour,
for instance. But the criterial attribute is colour in the following subset
of horse words: ‘bay’, ‘chestnut’ and ‘grey’. In the case of ‘piebald’ it is
the colour patterning which is criterial. In English we do not have
a single word for a black horse or a white horse but some languages do.
For more on criterial attribute theory see Aitchison (2012: 53–60).
Prototype Theory
However, not all words are so easy to define in terms of criterial
attributes. To stay with the superordinate term ‘horse’, what exactly are
the criterial attributes that make a horse qualify as a ‘nag’? Must a nag be
an old horse, a slow horse, a worn-out horse or a bedraggled-looking
horse, or what exactly? It is difficult to say, but we probably have a mental
picture of a horse which would definitely fit the bill. In other words we
know what a prototype nag looks like. Prototype theory holds that we
have an ideal or central example of a word’s meaning in our heads.
Individual examples we encounter may be more or less closely related to
that central prototype. When an example falls between two of our stored
mental prototypes, we may be uncertain what to call it. This uncertainty is
often reflected in everyday conversation. People may say, ‘I like those
shoes, or boots or booties or whatever.’ In other words, sometimes
meaning may have to be negotiated anew for some unusual new object,
quality or activity. Each use of a word in a particular context is unique but
it must not stretch the bounds of that word’s potential meaning range
184 Vocabulary
unduly. If it does, then communication is likely to break down. For more
on prototype theory see Aitchison (2012: 66–79) and Taylor (2003).
Words with Fuzzy Edges

It is in fact the case that some words are far more referentially specific than
others and the boundaries between words are often not as clear-cut as we
might imagine. Linguists talk of words whose meaning has ‘fuzzy’ or ‘blurred’
edges. The colours of the spectrum are the classic example. Thus, the border-
line between what we call ‘green’ and ‘blue’ in English is an artificial one. For
a physicist, green and blue merge into each other. Moreover, the borderline
drawn in English does not necessarily correspond to the borderline in other
languages, which may have two or more different colour words for what we in
English would describe as different shades of blue or green or perhaps as
‘greeny blue’ or ‘bluey green’. But although the borderline between ‘blue’ and
‘green’ is drawn somewhat arbitrarily, the two words are used systematically
to mark that arbitrarily drawn border. For more on fuzziness of word
boundaries see Aitchison (2012: 55–65) and Ullmann (1962).
Polysemy
Most words in English are polysemous, that is, they can have more than one
meaning or sense. Often metaphor is the basis of the polysemy and one can
distinguish a primary, or ‘core’, literal meaning around which secondary
meanings based on metaphorical extension cluster (Verspoor and Tyler
2013: 163). The basis of metaphor is comparison, and the reader or listener
has to grasp the implicit comparison underlying the non-core meaning.
Metaphorical extension thus imparts structure to the polysemy of words,
with some meanings being more transparent than others because they are
closer to the core meaning. If we hear the word ‘face’ out of context, we
think of a human face. This is the core, literal meaning, but by metaphorical
extension we can talk of ‘the face of a clock’, ‘a rock face’ and even ‘the
unacceptable face of capitalism’. ‘Face’ also combines with the verbs ‘lose’
or ‘save’, as in ‘lose face’ and ‘save face’, to form idioms. When a need arises
for new words, as with the advent of computer technology, for example,
then existing words may be used in a new, metaphorical sense. This
extension of meaning by metaphor helps users to infer the new meaning
based on their knowledge of the word’s existing meanings and the new
context in which it is now used. A classic example is the newer sense of
‘virus’ in a computer context compared to its older sense from medicine
and immunology. Other examples of metaphorical extension in computer
technology are ‘gateway’, ‘surf’ and ‘patch’. For more on metaphor and
polysemy see Aitchison (2012: 172–190), Carter (1998/2012: 137–144,
159–163), Gibbs (1994), Glucksberg (2001), McCarthy (1990: 22–30) and
MacLennan (1994).
Vocabulary 185
Referential and Associative Meaning
For any given sense of a word there are two broad meaning components:
• referential meaning
• associative meaning
Referential meaning or ‘denotation’ is primary and associative meaning

or ‘connotation’ is secondary. Often, two or more words may have the
same referential meaning but differ in their associations. In this way
partial synonymy arises, as in ‘thin’, ‘skinny’, ‘emaciated’, ‘slim’, ‘slen
der’, ‘lissome’ and ‘svelte’ as used to describe people. In this set, the
neutral word is ‘thin’ and its partial synonyms have positive (‘slim’,
‘slender’, ‘lissome’, ‘svelte’) or negative (‘skinny’, ‘emaciated’) associa
tions. Differences in associative meaning may occur along a number of
dimensions in relation to the core or neutral word. These include, among
others:
• colloquial language (‘conk out’ for ‘collapse’)

• intimate language (‘mummy’ for ‘mother’)
• technical language (‘sternum’ for ‘breastbone’)
• archaic language (‘maiden’ for ‘girl’)
• polemical language (‘bourgeois’ for ‘middle class’)
• poetic language (‘leas’ for ‘ fields’)
• complimentary language (‘slim’ for ‘thin’)
• derogatory language (‘skinny’ for ‘thin’)
• euphemistic language (‘glow’ for ‘sweat’)
For more on partial synonymy, see Collinson (1939), Ullmann (1962)

and Lyons (1977)
Sense Relations
A word’s meaning in a specific context derives in part from its sense
relationships to other words which could occupy its slot in the sentence
or utterance in which it occurs. These are paradigmatic or vertical relations.
Sense relations, such as synonymy, antonymy and hyponymy, operate
along the paradigmatic or vertical axis of language, that is, they are words
of the same class which are interchangeable in a particular slot of a sentence
(‘she drives a fast car’ versus ‘she drives a slow car’). Part of a proficient
speaker’s knowledge of a word consists of knowing which words can
replace the given word in its slot in the sentence or utterance and how the
replacement will change the meaning (de Saussure 1916/1986). For more on
sense relations see Lyons (1977, 1981)), Leech (1974), Yule (2017: 129–134),
Aitchison (2012: 99–112) and McCarthy (1990: 15–21).
186 Vocabulary
Some important sense relations are:
synonymy (‘quick’ – ‘fast’ – ‘swift’)

Synonymy is the relationship of semantic co-identity. However,
although it may be helpful to explain a specific instantiation of
a word to a learner by giving a near synonym for it that means
more or less the same in that context, there are actually very few
exact synonyms. The criterion for whether words do mean exactly
the same is whether you can substitute one for the other in all
contexts without changing the meaning. This is actually hardly ever
the case. The more usual relation is that of partial synonymy, where
the meaning of words overlaps, as with ‘quick’, ‘fast’ and ‘swift’.
antonymy (‘quick’ – ‘slow’)
Antonymy is the relationship of opposition. There are dichotomous
antonyms such as ‘off’ – ‘on’ and ‘in – out’, where there is no
intermediate point, and there are scalar or gradable antonyms, such
as ‘boiling’ – ‘freezing’, which mark the end points of a scale with
identifiable intermediate points. Intermediate points on the scale may
sometimes themselves be classifiable into antonym pairs so that,
working inwards from the extremes, we have ‘hot – cold’, ‘warm –
cool’, and, in the middle of the scale, ‘lukewarm – tepid’, which we
do not perceive as antonyms at all but rather as near synonyms.
hyponymy (‘bird’ versus ‘eagle’, ‘sparrow’, ‘lark’, ‘thrush’)
Hyponymy involves an umbrella word (a superordinate) which spans
a number of hyponyms, which are all examples of the set denoted by
the superordinate term, as in the case of the superordinate term ‘bird’
and its hyponyms ‘eagle’, ‘sparrow’, ‘thrush’, ‘lark’, and many more.
Another example is ‘vehicle’, which is the superordinate term for ‘car’,
‘bus’, ‘taxi’, ‘van’, ‘lorry’, ‘trap’, ‘coach’, ‘rickshaw’ and ‘cart’, among
others. They are all hyponyms of ‘vehicle’. Superordinate – hyponym
patterns may nest one within the other. So ‘car’ is a hyponym of
‘vehicle’ but itself a superordinate of ‘sports-car’, ‘limousine’, ‘hatch-
back’, ‘convertible’ and ‘roadster’, among others.
entailment (‘parent’ – ‘child’)
Examples of logical entailment are ‘parent’ – ‘child’, ‘teacher’ – ‘pupil’
and ‘murderer’ – ‘victim’. The first term entails, implies or requires the
other. In other words, you cannot logically have the first term without
the second term.
incompatibility (‘aunt’ – ‘son’)
The relationship of incompatibility means that the two words are
mutually exclusive; one cannot be an aunt and a son, and vice versa.
complementarity (‘child’ – ‘adult’)
The relation of complementarity involves two terms which are
mutually exclusive and which together constitute the full range of
options available to complete the set.
Vocabulary 187
converseness (‘aunt’ – ‘niece’; ‘teacher’ – ‘pupil’)
Converseness involves a relationship of reciprocity: if Anne is Betty’s
aunt, then Betty is Anne’s niece; if Bill is Tom’s teacher, then Tom is
Bill’s pupil.
Collocational Relations
A word’s meaning in context is also partly defined by its relation to other
words with which it co-occurs (or collocates) in the sentence or utterance.
Collocational relationships operate on the syntagmatic or horizontal axis of
language. They refer to the co-occurrence probability of a lexical word (the
source word) with another lexical word (its ‘collocate’) in a stretch of
connected language or ‘syntagma’ (de Saussure 1916/1986: 171–180). The
two words do not necessarily have to be adjacent to be in collocation
although often they will be. Thus, ‘red’ and ‘hair’ are in collocation both in
the sentence ‘she has red hair’ and in the sentence ‘her hair is red’ although in
the second case the word ‘is’ intervenes between the two words in colloca-
tion. Common collocations involve noun plus verb, verb plus noun and
adjective plus noun. For more on collocation see Firth (1957a, 1957b),
Benson et al. (1986), Gough (1996), Aitchison (2012: 113–118), Carter (1998/
2012: 61–74, 78–80), McCarthy (1990: 12–15), Nation (2001: 317–343).
Collocation is not so much a matter of right or wrong, as acceptable or
unacceptable. There is a scale of collocational acceptability from unac-
ceptable collocations such as ‘beige hair’ to highly acceptable colloca-
tions such as ‘brown hair’. ‘Beige’, as any hairdresser will tell you, is just
not a hair word, whereas ‘brown’ frequently co-occurs with ‘hair’.
However, ‘brown’ may be said to collocate acceptably but weakly with
‘hair’ because it can collocate with many other nouns as well, in contrast
to words which strongly collocate with ‘hair’, such as ‘blonde’, ‘auburn’,
‘flaxen’ and ‘mousey’. These words strongly collocate with hair because
they do not collocate with many other nouns. There is a strong prob-
ability they will collocate with hair. There is thus a scale of collocational
probability as well as a scale of collocational acceptability.
Sometimes the word that the source word collocates with actually
determines its meaning. Thus, ‘red’ collocates acceptably with ‘hair’ and
with ‘car’ as in, ‘Bill has red hair and a red car.’ However, we think of
two different colours. In other words the meaning of ‘red’ is determined
by the words it collocates with (‘hair’ and ‘car’, respectively). In some
languages different words would be used. The French, for instance, say
les cheveux roux for (natural) red hair but une voiture rouge for a red car.
Many words which are normally regarded as synonyms, for example
‘fast’ and ‘quick’, usually turn out to be only partial synonyms in the
sense that they are not mutually substitutable in all contexts. Thus, we
say ‘a fast car’ and ‘fast food’, but not normally *‘a quick car’ and
*‘quick food’ though these expressions are not grammatically wrong.
188 Vocabulary
On the other hand, we say ‘to take a quick look at something’ rather
than *‘to take a fast look at something’.
Creative language often involves the deliberate flouting of collocational
expectations for rhetorical effect, especially in jokes, puns, witticisms,
irony, and poetic imagery. In this way ‘nonce collocations’ may be
produced. A nonce collocation is a ‘one-off’ collocation, a special collo-
cation for a particular special context, as in ‘the wrinkled sea’ in
Tennyson’s poem ‘The Eagle’, where the expression is used to describe
a ‘bird’s-eye’ perspective of the sea. Another nonce collocation is ‘clean
silence’ in the lines, ‘Clean silence drops at night when a little walk/
Divides the sleeping armies …’ in Stephen Spender’s poem ‘Two
Armies’ (c. 1939).
Some collocations become fixed and take on a special meaning which
is often not necessarily deducible form the meaning of their constituent
words. They therefore receive their own entry in dictionaries and pre-
sumably in the mental lexicon too (see below). Their meaning is often
metaphorical and idiomatic. Examples would be ‘red herring’, ‘blue
funk’, ‘gunboat diplomacy’ and ‘shotgun wedding’, all of which are
listed in the Concise Oxford Dictionary. These are ‘firm’, ‘frozen’ or
‘lexicalised’ collocations (MacLennan 1994; McCarthy 1990: 27–28).
Knowing a word therefore involves building up knowledge of the
network of sense relations a word enters into on the paradigmatic axis
of substitutability as well as its collocational possibilities and restrictions
on the syntagmatic axis of co-occurrence.
The Mental Lexicon
The Container Model of Lexical Storage

The term ‘mental lexicon’ is used to signify the store of ‘lexical items’ in
the mind (Aitchison 2012: 11). Most items stored will be words, but the
term ‘lexical items’ rather than ‘words’ allows for the possibility that some
prefixes and suffixes as well as some multi-word units are also stored in the
mental lexicon (Juffs 2013: 183–184). It is highly likely that some very
productive prefixes such as ‘un-’, ‘anti-’, and ‘neo-’, as well as highly
productive lexical suffixes such as ‘-able’ and ‘-ness’ will be listed in the
lexicon (or perhaps in a separate lexical-morpheme lexicon). The evidence
for this is that it is possible for listeners and readers to deconstruct the
meaning of words they may never have encountered before provided they
know the word stem, for example, ‘un-put-down-able’, ‘de-inflat(e)-able’. It
is by the same token also possible to produce nonce words, such as ‘anti-
banana-eating’ and be confident that one’s listeners will understand.
Some multi-word units with a fixed meaning will also have their own
separate lexical entry in the mental lexicon (Levelt 1989: 186–187;
Garman 1990:239–242). These may include the following:
Vocabulary 189
formulaic greetings such as ‘good morning’
opening gambits such as ‘Could I have a word?’, ‘To cut a long story
short …’, ‘Unaccustomed as I am to speaking in public …’, ‘Much as
I would like to help you …’
stock metaphors such as ‘a lion in combat’, ‘at the crack of dawn’
stock similes such as ‘as sober as a judge’, ‘as poor as a church
mouse’, ‘as dead as a dodo’
all-purpose stock expressions such as ‘at the end of the day’, ‘when
all is said and done’, ‘that is a recipe for disaster’, ‘at this moment in
time’
stock expressions tied to a particular situation or register such as ‘I
declare the meeting open/closed’, ‘Many happy returns’, ‘Further to
my letter of …’
proverbs such as ‘Too many cooks spoil the broth’
lexicalised collocations/idiomatic phrases such as ‘a red herring’, ‘a
blue funk’
sentence-length idioms such as ‘to buy a pig in a poke’, ‘to kick the
bucket’, ‘to flog a dead horse’
allusive lexicalised quotations such as ‘We are not amused’, ‘To be
or not to be’
fixed-order binomial or trinomial expressions such as ‘bread and
butter’, ‘tea and biscuits’, ‘spick and span’, ‘knife, fork and spoon’,
‘hook, line and sinker’
For further discussion see de Saussure (1916/1986: 172–173), Yorio

(1980), Alexander (1984), McCarthy (1990: 6–11), Nattinger and DeCar-
rico (1992), Carter (1998/2012), Aitchison (2012: 113–118).
Jackendoff (1995) suggested that mature native speakers may have at
their disposal as many as 80,000 ‘multi-word units’ if one also includes,
in addition to the categories listed above, compounds, names, literary
and other titles, quotations and foreign phrases. Some multi-word units
are limited to speech or to writing, some to specific situations. Some are
metaphorical, some literal. Some may function as discourse organisers
and have a functional rather than a referential meaning (Nattinger and
DeCarrico 1992; Nation 2001: 210–216). Some may permit no lexical or
grammatical change, whereas some may permit limited variation.
Direct access to ready-made language chunks improves fluency, but
there must be a limit to the extent to which single-word entries are
duplicated by multi-word entries in the lexicon, as this would at some
point make the system inefficient and overload memory capacity. The
mental lexicon is not a phrase book. However, some expressions cannot
be decomposed and therefore must be stored as a prefabricated unit.
A case in point would be ‘spick and span’: ‘spick’ and ‘span’ do not
mean anything by themselves and so cannot be listed separately in the
mental lexicon. Lexicalised foreign phrases such as pied à terre may also
190 Vocabulary
be understood as a single lexical unit by many language users without
their knowing the meaning of the individual words. But what about
common idioms such as ‘to kick the bucket’? Although the constitutive
words must be listed separately in the lexicon, it seems that the idiom
also has its own special lexical entry. The evidence comes from studies
which show that people actually understand very well-known idioms
such as ‘to kick the bucket’ more quickly than they understand the
literal meaning of the phrase (Ortony et al. 1978; Gibbs 1980; Glucks-
berg et al. 1982; Tabossi and Cacciari 1988). This suggests that people
have direct access to the whole idiom.
The Spreading Activation Model

The idea of a store of words in the mind is a very old one. Nevertheless,
the term mental lexicon is nothing more than a metaphor for lexical
storage and it may not be a very good one. It implies a container and
contents, and the image is static. The container metaphor is also mis-
leading in that it implies safety. In fact, if a foreign language is not being
constantly used, we soon find it difficult to retrieve words. Even in the
case of the first language acquired, forgetting can occur with disuse, for
example in the case of emigration, and for children this can take place
rapidly (see Higby and Obler 2015).
The idea of ‘spreading activation’ within a connectionist view of
language use and language acquisition was mentioned in Chapter 1.
Spreading activation models of lexical processing emphasise the
dynamic processes involved both in recognising words when listening
or reading and accessing (retrieving) words when speaking or writing
(Obler and Goral 2007; Jarema and Libben 2007a; Libben and Goral
2015). They are dynamic models based on what is known about
neuronal interaction in the brain. They view the items of the mental
lexicon, mainly words, as interconnected nodes which form vast net-
works of associations. These nodes correspond at the physiological
level to the synapses or junctions between nerve cells (neurons) where
information is believed to be stored. The cortex or outer surface of
the brain is composed of a thin layer of neurons arranged in strata.
There are ‘billions of neurons, and many more billions of synapses
connecting them’ (Aitchison 2012: 246). Neurons are composed of
a cell body and projecting fibres (dendrites, which conduct inwards,
and axons, which conduct outwards). Via its dendrites the cell body
receives information from other neurons in the form of an electrical
charge and it transmits information electrically via its axons to other
neurons. The biochemically bridged gap across which electrical activity
is transmitted between the axon of one cell and the dendrite of
another is the synapse (Garman 1990: 50–51).
Vocabulary 191
It is through language use (listening, reading, speaking and writing) that
words are acquired and maintained. Each time a word is used or
encountered in input, neural connections of greater or lesser strength
are assumed to be established by a wave of ‘spreading activation’ among
interconnected words (Aitchison 2012: 245–247; Randall 2007: 115). The
idea is that ‘a multiplicity of nodes are excited by the arousal of a node
to which they are connected’ (Singleton 1999: 125). The strength of
activation will depend on how directly words are associated in memory
to the ‘priming’ word. These links among words include sense relations
(Lyons 1977) and relations of lexical collocation (Firth 1957a, 1957b),
already discussed in this chapter. Evidence for the psychological reality
of such links is provided by word-association tests (Jenkins 1970; Aitch-
ison 2012: 101–103.). Say ‘bread’ to people and they are likely to say
‘butter’ (collocation), say ‘bird’ and they are likely to say maybe ‘robin’
or ‘sparrow’ or ‘eagle’ or the name of some other bird (hyponymy), say
‘hot’ and they are likely to say ‘cold’ (antonymy). It is in fact the links
a word has to other words which largely define its meaning (de Saussure
(1916/1986: 179–180).
Unlike a printed dictionary, the mental lexicon is thus clearly not
organised alphabetically but, rather, thematically (Aitchison 2012: 11–17,
99–144). All the words associated with a specific situation such as ‘ship-
wrecked’ or a topic such as ‘language teaching’ or a semantic concept
such as ‘time’ comprise what is called a ‘word field’ and are closely
linked to one another in the mind. The mental lexicon is thus more like
an electronic dictionary, with hyperlinks and cross references, than
a printed dictionary, but it is far, far better. There are criss-crossing
links among the individual word entries within each word field but also
links which go out from each word to words belonging to different word
fields with which the source word may be semantically related. Some
words will therefore belong to multiple word fields, and word fields
themselves may overlap or nest within one another. The semantic links
among words form a network like a spider’s web which is modified or
updated every time a word is encountered and retained in memory. In
other words, the links among words are established and modified by the
individual’s own developing experience of the language (Libben and
Goral 2015: 643). Since each individual’s linguistic experience is unique
and may change over time, a word’s meaning can vary somewhat from
one individual to another at any one time and also change over time for
any individual.
What Information Is Stored for Lexical Items?

Whether the ‘container model’ or the ‘spreading activation’ model is
adopted, the question still arises as to what information is available in the
mind for lexical items. Levelt (1989:182–183) specified four sorts of
192 Vocabulary
information: meaning, syntax, morphology and phonology, which he illu-
strated with the example of ‘eat’. ‘Meaning’ would correspond to the
definition of the word ‘eat’ in a dictionary. ‘Syntax’ would include informa-
tion on the word class of ‘eat’ (verb) and its transitivity status (SV/SVO).
‘Morphology’ would include the information that ‘eat’ is the base form of
the verb and is a free-standing morpheme as well as information on its
inflection (‘eats’, ‘ate’, ‘eaten’ – or would ‘ate’ have its own separate entry in
the lexicon?). ‘Phonology’ would comprise information on how to pro-
nounce the word. To these four categories, we might add, for literate
speakers of languages which have a written form, ‘graphology’ or how the
word is written. Much of this information corresponds to the information
in a physical dictionary. However, the mental lexicon will usually not
contain the etymological information held in larger dictionaries since this
is not part of normal speaker knowledge. On the other hand, the mental
lexicon presumably contains much more information about the stylistic and
register status of a word than print dictionaries provide (Levelt 1989: 183).
Collocational information, including collocation-exclusion possibilities, is
presumably also provided, as well as idiomatic uses of the word.
There is some controversy about whether separate lexica exist for the
native language and for each foreign language known or whether one large
lexicon should be envisaged with entries marked for the language to which
they belong (Libben and Goral 2015). Lamendella (1977) suggested that
beginning learners of a foreign language will tend not to link new words to
their referents in the outside world but rather to the translation equivalents
of the words in the native language (see also Juffs 2013: 187–192). Beginning
learners will thus store and access words in the foreign language through
a native language ‘translation buffer’. As learners become more advanced,
the links between the words within the foreign language will, if all goes well,
become stronger than links between the foreign-language words and the first
language words. This will enable learners to operate outside the translation
buffer, so that their fluency will improve and negative interference will be
reduced (see Chapter 1 on negative interference).
It is helpful to distinguish between storage and access when discussing
the mental lexicon. Concerning storage, for all language users there will be
many words that they do not know, and some information stored for
individual words will be incomplete or erroneous. The processes of flesh-
ing out skeletal entries for individual words and revising and refining
existing stored information are never completed, especially as word mean-
ing and usage change over the long term. Concerning access, even profi-
cient speakers sometimes experience the ‘tip of the tongue phenomenon’
(difficulty in accessing a word which they do have stored). This may result
in slow, hesitant and laborious search for le mot juste using controlled
processes. Sometimes a number of different words come to mind and
compete for selection and a wrong selection may be made – very often of
a similar-sounding lexical item to the one required. This is malapropism.
Vocabulary 193
Stored Lexicalised Sentence Stems
Lexical and grammatical knowledge are linked, and in first language acquisi-
tion there is a close link between grammatical and lexical development (Bates
and Goodman 1997; Ur 2019: 22). It seems that proficient language users have
access to various partly fixed, partly productive lexico-syntactic patterns or
skeletal structures which serve as building blocks for language production and
make both production and reception easier. These skeletal structures have an
underlying syntactic framework but, unlike fixed phrases or idioms, they are
more productive on the paradigmatic and syntagmatic axes. They straddle the
border between grammar and vocabulary and can be varied by expansion,
deletion or substitution of lexical elements as well as by grammatical
transformations.
Pawley and Syder (1983) first drew attention to these ‘lexicalised
sentence stems’, as they called them, and argued that proficient speakers
draw heavily on them in order to speak both idiomatically and fluently.
Lexicalised sentence stems are often invested with specific pragmatic
meaning. Thus, the lexicalised sentence stem:
That’ll be the last time I’ll (ever) ask + [NOUN PHRASE] + to +

[VERB] + [NOUN PHRASE] (optional) + for me!
underlies the sentence:
That’ll be the last time I’ll ever ask Harry to do a job for me!
(Pawley and Syder 1983: 212)
This utterance, and others built on its underlying stem but with different
lexical items, expresses the pragmatic meaning of dissatisfaction, regret
and annoyance and of having learned one’s lesson from bitter experience.
Also implied is the idea of misplaced trust. This pragmatic meaning is
still present in the following sentence, for example:
That’ll be the last time I’ll ever ask Bill to clean my car for me!
For more on the links between vocabulary and grammar, see Hunston
et al. (1997).
Teaching Vocabulary
Vocabulary Acquisition: an Incremental Process

Getting to know words or other lexical items has been compared to
getting to know people. First you may just recognise them as a face in
the crowd, later you can describe them. If you get to know them better,
194 Vocabulary
you may know something of how they behave, the company they keep
and you may know something of their history. You will of course know
many people, and many words, only very superficially but know others
quite well, and some very well. You may even be on intimate terms with
a few and feel you can rely on them. On the other hand, people and words
you know well may, on occasion, suddenly surprise you by their beha-
viour and even let you down. Indeed our knowledge of people and words
is constantly changing in micro-increments with every encounter we have
with them. Ultimately, it may be said you can never know any person, or
any word, completely. And people and word-meanings change over time.
Not every word or other lexical item a learner encounters in speech or
writing finds its way into the mental lexicon; many are quickly forgotten.
Such words never make it beyond short-term memory. Normally,
repeated encounters with a word are necessary to ensure retention (see
Zahar et al. 2001). Each time a word is encountered afresh, its memory
trace is strengthened in the lexicon. Gradually a learner may start to
build up an idea of the word’s meaning and then start to use the word
(Nation and Gu 2007: 85). The danger of forgetting is highest soon after
the initial encounter with a word so that frequent repeated encounters
shortly after the first encounter are important. As time goes on, less
frequent encounters will be sufficient to stop the word from being
forgotten (Nation 2001: 74–81).
Some words are easier to learn than others. Word difficulty is due to
both word-internal factors such as word length and semantic transparency
of the word (Laufer 1990), as well as learner-internal factors, especially
interest in the thematic area or topic involved. If learners are familiar with
and interested in the topic, the new vocabulary item can be linked into
a semantic network which is already established in the mind. Further,
according to Schmitt (2000: 121), ‘[T]he more one engages with a word
(deeper processing), the more likely the word will be remembered for later
use.’ Particularly intense encounters are those which engage the emotions
and relate to the individual’s personal interests and communicative needs
(Nation 2001: 63; Laufer and Hulstijn 2001; Laufer and Nation 2012: 168).
There is an ascending scale of involvement with a word as follows:
• noticing a word as a known word on hearing or reading it

• retrieval of the word’s meaning from memory on hearing or read-
ing it
• retrieval of the word from memory when wanting to express mean-
ing oneself
• recognising that a word is being used in a new sense on hearing or
reading it
• using the word in what for the learner is a new sense or context
(adapted from Nation 2001: 66–69)
Vocabulary 195
Enhancing Informal Acquisition of Vocabulary in the Classroom
Learners can be expected to acquire more vocabulary than they are
explicitly taught and it is neither possible nor desirable that all vocabulary
be explicitly taught in the classroom. Much vocabulary can be acquired or
‘picked up’ by conversational interaction, by reading and by exposure to
the media both inside and outside the classroom. Nation (2001: 134–143;
394–404) describes a number of classroom speaking activities which may
be beneficial for the incidental learning of vocabulary, including retelling
stories, role-play, interviewing and information-gap activities.
In sessions which are not primarily devoted to vocabulary learning,
teachers can take certain steps to make it more likely that important
vocabulary is retained. These include various forms of ‘input enhance-
ment’, such as pre-teaching selected vocabulary items, highlighting key
vocabulary items in texts, asking questions about a text which involve
searching for the vocabulary item and ensuring that an important voca-
bulary item is not just used once but is repeated several times in a lesson
(Nation 2001: 64–73). Key vocabulary items which crop up can be briefly
focussed on by lifting them out of context and checking on learners’
comprehension of the word not only in its present context but in other
possible contexts. If the word itself is not new to the class but is used in
a new sense, it is worth pointing this out explicitly.
Explicit Teaching of New Vocabulary

However, since the classroom is an acquisition-poor environment, class-
room learners also require explicit (formal) teaching of new vocabulary
(Verspoor and Tyler 2013: 169). Wherever possible, new words should
be taught in semantically related sets and be linked to other words which
learners already know from the same semantic field. This increases the
likelihood of long-term retention by tying the new words into the
relevant neural semantic networks of the mind. Most vocabulary teach-
ing will focus on words in context. Words are not stored in isolation and
should not be taught in isolation but rather in relation to a specific topic,
theme or situation. Learners need to master both the paradigmatic sense
relations and the syntagmatic co-occurrence relations of a word to be
able to use and understand it. The very worst thing we can do is to teach
learners vocabulary disembodied from thematic context because the
sense relations that a word enters into differ according to its context
(Aitchison 2012: 108).
There are various techniques available for introducing new vocabulary
and some words are better suited to some techniques than to others. It is
advisable to employ as wide a range of vocabulary-presentation methods
as possible. Often it is best to use various techniques for any one word
rather than just relying on a single technique. Here are some suggestions:
196 Vocabulary
‘swim’ could be introduced by mime
‘wink’ by demonstration
‘pliers’ by showing the object
‘give up the ghost’ by acting it out
‘diner’ by showing a picture
‘rifle’ by drawing a rifle on the board
‘paint the town red’ by telling a story
‘abolish’ by giving examples
‘lukewarm’ by drawing a scale from ‘boiling’ to ‘freezing’ on the
blackboard
‘you’re welcome’ by describing a situation
‘portly’ by using related known words (sense relations)
Other techniques include:
encouraging learners to infer the meaning of the word from context

eliciting an explanation to the class from a learner who does know
the word
translating the word
asking a learner to translate the word
The last possibility is best used after a difficult word has first been
introduced by another technique as it serves as a final check on whether
learners have really understood. See also Scrivener (2011: 189–190).
When introducing new vocabulary to beginners, the ‘spoken form
first’ principle should be maintained for as long as possible in the first
years of learning to avoid ‘spelling pronunciations’ of words such as
‘clothes’ or ‘gloves’. It is best for the teacher to first say the word, then
ask the learners to repeat it in chorus and then individually. Only when
the word can be pronounced acceptably should beginning learners then
see it in written form and finally write the word themselves. Practice
activities in which learners then use newly learned words are important.
These include rhymes and chants as well as role-play such as ‘going
shopping’ and ‘at the bank’. However, thanks to modern technology, it is
now possible even for primary-school learners to learn the spelling,
pronunciation and meaning of new vocabulary autonomously if they
have access to a suitable electronic (or online) dictionary with an audio
track to provide the pronunciation (Freudenau 2012, 2017).
Vocabulary Practice and Consolidation

Vocabulary learning involves not just adding new words to the mental
lexicon but also fleshing out skeletal entries for words already present
(Kersten 2010: 57). At any point in time there will be a distinction
between a learner’s production vocabulary and recognition vocabulary.
Vocabulary 197
Production vocabulary – those words which the learner can use in speech
or writing – is always a subset of recognition vocabulary – those words
the learner recognises but cannot yet use (see Laufer 1998; Laufer and
Paribakht 1998). However, the distinction between recognition and
production vocabulary is really scalar rather than dichotomous. Schmitt
(2000: 119) notes, ‘For example, it is easy to find students who can
produce a word orally without any problems but cannot read it recep-
tively.’ Knowing a word includes knowing its meaning or meanings,
pronunciation, spelling, morphology (inflections), its syntactic co-
occurrence restrictions and its lexical collocational restrictions, and this
knowledge cannot be accumulated at one fell swoop. Language teaching
at all levels should therefore involve both teaching new vocabulary in
structured fashion as well as consolidation work on words which are
already partly known. This constitutes a cyclical approach, whereby the
teacher and the learner keep returning to already encountered and partly
acquired words.
Practice is important for vocabulary retention (Ur 2019). First
attempts to use difficult words may be accompanied by much word
searching, hesitation or mispronunciation as well as inaccuracies of
various sorts, including the wrong collocational and syntactic environ-
ment and incorrect morphology. Learners need time and opportunities
to experiment creatively with new words. In acquiring their first lan-
guage, young children indulge in monologues and self-talk, often repeat-
ing words recently heard over and over again and making up rhymes and
repetitive sequences (Weir 1962; Pinter 2017: 27; Saville-Troike 1988).
Such self-talk, or private speech, is important for first language acquisi-
tion from a Vygotskyan perspective (see Saville-Troike and Barto 2017:
121). Pinter (2017: 27) suggests that ‘playing with forms, sounds, rhyme
and rhythm, and creating imaginary or nonsense words’ is valuable
practice for young foreign language learners, too. Since nonsense words
allude to real words, they help to establish these latter words in memory.
Practice should, however, be creative rather than just mechanical (Light-
bown and Spada 2013: 117). It is important that recently learned words
are used in combination with other, better-known words. Drama activ-
ities and simple poetry writing work well for both younger and older
learners. Ur (2012: 154) has an exercise which requires learners to
compose rhyming couplets. Songs, jingles, poems, rhymes, slogans,
limericks, proverbs and quotations offer practice in pronunciation and
intonation and help to fix new words in memory.
Visual means can also be used to help vocabulary retention. Words are
remembered more easily if they are linked to other words in the same
word field. Learners can be encouraged to create wall posters of various
sorts, for example ‘mind maps’ which they constantly update, adding new
words as they are encountered. In a mind-map, a source word, for
example, ‘airport’, is written in the centre of the page or poster and
198 Vocabulary
related words are added around it with links shown by lines to the source
word and to one another. Mind maps can be generated by the class calling
out words for the teacher to add to the poster, or one of the learners can
take over the teacher’s role and ask other class members to contribute
words. For more on mind maps see Scrivener (2011: 190, 201–203).
Another graphical means of representing relations between words is
the use of ‘word steps’. This involves drawing a set of steps to represent
a semantic scale, for example from ‘freezing’ to ‘boiling’, with each step
being labelled accordingly (Ur and Wright 1992: 86). Word grids are also
a helpful means of aiding vocabulary retention. In a word grid a set of
near synonyms is placed on the vertical axis and a set of criterial
attributes is placed on the horizontal axis. The relevant attributes are
then ticked off for each word on the grid (Scrivener 2011: 203–204).
Collocation grids, to be discussed shortly, are another useful tool. All
these graphical devices can be used for activities in class and then stored
by learners either in an electronic vocabulary notebook, on file cards or
in ring binders, all of which are better for recording vocabulary than
traditional paper notebooks because they can be organised thematically
and new material added as it is encountered.
Vocabulary exercises of various sorts also help aid retention of voca-
bulary and there are many published collections available. Among the
exercises Ur (2012) provides are several which exploit sense relations
along the paradigmatic axis of language, for example ordering gradable
antonyms along a continuum (Ur 2012: 129–131); exploring differences
among near synonyms (idem: 162–166); identifying the criterial attributes
of semantically related words, for example co-hyponyms, using compo-
nential analysis (idem: 167–169); and finding hyponyms of a given super-
ordinate using thesauri, either print or online (idem: 175–176). Another
exercise exploits the idea of polysemy based on metaphorical extension
of a core meaning (idem: 172–174) and activities are provided which
encourage learners to explore word associations (idem: 177–178) and to
reflect on the stylistic and register appropriateness of words in context
(idem: 179–181).
Teaching Affixes (Prefixes and Suffixes)

Nation (2001: 269–278) suggests learners’ awareness of morphological
structure can be developed by asking them to decompose words. If
a learner knows the meaning of a particular prefix or suffix, he or she
can often predict how it will change the meaning of the word it is
attached to. The meaning of a word such as ‘disrespectful’, which may
be new to learners, can therefore become clear by breaking it down into
its component morphemes. However, in order to use lexical morphemes
correctly in speech and writing the patterns of possible combinations and
also the constraints operating on morpheme combinations have to be
Vocabulary 199
acquired, for example ‘disrespectful’ but ‘unhelpful’ and ‘happiness’ but
‘bravery’. There are interesting patterns to be discovered, for example
the underlying phonological patterns governing when the suffix ‘-ity’
rather than ‘-ness’ is likely to be used to make abstract nouns from
adjectives, as in ‘stupid-ity’, ‘insipid-ity’ and ‘rapid-ity’.
Games can be devised on the lines of Scrabble whereby learners have to
combine morphemes to form acceptable words. This will help them
discover restrictions on combinations. Cloze exercises can also be con-
structed in which instead of whole words being omitted, word parts are
omitted. In this way learners have to decide between, for example,
‘impossible’ and *‘unpossible’. If enough suitable examples are provided,
learners should be able to infer the pattern for themselves. After they
have done the activity, they can be asked to formulate the rule. Accept-
ability-judgement exercises with words consisting of affix plus stem can
also be designed to help learners discover the underlying patterns
(Nation 2001: 278–281). Nation (idem: 263–281) provides further teaching
suggestions and also a useful list of frequently occurring prefixes and
suffixes in English, graded according to difficulty level (idem: 268).
Ur (2012: 81–91) also has a number of suggestions for activities with
affixes and provides her own useful list of suffixes which transform
words into another part of speech in a systematic way (idem: 83–84).
One of her exercises for advanced learners involves giving learners boxes
with a slot for ‘noun’, ‘verb’, ‘adjective’ and ‘adverb’ (idem: 90–91). One
slot is already filled in, for example ‘ignore’ under ‘verb’, and learners
have to complete the others. In this way learners not only become aware
of the patterns of regularity but also of the gaps in the system (where
there is no word to fill the slot). Such activities can be done with or
without dictionaries. Nation (2001: 282–296) discusses the role of dic-
tionaries in vocabulary learning more generally.
Some suffix patterns are very productive. One example is the ‘-er’ suffix to
form ‘agent nouns’ from verbs as in ‘walk’ – ‘walker’. A lesson for adults of
upper-intermediate level or above might start by asking learners to generate
as quickly as they can agent nouns ending in ‘er’. They might be expected to
produce ‘runner’, ‘swimmer’, ‘eater’, ‘drinker’, ‘player’, ‘surfer’ but might
also sometimes overextend the pattern and produce erroneous forms, for
example *‘cycler’ for ‘cyclist’. Often an apparent exception to a pattern is in
fact part of another intersecting pattern. Such intersecting patterns can then
be explored so as to raise learners’ awareness of the morphological system.
Concerning ‘cyclist’, for example, learners can be encouraged (if necessary
with teacher prompts) to find other nouns on this pattern. They will
discover that ‘cyclist’ belongs to another pattern in which the suffix ‘–ist’ is
added to nouns and adjectives (not verbs), as in ‘cyclist’, ‘motorist’, ‘nudist’,
purist, ‘guitarist’, ‘violinist’, ‘defeatist’. This pattern cuts across the ‘-er’
pattern and is a potential source of error for learners, who might produce
erroneous forms such as *‘cycler’ and *‘defeater’. Further reflection,
200 Vocabulary
exercises and dictionary work may help learners to discover that sometimes
within a particular word field a certain pattern is preferred. For example,
within the word field of musical instruments the ‘-ist’ pattern is generally
preferred, as in ‘guitarist’, ‘violinist’, ‘trombonist’, but there are also other
competing patterns, as in ‘drummer’ and ‘trumpeter’.
Teaching Lexical Collocation

