LANGUAGE TEACHING

Language Teaching
Linguistic Theory in Practice

Melinda Whong

Edinburgh University Press

# Melinda Whong, 2011

Edinburgh University Press Ltd

22 George Square, Edinburgh

www.euppublishing.com

Typeset in Linotype Sabon

by Iolaire Typesetting, Newtonmore, and
printed and bound in Great Britain by
CPI Antony Rowe, Chippenham and Eastbourne

A CIP record for this book is available from the British Library

ISBN 978 0 7486 3634 1 (hardback)

ISBN 978 0 7486 3635 8 (paperback)

The right of Melinda Whong

to be identified as author of this work
has been asserted in accordance with
the Copyright, Designs and Patents Act 1988.
 CONTENTS

Acknowledgements ix

Chapter 1: Introduction 1
What is Language? 1
How does Language Develop? 9
Overview of the Book 20

Chapter 2: Historical Overview – Language and Language

Teaching 23
Introduction 23
Language Teaching: Shifts in Underlying Aims and Beliefs 23
Modern Foundations for Language Teaching 27
Early Approaches to Language Teaching 31
Second Language Acquisition and Language Teaching 34
Conclusion 38

Chapter 3: Language as a Biological Property 41

Language: A Generative View 41
Native Language Development 46
Second Language Development 48
Conclusion: What a Complete Theory of SLA Needs to Explain 63

Chapter 4: Language as Tool for Communication 68

Introduction 68
The Notion of Competence 70
Accounting for Meaning: A Functional Approach 72
Accounting for Social Context: A Sociocultural Approach 77
Accounting for Explicit, Procedural Learning:
A Cognitive Approach 82
One Comprehensive Framework: MOGUL 87
vi Contents

Chapter 5: Implications for Language Teaching 95

Implications for the Classroom 95
Observation 1: Exposure to Input is Necessary for SLA 96
Observation 2: A Good Deal of SLA Happens Incidentally 98
Observation 3: Learners Come to Know More Than
What They Have Been Exposed to in the Input 100
Observation 4: Learners’ Output (Speech) Often Follows Predictable
Paths with Predictable Stages in the Acquisition of a Given Structure 102
Observation 5: Second Language Learning is Variable in its Outcome 105
Observation 6: Second Language Learning is Variable
across Linguistic Subsystems 107
Observation 7: There are Limits on the Effects of Frequency on SLA 111
Observation 8: There are Limits on the Effect of a Learner’s First
Language on SLA 113
Observation 9: There are Limits on the Effects of Instruction on SLA 114
Observation 10: There are Limits on the Effects of Output
(Learner Production) on Language Acquisition 116
Conclusion 117

Chapter 6: Approaches to English Language Teaching 121

Introduction 121
‘Designer’ Approaches 123
Current Trend: Communicative Language Teaching 129
Post-Methods Era 134

Chapter 7: Putting Theory into Practice 142

Introduction and Context 142
A Post-Methods Lesson Plan 147
Discussion of the Post-Methods Lesson 151
Conclusion 162

Chapter 8: Practice and Practices – Responding to Student

Needs 164
Introduction 164
Low-Level Adult ESL Learners 164
Advanced-Level Adult EAP Learners 168
Young Learners 171
The Global Context: English as a Lingua Franca 174
Conclusion 178

Glossary 181
References 195
Index 209
 ACKNOWLEDGEMENTS

I would like to thank the students in my classes at the University of

Leeds for allowing me to try out on them many of the ideas found in
these pages, and to the staff at Edinburgh University Press who have
been so good to work with on this project. I am also grateful to friends
and colleagues who have supported me during the writing process.
Special thanks go to Cecile De Cat and Martha Young-Scholten who
have had so much confidence in this project from the start, and thanks
to John Burrell for making me put it aside once in a while. Thanks also
to Bonnie Schwartz to whom I owe my foundational training. I would
also like to thank those who have read portions of the book, provided
information which has found its way into the book or inspired me
during this project: in particular, Laura Bray, Kook-hee Gil, Randal
Holme, Chris Jenks, Julie King, Steve Kirk, Hui-ling Lin, Heather
Marsden, Bede McCormack, Diane Nelson, Roumyana Slabakova,
Hannah Sowden, Maggie Tallerman, Mike Sharwood Smith and a
very helpful anonymous reviewer. Of course, any faults, misrepre-
sentations or omissions are my responsibility alone.
 1
INTRODUCTION

WHAT IS LANGUAGE?

Like many of the most difficult questions, the answer to this question
may initially seem obvious because language is such a natural part of
our lives. But, what is language? If you find you cannot answer this
question very well, or if you find you have lots of possible answers to
this question then you are one step closer to appreciating the complex-
ity of language. As a language teacher, it is important to be able to
explain what language is. Yet language – that which we are trying to
teach – is not easily defined.
The first aim of this book is to discuss some of the many possible
answers to the question of what is language. The second is to explore
what we know about how knowledge of language develops. These
ideas will provide a foundation for the ultimate aim of this book: to
illustrate how an understanding of concepts in theoretical linguistics
can lead to a better understanding of language teaching in the class-
room. In this introductory chapter, we will briefly explore these three
aims as a preview to the remainder of the book, beginning with the
question of what language is.

The Form/Function Divide

Language can be characterised in terms of oppositions, a helpful
exercise for teasing apart its complexity. One common opposition
among researchers in linguistics is to see language as form versus
language as function. Language as form is concerned with the struc-
ture of the language itself, while language as function is more inter-
ested in the work that a language does to facilitate interaction among
people. The two views do not have to be seen as incompatible. In some
ways they are, in fact, complementary.1
Language as form is the more contemporary way of referring to a
view of language that traces its roots to what was called Structuralism.
2 Language Teaching

As indicated by the label, a structuralist view of language is one that is

concerned with the shape of language; it is a formalist view which sees
language as a system of rules and patterns. This approach is concerned
with categorising and labelling different aspects of language as a way
of understanding the patterns. This can be done from the very small
level of individual sound to the larger level of the sentence. Tradi-
tionally, this approach to language tries to break language down to its
smallest units. As a result of this reductionist approach, much dis-
cussion of language from a formalist view is concerned with detailed
and sometimes very subtle properties of language.
For example, a structuralist is interested in the individual sounds
that are possible in a language. This interest includes, for example,
phonetic descriptions, such as the voice/voiceless distinctions between
bilabial /b/ and /p/ or velar /g/ and /k/ in English. From the level of
sound a structuralist view moves to the level of morpheme to identify
the smallest bundle of sounds that have meaning. So, for example, the
word restatement has three morphemes:

(1) re-state-ment

The root, state, holds the core meaning, while the prefix re- is added to
give the meaning of ‘do again’. Interestingly in English, the addition of
suffixes often functions to derive one word form from another; so the
verb, state, through suffixation becomes a noun, statement; prefixes
do not have this ability. As noted in our example, the verb itself, state,
is a morpheme. From a formalist point of view, words themselves are
morphemes. The next level up for a formalist is the level of word
order, or syntax, which is concerned with how words combine to form
different patterns and meanings. This takes us up to the phrase and
then sentence level.
A cross-linguistic investigation of languages reveals very clear pat-
terns for word order within and across languages. Joseph Greenberg
studied hundreds of languages, looking for cross-linguistic regularities.
He established a set of forty-five properties of language that seem to be
universal (Greenberg 1963). So, for example, Universal number forty-
three says that if the nouns in a language show grammatical gender,
then the pronouns will also reflect gender. As an example, Spanish
generally uses the suffixes -o and -a to mark its nouns as masculine or
feminine, respectively; it also distinguishes between the feminine pro-
noun, ella, ‘she’, and the masculine one, el, ‘he’. Notice that this
implication is true in the direction stated. So, for example, a language
 Introduction 3

could have gender expressed on pronouns, but not on nouns – like in

English. Most of Greenberg’s Universals are implicational in this way.
This attention to universals paved the way for the development
from structuralist linguistics to Generative Linguistics. Noam Chom-
sky took this idea of universals one step further to argue that there is a
Universal Grammar (UG) that is a property of being human. Devel-
oping the ideas of the structuralists, UG is seen as a set of patterns or
rules. But for Chomsky these rules are a natural, even biological,
property of being human – from this set of rules, a person can generate
an infinite set of utterances. In both traditional structuralist linguistics
and the more current Generative Linguistics there has not been much
interest in language beyond the level of the sentence. This is one
criticism of the language as form, or formalist approach.
Adding to this criticism, some linguists argue that breaking language
down into bits may be an interesting exercise, but fails to address the
essence of what language is: a mode for communicating thoughts.
Linguists such as Michael Halliday argue that, in order to study meaning,
we need to look at the function of language. Functionalists, therefore, are
not interested in breaking language down into categories for the sake of
identifying patterns, but instead ask what work, what function an
instance of language performs. At the sentence level, a functionalist is
more interested in looking at the role of parts of a sentence. A function-
alist, for example, is interested in the connection between the following
two options for expressing the same idea in English in terms of how the
speaker is thinking about the scene in question:
(2) a. The clown chased the monkey.
b. The monkey was chased by the clown.
For a functionalist the choice is going to depend on the context
because it is the content of the message that is of primary importance,
not the form. So a speaker would choose (2a) if wanting to foreground
the activity of the clown, but (2b) if the monkey was the main
character in the event. Similarly, a functionalist is interested in how
words and phrases fit together. Consider the following examples:
(3) And then her mood changed completely – ever since leaving,
she’s been over the moon.
(4) The colleagues she left behind, by contrast, are all feeling a bit
down.
Phrases like and then and by contrast show the relationship between
ideas and help to hold them together. This is called textual coherence.
4 Language Teaching

Functionalists are also interested in how the message conveyed in

words extends beyond literal meaning. Expressions like over the moon
and feeling down in (3) and (4) illustrate how spatial notions associate
with emotion. In fact, there seems to be a cross-linguistic trend to
associate the location of ‘up’ or ‘on top’ with good, positive feelings
while ‘down’ or ‘low’ seems to be linked with negativity. Associations
such as these, between language and meaning, are the types of
linguistic features that interest functionalists.
While formal and functional approaches to language are clearly
quite different, they do not have to be seen in opposition. Instead, they
have different interests and perspectives in mind. So, one way to
answer the question of what is language is to say that language is both
a tool for communicating thoughts and emotions – function, and a
systematic set of rules and patterns – form.

Mentalist/Internal and Interactionist/External

Another opposition for viewing language is to think of language as the
internal mental process of individuals, as opposed to the external
interaction that occurs among people. This does not contradict the
above form/function divide, and in fact overlaps in some ways. But it is
also useful as another way of thinking about language.
The mentalist approach to language is often framed in terms of the
seventeenth-century philosopher, Descartes, who made a distinction
between the mind as the rational awareness unique to humans and the
body as the physical collection of cells, organs and tissue that is
characteristic of all living creatures. We can use this mind–body
dualism to distinguish between the brain, which refers to the physical
mass of neurons encased in the cranium, and the mind, which refers to
the more philosophical understanding of how mental processes work.
Recent developments in brain science can give us glimpses into the
physical workings of language in the brain. Computed axial tomo-
graphy (CAT) and functional magnetic resonance imagining (fMRI)
technology have been used to investigate which areas of the brain
respond to different types of stimuli. This has added weight to the
older observation by early brain researchers – most notably, Paul
Broca and Carl Wernicke – who found physical connections between
injuries to specific areas in brain’s left hemisphere and language
impairment. So-called Broca’s area in the frontal lobe is associated
with the more structural properties of language while Wernicke’s area
further back is known to be crucial for linguistic meaning.2 Current
 Introduction 5

technological developments may eventually lead to a more sophisti-

cated matching between linguistic behaviour and neural structure,
but, as we will discuss later, there are limitations on our understanding
of this link in current research.
The much older tradition of understanding language as a property
of the mind is a more abstract exercise that has made use of theoretical
models to explain relationships between various aspects of language.
There are two strands of linguistics that have developed different ways
of understanding how language in the mind works. Generative
Linguistics views language as a special kind of knowledge, distinct
from other kinds of knowledge. Generativists present a model of
language which separates language from other mental functions; the
separation also occurs within language itself as different aspects of
language are modelled as being distinct. This view is in direct opposi-
tion to a Cognitive Linguistics model, which sees language as the same
type of knowledge as other types of knowledge, and mental processing
as one process without special distinctions. We will come back to the
distinction between these views in Chapter 4. For now, the two views
are relevant in terms of how they differ in their psycholinguistic, or
mentalist, characterisation of language knowledge. In terms of the
brain, a cognitivist understands language processing to occur through-
out the brain using the same mechanisms as other cognitive processes,
while a generativist sees language processing as connected with
specific mechanisms in the brain dedicated exclusively to language.
Perhaps in time developments in brain science will determine which
view is more accurate. Until then, we are left having to choose between
these two opposing positions.
But how can we determine which position is correct? Mentalist
investigations explore language through two basic types of methodol-
ogy, a philosophical method and a psycholinguistic method. The
philosophical method depends on logical arguments regarding the
nature of language. Language was a concern even for Plato over two
thousand years ago in what has come to be known as Plato’s Problem:
how can people come to possess such a large store of knowledge when
only exposed to a limited set of examples? In the same spirit, gen-
erativists applied this question specifically to language, noting that all
people, whether schooled or not, have the ability to create perfectly
formed sentences (in their native language), unlike other complex
mental tasks such as mathematics, which have to be ‘learned’. A
cognitivist, on the other hand, will point out that our use of language
is dependent upon our understanding of the world around us and will
6 Language Teaching

note that schooling is vital for developing our language skills. This
type of argumentation is as old as the field of philosophy itself.
The second methodology, the psycholinguistic method, has developed
more recently. On-line processing tasks can give insight into the inner
workings of the mind. One example is a task which asks participants to
decide whether sentences presented on a computer screen are true or not.
While the participant is making their decision based on the truth value of
the sentence, the psycholinguist is actually interested in how long it takes
the participant to make that decision. Testing has shown, for example,
that the response time for grammatical sentences is faster than for
ungrammatical sentences. Thus mean response time studies can give a
glimpse into the inner workings of the mind.3 What that glimpse shows,
however, is debatable. Generativists have used psycholinguistic methods
like this to argue that language knowledge is automatic, operating
without our conscious control or awareness. Cognitivists have used it
to show that we categorise language based on associations among
concepts that come from our personal experience of the world.
Again we return to the question of what language is. For a brain
scientist, language equates to electric impulses in the brain. For a
psycholinguist it corresponds to behaviour in response to external
stimulus. Generative and cognitive psycholinguists interpret the same
responses as support for different views on how language is organised
in the mind. A different approach altogether is to view language as a
product of culture and as a tool for interaction among people. Taking
an external or interactionist point of view, language can be seen as an
integral part of culture, as a socially constructed set of practices. This
sociolinguistic view of language is more interested in how people in
society make use of language to signal identity, to negotiate interaction
and to exert power, for example. Sociolinguists are more interested in
the role of language in human interaction than in the structural proper-
ties of language divorced from context.
Other linguists see the properties of language in terms of the way
language is used in social contexts. One of the aims of Systemic
Functional Linguistics is to view language in terms of how it is used by
a group of people in a shared context. For example, a systemic
functionalist is interested in how texts can be identified by properties
of the text itself; the text of a political speech, for instance, will be
different from that of a story told to a group of children, not just in its
subject matter, but also in the language used to convey the ideas. For
systemic functionalists, the social context of human interaction will
critically define the shape that language takes.
 Introduction 7

Another linguistic approach to language which can also be seen as

‘external’ is the study of pragmatics. While a sociolinguist’s starting
point is the level of society or groups, and that of systemic function-
alists the level of text, pragmatics takes a particular interaction or
utterance as its starting point. Pragmatics is interested in what a
speaker/writer means and explores the possible range of meanings
an utterance can have depending on the context. A simple thank you,
for example, will have a completely different meaning and function if
uttered in the following situations:

(5) a. on receipt of a gift from a close friend

b. on receiving bad news from someone you do not know
very well
c. by an umpire in a tennis match at Wimbledon.

So now we can see that language covers a very broad spectrum

indeed: from the physical electric impulses in the brain to the way
individuals negotiate meaning in society. Because language spans this
internal/external spectrum, the field of linguistics has many subfields
that overlap with a number of disciplines, from brain science to
sociology, and, of direct relevance to us, to education.

Linguistic vs Metalinguistic
The field of education is important for this last opposition that we will
explore: the difference between linguistic and metalinguistic knowl-
edge. Linguistic knowledge refers to the knowledge of language that
we possess, whether we are explicitly aware of it or not. It is the
knowledge that develops without explicit instruction from a very
young age to allow us to communicate our intentions and thoughts.
Metalinguistic knowledge, on the other hand, is knowledge that an
individual has about the knowledge of language. In other words, it is
explicit awareness of linguistic knowledge. It is also awareness of
aspects of language that are part of the culture of a language group.
One type of metalinguistic knowledge closely associated with the
language classroom is the set of grammar rules that we can explicitly
teach and learn. These can be either descriptive or prescriptive rules.
Linguists hold fast to the belief that their task is to describe the
grammatical properties of a language in terms of what the speakers of
that language actually produce. Their aim is to capture language in
terms of the actual habits of its speakers without imposing any
8 Language Teaching

preconceived notions of what the language should be like. This type of

metalinguistic knowledge is known as descriptive grammar because it
captures the patterns and structures of the language that people
actually use. From a descriptive grammar point of view, it is better
to think in terms of the patterns that language users follow instead of
the more value-laden notion of grammar rules, which suggest a list of
what one should and should not do.
A descriptive grammar contrasts with a prescriptive grammar,
which is what is often found in textbooks used to teach a language.
A prescriptive grammar dictates what rules of language should be
adhered to, regardless of what the speakers of the language may
actually do in natural language production, whether part of so-called
standard speech or a dialect. The well-known example of prescripti-
vism is the admonition that no word should intervene between a verb
and the particle to in the infinitive form in English. It is unacceptable,
in other words, to ever insert an adverb within the infinitive construct,
even though English speakers do this routinely in speech and writing.
The reason for this prescriptive rule is historical; hundreds of years
ago, notions of ‘proper’ English were based on the properties of Latin.
In Latin it is impossible to interrupt the verb’s infinitive form because
the infinitive is signalled through a suffix on the verb itself; for
example, nothing can intervene between the infinitive suffix -ire
and the root aud- in the Latin infinitive form audire, ‘to hear’. The
power of tradition is so strong that some prescriptive grammarians
still impose this Latinate property on English infinitives. While this is
an extreme example, language textbooks often include ‘unnatural’
prescriptive rules, most of which are associated with the most pres-
tigious ‘standard’ forms of English.
A second type of metalinguistic knowledge is the knowledge re-
quired for literacy. In most alphabets in the world’s languages today,
there is an arbitrary relationship between letters and sounds.4 There is
no reason why two lines crossed to form a ‘t’, for example, should
correspond to the /t/ sound in English. Yet the association between
sound and letters allows us to record language on paper in a visual,
two-dimensional format. This association has to be explicitly learned.
It is possible for a person to have a fully formed grammar of a
language, with a very large vocabulary, without being able to read
or write that language. For this reason, reading and writing are not
considered ‘natural’ properties of language in the sense that they are
not abilities that normally developing children naturally acquire,
whether they try to or not. Instead, literacy-related aspects of language
 Introduction 9

such as spelling, punctuation, writing and reading must be taught and

consciously learned. This is the bulk of the hard work of the first few
years of schooling. Because of the integral role of reading and writing
in education, the ‘unnatural’ nature of literacy can be hard to accept.
Yet it is considered a feature of metalinguistic knowledge, not lin-
guistic knowledge, because of the level of conscious awareness that it
requires.5
So where does this leave us? So far we have considered a range of
ways to think about what language is. We can think of language as
lying on a continuum from form to function, or make a distinction
between internal mental processes and external social interaction. In
doing so, we include a full range of academic fields of inquiry. The
primary perspective taken in this textbook is an internal, mentalist, so
therefore psycholinguistic perspective. The reason for this is not
because of any belief that the psycholinguistic view is somehow better
or more important. Instead, it is because most of the literature on
language teaching that currently exists assumes a more external,
interactionist view of language.
Also important to language teachers is an understanding of peda-
gogical issues in general; fortunately, there are many texts on general
issues of pedagogy. This book, by contrast, is interested in thinking
about language teaching from the point of view of the more formal
properties of language itself, in hopes that teachers can be more
confident about their subject knowledge and thus make more prin-
cipled decisions when they teach. An expert teacher has their work cut
out for them. In addition to general principles of pedagogy, they need
to understand the complexity of language. Additionally, they need to
be familiar with the process of language development.

HOW DOES LANGUAGE DEVELOP?

Whether generativist, cognitivist or functionalist, linguists agree that

language develops, but are not in agreement on the question of how
language develops. By using the phrase language development I am
signalling the approach that I am taking in this discussion. Language
teaching and learning implies an interaction between a teacher and
students in a classroom. It is an activity that occurs in a particular
culture and society, and in a particular institution. Language devel-
opment, by contrast, is something that happens within an internal
mental system. Thus, by using the term language development I am
taking an internal perspective that emphasises mental processes.
10 Language Teaching

Nature vs Nurture?
The question of language development is often presented as one of
nature vs nurture. But this opposition is somewhat of a straw man
argument. It is as unreasonable to say that the development of language
is completely a matter of internal biological processes (nature), as it is to
say that language development only involves external linguistic input
(nurture). It is uncontroversial to say that both are necessary. The
debate arises when deciding on the balance between the two and how
the two interact. This nature/nurture opposition can be seen as similar
to the external/internal opposition we discussed above, but for lan-
guage development instead of language itself. Before thinking about
second language development, we will take a quick look at the native
first language context to ask how it is that a newborn child comes to
possess a fully developed language within a matter of a few years.
This question of language development was one of the central
motivators behind the Chomskyan language revolution of the
1950s. The mystery that inspired Chomsky was the problem of
how young children can develop a language system that enables them
to create an infinite number of sentences when they are only exposed
to a limited number of (often fragmentary) examples of language. For
Chomsky, the answer is that there must be some innate predisposition
for language already in the brain from birth. Nativism sees language
as an ability given by nature that exists at birth. But of course, whether
a child ends up speaking Turkish or Portuguese depends on the
language that the child is exposed to in infancy. Thus, even in the
Chomskyan nativist model, nurture in the form of language input
plays a key role. Thus it is more reasonable to talk about the nature
AND nurture debate than the nature vs nurture debate.
Most people, however, will agree that native first language devel-
opment and adult second language development seem to be very
different. So is there any use in discussing native first language
development in a book about second or foreign language teaching?
The answer is more obviously yes if you are a language teacher in a
primary school working with young children. So what if your teaching
exclusively involves adults? Even so, a general understanding of native
acquisition is still relevant. Despite the intuition that the two are
different, the distinction between native language development and
second language development (child or adult) has not been proven to
be entirely clear despite decades of research. Indeed, it may be that you
know someone (or may yourself be someone) whose second language
 Introduction 11

proficiency is seemingly native-like, yet with first exposure to the

language as an adult or late teenager. Arguably, if, as a teacher, you
want to make the best decisions about how (and what) to teach,
knowledge of all types of language development is important.

Native First Language Development and Adult Second

Language Development
Perhaps the most striking thing about native language development is
that all children manage to develop a complete grammatical system so
that they can speak the language that they are exposed to fluently –
that is, of course, unless there is some pathological reason for inter-
ference in the development process. Immediately you might find
yourself contrasting this with many of your own experiences within
the second language context. This intuitive difference is at the heart of
what Bley-Vroman (1990) calls the Fundamental Difference Hypo-
thesis: in first language development native fluency is always achieved,
while in second language development native-like fluency seems
exceptional. But the fact that native-like fluency can and does happen
makes for an interesting problem.
A second important point about native language acquisition is that
regardless of the type and amount of language that a child is exposed
to (unless entirely deprived of input), the child will develop language,
and will go through the same general set of stages that all children go
through. That children have a one-word stage that precedes a two-
word stage may not seem remarkable. Children will also develop a
large set of meaningful content words like dog and want before they
begin to use grammatical words like the or of. But less obvious is why
children will acquire certain grammatical words before others. As
shown in his now famous morpheme order study, Roger Brown found
that English-speaking children acquire grammatical morphemes in a
specific order. For example, they acquire the present progressive -ing
before they acquire the plural -s, both of which are acquired before the
regular past tense -ed and the 3rd person singular -s.6 Regular
development of this type is now widely accepted as characteristic
of native language development.
Also interesting is the regularity of errors that children make – as
well as the fact that certain types of error are never made. If you spend
time with young English-speaking children, you will notice that they
use words like breaked, falled and runned. These ‘errors’ suggest that
the child has acquired the regular rule for making the past tense in
12 Language Teaching

English. It is another stage of development altogether to discover that

the past tense -ed rule has exceptions. A perhaps less well-known
developmental error that many English-speaking children make is to
use the genitive pronoun my incorrectly where the nominative I form
is needed, to say things like My make a house.7
Contrast these developmental ‘errors’ with errors that are not made.
Children have not been found to make errors with nominative and
genitive case in the opposite direction to say things like *Look at I
house. Nor do they put the past tense -ed on words other than verbs to
say, for example, *happied, *noised or *latered.8 In other words,
native language development is not random; even the ‘incorrect’
language produced by children can often be traced to a rule in the
child’s language system, suggesting some underlying constraints on
the development of language.
These kinds of regularities are called on as support for the idea of a
Universal Grammar that limits what a language can and cannot do.
So, for example, a language will not allow a past tense inflectional
marker to be attached to a word that is not a verb. This is absent as a
property of English and absent from child language development as
well. In other words, UG constrains both what is possible in a
language and how language development will proceed. This con-
straint on language reduces the possible options, making it easier
for the child to develop language. Another striking property of native
language development is that its core properties do not seem to require
explicit teaching. And, in fact, young children seem largely unable to
take instruction on board, but instead seem to go through stages of
development at their own pace. The now famous exchange between
psycholinguist Martin Braine and his daughter, quoted in Pinker
(1994: 281), illustrates this point:

Child: Want other one spoon, Daddy.