Mastery of collocations is closely linked to the easing of language proces-
sing because words are ‘chunked’ into larger meaningful units (MacWhin-
ney 2012: 218–219). Learners need to build up collocational expectations
which match those of proficient speakers (see Nation 2001: 317–343).
There are various published collections of collocation exercises, for exam-
ple McCarthy and O’Dell (2005). Dictionaries list some strong and
frequent collocations, but learners may need guidance in how to under-
stand this information. Ur (2012: 60–63) has an exercise which trains
learners how to do this. She provides another exercise in which learners
access an online corpus to explore collocation. Two websites that provide
corpus-based examples of collocations are JustTheWord at www.just-the-
word.com and SkELL https://skell.sketchengine.co.uk/run.cgi/skell.
In classroom work on collocation the focus should be on identifying
strong collocations and collocation exclusions (words that a source word
cannot collocate with). Here is an exercise on adjective-plus-noun collo-
cations for advanced learners which could be done as a follow-up to
exploring agent nouns ending in ‘er’, as outlined above. This exercise is
best done in groups, using a dictionary, an online dictionary or
a concordance programme:
Exercise:
For each of the adjectives below, say which noun or nouns it best
collocates with, which other nouns it can collocate with and which nouns
it cannot normally collocate with:
Adjectives: attentive, beautiful, turgid, deep, efficient, clear, slick,
brutal, reckless, unsteady, stern, ardent, skilled, tough,
stylish, heavy, chain
Nouns: thinker, driver, talker, smoker, manager, supervisor(!),
teacher, fighter, waiter, murderer, speaker, cyclist, writer,
runner, joiner, singer, lover.
You might like to do this exercise and see how you get on. You might
also like to imagine you are a learner who is not sure of some of the
collocations. See if online resources can help you to solve the problems.
Collocation grids can also be used to explore and practise collocations.
In a collocation grid a set of source words is arranged on the vertical axis
Vocabulary 201
and a set of possible collocates is arranged on the horizontal axis of the
grid. Learners put crosses in the cells where collocation is possible. The
following exercise uses a simple collocation grid to focus on the colloca-
tion possibilities of the delexicalised verbs, ‘take’, ‘make’, ‘have’ and
‘do’. Delexicalised verbs are verbs which are used as place fillers so that
the verb itself is rather empty of meaning and the collocation possibili-
ties for following noun phrases are quite unpredictable:
Exercise:
Which, if any, of the items along the top of the grid does each of the verbs
collocate acceptably with? Put a cross to mark acceptable collocations. For
‘take’ this has already been done for you.
a holiday the shopping a chat a good time
take X
make
have
do
As the possibilities in the above case are ‘acceptable’ versus ‘unacceptable’,

a simple collocation grid is appropriate. However, in cases where there is
a scale of collocational acceptability a scalar collocation grid is more appro-
priate. A scalar collocation grid allows for degrees of collocational strength.
Instead of placing crosses in cells, learners place a value, say from 0 to 2, from
0 (does not collocate) to 1 (doubtful) to 2 (good) (see McCarthy 1990: 12).
Teaching Multi-Word Lexical Units

Multi-word lexical units of various sorts should be included in vocabu-
lary work (Ur 2019). Ur (2012: 64–80) provides exercises which involve
identifying lexical phrases and binomial expressions, as well as investigat-
ing the meaning of proverbs and idioms and exploring the underlying
meaning of allusive book, film and play titles. Green (2019) offers
detailed advice on how to use concordance programmes for teaching
multi-word units. Some fixed-order binomial and trinomial expressions
deserve special attention. For example, the sporting expressions ‘game,
set and match’ (tennis), ‘hook, line and sinker’ (angling) and ‘down and
out’ (boxing) are a rich source of metaphorical extension. Thus, ‘game set
and match’ can refer to any complete victory; ‘hook, line and sinker’ to
any complete deception, as in, ‘he swallowed the story, hook, line and
202 Vocabulary
sinker’; ‘down and out’ can be used alongside its literal sense from
boxing in its fixed metaphorical sense, as in the title of George Orwell’s
memoir Down and Out in Paris and London.
The meaning of multi-word units varies along a scale from transparent to
opaque. Relatively transparent set phrases can be understood from their
constituent words whereas relatively opaque ones cannot. These latter often
tend to be idiomatic expressions. Boers et al. (2004) suggest a teaching
approach to relatively transparent idiomatic expressions in which learners
are asked to deconstruct the underlying metaphor. In the case of opaque
expressions, sometimes there is a story behind them which dictionaries
explain. For example, ‘a red herring’ comes from the practice of hunt
disturbers using a raw herring to put the hounds off the scent of the fox
(Webster’s New Encyclopedic Dictionary 1995: 851). In such cases the teacher
can unravel the metaphor for learners. Learners will in this way acquire both
language and cultural knowledge, with possible positive motivational effects.
Teaching Lexicalised Sentence Stems

Work on lexicalised sentence stems should also be done, based wherever
possible on an example which occurs in the classroom, often in a reading
text. For advanced learners who are focussing on academic writing, such
a lexicalised sentence stem might be:
[PLACE ADVERBIAL] + alone (optional) + there are + [PAST

PARTICIPLE] + to be + [NP]
as in the sentence:
In England alone there are thought to be some two hundred different dialects.
Learners may understand this in context on the basis of understanding

the meaning of ‘England’, ‘thought’ and ‘two hundred different dialects’
but may not be able to use the sentence stem for themselves in other
contexts. This would be the justification for focussed work, as follows:
1. Learners work in groups, possibly using dictionaries or a concordance

programme, and find words which could be substituted for ‘thought’.
Learners might produce the following list: ‘guessed’, ‘calculated’,
‘estimated’, ‘supposed’, ‘known’, ‘imagined’, ‘believed’.
2. Learners then arrange these words along a scale of precision from
‘known’ to ‘supposed’.
3. An alternative construction is now introduced:
[PLACE ADVERBIAL] + alone (optional) + it is + [PAST PARTI-
CIPLE] + that (optional) + there are + [NP]
In England alone it is thought (that) there are some two hundred different
dialects.
Vocabulary 203
4. Learners now produce new sentences according to this pattern using
the above list of verbs and replacing ‘England’ and ‘two hundred
different dialects’ with different lexical content.
5. Learners are could then be given the following sentence stem:
[PLACE ADVERBIAL] + alone (optional) + there are +
[ADVERB] + [NP]
In England alone there are probably some two hundred different dialects.
They now consider which adverbs can be substituted for ‘probably’
and they order these adverbs on a scale of probability, perhaps as
follows: ‘certainly’, ‘undoubtedly’, ‘probably’, ‘possibly’, ‘supposedly’,
‘arguably’.
6. Learners create a new sentence stem using an adjective rather than an
adverb:
[PLACE ADVERBIAL] + alone (optional) + it is + [ADJ] +
that (optional) + there are + [NP]
In England alone it is probable that there are some two hundred different
dialects.
7. Learners generate sentences based on any of the sentence stems using
specific key words, for example ‘France – wines’, ‘Oxford – bicycles’.
A bit of research on the internet should provide appropriate examples
and numbers. Some classes may prefer to produce factually appro-
priate statements from their subject area. Others may wish to unearth
little-known statistics, so-called ‘nerd statistics’, and produce ‘true’
and ‘false’ statements. The class can then guess whether the statement
is true or false in terms of the probability index attached to it.
Food for Thought

Q. 1. (Group work) See if you can arrange some frequently occurring
prefixes into groups according to whether they are attachable to
nouns, adjectives or verbs and according to how they change the
meaning of the word. Is it possible to predict the combination
possibilities and combination restrictions operating on specific
prefixes?
Q. 2. (Group work) Based on some of the lexical suffixes mentioned in
the chapter, and any others you want to add, create an activity or
exercise to help learners explore word formation possibilities.
Q. 3. (Group work) Design an activity or exercise to help learners
understand and/or produce compound words such as ‘bookcase’.
Q. 4. (Group work) In groups, consider the meaning differences
between the following pairs of words:
river – stream; mountain – hill; shirt – blouse; pony – horse;
to question someone – to interrogate someone; bitter – sour;
to renovate – to redecorate; indignant – angry
204 Vocabulary
Each group member should now take one pair and in turn briefly
explain or demonstrate the meaning difference between the two words
to the rest of the group, who should assume the role of learners and
feel free to ask the ‘teacher’ if they do not understand. At the end of
the sessions, reflect as a group on what approaches were used and what
was effective and what was less effective. Consider whether some pairs
of words are better explained by examples (prototype theory), and
others better explained by giving defining characteristics (criterial
attributes). Finally, as a group, design a gap-filling exercise to check
whether learners have understood the explanations. Do the exercise as
if you were learners and consider whether you would get it right or not
on the basis of the explanations given.
Q. 5. Look at the following list of words and phrases and consider which
of the techniques listed below you would use to introduce them to
learners at an upper-intermediate level of proficiency.
Vocabulary: advertising campaign, catalogue, café, chase (verb),
exploitation, give up the ghost, hope (noun), hostage,
house, memory stick, often, paint the town red, put up
with, reduction, stapler, swimming, that’ll be the day,
vase, wellies, window sill, wink (verb)
Techniques: mime, facial expression, video, blackboard drawing,

graph or diagram, students act it out, tell a personal
anecdote, get students to infer meaning from a text,
point to the object, explain by examples, read out the
dictionary definition, translation, show a picture, bring it
into class to show to students, act out a conversation,
students who know explain to those who do not, stu-
dents look it up in their dictionary, create a story using
models or puppets, a guessing game (e.g. picture or
definition matching), tell a story using the item
Would you prefer any other method not listed for specific words?
Which methods listed would be particularly unsuitable for specific
words? Why is this?
Q. 6. The following adjectives are similar in meaning and the meaning
differences among them are best described by presenting them in
collocation with nouns with which they strongly collocate:
big, large, great, extensive, capacious, massive
Working individually, choose a noun which you think collocates
very strongly with each adjective. The best way to do this is to say the
adjective and then write down the noun that immediately comes to
your mind. Compare your choices with those of your colleagues and
see how similar or dissimilar they are. Create a sentence for each
adjective which would nicely illustrate the meaning of the collocation
Vocabulary 205
for learners. Finally, consider how you might teach these collocations
to advanced learners – or do you think it would be a mistake to teach
them explicitly at all?
Q. 7. As a group exercise, or alone, design a 4×4 scalar collocation grid to
help learners with using appropriate adjective and noun combina-
tions within the semantic field of English weather. Suggested adjec-
tives are ‘heavy’, ‘strong’, ‘bright’, ‘patchy’, ‘light’, ‘slight’. Suggested
nouns are ‘rain’, ‘showers’, ‘wind’, ‘sunshine’, ‘snow’, ‘breeze’,
‘sleet’, ‘hail’, ‘drizzle’. Choose four appropriate adjectives and four
nouns. Now complete the grid, as you think your learners might.
Note any problems that might arise and discuss these with your
colleagues. How effective or otherwise do you consider your grid?
Further Reading
Barcroft, J. (2015). Lexical Input Processing and Vocabulary Learning. Amsterdam:
John Benjamins.
McCarthy, M. and O’Dell, F. (2017). English Vocabulary in Use: Advanced. Third
Nation, I. S. P. (2008). Teaching Vocabulary: Strategies and Techniques. Boston, MA:
Heinle.
Redman, S. (2011). English Vocabulary in Use: Intermediate and Pre-Intermediate.
References
Aitchison, J. (2012). Words in the Mind: An Introduction to the Mental Lexicon.
Fourth edition. Chichester: Wiley-Blackwell.
Alexander, R. J. (1984). Fixed expressions in English: reference books and the
teacher. ELT Journal 38/2: 127–134.
Bates, E. and Goodman, J. C. (1997). On the inseparability of grammar and the
lexicon: evidence from acquisition, aphasia and real-time processing. Language
and Cognitive Processes 12/5–6: 507–584.
Benson, E., Benson, M. and Ilson, R. (1986). Lexicographic Description of English.
Amsterdam: John Benjamins.
Boers, F., Demecheleer, M. and Eyckmans, J. (2004). Etymological elaboration as
a strategy for learning idioms. In P. Bogaards & B. Laufer (eds.), Vocabulary in
a Second Language. Amsterdam: John Benjamins, pp. 53–78.
Carter, R. (1998/2012). Vocabulary: Applied Linguistic Perspectives. Second edition.
London and New York: Routledge. (Republished in Routledge Linguistics Classics
2012).
Collinson, W. E. (1939). Comparative synonymics: some principles and
illustrations. Transactions of the Philological Society 38/1: 54–77.
Crystal, D. (1996). Rediscover Grammar with David Crystal. Revised edition.
Harlow: Longman.
de Saussure, F. (1916/1986). Course in General Linguistics. La Salle, IL: Open Court
Publishing Company. English translation by R. Harris 1983 of 1972 edition of
206 Vocabulary
Cours de Linguistique Générale. Edited by Charles Bally and Albert Sechehaye.
Paris: Payot. (First published 1916).
Firth, J. R. (1957a). Modes of meaning. In Firth, J. R. (ed.), Papers in Linguistics
1934–1954. London: Oxford University Press, pp. 190–215.
Firth, J. R. (1957b). A synopsis of linguistic theory 1930–1955. In Firth, J. R. (ed.),
Studies in Linguistic Analysis. Oxford: Special volume of the Philological Society,
pp. 1–32.
Freudenau, T. (2012). Multimedia dictionaries in the primary school. In Lennon, P.
(ed.), Learner Autonomy in the English Classroom: Empirical Studies and Ideas for
Teachers. Frankfurt am Main: Peter Lang, pp. 95–122.
Freudenau, T. (2017). Wortschatzarbeit im Englischunterricht der Grundschule: Eine
Studie zum Autonomen Lernen mit Online-Wörterbüchern. (Dictionary Work in
English Teaching in the Primary School: a study of autonomous learning with
online dictionaries). Frankfurt am Main: Peter Lang.
Fries, C. C. (1952). The Structure of English: An Introduction to the Construction of
English Sentences. New York: Harcourt, Brace and Company Inc.
Garman, M. (1990). Psycholinguistics. Cambridge: Cambridge University Press.
Gibbs, R. W., Jr. (1980). Spilling the beans on understanding and memory for
idioms in context. Memory and Cognition 8: 149–156.
Gibbs, R. W., Jr. (1994). The Poetics of Mind: Figurative Thought, Language and
Understanding. Cambridge: Cambridge University Press.
Glucksberg, S. (2001). Understanding Figurative Language: From Metaphors to
Idioms. Oxford: Oxford University Press.
Glucksberg, S., Gildea, P. and Bookin, H. B. (1982). On understanding non-literal
speech: can people ignore metaphors? Journal of Verbal Learning and Verbal
Behaviour 21: 85–98.
Gough, C. (1996). Words and words: helping learners to help themselves with
collocations. Modern English Teacher 5: 32–36.
Green, B. A. (2019). Teaching and assessing multiword expressions using an open
educational resources academic English corpus. In Hinkel, E. (ed.), Teaching
Essential Units of Language: Beyond Single-word Vocabulary. London and
New York: Routledge, pp. 134–163.
Higby, E. and Obler, L. (2015). Losing an L1 to an L2. In Schwieter, J. W. (ed.), The
Cambridge Handbook of Bilingual Processing. Cambridge: Cambridge University
Press, pp. 645–664.
Huddleston, R. and Pullum, G. K. (2005). A Student’s Introduction to English
Grammar. Cambridge: Cambridge University Press.
Hunston, S., Francis, G. and Manning, E. (1997). Grammar and vocabulary:
showing the connections. English Language Teaching Journal 51/3: 208–216.
Jackendoff, R. (1995). The boundaries of the lexicon. In Everaert, M., van der Linden,
E.-J., Schenk, A. & Schreuder, R. (eds.), Idioms: Structural and Psychological Perspec-
tives. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 133–166.
Jarema, G. and Libben, G. (eds.) (2007a). The Mental Lexicon: Core Perspectives.
Amsterdam: Elsevier.
Jarema, G. and Libben, G. (2007b). Introduction: matters of definition and core
perspectives. In Jarema, G. & Libben, G. (eds.), The Mental Lexicon: Core
Perspectives. Amsterdam: Elsevier, pp. 1–6.
Jenkins, J. J. (1970). The 1952 Minnesota word association norms. In Postman, L. &
Keppell, G. (eds.), Norms of Word Association. New York: Academic Press,
pp. 1–38.
Vocabulary 207
Juffs, A. (2013). Second language acquisition of the lexicon. In Ritchie, W. C. &
Bhatia, T. K. (eds.), The New Handbook of Second Language Acquisition. Second
edition. Leiden: Brill, pp. 181–209.
Kersten, S. (2010). The Mental Lexicon and Vocabulary Learning: Implications for the
Foreign Language Classroom. Tübingen, FRG: Narr Francke Attempto.
Lamendella, J. (1977). General principles of neurofunctional organisation and their
manifestation in primary and non-primary language acquisition. Language Learn-
ing 27/1: 155–196.
Laufer, B. (1990). Why are some words more difficult than others? Some intralex-
ical factors that affect the learning of words. International Review of Applied
Linguistics 28/4: 293–307.
Laufer, B. (1998). The development of passive and active vocabulary in a second
language: same or different? Applied Linguistics 19/2: 255–271.
Laufer, B. and Hulstijn, J. (2001). Incidental vocabulary acquisition in a second
language: the construct of task-induced involvement. Applied Linguistics 22/1: 1–26.
Laufer, B. and Nation, I. S. P. (2012). Vocabulary. In Gass, S. M. & Mackey, A.
(eds.), The Routledge Handbook of Second Language Acquisition. London and
New York: Routledge pp. 163–176.
Laufer, B. and Paribakht, T. S. (1998). The relationship between passive and active
vocabularies: effects of language learning context. Language Learning 48/3: 365–391.
Leech, G. (1974). Semantics. Harmondsworth: Penguin.
Levelt, W. J. M. (1989). Speaking: From Intention to Articulation. Cambridge, MA:
MIT Press.
Libben, G. and Goral, M. (2015). How bilingualism shapes the mental lexicon. In
Schwieter, J. W. (ed.), The Cambridge Handbook of Bilingual Processing. Cam-
bridge: Cambridge University Press, pp. 631–644.
Lyons, J. (1977). Semantics. Two volumes. Cambridge: Cambridge University Press.
Lyons, J. (1981). Language, Meaning and Context. London: Fontana/Collins.
MacLennan, C. H. (1994). Metaphors and prototypes in the teaching and learning of
grammar and vocabulary. International Review of Applied Linguistics 32/2: 97–110.
MacWhinney, B. (2012). The logic of the Unified Model. In Gass, S. M. &
Mackey, A. (eds.), The Routledge Handbook of Second Language Acquisition.
McCarthy, M. (1990). Vocabulary. Oxford: Oxford University Press.
McCarthy, M. and O’Dell, F. (2005). English Collocations in Use. Cambridge:
Nation, I. S. P. (2001). Learning Vocabulary in Another Language. Cambridge:
Nation, I. S. P. and Gu, P. Y. (2007). Focus on Vocabulary. Sydney: NCELTR
Publishing.
Nattinger, J. and DeCarrico, J. (1992). Lexical Phrases and Language Teaching.
Oxford: Oxford University Press.
Obler, L. K. and Goral, M. (2007). Bilingual lexica. In Jarema, G. & Libben, G.
(eds.), The Mental Lexicon: Core Perspectives. Amsterdam: Elsevier, pp. 185–206.
Ortony, A., Schallert, D. L., Reynolds, R. E. and Antos, S. J. (1978). Interpreting
metaphors and idioms: some effects of context on comprehension. Journal of
Verbal Learning and Verbal Behaviour 17: 465–477.
208 Vocabulary
Pawley, A. and Syder, F. (1983). Two puzzles for linguistic theory: nativelike
selection and nativelike fluency. In Richards, J. & Schmidt, R. (eds.), Language
and Communication. London: Longman, pp. 191–226.
Randall, M. (2007). Memory, Psychology and Second Language Learning. Amsterdam:
John Benjamins.
Saville-Troike, M. (1988). Private speech: evidence for second language learning
during the ‘silent’ period. Journal of Child Language 15: 567–590.
Saville-Troike, M. and Barto, K. (2017). Introducing Second Language Acquisition
Saxton, M. (2017). Child Language: Acquisition and Development. Second edition.
London: Sage Publications.
Schmitt, N. (2000). Vocabulary in Language Teaching. Cambridge: Cambridge
University Press.
Teaching. London: Macmillan.
Singleton, D. (1999). Exploring the Second Language Mental Lexicon. Cambridge:
Tabossi, P. and Cacciari, C. (1988). Context effects in the comprehension of
idioms. In Proceedings of the Tenth Annual Conference of the Cognitive Science
Society, August 17–19, 1988, Montreal, Quebec. Conference Co-Chairs V. L. Patel
and G. J. Groen. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 90–96.
Taylor, J. (2003). Linguistic Categorisation. Third edition. Oxford: Oxford Univer-
sity Press.
Todd, L. (1987). An Introduction to Linguistics. Harlow: Longman.
Ullmann, S. (1962). Semantics Oxford: Blackwell.
Ur, P. (2012). Vocabulary Activities. Cambridge: Cambridge University Press.
Ur, P. (2019). Teaching multiword units: a teacher’s perspective. In Hinkel, E. (ed.),
Teaching Essential Units of Language: Beyond Single-Word Vocabulary. London
and New York: Routledge, pp. 19–35.
Ur, P. and Wright, A. (1992). Five-Minute Activities: A Resource Book of Short
Activities. Cambridge: Cambridge University Press.
Verspoor, M. and Tyler, A. (2013). Cognitive linguistics and second language
learning. In Ritchie, W. C. & Bhatia, T. K. (eds.), The New Handbook of Second
Language Acquisition. Second edition. Leiden: Brill, pp. 159–177.
Weir, R. H. (1962). Language in the Crib. The Hague: Mouton.
Yorio, C. (1980). Conventionalised language. TESOL Quarterly 14/4: 433–442.
Yule, G. (2017). The Study of Language. Sixth edition. Cambridge: Cambridge
University Press.
Zahar, R., Cobb, T. and Spada, N. (2001). Acquiring vocabulary through reading:
effects of frequency and contextual richness. Canadian Modern Language Review
57/4: 541–572.
Dictionaries
(1995). Concise Oxford Dictionary. Ninth edition. Oxford: Oxford University Press.
(1995). Webster’s New Encyclopedic Dictionary. Revised edition. New York: Black
Dog and Levanthal Publishers Inc.
8 Listening and Speaking
The Spoken Mode

Speech precedes writing in two ways: firstly, children learn to speak
before they learn to write and, secondly, all languages develop a spoken
form before they develop a written form. Of the four basic language
skills: listening, speaking, reading and writing, listening and speaking are
therefore regarded as primary and reading and writing as secondary
skills. Literacy, unlike speech, is not an acquired but a learned ability
and has to be taught (Ur 2012: 151; Harmer 2004: 3). Even with tuition,
success is not assured as witness the sizeable minorities of adults in
countries with universal education who nevertheless remain functionally
illiterate. The major world languages all exist in speech and writing, but
there are many less well-known languages in the world which have not
developed a written form, and may never do so. Conversely, a language
may in the course of time lose its speakers but be preserved in written
records, as in the case of Latin.
Another way of dividing up the four skills is into receptive skills
(listening and reading) versus productive skills (speaking and writing).
However, it is not helpful to regard listening and reading as ‘passive
skills’ and speaking and writing as ‘active skills’ (Johnson 2018: 240). In
fact, all linguistic communication is to a greater or lesser extent cryptic
and the receptive skills require active reconstruction of meaning on the
part of the listener or reader rather than mere reception of a signal (Field
2008: 209; Anderson and Lynch 1988: 11–15). Listeners no less than
readers have, as a matter of course, to ‘read between the lines’ (Sperber
and Wilson 1995; Clark 1996). Much can be left unsaid because it can be
inferred from the context of situation (Malinowski 1923; Firth 1957;
Halliday 1978: 28–31, 60, 109). Context of situation refers not just to the
physical context or setting but also to the psychological context. The
terms ‘field’, ‘tenor’ and ‘mode’ are sometimes used to specify three
aspects of the context of situation. ‘Field’ refers to the topic, ‘tenor’ to
the relationships among the participants and ‘mode’ refers to whether
the text is written or spoken and what sort of written or spoken text it is
(Halliday 1978: 33, 62–64). All these factors combine to define the
‘register’ of the spoken or written text, which is the particular brand of
language which marks its text-type or genre (Halliday 1978: 31–35;
Wallace 1992: 30–33).
A spoken text and a written text from the same field, for example
instructions for operating a washing machine, will each have a distinct
register. One reason for this is that written texts lack the suprasegmental
features of stress, rhythm and intonation available to the spoken mode.
Another reason is that in face-to-face speech the participants can take
some aspects of the physical context for granted (the washing machine
itself), can use pronouns and demonstratives (‘press this one’) rather than
complex noun phrases (‘press the large green button immediately below
the central dial’) or just point to the object. Pointing, or more generally
gesture of all sorts, is a paralinguistic device. Paralinguistic devices
include, as well as gesture: tone of voice, facial expression and various
non-speech sounds such as sighs, grunts, groans and chuckles. Since
speech usually relies both on contextual information and paralinguistic
features to a greater or lesser extent, listening to sound-only recordings
of interactive, highly context-bound speech is often a daunting – and
unnatural – task for learners. A video download is much easier to
understand and, more importantly, it gives learners the chance to learn
a lot about paralinguistic communication and appropriate social beha-
viour in various situations (Harmer 2015: 343). It is worth remembering
that misunderstanding can occur at the paralinguistic level if, for
instance, what is intended as a friendly smile by the speaker is taken for
a self-satisfied smirk by the listener.
Style is distinct from register and refers to the scale of formality along
which spoken and written texts will vary (Halliday 1978: 66). For any given
text-type, for example, an invitation to a party, furniture assembly instruc-
tions, a joke or a marriage proposal, a written version is generally more formal
than the spoken version if other variables are held constant. This is partly
because there is a relationship between spontaneity and formality: the more
spontaneous speech or writing is, the less formal it tends to be and vice versa.
Since writing in general tends to be less spontaneous than speech, writing also
tends to be, in general, more formal than speech. However, the register of
a highly formal spoken genre such as a legal address is likely to be much more
formal than that of a very informal written genre such as a postcard or
a memo (Johnson 2018: 254). The level of formality of an online chat may be
very close to that of conversational speech (see Ur 2012: 151–153).
There are also certain hybrid genres, which combine features of both
speech and writing, such as reading aloud and talking from notes. Tele-
phoning is an interesting case. Telephone conversations are conducted in
the spoken mode but with speaker and hearer being unable to see each
other. Children have to learn the special conventions of telephoning.
Young children may talk to you on the telephone assuming that you can
Listening and Speaking 211
see the toys they have around them without further description or
identification. Whether in one’s native language or in a foreign language,
telephoning is a special sub-skill which has to be acquired. E-mails are
another interesting hybrid form. They have their own conventions and
their own scale of formality. The smiley is an attempt to add facial
expression (a paralinguistic feature of speech) to the written mode.
For foreign language learners, proficiency in the spoken mode is not
necessarily a good guide to proficiency in the written mode and vice
versa. There are many people who can speak a foreign language quite
fluently but have not learned to read or write it. There are others who
can read and write in the foreign language quite well but who are not
fluent speakers and have problems understanding fluent speech. Broadly
speaking, classroom learners become proficient at those skills they
practise in the classroom and less proficient at those skills neglected in
the classroom. Nowadays, the emphasis is on striking a balance across
the four skills in teaching while seeking to give a communicative thrust
not just to listening and speaking, but also to reading and writing. In
some English for Special Purposes (ESP) courses a detailed analysis of
learners’ needs may be necessary so as to define their specific future
requirements in all four skills.
This chapter will focus on listening and speaking in the classroom and
the next chapter will consider reading and writing.
Listening
Why Listen?
In real life we often do not just listen, we listen and respond. Response is
not always verbal, for example we may nod, smile, shake our heads or
frown. Spoken genres vary according to whether they expect no response,
a non-verbal response or a verbal response. Sometimes a response is
mandatory, sometimes it is optional. Sometimes a response is expected
only after the speaker has finished talking, sometimes ongoing responses
are permissible or even expected: either verbal (‘OK. I’ve got that’ or
‘Sorry, I don’t quite follow that’) or non-verbal (nodding, smiling). In
some genres, such as the conversation, the listener can interrupt the
speaker so as to ask clarification questions. This is also possible in
seminars, but is less possible in lectures, quite impossible in sermons and
if you try it in political speeches, it is called heckling. Some other spoken
genres do not permit a verbal response for technical reasons – for example
radio news broadcasts, loudspeaker announcements in public places and
answerphone messages.
We also listen for different purposes and the purpose for which we listen
influences the way we listen, our listening style (Harmer 2015: 336; Scrive-
ner 2011: 252–253). There is a link between genre and listening style, but we
may also listen to the same genre in different ways for different purposes on
different occasions (Anderson and Lynch 1988: 30; Field 2008: 229–230).
Learners need practice in at least three sorts of listening:
• listening for gist: ‘Has anything of interest happened in the world?’

(radio news broadcast)
• listening for significant detail: ‘Is that my registration number?’
(loudspeaker announcement)
• listening for implications: ‘Why did she really phone me?’) (answer-
phone message)
Concerning listening for gist, Field (2008: 244) points out that listening-
comprehension lessons in which learners work through a list of questions,
each of which refers to a specific sentence, may actually deprive learners of
the opportunity to listen for gist. He suggests using authentic recordings
where the speaker rambles or digresses, as in some radio interviews, and
asking learners to decide for themselves what the essential points are (Field
2008: 245–246).
Listening Difficulties
Each genre presents its own special difficulties. Speech accompanied by
visuals is often easier to understand than speech without visuals. This is
especially true of commentaries, for example a TV sports commentary of
a football match or tennis match compared to a radio commentary of the
same match. The classic spoken genre which cannot rely on visual input
is the telephone conversation. Talking on the phone is difficult in
a foreign language.
Other things being equal, description is easier to understand than
narrative, which in turn is easier to understand than argumentation,
opinion and evaluation. It may be particularly difficult to distinguish
between fact and opinion if a speaker mixes the two (Anderson and
Lynch 1988: 60). Concerning narrative, if events are presented in
chronological order, this is easier than if there are flashbacks or flash
forwards. The more characters there are, the more difficult it is to
follow a story, especially if some main characters are easily confusable,
for example two boys rather than a boy and a girl. Less proficient
readers may also find it difficult to cope with pronoun reference and
listeners
the use of different nouns (for example, ‘vehicle’, ‘car’, ‘Ford’) to refer
to the same thing rather than repetition of the noun (Anderson and
Lynch 1988: 48–55).
Another difficulty for learners is that style shifting may occur within
a given genre for rhetorical effect. For example, lectures are a formal
genre and generally require formal language. However, in English-
speaking cultures, lecturers like to spice their lectures with informal
asides and even jokes in a way that learners from some other cultures
may not be prepared for (a cross-cultural difference). At worst, learners
may be thrown quite off track and lose the drift of the lecture because
they focus on trying to understand an aside which is actually not crucial
for understanding the lecture at all.
However, many learners have problems at a far more fundamental level.
The subjective impression they have is that people speak too fast for them so
that they quite simply cannot make out the words (Scrivener 2011: 249).
Learners do not necessarily need to make out every word, but if they can’t
make out word boundaries in general then they will be unable to use their
knowledge of English vocabulary and grammar to understand what is being
said. Learners often mention the following problems:
separating out individual words in the flow of spoken discourse

distinguishing between specific phoneme pairs
coping with a specific regional or social variety of speech
coping with rapid speech
coping with informal speech
coping with incomplete syntax, repetitions, non-standard grammar
and colloquialisms
coping with various forms of background noise
coping with speakers chipping in on another speaker, overlapping
speech turns and different speakers speaking at the same time
Field (2008: 157–161) notes that problems of word recognition often

arise because the rhythmic patterns of strong and weak syllables cut
across word boundaries in fluent speech, so that ‘went in’ may sound
like ‘when tin’ and ‘I scream’ like ‘ice cream’. More generally, learners
may not be prepared for weak-form pronunciations, for elisions and
assimilations (these things were discussed in Chapter 5). Learners may
therefore benefit from being asked to identify the word in each intona-
tion unit which bears the nucleus (tonic syllable) since these are the
words which are likely to be the important ones (Field 2008: 194–203).
Such awareness-raising will involve repeated listening to specific short
stretches of a listening text which learners have found difficult to under-
stand. Learners should also be encouraged to consider context so as to
exclude unlikely meanings.
Getting Started
In the early stages, listening texts should be about things with which
learners are familiar. For children in primary school, suitable topics are
school itself, the family, hobbies, pets, shopping, sports, clothes, food
and drink. Familiarity with the genre to which a spoken text belongs will
also make the listening task easier, as will background knowledge about
the speaker, the situation and the topic. Sometimes one can ‘understand’
the spoken text just by inferring meaning from prior contextual knowl-
edge. Here is a real-life example:
An English family with young children went into McDonalds in

Florence. After they had successfully ordered in English, the sales
assistant asked them something in Italian which they did not under-
stand. However, knowing that McDonalds in Manchester were run-
ning an offer of plastic toys and at this point the sales assistant would
have asked if the customer wanted the plastic toys or not, the mother
nodded, said, ‘Yes, please’ and promptly received the plastic toys.
This technique of predicting utterance meaning from what one already

knows is ‘top-down’ processing. The opposite, working out utterance
meaning from the known meaning of the constituent words and phrases
is ‘bottom-up’ processing (see Field 2008: 132–133; Scrivener 2011:
257–260; Harmer 2015: 337–339; Johnson 2018: 241–248). Genres with
which learners are unfamiliar will be difficult for them because they
cannot predict what is likely to happen next. Prediction skills can be
developed in learners of all ages by stopping in the middle of a story and
asking learners what they think will happen next. Anderson (2004:
82–85) presents a story with multiple cliff-hanger points and four options
to choose from at each point.
In the early stages learners can often understand more than they can
say, and they can benefit from listening activities which involve respond-
ing without speaking (see Hadfield and Hadfield 1999 for a collection of
such activities). A simple task along these lines is the ‘zoo task’ (Ur 2012:
112–113). It is suitable for young learners who have learned the names of
wild animals and how to express locations in space. Each learner is given
a plan of a zoo with cages numbered and spaces left to write in the names
of the animals. Various landmarks, such as a café, entrances and
benches, can be included. The teacher then describes a route she takes
through the zoo and names the animals she sees in each cage. Learners
write in the names of the animals while she is talking (appropriate pauses
being made for learners to write). The task can be varied to suit different
proficiency levels in terms of speed of delivery as well as frequency and
length of pauses. In some versions learners are allowed to ask questions
or can ask the teacher to repeat. To eliminate the writing element, which
may be slow and laborious for young beginners, learners can be given
pictures of the animals, which they merely have to place in the right
cages. In a further step, they can be given the names of the animals to
match with the pictures and they can then copy the names into their plan
of the zoo later.
A suitable listening task for young beginners who have just learned the
colours and parts of the body is the ‘my monster task’. The children
each receive an outline drawing of a humanoid ‘monster’. The teacher
tells them the colour of each part of its body, one sentence at a time
(‘The monster has green arms …’), and after each sentence gives learners
enough time to start to colour-in that part appropriately before she
moves on to the next part. At the end of the teacher’s talk the learners
can complete their colouring-in or amend their first attempt as necessary.
If the teacher sees there are many mistakes or uncertainties, she can
repeat her monologue. In a follow-up writing exercise the learners can
then label the different parts of the body on their picture, copying from
flash cards.
Storytelling is another means of giving learners of all proficiency levels
listening practice for real-life listening. It also makes it possible to
introduce speaking based on listening. For children one can use fairy
stories and fables. It is usually best if the teacher tells the story in her
own words rather than reading a written version verbatim or using a pre-
recorded story. Showing the children accompanying pictures can be
helpful in easing comprehension and providing practice in predicting
what might come next. Thus, a picture-book version of a story is often
an advantage. It is best if the children sit in a circle or horseshoe around
the teacher. Young children often like to hear the story over and over
again. They also like to say out loud in chorus certain lines which may
repeat themselves with variation at key points (‘Grandma, what big ears/
eyes/a big nose you’ve got’ in Little Red Riding Hood). Individual children
may then like to take over short pieces of dialogue themselves and even
to act out part of the story in front of the class. Often they want to
repeat this over and over again until they have all had a turn. At the end
of the story learners can be given a series of pictures which they have to
put in the correct order to represent the story.
Moving On
Beginning learners of all ages should be given ample time to get used to
their teacher’s voice before being introduced to new voices (Field 2008:
159–160). As learners become more proficient, however, they need to
move on to listening to a wide range of voices. Ur (2012: 101–106) points
out that to prepare learners for real-life listening, recorded material
should be used so as to provide learners beyond beginner level with
different varieties of English, different speech styles and different con-
texts, as well as practice in listening to monologues, dialogues and multi-
logues (more than two speakers). Harmer (2015: 340) distinguishes
between live listening and recorded listening. There are advantages and
disadvantages to both.
One advantage of live speakers is that they can adapt what they say to
the audience as they go along and learners can ask them questions or
engage in dialogue with them (Thornbury 2005: 57; Harmer 2015:
342–343). Live speakers include the teacher, other learners and invited
guests as well as people interviewed live via Skype. Live listening can
include listening to the teacher or other learners telling personal anec-
dotes, reporting items of news, including ‘gossip news’, telling short
stories with a ‘sting in the tail’ such as the Nasruddin stories (Maley
1993: 107–117) or telling jokes. To achieve the transition from listening
to speaking, the teacher can tell one group of learners a suitable story or
joke while the other groups are engaged in another activity. Individual
learners from the story/joke group can then, armed with a written list of
keywords, retell the story or joke to another group (see Anderson 2004:
81 for a list of jokes and keyword skeletons). If jokes are used, practice
in inferencing is provided since jokes often depend on implication.
One advantage of recorded materials is that they can be replayed over
and over again, in part or as a whole. Recorded material can be
subdivided into didactic recordings and authentic recordings. Didactic
recordings may either accompany a general course book or a book for
self-study. Authentic recordings include podcasts of TV or radio pro-
grammes as well as talks and lectures available on YouTube, such as the
TED Talks (for advanced learners). Quasi-authentic recordings can also
be made especially for the lesson, either by the teacher or the learners,
for example an interview with some person of interest. The defining
characteristic of authentic texts is that they are not intended for language
teaching. Authentic listening texts are generally preferable to didactic
listening texts since the latter are usually scripted and graded (simplified)
and therefore their language is contrived and unnatural (Field 2008: 284).
The linguistic difficulty of authentic listening texts can often be eased by
providing scaffolding in the form of maps, pictures, diagrams and charts.
Such scaffolds also enable learners to interact with and react to the
spoken text. It is often advisable to keep authentic listening texts short
but to give learners the chance to listen a number of times, each time for
a different purpose.
Checklist for Selecting Listening Texts

When selecting listening texts, it is important to appeal to learners’
interests and to their own perceived future listening needs. If learners
can relate the listening text to their own lives and linguistic needs, they
will mobilise all their resources to understand and will achieve far above
their usual levels. Conversely, if they lack interest in the topic and
perceive the text as irrelevant to their needs, they will mentally switch
off at an early stage. The following checklist can be used to help select
listening-comprehension texts. The more questions to which the answer
is ‘yes’, the more suitable the text is likely to be:
Is the text interesting enough for the learners (not the teacher!) to
listen to?
Is the text an authentic spoken text rather than a written text read
aloud or a contrived text recorded for teaching purposes?
Does the text sound natural as an example of its genre?
If the text originally had a visual track which has been removed, is it
still readily comprehensible despite the loss of the visuals?
Is it feasible to activate necessary background knowledge in the pre-
listening phase or a preceding lesson?
Will learners be able to understand enough of the text to feel a sense
of achievement rather than frustration?
If the answer to the previous question is ‘no’, is it possible to supply
sufficient scaffolding, especially visual material, to solve the problem?
Is the information content of the text suitable: neither too much nor
too little for learners to cope with in the number of hearings
planned for?
Can learners interact suitably with the text, either non-verbally or
verbally?
Will learners have the feeling they have learned something worth-
while from the text, not only linguistically but educationally?
A Three-Phase Lesson Structure