Father: You mean, you want THE OTHER SPOON.
Child: Yes, I want other one spoon, please, Daddy.
Father: Can you say ‘the other spoon’?
Child: Other . . . one . . . spoon.
Father: Say ‘other’.
Child: Other.
Father: ‘Spoon’.
Child: Spoon.
Father: ‘Other . . . spoon’.
Child: Other . . . spoon. Now give me other one spoon?
 Introduction 13

Have you experienced this kind of response to correction with your

classroom learners? Later we will look more closely at error correction
in second language teaching. Anecdotally, however, you might be able
to relate to the father in the dialogue when you think of some of the
errors that your students seem unable to overcome. Your learners’
errors may indicate a particular stage in development in which they
have reached some kind of plateau where development seems to have
come to some temporary halt. It is also possible that a learner reaches a
limit on development. It is the nature of second language development
that learners often fossilise, reaching a point where particular aspects of
their grammar stop developing such that the same error persists.
Notice that, both here and above, I have used the word error. A
useful distinction in the study of language development is the differ-
ence between an error and a mistake. Pit Corder (1967, 1971) is
attributed with making this distinction clear. A mistake is a one-off
blunder that anyone is capable of as a slip of the tongue or, perhaps, if
suffering from fatigue or if distracted when producing language. An
error, by contrast, is deviance from the target language that re-occurs
systematically. The above example of the over-application of past
tense -ed is considered an error because it is systematic and indicative
of a rule. Similarly, the tendency for some second language speakers of
English to use the pronouns he and she interchangeably because of an
absence of grammatical gender in their own pronoun system would be
considered an error and not a mistake.
Evidence of systematic errors is one of the core properties of second
language production that has emerged from second language acquisi-
tion studies. This realisation of systematicity led to the coining of the
term Interlanguage by Selinker (1972) in order to recognise the
grammar of the second language speaker as a proper grammar with
rules and regular patterns and not a random set of utterances that may
or may not resemble the target language in question. In this way,
second language development is like first language development; they
are both characterised by a degree of systematic development. How-
ever, it is important to recognise that it is a degree of systematicity.
While general trends and stages can be found among groups of
language learners, there is a certain amount of individual variability
that also exists. While this is true for native first language and second
language development, the degree of variability among second lan-
guage learners is much higher, a point we will discuss again later.
Variability is perhaps not surprising in second language develop-
ment, especially if one is to take into account the fact that, by
14 Language Teaching

definition, a second language learner already possesses a native

language. It is only reasonable to assume that the existence of the
first language is causing some kind of interference in the development
of the second language. So we might expect learners to show varia-
bility reflecting their particular native language. This is exactly what
early second language researchers who developed the Contrastive
Analysis Hypothesis thought. Contrastive Analysis is based on the
idea that the difficulty and ease of learning a second language depends
crucially on the extent of difference and similarity between the first
and second languages in question. Though, as we will discuss further
in Chapter 3, this hypothesis did not show itself to be entirely valid
when put to the test empirically, it has led to other developments in the
field. For example, the practice of Error Analysis supported the
important finding that the interlanguage of second language speakers
is systematic, thereby signalling universal properties, whether as a
result of the universal nature of language or universals in the process
of language development.
The other issue, already touched upon, is the difficult question of age
and second language development. Much has been made of the idea that
there is a critical period for language development. While the idea enjoys
much general agreement, there is very little consensus on the specific
details. Even establishing a particular age for the beginning and end of
the critical period has proven to be very difficult. There is a recognised
notion that the end of the critical period corresponds to puberty. Yet
many have been quick to point out that a meaningful connection
between language and puberty would be very strange indeed. The
seminal work of Johnson and Newport (1989) was for a time taken
to be evidence that the age of 7 was the cut-off for differences in language
development. But this age was quickly challenged by Bialystok and
Hakuta (1994) who reanalysed Johnson and Newport’s findings to
argue that the data indicate 20 to be the crucial age. Others have found
that the result may depend on what aspect of language is being tested.
Native-like pronunciation, for example, seems to require much earlier
language exposure than basic word order. But even so, no specific age
has been established for any particular aspect of language.
There has been much discussion of this idea of a critical period for
language development and many questions still remain to be answered.9
One generalisation that emerges from the research is that younger is
better. However, even this claim needs qualification since in instructed
settings it seems to be true in terms of what ultimately can be achieved,
rather than the amount of language development that is achieved when
 Introduction 15

young. It is also clear that the tendency for language development to be

more successful with youth is just that: a tendency. There does not appear
to be any single cut-off point. All studies show a distributed range that
spans from child to teenage to adult age ranges in a gradual way. We will
return to this discussion of critical period in Chapter 3.
Language development, then, is a process that occurs within the
cognitive system of the learner, whether as a first and native language,
or a second or foreign language. We have not mentioned the situation in
which a young child develops two languages simultaneously. For our
purposes, bilingualism will be considered qualitatively the same as native
monolingualism, the only difference being that two language systems are
developing concurrently. In contrast to bilingualism, the main interest in
this text is second language development, which refers to the develop-
ment of an additional language after (at least) one other language is
already a fully developed system. The question of child second language
acquisition will be only briefly explored in this text, though there is a
growing body of work showing this to be an area deserving of recogni-
tion in its own right.10 It is time now, however, to turn briefly to more
practical questions of relevance to the language teacher.

What are the Implications for Language Learning and

Teaching?
So far, this chapter has been concerned with ideas that may seem quite
abstract. Though I cannot imagine any language teacher thinking that
a discussion about language and language development is irrelevant to
language teaching, I can imagine a teacher wondering what the
connections are exactly between theory of language and the language
classroom. It is the main aim of this textbook to articulate these
connections. A flip through the table of contents will show that the
text initially concentrates on theory, gradually moving to issues of
classroom practice as the book progresses. But rather than make you
wait until midway through the book, let us briefly preview how we
might put theory into practice in terms of the ideas about language
and language development that we have encountered so far.
We have seen that language can be characterised by certain opposi-
tions. Hopefully, you will agree that there is no single ‘correct’ view of
language. But instead, by approaching language from different angles,
we can get a better sense of what it is, in all of its complexity. We can
then be more intentional about what aspects of language we are trying
to teach at any given time in the classroom. Let us begin with language
16 Language Teaching

as form. What implications might this have for you as a teacher?

Taking a structuralist point of view, you might view language as a
body of knowledge to be passed on to your students. This would take
the form of rules, paradigms and organised patterns. But surely this is
an outdated view of language, you might say. Perhaps, but if you think
of the linguistic/metalinguistic opposition, you might decide that there
is some value in trying to characterise the language into neat sets of
forms with clear and simple rules in order to develop your students’
metalinguistic knowledge, which they can tap into for more accurate
language production.
Another form-based view of language is the generative view in
which language is natural, biological. Thinking of our discussion of
nature/nurture, though it may be controversial how much ‘nature’ is
relevant in second language development, keep in mind that even first
language development is said to depend on external input (nurture). A
form-based generative view might encourage us to expose our learners
to some rich, natural input instead of just presenting learners with
simplified texts, as is often the practice, especially of textbook writers.
We might also be careful not to engage too much in teacher-talk in
which we over-simplify, or foreigner-talk in which we speak in a non-
target way. While we may think that this will aid in comprehension in
the moment, it certainly does not model the kind of language we hope
our learners to achieve in time.
Instead of a form-based view of language, we can take a view of
language as function. With functionalism in mind, we would present
language in terms of meaning, rather than grammatical rules. This
may involve establishing contexts for language and devising mean-
ingful tasks for learners which require the active use of language. A
functional approach to language teaching may have the same target
linguistic aims as a form-based approach, but will teach functional use
of language. To take the example from the above discussion, passive
and active sentence forms in English could by taught by presenting
learners with different scenes that highlight the different actors in the
event. Whether the grammatical properties of the two types of
sentences are ever explicitly taught may depend on whether this is
appropriate in terms of learner needs or needs imposed by a curri-
culum. The aim of this book is not to endorse any one view of
language teaching, but to clarify what the different positions are with
regard to language so that you have a principled basis from which to
choose what is most appropriate for your learners.
But how can we know whether we should draw on one view of
 Introduction 17

language over another in our classroom practice? As long as the

different views of language exist among theoretical linguists, it is
prudent for teachers to take on board all views for which there is
robust evidence. Thus, to continue with our active/passive example, a
thorough teacher may introduce the active/passive distinction through
contrasting differences in context before then explicitly drawing
attention to the grammatical differences between the two forms. In
order to reinforce the grammatical difference, students may be en-
couraged to practise by manipulating one form into the other through
traditional exercises in order to develop metalinguistic knowledge.
You might then have your students situate the contrasting forms by
discussing scenarios which naturally call for one form over the other in
order to put the new knowledge into use. Alternatively, the grammar
could be emphasised first and the meaning difference highlighted
second. A third option is to avoid explicit discussion of the gramma-
tical forms altogether and instead to focus exclusively on the func-
tional differences between the active and the passive. What matters is
that you recognise the differences between the options, and decide
how to teach the passive based on your understanding of language,
coupled with the needs of your students.
If you are thinking about students’ needs, you may be wondering
what we should make of the internal view of language as a mental
process. While it is dehumanising to think of our learners merely as
brains firing electric impulses in response to external input, it may be
worth acknowledging this aspect of language development to some
extent. If the brain is functioning by responding to external stimuli, it
only makes sense that we would stimulate the brains sitting in our
classrooms through plenty of examples and the repetition of examples,
but within a meaningful context if we want our students to remain
engaged. Hopefully, this will help mental structures to develop.
Considering the cognitive perspective which sees all knowledge
(including language) as linked by association, we would do our best
to present new language by making explicit links to previous or
existing knowledge. In short, a mental, internal view of language will
encourage a teacher to allow for and foster mental processing. This
view also tells us not only that active learning involves language
production on the part of the learner, but also that comprehension is a
critical part of language development.
Yet reception is clearly not enough. Our learners must also be able to
use language. The external view of language will urge us to focus on the
extent to which language is connected to social context. This view will
18 Language Teaching

remind us that the type of language we use changes depending on where

and with whom we use it. To continue once again with our example of
the active/passive, we may discuss with our students a genre of language
that makes regular use of the active/passive distinction. If we are
teaching students in higher education, for example, we might point
out that academic writing has traditionally taken advantage of the
passive voice to focus the action of the sentence while de-focusing the
actor, especially if the actor is the writer him/herself. This focus on the
active/passive in academic writing then allows for a discussion of how
academic writing is changing as academics are beginning to accept the
post-structuralist view that the writer is an integral part of the research
and should therefore be acknowledged as such instead of being ‘hidden’
from the reader. This kind of socially situated discussion will allow
students to participate in the target culture more effectively.
To summarise so far, a good language teacher will keep in mind the
full range of what a language is in order to foster the development of
all aspects of the language – that is, if native-like fluency is the ultimate
aim. It is also possible, however, that the teaching context is one in
which native-like fluency is not expected. Sometimes people need to
learn a language for a single specific purpose. A worker in a call centre,
for example, may need to have a competent grammar, a highly
developed specialised vocabulary and an awareness of culturally
conditioned telephone manner. Whether working towards complete
fluency or specialised skill, however, an ability to isolate and foster
particular aspects of language is valuable in teaching.
The flip side of teaching is, of course, learning. One would do well
also to keep in mind what we know about language development.
Even though the ideas presented so far need more discussion, we can
still draw some general implications for classroom teaching. The
extent to which UG is relevant to second language learning is con-
troversial. Yet it is uncontroversial that the target language you are
trying to teach can be characterised by grammatical principles, which
many linguists view as constrained by UG. Thus it is at least valid to
view language itself as a natural, rule-governed system, and to trust
your intuitions about language as you teach.
One common frustration for teachers is that, no matter the amount
of correction, some students continue to make the same errors. Instead
of being frustrated, you should first ask yourself whether it is an error
or a mistake. A systematic error might be traced to some universal
property of language, or it might be a property of the learner’s native
language. For example, a native speaker of French may consistently
 Introduction 19

place adverbs after the verb in English to say, for example, I finished
quickly my essay because this is the correct word order in French.
Mistakes, by contrast, as a natural part of language production, may
not be worth worrying about, especially if you are interested in
promoting fluency among your learners. High rates of mistakes,
however, may signal that the learning environment is causing undue
stress to the learner such that they are not able to perform to their
highest potential. If this is the case, you might need to consider making
changes to your classroom environment.
When viewing errors as a reflection of a student’s underlying inter-
language, it is useful to bear in mind the concept of UG because
interlanguage grammars are also ‘natural’ grammars that are restricted
to universal possibilities of language. After identifying an error, you
might try to determine whether it has perhaps occurred because the
student is in a stage where the language point in question is too advanced
for the learner. If so, then it may be best to let the error go unremarked,
knowing that this language point is not likely to develop until a later
stage in the learner’s overall proficiency. Alternatively, it may be that the
error persists, even though the learner is certainly proficient enough to
develop this aspect of language. This suggests that the learner has
fossilised where this particular aspect of language is concerned. In this
case, it may be useful to point this out to the student and to work together
to develop strategies to compensate for this fossilisation. For example,
the learner could be encouraged to edit any written work routinely for
the particular error in question. In spoken language, explicit monitoring
of production may be possible for very disciplined learners.
Whether the work of UG or not, second language learning holds to
a degree of systematicity. It may be useful, therefore, to look for areas
in which your students’ interlanguage indicates that the underlying
system is making use of a non-targetlike rule causing them to make a
particular error. If a non-targetlike rule is identified, it may be that in
time, with sufficient positive input, the student may overcome the
error. If the learner’s linguistic system is not able to develop in this
way, it might be possible to compensate by making use of their
metalinguistic knowledge. While a native speaker does not need
explicit awareness of underlying language rules, awareness of a rule
may help a non-native speaker to self-correct for more effective native-
like fluency. We will return to this discussion later in the book.
Finally, we ask what implication there is for the critical period
hypothesis. It may seem that there is little you can do in response to the
critical period generalisation that younger is better – unless you are a
20 Language Teaching

policy maker in a position of considerable power who is able to make

language learning part of the primary school curriculum. However,
keeping this younger-is-better tendency in mind will help you to
modify your expectations since young learners do not develop lan-
guage ability as quickly as older learners. This is probably to do with
higher levels of cognitive ability which develop with age. This also
provides guidance on questions of type of input. If you are working
with younger learners, you may want to provide ‘natural’ authentic
input in as large amounts as possible. If, on the other hand, you are
teaching at the university level, you may want to make more use of
explicit rule-based instruction to take advantage of the adult ability to
enhance production by making use of metalinguistic knowledge.
The implications drawn from our discussion so far have been
summarised in Table 1.1.

Language Implication

Form (Structuralist) Teach language as a set of rules

Form (Generativist) Nurture natural language development with rich input
Function Give attention to meaning
Internal/mental Emphasise processing and comprehension
External Teach cultural norms and expectations
Linguistic Provide rich authentic input
Metalinguistic Clarify language rules explicitly

Language development Implication

Nature Attend to the quality of input

Nurture Provide a sufficient quantity of input
Fossilisation Develop strategies for compensation
Systematic errors Identify underlying causes
Language mistakes Consider the classroom environment
Stages of learning Do not expect abilities far beyond current proficiency
Age Provide appropriate types of input

Table 1.1 Implications for language teaching

OVERVIEW OF THE BOOK

In this chapter we have considered the range of what might be included

under the broad category of ‘language’. At the most basic level, language
form includes the properties phonology, syntax and morphology. The
function of language, however, requires thought and memories as well
as physical motor control to receive and produce speech, along with an
 Introduction 21

awareness of the context in which language is needed for interaction. In

the remainder of this book we will explore these ideas in more detail and
then show that an understanding of language and language development
can help us to make decisions about teaching. Before we do this,
however, we will discuss the field of language teaching from a historical
perspective in Chapter 2, in order to trace the relationship between
language teaching and understandings about language.
Chapter 3 explores the biological view of language in much more
detail, discussing seminal research and presenting a set of generally
agreed generalisations that have come out of research in second
language acquisition from this perspective. In Chapter 4 we survey a
range of views about language which compensate for major omissions
in the biological approach, including the importance of language as a
tool for communication. We return to our set of generalisations from
second language research in Chapter 5 to begin discussing implications
for language teaching. In Chapter 6 we turn to the classroom, exploring
a range of approaches and methods for language teaching. This sets the
scene for an actual lesson plan, which is presented in Chapter 7. The
lesson is deconstructed in order to illustrate how it reflects theoretical
understanding of the complex nature of language, and facets of
language development. The final chapter of this book further explores
issues of language development by adapting the lesson from Chapter 7
to different contexts. In this way, Chapter 8 brings together some loose
ends, and allows for a summary of the importance of theory in practice.

FOR DISCUSSION

1. What did you think about language before you read this chapter?
Which of your existing views were supported and reinforced by
this chapter and which have been challenged? From the point of
language teaching, do you find some of the views more appropriate
and relevant than others?
2. Can you think of more examples of prescriptive grammar rules
which would contradict a descriptive account of what English
speakers actually do? Which type of grammar should be taught?
Discuss the extent to which teaching a prescriptive vs descriptive
grammar is appropriate in different teaching contexts.
3. Based on your personal understanding and experience, compare
and contrast native first language acquisition and adult second
language development. Are they entirely different and unrelated?
Or are there any commonalities?
22 Language Teaching

4. Recall a language teacher from your past experience. What view(s)

of language did they seem to hold? Think about what kinds of
activities you did in class. Did they focus on the structure/form of
language? Or the meaning/function? Was the teacher more con-
cerned with mastering language rules? Or did he/she prioritise the
culture associated with the language?
5. Make a list of all the different aspects of language that have been
discussed in this chapter. For each one, consider the extent to
which you feel as if you had to be explicitly taught and consciously
learned it in your native language. Then go through the list again
and consider each aspect in terms of whether it needs to be
explicitly taught in a second or foreign language classroom.
6. Find a language lesson. (If you do not have access to language
textbooks, try finding a language lesson on the Internet.) Consider
the lesson in terms of the oppositions about language that we have
discussed in this chapter. Where do you see evidence for each view?
Do you see evidence for any other ways to think about language
that we have not mentioned in this chapter?

NOTES

1. For an extensive discussion of the formalist versus functionalist per-

spectives, see Newmeyer (1998).
2. For a general overview of the brain, see Carter (1998/2000). For an
overview of the brain and language, see Ingram (2007).
3. For a classic and accessible introduction to psycholinguistics, see Aitch-
ison (1998). For more recent volumes written at a more technical level,
see Steinberg and Sciarini (2006) or Traxler and Gernsbacher (2006).
4. Note that originally most alphabets had some connection to the sounds
or meanings that they represent. See Coulmas (1989) or the edited
volume by Daniels and Bright (1996) for extensive discussion.
5. For a detailed discussion of literacy, see Holme (2004).
6. To read the original study, see Brown (1973). This well-known study is
described in most textbooks about first language acquisition. See, for
example, Ingram (1989). For a good collection of core readings in first
language acquisition, see Lust and Foley (2004).
7. Nina, age 2;0, as reported by Vainikka (1993/94).
8. The asterisk is used by linguists to show grammatical unacceptability.
9. For overviews of the issue, see Birdsong (1999), and Harley and Wang
(1997).
10. See Haznedar and Gavruseva (2008) for a recent collection of research in
child second language acquisition.
 2
HISTORICAL OVERVIEW – LANGUAGE AND
LANGUAGE TEACHING

INTRODUCTION

Language teaching is not, by any means, a new endeavour. In this

chapter we will take a brief look at the history of language teaching in the
European context. The theme of this historical review is that teaching
reflects the underlying beliefs of time and place, from beliefs about
education to beliefs about language itself. As our brief review of history
before the twentieth century will show us, many of today’s questions and
problems are very old indeed. There is evidence, for example, that
educationalists of fifteenth-century Renaissance Italy argued for the
importance of teaching language function in addition to language form.1
Looking at the more recent past, our discussion of the 1900s will focus on
the development of the academic disciplines that inform language
teaching today. A look to the past reminds us that many of the challenges
facing language teachers have not changed. What has changed, however,
is the way in which the challenges are understood.

LANGUAGE TEACHING: SHIFTS IN UNDERLYING AIMS AND BELIEFS

In thinking about language teaching in the past, it is important to keep

in mind the fact that systems for education were very different from
today. For instance, the specialisation of academic subjects into the
range of disciplines that we have today is itself a relatively recent
phenomenon. Language teaching for much of European history would
have been seen as a part of philosophy, which was itself subsumed
under theology.2 This is not an unnatural place for language if one
considers the legacy that can be traced back to Ancient Greece and
Rome to view language as a system of logic and rhetoric. As part of this
legacy, the languages that were deemed a necessary part of formal
education in the European Middle Ages were the classical languages.
24 Language Teaching

Knowledge of Latin was needed in order to study the classics of Ancient

Rome. Latin was also the language used not only for recording knowl-
edge of the past, but also for developing knowledge in theology,
medicine and law. The prestige of Latin was also maintained by the
role it played in the political powerhouse of that time, the Church.
Because Latin was not a native language in most of Medieval Europe,
it would have needed to be explicitly taught and learned. In fact, there
was a connection between changes in the use of Latin and the way it was
taught. Before the 1500s, Latin was the language not only of legal
documents, but also of educated people in the lands which correspond
to modern-day England, as it was in much of Europe. Though it was not
a native language, it was used for formal communication. Thus the
teaching of Latin included an emphasis on the ability to use Latin; in
other words, there was a functional approach to teaching Latin.
As the role of Latin in society changed, however, this functional
approach to teaching it also changed. When Latin was no longer spoken,
the practice of teaching Latin was limited to teaching the rules of the
language. Even in the distant past, then, there was a link between the
teaching of language and the needs associated with that language. Today,
when describing past teaching practices, we often assign the single un-
changing label ‘traditional’, and traditional teaching methods are often
equated with learning grammar rules and committing grammatical
paradigms to rote memory. Yet it is erroneous to label all of language
teaching historically as the same. While grammar instruction may char-
acterise language teaching practices in the more recent past of, say, the
nineteenth century, it was not like this throughout the Middle Ages.
Grammar teaching itself has also seen variability. The question of
whether to teach grammar inductively, encouraging learners to work out
rules themselves, or deductively, by presenting the rules for practice, also
has a long history. There was a shift in the Middle Ages from an inductive
approach to a deductive one. With the Renaissance, however, the
emphasis on discovery and exploration saw a move back to an inductive
approach to grammar again. In short, the past should be recognised as
undergoing constant fluctuation and change, with historical changes
leading to changes in language teaching. The growth of nation states,
with its beginnings in the 1300s, highlighted linguistic differences and
identity. With the rise of a French-speaking political powerhouse in
Europe, for example, came the use of French among English-speaking
ruling classes. The legacy of this French and Latinate influence is still seen
in modern-day English, especially at the lexical level.
The growth of a merchant class, improved modes of travel and
 Historical Overview – Language and Language Teaching 25

increased contact between peoples also led to different needs with regard
to languages. Thus, while Latin and French may still have held value
among the clergy and the ruling classes, the need for communication
between members of the middle classes within Europe was on the rise.
And in time, education became more and more open to members of
society outside the privileged classes. With these changes came a shift
away from Latin as the language of power, and with the interest in
culture during the Renaissance came an interest in local languages, or
vernaculars. As a result, while Latin was still taught in Great Britain, for
example, it was increasingly done through the vernacular of English.
Another factor affecting education has been innovation in technol-
ogy. The development of the printing press impacted education, as it
extended the domain of literacy to a much larger proportion of the
population. Along with this increase in printed documents in general,
the use of Latin was becoming restricted to writing only. In time, the
vernacular replaced Latin in official documents, a trend which oc-
curred across the whole of Western Europe. English did not replace
Latin as the official language for legal documents in Britain until the
mid-1600s. But tradition dies hard; for centuries after its use became
obsolete, knowledge of Latin was still held in high esteem.
The legacy of Latin also played a crucial role in the development of
linguistics in terms of understanding of how language works. Until the
1500s, Latin and Ancient Greek were the only languages whose
structure was explicitly known, analysed and written down in terms
of their grammatical rules, at least in Western Europe. In the Arab
world, linguistic study included both these Classical languages and
Arabic and other Asian languages. Yet in Europe, even when scholarly
interest began to recognise vernaculars, Latin was the benchmark from
which other languages were analysed and studied. So strong was this
influence that remnants of the Latin legacy still exist in English today,
the famous example being the prescriptive rule not to ‘split’ an infinitive
in English, which we discussed in Chapter 1. This rule illustrates the role
of Latin in traditional grammarians’ notions of ideal forms of language.
The shift from use of prestigious language to the vernacular is the
context in which the approach to language teaching known as Gram-
mar Translation developed. As Latin was not a language for interac-
tion, teachers did not use Latin as the medium of instruction, using
instead the native vernacular to explain how Latin works. This use of
the vernacular meant that Latin was taught in terms of its grammatical
rules and the aim limited to the ability to read Latin. From this came the
practice of translating passages from Latin into the vernacular, with
26 Language Teaching

teaching limited to the rules and structural paradigms of the language.

Also central to the Grammar Translation method was the idea of
transmitting cultural values to be found in the Classical literature being
translated. In other words, Latin taught through Grammar Translation
was more of an object of study than a tool for communication. In time
Latin itself became limited to a system of grammatical rules – a far cry
from the communicative and political tool it once was.
As the need to teach other European languages as foreign languages
developed, it is only natural that teachers looked to existing language
teaching practices, applying them to the teaching of other languages.
Thus Grammar Translation was adopted as the way to teach the living
languages of Europe, including English.3 It is perhaps not surprising that
this method achieved only limited success. After all, teaching a language
by translating literature and memorising structural rules is not likely to
result in fluent speakers or users of language. And by this time, grammar
teaching had returned to a deductive, rule-based approach. Addressing
the limitations of Grammar Translation in the late 1800s was the so-
called Reform Movement. In France, educationalists such as Gouin and
Marcel proposed breaking language into the four skills and focusing on
everyday language. In looking for a more effective approach, reformers
looked to developments in linguistics and psychology to establish a link
between language teaching and these newly emerging academic fields.
English reformer Henry Sweet, in his 1899 book, The Practical Study of
Languages, also identified language points to be taught and presented
them in terms of the four skills, sequencing them from simple to complex.
As we will explore in the next section, the idea of breaking down the
language into constituent parts can be traced to the practice of the early
linguists, while sequencing can be attributed to ideas of knowledge
development emerging from psychology.
Another influence on the Reform Movement was the observation
that children learn their native language quite early and easily. Noting
that infants acquire their native language in a seemingly effortless
fashion, applied linguists of the Reform Movement devised a radical
departure from existing methods, advocating the Direct Method, in
which instruction in the language classroom was limited exclusively to
the target language. Students were asked to restrict themselves to the
target language only, developing the language word by word. With
this came a return to inductive approaches to grammar learning as
students were expected to work out the patterns and rules underlying
language based on the real examples of language to which they were
exposed. The most famous example of the Direct Method is the Berlitz
 Historical Overview – Language and Language Teaching 27

School, which still exists today. It succeeded as a private language

school because its clientele tended to be highly motivated adults with a
specific need to become fluent quickly. This method, however, was
wholly unsuccessful in state schools with large classes of children who
were not likely to see the need for multilingualism and therefore
viewed language as just one more subject to be learned.
In looking at the history of language teaching in Europe, one theme
that emerges is the long history of some core questions which still
plague the field of language teaching. Another is the relationship
between development of thought in academia and movements affect-
ing practice in the language classroom. In the next section we explore
the scholastic foundations of language teaching with the development
of linguistics, psychology and the more recent fields of first and second
language acquisition.

MODERN FOUNDATIONS FOR LANGUAGE TEACHING

Scholars in Western Europe and North America in the late 1800s

became increasingly interested in the properties of language. The
legacy of the Classical languages, especially in Europe, fostered the
view of Latin as the most logical, and therefore superior, language and
linguists tended to use the structural properties of Latin as the norm
for studying other languages, imposing the rules of Latin on to other
languages. Because of differences between languages, the resulting
rules were, unsurprisingly, quite unnatural. None the less, language
lessons at this time largely involved learning rules like these, rules
which make up a prescriptive grammar.
Another change in the study of language in the eighteenth and early
nineteenth centuries came as a result of British Imperialism. Contact
with Asian cultures led to a great interest in Asian languages, espe-
cially as the British discovered a wealth of knowledge and scholarship
about Sanskrit in Asian linguistic traditions. This led to great efforts to
trace the relationships between languages, looking to identify family
resemblances among so-called Indo-European languages.
Thus there was shift in thinking away from the view of Latin as the
sole benchmark for other languages to an interest in a wider range of
Classical languages, which also led to an interest in studying languages
in their own right. From this came a new aim of describing language as
it is actually used. This commitment to a descriptive grammar is what
set the structuralists apart from earlier linguists. Ferdinand de Saus-
sure, the ‘father’ of structuralism, is noted for making the distinction
28 Language Teaching

between langue, a language, and parole, how language is used. The

basic premise of structuralism is that there are regular patterns within
any language and across all languages, an idea which was founda-
tional for the modern idea of universal properties in language. Because
their focus was on describing language as it is used, the structuralists
were mainly interested in spoken language.
There were some differences between the structuralism developed in
the US by Bloomfield and that developed in Europe by Jakobson of the
Prague School.4 The European structuralists were continuing a line of
inquiry handed down from Classics, and building on the work of Asian
linguists as they tried to trace the histories of the world’s known
languages, looking at similarities and differences at the level of linguis-
tic structure. They are perhaps best known for their contributions to the
study of phonology with the fundamental observation of phonemes as
sets of distinctive features such as voiced/voiceless, for example.
The North American effort was influenced by anthropologists like
Franz Boas who were interested in the languages of the indigenous
peoples of North America. This approach was also based on a struc-
turalist interest in language systems which reduced language to smaller
linguistic units, but with less of a legacy from the Classical languages of
Latin and Greek than linguists in Europe. Leonard Bloomfield was the
most influential figure in the development of structuralism in America
with his scientific approach to linguistic analysis. Despite difference
between European and American linguists, their approach was largely
the same: to create inventories of linguistic units, both sounds and
morphemes, with a broader aim of classifying languages into families.
A significant development at this time was the establishment of the
International Phonetic Association, whose early achievement was
devising the International Phonetic Alphabet, the IPA, in 1886. This
system for characterising all possible sounds in any language is typical
of the structuralist approach in two ways. It recognises sounds as
universal properties – any language could potentially make use of any
of the sounds. And it organises the sounds according to their structural
properties. The consonants, for example, are organised in terms of the
way they are produced, known as manner of articulation, and physical
position during production, or place of articulation.
The structuralists laid the foundations for other developments in
linguistics. Linguistic typology saw the work of Joseph Greenberg and
his forty-five Universals based on a study of thirty languages. Like the
example in the last chapter, many of his universals have survived the
test of time. To give another example, Greenberg observed that if
 Historical Overview – Language and Language Teaching 29

objects come before verbs in a particular language, that language is

also likely to have post-positions instead of prepositions. Take, for
example, Korean, an SOV language in which verbs follow their
objects. In Korean a prepositional phrase like in the house is expressed
as a post-positional phrase, jip-eh, with jip meaning ‘house’, and -eh
meaning ‘in’. Greenberg’s work can be seen as step towards the idea of
universal principles in language.
Taking a more decisive step in the direction of universals was the most
influential linguist of modern times, Noam Chomsky. Chomsky’s notion
of Universal Grammar developed out of this idea that all languages abide
by a single set of properties unique to humans. Of course, there is
variation from language to language, but that variation is itself con-
strained to a principled range of linguistic possibility. The fundamental
difference between Chomsky and the structuralists was his interest not
just in describing linguistic forms, but also in the linguistic operations
inherent to language. These operations were originally known as trans-
formations. For example, in order to form a yes/no question in English,
there is a transformation from the declarative to the interrogative form.
This involves an inversion of the subject and auxiliary verb; a sentence
such as John is going to London, for example, becomes Is John going to
London? Every language was understood to have a set of principles that
controlled transformations. These linguistic principles are said to be
generative because they allow for the infinite creation of structures. It is
for this reason that Chomsky’s approach came to be known as Trans-
formational Generative Grammar.
The basic principle constraining transformations is that they depend
on structure. Structure dependency is a key principle in formal linguis-
tics. Consider the above yes/no question transformation for illustra-
tion. If you had to come up with a rule that captures the relationship
between the declarative and interrogative forms, you might say that it is
a simple matter of moving the first verb to the front of the sentence. Try
turning a more complex example into a yes/no question, however, and
you will find that this ‘logical’ rule no longer works:

(1) a. John, my friend who was an artist for many years, is

going to London.
b. *Was John, my friend who an artist for many years, is
going to London?
c. Is John, my friend who was an artist for many years,
going to London?
30 Language Teaching

The transformation depends on structure, not on a linear rule; it

requires the main clause auxiliary verb to move to the front of the
sentence, not the first verb. This reliance on structure is known as the
principle of structure preservation because, even though there has
been a transformation, the structure of the sentence has been pre-
served, with the rules involved in the transformation respecting the
structure of the sentence. This is remarkable when you realise that
more ‘logical’ linear language rules not only are not correct in English,
but they are also unattested in the world’s language, while structure-
dependent rules can be found in all languages. In short, natural
languages have structure, and constraints or ‘rules’ operate on struc-
tures which are not as straightforward as a linear analysis would be.
Central to Chomsky’s understanding of language is the fact that
children acquire a complete and complex linguistic system in a
relatively short amount of time and at a very young age – in fact,
at an age when other cognitive processes have yet to develop. While
Chomsky’s ideas about language may have been a natural develop-
ment out of the structural tradition in linguistics, his claims about
language acquisition directly contradicted understood ideas of the
day. Accepted ideas about language acquisition were a product of
another academic discipline that developed in the early 1900s: psy-
chology. Psychology is concerned with the internal, mental develop-
ment of individuals. The development of language is only one of many
aspects of development of interest to a psychologist, and furthermore
was seen as akin to the development of other kinds of cognitive ability
and knowledge. In 1957 the prominent psychologist, B. F. Skinner,
published a book called Verbal Behavior, in which he argued that
children learn language through imitation and habit formation. The
stimulus–response view of learning in general, known as Behaviourism,
was argued to characterise not just general learning, but the learning of
native language by young children as well.
The young Chomsky made his mark in the academic world by
challenging the well-established psychologist on his views of language
development. His basic claim was that language knowledge is differ-
ent from other kinds of knowledge and, accordingly, develops differ-
ently. For Chomsky, Universal Grammar (UG) is the set of linguistic
principles that all humans are born with and which enables the young
child to acquire any language to which he or she is exposed. In
addition to this claim of nativism, Chomsky argued that children are
able to generate a range of linguistic structures that they will not have
heard. In this way, language acquisition is not merely a matter of
 Historical Overview – Language and Language Teaching 31

imitation, but instead is generative process; an infinite number of

utterances are possible because of a limited number of underlying
linguistic principles. In other words, the linguistic system of a child
will develop beyond the input to which the child is exposed. This
argument is called the poverty of the stimulus. There is, of course, the
need for a stimulus – the mother tongue that surrounds the child. But
that stimulus is poor in the sense that the child will be able to produce
linguistic forms that it will not have ever heard. We will return to this
poverty of stimulus idea in the next chapter.
The difference in views between linguists and psychologists estab-
lished a fundamental divide that still exists among researchers inter-
ested in questions of language and cognition. Even today, the
cognitivists see language as knowledge much like other kinds of
knowledge – all of which make use of a single set of cognitive
mechanisms. The generativists, by contrast, see language as distinct,
using a special or privileged set of cognitive mechanisms. These two
traditions will be explored in more detail in the next two chapters. But
first we will return to the language classroom to look more closely at
what was happening there in the first half of the twentieth century.

EARLY APPROACHES TO LANGUAGE TEACHING

As we have already seen, disaffection with Grammar Translation in

Europe inspired educational reformers to develop alternative ap-
proaches to language teaching. However, efforts by individuals were
not widely recognised at the time because there was no effective means
of disseminating ideas. This is because of a general absence of features
associated with professionalisation within the field of language teach-
ing. Until the late 1800s, there were no professional organisations,
conferences or journals to bring language-teaching professionals
together. A groundswell for change was more possible once profes-
sional organisations like the International Phonetic Association were
established and once linguistics and psychology were recognisable as
academic fields. With more formal methods for communicating
professional practice, the Reform Movement might have been able
to move language teaching away from Grammar Translation. Even so,
the ideas of these early reformists may have been too radical to take
hold in an effective way. Furthermore, the idea that learners should be
immersed in the language like native speakers was not supported with
an effective method for implementing this idea in teaching. Thus, after
taking two steps forward, there was a need to take one step back.
32 Language Teaching

In the first half of the 1900s, three sets of language teaching practices
developed in each of the three large English-speaking regions of the
world. The Oral Approach that developed in England had much in
common with the Situational Approach that developed in the Australian
context. We will consider these two approaches before looking at the
Audiolingual Approach of North America.5 Like other early applied
linguists, Harold Palmer and A. S. Hornby in England looked to
structuralist linguistics and psychological notions of language develop-
ment to inform their Oral Approach to language teaching. Accordingly,
language was identified as a system of patterns and structural paradigms
which could be sequenced in a graded way. And, as suggested by the
label, there was a strong emphasis on the spoken language.
Central to the Oral Approach was the idea that new language points
should be presented in context. In this way, it was much like the view of
teaching known as the Situational Approach, which developed in
Australia among applied linguists led by George Pittman. The Situa-
tional Approach made use of objects, pictures and other realia to present
new language points. There was also much emphasis on the concepts or
notions embodied in language, and not just lists of grammar rules. This
progressive view of what to teach was not matched with an equally
progressive view of how to teach, however. In both of these approaches,
language was taught through teacher-directed activities such as drills,
substitution exercises, whole-class repetition and dictation.
These practices were similar to those used by the Audiolingual
Approach, which developed in North America. Overlap between the
Oral/Situational Approaches and Audiolingualism can be found in the
emphasis on oral skills, with literacy a secondary concern. The
difference, however, was the extent to which speaking was seen as
dependent on listening. Proponents of Audiolingualism made much of
the belief that a learner needs to perceive language accurately before it
can be correctly produced. This emphasis on the aural was a direct
result of the influence of Behaviourism. The behaviourists viewed
language as primarily speech, not writing. Speech was seen as a pre-
condition for writing. This conclusion was based on observations of
native language acquisition combined with recognition of the exis-
tence of non-literate societies without written traditions.
Language development was understood to occur through mimicry
and repetition. Language could be seen as embodying a connection
between a stimulus and a response; an apple, for example, would
trigger the word apple. Language-learning, like other kinds of learn-
ing, was seen as a product of stimulus–response and reinforcement. In
 Historical Overview – Language and Language Teaching 33

practice, this meant a large proportion of class time was devoted to

intensive drill work. Through repetition, learners would develop a
new set of linguistic habits. In the early stages, language production
would be limited to imitated forms. Eventually, however, a person
would then develop enough structures to analogise from the learned
set of structures to similar structures. This would allow for the
production of more creative forms to express new meanings.
One reason for the development of Audiolingualism was the po-
litical context of the time. The 1900s saw two world wars, with new
global powers emerging in the mid-1900s. The development of
Audiolingualism came from funding provided by the United States
military, which had identified a great need for language specialists to
support the efforts of the emerging superpower. The need to develop a
multilingual military in a limited amount of time led to programmes of
very intensive study. The very structured and authority-bound nature
of Audiolingualism was no doubt compatible with a military environ-
ment. The degree of success found by these intensive courses drew the
attention of the wider language-teaching community.
In sum, the emphasis on intentional, explicit habit formation was in
direct contrast to the earlier beliefs of proponents of the Direct Method,
in which learners would naturally develop the language around them.
For Audiolingualism, learning required repeated drilling of one lan-
guage pattern at a time. Applied linguist, Charles Fries, contributed to
the development of Audiolingualism with his study of the structural
similarities and differences between languages. We will look more
closely at his idea of Contrastive Analysis in the next section. But first,
we conclude this section by summing up the similarities and differences
between Audiolingualism and the Oral/Situational Approaches.
Though stemming from different academic and historic traditions,
they were, in fact, quite similar in practice. They were largely tea-
cher-centred practices without any notion of autonomous or self-
directed learning. Teachers relied heavily on drilling, requiring endless
repetition of stock linguistic patterns. Dialogues, presented by teachers,
were memorised and recited by students. Repetition through drilling
allowed for a focus on pronunciation, rhythm and intonation.
Audiolingualism differed from the Oral and Situational Ap-
proaches, however, in its theoretical basis. The latter two approaches
had the intention of relating language to specific contexts. Thus,
structural patterns were identified and taught based on particular
social contexts. For example, a lesson might revolve around greetings
or requests. In this way, the Oral/Situational Approaches were more
34 Language Teaching

functionally inspired, at least in principle. That the implementation of

these approaches relied on teacher-centred lessons with repetition and
memorisation, however, undermined the intention of presenting lan-
guage as functional. The Audiolingual Approach also sought to
produce learners who could use the language, and saw its reliance
on habit formation as a legitimate means to that end. While both
approaches gave thought to the choice and sequence of structural
patterns for instruction, the Audiolingual Approach based this on the
linguistic nature of the patterns, in contrast with the Oral/Situational
Approaches’ consideration of function. The question of sequencing of
structural patterns owed much to developments in the newly devel-
oping field of second language acquisition.

SECOND LANGUAGE ACQUISITION AND LANGUAGE TEACHING

With the development of the academic disciplines of psychology and

linguistics came the beginnings of subdisciplines. Of most relevance to
language teaching has been the development of the field of second
language acquisition (SLA). This interdisciplinary field linked research
in linguistics and second language development. Since language in the
early 1950s was seen as a set of habits, learning a second language was
seen as an attempt to change existing language habits to a new set of
habits. This process was seen as involving what is known as language
transfer. Learners transfer their habits from their native language to
the second, or target, language.
Taking a habit formation point of view, the logical prediction is that the
more different a language is, the more difficult it would be to learn since
there would be more habits in need of change. This is, in essence, the basis
of Contrastive Analysis. Charles Fries is known for developing this
research agenda with the aim of improving language teaching and
learning. The agenda was to compare languages in terms of sound system,
syntactic system, morphological system and even pragmatic system in
order to identify which language habits would need to change in learning
the second language. This notion was further developed by Fries’s student,
Lado, who hypothesised that there is a direct connection between the
degree of similarity between native and target languages, and the extent to
which learners would acquire them. Lado’s Contrastive Analysis Hy-
pothesis proposed that there would be difficulties in language learning
where there are differences between the native and target language, and
ease where there are similarities between the two languages.6
Armed with the knowledge of similarities and differences between
 Historical Overview – Language and Language Teaching 35

native and target languages, then, a teacher could simply focus on

teaching the differences. The similarities should prove unproblematic
because transfer of native language habits meant there was nothing new
to learn (aside from the differing lexical items). Differences, on the other
hand, would be problematic, with errors resulting from native language
interference. In short, the logical conclusion of the Contrastive Analysis
Hypothesis was that language development could be predicted based
on knowledge of the languages in question. If true, this finding would
have been revolutionary for language teaching. Unfortunately, how-
ever, Contrastive Analysis did not prove itself to be quite so straight-
forward, especially for areas of syntax and morphology.
Early damage to Contrastive Analysis came from empirical studies
showing that learners did not necessarily make the errors predicted by
the Contrastive Analysis Hypothesis. One famous example is a study by
Zobl (1980) of French learners of English. This study focused on
pronouns in object position because of the differences between the
two languages. In English, object pronouns follow the verb, while in
French they come before the verb. Following the logic of the Contrastive
Analysis Hypothesis, French learners of English should, at an early stage
at least, produce sentences like I them know. But this contrastive analysis
error does not actually occur among French learners of English. In
addition to errors that learners do not make, research also shows that
many errors that learners do make cannot be traced to differences in the
corresponding structure in the native language.
There were also problems with the definition of ‘ease’ and ‘difficulty’.
While these notions seem intuitively reasonable, it is not entirely clear
how one might actually define or measure them. Difficulty is expected
where there are language differences, and differences are expected to
cause error. But a learner making an error may not perceive the linguistic
point as difficult or even be aware of the error. So in what sense is it
difficult? The heart of the problem is that the definition of difference is
linguistic while the understanding of the notion of difficult is psycho-
logical.7 Like comparing apples and oranges, this mismatch undermines
the validity of the hypothesis. While structural properties of language are
a product that can be spelled out in terms of typological similarity and
difference, feelings of ease and difficulty are part of a process, and as
such, do not have objective qualities that can be readily measured.
As a teacher, you may have had experiences that contradict the
Contrastive Analysis Hypothesis. For example, most language teach-
ing materials will include lessons on so-called false friends, cognate
words that can lead to incorrect use by learners. Two languages will
36 Language Teaching

often have developed words from the same word root either because
of similarity in language family or from borrowing which comes from
language contact. While the two words may be traced back to a
common meaning etymologically, they can easily develop over time to
have different meanings or uses in each language. This can lead to
more problems, not fewer, as learners are likely to think they know the
meaning of the word because of the similarity and have difficulty
modifying it, even though they have been told the difference. For
example, upon encountering the Spanish word simpático, an English
speaker is likely to assume that it means sympathetic, when in fact it
means ‘nice’ or ‘pleasant’. The opposite problem is also likely to be
familiar to you as a language teacher. A very different or unusual
aspect of language can be easy for learners to learn precisely because it
is so different. The difference can draw attention and spark the interest
needed for explicit learning to occur.
Echoing what many teachers will know from experience, empirical
research found that learners do not necessarily have difficulty and ease
where they are predicted to, at least in areas of morphology and syntax.
In the area of phonological development the Contrastive Analysis
Hypothesis model has been found to be empirically sounder. Within
the framework of the Contrastive Analysis Hypothesis, phonologists
looked to cross-linguistic typologies, the better to understand language
development. This led to other developments which were an improve-
ment on some of the original Contrastive Analysis Hypothesis ideas. The
Markedness Differential Hypothesis proposed by Eckman (1977) was
based on a classification of linguistic phenomena in terms of whether
they are marked or unmarked. While the concept of markedness has
been defined in different ways, in general the term refers to how common
or basic a feature is cross-linguistically.8 A linguistic form is marked if it is
unusual, in comparison with a commonplace, or unmarked, form.
The Markedness Differential Hypothesis was based on a typologi-
cal grouping of phonological features into sets much like the implica-
tional hierarchies we mentioned in Chapter 1. Within a particular
hierarchy, a learner is predicted to be able to learn features of a target
language that were marked if their native language also included these
features. Secondly, because the hierarchy is implicational, a less
marked feature within the hierarchy would be readily learned if the
marked feature was learned. By appealing to universal factors of cross-
linguistic typology, this approach was more principled, and, not
coincidentally, has been found to be more valid as a predictor for
the development of second language pronunciation.
 Historical Overview – Language and Language Teaching 37

The shortcomings of the Contrastive Analysis Hypothesis as a

general theory of language development do not mean that the native
language is irrelevant when learning a second language. Clearly there
are transfer effects. But the difficulty is that transfer is not as neat or
straightforward as early researchers might have hoped. One reason for
this is because the structure of language cannot be characterised as a
set of habits, with second language learning a matter of establishing
new habits. Another is to do with the fact that language is complex,
involving more than just grammatical structures. One line of research
which is interested in language beyond the structural level had
developed the notion of Conceptual Fluency, which depends on a
view of language in which concepts determine the structures of
language. In this view, a language is a reflection of the concepts
underlying it; thus the way that a culture organises its concepts will
determine the forms found in the language. This is most clearly seen in
that language’s metaphors. Danesi (1995) argues that a Contrastive
Analysis approach may be useful for understanding learners’ difficul-
ties at the level of conceptual fluency.
Mainstream research on Contrastive Analysis, however, was based
on a structural view of language. The main method used in this
research was to analyse errors. Error Analysis, a research agenda
first promoted by Pit Corder, directly challenged the behaviourist
basis of Contrastive Analysis, appealing instead to the Creative
Construction Hypothesis, which tried to apply Chomsky’s view of
natural, innate language development to second language learning.9
Arguing that second language development mirrored first, this hy-
pothesis was often called the L2=L1 hypothesis. Yet it is obvious that
this is much too strong a claim.
Before leaving the discussion of Contrastive Analysis and Error
Analysis it is worth considering it from the point of view of the
classroom. An approach based on errors is quite negative and clashes
with the progressive trend in education since at least the 1970s to a
more humanistic and supportive approach to learning and learners.
This difference in approach between SLA researchers and those
interested in classroom teaching may have contributed to the split
that began at this time between the field of SLA and the practice of
language teaching, a division which we will discuss in more detail in
the next chapter. But first, let us briefly consider what was happening
in the classroom during this time period.
The generally accepted structure of language teaching lessons in the
1970s and 1980s was PPP: Present, Practise, Produce. Accordingly, a
38 Language Teaching

lesson would begin with the presentation of some language point by the
teacher, followed by controlled practice by students, usually in the form
of exercises. Once the language point had been learned, the students
would be given a less controlled activity which required them to produce
the language point more freely. This very structured approach to a lesson
was essentially a culmination of structuralist views of language and
behaviourist views of language development. Notice that while great
changes in linguistics and the understanding of language development
were being proposed, language teaching continued to hold fast to older
academic traditions. One reason for this could be the lack of agreement
in the research paradigms, as we have seen with the abandonment of
Contrastive Analysis and Error Analysis.
Despite the abandonment of the approach, useful developments did
emerge from the Error Analysis research programme. Errors were
analysed in the context of the language the learner was trying to acquire
and not just in terms of the properties of the native language. The shift
away from the first language as the only explanation for difficulties in
second language development gave more prominence to the process of
second language development itself. It was with this shift that it became
clear that a learner’s developing grammar is systematic. Adoption of the
term Interlanguage legitimised learner language as an object of study in
its own right instead of seeing it as a faulty version of the target language
or as a transferred version of the native language. The recognition that a
second language learner’s grammar is systematic and not a collection of
faulty attempts to mimic the target language nor an inappropriate use
of native language ‘habits’ was supported by a growing body of
research. This severely challenged the behaviourist view of language
in SLA. Drawing from developments in linguistics, language was no
longer seen as a set of habits, but instead a dynamic system of abstract
rules, and the second language learner was seen to be developing a
dynamic system. We will take this idea further in the next chapter and
conclude here with the claim that the development of the idea of
systematicity in SLA is perhaps the most important legacy of the
Contrastive Analysis/Error Analysis era.