In general, formal listening-comprehension sessions using recorded mate-
rial are best structured into three phases: a ‘pre-listening’ phase, a ‘while-
listening’ phase and a ‘post-listening/reflection’ phase.
Pre-Listening Phase
A listening text should never just be sprung on learners as a surprise,
since listening in the real world usually takes place within the context of
background knowledge about the topic and genre as well as clear
expectations of the speaker and some motivation to listen. The pre-
listening phase should therefore aim to prepare the reader for the
upcoming text by:
arousing interest in the topic

priming latent topic knowledge
tapping into relevant vocabulary learners may already know
exploring key ideas which underlie the text and are crucial for
comprehension
relating the topic to the learners’ own world
The pre-listening phase is often best organised as whole-class brainstorm-

ing. A suitable picture or title is often a good lead-in and can be used as
a stimulus for speculation on the likely content of the upcoming text.
However, an inappropriate title or lead-in picture merely confuses lear-
ners and sets them off on the wrong trail. Harmer (2015: 352–353)
suggests sometimes using a word cloud (www.wordle.net) to help lear-
ners speculate on what a story might be about before they hear it. As
teacher, one should try to elicit information from the class rather than
giving a mini-lecture. Only if no one in the class can supply a specific
piece of necessary information should the teacher do this, and only as
a last resort. Any new vocabulary introduced should be kept to an
absolute minimum and should arise naturally from discussion. See also
Harmer (2015: 336), Johnson (2018: 249–250) and Field (2008: 214–220).
While-Listening Phase
Although in real-life listening there is usually no chance to listen for
a second or third time, in the classroom this is possible. Learners can
benefit greatly from repeated listening and in this respect the classroom
is an advance on the real world for foreign language learners. The more
interested a class is in the text, the more they will be prepared to listen to
it repeatedly. But fine judgement must be exercised here and skilled
teachers stop just before learners become bored. Learners should always
be given a distinct purpose for listening at each hearing, for example:
first hearing: listening for gist (global listening)

second hearing: listening for specific information
third hearing: listening critically
This procedure is similar to that suggested by Scrivener (2011: 254). It

can and should be varied for specific listening texts, however, and not
become ritualised. For example, in some cases it might be appropriate to
devote a first hearing to working out the role relationships of the speak-
ers or the variety of English being spoken or the likely setting of
a conversation. Harmer (2015: 346–347) describes a listening task in
which learners are asked to put a picture sequence into the correct
order while listening to a story.
Learner groups of limited proficiency cannot be expected to listen to
rapid speech for too long a stretch of time. Instead, the text should be
divided into sections and the recording stopped at various points to give
them a chance to ‘catch up’ and check that they have understood the
drift of the text so far before the listening task is resumed. These sections
should be of just the right length to ‘stretch’ learners’ comprehension
stamina but not to overstep the limits. The length of text to be listened
to between stops should then be gradually lengthened in the medium
term until learners can listen for several minutes at a time and enjoy the
experience.
Generally speaking, a video recording is preferable to an audio-only
recording. Harmer (2015: 344) suggests video viewing can be varied in
three ways: by including silent viewing, by using a ‘freeze-frame’ or by
using subtitles. English subtitles rather than native-language subtitles should
be used, however, since it is virtually impossible for learners to take in the
English soundtrack while reading subtitles in their native language. Switch-
ing between soundtrack and subtitles is easier if both are in English. If
a particular video is too difficult for learners even with English subtitles,
then it is unsuitable for them and an alternative video should be found.
A cautionary word: one should not try to extract every last grain of
meaning from listening texts, particularly at the lexical level. Vocabulary
is of interest only in so far as a particular lexical item may hamper or
hinder text comprehension at the level desired. It is not necessary and
often not even desirable that learners understand every single word.
Post-Listening Phase
In the post-listening phase learners should be encouraged to express their
own reactions to, and opinion of, the text. Beyond this, time might be
devoted to building on and extending any key vocabulary or grammar
encountered. As homework, further reflection and evaluation might be
done or further language work related to the listening text.
Songs as Texts for Detailed Listening

A special sort of listening task is listening to songs. Songs are particularly
suitable if video technology is not available in the classroom, since the
visual presence of the singer does not necessarily make it noticeably
easier to understand the song. Songs present their own special difficulties
of comprehension, however. In particular:
syllables may be prolonged (drawls) or shortened to fit the melody

words may be run together
in some styles of singing there may be systematic phonological
modifications, such as the replacement of ‘-ing’ by ‘-in’ and of
‘going to’ by ‘gonna’ in pop songs
non-standard grammar may also be used, for example ‘I ain’t’, for
‘I’m not’ and the use of ‘double negatives’ as in, ‘You don’t love me
no more’
some folk songs, nursery rhymes and traditional songs contain archaic
syntax, morphology and vocabulary, which may need to be glossed
On the other hand, some features of songs, especially rhythm and

rhyme, aid memorisation, while refrains, repetitions and modified repeti-
tions ease comprehension as well. Furthermore, listening to songs
represents an authentic, real-world task. Unlike most listening texts,
songs can be used profitably for detailed listening since learners often
want to know what songs mean right down to the last word. All these
advantages of songs far outweigh the negative points mentioned above.
The three-phase structure of ‘pre-’, ‘while-’ and ‘post-listening’ can be
applied to songs:
Pre-Listening Phase for Songs

As for any other listening text, a pre-listening phase can be used to prime
background knowledge, including knowledge of the singer or band and
the sort of songs they sing. Learners may be pleasantly surprised to find
they have a chance to use some international English terminology already
known to them, such as, for example, ‘boy group’, ‘drummer’ or
‘vocals’. This will build confidence and encourage them to contribute to
discussion.
While-Listening Phase for Songs

The normal procedure is for learners to listen to the song at least once,
and perhaps twice, with the focus on global understanding without the
song text. As a next step, learners can then be given a gapped version of
the song text (cloze text) and they can listen to the song again and fill in
the gaps (detailed listening). As a variation, advanced learners might be
allowed to work in pairs with the gapped song text and try to predict
words before the song is played and then listen to the song to see if their
predictions are confirmed.
Song texts are usually easily available on the internet, but teachers will
have to produce their own cloze versions. The classic cloze procedure is
to automatically delete, say, every seventh word. However, one can
choose specific words to delete instead. Another possibility is only to
delete words of a certain word class. This helps raise grammatical
awareness levels in learners. The first time one constructs a cloze song
text it is advisable to test it on a colleague in advance to make sure it is
‘doable’ for the learners. It should be challenging but not beyond them,
since this is a learning activity not a test. If the original cloze activity does
seem too difficult, there are many ways of reducing difficulty. For
example, two, three or four words can be provided at each gap from
which learners have to choose (multiple-choice technique). Another
possibility is to specify the number of letters for each word by using
dashes or providing the first letter or even the first syllable of the missing
words. During the cloze listening it is best to stop the recording after
each stanza to give learners time to catch up. After filling in the gaps,
learners can be given a chance to confer with other learners in pairs or
small groups. They can then be given the full song text to compare with
their own versions. Then they can listen to the song once again while
following the written text.
There are techniques other than cloze available. One possibility is to
insert extra (plausible) words here and there in the written text so that
learners have to listen closely and cross out those words which are not in
the song. A variation is to replace some words or phrases by others.
Another problem-solving activity involves learners reconstructing the
song text on the basis of a mutilated text (consisting of either half lines
or half stanzas) or learners can, instead, be given the lines of a song in
jumbled order. This is well-suited to a group activity either before or
after listening to the song. Harmer (2015: 358) suggests deleting the last
word of each line from a rhyming song, giving learners the words and
asking them to put them back in their correct places.
Post-Reading-Phase for Songs

This can be used for reflection on features such as rhyme, refrains,
repetition and repetition with variation, provided these are of interest
to the learners. Comparison can also be made with other songs learners
know. Rather than specifying the precise direction of this phase in
advance, it is advisable to have a number of options prepared and be
flexible enough to follow up on learners’ inclinations. In the spirit of
individualised learning, learners can work in groups, each group focuss-
ing on a different aspect, and their results can be presented to the whole
class afterwards. Role-play is also a possibility. Learners can take on the
role of music critics (some pro and some contra) and evaluate the song.
Another role-play activity involves a music critic interviewing the singer
or songwriter and asking what was meant by a particular word or phrase.
Speaking
The Limitations of Traditional Classroom Discourse

Learners need to practise using the spoken genres they will use in the real
world, even though these may not always be easy to predict. Traditional
teacher-fronted classroom discourse is a special genre in itself with its own
particular conventions and it does not necessarily prepare learners for the
real world. In the traditional classroom it is the teacher who decides what
is talked about, who talks to whom, when and for how long. The usual
pattern is that the teacher asks questions of the class to which he or she
knows the answer anyway (display questions) and learners have to give the
answers the teacher wants. Questions may be addressed to the whole class
or a specific learner may be nominated to answer by the teacher (Sinclair
and Coulthard 1975). As a result, learners may be at a loss for words in
real-world communication, where they may need to initiate conversation,
plan and steer discourse and adapt to the linguistic, paralinguistic and non-
linguistic reactions of their interlocutors and audience. As we will see in
Chapter 9, teachers’ display questions have a part to play in classroom
discourse. They provide scaffolding and guidance for learners. However,
the main focus should be on genuine communication in the classroom
among the learners themselves.
The following are some tell-tale signs of impoverished classroom
discourse:
Learners do not initiate speech themselves but only respond to

teacher questions or directives.
Learners give the answer they think the teacher is looking for.
When learners do speak, they rely heavily on the vocabulary and
structure of a teacher question or an available text.
Learners mainly answer display questions, retell stories which every-
one already knows or describe pictures everyone can see anyway (see
Scrivener 2011: 217).
Participation is limited to a few proficient and/or self-confident
learners.
Less self-confident learners tend to avoid answering questions
altogether.
Many learners do not speak at all.
Learners who do speak often produce only short isolated utterances.
One or two learners may try to answer almost every question
immediately regardless of whether they have anything meaningful
to say.
Spoken Classroom Activities

To make speaking activities reflect real-world communication, it is
necessary that learners talk mainly to one another in small groups,
which increases learner talk-time and gives more learners a chance to
speak. It is also less daunting and more natural than speaking out loud to
a teacher, standing in front of the whole class with everyone listening.
This is best done by introducing task-based spoken activities such as
presentations, interviews, simulations, role-play, mock trials, quizzes and
games as well as project work. These give learners the chance to speak
for more than just one short utterance at a time. Since task-based
activities are goal-oriented, they are usually more motivating than topic-
based activities such as discussions or debates (Ur 2012: 121). Activities
should be pitched so that they are cognitively rather than linguistically
challenging (no new structures and a minimum of new vocabulary). Some
preparation on the conventions of certain activities, for example a mock
trial, may be necessary. Learners may also need preparation for some of
the roles which they may be asked to adopt. Seating arrangements are
important in all spoken activities to ensure that learners can face one
another and are quite close to one another (Scrivener 2011: 216).
If possible, a choice of activities should be offered and the choice
varied. The class should be encouraged to continually try new activities.
Only if a class proves to be really averse to a particular activity after
several tries should it be dropped. Variation on the format of
a particular activity-type can give it a facelift and restore its waning
appeal. The stimuli used for sparking off an activity should also be
varied (not always brainstorming, not always a headline, not always
a picture, not always a YouTube video) and the sources should also be
varied (for example, not always the same website). To ensure the smooth-
running of group activities and the participation of all learners, appro-
priate roles can be assigned to some group members, such as language
watchdog, timekeeper, chairperson, speaker, secretary and referee.
Small-group work, or, even better, pair work, is particularly suitable for
anxious learners because it is less threatening than talking in public. For all
learners, speaking should proceed from pair-based activities to speaking in
small groups, then in larger groups, then presenting in pairs or small
groups to the whole class before individuals finally present before the
whole class. Notes, pictures and other aids to presentation may be used as
scaffolding and then gradually withdrawn as learners become more profi-
cient and self-confident. It is easiest for learners to talk about themselves,
and even beginners should be given the chance to introduce themselves to
a partner or small group and say a little about themselves.
The problem for many learners is translating language knowledge into
fluent performance (Johnson 2018: 232; Scrivener 2011: 213). To achieve
this they need the opportunity to communicate in English in an unthrea-
tening atmosphere. They should be put in situations where they want to
say something and are willing to risk incorrect English in order to
communicate. Practice in speaking should therefore be done largely
without error correction, although scaffolding may sometimes be pro-
vided (see Scrivener 2011: 224–228). The teacher’s primary role is that of
the orchestrator who sets up the situation and then during the activity
becomes an unobtrusive observer who usually tries to speak as little as
possible so that learners spend more time speaking. However, the teacher
has a number of secondary roles, including those of encourager, provider
of scaffolding and feedback, supportive listener, prompter, conflict-solver
and stage manager.
Provided the class has no objection, some oral activities can be video-
recorded and reflected on in a follow-up review session or in one-to-one
consultation with the teacher. Rather than focussing on errors made,
however, it is more profitable to focus on the pragmatics of speech, for
example conversational rules of turn-taking, interrupting, correcting,
disagreeing and chipping in. Following a role-play it might be useful to
examine how learners have used phatic language (language which oils the
wheels of social interaction) or to examine affective aspects of language:
do learners come across in their role as inappropriately and unwittingly
rude, over-polite, intimate, distant, jovial or downcast, for example.
Some groups of learners may voice discontent at the lack of error
correction (Johnson 2005; Scrivener 2011: 225). For such groups, some
formal feedback on error after the activity has been completed can be
provided (see Scrivener 2011: 226 for some suggested techniques).
Some major types of spoken activity will now be reviewed.
Discussions and Debates

These are both topic-based rather than task-based activities (Ur 2012:
121–122). The difference between a discussion and a debate is that the latter
is far more structured. It involves a motion, usually to abolish, introduce,
limit the availability of, ban or legalise something. The motion is proposed by
one speaker in a short opening speech and briefly seconded by another
person. It is then opposed by another person in a speech of similar length to
the proposer’s (time limits should be set). The opposing speaker provides
counter-arguments and is briefly seconded by another person as well. Then
the debate is thrown open to the floor (the class in our case). Those who wish
to speak must raise their hand and are called upon to do so by a chairperson
sitting at the front. The chairperson also generally keeps order, sees that no
one interrupts a speaker, ensures speakers speak one at a time, does not allow
speakers to wander off topic and does not permit personal abuse of another
speaker or the chair. Speakers who talk for too long are told by the
chairperson to come to a close (time limits can be set). Additional rules can
be negotiated by the class beforehand.
The success of a debate or discussion is always unpredictable, even if
one has prepared well for it. Without preparation, failure is pre-
programmed and for this reason unstructured discussions should never
ever be used as fillers. Structured debates focus on coming to a decision
and are therefore more likely to be successful than discussions, which
lack this focus (Scrivener 2011: 214). The chances of success for both
discussions and debates are enhanced if they are conducted as a follow-
up to a sequence of lessons on the topic in question. Success is even
more likely if preparatory sessions can be used for guided fact-finding
research by learners, involving ordering, sorting, sifting and selecting
background information, which may often be available online. For
cautionary words on discussions, see also Johnson (2018: 236), Scrivener
(2011: 211–212) and Ur (2012: 121–122).
Role-Play
Role-play is closer to improvised drama or free theatre than to conven-
tional playacting, for it is not pre-scripted. A group of learners act out
a scene, each learner taking on another identity. The characters whose
roles learners play have personalities, motivations and backgrounds
which are set out on a role card for each character and given to the
player. The outcome of the scene is, however, left open. The focus
should be on achieving things with language rather than on producing
correct language. Role-play should also be fun. Learners interact with
others in discourse which spans more than just one or two turns and
may develop in unpredictable ways. If role-plays are well chosen, parti-
cipants may have to defend a point of view, be conciliatory or be firm in
refusal. In this way they develop communication skills which will stand
them in good stead for real-world linguistic encounters. As in other
spoken activities, the teacher’s role is essentially that of facilitator.
One role-play that works well with adolescents is the situation of
a daughter or son who wishes to stay out later than parents wish for
a concrete, specified occasion (for example a birthday party). The role
cards should specify differences between the mother’s and father’s attitudes
to enable strategies of ‘divide and conquer’. The son or daughter may be
supported by, for example, an older sibling and opposed by, for instance,
a malevolent uncle or aunt. A compromise solution may be reached.
The collaborative effort involved in role-play is likely to enhance learner
solidarity and group coherence, thus creating a relatively safe environment
for language use compared to the real world, where there is always the risk
of losing face. Reticent, even anxious learners who might not otherwise
volunteer to speak are more likely to contribute actively when given a role
to play (Harmer 2015: 393; Scrivener 2011: 219). All learners will experience
the laying aside of their own personality and the assuming of another as
liberating and conducive to speaking. This effect can be enhanced if learners
dress up for the role, even if this only involves donning the appropriate hat,
wearing a jacket, knotting a tie round one’s neck or wearing sunglasses.
A store of costumes and props, ideally provided by the learners themselves,
can be kept in the classroom. Great attention should be paid to seating
arrangements to make the role-play as realistic as possible.
Role-play is suitable for all ages and for all proficiency levels beyond
absolute beginners. For less advanced learners, role-plays in simple
transactional settings such as restaurants, hotels, shops, the post office
or the bank are suitable. ‘The Restaurant’ role-play is a good one for
novice teachers to start with. The class is divided into groups of guests
seated at a number of tables. Additional roles are waiters and waitresses,
perhaps a head waiter and the chef, too. Guests each receive a menu.
Some specific language preparation will be required. The teacher can
build-in to the role-play a point where negotiation has to take place. For
example, one customer may have information on the role card which
says he or she is vegetarian and is particularly fond of a specific dish. On
the waiter’s role card it may state that, although this dish is on the menu,
it is not available today. In this way, as in real life, the problem crops up
without warning and negotiation has to take place.
Role-plays suitable for intermediate level learners include: ‘The
Shop-Lifter and the Store Detective’ (see Scrivener 2011: 221–222)
and ‘Buying a Family House’, which involves choosing a house from
a number of (preferably authentic) estate agents’ descriptions (see
Anderson 2004: 8–13). As in ‘The Restaurant’, conflict can be built
in to the role cards by giving different family members different
priorities (quiet and peaceful neighbourhood versus easy access to
shops, cinemas, bars and discos).
For advanced learners, more complex settings and tasks may be
appropriate, such as casting for a play or film, interviewing for jobs,
choosing a site for a facility such as a swimming bath in a town, deciding
among rival tenders for a construction contract or dealing with an
emergency such as a road accident or even a terrorist attack. A very
formal role-play is the mock trial, which has to be carefully prepared for
(see Anderson 2004: 50–53 for an example and a cautionary word about
possible pitfalls of an intercultural nature). For collections of published
role-plays see Anderson (2006) and Ladousse (1987).
Rather than holding a discussion in class, it is often better to treat the
topic within a role-play. This is the ‘committee makes a decision’ brand
of role-play. I once observed a very lively and successful role-play
discussion in a German school class of 16-year-old learners. It was
a follow-up to a unit in the course book on gun laws in the USA. The
situation for the role-play was that the local German town council was to
debate whether to allow more liberal gun laws on the American model in
their town (a highly improbable eventuality but one to which the learners
could easily relate). Roles were cleverly assigned to cover pro and contra
positions, including some guest speakers from the twin town in the
USA, among whom was the sheriff. There was a vote at the end. The
element of dramatisation much enlivened what might otherwise have
been a rather dry discussion.
Young learners will take to role-play naturally. Adolescents may need
some encouragement in order to overcome barriers of self-
consciousness. For adults it may be necessary to explain that there are
very good didactic reasons for role-play, and it is not just about being
silly. However, some individuals or groups may have inhibitions and no
one should ever be coerced.
A role-play consists of three phases: setting up, acting out and reflection/
evaluation. A role-play also involves a certain amount of pre-class prepara-
tion by the teacher. This includes:
choosing the topic

planning how much class time is needed for each of the three phases
assigning roles
preparing role cards
writing clear instructions for the class concerning the situation and
roles
supplying extra support (if necessary), such as large prompt cards,
realia, props
language preparation: identifying any key vocabulary or key struc-
tures to pre-teach
psychological preparation: planning how to sell the role-play to
learners
Roles should be assigned carefully. Generally, males should play male roles
and females female roles. An individual may be selected for a role because
of a positive appearance feature (for example beautiful hair, athletic build,
slim figure) but never for a negative one (for example, baldness, short-
sightedness, obesity). Learners should not usually be allowed to assign
roles themselves as they may not be sensitive to such matters.
Even at the very earliest stages, role-play should not involve learners
preparing written dialogues which they read out. Learners at lower levels
of proficiency may indeed need painstaking preparation before they can
even enact a scene at a café or a shop, but at most they should have some
keywords or phrases written down on a card they can hold. Even then, it
is better if the teacher or another class member has a set of large prompt
cards with key phrases on them and is ready to jump into the scene and
brandish the card (without saying anything) to help a stumbling learner.
Another possibility is to assign the role of prompter to a learner who
does not have a part in the role-play. In the manner of a theatre
prompter this learner should discreetly whisper prompts from the wings.
Setting up a Role-Play
The role-play has to be explained to learners, and often written instruc-
tions are advisable. Roles have to be distributed, learners given time to
study their role cards and seating arrangements made. Costume and
props may need to be distributed too. Role-play works best with
a small number of fixed roles (three to seven). A role-play can be played
by a single group performing to the rest of the class or the rest of the
class can have a collective role in the manner of a Greek chorus. This is
better than just being spectators. Alternatively, a number of groups can
perform the role-play in parallel. With classes where there are great
proficiency differences it may be a good idea to have a group of mainly
proficient learners first demonstrate the role-play to the class, and then
let the whole class do it in groups. In this case the members of the
original group can each be assigned to one of the new groups to function
as prompter and helper.
Basic information about each character (name, age, profession) should
be made generally available. Sometimes a nameplate helps, sometimes
costume makes it clear. Before the role-play starts, each character can
briefly introduce himself or herself. For more complex role-plays there
may be psychological information on the role card which is labelled
‘secret’ and is only to be revealed in the course of the role-play.
Acting Out a Role-Play

A fixed period of time should be allotted for the role-play and rigorously
adhered to, with the teacher firmly stopping the activity after the time
has elapsed and, hopefully, while learners are still excited and talking
a lot. A role-play does not have to come to a conclusion and is often best
broken off in medias res. (An exception is the ‘committee makes
a decision’ role-play.) The teacher should be available only as a consul-
tant during the role-play, as required by the learners.
Evaluation of and Reflection on the Role-Play

Each group reports on the progress of their role-play and how they dealt
with unexpected eventualities that arose. They should then also be given
a chance to evaluate the effectiveness of the role-play, or, for young
learners, to say whether they liked it or not.
Simulation
Johnson (2018: 234) notes that some writers use the terms role-play and
simulation synonymously. Other writers, for example Ladousse (1987: 5)
and Scrivener (2011: 224), view simulation as simply large-scale role-play.
It is true that a simulation is usually more elaborate than a role-play, but
the essential difference is that in simulation, unlike role-play, learners do
not assume a fictitious identity (Johnson 2018: 234; Thornbury 2005: 98).
Thus, learners who find it difficult to take on another identity may prefer
simulation.
Simulations should aim at solving some problem or other. Like role-
play, they should have a preparation phase, a performance phase and
a reflection phase. Usually the class will be divided into groups. In
contrast to role-play, each group must come to a solution (which is
recorded by a secretary). In the reflection phase the groups each present
their individual solutions and these are evaluated in plenum by the class.
Ideally, there should be a class vote for the best solution. The reflection
phase is important and is not something that can be postponed to the
next lesson because the tension of the moment will then be lost. This
means that time-keeping must be strict.
A well-known simulation suitable for advanced learners is ‘Lost
at Sea’:
Lost at Sea
You are drifting out of control on a yacht in the Indian Ocean. The yacht is
badly holed below its waterline, is sinking fast and cannot stay afloat for much
longer. Your compass has been damaged and most of your supplies and
equipment have been lost overboard or ruined by seawater. You are unsure of
your exact position but believe you are about 500 miles south-west of the nearest
land. Below are the only items that you have on board which are undamaged
and are in working order:
a sextant a flashlight with four batteries a five-gallon can of water
mosquito netting one case of emergency rations maps of the Indian Ocean
one life jacket a two-gallon can of petrol a small radio (reception only)
shark repellent a sheet of opaque plastic (20 sq. feet) a fifteen-foot length of nylon rope
a fishing kit a bottle of whisky two boxes of chocolate bars
Working alone, write down the five items you regard as most important for
survival and the five you regard as least important. When you have done this,
in groups of six decide on a group list of the most important five items and rank
these from most important to least important. Be prepared to explain your
group choice to the whole class and to justify your ideas.
The ‘Lost at Sea’ simulation exists in various guises, including ‘Desert
Dilemma’ (Harmer 1991: 130; Johnson 2007: 266–267) and ‘Lost in the
Forest’ (Scrivener 2011: 218). Anderson (2004: 54–57) presents a simula-
tion in which learners are shipwrecked on a desert island and have to
make a series of choices, receiving a pre-prepared card after each deci-
sion, which takes the story a step further and poses a new choice for
them. In this way they are led through a decision-making maze towards
survival but may lose some of their party in the process.
Other popular simulations involving group planning activities include
booking a holiday (Scrivener 2011: 218) or looking for a flat to share
(Johnson 2018: 230–231). On simulation and role-play see also Harmer
(2015: 392–393).
Interactive Language Games

Interactive language games are suitable for learners from lower-intermediate
level upwards. The idea is to get learners talking so that they have fluency
practice. Games need to be chosen according to the interests, proficiency
level, age-group and in some cases cultural background of the learner group.
They should function without constant teacher intervention.
One well-known game suitable for learners at lower-intermediate level
is the ‘Apple Pie Banquet’ (McCallum 1980: 9). This game provides
practice in holding language verbatim in memory while searching for
a new word. Learners are told that the previous evening they attended an
apple pie banquet where there were many sorts of food to eat. The game
starts by a learner saying, ‘Yesterday I went to a banquet and I ate apple
pie.’ He or she then draws out of a hat the name of another learner who
repeats the sentence and adds a food item beginning with the next letter
of the alphabet, for example, ‘Yesterday I went to a banquet and I ate
apple pie and baked beans ….’ And so it goes on through the alphabet
(excluding the letters ‘x’ and ‘z’). Learners are not permitted to make
notes. To ease the difficulty of the game, a poster displaying various
named or unnamed varieties of food in the manner of a collage can be
exhibited.
For advanced learners, the well-known BBC radio game ‘Just a Minute’
is a suitable way of promoting fluency (see Harmer 2015: 389). The rules
are simple. A speaker must speak for 60 seconds on a set topic without
hesitation, deviation or repetition. Initially, the teacher acts as chairper-
son/timekeeper/scorer, but learners can in time take over these roles. If
the speaker hesitates, deviates from the topic or repeats information (not
necessarily word for word) he or she may be challenged by another
learner, who says ‘Challenge.’ The timekeeper now stops the clock and
the challenger gives the reason for the challenge. If the challenge is
upheld by the chairperson, then the challenger takes over and proceeds
with the topic, having to talk only for the remaining time to gain points.
Points are awarded to whichever speaker is talking at the end of
the minute; additional points are awarded if a speaker speaks for the
entire minute; points are also awarded for a correct challenge and penalty
points for an incorrect challenge. The score is kept on the blackboard.
Three points might be awarded for talking for the whole 60 seconds, two
points for talking at the end of the 60 seconds, two points for a correct
challenge and one penalty point for an incorrect challenge. Instead of
awarding penalty points for an incorrect challenge, points can be
awarded to the incorrectly challenged speaker. For a class unwilling to
challenge, and this is often a problem, points for correct challenges can
be raised and points for incorrect challenges can be lowered (or even
abolished).
Games involving ‘taboo’ words can also be fun. Learners work in pairs
or small groups. One learner is interviewed and is told in advance that he
or she is not allowed to use a particular word, for example ‘coffee’. The
interviewer or interviewers then pose questions which try to lead the
interviewee into saying ‘coffee’ inadvertently. For more advanced lear-
ners, grammatical words such as ‘and’ or ‘but’ or ‘he’ can be the taboo
word.
Team quizzes are also a good means of stimulating discussion. Shuttle-
worth (2014) presents various possibilities. A variation on the quiz is the
group-ranking task (see Scrivener 2011: 218). In groups, learners have to
prepare their own ranking list of the best something: for example,
twentieth-century inventions, composers, singers or actors.
Guessing games such as ‘What’s My Line?’ (Thornbury 2005: 84) make
it possible to introduce an element of role-play into a game. A learner
assumes a particular occupational role and a panel asks yes/no questions
to find out his or her occupation. The panel may guess at any point.
After ten ‘no’s, including wrong guesses, the candidate has won. The
game can be adapted for hobbies, favourite food and many other items.
‘Mystery film star’ is another role-play guessing game. This time
a learner assumes the role of a film star (variations include pop star or
sports star) and is provided with a fact sheet of, say, ten items about the
star. The learner, as star, then reveals one fact about himself or herself (a
clue). The remainder of the class is divided into panels of, say, four
members. Each panel now in turn asks a yes/no question and may make
an identity guess. Points are lost for each ‘no’ and each wrong guess. The
star then gives another clue and so it goes on. The panel that guesses the
identity of the star first wins. If no one guesses after the tenth clue has
been given, then the star wins. The star should keep the most helpful
clues back for as long as possible. For a different version of this game,
see Anderson (2004: 32–35). Anderson (2014) provides a large selection
of speaking games, which he divides into four classes, namely, board
games, card games, guessing games and, finally, quizzes and puzzles.
Miscellaneous Problem-Solving Activities

Pairs or small groups of learners collaborate to solve a problem. The
motivation derives from the cognitive challenge involved. Such activities
may involve, for example, rearranging picture sequences to make a story
(Harmer 2015: 397–398; Thornbury 2005: 81–92) or spotting the differ-
ences between two almost identical pictures. In this activity two learners
are seated opposite each other and are separated by a screen. One learner
has one picture and another learner the other. Without looking at each
other’s pictures they have to find the differences by description and
discussion. This is an information-gap activity because it involves one
learner supplying another with vital information (Ur 2012: 123–124;
Scrivener 2011: 217–218; Thornbury 2005: 80–81; Johnson 2018:
225–228). A variation is when one learner has a picture which he describes
and the other learner has to re-draw the picture as accurately as possible.
A further variation is where one learner has, for example, the plan of
a furnished room and the other learner has sticky shapes representing
items of furniture to stick on to a felt board. For advanced learners more
elaborate problem-solving activities involving, for example, deciding
who was the murderer in a ‘whodunit’ scenario are a possibility. For
an example, see Anderson (2004: 86–93).
Bonding Activities
Some activities can serve the purpose of social bonding within the class.
They include ‘ice-breaker’ activities for use with a new class. One such
activity is for pairs of learners to find out how many things they have in
common (Ur 2012: 126). Another well-known bonding activity is ‘Persuad-
ing someone to part with an object’. Learners work in pairs. One holds up
a valued personal object (car keys, house keys, a ring or a watch). Objects
chosen should be important to the learner for practical or sentimental
reasons rather than monetary value. One learner tries to persuade the other
learner (in a set time) to part with the object until tomorrow (but not for
money). The owner of the object presents arguments why he or she cannot
do this. Another ice-breaking activity involves pairs of learners standing
back to back and describing each other. The inaccuracies can be amusing.
The teacher or another learner can press for more details, for example
about the colour of the glasses the partner is not wearing. Yet another
activity involves learners working in pairs or small groups and telling about
‘three strange things that happened to me’. One of the three is fictitious
and the other learners have to spot it (see Thornbury 2005: 96). A further
useful activity is ‘Room 101ʹ (Anderson 2004: 44–45). In groups, learners
decide on three things they hate and wish to ban to Room 101. Each group
then has to justify their choices to the class, and finally, on the basis of
a class vote, a class choice of three objects for Room 101 is decided on.
Post Hoc Evaluation of Spoken Activities

Immediately after any spoken communicative activity it is useful to ask
oneself:
Did learners talk a lot?

Did they speak without being called on to speak?
Was participation among the class reasonably evenly distributed?
Was the language used comprehensible, appropriate and natural?
If the answers to these questions are affirmative, then the activity was
probably successful. By contrast, an unsuccessful session is often char-
acterised by the converse of these positive features, namely, learners do
not talk much and do not talk without being called on to do so.
Participation may be very half-hearted, with long periods of painful
silence or, alternatively, just a small number of learners may monopolise
the talk. The language used may be unnatural, stylistically inappropriate
and incoherent so that communication is severely inhibited. In addition,
learners may appear to be uninformed about or bored by the topic. They
may appear to be shy, anxious or unwilling to initiate discourse. Some
learners may become disruptive. Some learners may abandon the activity
altogether and talk among themselves in their mother tongue. The
teacher may have to keep intervening to keep learners on topic and
even to solve conflicts.
Food for Thought

Q. 1. Imagine you are going to tell the story Dick Whittington and his Cat
to a class of learners. There are different versions of the story, as
you will see from a brief internet survey, including picture-book
versions. The story could be told to children or adults at various
levels of proficiency. Specify a group of learners and prepare
a three-phase lesson. First, draft the story as you wish to tell it,
building in devices such as repetition to aid comprehension. Check
that the length of your story is suitable for the time you have
available. Divide the story into thematic sections, at the end of
which you can pause and check on comprehension. Decide on any
scaffolding devices you may wish to use, especially pictures, and
then plan the three phases of your lesson.
The story is based on a real historical character, as you will see
from your internet survey. If you are intending to teach the story
to older children or adults, consider how you might or might not
use this fact, either in the pre-listening phase or in the post-
listening phase or for subsequent project work.
If possible, conduct a mock lesson with your colleagues
acting as learners (micro-teaching), or a real lesson with appro-
priate learners, ideally with some colleagues present as obser-
vers. Do not read out your draft verbatim but use it as a basis
for your spoken rendering. If possible, in a post-teaching ses-
sion discuss with your colleagues the success of the lesson and
opportunities for further improvement.
Q. 2. (Group exercise) ‘The Hot-Air Balloon’, is a well-known example
of a discussion role-play using celebrity roles (see Thornbury 2005:
103; Harmer 2015: 390; Johnson 2018: 236).
The Hot-Air Balloon Role-Play

This role-play is best played in groups of six to seven people. Learners
are each given a large card with the name of a (different) celebrity, and,
in one case per group, a rather mundane profession or trade, written on
it in large letters. Each learner pins their card on their chest and assumes
this role. They sit in a tight circle, facing inwards and it is explained to
them that that they are in a hot-air balloon over the Andes and that due
to a defect, the balloon is fast losing air and that they will crash and all
be killed if one of their number does not agree to jump overboard. Only
the lightened balloon will make it over the peaks of the Andes. As
a group they must decide (majority decision) within the next so many
minutes (specified by the teacher) which of them is least valuable to
society and can therefore be jettisoned.
Consider how you would manage this role-play in the classroom.