CONCLUSION

In this chapter we have discussed the practice of language teaching in

terms of developments in the academic understanding of language and
language development. In the pursuit of improving language teaching
practices, early applied linguists turned to advances in both language
 Historical Overview – Language and Language Teaching 39

and cognitive development. The lack of success in language teaching

practices may, with hindsight, be expected, given the lack of agree-
ment that emerged between linguists and psychologists. Moreover, as
discussed in Chapter 1, there has been and continues to be little
consensus on how we should understand what language is, whether
form or function, a property internal to individual minds or external
as a feature of society.
Language teaching practices have also varied depending on educa-
tion theory as well as social context. In the pre-modern period,
classical languages like Latin were taught for reasons of academic
pursuit rather than for real communication. With the need to teach
vernacular languages, the method of teaching initially remained the
same. Yet teaching language solely for reading ancient texts is clearly
not appropriate if the aim is for actual interaction with speakers of
another language. Grammar Translation, well suited for learning a
language as an academic subject, was not well suited for learning
language as a useful tool in society.
The first half of the twentieth century can be seen as a time of
exploration as applied linguists attempted to develop more effective
teaching practices based on developments in linguistics and psychol-
ogy. The second half of the twentieth century has been one of great
exploration in linguistics and psychology, but without straightfor-
ward recommendations for how we should teach language. If the
effect of the first language could be captured in a simple hypothesis
like the Contrastive Analysis Hypothesis, then teachers could come to
their task armed with knowledge of the native and target languages
and teach accordingly. Instead, what emerged was a very general
understanding that interlanguage is systematic. What needs to be
clarified is how this can help language teachers teach more effectively.
One claim coming out of this chapter is that application of theory
and research to real-life practice requires clear, agreed generalisations
from researchers. This may be one reason why change can sometimes
be slow. This brief historical overview shows that change in language
teaching practices has at times been quite slow. While radical attempts
to change teaching, like the Reform Movement, did not persist, they
did give way to incremental change. We also saw radical change in the
field of language study with the ideas of Chomsky. Perhaps these ideas
also need time before meaningful change can take place. In the next
chapter we will explore developments in Chomskyan linguistics, with
the ultimate aim of asking what implications there are for language
teaching.
40 Language Teaching

FOR DISCUSSION

1. Are you aware of language teaching traditions in other parts of the world?
How do they compare with the history of language teaching in Europe?
2. To what extent are the ‘traditional’ practices discussed in this chapter still
relevant today? Can you think of a current context in which traditional
teaching would be considered appropriate?
3. List three similarities and three differences between your native language
and any second language you have learned. Based on your experience as a
learner, do your examples seem to support the Contrastive Analysis idea
of ease and difficulty or go against it?
4. This chapter concluded with the claim that second language development is
systematic. What does this mean? Can you think of examples from your own
experience or knowledge that support or explain this claim of systematicity?
5. Do you prefer an inductive or deductive approach to grammar teaching?
Why? Would your answer differ depending on who your learners are?
Does it differ depending on the particular aspect of grammar?
6. Towhatextenthasyourlanguageteachingand/orlearningbeenbasedonaPPP
approach? What are the strengths and weaknesses of PPP, in your experience?
In what context is PPP an appropriate structure for language teaching?

NOTES

1. Musumeci (1997) documents this long and rich history in a short but
carefully researched volume.
2. See Rutherford (1987) for a discussion of the development of language
teaching within the context of academia and for a discussion of grammar
teaching in particular.
3. For extensive treatment of the history of language teaching in the English
context, see Howatt and Widdowson (2004).
4. For a classic collection of historical readings in linguistics, see Sebeok
(1966). Many classics, such as Bloomfield (1933), can also still be found
on the shelves of major libraries.
5. See Part 1 of Richards and Rodgers (2001) for a more detailed descrip-
tion of these approaches.
6. To read the original proposals, see Fries (1952) and Lado (1957). Many
introductory SLA textbooks provide clear discussions of the Contrastive
Analysis Hypothesis. See, for example, Gass and Selinker (2008).
7. For discussion of this point, see Long and Sato (1984).
8. See Haspelmath (2006) for a discussion of markedness which outlines
the various ways the term has been used and highlights the problems that
these differences in definition raise.
9. For Error Analysis, see Corder (1967, 1981). The main proponents of
the Creative Construction Hypothesis were Heidi Dulay and Marina
Burt. See Dulay and Burt (1975).
 3
LANGUAGE AS A BIOLOGICAL PROPERTY

LANGUAGE: A GENERATIVE VIEW

The middle of the twentieth century was witness to the Chomskyan

Revolution. Noam Chomsky’s view of language has had a lasting
impact, giving rise to a new field of linguistics known as Generative
Linguistics. From this, the subfields of generative first and second
language acquisition have also developed, though Chomsky himself
has had limited involvement in these areas. As we will discuss in this
chapter, the philosophical foundation of his theory depends quite
crucially on native first language acquisition. SLA, by contrast, goes
beyond Chomsky’s particular theoretical concern. Though Chomsky
himself has never extended his work to SLA, and certainly not to the
realm of classroom teaching, other researchers have explored these
areas from a generative point of view.
Though Chomsky’s generative theory has developed over time, its
aim has consistently been to move away from an item-by-item view of
language and language development to a model which captures
language as a system with universal configuration. While this endea-
vour initially led to a very complex and elaborate set of algorithms and
formulae, the current research is devoted to paring back the model to
its most simple and universal properties.
The two main tenets of Chomsky’s view of language are that
language is constrained by universal principles and that these prin-
ciples are an innate biological property of humans. The principles of
language are generative, allowing us both to interpret and to creatively
produce an infinite number of sentences. We will discuss these two
central points of Generative Linguistics in terms of what they mean for
the understanding of what language is and then what this means for
language development.
Chomsky’s ideas about language can be seen as a natural devel-
opment out of the work of the structuralists, as it is concerned with
identifying cross-linguistic patterns in languages at the level of the
42 Language Teaching

sound, the word, and the sentence. As we have seen, one criticism of
the Generative Approach is that it treats particular aspects of lan-
guage, especially syntax and phonology, in isolation, reducing them in
an atomistic way without proper regard for meaning. This is a valid
criticism, but it does not tend to bother generativists because their
main aim is to identify the structural principles underlying language
which give rise to meaning, not to explain meaning itself. For gen-
erativists, meaning is the product or result of linguistic structures, not
what determines structure. In other words, while our thoughts de-
termine what ideas we want to convey, the linguistic structure itself is
constrained by linguistic principles that are not altered by the mean-
ing. Instead, we have to choose from the range of structures available
in our language in order to convey our thoughts in the way which best
suits the intention of our message.
In the 1960s and 1970s specific language structures were associated
with ‘rules’, known as transformations. For example, there was a rule
for ‘passive formation’ which described how a passive sentence, like
The cat was chased by the dog, was derived or transformed from an
active sentence, such as The dog chased the cat. The fundamental
difference from the structuralists, however, was Chomsky’s observa-
tion that language patterns are not logical; they do not abide by rules
that can be determined by simple comparison of patterns. Instead, as we
saw in Chapter 1, rules are structure-dependent. One famous example
from Chomsky (1970) is the observation that the two sentences in (1)
may look the same, but they have different structural properties:

(1) a. John is eager to please.

b. John is easy to please.

They both have the shape of subject + copular verb + predicate

adjective + infinitival phrase, but you might have noticed that the
role John plays in the two sentences is different. John is the one who
wants to please in (1a), but the recipient of pleasure in (1b). It would
be difficult to explain this difference based on pattern alone; the next
set of examples shows that they are clearly different.

(2) a. John is eager to please Mary.

b. *John is easy to please Mary.

(3) a. *It is eager to please John.

b. It is easy to please John.
 Language as a Biological Property 43

From these examples we can see two basic generative notions.

Firstly, all native speakers will agree that John plays a different role
in these two sentences, but very few would be able to explain why. In
other words, speakers have knowledge of which they are not them-
selves aware. Secondly, speakers of a language know what is not
structurally possible. The sentence in (2b), for example, is not complex
– it does not include any difficult vocabulary, nor is it very long. Yet it
is clearly not an acceptable sentence. This is an example of what is
meant by constraints on language. There must be some rule or
principle that causes (2b) to be unacceptable; and this rule clearly
exists whether we are explicitly aware of it or not.
Over the years generative theory has undergone a series of changes,
with the name of the theory changing accordingly. The discussion of
language rules above is typical of the early theory, known as Standard
Theory, then Extended Standard Theory and then Revised Extended
Standard Theory! Each step in the development was a step away from
describing language structures like those in (2) towards explaining
them. This development in theory led to a move away from a growing
list of rules to a more general idea that all languages abide by a set of
universal constraints, known as principles, within a prescribed set of
options, known as parameters. The Principles and Parameters ap-
proach, proposed in the early 1980s, was part of the next major
revision in generative theory, known as government and binding.1 An
example of a principle common to all natural languages is that all
languages have a grammatical category for verbs with properties that
distinguish them from other categories, like nouns, for instance. The
ordering of verbs relative to the other elements in a sentence, by
contrast, is determined by a parameter. You will no doubt be aware
that languages differ in terms of whether the verb comes before or
after its object. There is a word order parameter with two possible
settings, usually referred to as the Head Parameter because the verb is
considered a head that dictates certain properties of the object. This
kind of parametric option can explain structural differences between
languages.
The idea that there are principles and parameters is still funda-
mental to the most recent version of the theory, known as Minimalist
Theory.2 As suggested by the label, Minimalism is concerned with
reducing the theory to as simple and unified a set of principles as
possible, to include only a few principles that constrain the core
elements of syntax and phonology, while cross-linguistic differences
are captured in parameters that are associated with non-core linguistic
44 Language Teaching

properties, like lexical properties. All of these versions of generative

theory, from Standard Theory to Minimalism, differ in terms of the
technical mechanics which regulate language, but they are all the same
in terms of the fundamental notions that we are considering here. For
generativists, knowledge like that illustrated in (1) to (3) above is
considered ‘core’ linguistic knowledge – subtle grammatical properties
that operate when speaking and hearing language, but whose proper-
ties the speaker does not explicitly understand. Somehow, each speak-
er develops a rich and complex set of abstract linguistic constraints of
which they may never be consciously aware.
A different type of complexity underlying language is illustrated in
another early example from Chomsky (1965):

(4) He decided on the boat.

If you think about the sentence in (4), you will agree that it is
ambiguous – it has two possible interpretations. One interpretation
is that someone made a decision while on a boat. The other is that a
decision was made about a boat (perhaps to purchase it), though the
person making the decision could be anywhere. If this second inter-
pretation is not easy to reach, consider sentence (5), which only has the
second interpretation:

(5) While on the train, he decided on the boat.

Ambiguous sentences and ungrammatical sentences are considered im-

portant evidence for the idea that core linguistic knowledge is innate. The
main argument is that these kinds of properties could not be learned
through imitation or repetition, as the Behaviourists originally argued.
The simple logic is that, if the source of this knowledge is not other
speakers, it must be that that this knowledge is innate. One property that
makes us human, then, is that we have a built-in predisposition for
language. We are born with linguistic knowledge that guides and con-
strains our use of language: in other words, Universal Grammar (UG).
UG is a set of universal properties that limit what a language can
and cannot do. It is these limitations that make a language a ‘natural’
language. As computer programmers and researchers in artificial
intelligence know, mimicking an existing language is very difficult;
designing a new language as intricate as any of the world’s natural
languages has proven impossible. Yet young humans develop lan-
guage with relative ease. Unless there is some pathological obstacle,
 Language as a Biological Property 45

every child develops what Chomsky calls linguistic competence. This

competence is knowledge of the core structural properties of one’s
native language, but as knowledge which is outside the explicit
conscious control of the language user. This idea of implicit linguistic
knowledge is not very controversial. Most linguists of any theoretical
background agree that we have knowledge that allows us to operate as
language users without explicit understanding of what that knowledge
is. The more controversial claim particular to the generative view is
that this knowledge is different from other kinds of knowledge. For
generativists, this distinct knowledge is modular.
Philosopher Jerry Fodor developed the idea of modularity, arguing
that modular knowledge is qualitatively different from general knowl-
edge. General knowledge is the knowledge you come to strictly
through experience. It can be deliberately altered and consciously
thought about. Fodor (1983) distinguishes modular knowledge from
general knowledge in terms of five properties. Firstly, it is specific to its
domain; examples include the five senses plus language. In other
words, the language module relies on mechanisms for language,
the vision module on mechanisms for sight, the smell module on
mechanisms for smell, and so on. Secondly, it is also different because
it is very fast. Consider, for example, some piece of general knowl-
edge, such as how many inches there are in a foot. While you will be
able to answer ‘12’ very quickly, it is not nearly as instantaneous as
your eyes’ ability to see the letters on this page, or your linguistic
ability to interpret the question. Thirdly, modular knowledge is
automatic – the module cannot not operate unless it is physically
impeded. You will know this if you have ever been trying to read on a
train with someone nearby who is talking on a mobile phone. Unless
you physically block your ears, it is very difficult not to hear what they
are saying. The fourth property of modular knowledge is that it is not
explicit metalinguistic knowledge. We saw this above with the am-
biguous and ungrammatical sentences. The fifth and final property is
that modular knowledge operates independently in such a way that it
cannot be altered by general knowledge. To test this, try speaking to
someone for five minutes without using the articles, a(n) or the. You
may be able to do it initially if you speak slowly and think carefully
about every word, but to maintain it would be very difficult. And,
most importantly, you would not be able to alter your language
system so that it no longer included articles. It is a property of the
(English) system which could not be changed even if you felt strongly
that articles were unnecessary and no longer wanted to use them.
46 Language Teaching

Before leaving this discussion of linguistic theory, it is important to

make clear that the theory of modularity is a theory about the mind.
We explained in Chapter 1 that mind refers to how cognitive processes
are represented and understood to operate. In this sense, it is a more of
a philosophical construct, in contrast with the brain, which refers to
the physical organ. As we will briefly discuss in Chapter 5, brain
studies by neurologists are beginning to investigate whether particular
locations in the brain correspond to language function at a more
specific level than the well-known Broca’s and Wernicke’s areas. This
is an area of cutting-edge research which follows a long history in the
development of science. It was well accepted, for example, that traits
pass from the parent to the child, long before the physical mechanisms
of genetics were identified. We turn now to child development to
discuss the development of language.

NATIVE LANGUAGE DEVELOPMENT

Perhaps the most revolutionary aspects of Chomsky’s ideas have been

to do with children’s acquisition of their first language (FLA). As with
the modifications of linguistic theory, a number of technical points
within FLA have developed over time, but the basic premise of the
generative view of FLA remains the same. Building on the view of
language discussed above, the innate predisposition for language
means that language development is going to occur with the help
of UG. This ‘help’ was initially conceived of as some kind of Language
Acquisition Device (LAD), but it quickly became clear that this kind of
terminology was misleading, as it suggested some kind of gear box in
the brain with cogs and switches. It is more accurate to say that the
mind responds to the input to which it is exposed by developing
language within a set of constraints.
The clearest expression of this idea comes from Chomsky’s (1981)
Principles and Parameters approach. A child is born with a set of
universal linguistic principles, along with a complement of unset
parameters waiting for linguistic input. Language acquisition is the
process of determining parameter settings based on language input. In
the last section we discussed parameters as having two options for a
single aspect of language. In fact, a single parameter is understood to
implicate a cluster of properties which may not readily seem related.
For example, if the Head Parameter is set for verbs to precede objects,
then automatically, prepositions will also precede their accompanying
noun phrases, for example, in the house, not *the house in. This is
 Language as a Biological Property 47

important because the child only needs to confirm the global setting of
a parameter in order to acquire a range of linguistic properties. If, by
determining one aspect of the target language, a child can come to
know a range of related subproperties of the parameter, this would
explain how a child can acquire the complex rules of a language before
other cognitive abilities have yet to mature.
In theory, all the formal properties of all languages can be identified
as either a principle or a parameter. These properties – that is, UG –
are innate, thus explaining how children can develop their native
language at such a young age and with relative ease. One generative
FLA agenda, therefore, has been to identify how these principles and
parameters account for development. In practice, this has not been
straightforward. But commitment to the idea of an innate UG that
constrains language development through some variant of principles
and parameter has not changed.
Within the generative claim of nativism there are some basic
arguments that enjoy general agreement despite debate on more
technical details. The first is the fact that all normally developing
children develop linguistic competence in a specific language. Of
course, they may or may not develop literacy skills, or the ability
to speak with the skill of a politician, or have a very impressive range
of vocabulary; but these aspects of language are outside the core
linguistic knowledge of interest to most generativists.
The second point is that there are stages of development in FLA that
all children go through in the same order (though often at different
rates). Brown’s morpheme order study, mentioned in Chapter 1,
exemplifies stages of development of grammatical morphemes among
English-speaking children. Additionally, stages in FLA show distinct
developmental patterns. We also saw examples of this in Chapter 1
with over-extension of the regular past tense suffix -ed rule to irregular
verbs like break and did by young children. But children are not
known to over-extend nouns or adjectives to say things like *I happied
yesterday, even though such a sentence would be perfectly under-
standable. This shows that when creating language based on rules, a
child is sensitive to the grammatical constraints on particular rules.
We also saw that children are resistant to change in their grammatical
systems until they are ready for the next stage.
The third point basic to the generative view is the observation that
child language development goes beyond what a child is exposed to in
their environment. The stimulus is said to be poor relative to ultimate
development of the child’s core competence. We illustrated a poverty
48 Language Teaching

of the stimulus phenomenon in the last section. A child does not need
to be told that John is easy to please Mary is ungrammatical. Nor do
they need to be taught that He made a decision on the boat has two
possible interpretations. These and many other subtle aspects of
language develop naturally as long as the child has been exposed
to language. This remarkable aspect of development is the main
reason why generativists believe that there must be some in-built
mechanism in the brain for language acquisition. It is important to
note that, though this discussion has been based on examples in
English, the theoretical points are valid for any and every language,
including sign language.
Much more could be said about child native language acquisition
and the development of linguistic theory. But we must leave these
discussions of formal linguistics to turn to the linguistic subfield of
more direct relevance to language teaching, that of second language
acquisition. What does this biological view of language and language
development mean for second language acquisition?

SECOND LANGUAGE DEVELOPMENT

The Current Gap between Theoretical and Applied Linguistics

The Chomskyan approach to language has been often criticised as
irrelevant to classroom language teaching, with many expressing
scepticism about the usefulness of the Generative Approach.3 A
rejection of formal linguistics by many applied linguists would never
have been contemplated during the era of pre-Chomskyan structural
linguistics. In fact, the term ‘applied linguistics’ first began to be used
in the mid-1900s, when there was considerable activity within lan-
guage pedagogy to find theoretical grounding for language teaching
practices. As noted in Chapter 2, prior to the 1950s, linguistics and the
new field of psychology appeared to be able to supply scientific
grounding which language teaching seemed to be lacking.4 But the
current situation is quite different. In addition to reluctance on the
part of applied linguists, generative SLA has developed without much
interest in questions of the language classroom.5 This contrasts with
early attempts to apply generative research to language teaching,
which had profound effects on pedagogy, but only in an abstract
and general way. For instance, the importance of ‘natural’ input to
language development is now widely accepted, but more specific
applications of how or when to supply this input is not.
 Language as a Biological Property 49

Unfortunately, since the late 1960s a rift has developed between

researchers studying the formal properties of language and those
interested in the teaching of language. A talk given by Chomsky at
the Northeast Language Teachers’ Association in 1966 marked a
pivotal point in the distancing of the two. In that talk, Chomsky was
widely understood to have said that linguistic theory is irrelevant to
the concerns of language teachers. In fact, what he actually said was:
‘the implications of these ideas for language teaching are far from clear
to me’ (Chomsky 1970: 59). The misconstrual of his remark can be
seen as a product of early frustration with the research agenda of
Generative Linguistics. The Chomskyan ‘revolution’ did not lead to an
equivalent revolution in the language classroom. There were early
attempts to introduce generative theory to language teachers,6 but
these proved inappropriate for the language classroom.
The Chomsky-led move away from habit formation to an innate
predisposition contributed to the distancing from teaching, since
‘natural’ language acquisition does not require teaching. Moreover,
Generative Linguistics modelled itself on the sciences, focusing on
abstract theoretical principles of language. Because their main aim was
to define the properties of linguistic competence, the focus was limited
to the abstract constraints of language. First language acquisitionists
followed this line of inquiry by looking for evidence of universal
constraints in the development of native language. It is only natural
that generative researchers interested in second language development
would follow suit. Since the 1980s, generative SLA researchers have
explored whether there is an innate predisposition for SLA. As
psycholinguists, their remit has been to investigate the internal, mental
processes implicated in second language development.
This research agenda is primarily interested in the properties of
language produced by learners, not learning as a product of teaching.
Those applied linguists who are interested in teaching tend to base
their research on functional, cognitive and/or sociocultural frame-
works, in part because these approaches have clearer compatibility
with questions of pedagogy. Sadly, academic debate can at times lead
to rigid divisions. There is a tendency for each approach to see itself as
the only correct approach. When fierce debates do not find any
resolution, it is easy for factions to develop. It is easier to work within
one’s own faction, debating minor points within an agreed general
theory, than to continue to engage with others whose basic theory is at
odds. In time, this can give rise to disparate subfields with separate
conferences and academic journals, resulting in a situation in which
50 Language Teaching

academic discussions between the factions are more and more rare.
This lack of cooperation within a discipline can leave the impression
that the ‘experts’ themselves do not have the answers, and result in an
atmosphere which cannot be said to be welcoming, supportive or
helpful for language teachers.
Another hurdle for teachers to overcome is the abstract nature of
linguistic formalisms. While the discussions of Generative Linguistics
in this book so far have hopefully been clear, you will probably agree
that they are relatively complex. In truth, they are carefully selected
points presented in a way that avoids the more complex formalisms of
the theory.
The most fundamental criticism of generative SLA, however, comes
from applied linguists who point out that there has been a lack of
concrete findings from generative research that can give guidance for
decision-making in the classroom. While it may seem a natural step to
apply findings from research on SLA to the teaching of language as they
emerge, meaningful application is only possible once there is a clear view
of what needs to be applied. Thanks to the commitment of SLA
researchers, there is now a large body of findings, with some generally
accepted trends beginning to emerge. This means that there are now
enough robust findings from a wealth of studies to begin making
connections to the language classroom. The aim of this book is to make
some of these connections. In the remainder of this chapter we will
consider some of the landmark research in SLA in order to explore the
implications for language teaching in subsequent chapters. Hopefully, in
doing so, we will also take a step towards closing the gap that currently
exists between generative linguistic research and language teaching.

Early Findings: Systematic Development

SLA research has a long tradition of looking for parallels between
second language development and native first language development.
If there are similarities between the two, then this could be support for
the claim that there are similar mechanisms at play in both processes.
From a generative perspective, any such similarities would be attrib-
uted to UG. The research comparing first and second language
acquisition in the 1960s and 1970s, however, did not rely heavily
on generative theory, but instead looked for general trends in second
language development. Because of the interest in comparisons with
native child development, much of the early focus was on children,
typically between the ages of five and nine in naturalistic settings.
 Language as a Biological Property 51

An important example of this kind of research is the morpheme

order studies that investigated L2 learners in order to look for parallels
between SLA and the findings of Brown (1973), which showed a
regular order in the acquisition of English inflectional morphemes by
native-speaking children. The first set of studies tested sixty Spanish
and fifty-five Chinese children learning English as a second language
(Dulay and Burt 1973; Dulay and Burt 1974). The results showed a
similar pattern of development between the child L2 learners and
Brown’s native speaker learners. Curious whether the result had to do
with age, Bailey, Madden and Krashen (1974) tested seventy-three
adult learners of English and also found a similar developmental
pattern. This research led to the conclusion that second language
development occurs in stages, and that the stages are much like those
in first language development.
Of course, SLA is different from FLA. For one thing, second
language learners have a native language to begin with. The role of
the native language, or L1 transfer, is another strand of SLA research
which has received much attention. In fact, Bailey, Madden and
Krashen (1974) were also interested in this second issue. In addition
to testing adults, they constructed their test so that forty of the test
subjects were speakers from eleven different native language back-
grounds, while thirty-three were Spanish speakers. When analysing
their results in terms of native language background, they found no
meaningful differences in the developmental patterns. Numerous
studies have replicated these findings, leading to a general consensus
that English morphemes are acquired in stages which are comparable
to those in native L1 acquisition, regardless of the age or native
language of the learner. The interesting question for researchers is to
explain why.
Though there are different explanations for why this regularity
occurs,7 the more general conclusion which shares wide agreement is
that second language learning is not a random process of decreasing
L1 transfer effects, but instead one of language development that
happens in stages that follow general trends, and that these stages are
similar for children and adults, regardless of their native language.
This conclusion, accepted by the 1980s, was compatible with the
general conclusion coming out of the work in Contrastive Analysis
and Error Analysis, which argued that L2 development is systematic.
The influence of generative theory on this conclusion was to suggest a
new understanding of the source of this systematicity. Researchers
hypothesised that UG was the constraining mechanism underlying
52 Language Teaching

second language development. If this was true, then the better the
properties of UG were understood, the more it would be possible to
predict and explain L2 development.
This recognition of the systematicity of L2 learners’ grammars
required some new terminology. A new label was needed because
it would be inaccurate to refer to the learner’s grammar as the target
grammar, or as an incorrect version of the target grammar. An English
speaker learning Spanish, for example, does not have a grammar of
Spanish. Thus, Larry Selinker (1972) coined the term interlanguage to
refer to the grammar of the L2 learner; an English speaker learning
Spanish will have a developing Spanish interlanguage. In addition to
being systematic, an interlanguage typically reflects both the target
language and the native language. It will change from stage to stage,
but does so in a principled way; it is not simply an imperfect attempt at
the target language. Most remarkably for generative researchers,
interlanguages have been shown to have elements which are natural
to language, but not a part of the learner’s native language nor the
target language they are trying to acquire. As will be discussed in the
next section, this has been taken as strong evidence for UG.
As the concept of UG has gained support and acceptance, the
question of the extent to which second language development is
‘natural’, depending on biological mechanisms, has arisen. After
all, it is quite obvious that first and second language acquisition
are different. This question became framed as one of ‘access to
UG’ in non-native language development, particularly for adults. Like
some of the early metaphorical language for native language devel-
opment, the use of the term ‘access’ suggests some active, explicit or
intentional process. For this reason, it is more accurate to use the term
UG-constrained development. We will explore the question of UG in
SLA in the next section.

UG-Constrained Development
If L2 development is constrained by UG, this entails that L2 devel-
opment can also be aided by principles and parameters. Guided by the
knowledge that L2 development seems different from native language
development, however, researchers wondered if the difference had
something to do with the fact that the L2 learner already has a
complete set of parameters which underlie their native language.
The question then becomes whether or not a learner can ‘reset’ their
parameters. Another possibility is that there is some maturational
 Language as a Biological Property 53

timeline which puts limits on the age span during which parameters
can be reset. The quest to determine the role of UG thus became
understood in terms of parameter resetting.
One of the most thoroughly discussed parameters is the so-called
Null Subject Parameter, which regulates whether or not a language
requires a subject to be expressed in order for a sentence to be
grammatical. As shown in (6a), subject-less declarative sentences
are ungrammatical in English, a language with a negative null subject
parameter setting. An example of a language that has a positive null
subject parameter setting is Spanish, shown in (6b).

(6) a. *Speak English.

b. Habl-o Ingles.
speak-1sg English
‘I speak English.’

According to linguistic theory, the null subject parameter includes a

cluster of properties, including the properties of modal verbs like must
and should. One well-known early study by Hilles (1986) was a case
study which followed the language development of a 12-year-old
Spanish boy called Jorge, learning English over a period of 10 months.
She found that, at first, Jorge produced English sentences without
subjects and without modal verbs. At the time when he began to
include subjects regularly in his sentences, he also began to use modal
verbs correctly. This result was very exciting, as it suggested para-
meters could be reset, indicating a role for UG in second language
development. But unfortunately, other research on the null subject
parameter found learners did not acquire all the ‘clustered’ properties
together. As a result, based on the studies conducted thus far, it is still
unclear whether L2 learners, particularly adults, can reset parameters.
One possible reason for mixed results found across studies is the
unfortunate fact that there has been considerable disagreement by
formal linguists over the exact set of properties which are seen to
cluster with the null subject parameter. Even more unfortunately,
there is considerable debate over the exact linguistic properties asso-
ciated with any of the parameters that have been proposed in the
parameter resetting literature so far. This messy picture has discour-
aged many, and is one legitimate reason for the warning by some
researchers that SLA results should be applied with caution.
More current research in SLA is taking a much finer-grained
approach, recognising that language includes a range of aspects from
54 Language Teaching

syntax and phonology to lexical and functional properties. This more

sophisticated approach is yielding a view in which different domains
within language are seen to develop differently. We will consider one
example later in this chapter. This continuous refinement of research
in formal linguistics may mean that L2 researchers will have to keep
refining their research as well. None the less, the results showing some
degree of parameter resetting are enough to suggest that it would be
unwise to rule out UG altogether.
There is other evidence of UG-constrained L2 development. As
mentioned above, there is evidence that the interlanguage of L2
learners sometimes includes elements of language which are found
in neither their native language nor the target language. Learners
sometimes go through a stage in which they seem to have a setting for
a parameter that is one of the options available for natural languages,
but which occurs in languages to which they have not been exposed. A
study by Clahsen and Hong (1995) of the null subject parameter
which investigated Korean learners of German seemed to find just this.
In order to make sense of this study, we need to know a few facts
about Korean and German, and a bit more about null subjects.
There are, in fact, two types of null subject languages. The first is a
language that allows null subjects because the verb form shows
enough information about the subject to allow the subject to be
omitted. We saw this in example (6) above for Spanish. The verb
hablar, ‘to speak’ in Spanish, has the 1st person singular -o ending,
allowing us to identify the subject as the 1st person singular, ‘I’, in
Spanish, even though it is omitted.
The second type of null subject language is exemplified by Korean
(and Japanese and Chinese). In these languages, verbs do not show the
person/number properties of the subject, but it is still possible to leave
out a subject – but only if the subject of the sentence is clear from the
context. So, for example, if you just explained that your friend, John,
was in a car accident, it would be grammatical to say (the equivalent
of) Got hurt. in Korean because it is clear from the context that it was
John who got hurt. So, now we have three options. There are
languages that do not allow null subjects, like English. Then there
are languages that allow them because of verbal morphology, like
Spanish (and Italian and Polish). Thirdly, there are languages that
allow null subjects if the discourse context makes the subject so clear
that it does not need stating, like Korean.
So what kind of language is German? German verbs show person/
number, like Spanish, but with regard to null subjects, German is like
 Language as a Biological Property 55

English; subjects are required in declarative sentences. So, Clahsen and

Hong were curious to see what Korean speakers would do with their
subjects when learning German. Thirteen of the thirty-three learners
tested seemed to know that German requires subjects, while two
seemed to be still using the Korean null subject setting. Because the
remaining eighteen do not seem to have either setting, Clahsen and
Hong conclude that parameters are not reset in L2 development, with
their ultimate conclusion being that UG has no role in second language
development.
These results have been looked at again by other researchers,
however. White (2003b) notes that five of these learners seemed to
have the setting not of Korean nor of German, but of a null subject
language like Spanish! This is seen in results of Korean speakers who
seem to know the verbal agreement properties in German, despite the
fact that Korean does not include verbal agreement, but not the null
subject properties. She interprets this as evidence that there is some
universal guiding principle for L2 development. This would account
for results that cannot be explained by the influence of the native
language, nor directly from the input from the target language. Of
course, this conclusion has been made based on only five learners of
English. However, there have been other studies of other parameters
that have also shown three sets of results – learners who have
developed the target settings, learners who still have their native
setting, and then an intriguing set of learners who seem to have a
setting that is known to exist in natural languages, but not the setting
of either language in question.8 These kinds of results have been taken
as evidence for a role for UG in L2 development.
As you can see, this kind of argumentation is complex, requiring
cross-linguistic study and a sophisticated understanding of formal
properties of linguistics. Ideally, teachers will be able to study enough
linguistics to make sense of this kind of research. Arguably, however,
it is the role of applied linguists and teacher trainers to make these
kinds of results clear for teachers. At the very least, it is important that
you as a teacher are aware of the overall conclusions, regardless of
whether you take in the more theoretical details of this kind of
research.
There is another line of research investigating the role of UG that
has also emerged. As in native FLA, researchers have asked whether
learners come to know subtle properties of a language without being
explicitly taught. In other words, can we find evidence that learners go
beyond the input, as they do in L1 acquisition, to acquire so-called
56 Language Teaching

‘poverty of the stimulus’ phenomena. The logic is that, if a learner

knows something that has not been taught and is not readily apparent
in natural input, this knowledge has to have come from somewhere.
Of course, the native language is one possible source. So these studies
have to take L1 influence into account as well.
The first study of this type was done with English learners of
Japanese by Kanno (1997, 1998), who studied a linguistic property
known as the Overt Pronoun Constraint, another property associated
with null subject languages.9 This constraint is to do with the inter-
pretation of the word ‘who’. Consider the English sentence below:

(7) Whoi/j said that hei bought a car?

In English, ‘who’ has two possible interpretations, either referring to

the same person as the person that bought the car, or referring to
someone else not mentioned in the sentence. In other words, (7) could
involve either one or two people – either the same person bought a car
and claimed to buy it, or one person bought it but someone else is
telling us about it. We can illustrate this dual interpretation on paper
by using subscripts. The little i is indexed to the pronoun he, as well as
to an unspecified referent outside the sentence, shown by the index j.
In null subject languages like Spanish and Japanese, the interpreta-
tion facts change depending on whether there is an overt subject in the
embedded clause or not. In null subject languages, if the embedded
clause has an overt pronoun then (the equivalent of) ‘who’ can only
refer to someone outside the sentence. This is shown for Japanese in (8):

(8) darej-ga karei-ga kuruma-o katta to itta no?

whoj-subj hei-subj car-obj bought that said Q
‘Whoj said that hei bought a car?’

If the embedded clause does not have a subject, (shown as Ø below),

then ‘who’ refers to the same person as the one who bought the car:

(9) darei-ga Øi kuruma-o katta to itta no?

whoi-subj hei car-obj bought that said Q
‘Whoi said that hei bought a car?’

You will probably agree that this Overt Pronoun Constraint is quite
subtle. Kanno argues that this UG constraint is not one of which
native Japanese speakers, not even teachers, are consciously aware.
 Language as a Biological Property 57

Importantly, it is not taught in Japanese language classrooms. This

qualifies it is as a poverty of the stimulus phenomenon. Kanno tested
English learners of Japanese and found that, despite never having been
taught the Overt Pronoun Constraint, about 80% of them knew it.
Because this constraint is so subtle, it is unlikely that learners would be
exposed to many (if any) examples of these limits on interpretation in
natural/un-instructed input. Therefore, the notion of poverty of the
stimulus seems relevant not just for native L1 acquisition, but (adult)
L2 acquisition as well. As learners from a negative null subject
language like English acquire a positive null subject language like
Japanese, this kind of UG principle becomes part of the learner’s
interlanguage. This research has inspired other poverty of the stimulus
research and is taken as straightforward evidence for UG-constrained
development in second language development.10

Instructed L2 Development
So if L2 development is constrained by UG, does this mean that there
is no place for explicit instruction? Most generative research has
avoided questions of classroom instruction. However, this question
of UG-constrained development vs explicit language learning was the
basis for a study by White (1992), who investigated French learners of
English, looking at adverb placement. French and English differ in
terms of where adverbs can occur:

(10) Les chats attrapent souvent les souris.

Cats catch often mice.
‘Cats often catch mice.’

White studied two classes of 10–12-year-old French-speaking chil-

dren. One class was explicitly taught the properties of adverb place-
ment in English. The other class was only exposed to lots of examples
of English sentences with adverbs in them. The class who was taught
explicitly scored higher on a test of adverb placement than the other
class, who did not know that adverbs need to go before the verb in
English. This experiment would have been clear evidence that teaching
is effective except for the unfortunate fact that these same two classes
scored equally badly on an adverb placement test a year later.11
From this study it seems that explicit teaching alone does not have
long-term value, at least for some aspects of language. Knowing that
learners do develop proficiency in classrooms, one way to interpret
58 Language Teaching

this result would be to say that some aspects of language development

depend on biological constraints that are not readily affected by
explicit instruction. This may be especially true of particular areas
of language. It is generally agreed that it is more difficult for learners to
know what is not possible in a language than what is possible. The
French children in White’s study will initially have a French rule for
adverb placement. When they learn English they may in the short term
be able to remember not to use the French rule if they are told that it is
not correct for English. But the fact that they no longer remembered a
year later suggests that their linguistic competence for this aspect of
English did not actually develop. What remains an open question is
whether this kind of knowledge could ever develop. Unfortunately,
research by generative SLA researchers on this and on the effect of
explicit instruction remains extremely limited.

Age and Language Development

Generative SLA researchers often argue that the main reason for
failure among adult L2 learners is input. L2 learning that takes place
in a classroom is very different in nature to learning in the home, both
in terms of quality and quantity of input, to say nothing of questions
of urgency and need. L2 development in more naturalistic settings is
also generally different. Many immigrant learners, for example, con-
tinue to live and associate with fellow native speakers even if they need
to operate in the target language for work. And there are the two other
fundamental differences between L1 and L2 development that we have
already mentioned: the question of age and the role of the native
language.
The question of age is usually referred to as the Critical Period
debate. Like many notions in language development, the Critical
Period Hypothesis was originally proposed for native language devel-
opment.12 The claim that natural language development is limited to a
particular age has seen the most research, however, in second lan-
guage studies. The Critical Period, as proposed by Lenneberg (1967),
is tied to biology with the time frame for language acquisition often
linked to maturation. Specific evidence for the start and finish of this
time frame, however, is hotly contested. The influential early study by
Johnson and Newport (1989) separated L2 speakers who arrived in
the US between the ages of 3 and 15 from those who arrived after the
age of 17. They found that, on all proficiency measures, those with an
earlier age of arrival outperformed those who arrived later.
 Language as a Biological Property 59

Like most influential early work, this particular study has been
carefully analysed, criticised and subject to many replications. As we
mentioned in Chapter 1, the generalisation that younger is better is not
straightforward since younger learners are in fact, not better as
regards short-term development; more mature learners learn faster,
at least initially.13 Indeed, both how young and how much better are
not agreed upon.
Importantly, this does not mean that older people cannot be
successful learners. In addition to examples that you may know of
personally, there are documented cases of people beginning to learn a
language as adults and managing to achieve what seems like a native
level of proficiency.14 For there truly to be a biological Critical Period,
it would have to be the case that no adults ever manage a native-like
proficiency. Because this is not the case, most researchers now think in
terms of a sensitive period, instead of a critical period. Learning in the
sensitive period makes higher ultimate attainment more possible, but
there is no universal cut-off point for language development. Though
adults have an advantage in terms of how quickly they can learn
language in the short term, in the long term younger learners can catch
up and pass the adults if given a sufficient learning environment.15
It also seems clear that age affects different aspects of language
development differently. Generally speaking, phonology seems con-
sistently more difficult for older language learners than syntax.
Certain aspects of syntax, such as verb–object word order or question
formation, by contrast, seem not to be affected by age. Yet other
aspects, such as the adverb–verb order we discussed above, seem
problematic. This finding that age effects vary in terms of aspect of
language has led to the conclusion that there are multiple sensitive
periods. As we will see in the next section, differences in development
from one aspect of language to another constitute a conclusion that is
clearly emerging from research in second language development.

Differences by Linguistic Domains in Second Language

Development
Given the complexity of language, it is not surprising that the process
of language development is complex as well. In the discussion so far
we have focused on one aspect of language at specific points of
development in order to understand the process better. Focusing on
one aspect alone, however, cannot reflect the reality of the complex
situation. We will try to put some of the pieces of the puzzle together
60 Language Teaching

as we discuss one set of well-known case studies which focus on an

area of language which does not require sophisticated understanding
of formal linguistics: the tendency for non-native speakers of English
to leave out inflectional morphemes when they speak.
Lardiere’s (1998) study of Patty, a Chinese adult leaner of English,
and White’s (2003a) study of SD, an adult Turkish learner of English,
focused on learners whose English developed naturalistically as adults
living and working in an English-speaking environment. These studies
provide empirical evidence for a phenomenon well known to English
language teachers. Recordings of their speech over many years docu-
ment that both these learners regularly omit articles, the and a, as well
as plural -s, 3rd person singular -s on verbs and past tense -ed, and
they make errors of form with the copular verb to be. While they both
make these errors, there are differences in amount. Patty makes
significantly more errors with inflectional morphology than SD. This
can be attributed to L1 transfer since the Chinese dialects that Patty
speaks have very little verbal morphology, while Turkish is rich in
morphology. At the same time, the very similar rates of errors on
articles bolster the argument that L1 affects interlanguage develop-
ment since neither Mandarin nor Turkish has articles.
Though the study of second language development has always
recognised the existing native language as a potential factor in devel-
opment, in the early days of generative SLA there was a tendency to
downplay this factor because it was associated with the behaviourist
ideas of habit formation. Language ‘habits’ from the native language
were seen as transferring to the second language context. While this
understanding of L1 transfer was refuted, the association between the
terminology and the concept meant early researchers hesitated to use
it. With the passage of time, however, this association has mostly
disappeared.
Yet much of the research in L1 transfer still takes on a Contrastive
Analysis Hypothesis type way of thinking. Though it was largely
refuted, it is hard to shake the Contrastive Analysis intuition that
problems will occur where there are differences between two lan-
guages and development facilitated where they are the same. And
results like Patty’s and SD’s continue to reinforce this intuition.
However, a closer look shows that the finding is more complex than
a simplistic Contrastive Analysis view. What is interesting in both of
these case studies is that, irrespective of L1 transfer, the lack of
accuracy tends to be a result of omission, not one of using the incorrect
form, apart from forms of to be. In fact, White found that SD uses the
 Language as a Biological Property 61

wrong verbal morphology only forty-nine times in 1,744 obligatory

verbal contexts; that is a rate of just 0.02%. That the lack of accuracy
is to do with verbal inflection being left out and not the use of the
wrong form is a systematic feature of development that suggests some
kind of production error.
Despite the errors relating to inflectional morphology, there is
strong evidence for mastery of related underlying rules of syntax.
Both Patty and SD are completely accurate with subject and object
pronouns, with accuracy rates of 99 to 100%. These findings are
significant because both obligatory subjects and pronoun forms are
understood to be a part of the same parameter as the one that requires
verbal inflection. Thus, the fact that subjects are supplied and pronoun
forms are consistently correct when verbal inflection is inconsistent is
evidence that underlying rules of syntax are present. The lack of
accuracy with inflectional morphemes like verb inflection and articles
must have something to do with some aspect of language production
and not knowledge of the grammatical constraints. The main con-
clusion, then, is that some aspects of grammar, such as pronouns and
subjects, seem to be easier to acquire and less affected by the learner’s
L1 than others: namely, articles or verbal inflection. Additionally, the
general finding that different aspects of language develop differently is
again supported.
Recent developments in technology have allowed us to explore this
idea of variable development from a physical point of view. Research-
ers in the 1990s were overly optimistic that research using new
technologies such as event-related potentials (ERP) studies, which
measure electrical activity in the brain, and fMRI scans, which
measure blood flow as a way of mapping brain functioning, could
pinpoint which areas of the brain were associated with which specific
aspects of language.16 It quickly became apparent, however, that even
this sophisticated technology was not enough to truly tease apart all of
the elements implicated in language. While research cannot exactly
pinpoint where different facets of language are stored, it can and does
show that the different aspects are processed differently. The clearest
result that is emerging is a dissociation between grammatical and
semantic processing. Study of second language speakers shows that
age has a direct effect on syntactic processing. Those learners who
began learning as young children show the same syntactic processes as
native speakers, while older learners do not. For semantic processing,
closeness to native-like processing seems to depend on proficiency in
the language, regardless of age.17
62 Language Teaching

Of direct relevance to our discussion of function words and mor-

phology, neuroimaging research seems to suggest that, as with more
core grammatical features of language, age seems to be the main factor
for determining whether second language speaker processing ability is
native-like or not. Yet what is shown by behavioural studies is that,
despite differences in processing as revealed by physical studies of the
brain, learners do seem to be able to know even very subtle gramma-
tical properties of language. What is emerging from the research,
therefore, is a view in which some areas of language, such as func-
tional morphology, seem to cause difficulty in language development.
Yet it is important to distinguish between knowledge as evidenced
through research and production ability as perceived in real-life
interaction. By knowing which aspects of language will develop
‘naturally’ and which need explicit attention, there is hope that, with
perseverance, even the most intractable aspects of language can be
overcome by diligent language learners with the help of highly
competent language teachers. Thus, even the most abstract theoretical
findings have useful implications for language teaching. We will come
back to this point when we discuss the implications of SLA research in
Chapter 5.
Returning to our case studies, the research on Patty and SD has also
been influential in discussions of fossilisation. Fossilisation may occur
because of some biological constraint on language development,
through some kind of sensitive period. It may also be a result of
insufficient quantity and/or quality of input. Additionally, the influ-
ence of the native language may keep the second language from
developing. The two case studies discussed here show that L1 transfer
can have persistent effects even at end-state levels of proficiency. This
somewhat pessimistic view does not sound like good news for the
language teacher. Would it be possible for these or any seemingly
fossilised learners to overcome this kind of inaccuracy in inflectional
morphology?
Fortunately, a recently proposed framework allows us to revaluate
this view of inevitable failure. The Modular On-line Growth and Use
of Language (MOGUL) proposal of Sharwood Smith and Truscott
allows for a different, and pedagogically more encouraging, under-
standing of what is happening in the case of Patty, SD and the many
other very proficient English speakers who none the less can be
somewhat inconsistent in supplying inflectional elements in their
language production. We will wait to discuss MOGUL until the next
chapter because it contains elements that are not strictly limited to a
 Language as a Biological Property 63

generative view. We will instead conclude this chapter on language as

a biological property by asking whether it can account for all that we
know about second language development.

CONCLUSION: WHAT A COMPLETE THEORY OF SLA NEEDS TO EXPLAIN

In a very useful book VanPatten and Williams (2007) outline ten

phenomena which have been observed in SLA. Their reason for doing
this is to challenge the different branches of linguistics to explain what
we know about SLA adequately. Those ten observations are:

1. Exposure to input is necessary for SLA.

2. A good deal of SLA happens incidentally.
3. Learners come to know more than what they have been
exposed to in the input.
4. Learners’ output (speech) often follows predictable paths with
predictable stages in the acquisition of a given structure.
5. Second language learning is variable in its outcome.
6. Second language learning is variable across linguistic subsys-
tems.
7. There are limits on the effects of frequency on SLA.
8. There are limits on the effect of a learner’s first language on
SLA.
9. There are limits on the effects of instruction on SLA.
10. There are limits on the effects of output (learner production) on
language acquisition.
(VanPatten and Williams 2007: 9–12)

The rest of VanPatten and Williams’s book consists of chapters

written by SLA researchers from different linguistic frameworks.
The chapter devoted to the generative perspective is written by Lydia
White. In that chapter, she explains how the biological view of
language addresses Observations 1, 3, 7, 8 and 9.
We have also noted that, despite any innate UG-constrained pre-
disposition for language that may still exist among second language
learners, input is crucial for language development (Observation 1).
The above discussion of the poverty of the stimulus phenomenon, in
which learners come to know more than what they have been exposed
to in the input, is evidence for Observation 3. The limits on frequency,
Observation 7, are also clear from studies like Kanno’s work on the
Overt Pronoun Constraint. With very few instances in the input,
64 Language Teaching

learners acquired this constraint. If you interpret this same observa-

tion in the opposite direction to say that, even with robust frequency,
there are limits, then the case studies of Patty and SD also support this
observation; despite the very frequent occurrence of inflectional
morphology in the input, these learners have not mastered them in
their own speech. The limits of the influence of the first language on
second language development (Observation 8) have been seen since
the early morpheme order studies showing similar orders of acquisi-
tion regardless of native language. The null subject parameter studies
discussed in this chapter also show this result, as learners from
different language backgrounds are able to re-set this parameter.
And finally, White’s own study of adverb placement clearly shows
that the effects of instruction are limited, Observation 9, at least in the
long term.
What about the other five observations? The observation that
learners’ output often follows predictable paths with predictable
stages in the acquisition of a given structure (Observation 4) has also
been supported by the generative view. We saw this in the early
morpheme order studies, and also in the parameter re-setting literature
showing that some learners go from their native language parameter
to a mid-point in which their interlanguage includes a setting that is
neither the native nor the target language setting.18,19 We have seen
that Observation 6, stating that second language learning is variable
across linguistic subsystems, is an area that has recently attracted a fair
amount of attention in the generative research. While it has long been
acknowledged that phonology and syntax seem to develop differently
among adult L2 learners, research on the subsystems of syntax vs
morphology vs the lexicon is also finding differences in development.
Emerging technologies suggest that the differences may reflect qua-
litative difference between types of linguistic knowledge. The ability
for modular linguistic knowledge to develop later in life may differ, for
example, when compared to non-modular knowledge. This is cur-
rently cutting-edge research with numerous hypotheses being pro-
posed and tested.
The remaining three observations (Observation 2, 5 and 10) are not
readily accounted for by the generative view of language. Note that,
for two of the three observations, however, it is not necessarily the
case that a biological view of language contradicts them; it is more the
case that generative linguists have not felt the need to address these
points. In saying that a good deal of SLA happens incidentally
(Observation 2), VanPatten and Williams are pointing to research
 Language as a Biological Property 65

that shows that, while engaging in tasks that explicitly highlight some
aspect of language, learners can come to know other aspects of
language that are not being explicitly highlighted. Because generative
research assumes this, there has been little direct study of the process
of incidental learning itself. Instead, the focus has been on the proper-
ties of the interlanguage that develop from incidental learning. Simi-
larly, generativist researchers have not been interested in the
observation that there are limits on the effects of output (learner
production) on language acquisition (Observation 10). This question
has not been relevant in a view which depends so crucially on a
process whereby an internal mental system reacts to external input.
Production in this view is simply a product of what the internal system
is capable of at any given moment in time.
That second language learning is variable in its outcome (Observa-
tion 5) may seem problematic for the biological approach because we
are all considered biologically the same in terms of predisposition for
language. We can point to differences arising from a difference in L1
transfer from one native language to another, but the variability
among learners with the same L1 would not be explained this way.
Perhaps answers will emerge from the current research showing that
different aspects of language develop differently. Aside from this,
generativists explain learner variability by looking more closely at
differences in the nature of the input.20 Taking a broader view, there
are other clear limitations to the generative view, including the
absence of research on the functional aspects of language use, as well
as discussion of the role of social context in language development.
These two charges form the foundation of Functional and Socio-
cultural views of language, respectively. We will briefly explore these
and other views of language in the next chapter before considering a
recent approach which tries to account for a fuller range of research
relevant to language learning and teaching.