Anticipate any problems that might arise and how you would deal
with these. For what sorts of learners would you regard it as
suitable and for what sorts of learners as less suitable? Can you
think of any variations on the procedure that could be introduced,
either to make it easier or more demanding according to the
learner group? Can you think of how the same idea could be
presented under another guise so that its shelf life could be
extended?
Q. 3. Try out the ‘Lost at Sea’ simulation with your colleagues and decide
on how you would use it in class, or whether you would not use it.
Q. 4. (Pair exercise) Play the ‘Persuading someone to part with an object’
activity with a colleague. After doing it, evaluate the activity. Do
you have any suggestions for refining and developing it?
Q. 5. Compare the features listed at the beginning of the ‘Speaking’ section
as characteristic of the impoverished classroom discourse found in
traditional, teacher-fronted classrooms (perhaps a ‘What did you do
at the weekend?’ teacher-led conversation class) and those mentioned
at the end of the chapter as characterising an unsuccessful commu-
nicative speaking activity (perhaps a problem-solving activity such as
a simulation). Which features occur in both discourse situations and
which only in one or the other situation?
Further Reading
Liontas, J. I. (editor-in-chief) (2018). The TESOL Encyclopaedia of English
Language Teaching. Volume 3. Teaching Listening (Renandya, W. A. And Hu,
G., Eds.), Teaching Speaking and Pronunciation (Christison, M. And Broady,
C., Eds.). Hoboken, NJ: Wiley-Blackwell.
Nation, I. S. P. and Newton, J. (2009). Teaching ESL/EFL Listening and Speaking.
New York: Routledge.
Powell, M. (2010). Dynamic Presentations. Cambridge: Cambridge University Press.
References
Anderson, A. and Lynch, T. (1988). Listening. Oxford: Oxford University Press.
Anderson, J. (2004). Teamwork: Interactive Tasks to Get Students Talking. Peaslake,
Surrey: Delta Publishing Ltd.
Anderson, J. (2006). Role Plays for Today. Peaslake, Surrey: Delta Publishing Ltd.
Anderson, J. (2014). Speaking Games. Peaslake, Surrey: Delta Publishing Ltd.
Clark, H. H. (1996). Using Language. Cambridge: Cambridge University Press.
Field, J. (2008). Listening in the Language Classroom. Cambridge: Cambridge Uni-
versity Press.
Firth, J. R. (1957). Personality and language in society. In Firth, J. R. (ed.), Papers in
Linguistics 1934–1951. London: Oxford University Press, pp. 177–189.
Hadfield, J. and Hadfield, C. (1999). Simple Listening Activities. Oxford: Oxford
University Press.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of
Language and Meaning. London: Arnold.
Harmer, J. (1991). The Practice of English Language Teaching. London: Longman.
Harmer, J. (2004). How to Teach Writing. Harlow: Pearson Education.
Pearson Education.
Johnson, D. (2005). It’s not what we expected! A case study of adult learner views
in ESL pedagogy. TESL Reporter 38/2: 1–13.
Second edition. Harlow: Pearson Education.
Ladousse, G. P. (1987). Role-Play. Oxford: Oxford University Press.
Maley, A. (1993). Short and Sweet: Short Texts and How to Use Them. Volume 1.
London: Penguin.
Malinowski, B. (1923). The problem of meaning in primitive languages. Supple-
ment to Ogden, C. K. and Richards, I. A., The Meaning of Meaning. London:
Kegan Paul.
McCallum, G. P. (1980). 101 Word Games. New York: Oxford University Press.
Shuttleworth, M. (2014). Classroom Games. Cologne, FRG: Anaconda.
Sinclair, J. and Coulthard, M. (1975). Towards an Analysis of Discourse. Oxford:
Sperber, D. and Wilson, D. (1995). Relevance: Communication and Cognition.
Second edition. Oxford: Blackwell.
Thornbury, S. (2005). How to Teach Speaking. Harlow: Pearson Education.
Wallace, C. (1992). Reading. Oxford: Oxford University Press.
9 Reading and Writing
The Written Mode

Producing and understanding speech is a natural human endowment, for
we are biologically driven to give expression to thought, to engage in the
exchange of ideas with other human beings and to make sense of the
world through language. The written mode, by contrast, is a secondary
symbolic system devised by human ingenuity. It renders the transient
symbolic sounds of speech into a graphic system of markings which can
be preserved on durable material such as wood, papyrus, parchment,
paper, or, latterly, stored electronically. In this way not only can the
limitations of time and space on oral communication be transcended, but
thoughts that were never expressed in speech can be recorded for poster-
ity in written form.
There are various writing systems used by the world’s languages.
A broad distinction may be drawn between alphabet-based systems and
character-based systems although there have been other systems, includ-
ing syllable-based systems, as used by the Phoenicians between 3,000 and
4,000 years ago (Yule 1996: 12). In alphabet-based systems, sounds are
represented by letters. In character-based systems each character repre-
sents, roughly speaking, a word. English uses the Latin alphabet, which is
the most widely used alphabet system in the world, the second most
widely used being the Arabic alphabet. Most European languages also
use the Latin alphabet, but not all: Russian and Bulgarian, for example,
use the Cyrillic alphabet and Greek uses the Hellenic alphabet. Outside
Europe, some of the world’s major languages, including Chinese, use
a character-based system of writing, not an alphabet-based system.
Japanese combines a character-based system and a syllable-based system
of writing.
Learners who can already read and write in a language which uses the
Latin alphabet have a head start in learning to read and write in English
compared to learners who have yet to learn the Latin alphabet. The
greatest challenge of all is faced by illiterate or only partly literate
learners. However, all learners who want to be able to read and write in
Reading and Writing 237
English, even those who already know the Latin alphabet, will have to
learn the system of English spelling, since each language has its own set
of significant sounds (phonemes) and its own system of matching letters
to sounds. In some languages there is a more systematic set of corre-
spondences between sounds and letters than in others, which means
some languages are easier to read out loud and easier to spell than
others. In this respect Italian is easier than French or English.
English spelling is difficult for a number of reasons. The first problem
is that there are 44 GB phonemes but only 26 letters of the alphabet.
Apart from this, the correspondence between letters and sounds is not
completely systematic. This is particularly noticeable for the vowel
sounds. Spelling conventions became established following the introduc-
tion of printing in the fifteenth century (Yule 1996: 14) but for a long
time in handwritten documents spelling conventions were quite fluid.
However, printing and universal education won out and standards were
imposed. However, since pronunciation goes on changing and evolving
naturally, the discrepancies between sounds and letters have become ever
greater over the centuries. The situation was worsened by the activities
of spelling reformers in the sixteenth century who introduced ‘etymolo-
gical’ spelling. For example, they changed the spelling of the word ‘dette’
to ‘debt’, thus causing spelling problems for millions of schoolchildren
for centuries thereafter, and with no end in sight. They did this for
etymological reasons, correctly tracing the word back to the Latin
debitum. However, what they did not know was that the word had not
entered English directly from Latin, but, following the Norman Con-
quest, from French. To this day the French word for ‘debt’ is dette. All
this makes reading and writing difficult, so that we all make spelling
mistakes when we write in English and mispronounce words we have
previously encountered only in written form. A good rule of thumb for
learners is that the pronunciation of a word is never a foolproof guide to
its spelling, and vice versa.
In this chapter it will be assumed that learners already know the Latin
alphabet, either because it is used by their native language or another
language they are literate in or because they have learned it for English at
an earlier stage of their education. Thus, we will not be concerned with
how best to teach the system of correspondences between the letters of
the Latin alphabet and English sounds or words. Harmer (2004: 44–48;
2015: 361–362) and Ur (2012: 153–157, 163–165) offer suggestions for
teaching handwriting and spelling. Collins et al. (2019: 98–110) offer a set
of spelling guidelines.
Writing is not just speech written down. Writing is generally denser
and less redundant than speech (Ur 2012: 150; Johnson 2018: 255). The
written mode frees writers and readers from the time and processing
pressures under which speakers and listeners have to operate. Writers
have more time to plan and revise their message than speakers. Readers
can choose their own processing pace, can go back, re-read and can skip
ahead to see what is coming. For these reasons, the written mode
facilitates the expression of more complex and differentiated ideas than
speech. It also makes it easier for writers to fine-tune their language and
use rhetorical effects. For the reader, the written mode encourages
reflection and critical thinking as well as appreciation and aesthetic
enjoyment of the written product. However, writing also makes it
possible to quickly scribble a note for the milkman and to chat online
in a written exchange which is rapid and in many ways like the sponta-
neity of conversation (Ur 2012: 151–152). Not only does each written
genre, like each spoken genre, have its own register, but there may be
differences between languages, so that learners cannot necessarily carry
over the conventions of a given genre from their native language into
English (Hudson 2007: 201–226).
Reading
Reading Purpose and Reading Technique

For teaching purposes a distinction is usually drawn between intensive
and extensive reading. Intensive reading involves short prose texts or text
excerpts read in class under teacher guidance. Extensive reading involves
longer texts, often whole books, read outside class. Consideration of
reading in this chapter will be limited to intensive reading of the sort
usually done in the reading comprehension lesson. Extensive reading of
fiction will be dealt with in Chapter 10, as will the intensive reading of
short poems in the classroom.
The aim of intensive reading sessions in the EFL classroom is to help
learners to become good readers. This means, firstly, enabling them to
read effectively a wide variety of authentic texts for various purposes
relevant to their real-life needs and, secondly, helping them to experience
reading in English as a pleasurable activity. Nuttall (1996: 31) describes
reading effectively as reading ‘at appropriate speed’ and ‘with adequate
understanding’. Appropriate reading speed does not depend only on
how difficult the text is, but also on how much readers want to get out
of a text. Learners have to know this and adjust their reading speed (and
technique) accordingly (Wallace 1992: 5). One can read any text too
carefully or too carelessly. Readers also need to be able to identify the
parts of a text which are important for their reading purpose so as to
read these parts carefully and to read in a more cursory fashion other
parts which are less central to their reading purpose.
All learners will need to read extensively outside class on their own
initiative if they are ever to become highly effective readers. This is
because practice seems to be of great importance in becoming an effective
reader. There is a link between reading effectively and enjoying reading.
Ineffective reading techniques mean that reading is not enjoyable, so
reading is avoided, which in turn means that reading skills do not
improve. By contrast, effective readers enjoy reading and so read more,
which improves their skills so that they enjoy reading even more. Nuttall
(1996: 127) refers in this regard to the ‘vicious circle’ of the weak reader
and the ‘virtuous circle’ of the good reader.
If learners experience reading comprehension lessons as painful rather
than pleasurable, they will never read extensively in English of their own
accord because intrinsic motivation will be lacking. This must be pre-
vented at all costs. Texts chosen should be carefully selected to engage
learners’ interest and the reading tasks performed should be challenging
but achievable. At the end of the lesson learners should feel they have
performed the tasks successfully under the guidance of the teacher. How-
ever, although the overt focus of the lesson is on comprehension of the
text at the desired level, the hidden agenda of the lesson is that learners
should be developing reading skills which they can transfer to other texts
they will encounter in the future, without a teacher to guide them.
Ineffective readers are inflexible readers. They tend to read all texts
word for word at the same slow laborious pace (Davies 1995: 138). They
may have no alternative because otherwise they will not understand
anything at all. For this reason it is counterproductive just to tell slow
readers to speed up. They will then merely rush through the passage and
become frustrated because they do not understand. Effective readers, by
contrast, are flexible readers. They can adjust their reading technique and
thereby their reading speed according to what they are reading and why
they are reading it (Davies 1995: 132–140; Wallace 1992: 6–7).
Some reading purposes/techniques are listed below. The list can easily
be added to.
Scanning for specific information

This involves going through a text to see if a particular item of
information is there, for example a name (Nuttall 1996: 48–49;
Davies 1995: 137, 140; Scrivener 2011: 265).
Skimming for gist
This involves going through a text to get a rough idea of what it is
about (Nuttall 1996: 49–50; Davies 1995: 137, 140; Scrivener
2011: 265).
Reading for information content
This is the neutral or default reading technique/purpose, which
Grabe and Stoller (2011: 8) call ‘reading for general comprehension’.
Critical or reflective reading for learning or appreciation
This involves pausing for reflection or memorisation and going back
and forth in the text (Davies 1995: 137, 140).
Empathetic or receptive reading (reading for pleasure)
According to Davies (1995: 140), this involves ‘a smooth, typically
unconscious reading of undemanding, familiar or narrative text, for
pleasure and “unconscious” language learning’.
Form-focussed reading
This involves reading a text for language learning and/or for formal
or linguistic analysis.
Word-for-word reading
This is the slow, laborious technique used by ineffective readers. Davies
(1995: 138–13) calls this ‘practice read’ and suggests it is a reading
technique in its own right, which even proficient readers fall back on to
cope effectively with difficult bits of texts. In word-for-word reading
learners may say the words under their breath (sub-vocalisation) or even
whisper them or say them out loud, and in some cases this can be helpful.
Particular techniques of reading are associated with specific text-types but

most text-types can be read in various ways (Davies 1995: 132). Thus, in
the following (invented) example, a reader might first scan newspaper
headlines and, if a particular headline, for example, ‘British men stingiest
in Europe’, strikes that reader as being of interest, may skim the article to
see roughly what it is about (for example, the amount of money British
men spend on birthday presents for their wives). The reader might read
the article for information content (how much do they spend and how much
do men in other countries spend?) or critically (how trustworthy are the
data for the contention?) or empathetically (for gleeful enjoyment, if the
reader is, for instance, divorced from a stingy British man). In a university
seminar, a newspaper article might even be read in form-focussed fashion,
that is, to analyse its language. An ineffective reader could not use any of
the techniques so far mentioned but would start with the first word of the
headlines and proceed word by word, trying to understand each word.
Not mentioned so far is reading out loud in class. Learners should not
be asked to read out loud a passage they have not yet had a chance to
prepare, since reading aloud to an audience is a daunting task, even for
advanced learners. And they should never ever be expected to read out
loud while seated in their place in class but should face their audience, or
the class should sit in a horseshoe formation. Inexpert readers find it
very difficult to read out loud to others and simultaneously understand
what they are reading. This also makes it difficult for the rest of the class
to understand. Because of cognitive overload, the reader’s speech is
likely to be unnaturally slow and halting; words will be mispronounced
and inappropriate stress, rhythm and intonation patterns will be used.
The result is that comprehension may be difficult even for highly
proficient listeners. Even a professional reader needs preparation and
rehearsal time. It can, however, be helpful for learners to listen to the
teacher reading or to hear a recording of another expert reader. This is
because the expert reader commands not only the pronunciation of
individual words, but, more importantly, can impose appropriate
intonation, rhythm and stress patterns on the text. For helpful discus-
sions of reading out loud, see Ur (2012: 136), Harmer (2015: 318–319),
Davies (1995: 138, 140), Nuttall (1996: 201–203) and Scrivener
(2011: 269).
A Three-Phase Lesson Structure

As with listening texts, reading passages should be chosen so as to fit in
with the learners’ interests and likely future language needs and they
should also be well-suited to follow-up discussion or writing activities.
Before deciding to use a text for reading, it is therefore worthwhile
asking oneself the following questions:
Can the learners relate the text to their own culture?

Can the learners write a prologue or epilogue to the text?
Can the learners respond to and evaluate the text from their own
point of view?
Can the learners contribute additional information of their own?
It is advisable to adopt a three-phase lesson structure comprising a pre-

reading phase, a while-reading phase and a post-reading phase (Nuttall
1996: 149–169). This procedure parallels that for listening
comprehension.
Pre-Reading Phase
The pre-reading phase aims:
to introduce and arouse interest in the topic

to motivate learners by awakening curiosity and giving them a reason
to read the passage
to provide some limited language preparation
Questions that the teacher might consider before going in to the

class are:
What do the learners already know about the topic?

How can I best activate this knowledge?
What aspects of the passage will interest the learners?
How can I trigger this interest?
Which relevant background cultural knowledge may the learners
lack?
How can I best supply this, perhaps with visual aids and realia?
Which absolutely essential key vocabulary should be cleared up in
the pre-reading phase?
A good way to start the lesson is to explore with learners the associations
that the text’s title has for them (Johnson 2018: 249–250; Nuttall 1996:
45, 155–156). This can be done as brainstorming. Learners should be
asked to infer from the title what they can about the topic of the text
they are about to read. Learners should never be given a text without
a title. If the text has no title or the title is inappropriate, for example if
the title of an original text is unsuitable for a short reading passage taken
from it, a suitable alternative title should be supplied. It is important that
the title is well-chosen since a well-chosen title helps readers to identify
the likely topic of the text. On the other hand, an inappropriate or
misleading title can be a great hindrance and lead readers up the garden
path (Davies 1995: 68). Paying attention to titles should become part of
learners’ strategic approach to texts they will read by themselves in
future. Without training they will not do this. Many learners never
read titles, headings and subheadings.
Having identified the likely topic, for example, a bus crash, a funeral or
a wedding, learners should then be encouraged to consider what back-
ground knowledge they already possess about it. Writers assume that their
readers share with them some background knowledge which does not have
to be explicitly stated. Relevant background knowledge is seen as being
encoded in the mind in the form of schemata (Bartlett 1932) based on the
individual’s past experience, including experience gained vicariously
through the media and by reading. All texts are incomplete in the sense
that readers always have to contribute background knowledge of their own
about the topic, the genre and perhaps even about the writer to understand
any text they read. On the basis of this background knowledge, readers then
make presuppositions about the content of the text before they even start
reading. These presuppositions may then be confirmed or disconfirmed as
the text develops and the updating process will continue while any particu-
lar text is being read. However, EFL learners may lack the requisite back-
ground knowledge or may have misconceptions if there are intercultural
differences involved (Johnson 2018: 247–248). It is therefore important to
identify any gaps in knowledge or wrong ideas learners may have. However,
only background information which is essential for understanding the gist
should be supplied, and only after attempts have first been made to elicit the
information from the class.
Concerning vocabulary, the only words which should be explained in
advance are those whose meaning is essential for understanding the gist of
the text and cannot be inferred. If this group of words is too large (more than
seven), this is an indication that the text is too difficult for the learners. Part of
becoming a good reader is learning how to infer word meaning from context
(top-down processing) and/or from the word’s morphological structure
(bottom-up processing). Every time we tell learners what a word means, we
are robbing them of a chance to develop this necessary procedural knowledge
(Johnson 2018: 241–248; Nuttall 1996: 16–18; Davies 1995: 57–62). The aim
of the reading lesson is not to teach vocabulary but to help learners develop
the necessary procedural knowledge (skills) to read a text effectively for
a specific purpose. As with background knowledge, so with vocabulary – it
is preferable in the pre-reading phase to elicit word-meanings from the
learners themselves if at all possible.
While-Reading Phase
The while-reading phase aims to give learners practice in employing
various reading styles and to help them understand:
the gist of the passage

the more detailed meaning of the passage at certain key points
something about the genre of which the text is an example
the structure of the passage
the writer’s purpose
In preparing for this phase, teachers might ask themselves:
Does the passage constitute a complete text, an abridged text or

a text excerpt?
What is the core meaning (gist)?
What is implied rather than explicitly stated?
To what genre does the passage belong?
What can be inferred from the passage about the writer’s stance?
What structural divisions, such as thesis and antithesis or introduc-
tion, development and conclusion, can be identified?
How will I divide the passage into sections for reading?
Which guiding questions (see below) can I draft for each section?
Which unknown language items occur for which learners can infer
a rough meaning?
Which unknown language items are not essential for global compre-
hension and are best ignored?
Which reading techniques can learners practise?
How long should a text be? The answer depends on the reading experi-
ence and proficiency of the learners. Ideally, the text should be short
enough for learners to read it through as a whole to get its gist initially.
Short passages (up to one page) can often be read as a whole a number of
times, each time for a different purpose. Thus, for the first reading
learners might be instructed to find out roughly what the test is about
(reading for gist or global understanding). This requires the technique of
skimming, which involves focussing on key words and skipping over less
essential words. A time limit which is sufficient to get the gist only, if
learners skim effectively, should be set.
Subsequent readings may focus on specific detail, and stylistic and
critical appraisal should come at later readings again (see also Scrivener
2011: 267). These later readings can often be best managed by reading the
passage in sections. Passages longer than a page and up to two or three
pages for more advanced learners will usually be read in sections of
perhaps one, two or three paragraphs after an initial skimming of the
whole passage. Before each section the teacher will provide the learners
with guiding questions (one to three) which they should bear in mind
while reading. The purpose of these questions is to direct learners’
attention to salient aspects of the passage. It is possible to have groups
of readers focussing on different guiding questions and pooling their
findings afterwards (see Nuttall 1996: 160–161). In general, questions
should not be sprung on learners as a surprise after they have read
a section. Rather, learners should know the questions before they read
the section so that they can listen for a purpose.
Questions may be arranged on a scale of ascending challenge (Nuttall
1996: 181–191). Just how this can be done will now be shown for
a passage for advanced learners from a novel by Donald E. Westlake.
This passage was used by Swan (1976: 43) as a passage for learners to
summarise. Swan supplied the title, ‘Just leave the keys in, sir’:
Just Leave the Keys In, Sir
Stan Murch, in a uniform-like blue jacket, stood on the sidewalk in front of the
Hilton and watched cab after cab make the loop in to the main entrance.
Doesn’t anybody travel in their own car any more? Then at last a Chrysler
Imperial with Michigan plates came hesitantly up Sixth Avenue, made the left
hand loop into the Hilton driveway and stopped at the entrance. As a woman
and several children got out of the doors on the right of the car, toward the
hotel entrance, the driver climbed heavily out on the left. He was a big man
with a cigar and a camel’s hair coat.
Murch was at the door before it was halfway open, pulling it the rest of the
way and saying, “Just leave the keys in it, sir.”
“Right,” the man said around his cigar. He got out and sort of shook himself
inside the coat. Then as Murch was about to get behind the wheel, the driver
said, “Wait.”
Murch looked at him. “Sir?”
“Here you go, boy,” the man said and pulled a folded dollar bill from his pants
pocket and handed it across.
“Thank you, sir,” Murch said. He saluted with the hand holding the dollar,
climbed behind the wheel, and drove away. He was smiling as he made the right
into 53rd Street; it wasn’t every day a man gave you a tip for stealing his car.
(From Bank Shot by Donald E. Westlake)
The least challenging questions are those where the answer consists of
explicit information to be found in the passage, for example:
Q.1. Where was Stan Murch standing?

A.1. On the sidewalk in front of the Hilton.
The learner can recover this verbatim from the passage. Rather more
challenging are questions where some minimal interpretation is required:
Q.2. Was the Chrysler being driven fast up Sixth Avenue?

A.2. No, it was being driven slowly.
In this case one has to understand that ‘hesitantly’ means ‘slowly’ in

context. The next question:
Q.3. Why did the man allow Stan Murch to drive his car away?
A.3. Because he thought he was going to park the car for him. (possible
acceptable answer)
is more challenging again. It requires inferencing or reading between the

lines because the answer is not stated explicitly in the passage. Even more
evaluation is required by a question such as:
Q.4. Why do you think Stan Murch looks at the man and says, ‘Sir?’ when
the man says ‘Wait’?
A.4. Maybe Stan thinks the man is suspicious and is going to ask him to
show some identification. (possible acceptable answer)
A still more demanding question requiring an evaluative personal

response would be:
Q.5. What are your feelings for the man whose car Stan Murch drives
away?
A 5. I feel sorry for him but I also think he was very naïve and must take
some blame himself for what happened. (possible acceptable answer)
The most demanding of all tend to be questions which require critical

reading and appraisal of authorial techniques, such as:
Q. 6. Can you identify some clues in the first part of the passage which
at second reading suggest Stan may not be employed by the Hilton to park
guests’ cars?
A. 6. On re-reading the text, I notice that Stan is not actually wearing
a uniform but a ‘uniform-like blue jacket’ and that he is not standing on
the hotel forecourt but ‘on the sidewalk’. (possible acceptable answer)
Questions should be varied and progress made from less to more

cognitively challenging. One should guard against asking questions
which focus on unimportant factual detail since they implicitly train
learners to read in an ineffective way and the lesson can easily degenerate
into a ‘numbers, names and colours’ quiz. Also to be avoided are
questions which focus on linguistic niceties not essential for understand-
ing and appreciation of the text. Questions may be given in written or
spoken form and may be answered either orally or in writing. It is often
good to build both written and oral questions and answers into the
lesson.
An appropriate balance of closed and open-ended questions should
also be struck. By a closed question is meant a question to which there is
only one right answer (as in questions 1, 2 and perhaps 3, above). For
open-ended questions (as in questions 4–6, above), by contrast, there is
a range of possible answers (Lightbown and Spada 2013: 147). Classic
closed questions take the form of ‘yes or no’ questions, ‘true or false’
questions, ‘either/or’ questions and multiple-choice questions. ‘Ques-
tion-word’ questions can in principle be open or closed, but ‘who’,
‘what’, ‘which’, ‘when’ and ‘where’ questions are more likely to be
closed, whereas ‘how’ and especially ‘why’ questions are more likely to
be open. Relatively easy closed questions (such as questions 1 and 2,
above) are often good to start with because they encourage learners to
participate. Open-ended questions should generally come later and are
valuable because they require learners to formulate longer answers
involving language planning at the sentence level and beyond.
The distinction between closed and open questions is not the same as
the distinction already touched on in Chapter 8 between display ques-
tions (where the teacher already knows the answer and wants to see if the
learner also knows it) and genuine questions (where the teacher genuinely
seeks either information or an opinion from a learner). This distinction
depends not on the form of the question but on the reason for asking the
question on the teacher’s part. Some teachers might have a relatively
open mind on questions 3, 4, 5 and 6, above, for instance, and accept
novel ideas from learners, whereas others might not. Most reading
comprehension questions are of course display questions designed to
help learners understand the text and learners will appreciate them for
this reason (Lightbown and Spada 2013: 146). Nevertheless, learners will
also appreciate it if they are sometimes asked genuine questions as well
(Nuttall 1996: 181–191). At the very best this may give rise to discussion
among the learners.
Post-Reading Phase
The post-reading phase, which may be the final phase of the lesson or
may extend over a follow-up session or include homework, aims:
to consolidate and build on what has been read

to give learners the chance to integrate the text into their own world-
knowledge so that it becomes part of their ‘mental encyclopaedia’
(Clark 1996)
In preparing for this phase it is advisable for teachers to ask themselves:
Is the passage open-ended (or one-sided)?

Does the passage invite resolution or complementation?
Does the passage invite a comparison with the life experience of the
learners?
Does the passage invite a rejoinder, a recommendation or a decision?
Post-reading tasks may be written or spoken or may combine speech and

writing. Conventional written post-reading tasks are summaries and
paraphrases, either done individually or in groups. Other possibilities
include writing a prologue or sequel to the passage, rewriting the passage
from a different perspective and rewriting the passage adopting a different
register or style. Prologues or sequels can be combined with a change of
perspective and/or with a change of style or register (Harmer 2004:
25–27). The ‘Just leave the keys in, sir’ passage, for instance, invites
a sequel in the form of the written statement the man from Michigan
might make to the police after reporting the theft (change of perspective
and register). Another possibility would be a prologue or sequel written
from the original authorial perspective (with no change of style or
register). Spoken post-reading tasks do not have to be limited to evalua-
tive discussion but can involve role-play. In ‘Just leave the keys in, sir’,
various sorts of role-play suggest themselves, including the reporting of
the theft to the police, an encounter between Murch driving the car and
a suspicious traffic policeman and Murch bragging to his friends the
following evening.
For advanced learners, a post-reading activity can lead into project work
supported by internet research. Following a post-reading discussion of
a text on the Spanish Armada, for example, project work could compare
the different accounts to be found on the internet and decide why the
Spanish plans failed. Another possible follow-up would be a formal debate
in which one side could support the view that the main problem was the
bad weather and (for the Spanish) uncharted waters, and the other side
could support the view that it was Drake’s naval strategy and his more
manoeuvrable ships which thwarted the Spanish plans.
Alternatives to Traditional Reading Comprehension Classes
Reading can be combined with writing skills from the very start. One
technique is that of text completion. This involves giving learners
a narrative text which breaks off at a ‘cliff-hanger’ point. For example,
‘Just leave the keys in, sir’ could be presented only up to ‘… and drove
away.’ Learners could then be asked to work in groups and bring the
story to a close using no more than perhaps 50 words. After doing this,
each group could present or display its version to the whole class for
evaluation and the various versions could be compared with the original.
In effect this is training in text-prediction skills (Nuttall 1996:12–16;
118–121). Very short stories with a twist in the tail are well-suited to
such cliff-hanger techniques, for example ‘The Appointment in Samarra’
as retold by William Somerset Maugham in his play Sheppey (Maugham
1933), which is also available online (www.k-state.edu/english/baker/eng
lish320/Maugham-AS.htm).
Related to the cliff-hanger approach are techniques of jigsaw reading,
which are based on the information-gap principle. In this approach the
class is divided into groups and each group gets a section of a passage to
read and then the passage has to be reconstructed and put together by
the class (Johnson 2018: 253; Harmer 2015: 331–334). Another problem-
solving approach is the sentence-reordering task, in which learners are
first given the sentences of the text in jumbled order and have to put
them in the correct order (Harmer 2015: 328–330). Another possibility is
providing learners with a passage which lacks an introductory, middle or
concluding paragraph and asking them to supply the most suitable one
from a selection. Alternatively, they can be asked to compose the
missing paragraph themselves. Afterwards they can then compare their
paragraph with the original one. Another variation is to insert an alien
paragraph or sentence into the text for learners to identify (Grabe and
Stoller 2011: 145).
Other possibilities include using matching tasks in which pictures are
matched to the text they best correspond to or vice versa. Picture-
matching is well-suited to descriptive texts, for example of houses,
rooms, towns and landscapes, but is not limited to these. Texts can also
be matched with other texts; headlines can be matched with newspaper
articles, and headings; introductions and conclusions can be matched to
texts. In a final phase, each learner group justifies its choice.
Yet another task-type is text expansion or reduction according to
certain criteria, for example adding adjectives or relative clauses, or
giving reasons, or providing flashbacks (Maley 1993, 1995). Other manip-
ulative operations include transferring a text into another genre as well as
restoring mutilated texts and completing incomplete texts (Harmer 2015:
323–325). All these activities are well-suited to pair work or small-group
work.
For a variety of reading-based activities suitable for young and low-
proficiency learners, see Hadfield and Hadfield (2000a). These include, as
well as matching tasks and sentence-reordering tasks, information-
transfer tasks where learners read a text and then complete a table or
chart. Other tasks the authors present are where learners correct factual
mistakes in a text, or draw an object described in a text, or guess what
object is being described in a text.
Writing
Written Genres
For some writing classes an examination syllabus will define the genres
learners are to master. In ESP (English for Specific Purposes) and EAP
(English for Academic Purposes) classes it will be possible by means of
needs analysis to identify the purposes for which learners will need to
write. For many other learners, especially those in TENOR (Teaching
English for No Obvious Reason) classes, however, it will often be
difficult to predict which genres learners will be required to write in and
which registers they need to command. Scrivener (2011: 238) suggests
a number of real-world genres that learners might practise. These include
letters and e-mails, newsletters, blogs, contributions to a class magazine,
advertisements for events, contributions to discussion forums on web-
sites, questionnaires, written applications of various sorts and project
work. Ur (2012: 157–159) suggests stories, personal anecdotes, instruc-
tions, route directions, descriptions of scenes, places and persons,
descriptions of a process – for example an experiment – summaries and
critical reviews of books or films, recommendations, advertisements,
leaflets, newspaper reports and short research papers. She also specifies
various sub-genres of letters/e-mails, including job applications, com-
plaints and replies.
Four Major Approaches to Teaching Written Genres

There are four major approaches to teaching written genres:
the product approach

the process approach
the genre approach
the process-genre approach
The Product Approach

This is the traditional approach and comprises four stages:
familiarisation of learners with the specific features of the genre
usually by means of an example text (see Scrivener 2011: 241–242)
controlled writing, in which the learners practise the skill at the
micro-level, first within strict guidelines and with the provision of
scaffolding and then more freely as support is progressively
withdrawn
guided writing, in which learners produce a complete text guided by
the provision of a skeleton structure, including perhaps some key-
words or phrases, or even topic sentences
free writing, in which students produce their own text without
guidance (in groups or individually)
The product approach is well-suited to genres which are highly rule-

governed at the formal level, for example business letters. It is also well-
suited to learners at lower levels of general proficiency who may lack
some necessary language at the level of declarative knowledge; it is also
suitable for learners who do not have any experience of the genre
(Harmer 2004: 55–59, 63). Additionally, the product approach is helpful
if the teacher wants to focus on specific vocabulary, syntax and linking
devices. It aims to supply learners with the necessary linguistic knowl-
edge and linguistic items for the task at hand. The approach in its
strictest form might be described as ‘assisted imitation’. The emphasis is
on breaking down the productive process into stages so that a good
linguistic product will finally be produced by the learners in autonomous
fashion. Although the product approach tends to be associated with
individual writing, collaborative writing is also possible (Harmer
2015: 367).
Hadfield and Hadfield (2000b) offer many examples of guided writing
tasks suitable for young learners and learners of low proficiency. These
include blank-filling exercises, linking sentences by using an appropriate
co-ordinating conjunction (‘and’ versus ‘but’) and reordering tasks, invol-
ving either reordering jumbled words within the sentence, or reordering
jumbled sentences within a paragraph. Sometimes learners are given
skeleton sentence structures which they can vary by replacing the origi-
nal words with appropriate substitutes at various points (substitution
frames). Other activities involve giving learners notes to expand into
a text. Another approach the authors use is providing ‘split halves’
sentences, where the first halves of a set of sentences are presented in
a list and the second halves in another list, but in jumbled order.
Learners have to match up the sentence halves correctly. The texts or
sentences produced then serve as a basis for learners to produce their
own texts (free writing).
One objection sometimes made to the product approach is that
written teacher feedback tends to focus on error correction and is often
received some days after the writing task was done, so that the chance to
learn may have been missed. Even if learners are able to make sense of all
those arrows, letters and numbers as well as straight and wiggly lines – all
in aggressive red ink – which they see when their work is returned to
them, it may be difficult for them to remember what they had originally
wanted to express anyway. Scrivener (2011: 244) suggests that teacher
feedback should focus not just on error correction but on appropriate-
ness in terms of style and register. Scrivener (2011: 245–248), Ur (2012:
161–163) and Pawlak (2014: 143–148) discuss a number of ways of
providing feedback on written work.
The Process Approach

The process approach is quite different. It places emphasis on helping
learners to develop skills such as planning, drafting and editing a text
rather than on supplying learners with a model to imitate. Drawing on
the distinction between declarative knowledge (‘knowing that’) and pro-
cedural knowledge (‘knowing how to’), the approach assumes that each
text is unique and poses its own challenges and problems. For this
reason, it is held that model texts written for one situation will never
work perfectly for another. The underlying premise is that the develop-
ment of writing ability is an unconscious process, which can only take
place when the learners are engaged in the writing activity itself (proce-
duralisation of knowledge).
Typically, four stages are involved, although these are not sacrosanct. The
first phase may be done with the whole class, the subsequent three phases are
usually done in small groups of four to six persons (collaborative writing):
a pre-writing phase, often in the form of brainstorming (see Scrivener

2011: 239)
a drafting phase, in which a rough version of the complete text is
produced
a revising stage, in which large-scale additions, deletions and reorder-
ings at the macro-level may be made
an editing phase, in which the micro-level of language is scrutinised
and modified
In this approach the teacher’s role is essentially that of facilitator,

motivator, resource and feedback provider (see Harmer 2015: 369). He
or she moves among the groups offering help and advice, but the
processes of drafting, revising and editing are supposed to be carried
out by the learner groups themselves. Sometimes groups may be shuffled
or rearranged between phases, or one group may send a messenger to
another to make for an interchange of ideas. The messenger may then
return to the original group to provide feedback from the group just
visited.
Each group should aim at producing a final group version. This can be
produced by one member acting as scribe, but it is probably better for
the final group version to be written up by each learner separately so that
they are all actively engaged (Ur 2012: 160). The process approach does
not insist on a finished product and holds that ongoing feedback (from
the teacher and peers) during drafting is more likely to result in learning
than post hoc written teacher corrections. The rationale behind this is
that the provision of ongoing feedback operates within the zone of
proximal development (ZPD) to tackle an immediate problem and thus
offers affordances for learning (see Chapter 1 on the ZPD). Ur (2012: 162)
notes, however, that learners usually prefer teacher feedback to peer
feedback. On the other hand, Scrivener (2011: 244) suggests peer feed-
back may be preferable to traditional written teacher feedback since it is
more likely to focus on appropriateness than error.
The process approach is often more suitable than the product
approach for more advanced learners. It also lends itself to genres
which do not have rigid conventions and permit of considerable stylistic
variation.
The Genre Approach

A third approach is the genre approach. This approach follows the
product approach in placing emphasis on imparting information to
learners, since for any given genre there may be specific conventions
concerning layout, organisation and sequencing of information, style and
level of formality (see Scrivener 2011: 241–242). Furthermore, each
genre, for example application letters, reports, essays and reviews, con-
sists of various sub-genres, each of which will have a more specific
register than that of the superordinate genre, for example job-application
letters, research reports, descriptive essays and film reviews (see Harmer
2004: 16–22). Learners may therefore require special language prepara-
tion before writing in an unaccustomed genre or subgenre. This prepara-
tion may include lexis, for example legal or medical vocabulary; syntax,
for example use of the passive in reporting experiments in the natural
sciences; and pragmatics, for example not explicitly stating negative
points about a person but implying them by what is left unsaid in
personal references (Grice 1975).
In other words, genres cannot be lifted out of the social context in
which a particular written text is produced: especially the purpose of the
writer, the specifics of the subject matter and the writer’s relationship
with the intended audience. These variables make for the uniqueness of
each text. Thus learners are not given a de-contextualised business letter,
for instance, as a model for their own production, but rather a particular
business letter written in a specified situation to a particular addressee
about a subject which is clearly specified too. Additional relevant
information may be provided about the relationship between the corre-
spondents. Having studied the model letter, learners may produce
a reply to it or to a similar letter in a different but comparable context
which is specified.
There are three stages involved:
modelling the target genre, usually by an example with focus on

possible permissible variations (teacher-fronted)
the joint construction of a text by learners and teacher (with the
whole class)
the independent construction of a text by the learners (individually
or in groups)
It is the introduction of the social element which distinguishes the genre

approach from the product approach. Harmer (2015: 378–379) shows
how a genre approach can be used for writing a guidebook entry.
Scrivener (2011: 243) emphasises the importance of clearly defining in
advance the intended audience for, and purpose of, any piece of writing,
not just letters, as this will influence how the piece is to be written.
Otherwise learners are just ‘writing for teacher’.
The Process-Genre Approach

The process-genre approach represents a synthesis of the genre and
process approaches (see Harmer 2004: 86–87). It holds that writing
involves specific language knowledge, so that appropriate language input
which learners internalise will help to improve writing ability (as in the
product approach). Also required, however, are knowledge of context
and purpose (as in the genre approach). Further, it recognises that writing
is a skill for which practice is required to develop procedural knowledge
unconsciously (as in the process approach). The stages are:
setting up the context for writing by means of authentic texts,

pictures, brainstorming and teacher provision of situation-specific
language
a drafting phase, in which a rough version of the complete text is
produced
a revising stage, in which large-scale additions, deletions and reorder-
ings at the macro-level may be made
In the process-genre approach even more background information is

provided than in the genre approach. A scenario is presented which
provides the entire social context within which the piece of writing has
to be written. This is often built up by providing authentic texts or
pictures and may involve an element of role-play or simulation. Learners
are informed about what has been the stimulus to write (past history),
what the writer’s intention is, what can be maximally achieved by writing
and what unwished-for outcomes are to be avoided. Often this would be
far too much for one single writing lesson alone. Thus, the approach
lends itself well to an integrated-skills classroom, where the writing task
could follow on from previous theme-based work. The whole session
can constitute a role-play within which the writing is produced and
feedback on the written product can then be provided by the learners
themselves within the role-play. Such peer feedback is likely to be more
successful than it would be outside the role-play framework since
learners may otherwise not wish to criticise their peers’ work nor receive
criticism from them (Ur 2012: 162).
Role-play usually involves learners writing within a clearly specified
role to a clearly specified (but fictitious) addressee (Scrivener 2011: 244;
Harmer 2015: 376–378). However, by using authentic texts and authentic
writers as models, role-play can be brought closer to real-world writing.
Two suitable authentic genres available in newspapers and online are
‘lonely hearts’ advertisements and ‘agony aunt’ columns. These two
genres provide practice in writing within the superordinate genres of
‘written personal self-presentations’ and ‘asking for and giving written
personal advice’, respectively. The following two teaching ideas are
suitable for adolescent and adult intermediate-level learners. Learners
unfamiliar with these genres in English would need to study authentic
examples to learn the conventions first. The teacher circulates while
learners are writing and offers advice and support.
The first example uses lonely hearts advertisements to elicit written
self-presentations. The class is divided into a number of small groups
and each group is given a different role card, which has been prepared in
advance by the teacher. This provides biographical data of a fictitious
person and his or her wishes concerning a future partner. In the first
phase each group studies their role card. In the second phase the same
authentic selection of lonely hearts advertisements from newspapers or
from the internet is given to each group, and each group has to decide
which advertisement to reply to. In the third phase each group writes
a letter of introduction to their chosen lonely heart. The role cards
should be so drafted that there is no ideal match with a specific lonely
heart advertiser, but some partial matches. In a final phase the letters of
introduction can then be presented or displayed to the whole class and
evaluated both in terms of appropriateness of prospective partner chosen
and language appropriateness.
The second example uses agony aunt columns to elicit letters asking
for and giving personal advice. The teacher selects a number of authentic
letters to agony aunts and the corresponding replies and hands these out
randomly. Learners work in groups and in the first phase match the
answers to the appropriate requests for advice. In a second phase
learners can evaluate both the requests for advice and the answers given
and work out a catalogue of criteria for good request letters and good
agony aunt replies. In a third phase, some new requests for advice are
distributed and each learner group assumes the identity of the agony
aunt and writes a reply. In the fourth phase these replies are evaluated by
the whole class according to the criteria already agreed on. In a fifth and
final phase the replies actually written by the agony aunt can be handed
out and compared with the learners’ replies.
Some care should be taken to select letters with topics appropriate for
the group of learners in question with respect to age and cultural back-
ground. An authentic example I have seen successfully used in school at
intermediate level was a letter from an elderly couple in their seventies
who appealed for advice about how to deal with their two lazy sons,
both in their mid-thirties, who refused to leave the parental home,
smoked and drank a lot, provoked complaints from the neighbours and
generally made life unpleasant for their parents.
Beginning Writing
But how do we get young learners started on writing? From the early
stages of learning English learners can be asked to label a picture accord-
ing to teacher instructions, as in the ‘my monster task’ and ‘the zoo task’,
both introduced in Chapter 8. The zoo task can be supplemented by
a gapped written description of the zoo (a zoo guide) so that learners can
fill in the names of the animals. The final one or two sentences can be
supplied by the learners themselves. In the same way the ‘my monster’
task can lead into a controlled writing task in which the learners fill in
a gapped text which describes the monster. In some sentence the relevant
part of the body is omitted (‘It has blue _____’) and in some the relevant
colour is omitted (‘It has a _____ nose’). They can complete the text for
the final two colours and parts of the body by writing the sentences in
full on the model of the supplied gapped sentences. The principle is
always that learners first hear a word and then see it in written form
before they write it.
For older children, as well as adolescents and adults of limited English
writing proficiency, a good way to get them writing is to focus on the
genre of authentic notices, for example ‘Silence’, ‘Private’, ‘No milk
today’, ‘Keep out’, ‘Gone away’, ‘Reserved shelf’, ‘No entry’, ‘Through
traffic only’, ‘No trespassing’, ‘No food and drink beyond this point.’
Initially, the teacher can provide a selection of notices or photographs of
notices, but then learners can be asked to photograph authentic notices
and bring them to class. A first step might be for learners to put notices
into thematic groups, for example, road signs, signs in public places,
signs in libraries. They can also put notices into language-pattern groups,
for example, notices beginning with ‘no’, with ‘please’, notices ending in
‘prohibited’. A more demanding task is to modify the language of
a notice (for example, ‘No smoking’ to ‘Do not smoke’ to ‘Please do
not smoke’ or even ‘Please refrain from smoking’ so as to make it more
polite). Learners do not have to understand every word of a notice, just
to understand its pragmatic meaning. Cognitive challenge can be intro-
duced by showing learners, for instance, a picture of a notice outside
a walled estate saying ‘Trespassing prohibited’ and asking them to guess
what it means. If they cannot do this, they can be shown another notice
as a clue which says ‘Private property’ or ‘Keep out’, or both. Since
notices and messages are often elliptical, learners can also profit from
writing out a full-sentence version of the notice. Finally, learners might
be asked to write suitable notices to display in their classroom, or in the
school, including English versions of notices already displayed in the
local language. This will involve copying and copying with variation.
Another writing activity suitable for lower-intermediate level is to have
learners create their own greetings cards, especially Christmas cards,
Easter cards and Valentine cards, as well as invitation cards for special
occasions such as birthday parties, all based on a limited number of
models. It is best to use authentic greetings cards as models: either
printed or electronic ones.
Writing Short Descriptions and Narratives

As a next step, learners might be introduced to writing short descrip-
tions and narratives. This involves developing coherence and cohesion
in writing. Coherence refers to the implicit logical and epistemological
structure of a text and cohesion refers to the linguistic means
employed to make that structure explicit for the reader (Halliday and
Hasan 1976; Harmer 2004: 22–24; Wallace 1992: 11–14; Hudson 2007:
173–178). Halliday and Hasan (1976: 13) identified five sorts of cohe-
sive devices, namely: reference, substitution, ellipsis, conjunction and
lexical repetition/reiteration, Of particular relevance for learners are
pronoun reference (‘a boy and a girl … they’) and article reference (‘a
boy and a girl … the boy’). Also important are conjunctions such as
‘because’ and ‘although’ and conjunctive adverbs such as ‘therefore’
and ‘however’ as well as direct lexical repetition (as in ‘the boy … the
boy’) and what Halliday and Hasan (1976: 277–282) called ‘lexical
reiteration’, that is the use of near-synonyms (‘the boy … the young-
ster’) or of words at different levels of specificity (‘car … vehicle’) to
refer to the same thing.
In descriptive writing, cohesion often involves use of pronoun refer-
ence and place-adverbial reference across sentence boundaries, as in
‘A girl is sitting on a chair in a garden. She is reading a book. Behind
her there is a field ….’ One simple activity is for each learner (or learner
group) to write a description of a slightly different picture and then hand
it to a learner (or group) who has not seen the original picture. All the
pictures are then displayed on the wall and each learner or group decides
which picture matches the description they have been given (a matching task).
For the establishment of cohesion in narrative writing, ‘sequencing’
time adverbials, for example, ‘first’, ‘then’, ‘finally’, as well as tense
relations, especially simple and continuous past and past perfect tenses,
are important. Learners can be made aware of this by doing paragraph-
reconstruction tasks. Learners work in small groups and are first given
the constitutive sentences of a narrative paragraph in jumbled order and
asked to put them back into the right order and justify their decision (see
Harmer 2015: 371–372 for an example). Their attention can be drawn to
relevant cohesion markers and they can then write their own story based
on the model and incorporate cohesive features into it.
In this way they will learn that in English the simple past tense is the
normal or default tense for written narrative and the sequencing of verbs
in the text represents the sequence of events being recounted as in,
‘(First) I came, (then) I saw, (and finally) I conquered.’ They can be
shown how adverbs not only make the time order clearer but can also
express other relations such as cause and effect. For example, ‘so’
indicates cause and effect in, ‘She finished her meal. Then the doorbell
rang. So she went to the door.’ As learners progress, they can be shown
stories in which tenses are used to refer back to an earlier event (past
perfect tense) or to show one event interrupting another (simple past and
past continuous tenses). In such cases, appropriate use of adverbs makes
these time relations even more fine-grained as in, ‘She had just finished
her meal and was already clearing the table when suddenly the phone
rang.’ Learners can be encouraged to build these features into their own
stories. Wright and Hill (2008: 91–94) suggest some ways of doing this.
Picture sequences which form a narrative are another stimulus for
narrative writing and there are many published collections of picture stories
available. A problem-solving approach can be adopted by putting the
pictures in jumbled order, having a blank picture frame in the sequence or
leaving the final picture out, so as to have an open-ended story. Another
refinement is to provide evidence of time elapsed from one picture to
another (a clock on the wall in the picture, for example). Learners can then
be asked to incorporate this information into their rendering of the story.
Another way of eliciting written narrative is to tell the learners a story
using appropriate time adverbials, and then ask them to recall these
words. The list is then displayed on the board. Next, learners are given
the sentences which make up the story, but in jumbled order. They then
reorder these appropriately. A variation to increase difficulty is to omit
the time adverbials from the sentences in jumbled order and ask learners
to add these themselves. In a final phase, learners then produce a story of
their own along similar lines to the model story (Wright and Hill 2008:
89, 161).
Yet another well-known activity is writing a ‘chain narrative’ (Ur 2012:
160). The principle is that a group of, say, six learners writes a six-
sentence narrative paragraph on a topic they are given or which they
choose themselves. Usually the first sentence is specified. One learner
then writes a second sentence which follows on from the preceding
sentence, another learner writes the third sentence and so on. The final
writer has to make the paragraph in some way conclusive. An elegant
way of organising the chain narrative is the ‘story circle’, which makes it
possible for all the group members to be working at the same time (on
separate stories) and for multiple stories on the same topic to be
produced and evaluated afterwards. The group of, say, six learners sits
in a circle. Initially, each learner in the group has the same first sentence
in front of them. They each write a (different) second sentence, then pass
their papers to the learner next to them, who writes a following sentence
and so it goes on until six versions of the story are completed. Various
conditions can be introduced. For example, to practise discourse con-
nection it can be specified that each sentence must begin with
a connecting adverbial. A list of adverbials can be provided for learners.
The narratives can finally be displayed for the whole class to evaluate.
The ‘dictogloss’ technique (Thornbury 1999: 82–85) can be used to
introduce learners to writing stories with multiple paragraphs. The
teacher tells learners a story, which is clearly divided into, for instance,
three or four ‘oral paragraphs’. Suitable stories to choose are those
where the breaks between paragraphs mark a change of scene, or a leap
forwards or backwards in time. This will often be signalled by an initial
adverbial of time or place at the beginning of the new paragraph, some-
times by both if a shift of time and scene is involved, for example
‘Meanwhile, back at the ranch …’. While telling the story, the teacher
indicates the paragraph boundaries by means of pause and intonation.
Learners listen a number of times, they can make notes and then in
groups they reconstruct the story, write down a group version and
decide where to make the paragraph breaks. After the different versions
have been compared by the whole class, learners can listen to the story
again or see a written version of the original story and compare their
versions with the original (see Harmer 2004: 74–76, 2015: 380–381 for
examples).
The Transition to Discursive and Expository Writing

At intermediate level and above, learners need to make the transition to
discursive and expository writing. By this is meant all writing which
deals with abstract ideas and thus employs the language of written
argumentation in English. Many learners will one day need to write
essays, reports, recommendations, expositions and the like, whether in
an educational or business context. As a very first step, learners need to
be introduced to the idea of the topic sentence of a paragraph and to
practise identifying the topic sentence in well-chosen authentic para-
graphs related to their needs and interests. Halliday and Hasan’s (1976)
five sorts of cohesive devices, namely: reference, substitution, ellipsis,
conjunction and lexical repetition/reiteration, mentioned above, provide
a framework for understanding how ideas are linked in discursive writ-
ing. Particularly important are conjunctions which express abstract epis-
temological relations between two clauses within a sentence. These
include reason (‘because’, ‘since’, ‘as’), purpose (‘so that’, ‘so as to’, ‘in
order to’, ‘with the intention of’), result (‘so’, ‘so that’), condition (‘if’,
‘unless’, ‘in case’), concession (‘although’, ‘while’, ‘whereas’) and excep-
tion (‘except that’) (see Greenbaum and Quirk 1990: 316–324). Equally
important are conjunctive (linking) adverbs and adverbials which estab-
lish the relation between a new sentence and the previous one, for
example: ‘consequently’, ‘therefore’ and ‘as a result’, (result); ‘however’,
‘nevertheless’ and ‘despite this’ (concession); ‘on the other hand’ (con-
trast) (see Greenbaum and Quirk 1990: 184–187).
Cloze techniques are one way of drawing learners’ attention to para-
graph organisation and connectivity (Johnson 2018: 256). This can be
done by removing every nth (usually 7th) word from an authentic
passage. It can also be done by removing linking adverbials and other
cohesive devices. Another approach is not to leave blanks, but to switch
the linking adverbials around, as in this example for university students
of economics:
The techniques of modern large-scale manufacturing are based upon

economies of scale. RATHER, many operations can be carried out at
a much reduced cost per unit if the operation is to be carried out
thousands of times rather than just a few times. THIS IS BECAUSE
placing a large amount of manufacturing capacity under just one
control does not guarantee economies of scale. CONSEQUENTLY,
it furnishes the conditions for attaining them. HOWEVER, it would
be a mistake to view economies of scale as an advantage which
inevitably results as the extent of business operations is expanded.
Learners are given the task of rearranging the capitalised expressions

appropriately. When they have done this, they can then be asked to
decide which sentence is an explanation, a qualification, a result,
a contrast and a general statement, respectively. In a final step, learners
can be asked to reorganise the individual sentences. They are given the
beginning of each sentence and have to complete it, as follows:
Economies of scale … For many operations a much reduced cost per

unit … Economies of scale … The conditions for attaining them …
To believe that …
In a final phase, learners can write their own paragraphs on a topic
related to their own areas of expertise but which follows the discourse
structure of the model paragraph (parallel writing).
Using Pictures as a Stimulus for Writing

Learners need a lot of practice in writing, and the challenge for the
teacher often resides in finding stimulating material which engages lear-
ners’ interest and motivates them to express themselves in writing. One
way of stimulating written expression is by using pictures rather than
texts. This is particularly advantageous for less proficient learners since
time is saved that would otherwise be spent on understanding the
stimulus text before any writing could begin. Pictures can be used to
lead in to various sorts of writing (see also Harmer 2004: 67–69).
Single pictures of landscapes, buildings and interiors of buildings, as
well as various forms of ground plans and aerial views can be used to
stimulate description. For learners of intermediate and lower-intermediate
level, guidance will be needed on how to structure the description and in
what manner it proceeds. A certain amount of technical vocabulary (‘in
the background’, ‘in the foreground’) will need to be supplied. It is
advisable, if at all possible, not to do the description as just a pedagogical
exercise, but as a piece of functional writing (writing with a purpose) in
a particular situation so that a specific genre is practised, for example an
estate agent’s description of a house or a police officer’s description of the
scene of a crime. This can be done within the context of role-play.
To stimulate narrative writing, picture sequences in the manner of
a comic or cartoon sequence can be used, as already mentioned. The
difficulty of the task can be eased by providing support, as follows:
An accompanying gapped narrative text (cloze text) or a skeleton text

(keyword text) can be provided so that learners just have to fill in the
missing information rather than write their own text from scratch.
Captions for each picture can be provided, and learners can use these
as subheadings to structure their narrative.
Speech bubbles can be provided for each picture so that learners can
incorporate these as dialogue into their narrative if they wish.
Task difficulty can be increased, if appropriate, as follows:
The captions or pictures can be presented in jumbled order.

Instead of incorporating direct speech from the speech bubbles into
their narrative, more advanced learners can transform the direct
speech into indirect speech.
Blank or gapped speech bubbles can be provided for learners to
complete.
Reflective Writing (an Enigmatic Picture)
Enigmatic pictures which admit of different interpretations can be used
to stimulate reflective writing. One possibility is to use a picture which is
unambiguous for its intended audience but for intercultural reasons is
enigmatic for the learner group, who can be asked to interpret the
picture in writing. A newspaper picture I have often used is that of an
elderly lady who has just won the National Lottery. Some learners tend
to mistake the glass of champagne she is drinking for a glass of beer,
which leads them to miss the whole point.
Another way of using pictures to stimulate reflection is to present action
pictures which tell a story in the manner of a ‘still’ from a film. Pictures of
train derailments, plane crashes, mass motorway pile-ups, floods, fires and
the like are all suitable. Pictures of humorous mishaps or people being
caught in the act of doing something they should not be doing are also
effective stimuli. Learners can be asked to interpret the picture and write
about what they speculate has just happened before the picture was taken
and what will immediately follow. In all these cases, fine-grained use of
a wide variety of tense forms – especially past continuous and future
continuous forms – as well as adverbials and modal verbs to express various
degrees of probability and possibility will be required. For practice in
writing about the hypothetical, a question which is counterfactual to the
picture can be posed (‘What would have happened if …?’).
Evaluative Writing (Two or More Pictures for Comparison)

Pairs or groups of related pictures can be used to stimulate evaluative
writing. Pictures of people are well-suited to this. Learners can be told
that they have to select someone to do a particular job (for example,
looking after someone’s dog while the person is on holiday) based just
on their appearance and justify their choice to the dog owner in writing.
Pictures of real individuals unknown to the learners can be chosen,
including criminals, highly praiseworthy, exceptional citizens and some
ordinary individuals, for example, the teacher’s brother in-law.
Writing Which Accompanies a Picture

Pictures can also be used to stimulate specific genres of writing tied to
the picture. These include ‘Wanted’ pictures of criminals, pictures of
missing or lost pets, estate agents’ pictures of property and pictures of
cars for sale. Learners are given the picture and have to write an
accompanying text based on examples of similar texts of the same
genre. Learners can also be asked to write the accompanying story to
a newspaper picture, for example of a plane crash. This can be done in
groups, and afterwards learners can compare their productions with the
original article. To be able to do this task, learners will need some
preparation on the special conventions of newspaper discourse, for
instance that in newspaper articles the most important information
comes first and less essential detail is added later, which enables news-
paper readers to break off if they feel they already have enough informa-
tion after the headline, the sub-headline or any of the paragraphs (van
Dijk 1988). A refinement is for half the class to write an upmarket
paper’s report and the other half a tabloid report. This is suitable for
advanced learners working within a theme-based syllabus which has
a unit on newspapers and newspaper language.
Food for Thought

Q. 1. Display questions are usually not asked outside classrooms and
have come in for a lot of criticism from communicatively oriented
TEFL theorists. However, display questions have a long and
highly respectable history in education more generally, within the
Socratic dialogic
Sophoclean dialogicapproach
approachbybywhich
which the teacher’s carefully
the teacher’s
chosen questions guide the learner step by step towards under-
standing. How do you see the role of display questions in reading
comprehension?
Q. 2. Reading texts may be difficult for learners for various reasons. Find
four difficult texts or passages, one of which is lexically difficult,
one of which is grammatically difficult, one of which is pragmati-
cally or culturally difficult and one of which is difficult because of
its subject matter or argumentation. Consider how you would
prepare learners for their respective problems in the ‘pre-reading’
and ‘while-reading’ phases.
Q. 3. Imagine you have a class of lower-intermediate learners and you
want to teach them to write personal and business e-mails in
English. Which formal conventions/specific language will they
need to be aware of? Which of the approaches mentioned in the
chapter would you choose to teach them? Or would you employ
a different approach?
Further Reading
Grabe, W. (2009). Reading in a Second Language: Moving from Theory to Practice.
New York: Cambridge University Press.
Lems, K., Miller, L. D. and Soro, T. M. (2017). Building Literacy with English
Language Learners: Insights from Linguistics. Second edition. New York: Guilford
Publications.
Liontas, J. I. (editor-in-chief) (2018). The TESOL Encyclopaedia of English Language
Teaching. Volume 4. Teaching Reading (Anderson, N. J., Ed.), Teaching Writing
(Belcher, D. B. And Hirvela, A., Eds.). Hoboken, NJ: Wiley Blackwell.
Manchón, R. M. and Matsuda, P. K. (eds.) (2016). Handbook of Second and Foreign
Language Writing. Boston and Berlin: Walter de Gruyter.
Nation, I. S. P. (2009). Teaching EFL/ESL Reading and Writing. London and
New York: Routledge.
References
Bartlett, F. C. (1932). Remembering: A Study in Experimental and Social Psychology.
Collins, B., Mees, I. M. and Carley, P. (2019). Practical English Phonetics and
Phonology: A Resource Book for Students. Fourth edition. London and New York:
Routledge.
Davies, F. (1995). Introducing Reading. London: Penguin.
Grabe, W. and Stoller, F. L. (2011). Teaching and Researching Reading. Second
edition. Harlow: Pearson Education.
Greenbaum, S. and Quirk, R. (1990). A Student’s Grammar of the English Language.
Harlow: Longman.
Grice, H. P. (1975). Logic and Conversation. In Cole, P. and Morgan, J. L. (eds.),
Syntax and Semantics, Volume 3: Speech Acts. New York: Academic Press, pp.
41–58.
Hadfield, J. and Hadfield, C. (2000a). Simple Reading Activities. Oxford: Oxford
University Press.
Hadfield, J. and Hadfield, C. (2000b). Simple Writing Activities. Oxford: Oxford
University Press.
Halliday, M. A. K. and Hasan, R. (1976). Cohesion in English. London: Longman.
Harmer, J. (2004). How to Teach Writing. Harlow: Pearson Education.
Pearson Education.
Hudson, T. (2007). Teaching Second Language Reading. Oxford: Oxford University
Press.
Maley, A. (1993). Short and Sweet: Short Texts and How to Use Them. Volume One.
Maley, A. (1995). Short and Sweet. Short Texts and How to Use Them. Volume Two
Maugham, W. S. (1933). Sheppey: A Play in Three Acts. London: Heinemann.
Oxford: Heinemann.
Pawlak, M. (2014). Error Correction in the Foreign Language Classroom: Reconsidering
the Issues. Berlin: Springer.
Swan, M. (1976). Understanding Ideas: Advanced Reading Skills. Cambridge: Cam-
Thornbury, S. (1999). How to Teach Grammar. Harlow: Pearson Education.
van Dijk, T. A. (1988). News as Discourse. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Wallace, C. (1992). Reading. Oxford: Oxford University Press.
Wright, A. and Hill, D. A. (2008). Writing Stories: Developing Language Skills
through Story Making. Poole, Dorset: Helbing Languages.
Yule, G. (1996). The Study of Language. Second edition. Cambridge: Cambridge
University Press.
10 Literature in the Language
Classroom
A Wide Definition of Literature for the EFL Classroom

For TEFL purposes it seems sensible to adopt a wide rather than
a narrow definition of literature, embracing not only the ‘high’ literature
of the classical canon, but also those ‘low’ forms of popular or folk
literature in verse or prose to which Bakhtin (1981) drew attention. This
includes, alongside novels, short stories and poems, genres such as
nursery rhymes, pop songs, bedtime stories, fairy tales and comics as
well as films and TV drama (see also Griffiths 2016 for more suggestions).
The language of contemporary popular genres has the advantage that it
tends to be closer to the vernacular than the more stylised language of
high literature, especially older literature. It is also popular children’s
and youth genres rather than the classical canon which provide input for
first language development. For those learners who find reading difficult
or uninspiring for whatever reason, then TV and film provide an alter-
native means of access to English-speaking culture. These media are
readily available on the internet nowadays and advantage can be taken
of such popular, vernacular genres as TV ‘soap operas’.
For children of primary school age, nursery rhymes and picture books are
the relevant genres. Ghosn (2013: 135–136) discusses the selection of suitable
story books for young learners of primary school age under consideration of
a book’s literary and aesthetic merit, its enduring value, its developmental
appropriateness, the universality of its theme or topic, its plot, its illustrations
and its language. On the language of picture books, she comments:
The language in picture books should be rich and expressive but

with some amusing repetition that allows for reader/listener predic-
tions and confirmations. Ideally, grammatical structures and useful
formulaic expressions will be repeated and made clear in the story
context. Look also for vocabulary that provides synonyms and
alternative expressions.
Ghosn (2013: 136)
266 Literature in the Language Classroom
For secondary school learners, Matz and Stieger (2015) discuss the
advantages of using what are variously called ‘young adult’, ‘teenage’ or
‘youth’ novels, that is novels written for and about young people. In
these novels the protaganists tend to be young people of similar age to
the learners and the story often deals with concerns which are accessible
to young people. Some attention should be given to ensuring that the
novel will appeal to both sexes. In this respect Hesse (2009: 13–16) notes
that whereas girls are generally willing to read ‘boys’ books’, boys are
usually not willing to read ‘girls’ books’. For an annotated bibliography
which includes picture books, children’s, youth and young adult fiction
as well as poetry, graphic novels and story apps, see Bland (2018a:
277–300). For a discussion of story apps, see Brunsmeier and Kolb
(2018).
In this chapter, however, focus will be on both verse and prose for
adults, adolescents and children. Verse will include short poems as well
as children’s verse, and the terms ‘verse’ and ‘poetry’ will be used
interchangeably without any value judgement being attached to them.
Concerning prose, focus will be on short novels and youth novels.
Nothing will be said about reading plays in class because plays are
written to be performed rather than read, are generally more difficult to
deal with in the EFL classroom and are best reserved for mature,
advanced learners interested in the genre.
Three Sets of Reasons for Reading Literature in the Language

Classroom
The benefits of using literature in the language classroom include lan-
guage-learning benefits, motivational benefits and intercultural benefits
and these will now be explored.
Language-Learning Benefits
Reading literature provides a great variety of authentic, situatio-
nalised language in a wide range of clearly defined settings and
role relationships (O’Sullivan 1991: 3; Koutsompou 2015: 75).
Reading literature provides practice in a variety of reading techni-
ques, especially critical reading and empathetic reading (see Chapter
9 on reading techniques). Practice in empathetic reading can only be
acquired by extensive reading of enjoyable books outside the
classroom.
Readng literature provides opportunities for developing inferencing
and interpretative abilities so as to understand implicature, irony,
imagery and other forms of indirect meaning (Griffiths 2016; Lazar
1993: 19: Collie and Slater 1987: 5).
Literature in the Language Classroom 267
Motivational Benefits
Reading literature can be a powerful intrinsic motivator. This is
because literature is written to be read for its own sake, for pleasure.
The reader escapes from the real world and enters a fictional world
with which he or she becomes psychologically involved (Collie and
Slater 1987: 5–6; Lazar 1993: 15; Koutsompou 2015: 75). At best, the
reader does not want to put the book down.
Reading literature for pleasure can form the foundation for becom-
ing a lifelong autonomous reader, which in turn contributes to the
development of the personality (Lazar 1993: 19; Koutsompou
2015:76). Autonomous readers are willing, able and eager to read
without having a teacher available. If this is to be achieved, it is
important to select literature that is interesting enough to make
learners want to read it and easy enough for them to understand
without too much trouble. It is more important that the learner is
interested enough to continue reading a short novel to the end than
it is that every scrap of meaning has been wrung out of it – for every
abandoned book is a demotivator for reading further books in
English.
Intercultural Benefits
Reading literature makes it possible to experience English-speaking
cultures vicariously (Lazar 1993: 16–17; Collie and Slater 1987: 4). By
‘culture’ is meant the prevalent forms of social interaction and the
values which underlie them in a speech community. At the most
superficial levels these include rituals of greeting, conventions of
eating and drinking, politeness conventions, behaviour within the
family and in formal and informal situations at school, at work and
in leisure and in social environments. At more subtle and complex
levels, culture involves assumptions about human social interaction,
social behaviour and value systems, which may be manifested in
rituals and taboos (Grimm et al. 2015: 159–162). Language and
culture are intertwined (Kramsch 1998). The child acquiring its first
language is also simultaneously acquiring the cultural mores of its
society. Any speech community shares not only a mental lexicon but
also a mental encyclopaedia of cultural knowledge which lays the
foundation for ‘cultural literacy’ (Hirsch 1988; Clark 1996). The
mental encyclopaedia includes not only explicit knowledge learnt at
school, for example that Henry VIII had six wives, but also implicit
knowledge acquired indirectly through socialisation, for example
when to say ‘please’ and ‘thank you’.
Reading literature also provides EFL learners with an alternative
perspective on the world and on their own native culture, for
initially we all view the world through the spectacles of our own
culture. Readers are encouraged to adopt this alternative perspective
while reading, to reflect on the points of difference and, importantly,
similarity between the two cultural perspectives and, in a further
step, to see themselves as others see them (Ur 2012: 219–220).
The repeated experience of reading successfully, and largely unaided,
an authentic piece of literature can give learners the feeling that they
belong to the community of English-speaking readers for whom it
was written. This is not only an intercultural but also a motivational
benefit.
Short Poems
Suitable Poetic Genres

Short poems encourage various reading techniques for different pur-
poses, including empathetic reading, critical reading and form-focussed
reading (all discussed in Chapter 9). Appropriate short poems may be
read by intermediate or even lower-intermediate learners who are not yet
in a position to read a short story or short novel. Short lyrical poems,
including love poetry, descriptive poetry, reflective poetry and polemical
poetry, are often suitable for use in the EFL classroom, as is humorous
verse, including limericks, doggerel verse and riddles. For young lear-
ners, children’s verse, including nursery rhymes, is suitable.
In selecting poems, the most important criterion, and the one often
most difficult to gauge, is that of personal appeal to the learners. This
will depend, among other things, on the learners’ cultural background,
age, maturity, interests and level of language proficiency. If possible, the
poem should also fit in with the topics being covered in class at the time,
which may be laid down by the syllabus or course book. A suitably
chosen short poem is indeed often a good way of enriching the course
book. Once a piece of verse which seems suitable has been found, it can
be scrutinised more closely by considering how many of the following
positive features it possesses:
a personal, intimate tone, evocative of the speaking voice (as in much

lyrical poetry and many dramatic monologues)
the expression of personal emotions
musicality imparted by rhythm, metre and rhyme
language repetition with variation (as in refrains)
formal brevity so that the poem can be read in a single classroom
session
inclusion of a riddle-like element, a sting in the tail or a surprise
ending.
clear physical division into stanzas
No single poem is likely to score highly on all these features. However,
subsets of these features tend to co-occur in certain genres. For example,
lyrical poems often display a personal intimate tone, musicality, formal
brevity, emotional language and division into stanzas. Many humorous
poems, on the other hand, are short and riddle-like in that the reader has
to solve a problem while reading them. The riddle itself is an ancient
literary form, popular in Anglo-Saxon and Latin literature. It survives
today in popular culture and in children’s culture, and lends itself well to
exploitation in the EFL classroom.
However, the poem under consideration should also be scrutinised for
the following features which, if present, may make comprehension difficult:
considerable deviation syntactically and/or lexico-semantically from

the language of prose
greatly compressed syntax, ellipsis and lack of cohesive links, which
may require the reader to make difficult bridging inferences
double meanings, plays on words, reliance on secondary associative
meaning, metaphor, puns and allusion
stylised poetic diction, archaic language, word coinage or use of
a word in a nonce sense
These features in themselves do not rule out a poem. Indeed, many of

these features offer possibilities for exploitation in the classroom. It is
a question of matching the difficulty level of the poem to the proficiency
level of the learner group so that learners are challenged but not over-
challenged. Learners may need teacher support to cope with the language
within their zone of proximal development (ZPD) but this zone should
not be exceeded and learners should be able to solve the problems with
teacher support within the time available rather than just being given
explanations which are beyond their grasp. (You might like to look back
at Chapter 1 for what was said there about the ZPD.)
A Three-Phase Teaching Approach

In Chapter 9 a three-phase approach to reading short prose texts in class
was presented. This same approach with some modifications can be used
as a basis for reading short poems in the classroom. The following
remarks on each of the three phases should therefore be read as
a supplement to the guidelines given in Chapter 9, and it is probably
a good idea to look at them again at this point before reading on.
Pre-Reading Phase
In the pre-reading phase the learners might be given the title of the poem
and asked to infer what they can from it. Alternatively, the teacher might
recite the opening lines of the poem to the learners and ask them to
predict how they think it will continue, what it will be about, even, if
feasible, where and when it was written. This can be done by means of
brainstorming and probing questions. Another approach is to start with
a relevant picture related to the topic of the poem or with a picture of
the poet. If the name of the poet is likely to be familiar to the learners,
further brainstorming may elicit what the class knows about the poet’s
life and times. The teacher may introduce into the discussion a small
number of crucial words from the poem which might be cleared up in
advance and then displayed. To conclude the ‘pre-reading phase’ and
lead in to the ‘while-reading phase’, the poem might then be recited to
the class by the teacher, first without learners having access to the text
and then with them having access to the text.
While-Reading/While-Listening Phase
In the ‘while-reading/while-listening phase’, the poem will be read
silently, stanza by stanza. Some classes, however, may prefer to first
read the poem in its entirety before going through stanza by stanza, and
very short poems of a few lines can be read all at once anyway. The
teacher will provide guiding questions in advance of each stanza. These
will very often focus learners’ attention on key vocabulary and syntactic
aspects, especially where these differ from the standard language and are
difficult to understand. At the end of each stanza a rough meaning for the
stanza should be arrived at. This may involve focus on rhetorical
features, including metaphor, double meanings and the like. The aim
should be for the class to solve the problem to their own satisfaction.
They may first work in groups, then the results from the individual
groups can be pooled for the whole class. Residual problems can be
cleared up by the teacher, but those places in the poem which are
ambiguous and open to interpretation should be left open.
Poems are meant to be recited and recitation might conclude the
while-reading phase. The teacher can recite the poem in stanzas or as
a whole to provide a model for learners to follow. Alternatively, it is
often possible to use a recorded recitation, even a recitation by the poet,
which may be available, for instance, on YouTube. Learners can follow
a recitation while reading silently, can whisper the words during the
recitation or can read the poem out loud in chorus to accompany the
reciter. Young learners will enjoy reciting poems, first in chorus as
a whole class, following the teacher as model and then individually,
perhaps one stanza per learner. For older children and adults, too, who
may be more reticient, this procedure is also recommendable. Poems
which have embedded dialogue are particularly suitable for recitation
since roles can be assigned to individual learners. Bland (2015) stresses
that children’s poems and rhymes should be performed in the classroom.
She argues that children will delight in the rhythm, rhyme and repetition,
all of which further memorisation and provide formulaic patterns which
serve as a basis for the acquisition of grammar and vocabulary.
Post-Reading Phase
In the post-reading phase a task-based approach is often advisable, espe-
cially for school language learners who may not like literary discussion of
a poem. Tasks can include matching the poem to pictures or matching the
poem to other poems, which may, for example, treat the same topic from
another perspective. Other tasks include writing sequels or prologues to
the poem in prose or in verse. It is also possible for learners to write their
own short poem on the model of the poem which has been read. Albers
(2015: 106–110) reports using Carl Sandburg’s six-line poem ‘Fog’ (www.
poetryfoundation.org/poems/45032/fog-56d2245d7b36c) which starts ‘The
fog comes/on little cat feet’, with a school class of 14–15-year-olds. In the
post-reading phase the learners were asked to write their own short poem
in which they were to compare a natural phenomenon to an animal.
Albers presents a poem produced by one learner which starts, ‘The rain
is coming suddenly/like bees.’
Humorous Verse for Children

For young learners of limited proficiency, children’s verse, especially
humorous children’s verse, is a valuable way of introducing theme-
based vocabulary. An example would be Kenn Nesbitt’s poem ‘School
Supplies’ (www.poetry4kids.com/poems/school-supplies/). In this poem
the speaker is ostensibly a young child of school age who lists all the
equipment a child needs for school. The metre and rhythm aid memor-
isation and children will enjoy reading the poem out loud. It can be fun
to successively speed up the pace of reading out loud to get the best
crescendo effect, which is implicit in the metre. This makes for excellent
articulation practice. There is a joke in the tail of the poem as the child
adds to its list of necessary school supplies items such as a new compu-
ter, a motor scooter and a giant screen TV and notes that mom is
looking ‘nauseated’. To understand the joke, learners not only have to
work out what the new, rare and difficult-to-pronounce word ‘nauseated’
might mean in context, but also to understand how the word field of
‘school supplies’ has been overextended. If all goes well, the children will
rise to the cognitive challenge and get the joke, which is a first step on
the road to appreciating humour in English. The fact that this is
a children’s poem ensures that the humour is suitably pitched at children
rather than adults.
Another Kenn Nesbitt poem with the joke in the tail which is suitable
for young learners of limited proficiency is ‘My dog does my homework’
(www.poetry4kids.com/poems/my-dog-does-my-homework/). This short
poem is also written in the first person and the speaker is a child who
tells how the dog not only does the homework every evening but gets all
the answers right. The syntax is simple and conforms to prose, the tone
is personal and the vocabulary unproblematic except for the word
‘slobbered’, which is crucial for understanding the joke in the tail,
namely that the homework cannot be handed in because the dog slob-
bers all over it. Learners have to infer the meaning of the word ‘slob-
bered’ from context (top-down processing) and from the word’s sound
symbolism (bottom-up processing). If all goes well, young learners will
be motivated to solve the riddle-like problem and to remember the
word, which is a useful one to know.
Humorous Verse for Adults

Maya Angelou’s poem ‘The Health Food Diner’ (https://allpoetry.com/
the-health-food-diner), which satirises excessive preoccupation with
healthy eating, is a good example of a humorous poem which could be
integrated into a thematic unit on ‘food’ or ‘health’ for adolescent or
adult learners. However, as tends to be the case in ‘listing’ poems, nouns
tend to predominate and the lack of relevant verbs, adjectives and
adverbs would need to be corrected for at other points in a teaching
unit on food.
For more advanced adult learners, an example of a motivating piece of
humorous verse would be ‘The Purist’ by Ogden Nash (https://westegg.
com/nash/purist.html), where the joke is once again in the tail. The poem
is about a certain Professor Twist, a scientist who ‘never bungles’. When
told that his wife has been eaten by an alligator, he ‘… could not but
smile,/“You mean,” he said, “a crocodile.”’ The syntax of the poem
conforms to prose and the only difficult word is ‘bungle’, which could be
covered in the pre-reading phase and is a useful word to learn. The word
‘purist’ does not occur in the poem itself, and it is probably best to let
learners infer the meaning from the joke in the last line and discuss this
in the post-reading phase. Follow-up work could involve research on the
differences between alligators and crocodiles.
Descriptive Poems
A suitable children’s descriptive poem which could also be used with
adults of limited proficiency is ‘Four Seasons’ by Cecil Frances Alexan-
der (www.dltk-holidays.com/spring/poem/m-4seasons.htm). Each of its
first four stanzas is devoted to a description of spring, summer, autumn
and winter, respectively, without naming the season concerned, while the
fifth stanza mentions the cyclical nature of the four seasons. This
structure provides a basis for structuring the lesson. The poem
constitutes a riddle and is well-suited to a matching exercise in which the
seasons are matched to the stanzas. The riddle element can be further
increased by giving learners the poem without its title and asking them to
supply a title or, for less proficient learners, to choose the correct title
from a choice of, say, three or four options. Alternatively, the learners
could be given the poem with the title and asked to label each of the first
four stanzas with the correct season. In the post-reading phase learners
can say what their favourite season is and justify their choice to their
fellow learners. This could also be done as written homework.
This poem is also well-suited to recitation and to analysis of rhyme
and metre. The rhyming couplets (aabb), are repeated from stanza to
stanza, which is a nice easy rhyme pattern for novice poetry readers to
identify. There are various instances of alliteration which can also be
focussed on (‘a coat of clover clothes the hills’, ‘slush and sleet’) and of
structural repetition (‘And I must dance and I must sing’). Directing
learners’ attention to these can constitute a first step towards their
becoming aware of poetic devices. Because the rhyme scheme makes the
lines rather predictable, the poem is well-suited to an exercise in which
words are blacked out and learners have to insert a suitable item, relying
on rhyme, rhythm, semantics and structure, for example:
The ground is thick with slush and sleet/And I can barely feel my _____.
If this is too difficult for a particular group of learners, then scaffolding

can be provided, as in:
And I can barely feel my f___.
or even:
And I can barely feel my f__t.
Alternatively, the final word of all the second lines of each couplet can
be omitted and given in a jumbled list so that learners have the task of
putting the jumbled items in the correct place in the poem.
Love Poetry for Advanced Learners

One great advantage of short poems is that they often deal with universal
topics to which most people can relate and which transcend cultural
differences, even though the setting may be strongly culturally specific.
One of the great universal topics is love, and all the things that can go
wrong with it. Humorous love poems such as John Betjeman’s ‘A
Subaltern’s Love Song’ are suitable for the EFL classroom. In this case
a reading is available on YouTube so that learners can appreciate the
metre and rhyme and of course use the rendition as a model for
recitation (www.youtube.com/watch?v=Zh5YmyH3EGI).
Rather more sardonic treatments of love can also work well, such as
Roger McGough’s ‘40 Love’ (http://a-poem-a-day-project.blogspot.de/
2012/07/day-25-40-love.html), a poem about a middle-aged couple playing
tennis. The poem is set out on the page with a gap in the middle of each
line and the iconographic meaning of this layout can be focussed on, as
can the double meaning of the title and the literal and figurative uses of
the word ‘net’. The poem is thus well-suited to a cognitively based,
problem-solving approach, with the emphasis being on guiding the
learners towards solutions rather than imposing interpretations on
them. In the pre-reading phase the relevant language of tennis scores
can be explored. However, the unravelling of the double meaning of the
title is probably best postponed to the post-reading phase. The post-
reading phase also invites discussion about the idea of lovers growing
apart in middle age: an idea which should be familiar to all adolescent
and adult learners. As follow-up work in the ‘post-reading phase’, role-
play of a counselling agency for the couple can be conducted. Learners
may like to create retrospective biographies for the couple, too, to make
the counselling more realistic.
Advanced learners may find it motivating to find they can read older
poetry. Some older love poems are quite accessible for advanced learners
precisely because the theme is universal. A very accessible love poem,
the language of which is simple but emotionally evocative, is ‘First Love’
by John Clare (1793–1864) (www.poetryfoundation.org/poems/50263/
first-love-56d22d33757cd). The poem deals with the experience of being
in love. Poetic and archaic language features do not greatly impede
comprehension. As a post-reading task learners can be asked to moder-
nise the language of the poem, making only minimal changes, and to
write it down in prose. They will be surprised at how modern it sounds.
In further follow-up exercises, prologues and sequels in prose can be
written. The poem ends with the couplet ‘My heart has left its dwelling
place/And can return no more.’ This image of the separation of the heart
from the suffering lover is very old and learners can do follow-up
research to trace the image in literature and songs through the ages and
across cultures. One line of research would be to see how it continues to
thrive in popular culture, for example in the lyrics of pop songs.
Another line of research would be for learners to find out about the
poet on the internet. They will discover there was indeed a woman with
whom Clare was unhappily in love for many years, and that it may have
been this love which drove him to the madness which overcame him in
his final years. Rather more difficult but still accessible is Clare’s short
poem ‘I Am!’ (www.poetryfoundation.org/poems/43948/i-am), which
starts ‘I am—yet what I am none cares or knows’, which deals with
alienation and rejection.
An even older love poem which is accessible for advanced learners is
‘To My Dear and Loving Husband’ by Anne Bradstreet (1612–1672)
(www.poets.org/poetsorg/poem/my-dear-and-loving-husband). The poem
starts with the idea of the loving couple being one, ‘If ever two were one,
then surely we,/If ever man were loved by wife, then thee,’. Because this
image of the two lovers being inseparable is so familiar, not only in
English-speaking cultures, comprehension should be unproblematic for
advanced learners despite the archaic pronouns and the complicated
syntax.
Polemical and Reflective Poems for Advanced Learners

Polemical poems are suitable for advanced learners because they invite
a rejoinder or reply to the argument. It is sometimes possible to find two
polemical poems which take up opposing positions on the same topic.
For example, on the topic of the First World War, ‘Peace’ by Rupert
Brooke (https://www.poetryfoundation.org/poetrymagazine/poems/
13074/peace), on the one hand, versus ‘Anthem for Doomed Youth’ by
Wilfred Owen (www.poetryfoundation.org/poems/47393/anthem-for-
doomed-youth) or ‘Break of Day in the Trenches’ by Isaac Rosenberg
(www.poetryfoundation.org/poetrymagazine/poems/13535/break-of-day-
in-the-trenches), on the other, could be contrasted.
Another polemical poem on war, suitable for advanced learners, is
Carl Sandburg’s ‘Grass’ (www.poetryfoundation.org/poems/45034/grass-
56d2245e2201c). Towns and cities are another topic on which many
suitable polemical poems can be found, for example ‘Chicago’, also by
Carl Sandburg (www.poetryfoundation.org/poetrymagazine/poems/
12840/chicago). A polemical poem which unites the themes of towns
and war is ‘Slough’ by John Betjeman (www-cdr.stanford.edu/intuition/
Slough.html).
Reflective poems, particularly those which invite completion or addi-
tion, are also suitable for the advanced EFL classroom. An example is
Robert Frost’s poem ‘The Road Not Taken’ (1916) (www.poetryfounda
tion.org/poems/44272/the-road-not-taken). The idea of choosing which of
two roads to take is a familiar one. The syntax of the poem is that of
prose and the vocabulary is everyday. A picture of two paths diverging
in a wood in autumn with the leaves already turned to yellow is a good
lead-in to the poem in the ‘pre-reading phase’ since the wood is described
as ‘yellow’ in the first line. Guiding questions can alert the learners to the
figurative meaning of the poem and to the switch to the future contin-
uous tense (‘I shall be telling this with a sigh’) to express the prediction
with which the poem closes, namely that in the far-distant future (‘ages
and ages hence’) the poet will be recounting how he took the less
frequently used road in the wood ‘and that has made all the difference’
(final line).
The nicely idiomatic phrase ‘to make all the difference’ and what it
implies can be focussed on in the post-reading phase. The poem is also
well-suited to structural analysis, and further follow-up work might focus
on the five-line stanzas, the rhyme scheme (abaab) and the metre (stress
patterns). Finally, the poem might be used as a springboard for learners
to write about a choice of roads (literal or metaphorical) they were once
confronted with and the consequences of their choice. This will require
good command of tenses.
Short Novels
Selecting a Short Novel

Once learners have reached upper-intermediate level, the first major step
on the road to becoming an autonomous and extensive reader in English
should be taken, namely the selection of a short story or short novel to
be read by the class as a whole. Careful selection is important and the
following catalogue of questions may be helpful (see also Nuttall 1996:
170–180.; Lazar 1993: 52–56; Hesse 2009: 12–13; Bland 2018b: 12):
Are the plot, characters and setting appropriate to the age and
interests of the learners? (Lazar 1993: 53, 56; Nuttall 1996: 170–171;
Bland 2018b: 12)
Are the plot, characters and setting appropriate to the literary and
cultural background knowledge of the learners? (Lazar 1993: 54;
Bland 2018b: 12)
Is the language appropriate to the learners’ level of proficiency?
(Lazar 1993: 53)
Is the length of the book appropriate to the time available? (Lazar
1993: 55)
Is the plot relatively straightforward?
Is the narrative technique straightforward and chronological?
Is there a small number of easily identifiable main characters?
Can learners, especially young learners, empathise with any of the
main characters? (Bland 2018b: 12)
Is there variety in terms of characters, settings, events, actions and
emotions?
Are there some clearly defined scenes which are exploitable for
language-learning activities? (Lazar 1993: 55; Nuttall 1996: 171–174)
Is the book authentic (neither simplified, abridged, nor a parallel
text)? (Nuttall 1996: 177–178)
Is the physical presentation of the book attractive (book cover,
typesetting and illustrations)? (Nuttall 1996: 178–179; Bland
2018b: 12)
Is the book compatible with the syllabus? (Lazar 1993: 55)
It is rare that any single novel or short story fully satisfies all these
criteria, but if a book fails to satisfy most of them, then it is probably
best not to use it. If at all possible, learners should be involved in the
selection process although this may not be advisable for novice readers,
who may not be able to make a principled choice. The more cognitively
mature the class is and the more experience they have of reading, both in
their own language and in English, the more they will be able to
participate in the selection process. However, it is not advisable to give
a class a completely free choice. Rather, the teacher should preselect
between two and six books which satisfy most of the above criteria and
present this short list for the class to choose from. In this way nothing
can go seriously wrong. A whole classroom session should be devoted to
the selection process and this should be carefully planned, since the
choice made by the class must be abided by. It is worth bearing in mind
that the physical appearance of the books shown for selection may
influence learners’ choices and it is clearly undesirable for the books to
be literally judged by their covers. For this reason it is worth ensuring
that all the books are reasonably new, clean, attractive and typographi-
cally reader-friendly.
The procedure for an advanced class would be to work in groups: one
group for each book. Each group would browse through its book (as one
might in a bookshop) and finally produce a group evaluation of it based
on a catalogue of criteria provided by the teacher, including aspects such
as overall length, characters, setting, subgenre (for instance, detective
story), language difficulty and likely interest value of the book. After an
evaluation has been arrived at, groups would be shuffled with at least one
person per re-formed group from each of the original groups so that each
book is represented. Within each group the evaluations of each book are
compared and a group decision is made on which book to recommend.
Then the whole class can discuss the decisions of each of the groups and
a final decision can be reached on which book to read. It is important
that each group has adhered to the criteria catalogue so that evaluations
are easily comparable.
Less advanced classes should only be given two or three books to
choose from, and more than one group should assess each book.
They should be provided with one or more of the following:
short excerpts from the book to read

the publisher’s blurb for the book
a brief outline of characters, setting and situation which does not
give away the plot
pictures or maps of the setting
Often there are various editions of a book available and some thought
should be given to which edition to choose. Generally, learners will prefer
clear, reasonably large print on bright white pages with clear spaces between
lines and wide margins rather than a lot of small-font text crammed on to
pages with meagre margins and close spacing between the lines. Illustrated
editions are generally preferable to editions without illustrations and
a bright, illustrated front cover is preferable to a drab front cover without
visual material. So as to make the reading experience as authentic as
possible, an edition produced for native speakers rather than a didactic
edition for foreign learners is advisable. Learners themselves can be
involved in the decision process of which edition to choose. The issue of
ownership is important and for this reason each learner should possess his
or her own personal copy. However, it is essential that everyone has the
same edition. Everyone must be on the same page, as it were!
What Makes a Short Novel Too Difficult?