FOR DISCUSSION

1. The examples of linguistic ambiguity given in this chapter are based

on English. Can you think of examples of language ambiguity in
other languages? Try to think of examples in which the ambiguity
lies in the structure of the sentence as opposed to word-level
ambiguity (in other words, one word that has two meanings).
2. Consider the five properties of modular knowledge. Are they
relevant to knowledge of a second language?
66 Language Teaching

3. Some people argue that if, as argued here, there is evidence that
SLA is constrained by UG, then second language learners should be
able to acquire language just like native language speakers. Explain
why this logic is flawed.
4. All of the SLA research presented in this chapter takes a mentalist,
or psycholinguistic, point of view. What effect, if any, would non-
mentalist factors such as the following have on the claims in this
chapter: motivation, aptitude, level of education, gender, cultural
background?
5. Consider each of the ten Observations about second language
development in terms of your own personal experience. Do you
have any personal examples to support them? Do you have any
experiences which seem to contradict any of them?
6. Discuss each of the ten Observations in terms of classroom lan-
guage teaching. What implications can you draw?

NOTES

1. The original version can be found in Chomsky (1981) with an updated

version in Chomsky (1986).
2. To read the original proposal for Minimalism, see Chomsky (1995). For
an introductory treatment of Minimalism, see Radford (1997). For a
more technical treatment, see Adger (2003).
3. For an early example, see Diller (1971). More recent examples include
Ellis (1997) and Kramsch (2003).
4. For more discussion, see Lado (1957) and Sharwood Smith (1994).
5. One notable exception has been Patsy Lightbown; see Lightbown (1985,
2000, 2003) and Lightbown and Spada (2006).
6. See, for example, Thomas (1965) or Rutherford (1968).
7. While generativists such as Krashen or Dulay and Burt (1973, 1974) see
this as evidence for an in-built syllabus guiding language development,
cognitivists such as Goldschneider and DeKeyser (2001) look to the
nature of the input itself to explain the regularity.
8. Other work exploring parameter values that are not found in the L1 or
the target grammar has focused on the properties of reflexives (Finer
1991; Finer and Broselow 1986; and MacLaughlin 1998) and case
checking (Schwartz and Sprouse 1994).
9. See Rothman and Iverson (2007) for research on the Overt Pronoun
Constraint in Spanish.
10. Other poverty of the stimulus studies have focused on other subtle areas
of semantics. See, for example, the work of Laurent Dekydtspotter and
colleagues (Dekydtspotter, Sprouse and Anderson 1997; Dekydtspotter,
Sprouse and Thyre 1998; Dekydtspotter, Sprouse, Swanson and Thyre
 Language as a Biological Property 67

1999; Dekydtspotter 2001; Dekydtspotter and Hathorn 2005) and

Marsden (2005, 2008, 2009).
11. Schwartz and Gubala-Ryzak (1992) reanalyse White’s data, arguing that
there is evidence that, even in the post-test done immediately after
teaching, the learners are applying a linear rule and not showing knowl-
edge of the linguistic constraints on adverb placement.
12. For obvious ethical reasons, it is not possible to design Critical Period
studies in native language acquisition. Unfortunate ‘natural’ situations
that have arisen often include serious abuse or neglect which makes it
difficult to compare language development to usual native development.
13. For more discussion of the Critical Period Hypothesis and relevant
studies, see Hensch (2004) and Singleton and Ryan (2004).
14. For specific studies, see Birdsong (1992); Ioup, Boustagui, El Tigi and
Moselle (1994); Bongaerts, Planken and Schils (1995); White and
Genesee (1996); and Bongaerts (1999).
15. See note 7 of Chapter 5.
16. See Mueller (2005) for a review of ERP studies.
17. See Chapter 4 of Slabakova (2008) for a discussion of neuroimaging
studies and second language development.
18. For a well-known case study showing parameter re-setting in child L2
acquisition, see Haznedar (2001).
19. Many other studies from the late 1970s into the 1990s have supported a
view of developmental stages, with researchers offering explanations
from a range of theoretical perspectives from generative to cognitive
linguistic frameworks. Well-known examples include the very large
European Science Foundation (ESF) project of 1982–8 (see Klein and
Perdue 1997) and the Zweitspracherwerb, Italienischer, Spanischer und
Portugiesischer Arbeiter (ZISA) project (Meisel, Clahsen and Pienemann
1981).
20. See Piske and Young-Scholten (2009) and papers within for a discussion
of input.
 4
LANGUAGE AS A TOOL FOR COMMUNICATION

INTRODUCTION

We began this book by pointing out that there are many perspectives on
how to characterise language, and we have discussed how language
develops. The main thrust of the book, however, has been to present a
mentalist view, and the generativist view in particular, both because it is
such an influential view in theoretical linguistics and because it is the view
that receives the least amount of discussion in terms of the language
classroom. One reason for this is the conceptual distance between
abstract linguistic theory and the practice of teaching. Hopefully this
book is able to narrow this distance somewhat. It is not a coincidence
that the most formal view of language has had the least application in the
language classroom. As we began to see in Chapter 3, Generative
Linguistics relies on highly abstract constructs and formalisms which
can be daunting for anyone without thorough training in theoretical
linguistics. As may become apparent in this chapter, other approaches
have been more accessible. And more importantly, other approaches
consider application to the classroom as a natural objective for their
research. Generative linguists, on the whole, have not taken this view.
Unfortunately, there are limits on the extent to which we can
explore all the different approaches to language in a single book.
This chapter presents an overview of the major approaches, cate-
gorised here under the broadly defined labels of Functional, Socio-
cultural and Cognitive Approaches. To some extent, these all stand in
opposition to the generative view. Some are opposed to what they see
as a strictly rationalist approach whereby language is decomposed
into its smallest possible units. This kind of atomistic method invites
the criticism of not seeing the forest for the trees. The other difference
is deeply philosophical, as these approaches see language and all of the
grammatical principles associated with language as the outcome of
meaning and thought instead of viewing grammar as its own system
independent from meaning.
 Language as a Tool for Communication 69

In other words, these approaches differ from the Generative Ap-

proach in terms of the importance they attach to meaning. While
generativists view meaning as a by-product of what can be expressed
though formal linguistic rules, these approaches see meaning and the
making of meaning as the driving force for language. To the extent
that all of these approaches have an interest in the way we use and
understand language, they could all, broadly speaking, be considered
functional. However, the label ‘Functional’ will be reserved for a more
specific view with Cognitive and Sociocultural approaches distinct
enough to warrant separate labels.
Because they are all interested in the way we use language they can
be (and often are) viewed as in fundamental opposition to a generative
view, as suggested in our brief discussion of the form–function divide
in Chapter 1. The difference is a real one for linguists concerned with
fundamental questions of how language is organised. As we have seen,
for formalists, the building blocks of language are grammatical
constraints like subject–verb agreement. For functionalists, meaning-
ful concepts like time and space form the foundation of language.
While for most linguists these are irreconcilable differences, the view
taken in this book is that there is no reason why we should not
research and understand both form and function/use.
In fact, we can go further to promote a more conciliatory view that
does not dwell on opposition, but instead focuses on the extent to
which the different approaches are asking different questions. An even
more optimistic view is that there is a degree of convergence which is
beginning to emerge among the opposing theories. Regardless, the fact
that a range of views exists points to the reality that no single
approach seems to be able to account for all of the factors relating
to language, and its development and use. Because the different
approaches have different useful contributions for understanding
language development, there is every reason why a good language
teacher should be well aware of the range of ideas.
Though Generativism has made great gains in explaining cross-
linguistic regularities in language and language acquisition, there are
some very large areas that are not included in the Generative Ap-
proach. The function of language as a communicative tool is a glaring
omission. In this chapter we will discuss how Functionalism is an
approach that takes language use as its primary emphasis. Questions
about language function inevitably bring us to questions about social
context and societal norms. This is what interests researchers taking a
Socioculturalist Approach. The final approach to be discussed in this
70 Language Teaching

chapter, the Cognitive Approach, is one that has seen much influence
from psychology. It is similar to the Generative Approach in its
interest in the mental processes underlying language. But as we will
see, the understanding of these processes is fundamentally at odds and
much more in concert with a Functionalist Approach instead. After
exploring each of these three approaches, we will consider a recent
proposal that provides a framework that accommodates a wide range
of views. We will use this framework to bring all the discussions of
language and language development together, providing a unified
basis from which to consider specific implications for the language
classroom in the remainder of this book.

THE NOTION OF COMPETENCE

In the last chapter we referred to the ‘competence’ that speakers have

of their (native) language. The idea presented there was that all
humans develop a perfectly formed grammatical competence – unless
there is some kind of pathological problem. The variation among
speakers in their ability to use language, by contrast, has been
attributed to differences in ‘performance’. Chomsky’s now famous
claim was: ‘We thus make a fundamental distinction between com-
petence (the speaker-hearer’s knowledge of the language), and per-
formance, the actual use of the language in concrete situations’
(Chomsky 1965: 4). Chomsky’s terms developed the contrast made
by Swiss linguist, de Saussure, in the late 1800s between langue, which
is the system that regulates language, and parole, the actual utterances
produced by speakers. For generativists, this distinction provides a
useful way to limit the scope of their research. For many others,
however, this distinction has been seen as not useful, to the point of
being absurd. Dell Hymes countered Chomsky’s distinction by pro-
posing what he called communicative competence to characterise
what speakers need to know in order to be communicatively compe-
tent in a speech community.1 This includes the social and cultural
knowledge needed to understand and use linguistic forms in context,
and in line with cultural norms.
The idea of communicative competence was well received by
researchers interested in developing more effective means for teaching
language. Canale and Swain (1980) refined Hymes’s ideas by propos-
ing four dimensions of communicative competence. Firstly, they
acknowledged that grammatical or linguistic competence is an essen-
tial part of being communicatively competent. This type of compe-
 Language as a Tool for Communication 71

tence includes the linguistic forms which are traditionally subsumed

under the category of ‘grammar’, including rules of sentence structure,
word formation and pronunciation.
The second dimension is sociolinguistic or pragmatic competence.
This competence refers to the ability to use language to achieve
communicative goals, drawing on the norms and conventions of the
culture in question. This kind of competence depends on a knowledge
of culture and the way we use language to participate in social context.
For example, if you were invited to someone’s home for dinner and
asked, prior to the meal, whether you would like to use the bathroom,
chances are you would understand that your host was inviting you to
wash your hands. This product of cultural experience was not the
communicative intent understood by a friend of my father’s, a recent
immigrant. When he did not return, my father went and discovered him
in the bath – he did not want to offend the hosts by refusing their
generous offer! In order to understand the propositional content of
many of the exchanges that we have every day correctly, we must have a
shared cultural knowledge, based on the norms of the culture.
In addition to these competences, Canale and Swain proposed
discourse competence, which regulates the way in which language,
whether spoken or written, is interconnected. Discourse competence
in writing is understood in terms of coherence. Words and phrases
such as however, nevertheless and therefore, for example, connect
ideas, helping to hold the text together in a more meaningful way.
Other more ‘grammatical’ discourse devices are personal pronouns
like him or her, which are used to refer to a person already mentioned,
and demonstrative pronouns like this and that, which also pick out an
aforementioned referent. You can find an example of this latter type of
discourse marker in the opening five words of this paragraph (as well
as the examples in this very sentence).
In speaking, discourse markers signal a range of functions including
interruption, disagreement, change of subject, and so on. Examples of
phrases that serve these three functions, respectively, are given in (1):

(1) a. Sorry, but could you say that again?

b. Maybe so, but couldn’t we X instead?
c. So anyway, what do you know about . . .?

You may have noticed that two of these examples use the word but and
two use the word so. These little words are very hard to define in
isolation, but serve a range of functions in discourse, making it difficult
72 Language Teaching

to articulate particular ‘rules’ for specific discourse markers. It is these

kinds of linguistic phenomena that have inspired some linguists to insist
on the importance of function in language. Systemic Functional Lin-
guistics has developed very sophisticated analyses of how texts, both
spoken and written, cohere to convey particular meaning. As we will
see in the next section, for functionalists, the meaning of words like so
and but depends on both neighbouring words/phrases and the wider
context of discourse in which these words were uttered.
Before moving to our discussion of Functionalism, however, we
need to consider the fourth and final competence proposed by Canale
and Swain. Strategic competence refers to the ability to keep com-
munication going, an important skill for language learners. A knowl-
edge of strategic competence will help a speaker to negotiate language
situations when there is a breakdown in communication. This may
require requests for clarification on the part of the hearer or a repair
by the speaker if it becomes clear that there has been a miscommu-
nication. In discussing strategic competence, Canale and Swain point
out that speakers use both verbal and non-verbal strategies. While this
may sound like a step away from ‘language’, if the basis of language is
communicative function, then all aspects of communication are re-
levant, from gesture to context to shared cultural norms.

ACCOUNTING FOR MEANING: A FUNCTIONAL APPROACH

Communicative function lies at the heart of the Functionalist Approach

to language. While communicative function may sound like commu-
nicative competence, Functionalism actually rejects the notion of
competence because it sees no value in any division between ‘compe-
tence’ and ‘performance’. For functionalists, the one aspect of language
that is explicitly sidelined in the Generative Approach is, in fact, the
defining feature of language: meaning. The functional view sees lan-
guage as a system for expressing meaning, and thus the primary
purpose of language is to facilitate interaction and communication.
This does not mean that language does not have structure, but the
structure of language reflects functional or communicative uses, not
some abstract, pre-determined set of grammatical constraints.
The ‘father’ of functionalism, British-born Michael Halliday,2 was a
student of John Firth, whose ideas about language grew out of the
tradition of European linguistics known as the Prague School, a
linguistic tradition rooted in literary studies and semiotics. Semiotics
is the study of the relationship between signs and meaning. A word,
 Language as a Tool for Communication 73

for example, is a symbol which has an arbitrary relationship with the

thing that it signifies. The interest in meaning through semiotics, and
culture through literary studies provided a natural springboard for a
functional view of language, in contrast with the more anthropolo-
gically oriented interests of early American linguists who focused on
the structure of language. Functionalists are not necessarily dismissive
of structural forms, but see forms as a realisation of function. Their
interest is in exploring the mapping from function to form.
In fact, within what can be broadly labelled ‘functionalism’ lies a
wide range of approaches to the question of grammar and structure.3
At the one end, there are theories which may seem very much like the
structuralist Generative Approach in the sense that they also view
grammar as a system. These so-called ‘structural-functional’ ap-
proaches include Functional Grammar (Dik 1978) and Role and
Reference Grammar, as originally proposed by Van Valin and Foley
(1980). These contrast with other Functional Approaches which pay
less attention to grammatical structure, focusing on language at the
level of discourse and interaction instead. Because these approaches
are less interested in what is traditionally understood as ‘grammar’,
they can be viewed outside of the range of ‘formal’ theories about
language which limit themselves to the more narrow level of the word
and phrase. Going to the furthest end of the spectrum, there are
functionalists who believe that there is no grammar at all, that instead
grammar is a set of patterns that linguists have imposed on language in
order to make sense of it. Before looking at the ‘Construction
Grammar’ view, we will briefly explore a more middle-ground view
within Functionalism, as exemplified in Halliday’s Systemic Func-
tional Grammar (SFG).
Because the purpose of language is communication, for SFG a
sentence is not seen so much as a combination of subject, verb and
object, but as a vehicle for conveying information. As we saw in
Chapter 1, it is the roles that different parts of a sentence play, and the
relationship between these roles that are of interest to a functionalist.
In a verbal exchange, for example, there will typically be given
information that is already known to the speaker and the listener,
and new information; the two combine to make the sentence mean-
ingful. Functionalists use the term theme to refer to the starting point
of a message, and rheme, for the idea that the speaker (or writer)
moves to in order to develop or add to the theme. As in many
languages, in English the theme tends to occur at the beginning of
the sentence, while the rheme occurs at the end.
74 Language Teaching

(2) Our neighbour has not yet returned our lawnmower.

theme rheme

By nature, this idea of theme and rheme relies crucially on context,

perspective and meaning.
The most notable aspect of Halliday’s SFG approach is the analysis
of language at the level of text. SFG tends to take the text as the
starting point from which to work ‘down’, looking at sentences,
phrases and words, as well as ‘up’, looking at the context in which
the text is produced.4 While a lay person’s understanding of the word
text is usually limited to pieces of writing, for linguists text refers to
both written and spoken forms of discourse. SFG has provided a
framework to analyse texts in their broader contexts. For example, a
text depends on the nature of the activity with which the text is
associated, or field. Consider, for example, a university lecture. The
nature of a university lecture is very much defined by the academic
conventions of the university. Also important is the relationship
between the participants associated with the text. Given the label
tenor, this would, in our example, refer to the relationship between the
lecturer and the students. Additionally, there is the question of mode,
which is the medium through which the text is communicated. In our
example, a lecture is a spoken text of a relatively formal variety.
Within the text, the forms that are chosen by the writer/speaker can
reveal the communicative intent of the text as well. One linguistic
mechanism in SFG is the notion of grammatical metaphor. By using
the word metaphor, this term captures the way that grammatical
forms can be extended beyond their literal meaning to have a parti-
cular function. One well-known example of grammatical metaphor is
that of nominalisation. Consider the examples in (3):

(3) a. The great migration from Africa saw early humans

reach Europe about 1.8m years ago.5
b. Early humans migrated from Africa to Europe about
1.8m years ago.

The option of using a nominalised form of an action as the subject of

the sentence in (3a) serves a function beyond just acting as a gram-
matical subject. In (3b), the migration event invites disagreement
because it is presented as a claim. In the normalised form presented
in (3a), by contrast, the migration event itself is presented as known
information suggesting that it is an indisputable fact, with the time of
 Language as a Tool for Communication 75

the event offered as new information. By choosing to use the nomi-

nalised form, the writer reinforces a particular view of the historical
events, a very subtle but effective use of a grammatical form.
A functionalist view of language is interested in the work that
language does and the effect that it achieves. Words and phrases are
seen in terms of action. The example given in (4) illustrates this point:

(4) Next?

Even without context, this single word promises action, especially

with the added question mark. With context, this simple word is
capable of triggering a series of events. Imagine a stony-faced immi-
gration official looking beyond an exasperated traveller to the next
person in the queue. The speech act performed by this simple word is
to indicate that the communicative event has been concluded and that
the traveller should move on to allow the next person to approach the
desk. John Searle’s Speech Act Theory develops the idea that language
is functional.6 The importance of context is clear in this kind of
approach to language. Additionally, both spoken utterances and
written text are interpreted with reference to the listener/reader’s store
of cultural understanding. While some of these norms may be similar
across cultures, others will vary just as cultures vary. In this way,
language is very much seen as social behaviour.
Because language depends on social norms and context, a func-
tionalist analysis of language is often couched in terms of tendencies
and not rules. After all, it would be difficult to articulate hard and fast
rules when it comes to human behaviour. For many functionalists, an
aversion to rules extends to what may be seen as more core properties
of language as well. Regularities which have traditionally been cap-
tured by grammatical rules are seen by some as well-used patterns that
have developed within speech communities.
This notion is at the heart of Construction Grammar.7 A construction
is an identifiable linguistic pattern associated with a particular meaning;
it is a mapping between form and meaning. This mapping can be as small
as a morpheme or a word and as large as a phrase or even a whole
sentence. Building on seminal work by Charles Fillmore (1988), Gold-
berg (2006: 5) gives the following examples of constructions:8

(5) a. Morpheme e.g. pre-, -ing

b. Word e.g. avocado, anaconda, and
c. Complex word e.g. daredevil, shoo-in
76 Language Teaching

d. Complex word (partially e.g. [N-s] (for regular plurals)

filled)
e. Idiom (filled) e.g. give the Devil his due
f. Idiom (partially filled) e.g. jog <someone’s> memory
g. Covariational conditional The Xer, the Yer
(e.g. the more you think
about it, the less you
understand)
h. Ditransitive Subj V Obj1 Obj2
(e.g. he baked her a muffin)
i. Passive Subj aux VP (PP)
(e.g. the armadillo was hit by
a car)

Constructions like these become conventionalised by speech commu-

nities. This means that it is the users of language who determine the set
of constructions that belong to the language, and the degree to which
each construction is considered an acceptable part of the language. In
this way, constructions are a product of language use and not the
product of grammar rules. Thus, it is possible to take the next step and
claim that grammar itself does not exist.
Support for this view comes from work in Corpus Linguistics which
shows that much of language use is formulaic, consisting of set phrases
that are repeatedly used. The growing body of corpus data shows
regular use of collocations and colligations, words and phrases that
tend to co-occur. As shown in the examples below, there are alternative
ways of expressing the same concept. Corpus studies, however, show
that as language users we tend to use the more formulaic alternatives.
Collocations:

(6) a. weather forecast (compare with: weather prediction)

b. marked contrast (compare with: large contrast)
c. up and down (compare with: down and up)

Colligations:

(7) a. to put it another way (compare with: to say it in a

different way)
b. an endless stream of people (compare with: an endless
line of people)
c. incite a riot (compare with: incite a protest)
 Language as a Tool for Communication 77

While a view of language which says that there is no grammar may

seem problematic for language teaching, it is well known to both
teachers and students that most ‘grammar rules’ have exceptions. This
makes sense if there are not, in fact, any rules, but instead tendencies
that emerge from shared use of a language. This view finds support in
the observation that members of a linguistic community routinely
change the way they use language. The constantly changing slang of
young people provides one example, with words like wicked or cool
used to convey a meaning very different from the original.
Shifts in language occur at the more grammatical level as well. And
geographical differences can be traced through changes in accents as
dialects continuously diverge. One grammatical shift exemplified in
(8) shows how it has become fairly common in English to use the third
person plural pronoun they in a context in which the traditional
grammar ‘rule’ would require a singular form. Notice that the verb
also agrees with the plural their, and not the main clause subject
everyone, as expected by traditional grammar:
(8) . . . and nearly everyone who’s bought their houses have
altered the doors.9
Use of the generic they is a result of a change in cultural norms; people
in English-speaking countries have become uncomfortable with the
use of the masculine form to refer to both men and women. The
preference for appropriate language has over-ruled the structural
requirement for number agreement in (8).
Our discussion of the functionalist view has brought us to a radical
point: that there is no grammar. This conclusion, in fact, even goes
further than many functionalists would. What is agreed upon, how-
ever, is that language is a system for communication which is central
for explaining what it means to be human. The functionalist view of
language also sees the development of language as dependent on
language use. Language development, whether native or non-native, is
driven by a need to communicate. A learner develops a repertoire of
language through interaction with members of the speech community.
This idea of speech community is even more important for the
approach discussed in the next section.