A major concern is that the novel should not be too difficult for learners.
However, books may be difficult for various reasons. Some of the major
sorts of difficulty are conceptual difficulty, intercultural difficulty, topic-
related difficulty and language difficulty:
conceptual difficulty
Conceptual difficulty arises when the ideas being expressed are themselves
complex. An example might be the philosophical reflection one finds in
George Eliot’s novels, for instance. This sort of difficulty can be identifed
by asking the question. Could learners understand this book if it was
written in their mother tongue? As a rule of thumb, learners should never
be asked to read something which would be too difficult for them in their
mother tongue.
intercultural difficulty
Intercultural difficulty in reading arises when the writer takes for
granted certain knowledge about the culture which EFL learners may
lack. EFL learners may, for example, find the cultural mores repre-
sented in the story puzzling if these are very different from those of
their own native culture or they may view the novel from their own
cultural perspective, which can lead to misunderstanding. These
problems of spatial dislocation can be compounded by problems of
temporal dislocation if the novel is set in, or was written in, a bygone
age. Nineteenth-century novels with their social mores in the area of
marriage conventions and parental authority, for instance, are likely
to be difficult even for native-speaker readers, especially young read-
ers, since this is a world far removed from their own experience. In
pre-selecting a class reader, it is therefore often worthwhile for the
teacher to go through the book and mark aspects of setting, plot and
characters that are alien to the learners’ contemporary native culture
and, conversely, aspects which are readily culturally transferable.
topic-related difficulty
Topic-related difficulty arises when learners lack background knowledge
about the area of life with which the novel deals. A famous example is
Herman Melville’s (very long) novel Moby Dick which deals with whaling.
The problem is not limited to the inevitable occurrence of difficult
technical words. The real difficulty will lie in the fact that much back-
ground information may not be explicitly stated but left implicit. There is,
however, a positive side to topic-related difficulty for the foreign reader,
namely that topic knowledge can in part compensate for limited language
competence. If the learner is not only informed about the topic but also
interested in it, intrinsic motivation is likely to operate. For this reason
topics should, wherever possible, be matched to learners’ interests.
language difficulty
Language difficulty may involve pragmatics, syntax and lexis, as discussed
in Chapter 9. However, provided the gist of the story can be understood
and learners are sufficiently interested and motivated, then there is no
reason to reject the book. Learners’ ability to infer syntactic and lexical
meaning from context should not be underestimated. This is all that is
required for comprehension. If appropriate, but only if appropriate,
follow-up work on lexis and syntax can focus on productive use.
Screening Short Novels for Language Difficulty

There are various measures for assessing lexical and syntactic difficulty.
For lexical difficulty these include ‘lexical density’ and the ‘type–token
ratio’ of lexical words (see McCarthy 1990: 71–73). By lexical density is
meant the percentage of a text’s total words made up of lexical words
(nouns, verbs, adjectives, adverbs). The greater the percentage, the more
lexically dense and potentially more difficult the text is. To calculate
lexical density, the procedure is as follows:
1. Take a number of pages chosen at random from the book.

2. Count the total number of words.
3. Count all the lexical words (nouns, verbs, adjectives and adverbs).
4. Divide the total number of words by the number of lexical words
and express this as a percentage.
However, another factor also plays a role in lexical difficulty, namely how
many of the lexical words are different words (types) rather than repeat
occurrences of the same word (tokens). This figure can be obtained by
conducting a type–token ratio test of the lexical words. To calculate the
type–token ratio:
1. Count the number of lexical words which are different words (types),
ignoring repeat occurrences of the same word (tokens). In
performing this count, word types are defined as a common stem
plus inflectional morphemes. Thus ‘wash’, ‘washing’ and ‘washed’
are all tokens of one type, not three separate types.
2. Then divide the total number of lexical words (tokens) by this
smaller number of different lexical words (types) and express this as
a ratio of types to tokens.
For example, if there are 100 occurrences of lexical words (tokens) in the
passage, but these consist of only 20 different words (types), then 100
divided by 20 equals 5 and the type–token ratio is 1:5, that is, 1 type to 5
tokens.The lowest mathematically possible type–token ratio is 1:1, which
would mean that there is no lexical repetition, so the text is potentially
lexically very difficult. Higher values of the second element in the ratio
mean there is more and more lexical repetition.
Measures of syntactic difficulty include number of ‘words
per sentence’, ‘words per T-Unit’ and ‘subordinate clauses per T-Unit’.
To calculate words per sentence:
Take a number of pages chosen at random from the work and find
the average number of words per sentence for each page by dividing
the total number of words by the number of sentences.
Generally speaking, comprehension becomes more difficult as sentence

length increases, so the higher the average number of words per sentence,
the more difficult the text is likely to be.
However, sentence structure also plays a role, since coordinated clauses
joined by ‘and’ produce long sentences, but because there is neither sub-
ordination nor embedding, the text is still quite easy to process. For this
reason it is more accurate, if a little more time-consuming, to calculate not
words per sentence, but words per T-Unit (‘minimal terminable unit’).
A T-Unit is the shortest unit which it is grammatically allowable to punctu-
ate as a sentence. It is defined as ‘one main clause plus any subordinate clause
or non-clausal structure that is attached to or embedded in it’ (Hunt 1970: 4).
A clause in turn is defined as ‘any expression containing a subject or
coordinated subjects and a finite predicate [verb] or coordinated predicates’
(ibid.). As examples of clauses, Hunt gives, ‘Jim and I went home’, ‘I went
home and rode my bike’, ‘Jim and I went home and rode our bikes.’
However, ‘Jim went home and I rode my bike’ would be two clauses and
also two T-Units. To calculate words per T-Unit:
1. Count the number of T-Units.

2. Count the total number of words, excluding ‘and’, ‘or’ and ‘but’
between T-Units.
3. Divide the number of words by the number of T-Units.
The higher the average number of words per T-Unit, the more difficult
the text is likely to be.
An even more sensitive test for syntactic complexity is to calculate the
average number of subordinate clauses per T-Unit, as follows. To do
this:

2. Count the number of subordinate clauses.
3. Divide the number of T-Units by the number of subordinate clauses.
The more subordinate clauses per T-Unit, the more complex is the
syntax and potentially more difficult to understand.
Particularly difficult syntax is characterised by the subordinated clauses
being embedded within the main clauses rather than attached to them.
To measure this, a test for embedding can be carried out, as follows:

2. Count the number of embedded subordinate clauses.
3. Divide the number of T-Units by the number of embedded subordi-
nate clauses.
The more embedded subordinate clauses per T-Unit, the more difficult
the text is likely to be, syntactically.
Concerning pragmatic difficulty, this is best assessed by taking
a number of passages at random from the novel and examining the
cohesion markers discussed in Chapter 9. In particular, if sentence-
linking adverbials tend to be lacking and learners have to make difficult
bridging inferences from one sentence to another, and if pronoun
reference across sentences is not clear, then the text is likely to produce
difficulties for learners.
Plannning a Three-Phase Reading Course

How quickly a class will read a novel depends on many variables,
including the proficiency and reading experience of the learners, their
motivation, how much class time per week is available, how much
time for reading out of class is available and what other commitments
learners have. It is, however, not a good idea to have a class reader
drag on for too long, since motivation and interest may flag, the plot
may be forgotten and outside events may interfere with completing the
novel.
A three-phase plan of pre-reading, while-reading and post-reading
can be applied. The novel will usually be read mainly out of class. At
the end of each classroom session learners should be given a section of
the book to read for next time, starting at the end of the pre-reading
phase (see below) and extending up to the end of the while-reading
phase (see below). In the post-reading phase writing activities based on
the book can be incorporated. These can be used for assessment
purposes, should the syllabus or curriculum require this (see O’Sulli-
van 1991: 5). In the while-reading phase, classroom sessions should
generally not be used for reading a new section of the book but for
consolidating what has just been read at home and for preparing
learners for the next piece of reading to be done. This is part of
training learners to become autonomous readers. However, learners
should not just be left alone to read the book. The classroom sessions
are vital for providing learner support and function as bridging struc-
tures between out-of-class reading phases.
Pre-Reading Phase
The aim of the pre-reading phase is to arouse expectations about the
book. In the first introductory session, a possible point of departure is
the book’s cover and the book’s title. Learners can be encouraged to
brainstorm possible literal and metaphorical meanings of the title. They
can then speculate on what the book may be about. Any ambiguity or
a secondary allusion in the title which the class does not spot should not
be gone into at this stage but taken up after the whole book has been
read. Instead of this approach, or as a follow-up to round off the first
session, the teacher might read aloud the opening of the book to the
class, without the class having access to the printed page. The class can
then speculate on what they think might happen next. The teacher
should be careful not to give away what actually does happen next! The
class’s predictions can be discussed in later sessions after they have read
some of the book.
In a second introductory session, brainstorming about the time and
place in which the book is set might be done.The teacher might supply
appropriate pictures of the setting. Key background information, for
example about the profession of a major character, may be provided by
the teacher if probing indicates that learners do not possess this knowl-
edge already. As much relevant information should be elicited from the
class as possible and the teacher should supplement or correct any wrong
assumptions parsimoniously. Only in areas where it appears that learners
are vastly uninformed or actually misinformed compared to the intended
native-speaker readership, should the teacher provide copious information
to redress the balance. Concerning vocabulary, only a very limited
number of key words (no more than five or six) which are vital for getting
started with the book should be discussed at this pre-reading stage.
Learners should start reading the book after the first or second pre-
reading sessions and be asked to get to a specific point for the following
session. Rather than saying, ‘Read to page 10 for next week,’ it is better
to say, ‘Read to the end of Chapter 1’ for next week, or ‘Read up to “he
stepped off the train and looked at the sleepy little town he had said
goodbye to all those years ago” on page 11, line 7.’ Where possible,
learners should be asked to read to a cliff-hanger point.
Starting in this session and continuing for each session of the ‘while-
reading’ phase, a small number of guiding questions, perhaps between
three and seven, should be set to help learners through their allotted out-
of-class reading (see Chapter 9 on ‘guiding questions’). Answers to these
questions will form the basis of the first half of the following in-class
session each week (see ‘while-reading phase’, below). If learners are
keeping a reading log, the guiding questions can help them to structure
their log entries (see ‘reading logs’ below).
While-Reading Phase
Each classroom session in the ‘while-reading’ phase falls into two parts.
The first part consists of consolidation work on the section of the book
which has just been read out of class since the last session and the second
part consists of preparation for the next section of the book to be read,
including the provision of guiding questions. For the consolidation part,
learners work in small groups and compare their answers to the guiding
questions. Any difference of opinion within groups can then be dis-
cussed later in the session by the whole class, or the teacher can be
asked for help. Every effort should be made to ensure that the classroom
sessions do not just degenerate into re-reading the last section assigned
for reading at home. If some backtracking is absolutely necessary in
class, then it is better to concentrate on one or two key scenes from the
previous section. If possible, the teacher should refrain from beginning
each session by summarising the plot to date. This might encourage some
learners not to read at home and simply to rely on the teacher’s plot
updates.
This approach is not without its practical problems, for which it is
best to be prepared. In particular, some class members (the high-fliers)
will tend to do the reading and come to class already armed with full
answers to the guiding questions, while others (the laggards) will tend to
fall behind in their reading, and, if left to their own devices, would fall
hopelessly behind and be lost for ever after a few sessions. The aim of
the group work is to iron out these differences after each reading assign-
ment. If all goes well, discussion of the guiding questions in groups will
not only bring the laggards up to date but will have a disciplining effect
on them since they will not want to be found lacking by their peers. In
this regard, at least for school classes, learners should generally not be
left to form their own working groups since there is then the very real
danger that the laggards will choose to stick together within their shared
comfort zone and the high-fliers may also stick together. Rather, the class
should work in mixed groups of high-fliers, average-pace readers and
laggards. A good policy is for groups to be selected by the teacher at the
beginning of each session with one member of each group being
exchanged each week.
The second part of each ‘while-reading’ classroom session is devoted
to preparation for the next section of the book to be read. This may
include the teacher reading the opening of the next section or chapter
out loud, and, if a change of scene or a shift in time is involved, probing
to see if learners have grasped this. Relevant character developments
should also be discussed and expectations about pending new develop-
ments in the plot elicited. Finally, as already mentioned under ‘pre-
reading phase’, new guiding questions should be set to guide learners
through their next chunk of reading. These questions should focus on
possible surprises learners may experience while reading and should be
so formulated that they do not actually give the plot away in advance.
For example, ‘Why do you think character X reacts the way she does
after she learns character B’s secret?’ (This question neither gives away
the reaction nor the content of the secret.) Learner reactions to guiding
questions should always be taken up in the following classroom session.
More advanced classes can be asked to formulate their own guiding
questions (see Kucza 2012).
Post-Reading Phase
How much classroom time is available for this phase will vary, but after
the book has been read learners should produce something of their own
which goes beyond the book and encourages them to reflect on what
they have read. This may be done individually or in groups. It is best in
the final weeks of the while-reading phase to give learners a choice of
possible activities to choose from, but to be open to any further sugges-
tions learners may make themselves. Possible post-reading activities
include:
analysing the language of a key passage

rewriting a chapter or scene from the perspective of another
character
rewriting a key dialogue from the book, replacing one of the partici-
pants by oneself
rewriting a dated scene or dialogue (perhaps an argument between
parent and child or two lovers) so as to bring it up to date
presenting the supposed viewpoint of a major or minor character on
something which occurs in the story
importing a character from another book in place of an existing
character and rewriting a scene as this character might act
dramatising a scene (see Bamford and Day 2004: 115–121)
contacting the author via the publisher by e-mail and posing ques-
tions (whole class activity)
writing a prologue or a sequel to the book
dividing the story into a certain number of episodes for TV drama-
tisation purposes, trying to close each episode at a cliff-hanger point
rewriting a section to change the plot development
writing an alternative ending, perhaps providing a happy rather than
a sad ending or vice versa (see Bamford and Day 2004: 164)
filling in something not included in the narrative, such as what
character X did while in place Y
adding a fictitious dialogue of the pattern, ‘What would characters
X and Y have said if they had met at point Z in the story?’
writing an inner monologue for a character expressing what he or she
was thinking when he or she did this or that (see Bamford and Day
2004: 148)
writing a psychologist’s report on one of the characters
writing a review of the book
(for suitably talented and interested learners) making a set of illustra-
tions for the book
(for suitably talented and interested learners) designing an improved
book cover
writing an introduction to the book aimed especially at foreign
learners
writing a blurb for the book
Many of the activities mentioned above are well-suited to group work.

Some can be done in classroom sessions, some out of class as project
work and some can be done within the context of keeping a reading log
(see also Koutsompou 2015: 77–78 for some further suggestions). For
assessment purposes written tasks of the ‘personal response’ sort sug-
gested in Bamford and Day (2004: 148) can be set. These include writing
an appraisal of the character the learner most or least identifies with, or,
alternatively, assuming the role of a character and writing either a letter
to a friend or a diary entry at a particular point in the story. A further
variation is for learners to imagine that one of the characters has moved
in next door to them and to describe a day spent with the character.
Alternatively, learners can imagine that a character has written to them
for advice on some point and they should write a reply (see also Bamford
and Day 2004: 155). Yet another idea is for learners to mention any
lesson for their own lives they have learned from the novel.
Reading Logs
Class readers do not necessarily have to be accompanied by a reading
log. However, a reading log provides the learner with a personal, tangible
record of his or her reading experience. It records the learner’s emotional
interaction with the book, encourages ongoing reflection on reading and
promotes individualised learning and learner autonomy. This is far
better than doing tasks which have been set by the teacher. Kaupmann
(2012), working with children aged 13–14 years in mixed ability classes in
a German secondary school, stresses that the reading log caters for
individual interests, different levels of reaction, quantitative variation in
production and different speeds of reading. For these reasons a reading
log is particularly valuable for learners embarking on reading their first
short novel as a class. If a reading log is to be kept by novice readers, the
pre-reading phase will have to be extended by one or two extra sessions
devoted to explaining how to keep a reading log. The extra time and
effort spent in explanation and setting things up at the beginning will be
amply repaid by the benefits which accrue in the following weeks.
It is, incidentally, a good idea to use the term ‘log’ rather ‘diary’ or
‘journal’ with learners. It suggests a journey or voyage, as in a ship’s log.
One way of introducing the term in the ‘pre-reading’ phase is to explain
what a ship’s log is and point out that certain features of a ship’s log are
relevant for the reading log. In particular, the ship’s log is written up at
the end of each tillerman’s turn of duty and, similarly, the reading log
should be brought up to date as soon as possible after reading each
allotted section of the novel so as to mirror the reader’s immediate
reactions. Furthermore, the ship’s log charts the ship’s course but also
records anything out of the ordinary which attracts the tillerman’s
attention and, similarly, the reading log should chart the course of the
plot but also record things which engage the reader’s attention. This
means it should not be limited to chapter summaries. Learners will
appreciate being given a short list of possible items to include in their
reading logs and they should be encouraged to vary their choice of items
appropriately (not just mechanically) from entry to entry.
Possible reading-log entries include:
personal reactions to new characters, places and institutions intro-

duced in the book as well as plot developments or actions of
a character which surprised, shocked, disappointed or favourably
impressed the reader
problems or difficulties the reader had in understanding aspects of
the plot, setting or a character’s motivation
updates to previous entries concerning things previously problematic
which have now become clear as well as things the reader has now
changed his or her mind about
predictions as to how the plot will go on in the next section to be
read
one or more new words that the reader liked or found useful and
why this is so
one or more words the reader found tricky and had to look up in the
dictionary
extra background material or personal creative material, including
pictures, verse or a comic strip (see Bamford and Day 2004: 148,
158–162)
Reading-log entries should look both forwards and backwards. Looking

forwards involves making plot predictions but also expressing hopes and
fears. Looking backwards involves confirming or revising predictions,
hopes and fears made in previous log entries. In the post-reading phase
additional entries which look back at the novel as a whole can be
incorporated into the reading log. Hesse (2009: 102–103), however,
cautions that some learners in schools, especially boys, may not like the
element of personal reflection involved in many reading-log tasks and
may prefer more hands-on, practical tasks such as doing web research,
drawing or even making a model of the landscape of the novel’s setting
or dramatising parts of the novel. Learners can also assume the role of
a film director who has to reduce the length of the novel by one third.
The reading log can provide teachers with ongoing feedback on how
learners are progressing, but teachers should consider carefuly whether
they want to avail themselves of this opportunity or not. It may be better
for the teacher not to have access to the log and to regard it not as
a display document but as a private document which is the personal
property of the learner (like a personal diary). At the very least, the
teacher should not read the log without the consent of the learner and
probably should not write on it, that is, should not correct language or
make written suggestions for improvement in the margin.
It is regrettable that reading logs are sometimes used as part of the
assessment procedure, since this encourages learners to write for the teacher.
If a reading log becomes nothing more than another testing instrument,
many of its potential motivational and educational benefits will be vitiated.
If the syllabus or the institution does require a reading log to be used for
assessment, then it is best to do this on a portfolio basis so that learners can
select specific parts of their log for assessment. The precise details of the
assessment procedure must also be made clear to learners in advance.
Project Work
Links between Reading Logs and Project Work

Reading logs can lead in to project work in which learners work largely
autonomously over a period of several weeks. However, whereas reading
logs are a personal product, project work is often best done in groups of
two to six persons. Project work will normally involve groups of learners
answering a question which goes beyond the content of the novel and for
which it is necessary to gather additional information. For example, after
reading Black Beauty by Anna Sewell, learners might consider, ‘How well
or how badly did the Victorians treat their horses?’ Like keeping
a reading log, project work, whether done individually or in groups,
involves personal involvement, responsibility and ownership. Both result
in an end product in which learners can take pride and experience
a sense of achievement. Skills are integrated around the task at hand in
a natural way so that the gap between language study and language use is
effectively bridged. Both reading logs and project work further learner
self-direction and autonomy as well as cooperation. In project work in
particular, learners are given the opportunity to draw on knowledge
from different areas of their lives. This means that each individual may
have something special to contribute and that all can benefit from the
specialised knowledge or experience of individual learners. Learners also
have the opportunity to contribute what they have learned in their
profession or in other academic subjects as well as from relevant books
they may have read in their native language or in another foreign
language they command. This gives them a sense of the unity of learning.
Within a school setting, opportunities for interdisciplinary learning also
present themselves.
A Three-Phase Plan
For project work, a three-phase plan is advisable, consisting of in-
classroom planning, extra-classroom execution and presentation:
in-classroom planning
In this phase learners discuss the content of their project with the
teacher and decide on how its scope is to be limited. Methods of
data collection such as interviews, visits, literature research and web
research are discussed. A sufficient number of in-classroom planning
sessions must be held before the execution phase starts, because then
learners should work independently of the teacher.
extra-classroom execution
In this phase learners now carry out their information-gathering
activities, sifting material, making notes and writing up the project.
Students may report back to the teacher for consultation if they
wish, and an office hour should be available for this eventuality. This
makes it possible for the teacher to spot in good time any procedural
errors learners may make. Learners should, however, self-monitor
their activities at all times in the execution phase precisely because
continuous teacher monitoring is not available.
presentation
This is the phase in which each group then presents its project to the
whole class for critical review and feedback. Teachers should make sure
that a clear time schedule for presentations is drawn up in advance and
adhered to. Otherwise, there is always the danger that presentations
take too long, insufficient time for critical review is available, presenta-
tions run late and the last few presentations and reviews have to be
hurried through. This should be guarded against at all costs.
Four Possible Problem Areas

There are four broad areas in which problems may arise in project work
before the presentation phase is reached. These are poor time manage-
ment, lack of perseverance, lack of self-direction and use of the mother
tongue:
poor time management

It is often difficult for learners to predict how much time it will take
to establish contact with individuals or institutions and get a reply or
to find specific material. For this reason the worst case scenario
should be sketched out in the planning phase and contingency plans
formulated in advance so that learners can quickly switch to a plan
B if necessary. These plans should be formulated in the form ‘If
X does not happen by DATE, then do Y ….’ It is very important to
do this in advance, since once students are in the execution phase the
teacher cannot intervene so quickly and so easily. Since it is unfortu-
nately often the case that precisely those groups who are experien-
cing problems do not choose to consult the teacher, it is sometimes
advisable to require each group to send an e-mail by a certain date
indicating their current state of progress.
lack of perseverance
Individuals may not develop sufficient interest in the topic or may be
discouraged by setbacks. For this reason it is important during the in-
class planning phase to make certain that learners are under no delusions
about the demands of their project. Another problem is that conflicts
within the group may lead to a loss of perseverance. This is one reason
for providing an office hour and learners should be strongly encouraged
to go to the office hour if they are experiencing intra-group conflicts.
lack of self-direction
The downside of freedom to choose is freedom to do nothing.
A small minority of learners may believe that by adopting a passive
stance and not acting on their own initiative the task will be done for
them by the teacher. These learners may turn up in the office hour
and expect to be provided with materials and information. To do
this would be counterproductive. The teacher should offer advice
but not do the group’s work for them. The problem can be mini-
mised in advance by ensuring that each group contains some self-
directed learners.
use of the mother tongue
In the case of groups who all share a single mother tongue, there is
always the danger that, once the teacher is not present, the mother
tongue will become the language of work. Learners should be
encouraged to always work in English. In multilingual groups, if
there are two or three mother tongues among the class which are
over-represented, then the teacher should ensure that mother tongue
subgroups do not form.
Nine Guidelines for Supervising Project Work

Many problems can be eliminated before they arise if certain simple
precautions are taken from the very start:
Stress throughout that learners are responsible for their own work.
Provide learners with checklists of things they need to do in each
phase.
Give clear instructions in advance of each new step so that learners
know exactly what to do next. It is best if these instructions are in
written form.
Ensure learners are fully briefed by the end of the planning phase
before they are sent off into the execution phase in their groups.
Be very firm during the in-class planning phase about limiting the
scope of the project since learners often tend to overestimate how
much they can achieve in the time available.
Instruct learners to draw up a strict time plan and to keep to it in the
execution phase.
For the presentation phase, ensure that learners have a realistic time
plan.
Impress upon learners that they should be in a position to shorten
their presentation if they start to run out of time, rather than having
to break off in medias res. An abridged presentation is preferable to
an incomplete presentation.
Be prepared to step in firmly during presentations which are taking
too long and urge learners to come to a conclusion.
Food for Thought

Q. 1. Which of the following advantages of using literature in the language
classroom, suggested by Collie and Slater (1987: 3ff.), do you find
most valuable: authenticity, cultural enrichment, language enrich-
ment, personal involvement? Which of these advantages do you
find least convincing? Do you think there are any other advantages
which are important?
Q. 2. Read the poem ‘The Road Not Taken’ by Robert Frost, mentioned
earlier in the chapter, and decide on how you would introduce it in
the pre-reading phase for a group of advanced learners.
Q. 3. Take any one of the poems mentioned in the chapter and decide on
some suitable post-reading activities for it.
Q. 4. (Group activity) Take either a short novel you have read yourself or
a youth novel you know or a children’s story-book you know. Based
on the suggestions made in the chapter, set up a catalogue of criteria
for assessing its suitability for a particular group of EFL learners you
are familiar with. How well would the results of an assessment
based on the catalogue chime in with your own subjective assess-
ment of the book’s suitability for the group of learners?
Q. 5. (Group activity) Imagine a teacher has asked you for advice about the
suitabilitiy of selecting The Quiet American by Graham Greene as
a class reader for a group of young, female, European EFL students
(aged 16–20) at an upper-intermediate/advanced level of proficiency.
Familiarise youself with the novel, test its suitability for these learners
against the criteria for selecting a short novel as class reader presented
in the chapter and decide what advice you would give the teacher.
Further Reading
Bland, J. and Lütge, C. (eds.) (2013). Children’s Literature in Second Language
Education. London: Bloomsbury Academic.
Hall, G. (2015). Literature in Language Education. Second edition. Basingstoke:
Palgrave Macmillan.
Schmitt-Egner D. (2007). Authentic Texts and Real-World Activities in the Classroom: An
Approach to Improving English Language Skills. Norderstedt: Books on Demand.
References
Albers, C. (2015). Poetry in the intermediate EFL classroom. In Delanoy, W.,
Eisenmann, M. and Matz, F. (eds.), Learning with Literature in the EFL Classroom.
Bakhtin, M. M. (1981). Discourse in the novel. In Holquist, M. (ed.), The Dialogic
ImaginatIon: Four Essays. (Translated by Emerson, C. and Holquist, M.) Austin,
TX: University of Texas Press, pp 259–422.
Bamford, J. and Day, R. R. (2004). Extensive Reading Activities for Teaching
Language. Cambridge: Cambridge University Press.
Bland, J. (2015). Performing poems in the primary school. In Delanoy, W.,
Bland, J. (2018a). Annotated bibliography: literary texts recommended for children
and young adults in ELT. In Bland, J. (ed.), Using Literature in English Language
Education: Challenging Reading for 8–18 Year Olds. London: Bloomsbury, pp.
277–300.
Bland, J. (2018b). Introduction. In Bland, J. (ed.), Using Literature in English
Language Education: Challenging Reading for 8–18 Year Olds. London: Blooms-
bury, pp. 1–22.
Brunsmeier, S. and Kolb, A. (2018). Story apps: the challenge of interactivity. In
Bland, J. (ed.), Using Literature in English Language Education: Challenging Read-
ing for 8–18 Year Olds. London: Bloomsbury, pp. 105–119.
Collie, J. and Slater, S. (1987). Literature in the Language Classroom:. A Resource Book
of Ideas and Activities. Cambridge: Cambridge University Press.
Ghosn, A.-K. (2013). Storybridge to Second Language Literacy: The Theory, Research
and Practice of Teaching English with Children’s Literature. Charlotte, NC: Infor-
mation Age Publishing Inc.
Grimm, N., Meyer, M. and Volkmann, L. (2015). Teaching English. Tübingen,
FRG: Narr Francke Attempto.
Hesse, M. (2009). Teenage Fiction in the Active English Classroom. Stuttgart, FRG:
Klett.
Hirsch, E. D. (1988). The theory behind the dictionary: cultural literacy and
education. In Hirsch, E. D., Klett, J. F. and Trefil, J. (eds.), The Dictionary of
Cultural Literacy. Boston, MA: Houghton, Mifflin and Co, pp. xi–xvi.
Hunt, K. (1970). Stylistic maturity in schoolchildren and adults. Monographs of the
Society for Research in Child Development 35/1.
Kaupmann, K. (2012). Suggestions for the effective use of reading logs in the ninth
grade. In Lennon, P. (ed.), Learner Autonomy in the English Classroom. Frankfurt
am Main: Peter Lang, pp. 123–146.
Koutsompou, V.-I. (2015). The use of literature in the language classroom: methods
and aims. International Journal of Information and Education Technology 5/1: 74–79.
Kramsch, C. (1998). Language and Culture. Oxford: Oxford University Press.
Kucza, J. (2012). Promoting learner autonomy by training students to generate their
own reading comprehension questions. In Lennon, P. (ed.), Learner Autonomy in
the English Classroom. Frankfurt am Main: Peter Lang, pp. 299–329.
Lazar, G. (1993). Literature and Language Teaching: A Guide for Teachers and
Trainers. Cambridge: Cambridge University Press.
Matz, F. and Stieger, A. (2015). Teaching young adult fiction. In Delanoy, W.,
McCarthy, M. (1990). Vocabulary. Oxford: Oxford University Press.
Oxford: Heinemann.
O’Sullivan, R. (1991). Literature in the language classroom. The English Teacher
20: 1–7.
Online reference
Griffiths, C. (2016). Using literature in the language classroom: an integrated
approach. Humanising Language Teaching 18/4. www.hltmag.co.uk
11 Language Testing
Approaching Language Testing
Teachers and Testing

Language testing is the measurement of language-learning development
and is for this reason alone of great importance to both language teachers
and learners. Beyond this, however, language testing is situated at the
interface between language teaching and society at large. This is because
in addition to teachers and learners there are other important stake-
holders involved, including parents, employers, governments and, ulti-
mately, the taxpayer. Parents, schools and universities all use language
tests to make important decisions which directly affect language learners.
Most language teachers will be involved with language testing in some
way. They may have to set, administer and mark tests for their own
classes. They may have to collaborate with colleagues in setting, admin-
istering and marking tests for their own institution. They may have to
prepare learners for external examinations which others set and mark.
They may have to invigilate at written examinations. They may have to
conduct oral examinations and assess oral performance, either alone or
in collaboration with colleagues.
It is therefore important for language teachers to know what language
tests can and cannot achieve, how far they can be trusted and what their
inherent limitations are. Parents and employers sometimes want to know
whether a specific language test is good or bad. The question, however,
should be whether the test is good or bad for a particular purpose,
namely the purpose for which it was designed. Problems arise when tests
are re-used for purposes for which they were never intended, with
learners for whom they were never designed and in situations not
originally envisaged. Problems also arise when the results of a test are
used to make decisions for which the test was not intended. In designing
one’s own language tests and in assessing language tests designed by
others it is very important to choose horses for courses rather than
trying to spot the winner.
294 Language Testing
Before setting any language test, test setters should always ask them-
selves the following questions:
Why am I testing?
What am I testing?
How am I going to test?
What will the consequences of the test be?
Testing Purposes
A broad distinction can be drawn between proficiency tests, which aim
to measure learners’ general language ability independent of any course
they have attended, and achievement tests, which aim to measure lear-
ners’ success on a particular language course (see Hughes 2003: 11–15;
Ur 2012: 167; Bachman 1990: 70–71; Johnson 2018: 262–263; Harmer
2015: 411). Also worthy of consideration are placement tests and diag-
nostic tests.
Proficiency Tests
Proficiency tests are, as a rule, standardised tests developed by some
external body. Proficiency tests may be limited to a particular skill, for
example, reading proficiency, or may aim to test proficiency across
a number of skills and tasks by means of a battery of different tests.
Well-known standardised English proficiency tests which use test bat-
teries include the Test of English as a Foreign Language (TOEFL) and
Cambridge Proficiency tests. They aim to rate learners’ language ability
for comparison with other learners anywhere in the world who may have
learned under quite different circumstances, may have various mother
tongues and may be of diverse social, cultural and educational back-
grounds as well as being of different ages. Such tests have a standard
design, are always intended to be taken under the same conditions and
always to be marked according to the same set procedure. Using a test
battery makes it possible to give learners discrete marks for each sub-
component of the battery, for example reading, writing, listening and
speaking. These discrete marks are valuable for diagnostic purposes since
they will reveal the relative strengths and weaknesses of the learner. For
detailed discussion of standardised tests see Brown and Abeywickrama
(2010: 103–121).
Achievement Tests
Achievement tests include progress tests, end-of-course tests and revision
tests. They differ from proficiency tests in that they are part of classroom
teaching. It is this sort of test in its various forms which classroom teachers
Language Testing 295
are most likely to be concerned with setting, administering and marking
(Hughes 2003: 12–13; Harmer 2015: 411). The defining characteristic of
achievement tests is that they are always based on a specific course or
syllabus that learners have followed or are following. They aim to establish
whether learners have learned what the course or syllabus has aimed to
teach them. This is usually done by looking at what has been covered in
class (course content) and testing it, but Hughes (2003: 14) argues for basing
achievement tests on course objectives rather than course content.
Placement Tests
Placement tests, sometimes called admission tests or entrance tests, aim to
place learners within broad bands for admission to particular courses
(Harmer 2015: 410; Johnson 2018: 263). If candidates come from a variety
of learning backgrounds, then a standardised proficiency test may be used
or the admitting institution may set its own proficiency test. However,
placement tests do not need to be as fine-grained as proficiency tests. They
only need to distinguish those candidates who fall within the proficiency
band stipulated for the course from those candidates below and above that
band. Hughes (2003: 16–17, 70–72) discusses the setting of placement tests.
For placement tests within an institution or where the candidates have all
followed the same syllabus or course, then the end-of-course achievement
test may serve as a placement test for the next course.
Diagnostic Tests
Diagnostic tests aim to identify learners’ strengths and weaknesses so that
future teaching can be matched to learner needs. This is also a test which
teachers are likely to set, administer and mark; for example, if they take
over a new class previously taught by another teacher. In a school context
the test might be based on the syllabus learners have been following or the
course book they have been using (an achievement test). However, if the
new class is very heterogeneous, with learners having followed different
syllabuses, used different course books and been taught according to
different methods, then a proficiency test which assesses specific skills
such as reading or writing, or domains such as grammar or vocabulary,
may be used as a diagnostic measure. For discussion of the possibilities for
computer-based diagnostic tests see Hughes (2003: 15–16).
Summative versus Formative Testing