ACCOUNTING FOR SOCIAL CONTEXT: A SOCIOCULTURAL APPROACH

There are many researchers who have been critical of the more
mentalist approaches to language learning, arguing for the importance
78 Language Teaching

of social context. These include sociocultural theory, cultural histor-

ical activity theory, activity theory, conversational analysis and eth-
nomethodology. As it will not be possible to discuss each of these in
turn, this section will bring together some of the main ideas behind
these views. I will use ‘Sociocultural’ as an umbrella term and will refer
to this as an ‘approach’ because the ideas here go beyond the specific
claims of sociocultural theory.
Proponents of Sociocultural Approaches to language have been
vocal in their criticism of researchers who reduce language speakers to
their mental processes. In a well-known paper challenging the field of
linguistics, Firth and Wagner argue that meaning is not a brain-to-
brain activity, but rather ‘a social and negotiable product of interac-
tion, transcending individual intentions and behaviours’ (1997: 763).
For a sociocultural theorist, a learner is much more than their
grammatical competence, and language is much more than input.
Language is a resource for participating in everyday life; it is an
integral part of social processes. Socioculturalists speak of language as
a cultural artefact, both as a product of and contributor to language
socialisation. In short, language is inextricably tied to social context.
Because of its emphasis on the social nature of language, this view
often takes a more political stance, acknowledging the power relations
that exist among members of a community. It sees language users as
both contributing to and affected by language norms. For this reason,
much of the work of language learning is to become competent in the
cultural, social and political norms of particular communicative
contexts. In this way, language is a tool of mediation, connecting
the individual and his/her social environment, or to put it another
way, language mediates between the self and the external world,
allowing an individual to translate their thoughts into social action.
But this mediation is a two-way process; language also allows humans
to think and talk about ideas and events. In other words, language also
mediates in the opposite direction – from the external to the internal. This
latter point is one of the more controversial aspects of this approach.
Language is not an abstract system distinct from use or distinct from
cognition. In addition to being a means for communicating with others, it
is seen as a tool for thought. In other words, it is viewed as a means for
mediating cognition, as a tool that allows for mental activity. What
makes this controversial is the idea that this kind of linguistic mediation
is required in order for humans to make sense of the world. Moreover,
language is understood to be a prerequisite for the development of
consciousness. Thus language is not a set of inborn or innate predis-
 Language as a Tool for Communication 79

positions, but instead a socioculturally constructed mediational means

both for participation in society, and for basic cognition.
While the best-known contemporary proponent of the sociocultural
view is James Lantolf,10 the original ideas date back to the Russian
psychologist, L. S. Vygotsky (1896–1934) and the earlier part of the
twentieth century. Vygotsky’s ideas were rejuvenated and applied to
the second language learning context by applied linguists in the early
1990s. For this reason, the sociocultural view is sometimes also called
the ‘neo-Vygotskyan’ view. Vygotsky was concerned with develop-
ments in psychology at that time, and he explored questions of language
development primarily among children. In the context of development
and learning, Vygotsky argued that learning is socially mediated,
requiring interaction. He went on to propose principles for language
development, a number of which have been taken up by contemporary
researchers interested in language learning and teaching.
Early stages of development depend on so-called other regulation,
which means looking to someone else who is more skilled. Through
supportive interaction, the learner will be able to appropriate knowledge
into consciousness. This supportive interaction has come to be known as
scaffolding, which is the idea of providing assistance and direction for the
learner to support their development. While this notion is relevant to
learning in general, it has been seen as especially useful in the context of
language development.11 Through scaffolding, a learner has the optimal
conditions for development. The range of potential development is
known as the zone of proximal development (ZPD). Within the ZPD,
the learner moves from a dependent to an independent state.
Perhaps one reason why this idea has many supporters is because it
is a view that looks forward to what can be achieved, instead of back
to what has already been achieved. And it highlights what an in-
dividual learner may be able to achieve independently instead of
assessing the learner’s current abilities (or lack thereof). It also avoids
the common practice of measuring the learner against some idealised
target. Another reason why this view has been well received is its
emphasis on the value of collaborative learning. The accomplishments
of the individual are very much dependent on collaboration with
others through scaffolding and other regulation.
For this reason, the process of learning is seen as one of co-
constructing knowledge. As such, learning is by definition a social
process, especially in the early stages. One aspect of the learning process
seen as crucial by Vygotsky is imitation. From a contemporary
perspective, imitation is viewed as somewhat suspect because of more
80 Language Teaching

recent developments in our understanding of FLA. However, in the neo-

Vygotskyan view, imitation is not seen as a passive, behaviourist
phenomenon, but instead as an active process. Much emphasis is put
on delayed imitation, as a kind of personal rehearsal. This is known as
private speech, a type of inner speech directed to oneself to mediate
mental behaviour. This private speech is seen as a tool for thought.
Mediation through private speech can lead to self-regulation, which is a
developmental stage beyond other regulation. Self-regulation is an
advanced stage which will come through sufficient scaffolding. Another
process that can lead to self-regulation is object regulation, whereby
learners use physical objects to help them to develop their knowledge.
These various processes all contribute, in time, to the more indi-
vidual stage of self-regulation. Throughout, development relies on
internalisation, which is the development of psychological functions.
This process regulates the relationship between the individual and the
social environment, while at the same time the social environment
drives internalisation. The whole system is seen as dynamic – language
is a process of developing social norms, while at the same time social
norms are shaped by the use of language.
One very recent development from this line of thinking has seized
on this dynamic notion to propose that language should be seen as a
system that is constantly emerging. Diane Larsen-Freeman and Lynn
Cameron have applied the idea of Complexity Theory to language,
viewing language as in a perpetual state of adaptation.12 As such,
complex systems are characterised by a constantly changing interplay
between the existing system and the pressure points affecting the
system. This can give rise to an appearance of stability, but in fact such
a state does not really exist. In this dynamic systems approach, an
individual’s brain is seen not as a container for language, but as
shaped by language. This very new view is quite different to the other
approaches we have explored so far.
Another outgrowth from the sociocultural view has been an increased
exploration of the social context. Within a Situated Learning Theory
approach, some researchers focus on language in terms of so-called
communities of practice, which are identifiable groups with shared aims,
norms and functions.13 As members of society, we typically belong to a
number of communities of practice, each with habitual ways of doing
things based on a set of symbolic resources. There is a long tradition of
viewing learning in formal educational settings as a type of socialisation.
The community of practice view acknowledges this function of educa-
tion and seeks to teach learners explicitly to be able to participate
 Language as a Tool for Communication 81

successfully in the communities of practice that are relevant to them.

Many also take one further step towards a politicised view to explore the
sociopolitical implications of becoming a member of any particular
community of practice. This Critical Theory approach is interested in the
importance of power relations within particular communities.14 The
ultimate aim in a classroom setting, then, is to empower learners to be
able to navigate successfully through the complexity of the communities
of practice which they are likely to encounter.
In sum, sociocultural views of language are concerned with the
ways in which language affects society and society affects language.
This has led to research in social context and the interplay between
that context and language. Focus on the individual explores how the
learner can become an active part of this social network. In doing so,
there has been concern with the process of language development, but
at the individual level this process is seen as tied up in the social
context. Even the most clearly psychological process – that of inter-
nalisation – rests on the need for interaction, as self-regulation comes
out of collaboration and reliance on others.
As the emphasis in this approach has been primarily on development,
one criticism of this approach is that the description of what language is
has been limited to the idea that language is a mediating tool. This
somewhat abstract view of language may not be helpful for determining
the linguistic aims within a language teaching lesson, but it does provide a
starting point for asking how language should be taught since it relies
crucially on collaboration and interaction with others. Those socio-
culturalists who are interested in the more structural properties of
language tend to align themselves with functionalist characterisations
of linguistic structure since language is seen as a tool for making meaning.
Another criticism of the Sociocultural Approach is that there is not
enough emphasis on the cognitive processes implicated in language
use.15 Responding to this, the Sociocognitive Approach recognises the
role of society in meaning-making while also focusing more on the
cognitive side of the equation.16 The premise is twofold: that the
structure of language affects how speakers think and at the same time
is limited by the constraints of human processing. An example of
processing constraints is the tendency for languages to limit the
number of clauses that occur in a single sentence. The idea is that
there are limitations in how much complexity a person can process
during communication. This brings us to developments in psychology.
In the next section we consider the view of language that holds
cognition to be central its approach.
82 Language Teaching

ACCOUNTING FOR EXPLICIT, PROCEDURAL LEARNING: A COGNITIVE

APPROACH

The Cognitive Approach to language is based on research in psychol-

ogy showing that the brain is a system of neural connections. This
view of mental processing underpins the cognitive understanding of
how language is organised in the mind. While both cognitive and
generative linguists are interested in the mental processes underlying
language, the fundamental difference lies in their understanding of the
nature of linguistic knowledge. As we have seen, for generativists,
language knowledge is unique, distinct from other kinds of knowl-
edge. Cognitivists, by contrast, view language as the same type of
knowledge as other kinds of knowledge. In other words, language
makes use of the same types of processes and knowledge stores as
other kinds of knowledge do.17 In this view, language is a system of
connections, not an inventory of symbols and abstract rules. For this
reason, this approach can also, broadly speaking, be referred to as
Connectionism. Because language has been seen as just another type
of knowledge, there has been less of a tradition for research on
properties of language itself within the Cognitive Approach, yet recent
work in construction grammar is quickly changing this. We discussed
the construction grammar approach in the section on Functionalism.
This is one example of how these two approaches overlap.
Yet, the Cognitive Approach has other influences that make it
distinctive enough from Functionalism to treat it as its own approach.
In particular, there is a rich tradition of research on the computational
tools implicated in language use and development. Cognitivism owes
much to the seminal work of early psychologists like Jean Piaget, in
addition to early linguists like Benjamin Whorf. Evolving ideas from
developmental psychology, just as language is seen as part of a larger
set of knowledge, language learning is understood to develop like
other kinds of learning. It is characterised as explicit and procedural.
Through associative learning stores of language knowledge build up
over time, based on the input and experiences to which the learner is
exposed. Additionally, the processes available to human cognition
shape the way knowledge is organised.
One crucial process is captured in Prototype Theory, a theory that
addresses the basic human cognitive ability to classify as a way of
making sense of the world. A prototype is a kind of ‘best example’ or
most typical representation of any given category.18 To take a com-
monly used example, the prototypical ‘bird’ is an animal with specific
 Language as a Tool for Communication 83

attributes such as wings, feathers and a beak, which lays eggs and builds
nests. Yet, categories are not black and white. Instead, there are ‘good’
and ‘less good’ examples of any category. Compare a robin with a
penguin. The penguin does not fit the prototype of a bird very well as it
lacks some of the core attributes of the bird category, but it still qualifies
as a member. This blurring at the edges of a category, exemplified by
our penguin, has led to the notion of fuzziness. The lack of concrete
definition for notions such as categories is the type of thing that draws
criticism from generative linguists who, historically, have insisted that
language does have black and white categories.19
Eleanor Rosch was the early pioneer who developed the term
prototype in the 1970s. Her original aim was to address the large
philosophical question that asks which is more primary, concepts
themselves or language? The concept that ideas, or the way that the
mind conceptualises, might be affected by the way a person’s language
characterises the ideas is attributed to the linguist Benjamin Whorf of
the early 1900s, who, along with Edward Sapir, proposed ideas which
came to be known as the Sapir–Whorf Hypothesis. In its most extreme
form, this view states that language determines how people think. Less
extreme than linguistic determinism is linguistic relativism, which
claims that language affects how people think. Research in Prototype
Theory investigates whether there are connections between the way a
language labels categories and the way people think about the
associated concepts. The seminal work of Rosch has been developed
further by other well-known researchers, such as Labov and many
others, who used experimental methods from psychology to test the
notion of prototypes and categorisation.20
For cognitive linguists, categories stored in the mind find labels
through words. Thus research on the mental lexicon is fundamental to
cognitive research. What this research shows is that there is no one-to-
one relationship between concepts, categories and individual words.
Instead the picture is much more complex, a complexity exemplified
by metaphorical language. Cognitive linguists have identified lexical
metaphors as a window into how the way we make sense of concepts
is reflected in language. The extension of the meaning of a word
beyond the literal can give insight into the way the mind makes
connections between ideas. In Chapter 1 we mentioned the tendency
to associate that which is good with the spatial notion of ‘up’ or
‘above’. So in English, for example, we have expressions like feeling
high, upbeat or on top of the world. Similarly, we have negative
expressions like under the weather and downtrodden. I am sure you
84 Language Teaching

can think of equivalent examples in the other languages that you

know. Metaphors like these are seen as a natural by-product of the
natural human tendency to associate the positive or negative with
spatial positioning. Examples like these are used to argue that the
primary units of language are conceptual, having meaning at the core,
and not structural, devoid of meaning. This discussion of lexical
metaphor should bring to mind our earlier discussion of grammatical
metaphor. Once again we see overlap between cognitive and func-
tionalist approaches to language. Because both take meaning as their
starting point for understanding language, this is not surprising.
In addition to the complexities of human cognitive processes, there
is a complexity that comes from the fact that we are all unique, with a
unique set of experiences shaping our stores of knowledge. Each of us
has an individual set of representations or schemas associated with
concepts that are shaped by our individual personal experiences. A
schema is a template or a framework for a concept or idea. Though we
are likely to share a prototype for bird, for example, each of us will
have a more individualised schema with lesser or greater detail for
particular aspects of birds and birdlife, based on our experience with
and knowledge of birds. Additionally, schemas will be connected with
a range of categories which are associated by meaning networks.
One way to think about learning, then, is as the development of
schemas, a process which relies on the human tendency to categorise.
As we try to make sense of new information to which we are exposed,
we activate all relevant schema, looking for a match. If there is a
schema which matches the information, then that schema is con-
firmed, a process which reinforces or strengthens the existence of the
schema. Similarly, if the information does not fit a schema exactly but
is a close match, the existing schema may be modified or added to,
based on the new information. If there is no match among the store of
existing schemas, then a new schema may be created.
This simplified description of schema gives a sense of how asso-
ciative learning works. Crucial to this view of learning is the im-
portance of frequency. A new signal that is only received once is not
likely to create a fully formed or prominent schema. Instead, those
existing schemas that are regularly activated will become more and
more robust, or entrenched. Thus, learning depends on quantity and
regularity in the input. The mind is able to construct schemas, or
knowledge, based on probabilistic reasoning. While this is true for all
kinds of knowledge, of interest to us is the development of language
knowledge. In response to a visual or auditory linguistic signal – a
 Language as a Tool for Communication 85

word, a phrase or a sentence – there is activation of a range of possible

linguistic representations. Cues such as context, existing knowledge
and the immediately preceding linguistic signal help to determine the
most probable match. Also important are current activation levels for
each schema. The activation of a linguistic representation raises what
is known as the resting level of the schema in question.
This also leads to what known as priming. Representations that are
activated more frequently have higher resting levels. If a schema has a
higher resting level, then it is primed, or more readily activated, and it
requires less robust evidence for subsequent activation. If at a lower
resting level, a stronger, more robust signal is needed to activate it.
Moreover, because schemas are organised in sets of related represen-
tations, creating networks based on association, when any single
representation is primed, the network of representations associated
with it is also primed.
In this way, language development can be seen as the building up of
linguistic patterns based on associative learning, and not the develop-
ment of abstract linguistic rules.21 In this process of raising and low-
ering activation levels, there is a crucial role for time. If a representation
is not activated, in time the resting level will gradually diminish, while
repeated activation will strengthen the schema over time. If, however,
activation of a particular representation implicates a slightly altered
form–meaning mapping, the stored representation will also alter ac-
cordingly. This is learning. Notice that this view of learning means that
the system is constantly fluctuating; it is dynamic. Proponents of
Complexity Theory, discussed earlier, emphasise how this change
means that language is constantly evolving. Some take this view one
step further to claim that this constant change implies that language
itself is a system that is emergent.22 While this may go further than some
cognitive linguists are comfortable with, it illustrates how similar views
of language can be linked with different research traditions.
This discussion of learning has presented the development of
schemas as a product of receptive processes, yet schemas also become
strengthened with use. Those who emphasise this aspect of the
Cognitive Approach refer to it as a usage-based approach. Language
knowledge strengthens and builds in the learner’s memory banks as a
result of repeated usage. And learning is exemplar-driven, meaning
that stores of language knowledge grow based on examples. As
implied in our discussion of prototypes, these examples are not stored
randomly, but instead made sense of in an orderly fashion, because it
is a basic human mental function to categorise, allowing us to make
86 Language Teaching

better sense of the world. There is a rich body of research in

psychology supporting this human tendency to categorise. Human
brains seemed programmed to extract or impose patterns. In this way
learners, including language learners, are rational. They are statistical
problem-solvers, working out patterns and constructions from the
input. As part of the learning process, they apply a probabilistic
methodology, determining what is likely based on activation levels
of their repertoire of stored representations.
This raises the question of intentionality. Is this statistical problem-
solving an explicit, conscious process, or does it happen implicitly?
Within the cognitive tradition there is a clear recognition of a dis-
tinction between explicit and implicit learning. Yet the distinction is a
problematic area for cognitive researchers, as it is difficult to differ-
entiate empirically between the two. Nevertheless, there is an under-
standing that learners are capable of and variously prone to use both
types of learning. Moreover, both types are subject to proceduralisa-
tion of activation to the point of automatisation. In other words, a
schema, or set of schemas, can be repeatedly activated to the point
where the activation does not require explicit effort or thought. The
example that is often given to illustrate proceduralisation is learning to
drive a car or ride a bike. Initially, one needs to think about every
move, but eventually, the actions become automatic even to the point
where one might have difficulty explaining what one does.
Language production, then, involves the stringing together of
linguistic structures, choosing from the store of constructions that
have developed as a result of repeated activation. From a processing
point of view, stringing together constructions is in some ways more
cognitively efficient than generating new and original utterances. The
important work in psychology has led second language acquisitionists
from a range of traditions to realise the importance of processing in
language development.23 Processing approaches recognise the need
for a transition theory to explain how language development proceeds
from one stage to another. Such a theory is needed to complement a
property theory like that of the generative tradition, which concerns
itself with explaining the properties of language. In the next section we
will explore a framework that tries to combine both aspects in order to
provide a more comprehensive understanding.
To summarise, in the Cognitive Approach, language learning is a
matter of establishing form–meaning mappings and working out the
probability of their use and distribution. Language ‘rules’ can be seen as
regular, recurring patterns. As for the more radical functionalists, this
 Language as a Tool for Communication 87

allows for a view in which rules themselves are not basic to language, but
a product of a built-up store of linguistic representations. The human
tendency to categorise systematically, in other words, can explain why
language appears to abide by rules. From a cognitive point of view,
language can be seen as no more than a set of words and patterns, stored
by association in complex networks, and reinforced through activation.
Yet a view in which there is no grammar is controversial and, for most
language teachers, not likely to sit very well. It is possible, instead, to take
aspects of each of the views of language represented by the range of
linguistic approaches to find compatibility in such a way as to make sense
of the full complexity of language. As a step towards finding compat-
ibility, consider the diagram in Figure 4.1, which tries to situate the
different linguistic approaches in relation to each other and in terms of
the basic internal/mentalist vs external/interactional opposition intro-
duced at the start of this book.

Figure 4.1 Linguistic approaches to language

A consensus-based approach is the one taken in the next section and

the rest of this book as we try to make sense of (all) principles of
linguistics in the context of language learning and language teaching.

ONE COMPREHENSIVE FRAMEWORK: MOGUL

The work of cognitive linguists has influenced the development of the

framework that we will explore in this section. The Modular On-line
88 Language Teaching

Growth and Use of Language (MOGUL) framework developed by

John Truscott and Mike Sharwood Smith draws from a range of
research from differing linguistic approaches to try to make sense of
the complex process of language development.24 It is because of its
broad base that we are going to explore it here. As originally proposed,
MOGUL is primarily interested in mental, or psychological, processes,
taking a generative view of language and a cognitive view of language
development. While more ‘external’ aspects such as society and context
are also important we will not explore them in any depth, as the primary
approach in this book has been an internal/mentalist one. In this section
we will look at how MOGUL approaches the nature of language and
language development in order to have one unified theory in which to
situate our discussion of language teaching in the rest of this book.
MOGUL develops the idea of the modularity of mind as discussed in
the last chapter. It specifies that the language module is limited to ‘core’
aspects of language such as phonology, syntax and semantics. Since one
distinguishing feature of modular knowledge is that it is not available to
conscious processing, the processing of core syntax and phonology
happens implicitly without our control or even our understanding. Of
course, the language module interacts with real-world, or general,
knowledge. Unlike modular knowledge, general knowledge is available
to conscious processing. As made explicit in MOGUL, language requires
both types of knowledge, modular linguistic knowledge and (non-
linguistic) general knowledge. After all, we have to have an under-
standing of the world in order to make reference to it linguistically.
Additionally, however, it is possible to have non-modular general
knowledge about language, or metalinguistic knowledge. In MOGUL
terms, we will refer to this as non-modular linguistic general knowledge.

Figure 4.2 MOGUL: Types of knowledge necessary for language

 Language as a Tool for Communication 89

These three types of knowledge are shown schematically in Figure

4.2. To illustrate these types of knowledge, consider what is required
in order to make sense of an utterance like (9).

(9) He knows if you’ve been naughty or nice.

If you heard this sentence, in order to process it, you would need to
know principles of (1) English syntax, such as word order and subject–
verb agreement, (2) English phonology, so that the sounds could be
segmented into words, and (3) English semantics, in order to under-
stand, for example, that the pronominal subject refers to some entity,
not named in the sentence yet known to the speaker. These are all
examples of the type of knowledge that is stored in the language
module, or modular linguistic knowledge.
Beyond this linguistic knowledge, however, is additional informa-
tion that speakers from a particular tradition will derive because of
cultural knowledge. The use of the phrase naughty or nice, combined
with the idea of one’s behaviour being known, will, for most English-
speakers in Western cultures, conjure up some idea of Santa Claus and
the delivery of gifts at Christmas. Yet there is nothing inherent to the
individual items in the above sentence to give rise to this larger
meaning. The knowledge required to come to this larger meaning
is stored in the non-modular, or general knowledge, component of the
mind. It is mapping between form and meaning like this that would
qualify (9) as a construction as well.
While we have explained the more straightforward differences
between modular linguistic knowledge and other general knowledge,
we have not discussed the third type of knowledge: non-modular
linguistic general knowledge. One crucial aspect of the MOGUL
framework is that the ‘non-modular’ component of the mind is also
capable of storing knowledge about language. We need to clarify more
specifically the definition of linguistic knowledge. To do so, let us take a
closer look at the two words, naughty and nice. The fact that naughty
and nice behave grammatically as predicate adjectives is modular
linguistic knowledge, knowledge that will lead every native English
speaker to agree that the following sentences are ungrammatical,
whether or not they can explain that they are ungrammatical because
naughty and nice cannot be used as verbs or prepositions, respectively.

(10) a. *I will naughty/nice my gifts for Christmas.

b. *Put the gift naughty/nice the tree.
90 Language Teaching

The meanings associated with naughty and nice, on the other hand,
are non-modular general knowledge. The third type of knowledge –
knowledge about language – is linguistic in the sense that it is about
language but, because it is explicit knowledge, it is considered non-
modular general knowledge. In other words, MOGUL claims that
‘linguistic’ knowledge exists in both the language module and in extra-
modular stores, but despite being ‘linguistic’, the quality or type of
knowledge is different.
Because we are sophisticated users and perhaps teachers of English,
we have metalinguistic knowledge that tells us that in (9) naughty and
nice are predicate adjectives. We can know this because we can
analyse the properties of these words, working out their distribution
patterns to see that they occur in the predicate and modify the
pronominal subject. It is this explicit general knowledge that enabled
me to create nonsense sentences like those in (10) for illustration. Yet
when I use the words naughty and nice spontaneously, I do not attend
to the grammatical or distributional properties of the words. Instead,
just like you, I choose these words in terms of their meaning in order to
discuss whether Santa Claus will bring presents this year. My modular
linguistic knowledge ensures that I abide by correct syntactic struc-
ture. My training as a linguist allows me to explain the properties of
the sentence, or to talk about language; in other words it is linguistic,
but explicit, so part of my general knowledge stores.
While there are different types of knowledge, there is only one
process in MOGUL by which knowledge develops. Language devel-
ops through the on-line use of language. Also known as the Acquisi-
tion by Processing Theory (APT), learning occurs through the
reinforcement of input, as described in the section above on the
Cognitive Approach. When, for example, a new linguistic item occurs
in the input, it registers in the learner’s mind. If that item is never
encountered again, it is not likely to become a permanent part of the
knowledge store (linguistic or non-linguistic). With repeated reinfor-
cement, however, that item will in time become a part of stored
knowledge. Based on psycholinguistic research, this process of lan-
guage ‘growth’ is the same as all kinds of knowledge development,
regardless of what type of knowledge it is. Thus, through activation
and reinforcement, formal linguistic aspects of input will become part
of modular linguistic knowledge while knowledge of meaning and use
builds in general knowledge.
You may be wondering what advantage there is for maintaining
two types of knowledge, modular and general. From a theoretical
 Language as a Tool for Communication 91

point of view, the distinction is needed because of research showing

differences between different types of knowledge. In language acquisi-
tion, it can help to explain why SLA seems to be fundamentally
different from FLA. For native language acquisition, the language
module seems to develop quickly and relatively easily. For adults,
language learning seems to depend much more on deliberate, con-
scious learning. This does not mean that the language module is
irrelevant for adult language learners, but it does seem that certain
aspects of a language are not facilitated in the same way for young
native learners as they are for adult second language learners.
The distinction is also useful in terms of language teaching. It is
completely feasible for a person to have both modular linguistic
knowledge and a sophisticated non-modular understanding of a
linguistic principle – as I illustrated above, showing that I have both
kinds of knowledge for naughty and nice. Because language proces-
sing can be based on both types of knowledge, this means that there is
the possibility that general linguistic knowledge can be used to
compensate for modular linguistic knowledge. If a learner is not able
to develop modular linguistic knowledge as children do at an early
age, they can potentially develop non-modular general linguistic
knowledge through hard work and frequent exposure. In other words,
there is on-line growth and use of language, whether modular or
non-modular. And most encouragingly, there is no reason why non-
modular knowledge should not lead to native-like language produc-
tion – though it may require lots of practice and use.
This last point means that teachers should know which aspects of
language can develop as modular knowledge for their learners and
which cannot. Then, they should provide the correct type and an
ample amount of input for language to develop. It is when thinking
about how to provide input that an external view of language becomes
useful. Functional and Sociocultural Approaches remind us that
learners are more than their mental processes. Any teacher will agree
that learners thrive in a supportive atmosphere. Moreover, encoura-
ging learners to achieve their potential, or to work within their Zone
of Proximal Development, instead of holding out fluent native speak-
ers as the target is much more pedagogically sound. In addition to a
supportive environment, there are culturally based aspects of language
that learners need to be taught in order to become competent parti-
cipants in relevant communities of practice.
A functionalist view is also indispensable for teachers trying to
explain language in terms of its use. The functional view also offers an
92 Language Teaching

accessible and meaningful way to understand language. There are

times in which teaching language by drawing attention to function is
more appropriate for explicit instruction than dwelling on structural
properties. Moreover, embedding language in context will make it
more meaningful. If learners are more engaged because the teaching
emphasises the more meaningful aspects of language, this will result in
much higher levels of engagement. More language engagement,
whether explicit or implicit, means more opportunities for activation
and raised resting levels for new knowledge. In other words, there will
be language development – both modular and non-modular.
In short, MOGUL is useful for language teaching because it has room
for the current range of views in linguistics. Additionally, this frame-
work allows for a complex understanding of what language is, both
structurally and functionally; and it acknowledges important findings
in psychology that point to how knowledge develops and how it is
processed. In other words, it provides a single framework for under-
standing the complex processes involved in language and language
development. We therefore turn to the language classroom, asking
what implications we can draw from these theoretical discussions.