Another perspective on test purpose is provided by the distinction
between ‘summative’ and ‘formative’ testing. Summative tests ‘provide
only a single mark, often expressed as a percentage, offer no specific
feedback on aspects of performance, and are designed to summarise or
conclude a period of learning’ (Ur 2012: 167–168). Formative testing, by
contrast, ‘happens in the middle of a period of learning rather than at the
end, provides clear feedback in the form of error correction and sugges-
tions for improvement, and has the primary aim of enhancing future
learning’ (Ur 2012: 168). According to Harmer (2015: 408), ‘Formative
assessment is part of the learning process itself and looks to the future,
rather than focussing exclusively on what has been achieved up to a given
point in time.’ There is actually a formative–summative scale along which
any test can be placed. A diagnostic test which does not give an overall
mark but rather identifies areas of strength and weakness for each candi-
date in certain skills and domains is at the formative end of the spectrum.
A proficiency test or an end-of-course assessment test which provides just
a single mark is at the summative end of the spectrum. However,
a proficiency test which consists of a number of test batteries and provides
separate marks or proficiency bands with descriptors for various domains,
such as grammar and vocabulary and certain skills, such as speaking and
writing, offers test takers and their teachers some differentiated feedback on
performance and has formative value. For examples of different forms of
test feedback see Brown and Abeywickrama (2010: 80).
Classroom achievement tests, especially progress tests and revision
tests, have a number of didactic or formative functions, as follows:
to make apparent the structure of the syllabus

to offer opportunities for language learning
to provide feedback on progress to learners and thus further learner
autonomy
to help maintain learner motivation
to provide feedback to teachers for fine-tuning of future teaching
Thus, in designing the test every effort should be made to ensure that the
test corresponds to the structural design of the syllabus, constitutes
a positive learning experience for learners, provides them with feedback
on their learning and will have positive motivational effects for them. It
should also provide the teacher with valuable feedback on progress (see
Ur 2012: 168). Even in the case of end-of-course achievement tests,
which are primarily summative, it is advisable to set the test for the
penultimate session and then hold a feedback session in the final session.
Even if the test marks the end of formal teaching, learners can still
benefit from test feedback and teacher advice for the future (see Brown
and Abeywickrama 2010: 7–8, 39).
The Three Phases of Testing

All testing, but especially formative testing, involves three phases for
learners:
preparing for the test (content and technique)
taking the test
reflecting on the test and obtaining feedback
Preparation for the test should build ‘anticipatory motivation’, doing the
test should generate ‘achievement motivation’ and reflection/feedback
should be conducive to ‘reinforcement motivation’. As a result, all
learners, not just the top 10%, should be positively motivated by the
test to go on learning. They should all feel a sense of achievement in
having successfully prepared for and completed a test. Tests should
never be used to show learners how little they know or how incompetent
they are! Every failed test, or test perceived as failure, is a nail in the
coffin of continuing language-learning motivation.
Young and beginning learners will need in-class training in test-
preparation techniques, while older and more advanced learners should
increasingly be able to take charge of their own preparation, which can
be done mainly out of class. Doing the test is itself a learning experience.
For this reason, as already mentioned, a feedback session should be held
as soon as possible after tests have been marked and scripts returned to
learners. Feedback should be both positive and negative. Emphasis
should be on individual learner progress rather than on comparisons
among learners. Ideally, this would be based on ‘individual-referenced’
marking criteria, which will be discussed later in this chapter. Sometimes
brief individual feedback sessions with each learner rather than a long
session with the whole class may be preferable. These individual sessions
can be held during class time while the class is engaged on other
activities.
Testing Techniques
Direct versus Indirect Testing

Ideally, one would want to test directly the precise real-world skills
learners require (see Hughes 2003: 17–19). Much real-world language
use involves drawing on a number of skills for a particular task, as
for example in interviewing (listening and speaking), note-taking
(listening/reading plus writing), summarising (reading and writing).
‘Integrated skills testing’ aims to get close to real-world language
needs by assessing performance directly on such complex tasks. In
some cases, however, setting up a direct test would be time-
consuming, difficult to administer and expensive. Therefore, testers
often have to find indirect methods of testing. Thus a skill is tested
which is judged to be a good guide to the target skill and which is
easier to test. In practice, most testing is more or less indirect, but
this always involves a risk since the learner is not being asked to
perform a real-world task under real-world conditions. An example
of an indirect test would be a written conversational cloze test to
assess conversational ability. Another sort of indirect test of language
skill is a test of language knowledge, for example of grammar or
vocabulary. Knowledge-based tests tend be easier to set and to score
than skill-based tests, but they are, unfortunately, poor predictors of
real-life language use because knowledge about the language may not
be directly transferable to performance in the language. (This is the
distinction drawn in cognitive science between ‘declarative knowl-
edge’ or ‘knowing that’ and ‘procedural knowledge’ or ‘knowing
how to’, discussed in Chapter 1.)
Discrete-Point versus Integrated Testing

A further distinction can be drawn between ‘discrete-point tests’, for
example yes/no questions, true/false statements, gap-filling exercises and
multiple-choice questions, on the one hand and ‘integrated tests’, such as
an essay, a translation, an oral presentation or interview, on the other.
Discrete-point tests consist of separate test items and test one thing at
a time (Harmer 2015: 412; Johnson 2018: 263; Hughes 2003: 19), whereas
integrated tests seek to test language performance as a whole (see Hughes
2003: 19; Johnson 2018: 264–267; Harmer 2015: 414–417). In integrated
tests there is no limit to the number of possible mistakes test takers can
make, but they also have the chance to perform positively without the
test explicitly demanding this of them (by using a particularly apt
idiomatic phrase, for example). Discrete-point tests are generally easier
to administer and mark and generally less time-consuming than inte-
grated tests. On the other hand, one obvious disadvantage of discrete-
point tests such as yes/no questions and true/false statements is that the
test taker has a 50% probability of being correct by chance, and in
multiple-choice tests with four options the probability is still 25%
(Hughes 2003: 76–77, 79).
Discrete-point tests lend themselves to objective marking procedures
and integrated tests invite subjective marking. Discrete-point tests with
one and only one correct answer per item require no judgement on
the part of markers and marking is therefore quite objective. All that
is needed is an answer key. If the test is laid out appropriately, a mask
can be placed over each completed test for quick manual marking.
Online versions of discrete-point tests can be computer-marked. Inte-
grated tests, by contrast, usually require more or less subjective
judgement on the part of markers. The more subjective marking is,
the more time-consuming and less economical it will be and the
greater is the likelihood of disagreement among markers. This means
that objectively marked tests are more statistically reliable than sub-
jectively marked tests.
Here are some well-known testing techniques:
Gap-filling (isolated sentences)

Example: Fill in the gap with the right word.
If it rains, you can get _____.
Matching of words or sentences (usually synonyms or opposite meanings)
Example: Match each word from the first list with a word with the
same meaning from the second list
large, unhappy, a lot of, little
small, many, big, sad
Multiple-choice
Example: Fill in the blank with a suitable word from the list below.
(1) hot (2) wet (3) bored (4) frozen
If it rains, you can get _____.
Cloze (a passage)
Example: Fill in the gaps with suitable words.
A cold front is approaching from _____ west and we can
therefore expect lower _____ tomorrow throughout the
country. There is _____ to be some rain in the morning,
_____ it will become brighter later in the _____.
Syntactic transformations
Example: Put the sentence into the passive
The dog bit the man.
Dictation
The tester dictates a passage and the test takers write it
down. Usually the passage is read through once first at
normal speed before candidates write, is repeated
slowly in sections with repetition while they write, and
is finally read through again at normal speed. (For
specific focus on spelling, a set of isolated words can be
dictated.)
Listening or reading comprehension questions
Example: Was the door open or closed when Jim arrived at the
house? (Yes/No)
The jacket was lying under the chair. True or false?
(True/False)
Where was the jacket when Jim entered the dining
room? (Factual question)
Why do you think Jim shuddered when he saw the
jacket? (Opinion question)
Guided writing (sentences or passages)
Example: (Based on a written text) Rewrite the passage from the
perspective of the woman referred to in line 3, making
any necessary changes.
Translation
The candidate is required to translate words, expressions,
sentences or a passage into or out of the foreign language.
Sometimes dictionaries may be provided, sometimes not.
Task-based writing
A simulated real-life piece of writing, often a letter or
report, is to be produced with a specified goal, addres-
see and context.
Essay writing
Example: Write an essay of approximately 2,000 words on
EITHER ‘The role of the monarchy in Britain today’
OR ‘The role of drugs in sport today’.
Gap-filling, matching and multiple-choice are classic discrete-point test

techniques. Cloze is claimed to combine elements of discrete-point and
integrative testing because although it consists of separate items, these
items are not completely independent of one another and an under-
standing of the passage as a whole is required. Dictation of a passage is
an integrated test but dictation of single words is a discrete-point test.
Listening and reading comprehension tests often combine discrete-point
techniques (yes/no, true/false and factual questions) with integrative
techniques (evaluative or opinion questions). Guided writing, translation,
task-based writing and essay writing are integrated tests. All the tests
listed above are written tests and for this reason are better guides to
written proficiency than to oral proficiency, which is usually tested
directly. For discussion of these and other test techniques see Ur (2012:
174–177), Hughes (2003: 75–82), Harmer (2015: 412–416) and Johnson
(2018: 263–268, 272–276).
Tests Which Focus on Receptive Skills

Sometimes one may wish to test mainly listening or reading comprehen-
sion, respectively, while minimising the use of speaking or writing. This
is often done by setting yes/no or true/false questions or factual ques-
tions requiring a short answer based on a recorded spoken text or
a reading passage, respectively. The longer the answers the questions
require, the more the test becomes an integrated test of both receptive
and productive ability (which may well be desired). Evaluative (opinion)
questions usually require longer answers than factual questions.
Multiple-choice questions based on a reading passage or a recorded
spoken test also test comprehension without demands on productive
ability. Multiple-choice tests are quick and easy to mark and they are
objective, but these advantages are partly offset by the difficulty of
constructing the test. The great problem is to produce options which to
the learner should all seem equally plausible. This is in fact very difficult
to achieve, particularly if the test writer has no assistance. A small team
of test designers is much more effective than one person. For some
criticisms of multiple-choice tests see Hughes (2003: 76–78), Johnson
(2018: 263–264), Harmer (2015: 412).
When designing multiple-choice tests there are therefore a number of
principles to abide by:
There should be only one correct option per item.

There should be no fewer than four items to choose from (other-
wise it may be too easy to guess) and no more than five (because in
practice it is often difficult to find more than five plausible
options).
The sentence stem should provide the full context necessary to solve
the task and conversely the options should be dependent on the
context created by the stem.
All the options should be formally comparable so that test takers
cannot make shrewd guesses at the odd one out.
The stems and the options should also be kept reasonably short.
(The form of the stem can be varied, however. It may be an
incomplete statement, a question, a gapped sentence or a sentence
in which an element is to be replaced by one of the options.)
For more detailed guidance on how to design multiple-choice items, see

Brown and Abeywickrama (2010: 67–78).
Before drafting questions for reading and listening comprehension,
one should first decide on which items of information one wants to ask
about. If one simply goes through the text, just drafting questions
spontaneously, it is likely that the focus will be on unimportant or
inessential information which it just happens to be easy to write
a question for. The items of information sought for should also be
sufficiently far apart in the passage to enable learners to answer one
question before having to focus on the next. In the case of listening
comprehension, candidates should also be given sufficient time before
the recording is played to familiarise themselves with the questions.
Otherwise, what is tested will be multi-tasking ability, involving reading
and listening simultaneously. Questions should take up key words from
the passage so that learners can use these as a means of orientation,
unless one intends to make the task extremely difficult (for advanced
learners, for example).
Information-transfer tasks, not included in the list above, are another
way of testing receptive skills while minimising productive skills, and may
be particularly useful for learners at low-levels of proficiency, especially in
classroom tests where the emphasis is on formative rather than summative
testing and on maintaining motivation. Information-transfer tasks include
picture-cued tasks, involving at the simplest level identifying items in
a picture. Picture-cued tasks can also be combined with multiple-choice,
either deciding which of two or more pictures corresponds to a spoken or
written cue word or sentence, or deciding which of a number of cue words
or sentences corresponds to a picture. Other forms of information-
transfer include labelling a diagram, completing a form and marking
a route on a map based on spoken or written instructions.
For examples of information-transfer techniques to test reading ability
see Nuttall (1996: 194–200) and Brown and Abeywickrama (2010:
239–240). For a similar approach to listening comprehension, see Brown
and Abeywickrama (2010: 169–171). Scanning tasks in which test takers
have to locate a date, a name, a number or a price, for example, also test
reception without any element of language production (Brown and
Abeywickrama 2010: 249).
Criterion-Referenced versus Norm-Referenced Testing

This distinction refers to the criteria against which tests are marked (see
Hughes 2003: 19–22; Bachman 1990: 72–76; Brown and Abeywickrama
2010: 8–9; Johnson 2018: 268). They should always be made clear to test
takers.
Criterion-Referenced Testing
A criterion-referenced test assesses learners relative to a set of predeter-
mined criteria, for example, ‘the ability to ask for information and
understand simple instructions’ or ‘the ability to write a letter of
recommendation which is understandable, stylistically acceptable and
not seriously marred by language error’. Theoretically, all the learners
taking the test could pass, assuming they satisfy the criteria, or they
could all fail, assuming they were all unable to satisfy the criteria. The
advantage of criterion-referencesd testing is that learners who achieve the
criteria receive positive reinforcement, which will be motivating, and
learners who do not achieve the criteria can be shown in what ways and
by how much they have fallen short.
Norm-Referenced Testing
A norm-referenced test does not assess learners with reference to pre-
determined performance criteria but with reference to how other
learners perform. A classic example of norm-referenced testing is IQ
testing. An IQ of 100 defines the intelligence of the average person. From
experience of testing many people, IQ scores above and below 100 become
associated with certain levels of capability. The same is true of scores on
a proficiency test such as TOEFL. The test itself does not say that with
a particular score the candidate is able to follow lectures in English or with
a higher score is able to write a Master’s thesis, for instance, which would be
done by a criterion-referenced test. Rather, based on past experience,
institutions set certain TOEFL target scores for admissions because they
know that candidates who score above a certain level have been able to, for
example, follow lectures, and above a certain higher score have been able to,
for instance, write a Master’s thesis (see Hughes 2003: 21).
The teacher who applies norm-referenced testing to his or her class for
achievement tests awards marks based on the achievement of
a hypothetical average learner for that class. In practice, the experienced
teacher probably has an idea in mind of what level of performance can
be expected of the average learner based on the past performance of
average learners over the years. This hypothetical average learner’s
performance becomes the norm against which the performance of other
learners is judged. This is why children who move from a less academic
school to a more academic school may suddenly find that their marks are
much lower than they used to be. It is not that their English has got
worse, just that a higher standard is expected.
Individual-Referenced Testing
Proficiency tests must be either norm-referenced or criterion-referenced to
fulfil their purpose. However, for classroom achievement tests, especially
progress tests and diagnostic tests, there is a third possible approach, namely,
‘individual-referenced testing’ (see Ur 2012: 170). This approach sets test
results which are neither norm-referenced nor criterion-referenced, but are
based on comparison with the individual’s prior performance. Thus, indivi-
dual-referenced testing aims to measure a learner’s individual progress. This is
likely to be far more motivating for most individuals and also provide them
with more fine-grained feedback to help them progress further.
Test Selectivity
All language testing is necessarily selective along a number of dimen-
sions, especially the following:
‘language selectivity’
This refers to the range of grammatical structures, vocabulary, prag-
matic functions and genres as well as the number of language
varieties, styles and registers which are tested.
‘mode selectivity’
This refers to whether the test is a written or a spoken test.
‘skill selectivity’
This refers to which of the four skills: listening, speaking, reading
and writing, are tested.
‘test design selectivity’
This refers to the tasks the test taker has to perform, for example
a gap-filling exercise, translation or a dictation.
‘topic selectivity’
This refers to what the thematic content of the test is, for example an
essay on ‘Drugs in Sport’ versus an essay on ‘The Role of the
Monarchy in Britain Today’.
‘test-conditions selectivity’
This refers, for example, to whether the test is an open-book exam
or not, whether it is done under time pressure or not and, for an oral
presentation, whether preparation time is allowed or not.
When a test is highly selective on a number of these dimensions, great

care must be exercised in generalising its results to other contexts.
Thus, learners who score well on a grammar test which tests only
prepositions might not have done so well if the test had been on the
passive and vice versa. A written test is not necessarily a good guide to
spoken ability and vice versa. A test based on receptive skills (listening
and reading) may not always be a good guide to proficiency in the
productive skills (speaking and writing) and vice versa. Transferability
of proficiency from one task to another should not be assumed either.
For instance, good translators are not necessarily good letter writers
and good conversationalists are not necessarily good negotiators or
good lecturers. Discrete-point tests may not always be the best of
guides to integrated language use, and knowledge about the language
may not be transferable to performance in the language. Tests done
under time pressure may not be a good guide to using the language
without time pressure and vice versa. Oral ability tested in a one-to-one
interview may be a poor guide to conversing against background noise
or chairing a meeting.
Where a number of discrepancies between the test and the future real-
world language needs of the test takers exist, one should be cautious
about expecting it to be a good predictor of real-world proficiency. That
is to say, it may have limited validity– something about which more will
be said soon. Thus, for example, a written test consisting of multiple-
choice questions on verb morphology might be expected to be a rather
bad guide to ability to handle a complaints phone-line for a washing
machine manufacturer. This does not mean that the test does not
measure mastery of verb morphology well, but rather that command of
verb morphology is not the only or even the most important aspect of
task proficiency for handling telephone complaints lines. In general, one
should be cautious about relying solely on indirect, discrete-point tests
for purposes which may have far-reaching consequences for learners.
This is the issue of ‘ethical validity’, about which more will also be said
shortly.
Test takers and Tests

Finally, one should always bear in mind that test takers are people and
not machines. There is variation among EFL learners themselves con-
cerning the way they react to tests in general, to the specific form of test
being taken and, in the case of oral exams, to the specific examiner.
Some of the personality factors discussed in Chapter 3, such as self-
confidence, anxiety, willingness to take risks and, in the case of oral tests,
willingness to communicate, are likely to influence test performance.
Test anxiety in particular can distort test results, particularly in oral
tests. All that was said in Chapter 3 about anxiety is acutely relevant in
test situations, especially in oral tests.
Furthermore, certain learners are able to apply general cognitive
strategies, such as dividing their time appropriately among questions,
while some learners may be rather inept at managing the cognitive
challenge of the test, including time management. These are differences
in ‘test-wiseness’ (Bachman 1990: 214; Harmer 2015: 412). Some learners
will grow in ‘test-wiseness’ over time as a result of practice in taking
tests. Some learners may receive special training and practice in specific
test-taking techniques, for example intelligent guessing for multiple-
choice questions, so that they learn the tricks of the trade and become
‘super test-wise’. This may improve their test result but it does not
necessarily mean their English has improved.
Evaluating Language Tests

Any test can be evaluated according to certain criteria but the evaluation
must always consider the context in which the text is conducted and the
purposes it is intended for. Lado (1961: 30–37) set up five criteria for
evaluating a test, namely: ‘validity’, ‘reliability’, ‘scorability’, ‘economy’
and ‘administrability’. To these may be added ‘discriminability’ and
‘backwash on teaching’ (see Hughes 2003: 53–57). This makes a total of
seven criteria for assessing a test.
Validity
By ‘validity’ is meant the extent to which the test measures what it aims
to measure and nothing else. It is possible to look at validity from
a number of angles (see Harmer 2015: 409; Hughes 2003: 26–35;
Bachman 1990: 236–295; Brown and Abeywickrama 2010: 29–36). These
include the following five angles.
Construct Validity
Construct validity can be regarded as the overarching idea of validity
(Hughes 2003: 30–32; Bachman 1990: 256; Brown and Abeywickrama
2010: 33–34; Harmer 2015: 409; Johnson 2018: 270). By ‘construct’ is
meant things such as ‘proficiency’, ‘fluency’ and ‘communicative compe-
tence’, which we are trying to get at in the test. As Brown and Abey-
wickrama (2010: 33) put it, to check construct validity we must ask the
question, ‘Does this test actually tap into the theoretical construct as it
has been defined?’ Weigle (2002: 41) defines ‘construct’ as ‘the ability we
want to measure’. She continues that for a language test this involves
‘determining what factors are involved in real-world language use, and
which of those factors are essential to what we want to measure and what
we do not.’ Construct validity therefore entails ensuring that ‘the test is
consistent with current theories of language and language learning’
(Johnson 2018: 270) and ‘performance … is consistent with predictions
that we make on the basis of a theory of abilities, or constructs’ (Bach-
man 1990: 254–255). Thus, a test of conversational ability that consisted
of candidates introducing themselves and saying something about their
hobbies and interests would lack construct validity because this is
a monologue and conversation involves two or more participants.
Face Validity
By ‘face validity’ is meant whether the test ‘looks right’ in terms of the
language skills it aims to measure. Of course, this is no guarantee that the
test really is a valid test (Bachman 1990: 286). Face validity is thus not
a scientific concept. However, it is difficult to persuade testers and test
takers to believe in tests with poor face validity, for example paper-and-
pencil tests of oral ability (see Hughes 2003: 33; Bachman 1990: 285–289;
Brown and Abeywickrama 2010: 35–36; Johnson 2018: 269–270).
Empirical Validity
By ‘empirical validity’, sometimes called ‘criterion-related validity’, is
meant how well the results of the test correlate with other recently
obtained measures of assessment, for example performance on another
language test or teacher ratings (concurrent empirical validity) and/or with
the learners’ subsequent performance, for example in a later final exam-
ination (predictive empirical validity). See Hughes (2003: 27–30) and Brown
and Abeywickrama (2010: 32–33). Empirical validity can only be estab-
lished after the test has been taken.
Content Validity
By ‘content validity’ is meant the extent to which the test covers
a balanced sample of the course content and language skills taught and
no others (for achievement tests). Or it means that the test covers the
domains of language and language skills it aims to test and no others (for
proficiency tests). Like face validity, content validity can and should be
established before the test is taken (Hughes 2003: 26–27; Brown and
Abeywickrama 2010: 30–32; Johnson 2018: 269).
Ethical Validity
By ‘ethical validity’, sometimes called ‘consequential validity’, is meant
whether the test provides a solid enough foundation to justify the
personal, educational, financial and career consequences it may have for
test takers (Bachman 1990: 279–285; Brown and Abeywickrama 2010:
34–35). Relevant questions to ask are, ‘Who will know the results of the
test, to what uses may these parties put the results and what control or
influence do I, the test setter, have over those uses?’ Also of relevance is
whether it will be possible for test takers to retake the test at a later date.
Concerns of ethical validity become particularly acute if the test consti-
tutes the sole criterion on which decisions will be based, unaccompanied
by other instruments, such as continuous assessment and teachers’
reports.
Reliability
‘Reliability’ is concerned with the stability and consistency of results
obtained on the test (see Hughes 2003: 36–38; Johnson 2018: 270–272;
Bachman 1990: 160–235; Brown and Abeywickrama 2010: 27–29;
Harmer 2015: 409). Bachman (1990: 160) describes reliability as being
concerned with answering the question, ‘How much of an individual’s
test performance is due to measurement error, or to factors other than
the language ability we want to measure?’ A completely reliable test
would be unaffected by:
the time and circumstances of testing (‘test-retest reliability’)

the identity of the marker and repeat marking by the same marker
(‘marker reliability’)
fluctuations in the test taker’s mood (‘test taker reliability’)
the topic of the test (‘topic reliability’)
(for oral tests) the identity of the tester (‘tester reliability’).
Test reliability can be measured statistically by calculating the ‘re-testing

coefficient of reliability’ of a test (see Hughes 2003: 38–40). A perfectly
reliable test would mean that the same set of candidates would theoreti-
cally score exactly the same on it every time (a reliability coefficient of
1.0). If, on the other hand, there were no relation at all between how the
candidates would score each time they took the test, then the reliability
coefficient would be 0. This would mean it was a completely useless test.
There are three methods of obtaining reliability coefficients, all of which
have obvious limitations (see Hughes 2003: 39–40):
the test-retest method

This involves a sufficiently large number of candidates taking the
exam twice. The two sets of results are then compared to obtain the
‘re-testing coefficient of reliability’. The disadvantage of the method
is that one would expect test takers to score better at a second
attempt on the same test (a learning effect).
the alternate forms method
This involves having the candidates take two different forms of the
same test. How closely the results of the two versions of the test
correlate or match gives an indication of the reliability of the test. The
problem is that a shadow version of the test has to be produced which
differs from the original version only in the content of each question.
the split halves method
This involves the candidates taking the test just once, but separate
scores are computed for each half of the test. Performance on each
half of the test is then compared. The correlation coefficient for
the two sets of scores is computed. This is the ‘coefficient of
reliability for one half of the test’. A statistical formula is available
to compute, from the reliability of one half of the test, a close
estimate of the entire test’s reliability. This is in fact the method
most frequently used to calculate test reliability. It is, however,
not always easy to split the test into two halves which are exactly
comparable (see Bachman 1990: 172–174).
Some types of test are inherently more reliable than others. Reliability
coefficients as high as 0.9 have been obtained for some highly structured
test-types such as vocabulary tests and reading comprehension questions.
Reliability coefficients for listening comprehension tests are unlikely to
be higher than 0.8, perhaps because of the acoustic element, which may
be subject to various external influences, including outside noise and
how well-placed the test takers are relative to the recording equipment.
Slightly lower reliability coefficients (0.7) are found for oral production
tests (see Hughes 2003: 39).
As a rough guide to interpreting reliability coefficients, one should
remember that the square of the coefficient gives a percentage measure
of the variance between the two sets of test scores being compared,
which is accounted for by the differences in ability the test is trying to
measure. The remaining variance is due to other factors. Thus,
a reliability coefficient of 0.7 means that the test accounts for 49% of
the variance and a coefficient of 0.5 means that only 25% of the variance
is accounted for by the test. These figures do not mean that 49% or 25%,
respectively, of test takers would have the same score if they took the test
on a different occasion. The coefficient is, rather, an overall measure of
statistical agreement between two data sets (Hughes 2003: 29).
If a published test reports a reliability coefficient, this gives an indica-
tion of how trustworthy the results of the test are, but it does not say
how close scores on the test are to candidates’ ‘true scores’. By ‘true
score’ is meant what candidates’ average score would be if they took the
test many times (Hughes 2003: 40–42; Bachman 1990: 166–167). Of
course, this would be impossible in practice for a language test for all
sorts of reasons. But the difference between the ‘true scores’ and ‘actual
scores’ is very important when decisions are being made about border-
line ‘pass’ or ‘fail’ candidates.
In fact, a further statistical technique is available which uses the reliability
coefficient of a test and the standard deviation (SD) of the test’s scores to
obtain a ‘standard error of measurement’ (SEM) for a test (see Brown 1999;
Bachman 1990: 197–202; Hughes 2003: 40–42, 223–224). This is expressed as
a single number and the more reliable a test is, the lower the number will be.
If, for example, the SEM of a test is 2, this means that an individual’s ‘true
score’ is within the range ±2 of his test score, with a probability of about 68%
(= 1 standard error). And we can be 95% sure that his/her true score lies
within ±4 of the test score (= 2 standard errors). These relationships apply
quid pro quo for any SEM value. According to Hughes (2003: 41), published
tests should give not only a reliability coefficient but also an SEM.
The SEM can thus be used to identify a sensitive band of scores around the
pass/fail borderline where candidates should be given special attention, with
consideration perhaps being given to teachers’ reports, or other work pro-
duced by these learners, before a decision is made. For instance, if the pass
mark for the test is 40 and the SEM is 2, as in the above example, then if you
want to be 68% certain that you are passing and failing the right people, you
should recognise that within the band of scores above 38 and below 42 this
degree of certainty is not guaranteed. If you want to be 95% certain that you
are passing and failing the right people, then this degree of certainty is only
guaranteed for those scoring below 36 (fail) and above 44 (pass). In such
circumstances, thought has to be given to the consequences of passing and
failing for learners. These are considerations of ethical validity, which have
already been discussed.
Perfect test reliability is a pipe dream, but often reliability can be
improved by ensuring that:
Test items or questions are independent of one another.

Questions are unambiguous.
Instructions are clear.
Candidates are familiar with the type of test being used.
There is little or no scope for successful guessing.
Marking is objective.
Topics, techniques and question types are varied.
Learners write or speak on more than one topic.
For further suggestions for improving test reliability, see Hughes (2003:
44–50).
Discrete-point tests tend to be more reliable than integrated tests because:
they are easier to score and their scoring is more objective (marker
reliability)
results are possibly less influenced by the test taker’s mood (test
taker reliability)
results are probably less influenced by the questions (topic reliability)
However, integrated tests are usually much closer to real-world skills and
may therefore often have higher validity for the testing purpose at hand
(Johnson 2018: 271–272).
Scorability
By ‘scorability’ is meant how easy or difficult it is to score (mark) the test.
More precisely, scorability is concerned with how much time it takes to
mark the test, whether markers need special training and briefing and
whether multiple markers are required. As already mentioned, if scoring is
objective, as in discrete-point tests with one correct answer, then scoring is
made very easy and can often be done by computer. Scorer-reliability
coefficients can be obtained in much the same way as test-retest reliability
coefficients. One would expect the scorer-reliability coefficient to be higher
than the overall test-reliability coefficient, which includes other sources of
unreliability. Hughes (2003: 43–44) cites a case where the scorer reliability
on an essay-writing test was 0.92 and the test-reliability coefficient was 0.84.
Economy
By ‘economy’ is meant the extent to which the test justifies the amount
of time, money and other resources invested in it. Other resources
include not only technical equipment but also human resources. Tests
conducted individually are much less economical than those done in
large groups. For this reason written tests tend to be more economical
than oral tests. A re-usable test is also an economical test. Discrete-point
tests which can be taken online are extremely economical once they have
been set up for the first time.
Administrability
By the rather cumbersome term ‘administrability’ is meant whether the
test can feasibly be administered in the specific situation. Administrabil-
ity is thus concerned with issues such as the availability of suitable
testers, adequate rooms, appropriate technology and adequate ancillary
personnel.
Discriminability
By the equally cumbersome term ‘discriminability’ is meant to what
extent the test effectively distinguishes among performance differences
to the degree required. It sometimes happens in discrete-point tests, for
example, that scores bunch at the top end of the scale (a ceiling effect).
Generally, a test should give weaker learners scope to show what they
know, while stretching stronger learners to their limits. However, in
placement tests which merely wish to establish a cut-off point as
a criterion for entrance to a course, discrimination across the whole
spectrum may not be necessary. In a criterion-based progress test,
bunching of scores at the top might merely show that learners have
mastered what the course aimed to teach them, which is fine.
Backwash on Teaching
By ‘backwash’ on teaching is meant the influence the test will have on the
teaching carried out to prepare candidates for it (see Hughes 2003: 1–2,
53–57; Brown and Abeywickrama 2010: 37–39; Johnson 2018: 261–262;
Harmer 2015: 410; Bachman 1990: 283). Generally speaking, direct tests
have positive backwash effects. For example, direct oral testing of spoken
ability to conclude a conversation course will tend to mean that a lot of
speaking will be done in class (Bachman 1990: 283). Indirect tests may
have negative backwash effects. For example, assessment of reading
comprehension solely by a multiple-choice test may lead to too much
class time being spent on practising test-taking strategies (Johnson 2018:
264). The issue is tricky, for unfamiliarity with a particular type of test
can depress performance and lower reliability. Thus, some test-taking
practice will be necessary and desirable to make learners familiar with
the type of test they have to take (Harmer 2015: 421–422).
Written Language Tests
Piloting the Test

Ideally, a written test should be piloted in advance with another group of
learners comparable to the target group. If this is not possible, one or
more colleagues should be asked to read through the test in advance and
see if they find anything problematic. If this is not possible either, then
the test setter should assume the role of a test taker and take the test in
advance under exam conditions to assess its feasibility (see Brown and
Abeywickrama 2010: 77–78; Harmer 2015: 418).
Advance Information for Test Takers

Test takers need to be informed of the following things in advance:
the time and place of the test

the test conditions (for example, the availability or not of aids such
as dictionaries)
what they must, may and may not bring into the examination room
the length of time allotted for the test
how unanswered questions, incomplete answers and ‘off-topic’
answers will be marked
(in achievement tests) what parts of the syllabus are being tested
(for novice takers of written exams) the rules of maintaining silence,
of non-consultation with one’s neighbour, of not looking at one’s
neighbour’s script and of asking the invigilator if one wishes to leave
the room
Invigilation
On the day of the test the invigilator should exude calmness, confidence
and relaxation and, if he or she is also the classroom teacher, should on
no account adopt a different persona from the normal classroom persona
the learners know. If the invigilator seems tense, then the learners will
get tense, too. Invigilators should plan in advance exactly what they will
say in their introductory remarks and try to strike a note of supportive
encouragement. Immediately before the papers are given out, learners
should be reminded that there is to be silence now. Learners should also
be told that if anyone does not understand the wording of a question, he
or she can at any point in the exam raise a hand and the invigilator will
come and explain the language only. Otherwise, examinees should
receive no help of any sort during the exam.
Further, invigilators should decide in advance at what point or points
in the exam they will tell learners how much time they have left. When
there are only five minutes left, this should in any case be announced.
When it is almost time to finish, the invigilator should then say some-
thing like, ‘Your time is up now. Please finish the sentence you are
writing and put down your pens.’ Then, when time is up, the invigilator
should say, ‘Stop writing now. I shall now collect your scripts. Make
certain your pages are numbered and your identification number is on
every page.’ The invigilator should then immediately start collecting the
scripts. For further guidance on administering written tests, see Hughes
(2003: 215–217) and Brown and Abeywickrama (2010: 78).
If possible, candidates should be identified by numbers rather than
names on their scripts, so that knowledge of the candidate’s identity,
gender or ethnicity does not influence the marker.
Marking (Scoring) Procedures

For integrated tests such as reports and essays, it must be decided in
advance whether marking is to be ‘impressionistic’, ‘holistic’ or ‘ana-
lytic’. Impressionistic or ‘general impression’ marking means that the
marker reads the script and assigns a mark without any criteria being
explicitly stated (Weigle 2002: 112). However, usually a decision has
to be made between holistic and analytic marking. For holistic mark-
ing, a rating scale comprising usually about six bands of achievement
is set up and a brief description of the expected level of performance
for each band on the scale is formulated in writing. Markers then
assign each script to the appropriate band on the scale in the light of
these descriptions. It is important to explain the descriptors for each
band to markers so that there will be no discrepancies of interpreta-
tion among markers. For sample rubrics for holistic marking of
integrated writing tests, such as essays, see Hughes (2003: 87–89) and
Weigle (2002: 113).
For analytic marking instead of a single banded scale, a number of
specific aspects of performance are focussed on, for example, ‘content’,
‘organisation’, ‘grammatical command’, ‘lexical command’, ‘spelling and
punctuation’. For each aspect of performance, a rating scale comprising,
say, six achievement bands is set up and provided with a description
specific to that aspect. Markers then assign each script to a band on the
scale for each performance aspect (content, organisation etc.). If an
overall mark or grade is required, then grades for each performance
category are averaged. A weighting factor can be applied, however, so
that, for example, ‘spelling and punctuation’ counts for less than the
other categories (see Weigle 2002: 72–73, 112–127). For sample rubrics
for analytic marking of integrated writing tests, such as essays, see Brown
and Abeywickrama (2010: 114–115), Ur (2012: 182), Weigle (2002:
116–119) and Hughes (2003: 91–93; 95–96).
The argument advanced in favour of analytic scoring is that each
learner has his or her own profile of individual strengths and weaknesses:
a fact not accommodated for by holistic scoring (Hughes 2003: 100;
Weigle 2002: 120–121). Often a candidate is strong in some respects,
say content and organisation, and weak in others, say grammatical and
lexical command, so that it is difficult to assign him or her to a specific
holistic band. Those who favour holistic scoring argue, however, that it
is easier and perhaps more realistic to judge learner proficiency holisti-
cally than to analyse a limited number of selected performance features.
They make the point that it is by no means clear which particular
performance features are the most relevant for inclusion anyway, let
alone what their relative weightings should be. Holistic and analytic
marking can be combined by providing markers with both analytic
rating scales and a holistic rating scale. However, a procedure then has
to be set up concerning how to deal with possible discrepancies between
the two sets of scores.
For both holistic and analytic marking procedures, markers must be
provided with a scoring key or rubric, guidelines for marking and
examples of typical answers for each category on the rating scale. It
may be necessary for test setters to train markers in the procedure in
advance (see Hughes 2003: 49–50). In the training sessions, sample
answers can be discussed and it can be shown how these should be
scored. Markers also need guidance in how to treat ‘off-topic’ answers,
incomplete answers and the like. For all integrated tests, however they
are scored, at least two markers who mark independently of each
another should be used and the final mark should be an aggregate of
their two marks (Hughes 2003: 50). It is advisable to lay down in
advance that in cases of a (specified) great discrepancy between marks,
for example more than one band on a six-point scale, or a difference
between a passing and failing grade – even if this is only one band
difference – then a third marker will independently mark the paper and
an aggregate of all three marks will be taken as the final mark. If
holistic rather than analytic marking is adopted, Hughes (2003: 95)
suggests using four markers to ensure high reliability. (See Weigle
2002: 127–134 for more detailed discussion of marking integrated
written tests.)
Techniques are available for computerised marking of integrated writ-
ten tests such as essays taken online. One approach is to use the
statistical technique of regression analysis to find which quantifiable
surface variables are the best predictors of scores given by expert
markers for a fairly large subset of a set of essays (see Weigle 2002:
234–237). The variables investigated might include some of those already
mentioned in Chapter 10 for assessing linguistic difficulty, for example
sentence length, syntactic embedding, type-token ratios and lexical den-
sity. The best-fit model obtained can then be used to score the rest of the
essays in the set.
Van Moere and Downey (2016: 346) report that for the online version
of the writing section of the Test of English as a Foreign Language
(TOEFL iBT) two scores are calculated, one a computerised score and
the other a score produced by an expert human marker. Correlations
between the two scores are apparently comparable with or even slightly
higher than those between pairs of human markers.
Oral Language Tests
Techniques
If at all feasible, oral proficiency should be tested directly rather than
indirectly. Brown and Abeywickrama (2010: 205) list various direct
techniques, the most frequent of which include:
describing a picture
telling a story from a series of pictures
giving an oral presentation on a specific topic (prepared or
unprepared)
a conversation with one or more examiners (the oral interview)
a group conversation among, say, four test takers (the peer
conversation)
The first three options involve monologue, and Ur (2012: 181) suggests
picture-cued tasks are more appropriate for young learners or beginners
and an oral presentation is more suitable for more advanced learners.
(For various examples of picture-cued oral testing see Brown and Abey-
wickrama 2010: 193–100, 220–221.)
Oral presentations are not without their problems. It is a very stressful
experience to be told to speak extempore on a topic. Yet, if candidates
are given their topic well in advance of the test, they are likely to deliver
a prepared and memorised speech. It therefore seems preferable to tell
them their topic just before their test and give them a few minutes to
prepare (Ur 2012: 181). Decisions then have to be made about whether
candidates should have access to a dictionary or not. Another problem
which arises is that if all candidates are to be given the same topic or
topics, or the same set of topics to choose from, it is likely that
candidates who have just done the test will ‘leak’ the topics to candidates
waiting to take the test, who thus have more time to prepare. Yet
another problem is that it is difficult to avoid topic bias, which will
lower reliability. A single set topic will inevitably suit some candidates
better than others, but offering a choice of topics does not solve the
problem of topic bias, since complete comparability of topics cannot be
ensured. For these reasons it may be best to ask candidates to speak on
a range of topics, although this involves mini-presentations, which may
not be desired if the test aims to assess whether candidates can speak at
length. For discussion of various ways in which bias can enter a test, see
Brown and Abeywickrama (2010: 96–98).
In view of the problems involved with oral presentation, it is perhaps
not surprising that the oral interview is often preferred as a technique of
oral assessment. A further advantage is that, since most learners will
need to interact in English in the real world, the interview will often have
greater construct validity. However, the fact that the interviewer has to
assess the test taker’s performance while participating in the conversation
will tend to lower reliability compared to non-interactive oral assessment
where the tester can devote complete attention to assessing the test
taker’s performance (Ur 2012: 181). One way of obviating this problem
is to assign the roles of interviewer and assessor to two different persons
(Harmer 2015: 420).
In all oral testing involving interaction between the tester and the
test taker, the latter is always in the uncomfortable position of having
to interact with someone in the superior or dominant role position
(Brown and Abeywickrama 2010: 211). This may affect performance.
Furthermore, it is difficult, perhaps impossible, for testers not to be
influenced positively or negatively by personal characteristics of the
test taker, which lowers reliability. Conversely, test takers may be
influenced by personal characteristics of the tester and this may affect
their performance in the test. These factors include age, sex, ethnicity,
social status, variety of English spoken, accent and even style of dress,
as well as perceptions of psychological characteristics such as domi-
nance versus submissiveness, helpfulness versus unhelpfulness and
tolerance versus intolerance. There may also be intercultural attitudi-
nal differences which operate, too, particularly when men are inter-
viewed by women and vice versa. Sometimes peer conversations,
rather than the traditional interview, may help to reduce such pro-
blems, but not always.
The advantages of the peer conversation over the interview are, firstly,
that it is more economical, particularly in terms of time, than a one-to-one
interview. Two assessors can sit in with a group of, say, four test takers and
concentrate entirely or mainly on assessment, according to how much they
participate. Secondly, if both testers assess all four candidates, then two
marks are available to be aggregated for each candidate, which should
improve reliability. Thirdly, rather than just asking test takers to discuss
a topic, tasks can be set up for learners to solve in the manner of
a simulation or role-play (see Brown and Abeywickrama 2010: 214). If
these are chosen to correspond to the candidates’ likely future uses of the
language, for example in their professional lives, then the test acquires
greater validity. Fourthly, it is also claimed that, compared to the one-to-
one oral interview in which the interviewer asks questions and the test taker
gives answers, the language produced is likely to be more varied and less
constrained. This should improve construct validity since most real-life
speech does not consist solely, or even mainly, of answering questions.
However, one has to guard against the danger of unequal distribution of
speech among the test takers and group dynamics may be unpredictable
(Brown and Abeywickrama 2010: 211). For discussion of a testing project
which focussed on peer conversations involving collaborative group inter-
action, see Little and Ushioda (1996).
Marking (Scoring) Procedures
As in the case of integrated written tests, a decision has to be made whether
to adopt holistic or analytical scoring techniques or to combine the two
approaches (Harmer 2015: 419–420). For marking purposes, a scoring sheet
should be designed for use by all testers. The aim should be for testers to
unobtrusively keep score, for if they are constantly having to make elabo-
rate notes this could disturb the test taker. Anxious candidates in particular
may assume that the tester is just counting and noting their mistakes. This
problem is most acute in one-to-one interviews. If resources allow the
possibility of having a scorer and an interviewer, then the scorer can sit
out of the direct line of vision of the test taker.
Holistic Scoring
Various banded rating scales are available, for example, those devel-
oped by the CEFR, to cover the complete range of proficiency from
the near-native speaker down to someone who cannot communicate at
even a basic level in English (see Ur 2012: 12; Johnson 2018: 272–274).
For each band, a descriptor of the typical performance for someone in
this band is provided. Problems of some candidates not fitting easily
into the holistic descriptor bands, as noted for assessing written tests,
apply – perhaps even more acutely – to spoken performance. For
instance, some learners may be fluent but produce lots of errors
(under-monitoring), while others may speak largely without errors but
in a halting and hesitant manner (over-monitoring). For sample rubrics
for holistic scoring of spoken tests, see Brown and Abeywickrama
(2010: 116, 206, 217, 219), Harmer (2015: 419) and Johnson (2018:
272–273).
Analytical Scoring
In assessing spoken language it is very difficult for a tester to concentrate
on more than about six features at once, so that it is important to
carefully choose the best set of features. They should be discrete,
relevant and comprehensive. Brown and Abeywickrama (2010: 212–213)
provide an analytic scoring rubric comprising six features, namely:
‘grammar’, ‘vocabulary’, ‘comprehension’, ‘fluency’, ‘pronunciation’ and
‘task’ (= success in doing the task required). Each feature has five bands
or levels of attainment. Harmer (2015: 420) gives an example set of
features comprising ‘pronunciation’, ‘fluency’, ‘use of vocabulary’, ‘use
of grammar’, ‘intelligibility’, ‘repair skills’ and ‘task completion’ with
a six-band scale of attainment for each feature.
It is clearly important to define exactly what is meant by a particular
feature label to avoid different testers interpreting it differently. My own
research has indicated that even trained and experienced native-speaker
EFL teachers may interpret criteria such as ‘fluency’ in various ways if
these are not specified for the task at hand (Lennon 1995). For example,
if a test taker does a lot of self-correcting, some testers may perceive this
as lack of language correctness, while others may perceive it as lack of
fluency, while others again may actually double-count self-correction as
contributing both to lack of correctness and to lack of fluency. Further-
more, some testers may react positively and some negatively to commu-
nication strategies such as circumlocution, paraphrase, syntactic
reformulation or use of a series of near synonyms to convey a word not
available in lexical storage. It is also very difficult to ensure there is no
overlap among features. For example, with reference to ‘intelligibility’ in
the Harmer (2015: 420) list above, lack of intelligibility can have various
causes, including grammar, vocabulary and especially pronunciation, so
that there is some category overlap here.
Not only the feature labels themselves, but also the descriptors for
each attainment level, will usually need elucidation. To stay with fluency,
descriptions which use expressions such as ‘speaks quickly with occa-
sional hesitations’ or ‘speaks slowly with many pauses’ are by no means
unproblematic, since speed of speech is not only a function of fluency
but of other factors, including temperament and affective factors. Some
intellectuals speak slowly and deliberately and excited or nervous candi-
dates may speak unnaturally quickly. Other terms frequently used in
descriptors, such as ‘few’, ‘frequent’, ‘quite a lot of’ are not self-
explanatory and need to be defined more closely. For these reasons
there should at very least be a meeting of testers in advance and
consensus should be reached on how to interpret the feature labels and
descriptors. If time and resources allow, one or more tester-training
sessions may be advisable so that testers can listen to selected recordings
from previous oral tests, score them and compare their scores with the
scores originally given.
Test takers themselves should also be informed well in advance about
how their spoken performance is to be assessed. They may otherwise
make incorrect assumptions about what the scoring criteria are. In
particular, some candidates may assume that language correctness is the
sole criterion and adopt message reduction and avoidance strategies. In
this regard it is advisable to reward successful use of communication and
repair strategies and to inform candidates about this well in advance.
Similarly, it should be explained to learners that over-monitoring and
under-monitoring will be equally penalised and that the optimal monitor
user will be rewarded.
As for written language testing, procedures for computerised scoring
of oral tests have been developed, based on using regression models to
establish which sets of measurable features associated with, for example,
fluency, pronunciation, vocabulary and grammar, correlate best with
scores given by expert human markers. High correlations between com-
puterised scores and scores given by human markers are often obtained
(see Van Moere and Downey 2016: 347–350).
Six Rules of Thumb for Oral Testing

There are a number of rules of thumb to bear in mind when conducting
oral examinations:
Use the first few minutes as a warm-up phase which does not count
as part of the test, and make this clear to the candidates.
Try to create a friendly, unintimidating atmosphere both in terms of
the room chosen, seating arrangements and one’s own manner as
examiner. This will contribute to a more natural conversation and
discourage excessive monitoring by the examinee. It is not advisable
to waste time formally introducing oneself as examiner to each
candidate, much less shaking hands because this will only make test
takers more nervous.
Use a variety of tasks, topics and techniques to contribute to a more
rounded picture of proficiency.
Two testers (or an interviewer and a scorer) are better than one.
Use both formal and communicative criteria for assessment and
consider combining analytical and holistic assessment. Analytical
criteria should include both positive as well as negative features.
Agree in advance on the scoring criteria with any other testers and
make sure these criteria are made known to candidates in advance.
Some Alternatives/Supplements to Traditional Tests

Concern about the ethics of making learners’ future academic and profes-
sional careers hinge on a single test have led to increasing conviction that
tests are best supplemented by alternative assessment instruments (see Ur
2012: 169–171). Additionally, some alternative methods of assessment may
be less culturally biased than traditional testing. The term ‘triangulation’ is
used when a number of different perspectives or information sources are
combined for assessment purposes, for example a traditional language test,
a teacher report and peer assessment (see Brown and Rodgers 2002:
243–245). Some varieties of alternative assessment include the following.
Teacher Reports
One long-established supplement to the test is the teacher report. In the
report the teacher may state the mark expected of the learner in the test. This
projected mark may even be used in conjunction with the mark achieved on
the test to arrive at the learner’s final mark. Traditional teacher reports may
be supplemented by teacher observations on work done and activities
performed. Teacher judgements are, however, necessarily subjective.
Continuous Assessment
Another alternative to traditional testing is continuous assessment, in
which various pieces of work during the year are marked by the teacher
and these marks are amalgamated to form a final mark at the end of the
course. This may supplement a final test or replace it.
Portfolio Assessment
An approach that encourages the development of learner autonomy is
the compilation of a portfolio of work done over the course. Learners
should be permitted to include only those examples of work that they
wish to include. Portfolios may include audio and video material as well
as self-evaluation questionnaires and peer evaluations. In these ways
a composite, rounded and multifaceted picture of a learner’s progress
can be built up. For a discussion of portfolio assessment as an alternative
to traditional tests, see Brown and Abeywickrama (2010: 130–134),
Harmer (2015: 411) and Weigle (2002: 197–229).
Peer Assessment
Where the results of a test are only for classroom-internal use and do
not have any significance for learners’ future academic or professional
careers, peer assessment as a supplement to a test should be consid-
ered. The more experienced and mature the learners are, the more that
peer assessment is likely to be a viable option. It can, for example, be
used for assessing learners’ oral presentations to fellow students. Peer
assessment requires careful preparation, however, and a clear catalogue
of scoring criteria has to be agreed on in advance. One advantage of
peer assessment is that learners may be more willing to accept the
criticism of their peers than that of their teacher. In this regard, the
didactic value of peer assessment will be maximised by the inclusion
of a final reflection and feedback phase. For an experimental study of
peer assessment of essays with school learners aged 15–16 years see
Bärenfänger (2012).
Self-Assessment
Self-assessment is another supplement to testing, worthy of considera-
tion for classroom-internal purposes. Learners need practice in this
technique if they are to develop a measure of autonomy and self-
direction, to become less reliant on external feedback from exam results
and be able to monitor their own progress more accurately. Self-
assessment should ideally be a continuous process which accompanies
learning. Learners should be encouraged to reflect on their performance
and to document their reflection in written form so that a longitudinal
record of their progress emerges. Such documentation may take the form
of a learning log or diary. In particular, learners should focus on what
they can do and not just record perceived deficits. Ideally, the log should
be the property of the learner and the teacher should not necessarily
have access to it unless the learner so wishes. Self-assessment can be
combined with individual-referenced testing. For further discussion of
self-assessment and peer assessment, see Brown and Abeywickrama
(2010: 144–154).
Food for Thought

Q. 1. Look at the list of well-known test-types in the chapter again. Do
you know of any other test-types to add to the list? If so, how
would you categorise them along the spectrum of discrete-point to
integrative, and how objective or subjective is marking likely to be?
What would you regard as their advantages and disadvantages for
specific learner groups in particular situations?
Q. 2. Have you experienced as a learner, or used as a teacher, individual-
referenced testing techniques? If so, explain to the group how it
was done and say how effective or otherwise you found the
technique. If not, would you be interested in using them and do
you see any concrete applications for them in your own teaching?
Q. 3. What, in your opinion, are the advantages and disadvantages of
supplementing or replacing traditional tests with (a) teacher
reports, (b) continuous assessment (c) portfolio assessment (d)
peer assessment? Limit your answer to the specific groups of
learners whom you teach or will teach.
Q. 4. (Group task for practising teachers) Do you see any possibilities
for integrating self-assessment into classroom practice with your
own groups of learners. Do you think this would be desirable?
Q. 5. With reference to what you have read in the chapter, what
suggestions could you make to an institution which only uses
a traditional essay to assess proficiency at intermediate and
advanced levels? Or do you think this is a good idea and should
not be changed?
Q. 6. (Group task) Imagine you have to design a placement test for
a large number of adults of vastly differing proficiency levels and
whom you have to assign to three broad groups – beginner,
intermediate, advanced – in the following skills: speaking, reading,
writing. How would you do this as economically as possible?
Further Reading
Tsagari, D. and Banerjee, J. (eds.) (2016). Handbook of Second Language Assessment.
Boston, MA: De Gruyter Mouton.
Underhill, N. (1987). Testing Spoken Language: A Handbook of Oral Testing Techni-
ques. Cambridge: Cambridge University Press.
References
Bachman, L. F. (1990). Fundamental Considerations in Language Testing. Oxford:
Bärenfänger, M. (2012). Peer correction in the English language writing classroom:
an empirical study in the ninth grade of a Realschule. In Lennon, P. (ed.), Learner
Autonomy in the Language Classroom: Empirical Studies and Ideas for Teachers.
Brown, H. D. and Abeywickrama, P. (2010). Language Assessment: Principles and
Classroom Practices. Second edition. White Plains, NY: Pearson Education.
Brown, J. D. (1999). Standard error versus standard error of measurement. JALT
Testing and Evaluation SIG Newsletter 3/1, April: 20–25.
Brown, J. D. and Rodgers, T. S. (2002). Doing Second Language Research. Oxford:
Pearson Education.
Hughes, A. (2003). Testing for Language Teachers. Second edition. Cambridge:
Lado, R. (1961). Language Testing: The Construction and Use of Foreign Language
Tests. London: Longman.
Lennon, P. (1995). Assessing short-term change in advanced oral proficiency:
problems of reliability and validity in four case studies. ITL Review of Applied
Linguistics 109/110: 75–109.
Little, D. and Ushioda, E. (1996). Testing communicative oral skills through
collaborative group interaction. In Aub-Buscher, G. (ed.), The Linguistic Chal-
lenge of the New Europe: Papers from the 3rd CERCLES Conference, Hull,
23–25 September 1994. Plymouth: CERCLES, pp. 123–138.
Oxford: Heinemann.
Van Moere, A. and Downey, R. (2016). Technology and artificial intelligence in
language assessment. In Tsagari, D. and Banerjee, J. (eds.), Handbook of Second
Language Assessment. Boston, MA: De Gruyter Mouton, pp. 341–357.
Weigle, S. C. (2002). Assessing Writing. Cambridge: Cambridge University Press.
Index
Note A page no. in bold indicates reference to a Table. A page no. in italics
indicates reference to a Figure. A page no. followed by a ‘Q’ indicates
reference to a question in the ‘Food for Thought’ section.
active voice 159–160 blended learning 101, 104, 108

adaptive learning 101 blogs 103–104, 108–109
adjective comparison see inflection Bloomfield, L. 34
admission tests see language tests: Bombay TV 109
placement bonding activities 232, 234Q
adverbials 161, 163, 257–259, 281 bottom-up processing 82, 172, 214,
affective filter 12 242, 272
agony aunt 254–255 brainstorming 105, 217, 242, 251, 253,
alien paragraph 248 270, 282
alliteration 273 breathing 127–128
allophone 125–126, 146 Bristol Language Project 65
alternative assessment 319–321Q
amotivation 72 CANAL-F test 64
Animoto 109 Carroll, J. B. 59, 62–63, 65
antonymy 186 catenative verbs 153
anxiety 76–79, 86Q CEFR see Common European Frame-
approximant 134, 137 work of Reference
aptitude 59–65 chain narrative 258
archaic language 185, 269, 274 child-directed speech 3
aspect 158–160 children’s verse 271–272
assimilation 126–127, 146, 213 Chomsky, N. 3, 10, 43
attribution theory 73–74, 86Q chunks 16, 42, 82, 89
audio material see recordings class reader 276–287, 291Q
Audiolingual Method 34–37 classroom discourse 221–222, 234Q
Austin, J. L. 43 classroom interaction 90, 109,
automatisation 13–15, 113, 168, 174 116–117Q
awareness 4, 49, 85, 174, 176, 198–199, classroom roles 109–110
213, 220 CLIL see Content and Language Inte-
grated Learning
backsliding 15–16 cloze 199, 220, 259–260, 298–300
backwash 311 code-switching 20
Berlitz Method see Direct Method coefficient of reliability see reliability
Bilingualism 5–7, 23Q cognitive overload 112–114
binomial expressions 189, 201 cognitive style 81–84
324 Index
cohesion markers see cohesive devices Co-operative Langage Learning (CLL)
cohesive devices 256–257, 259, 269, 281 47–48, 54, 100
collocation 107, 187–189, 191–192, copular verbs 163
197, 200–201, 204–205Q Corder, S. P. 9–10, 44
colloquial language 11, 43 185, 213 creative language 188
Common European Framework of criterial attribute theory 183, 198,
Reference (CEFR) 52, 95, 317 203–204Q
communication strategies 20–21, cultural literacy 267
24Q, 318 Curran, C. A. 41
communicative competence see
competence dark ‘l’ 83, 126, 137
Communicative Language Teaching debates see discussions
(CLT) 42–46, 54 declarative knowledge 13, 168,
Community Language Learning (CLL) 251, 298
41–42, 98 declarative sentences 156, 161
competence 10; balanced 38; deductive reasoning 32, 82–83, 169–171
communicative 43, 306; current 114; demotivation 71–72
errors of 113; functional 53; denotation see word meaning
grammatical and lexical 62; output derogatory language 185
95; strategic 21, 80; transitional 10 determiners 157, 182
Competence-Based Language Teaching dictation 299
(CBLT) 51–53, 94 dictionaries 98–101, 106–108, 188,
competence-based syllabus see output- 191–192, 196, 199–200, 202, 204
based syllabus dictionary meaning see word meaning
complementarity 186 dictogloss 258
complex sentences 4, 166 diphthongs 124, 124–125, 129–130,
complex transitive verbs see verb 132, 132–133
transitivity Direct Method 32–33, 92–93
complimentary language 185 discussions 224
compound sentences 166 ditransitive verbs see verb transitivity
comprehensible input 12–13, 23–24Q drills 35–36
comprehension questions 299
computerised marking/scoring 314, egocentric speech 23
318–319 electronic portfolios 108
concordance programmes 107, 171 elision 126
conjunctions 4, 166, 182, 256, 259 e-mail 51, 103–104, 211, 262Q
conjunctive adverbs 256, 259 empathy 81
connectionism 16–17, 190 empirical validity 306–307
connotation see word meaning English for Academic Purposes (EAP)
Conrad, J. 6 249
consonants 125, 133–138 English for Specific Purposes (ESP) 49,
Conspicuous General British (CGB) 52, 172, 249
123, 125, 130–131, 133, 136 entailment 186
construct validity 306, 316 entrance tests see language tests:
container model see mental lexicon placement
Content and Language Integrated error see language error
Learning (CLIL) 55, 93 ESP see English for Specific Purposes
content validity 307 essays see writing
continuous assessment 52, 320, 321Q Estuary English 122–123, 138
contrastive analysis 8, 34 ethical validity 305, 307, 309
controlled processing see euphemistic language 185
automatisation exemplar-based learning 175
converseness 187
Index 325
face validity 306 Grammar-Translation Method 31–32,
facilitator 45, 50, 55Q, 109, 225, 251 36, 91, 169
false fluency see fluency grammatical words 2, 140–141, 182
false friends 35 greetings cards 256
field, tenor and mode 51, 209–210 group conversation see peer
field-dependency/independency 82 conversation
finite verb forms 152, 166, 280
first language acquisition 1–5, 23Q Halliday, M. A. K. 43, 45, 51, 209–210,
Firth, J. R. 43, 187, 191, 209 256, 259
Flickr 108 hand-held devices 108–109
fluency: and accuracy 12, 15, 110, 115, hierarchy of needs 65–66
118Q; false 15; and games 229–230; Hot Air Balloon 233–234
and good language learners 17; and humanistic education 41
language teaching methods 42, 45; Hymes, D. 43
and language testing 306, 317–318; hyponymy 186, 191
and the mental lexicon 189, 192
fossilisation 11, 167 idiomatic phrases see lexicalised
Fries, C. 34, 182 collocations
Frisby, A. W. 38 idioms 184, 189–190
ice-breaker activities 232
gambits 189 ideal self 67
games: grammar 174–175; vocabulary Illustration, Identifcation, Inferencing
199; interactive 229–231 (III) 170
gap-filling 172, 204Q, 220, 255, imperative sentences 165–166; and
299–301; see also cloze Total Physical Response (TPR)
Gardner, R. C. 66–67, 77–78, 114 171, 174
General British (GB) 122 incidental learning 4, 46, 55, 86,
genitive 34, 154–155 174, 195
genres: and language testing 303; and incompatibility 186
register 210; and teaching grammar inductive reasoning 38–39, 59–60, 62,
172; and teaching listening 211–214, 82–83, 169–171
217; and teaching literature 265, inferencing 17, 170, 216, 245, 266; see
268–269, 277; and teaching reading also top-down processing
238, 242–243, 248; and teaching inflection 150–156; of adjectives 155,
speaking 221; and teaching writing 177Q ; of nouns 153–155; of
45, 249–250, 252–255, 260–261 pronouns 155–156; of verbs
glottal stop 122, 138 150–153, 177Q
goals: and classroom interaction Information and Communication
109–110; and ICT 118Q; and Technology (ICT) 100–109
language teaching methods 32, 37, 43, information gap 45, 195, 231, 248
47, 50–55, 56Q; and learner information transfer 249, 302
autonomy/self-regulation 19–23, inhibition 12, 80, 226
24Q; and motivation 65–66, 69–70; input enhancement 195
and needs analysis 96, 98; and the input hypothesis see comprehensible
syllabus 90–91, 94–96, 117Q input
good language learners (GLL) 17–20, input modification 110
82, 86Q instrumental orientation 67, 86Q
grading of language material see selec- integrated skills 254, 297
tion and grading integrative orientation 67–68, 70, 72,
grading of language tests see marking 75, 86Q
grammar teaching 167–177; see also integrative testing see language tests
games; see also songs interactional modification 111
326 Index
interactive language learning see socio- language teaching materials:
cultural theory classification 98–99; and language
interlanguage 11; see also language lear- teaching methods 38, 41–42, 44, 47,
ner language 49–51; and learner co-operation/self-
internet research 107 direction 112; for listening 216; and
interrogative sentences 164–165 needs analysis 96; online 100–109,
intimate language 185 118Q; and project work 289; and the
intonation 142–144; and classroom syllabus 91; see also online tools
practice 146–148Q see also presentation software
intransitive verbs see verb transitivity language tests: achievement 294–295;
invigilation 312–313 criterion-referenced 302; direct versus
irregular verbs 92, 98, 151–152, 177Q indirect 297–298; discrete point
298–300, 321Q; evaluation criteria
jazz chants 147 for 305–311; formative 295–296;
Jespersen, O. 1 individual-referenced 297, 303,
joint attention 2 321Q; integrative 298–300, 319 (Q);
jokes 4, 127, 147, 188, 213, 216 norm-referenced 302–303; oral
Jones, D. 121 315–319; piloting 311–312; placement
Just a Minute 230 295, 321Q; proficiency 294, 321Q;
progress 295; revision 295; selectivity
kinaesthetic learning 40 of 303–305; summative 295–296;
Kissinger, H. 6 techniques 297–302, 315–316, 321Q;
written 311–314; see also marking; see
Lado, R. 8, 35, 305 also multiple choice
language acquisition: in bilingual learned helplessness 74
settings 5–7, 23Q; in classroom learner autonomy 19, 49–50, 112–113,
settings 7–8, 23Q; and error 286–288, 296, 320
8–11,23Q; in first language settings learner empowerment 46, 48
1–5, 23Q; and grammar teaching 167; learner identity 20, 49, 67
and language learning aptitude 65; learner self-direction see learner
and lexicalised sentence stems 193; autonomy
and private speech/self-talk Learning Management System (LMS) see
197; see also automatisation; see also Virtual learning Environment (VLE)
communication strategies; see also learning style 84–86
connectionism; see also language Lenneberg, E. 4–5
learning strategies; see also restruc- lexical access 192
turing; see also socio-cultural theory lexical density 279, 314
Language Aptitude Battery (LAB) lexical reiteration 256
63–64 lexical storage 191–192
language error: and anxious/ lexical words 140, 182, 279–280
demotivated learners 78; of lexicalised collocation 188–189
competence versus performance 10, lexicalised sentence stems 189–190,
113;and correction/feedback in the 193, 202
EFL classroom 112-118Q; and light ‘l’ 83, 126, 137
language acquisition 3, 7–11, 15, listening 211–221; critically 218;
23Q; and language teaching methods difficulties 212–213; for gist 212, 218;
36–37, 39, 41, 44–45; and language for implications 212; to live
testing 296; and speaking activities speakers 215–216; to recorded
223–224; and writing 250–252 materials 216; for significant detail
language laboratories 36 212; to songs 219–221; for specific
language learner language 10–11, 23Q, information 218; style 211–212; see
92, 167 also questions
language learning strategies 17–19, 23Q literacy 200
Index 327
literature: definition for the EFL negotiation of meaning 20–21, 183
classroom 265–266, 290Q; reasons non-finite verb forms 152–153
for reading 266–268; see also noun phrase (NP) 157
novels: see also poems; see also project noun plural 154, 173
work novels 276–287, 291Q; see also reading
lonely hearts 254 logs
Lost At Sea 229, 234Q nucleus see intonation
nursery rhymes 99, 176, 219, 265, 268;
marking: of oral tests 317–319; of see also songs
written tests 313–314
Maslow, A. H. 65–66 observer’s paradox 11
matching tasks 60, 63, 204, 248, 257, OK Corral 108
271, 299 online audio and video files 104–105;
materials see language teaching materials see also recordings
mental encyclopaedia 247, 267 online concordance programs 106–107
mental lexicon 188–193 online corpora 106–107
metaphor 184, 188, 189, 198, 201–202 online dictionaries 106–107
Meyer-Briggs Type Indicator (MBTI) online poster walls 105
74–76, 86Q online tools 103–107
Michigan English Language Proficiency oral activities see speaking activities
Standards (MELPS) 95 oral interview 315–316
mind maps 197–198 oral presentation 304, 315, 320
minimal pairs 123, 126, 133,145 oral tests see language tests
minor sentences 166–167 output-based syllabus 94–95
mistake see language error output hypothesis 21
MLAT 59–63 overextension 2, 15, 18
modal auxiliary verbs 153, 173–174 overgeneralisation 20
mode: spoken 209–211; written
236–238; see also field, tenor and paradigmatic axis 185, 188, 193, 195, 198
mode paragraph reconstruction tasks 257
model dialogue 35 paralinguistic features 123, 210
Modern Language Aptitude Test see paralinguistic strategies 21
MLAT passive voice 159–160
Monitor Model 12–13, 23Q past perfect tense 158–159, 257
monophthongs 124, 128–131 peer assessment 49, 319, 321Q
monotransitive verbs see verb peer conversation 315–316
transitivity peer feedback 22, 252, 254
morphemes 150–156, 175, 180–181, peer teaching 50
188, 198,279–280 performance slip see language error
motivation 55, 65–74, 86Q, 297; see also personality: of the learner 63, 74–81,
instrumental orientation; see also 84, 267, 305; of the teacher 55Q
integrative orientation phatic language 43, 223–224
multi-media text-creation software phoneme 123, 124–125, 129, 133, 136,
106, 118Q 144–146, 148Q; and spelling 237
multi-word units 188–189 picture books 215, 233Q, 265–266
multiple choice 61–63, 220, 246, picture stories/sequences 99, 106, 218,
298–302, 305, 311 231, 257, 260, 315
musical ability 62 pictures: and grammar teaching 170;
and ICT 102–103, 105–108; and
nasal cavity 127, 128, 133 language teaching methods 38–39, 41;
nasal sounds 128,133, 134, 135, 145 and language testing 302, 315; and
needs analysis 52, 96–98 listening 214–218; and literature-
negative sentences 164–165 based activities 270, 275, 277, 282,
328 Index
287; and reading activities 248; and opinion 299–300; factual 299–300;
speaking activities 223, 231, 233Q; and games 230–231; genuine 246;
and writing activities 257, 260–262 guiding 244, 270, 275, 283–284; and
pidgin 11; in the classroom 115 ICT 103–107; and intonation
Pimsleur, P. 63 patterns 144; and language teaching
pitch 139, 142–143 methods 33, 35–36, 39; and listening/
Pittman, G. 38 reading comprehension 299, 301; and
Plan, Perform, Reflect (PPR) 47, 112 songs 176; unanswered 312; yes/no
podcasts 104, 216 299–300; see also multiple choice
poems 268–276, 291Q questionnaires: and needs analysis
poetic language 185 96–98; and portfolios 320
polemical language 185
polysemy 184, 198 reading 238–249; aloud 240–241;
portfolios 47, 52, 108, 287, 320–321Q alternatives to traditional classses
possible selves see ideal self 248–249; cliffhanger technique 248,
post-methods era 53–55 283; difficulty 262Q; jigsaw 248;
PPP see Presentation, Practice, purpose and style/technique
Production 238–241; lesson structure 241–247;
practice-makes-perfect principle 39 sample passage and questions
pragmatic difficulty 281 244–246; word for word
pragmatic meaning 256 239–240; see also questions
pragmatics 8, 43, 45, 54 reading logs 285–287
prediction skills 105, 214, 220, 248, real world language 297–306
265, 282, 286–287 realia 100
prefabricated unit 189 Received Pronunciation (RP)
Prefixes 180–181, 188, 199, 203Q 121; see also Conspicuous General
present progressive tense 38, 158–159, British (CGB)
176 recasts 3, 115
Presentation, Practice, Production receptive skills 209, 300–302, 304
(PPP) 37–39, 169, 174 recordings: audio 99, 109, 172, 210;
presentation software 102–103 authentic versus didactic 212, 216;
private speech 19, 197 video 72, 99 102, 104–106, 109, 210,
problem-solving activities 221, 228, 219
231–232, 248, 257, 274; see also writ- reference see word meaning
ing: guided Regional General British (RGB) 123
procedural knowledge 13, 168, 174, register 45, 51, 210, 238
243, 251, 298 reliability 307–310
productive skills 209 reordering tasks 248–251
project work 287–290; and ICT 108; restructuring 15–16, 20, 168
and individual learner differences 75; rhythm 141–142
and language teaching methods risk-taking 79, 80
46–48; and speaking 222 Rogers, C. 42
pronunciation practice 116, 144–147 role-play 221, 224–228, 233–234Q,
prototype theory 183–184, 203–204Q 247, 254–255; and pronunciation
proverbs 189 practice 147
Public School Pronunciation (PSP) see rote learning 49, 59, 86
Received Pronunciation Rousseau, J.-J. 41
RP see Received Pronunciation
questions: clarification 3, 111, 211; and
classroom interaction 113, 116; scaffolding 22, 72, 216–217, 222–223,
closed versus open 246; Display 233Q, 273
221–222, 246, 262Q; and ‘do’ scanning 239, 302
auxiliary 164–165; evaluative/ scorability 310
Index 329
scoring see marking spreading activation see connectionism
Searle, J. R. 43 standard error of measurement
seating arrangements 46, 222, 225, (SEM) 309
227, 319 stock expressions 189
Second Life see virtual worlds story apps 266
selection and grading 34, 38, 39, 44 story books see picture books
self-actualisation 66 story circle 258
self-assessment 49, 319, 321Q story-telling 215, 233Q, 257;
Self-Directed Language Learning multimedia 106, 118Q
(SDLL) 48–51 stress patterns 138–141
self-efficacy 76 structuralist linguistics 34
self-esteem 66, 76 substitute frames 34
self-regulation 19–20, 22–23 subtitles 105, 109, 219
self-talk see private speech suffixes 150–156, 181, 188, 198–200,
SEM see standard error of 203Q,; and word stress 140, 146
measurement superlative see inflection
sense relations 185–187, 191, 198 suprasegmental features 210
sensitivity to rejection 76, 80 Survival English (SE) 66, 98
sentence frame 173 syllabus 90–96, 117Q: analytic 94;
sentence reordering tasks 248 cyclical/spiral 53, 95–96; grammatical
sequential monolingualism 7 91; inbuilt 10; integrated/multi-track
silent period 39 95; lexical-grammatical 92–93;
simple past tense 157, 164, 176, 257 negotiated 102; notional-functional
simple sentences 1–2, 166 44, 94; output-based 94–95;
simulation 228–229 situational 93; synthetic 94; topic-
Situational Language Teaching (SLT) based 93
37–39, 93 synapse 190
skeletal structures 173, 193 synonymy 186
skill mastery see Competence-Based syntactic transformations 299
Language Teaching syntagma 187
skimming 239, 243–244 syntagmatic axis 188, 193, 195
Skype see video communication syntax 1–3, 156–167, 192
platforms
slot and filler 173 Task-Based Language Teaching (TBLT)
social learning 48 46–47, 54, 95, 100
social media platforms 106 teacher reports 319–321Q
socio-cultural theory 21–23, 47 teaching materials see language teaching
songs: for grammar practice 175–177Q ; materials
for listening comprehension technical language 185
219–221; on video 105 TED talks 216
Spanish Armada 108, 247 teenage novels see young adult novels
speaking activities 222–224, 233–234Q; telegraphic speech 1–2
see also bonding activities; see also tenor see field, tenor and mode
discussions; see also games; see also TENOR (Teaching of English for No
problem-solving activities; see also Obvious Reason) 52–53, 249
role-play; see also simulation Tenses 157–164, 171, 176, 177Q, 257,
speech act 43, 53, 174 261, 297
speech bubbles 260 tests see language tests
speech style 3–4, 11, 14, 141, 210, test-wiseness 305
212–213, 219 text type see genres
spelling 237 TOEFL 294, 303, 314
spelling pronunciation 196 tolerance of ambiguity 81
‘spoken form first’ principle 196 tonality see intonation
330 Index
tone group/unit see intonation vowel gradation 150
tongue twisters 147 vowels 123–133, 124, 130,
tonic syllable see intonation 132, 147Q
tonicity see intonation Vygotsky, L. 22–24Q, 47, 168
top-down processing 214, 242, 272
topic sentence 259 Watch with Mother 122
Total Physical Response (TPR) weak forms 140–141
39–41, 171 webinars 106
transitive verbs see verb transitivity What’s My Line 231
translation buffer 192 wikis 104
treasure hunt 107 willingness to communicate (WTC)
triangulation 319 80, 305
trinomial expressions see binomial word difficulty 194
expressions word formation 180–182, 203Q
T-Unit 280 word meaning 182–188; associative/
type-token ratio 279 connotational 185; citational/
dictionary 182–183; contextual
universal grammar 3 182–183; denotational/referential 185
U-shaped behaviour 15–16 word clouds 105, 218
word field. 93, 173, 191,197, 200, 271
validity 304, 305–307 word grids 198
variable rules 11 word steps 198
verb complementation 161–164, 173 words with fuzzy edges 184
verb phrase (VP) 157 writing 249–262Q; for beginners and
verb transitivity 162–164 lower intermediate level 255–256;
verse see poems collaborative 22, 102, 104, 174,
video material see recordings 250–251; description and narrative
video communication platforms 105 256–258; discursive/expository
video tours 109 258–260; e-mails 262Q ; essay 252,
Virtual Learning Environment (VLE) 300, 310, 313–314, 321Q; greetings
102–103 cards 256; guided 250, 300; parallel
virtual worlds 106 260; picture-based 257, 260–262; real
vocabulary teaching 193–205Q; affixes world genres 249; sample passage
198–200, 203Q ; collocations 259; task-based 300; teaching
200–201, 204–205Q; enhancing approaches 249–255
informal acquisition 195; explicit writing systems 236
teaching of new vocabulary 195–196, written tests see language tests
203–204Q; games 199; lexicalised
sentence stems 202–203; multi-word young adult novels 266
units 201–202; practice and youth novels see young adult novels
consolidation 196–198; techniques YouTube 99, 104–105, 177Q, 216, 273
195–196, 204Q
vocal chords 127, 127–128, 133–134, zone of proximal development (ZPD)
138, 145 22–24Q, 252
vocal tract 127–128, 133 zoo task 214
voice onset time (VOT) 134
voicing 128

The Foundations of Teaching English As A Foreign Language (Paul Lennon) (Z-Library)

Uploaded by

Copyright:

Available Formats

The Foundations of Teaching English As A Foreign Language (Paul Lennon) (Z-Library)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Foundations of Teaching English As A Foreign Language (Paul Lennon) (Z-Library)

Uploaded by

Copyright:

Available Formats

The Foundations of Teaching English

An introductory textbook that assumes no prior knowledge of linguistics

Paul Lennon is Professor of TEFL, Bielefeld University, Germany

ISBN: 978-0-367-25095-9 (hbk)

Detailed Table of Contents vii

1 Language Acquisition and Language Learning 1

2 Language Teaching Methods 31

3 Individual Learner Diﬀerences 59

4 Facilitating Classroom Learning 90

8 Listening and Speaking 209

9 Reading and Writing 236

10 Literature in the Language Classroom 265

11 Language Testing 293

1 Language Acquisition and Language Learning 1

2 Language Teaching Methods 31

3 Individual Learner Diﬀerences 59

4 Facilitating Classroom Learning 90

8 Listening and Speaking 209

9 Reading and Writing 236

10 Literature in the Language Classroom 265

11 Language Testing 293

5.1 The organs of speech 127

5.1 Transcriptions for the 20 GB vowels 124

This is an introductory book about Teaching English as a Foreign

The passage from Bank Shot by Donald. E. Westlake on pages 244–245 is

First Language Acquisition Settings

 explaining how the language works (instruction)

The higher level of ultimate achievement which younger starters are

Contrastive versus Non-Contrastive Error

Global versus Local Error

 Errors are a necessary part of linguistic development and represent

… a temporary activation of nodes in a sequence that is not yet

… is triggered by appropriate inputs and then operates indepen-

Speech production necessarily involves an optimal blend of controlled

The Good Language Learner

 They focus on language both as a formal system and a means of

Strategies of Young Children in Naturalistic Environments

Strategies of Classroom Learners

Self-Regulation, Autonomy, Motivation and Identity

Communication Strategies and Negotiation of Meaning

1. Using the native language or another language. This includes:

2. Using one’s available English language resources. This includes:

4. Paralinguistic strategies of gesture, mime and facial expression (for

Socio-Cultural Theory and Interactive Language Learning

the distance between the actual developmental level as determined

Food for Thought

The Grammar-Translation Method

 The mother tongue was the language of instruction.

The Direct Method

 The target language was the language of instruction.

See Richards and Rodgers (2014: 11–14), Howatt with Widdowson

Often ‘substitute frames’ furnish a simple device for the introduc-

The syllabus for Audiolingual courses was therefore grammatically

 ‘the fundamental law of contiguity’: when two experiences have occurred

In the Audiolingual Method, emphasis was on listening and speaking

 The goal was automatic, unconscious mastery of the language.

Situational Language Teaching (SLT)

See Richards and Rodgers (2014: 44–57), Howatt with Widdowson

Total Physical Response (TPR)

explaining how the language works (instruction)

Errors are a necessary part of linguistic development and represent

They focus on language both as a formal system and a means of

The mother tongue was the language of instruction.

The target language was the language of instruction.

‘the fundamental law of contiguity’: when two experiences have occurred

The goal was automatic, unconscious mastery of the language.

The emphasis is on making learning an enjoyable experience in

Was language used to successfully transmit meaning?

The emphasis is on social interaction, the exchange of information

At the lowest levels learners can be oﬀered a restricted number of

The ultimate goal is complete learner autonomy.

‘Field’ refers to the physical and psychological setting of the

Part One (43 items), ‘Number Learning’, aims to test inductive

selective encoding (focussing on relevant information for the task)

an integrative orientation (which did not exclude recognition of

They are not looking forward to learning the language as an enjoy-

attributing success to eﬀort and failure to lack of aptitude. As long as

extroversion versus introversion

some prefer to work alone, others in cooperation with their fellows

What are the goals of the course? (goal-setting)

The prospective learners may be observed in their professional or

didactic (contrived) materials