FOR DISCUSSION

1. Now that we have considered each of the four basic views of

language, discuss the extent to which your own understanding of
language prior to reading this book was informed by each of the
views.
2. Compare and contrast the four basic views of language, identifying
areas in which they are in opposition and areas in which they are
compatible.
3. Discuss the extent to which it is important as a language teacher to
have a clear view of what language is in the linguistic sense.
4. To what extent does relevance of each view of language depend on
the language teaching context? Specifically, are there differences in
terms of age or proficiency level? Second vs foreign language
teaching? Personal vs academic or professional use?
5. Go through the section on MOGUL and identify explicitly where
the different views about language emerge. Are there any major
concepts that are not covered in the MOGUL framework?
6. What is your view of language now? Consider your understanding
of language in terms of language teaching. What implication could
your views have for the way you approach language teaching?
 Language as a Tool for Communication 93

NOTES

1. The influential original discussion can be found in Hymes (1972).

2. Halliday has been very prolific, producing many works in Systemic
Functionalism. The first version of his seminal work, An Introduction to
Functional Grammar, was published in 1985. It has been re-issued, with
the third edition, revised by Christian Matthiessen, appearing in 2004. A
ten-volume series comprising his collected work has also been published
recently, edited by Webster (2007).
3. See Butler (2003) for an extensive discussion of the branches of the
various structural-functionalist branches within Functionalism.
4. As Halliday’s SFG has mainly focused on the textual level, a branch of
SFG has developed advancing the theory at the level of the sentence and
below. This branch is now known as Cardiff Grammar because Fawcett,
its foremost proponent, has done most of his work with colleagues in
Cardiff since the 1980s. See Fawcett (2000). The Hallidayan approach,
as a result, is now sometimes referred to as the Sydney approach.
5. This example comes from The Guardian Weekly, 23.07.10, in an article
entitled ‘Britain’s human history revised’, pp. 30–1.
6. The basic components of Speech Act Theory were proposed by Austin
(1962), but then developed by Searle (1969).
7. In addition to Goldberg’s approach to Construction Grammar, there is
the Radical Construction Grammar approach of Langacker (1987),
Lakoff (1987), and Croft (2001).
8. Many cognitive linguists also take a Construction Grammar view of
language. In fact, there is a degree of debate as to whether Cognitive
Linguistics is a form of Functional Linguistics. See Butler (2003), who
includes Cognitive Grammar as a structural-functional approach, while
acknowledging that many cognitive linguists prefer to see themselves as
distinct from Functionalism.
9. Example found on 28.07.10 using the Compleat Lexical Tutor (http://
www.lextutor.ca/), to search the British National Corpus (BNC) Spoken
Corpus.
10. For posthumously published work, see Vygotsky (1979, 1986). For more
current developments in sociocultural theory and language learning,
see the edited volume by Lantolf (2000), as well as Lantolf and Thorne
(2006).
11. The term scaffolding was introduced by education psychologist Jerome
Bruner. See Wood, Bruner and Ross (1976).
12. For more comprehensive discussion of a Complexity Theory as applied
to language and language learning, see Larsen-Freeman and Cameron
(2008) and de Bot, Lowie and Vespoor (2007).
13. For early articulation of Situated Learning Theory, see Lave and Wenger
(1991), and Wenger (1998) for more on the communities of practice idea.
94 Language Teaching

14. For a collection of papers, see the edited volume by Norton and Toohey
(2004), including an introductory chapter with an overview of the
development of a Critical Theory approach to language pedagogy.
15. For an enlightening discussion of the sociocultural view as compared to a
cognitive view, see Zuengler and Miller (2006).
16. See, for example, Atkinson (2002).
17. For an overview of Cognitive Linguistics, see the introductory texts by
Croft and Cruise (2004) or Evans and Green (2006).
18. For early original work, see Rosch (1973, 1975). For an overview of
Prototype Theory as applied to language, see Taylor (1995). The idea has
seen much development in Cognitive Linguistics, most notably by Lakoff
(1987) and Langacker (1987).
19. See, however, Aarts (2007), who argues for what he calls gradience in
syntax.
20. See, for example, Labov’s famous 1973 study on cup-like objects.
21. For discussion of the Cognitive Approach in FLA, see Tomasello (2003).
Nick Ellis is one of the best-known researchers in second language
development. See Robinson and Ellis (2008) for a useful overview.
22. There are a number of emergentist approaches to language development.
For example Ellis (2003), O’Grady (1999, 2003) and MacWhinney
(1999).
23. In addition to MOGUL, there are other processing approaches to second
language development. O’Grady (2003) attributes difficulties in second
language development to so-called computational complexity such that
learners are limited by the current level of their interlanguage. Carroll
(2001) looks to instances in which the parser has problems with the input
as the time when acquisition occurs. VanPatten’s input processing model
takes more of a cognitivist view, relying on the learner’s need to match
language form with meaning, drawing from the forms that the learner
currently holds. See VanPatten (2009) and references therein.
24. To read about MOGUL as originally proposed, see Sharwood Smith
(2004) and Truscott and Sharwood Smith (2004 a, b).
 5
IMPLICATIONS FOR LANGUAGE TEACHING

IMPLICATIONS FOR THE CLASSROOM

So far in this book, we have explored the fields of theoretical linguistics

and language acquisition. And we have seen some very different views
of language with linguists who view language as an internal mental
property of individuals and others who view language as intricately tied
up with the norms and habits of particular speech communities external
to the individual. But we have also said that these views are not
necessarily incompatible; a psycholinguist can analyse the internal
processing and production of the same language that the functionalist
considers from the point of view of meaning and the sociolinguist
analyses in terms of adherence to cultural norms of behaviour.
Within psycholinguistic research, however, there is fundamental
opposition, as generativists see language as modular, while cogniti-
vists equate it with other kinds of knowledge. Though this view is
irreconcilable, the fortunate point for us is that it does not necessarily
matter for language teaching. The common aim in the language
classroom is for learners to be able to produce and comprehend
language effortlessly and fluently. Moreover, there seem to be areas
in which the two views can find consensus, as illustrated by MOGUL,
which draws on aspects of both views. Whether language is ‘modular’
or not, psycholinguists of any perspective agree that there is a
difference between automatic linguistic abilities and the more la-
boured efforts associated with metalinguistic skills. As teachers, we
want to move our learners continuously toward the automatic,
linguistic end of this performance spectrum, regardless of what the
theoretical explanation for the difference is.
Other ‘oppositions’ can be reconciled if they are seen as opposite
ends of a spectrum which, taken together, present a more complete
picture of the complexity of language. The ‘formalists’, for example,
are attempting to define the abstract rules underlying language, while
the ‘functionalists’ recognise language as an embodiment of meaning
96 Language Teaching

whose core properties are grounded in experience. These are not

necessarily contradictory – we can abide by linguistic constraints
when we express meaning, drawing on our life experiences. Similarly,
the ‘mentalists’ are busy researching the inner workings of the mind,
while the ‘interactionists’ have focused on language in society. A
conscientious teacher will recognise all of these facets of language and
work to develop them depending on the particular needs of the
students. In this chapter, we will extend the theoretical discussions
that we have had so far to the classroom and begin to explore their
practical implications. Then, in the remainder of the book, we will
turn to more specific approaches and trends in language teaching in
attempt to see how a knowledge of underlying theory can inform even
the more practical, everyday decisions made in the classroom.
Our application of theory will be structured around the ten observa-
tions about second language development presented by VanPatten and
Williams (2007), which we outlined in the conclusion of Chapter 3.

OBSERVATION 1: EXPOSURE TO INPUT IS NECESSARY FOR SLA

While the observation that exposure to input is necessary for language

development may seem obvious, the implications for language teach-
ing are not so straightforward. There are two main issues to be
considered. Firstly, there is the question of quality: namely, what
type of input. Secondly, there is the question of quantity.
One way in which we can divide input is to consider implicit and
explicit aspects. Implicit input can be associated with the natural
language of spoken language and ‘authentic’ written materials. As we
saw in Chapter 2, one of the main complaints about Grammar Transla-
tion was the lack of exposure to natural, spoken language, a complaint
which led to the development of the Direct Method. If we thought that
second language learning was the same as native first language devel-
opment, we might choose to restrict all language spoken in the classroom
to the target language, as done in the Direct Method. But attempts have
shown that this is not a reasonable choice. One reason for this is input.
The nature of classroom learning is fundamentally different; both the
amount of input and the reasons for using language are different in a
classroom as compared with a family home environment. Given class-
room restrictions, teaching also needs to include explicit input.
Explicit input refers to explanations about language, or metalin-
guistic input. This is sometimes also referred to as negative evidence
because it includes explanations about what is not correct in the target
 Implications for Language Teaching 97

language, in addition to explanations of the rules of language. While

language ‘rules’ usually refer to grammar rules, such as how to form a
passive sentence, there are also rules about language use that can and
should be taught. Every language, for example, has mechanisms for
expressing politeness. The use of modal verbs such as might and may
or adverbial expressions like if possible is a grammatical way to
express politeness in English. And beyond grammaticality, a learner
needs to know which social contexts expect polite forms as well as the
appropriate level of politeness. Explicit rules, then, need to cover
language in terms of both form and function.
So should a teacher provide implicit or explicit input? The answer, of
course, is both. As illustrated in MOGUL, a learner can develop both
implicit linguistic knowledge and explicit metalinguistic knowledge.
And both can develop to high levels of automaticity, which will show
itself as language fluency. The difficult question is how to know when to
provide implicit or explicit input. The answer lies in understanding how
different aspects of language develop. We will return to this question
when we explore the variability of language development across
linguistic subsystems (Observation 6). For now we will say that there
are times when learners should be immersed in as much authentic input
as possible and times when explicit explanation is necessary.
Whether it is implicit or explicit, learners need to get as much
exposure to the target language as possible. The amount of language
exposure a learner gets can be affected by so-called affective factors,
such as motivation, personality, attitude and other emotional factors.
Much has been made of the claim that second language speakers do not
have the same motivation for learning as native first language speakers.
Wanting to succeed on a course in class cannot equate with needing to
express yourself to satisfy basic everyday needs. This difference means
that teachers need to inspire their students so that they have a healthy
and inquisitive attitude towards the target language.
Another aim of language teaching is to facilitate exposure outside the
classroom, increasing the amount of input each learner gets, regardless
of the differences in ability and motivation among students. While an
outgoing, risk-taking learner may feel comfortable participating in on-
line chat rooms, for example, even a learner with a shy and reticent
personality can find ways to increase the amount of input they are
exposed to if properly motivated. Since, as we will see in Observation 2,
learners can develop language incidentally when their primary focus is
on something other than the explicit learning of language, even the
shyest learner will benefit from more ‘passive’ activities like reading in
98 Language Teaching

the target language. In fact, one line of SLA research argues that reading
for pleasure can lead to great improvement in language proficiency.1
In short, the observation that plentiful exposure to input is neces-
sary for SLA means that teachers should be sure that learners get both
implicit and explicit input and that the amount of input is maximal,
fostering learners’ motivation and interest so as to maximise the
amount of exposure.

OBSERVATION 2: A GOOD DEAL OF SLA HAPPENS INCIDENTALLY

This observation builds on the first, addressing the effect of input. A

growing body of research suggests that learners come to know
particular features of language without explicitly engaging with those
features. So, for example, while trying to understand a text in order to
answer comprehension questions, a learner can also come to know
new vocabulary words or syntactic constraints. Drawing from the
range of research, this finding seems to include other aspects of
language like word order, derivational morphology and word for-
mation, as well as more extra-grammatical features such as politeness
strategies and appropriate register as well. As pointed out by Nick
Ellis (2007: 88), ‘We ‘‘know’’ far too many linguistic regularities for us
to have explicitly learned them.’ This observation brings to mind
Krashen’s distinction whereby implicit development, or acquisition,
seems to occur alongside explicit development, or learning. While
Krashen proposed these two types of knowledge as distinct and
dissociated from one another,2 what matters for the language teacher
is the comprehension and production abilities of the learner, not
theories about the knowledge underlying these abilities.
One implication of incidental learning for the classroom is what has
come to be known as Focus on Form. Michael Long has made a useful
three-way distinction between Focus on Meaning, Focus on FormS
and Focus on Form. A Focus on Meaning lesson will limit itself
exclusively to the function of language, or strictly communicative
teaching, without explicit metalinguistic teaching. Focus on FormS
captures the other extreme whereby language forms (alone) are
explicitly taught in isolation from their use, and drilled without
any meaningful context. This often takes the form of grammar
exercises. The capital ‘S’ on Form is used to make a clearer distinction
with the more appropriate middle ground, Focus on Form.3
One rationale for Focus on Form is the observation that much
learning happens incidentally. With this in mind, teachers are en-
 Implications for Language Teaching 99

couraged to devise a meaningful task that will require the learners to

engage genuinely with language. The task should include language
which is challenging to the learners’ current levels of proficiency, with
forms which the learner has not yet mastered. When these forms
interfere with the activity of the lesson, the teacher should call explicit
attention to them, explaining them clearly. In other words, the teacher
should focus on forms, but only in the context of meaningful engage-
ment with language. Thus, Focus on Form is a way of teaching form,
but in the context of use or function.
The degree to which the task involves challenging forms, and the
extent to which the teacher orchestrates attention to them, is a matter
of professional judgement, judgement which should not be random,
but instead based on a range of factors, both in terms of theoretical
understanding of language/language development and the practical
constraints of the classroom environment. Knowing that development
occurs through repeated exposure, however, means that a good
teacher will be attuned to frequency, supplying sufficient amounts
of examples of the forms in question. Additionally, an expert teacher
will pay attention to the salience of forms in the input. In other words,
new forms need to be prominent enough to be picked up on. If new
forms do not stand out and learners can understand the input without
having to make sense of them, then Focus on Form is less likely to
occur.
The discussion of ‘forms’ generally brings to mind inflectional
morphology or specific grammatical structures. Yet there is no reason
why Focus on Form should not include a focus on extra-grammatical
forms as well. At the level of text, for example, students could be asked
to read a short academic essay and a newspaper article on, say, the
effect of litter on the environment. This could be done in preparation
for a letter which the student is asked to write to their local council
about litter in their street. Within this task, the teacher could focus not
only on new vocabulary and any complex grammatical points, but
also on register, pointing out the features of journalistic writing that
differ from those of academic writing and formal letter writing. Focus
on Form could likewise be extended to focus on cultural norms for
registering complaints in the local setting. In fact, there is no limit to
aspects of language and interaction that can, in principle, be explicitly
taught by language teachers. In sum, teachers would do well to keep
learners engaged in tasks that interest them, knowing that they can
and will be learning other features of language in the mean time.
100 Language Teaching

OBSERVATION 3: LEARNERS COME TO KNOW MORE THAN WHAT THEY

HAVE BEEN EXPOSED TO IN THE INPUT

It may seem that this observation is quite similar to Observation 2, but

there is a difference. Incidental learning means that learners will come
to know backgrounded properties of language while they are focusing
more explicitly on other properties or activities. Observation 3 refers
to an even more implicit phenomenon whereby learners come to know
properties of language that they are not even exposed to directly. In
other words, there are ‘poverty of the stimulus’ effects in language
development.
In Chapter 3, we explored how poverty of the stimulus is the basis
for the nativist conceptualisation of language acquisition and comes
from the overwhelming evidence that children come to know more
about the ambient language than the subset of language to which they
are actually exposed. We then saw that there is evidence for poverty of
the stimulus effects among second language learners as well, looking
at research on the Overt Pronoun Constraint as an example. This has
led to the conclusion that UG-constrained development is relevant in a
second language context. Of course, adult second language learners
do not seem to benefit from UG-constrained development as thor-
oughly as first language learners, but the fact that such a phenomenon
is possible at all is important and should be a factor for decision-
making in the language classroom.
The most general implication of Observation 3 is further support that
teachers should not be afraid to include authentic, natural input – in all
of its complexity – in their classroom. However, in considering class-
room application, the obvious question is how much authentic input
should there be? Should teachers only ever use the target language?
Should all classroom material be taken only from native sources? In the
excitement of the ‘Chomskyan revolution’, early applied linguists
encouraged teachers to take this kind of ‘native’ approach to their
second language teaching. Krashen and Terrell (1983) promoted the
so-called Natural Approach, in which teachers were to foster language
development as much like the native child experience as possible. They
were even to allow for a silent period, in which their learners were
exposed to large quantities of authentic input, but not expected or
asked to produce their own language. Advances in the new scientific
approach to language revived the idealistic notions which had already
failed Direct Method proponents years earlier. Though a completely
‘natural’ approach may not be appropriate, these ideas have led to the
 Implications for Language Teaching 101

development of content-based or immersion approaches to language

teaching which can be found in a range of settings from primary schools
to high-level preparatory academic English programmes. In these
courses students are taught academic content in the target language
itself as a way of teaching the language.
The fact that second language development is different to first is, of
course, obvious to teachers and learners alike. Exploring the reasons
may help to determine how to navigate the differences. Age and the
learner’s existing knowledge of a first language are two obvious
reasons for the difference, but another point that is starting to emerge
clearly from the research is that different domains of language seem to
develop differently. As we mentioned in Chapter 3, there is evidence
from new technologies that different aspects of language are processed
differently in the brain. Thus it is not surprising that there might be
developmental differences as well. While Generative Linguistics has
always differentiated between modular and non-modular develop-
ment, the growing body of SLA research is telling us that a more
refined view of language in second language contexts is needed as well.
The research showing equivalence between native and non-native
processing on semantic elements of language suggests that learners
might respond well to natural, authentic input within the realm of
meaning. However, differences in processing syntactic aspects of
language suggest that a more explicit approach may be appropriate
for the more core grammatical features of language like syntax and
phonology. Interestingly, the research showing poverty of the stimulus
effects in second language development so far are mostly to do with
semantic knowledge. The Overt Pronoun Constraint, for example, is
to do with knowing the possible meanings of a pronoun in different
grammatical contexts. Other poverty of the stimulus studies also
investigate subtle points of interpretation.4 Again, this suggests that
meaning-based aspects of language, including some very subtle aspects
of language of which even native speakers are not explicitly aware, can
develop naturally without explicit instruction.
This distinction between syntax and semantics also suggests that we
should promote active engagement with authentic language, or mean-
ing-based instruction, while also allowing for explicit engagement
with language form. In other words, it provides more support for a
Focus on Form approach to language teaching. But we can go even
further. The MOGUL framework can help us to determine how we
should teach the forms on which we decide to focus. If acquisition
happens by processing, then students should be given activities that
102 Language Teaching

require them to process particular forms actively, remembering the

crucial role of frequency, since learning occurs as a result of repeated
activation. This tells us that there is a clear place for practice and
repetition in language teaching. While this might sound like a return to
the more rote practices of drilling, we must resist a strict one-or-the-
other mentality. Learners should practise, but in the context of
meaningful engagement. If repetitive practice is meaningless, it is
no longer requiring active processing, and learning is compromised.
In sum, there is a degree of modular development that seems to be
possible for adult second language learners. Because learners seem to
come to know more than what they were exposed to, teachers should
readily expose learners to natural language, in all its complexity.
However, this should be tempered by explicit focus on some forms to
allow for repeated engagement. The question of which forms will be
addressed further when we discuss Observation 6.

OBSERVATION 4: LEARNERS’ OUTPUT (SPEECH) OFTEN FOLLOWS

PREDICTABLE PATHS WITH PREDICTABLE STAGES IN THE ACQUISITION
OF A GIVEN STRUCTURE

This observation refers to the systematic nature of interlanguage

development, based on research on specific structures, such as ques-
tion formation and negative sentences. As we have seen previously, the
morpheme order studies show stage-like development in both native
and second language development. From this research there is a
tendency to generalise to say that language develops in systematic
stages. But as we have said repeatedly, language is a complex and
multi-faceted phenomenon. While it is uncontroversial that there are
developmental stages in a very broad sense, there is no documented
research showing that all of language is entirely systematic. In other
words, though this observation seems to hold at very general and at
some very specific levels, there is a rather large middle ground which
research has yet to address in this regard.
To illustrate more specifically, at the very general level it is well
documented that learners acquire contentful words like table and run
before they learn function words like the and of. Similarly, they can
produce words and phrases before they are able to produce gramma-
tically well-formed clauses, which, in turn, precede complex clauses.
And at the other end of the spectrum, there is research on some very
specific structures that shows developmental stages. When acquiring
negation in English, for example, learners will first add the negative
 Implications for Language Teaching 103

word no to the front of a predicate to say things like no like coffee

before they systematically add a subject, I no like coffee, which
precedes the final stage in which they correctly include an auxiliary
verb: I don’t like coffee.5 So for some specific structures there are
predictable stages, but there is no research (yet) to show clear devel-
opmental stages for all aspects of language. Moreover, the interest in
stages has been limited mostly to formalist research, with very few
researchers even asking whether this idea of definable ‘stages’ applies
beyond grammar to areas of language function and use.
So what is the use of this observation in the language classroom? If
the observation could be empirically shown for all aspects of language
at all levels, then the implication would be to strictly follow the
developmental stages. Since it is not, teaching in strict stages is only
reasonable in a broad sense of moving from the more simple to the
more complex and for the limited range of specific language structures
which have been researched, like negation.6 Yet, systematicity in
language development, generally speaking, has implications for error
correction. If you know that a particular language point is not likely to
develop until a later stage in the learner’s proficiency, this would
suggest that some errors are better left unremarked, with focus instead
on errors at the current stage. If, for example, your learners say no
like, it would make more sense to encourage them to add subjects to
their negative statements than to insist on statements with subjects and
auxiliary verbs since we know each of these represents a different stage
in the development of negation.
You may wonder whether teaching a more complex, final stage
might help the learner to go through the intermediate stages more
quickly. There is research that investigates whether learners can know
a less ‘marked’ structure if they are taught the more complex corre-
sponding ‘marked’ forms. Results of research in markedness show
that some learners who were taught more complex forms first were
able to learn the simpler forms in the process. Unfortunately, however,
the results of this kind of research are mixed, suggesting that it may be
unrealistic to expect this from all learners in all contexts or for all
aspects of language.7 Returning to our main point about error
correction, you may want to expose your learners to complex forms,
but it may be best to refrain from correcting those errors that reflect a
natural stage in the learner’s development.
It is also worth considering that an error may not be an error from
the point of view of the learner. The learner is developing an ever-
changing interlanguage grammar, one that is a system and not a
104 Language Teaching

random set of words and fragments. If the learner’s current inter-

language system includes I no like coffee for negative statements, then
for this learner, this is the way to make negative statements. It is only
an error from the teacher’s point of view. Of course, you may choose
to appeal to a student’s metalinguistic ability to explain that this is an
error as a strategy for compensating for their current interlanguage
stage. But if you know that this is a natural step in the learner’s
development, you may instead want to use the limited time you have
with your learners to focus on something else and let the learners’
systems develop in their own time as they get more input.
Unfortunately, we also know that sometimes errors persist, even
though the learner is considered proficient enough to develop a
particular aspect of language. Fossilisation is an unfortunate outcome
for many learners on varying aspects of language. From a MOGUL
perspective, however, if an error persists beyond the level at which we
would expect it, the option is to try to compensate for this lack of
modular linguistic development by explicit teaching to allow for
metalinguistic development of the language point in question. A
teacher can bring the point to the learner’s attention and work
together with the learner to develop strategies to compensate for
fossilisation, drawing on the learner’s extra-linguistic general intelli-
gence. With practice, a dedicated learner may be able to automatise
this knowledge so that his/her production appears native-like. In other
words, for any point of language, a motivated learner should be able
to overcome shortcomings in their implicit linguistic knowledge by
making use of their general intelligence. This kind of metalinguistic
monitoring is particularly useful in written production, as both time
and visual representation allow a careful writer to edit their language.
With enough dedication and hard work, such self-correction may, in
time, be achieved in oral production as well. This will, of course, be
helped by experienced teachers who can clearly explain shortcomings
and remedy them in a supportive and constructive way.
This view of errors as indicators of a learner’s current stage of
development is more in concert with Sociocultural Approaches than
earlier views of errors as indicative of distance from the native speaker
target. Identifying a learner’s current and potential stages allows for a
Zone of Proximal Development perspective which recognises aims
which are possible and realistic for the learner. A learner should be
supported to enable him/her to achieve within their Zone, instead of
being constantly corrected or criticised for their current abilities. In
addition to errors, we should not completely discount mistakes. As
 Implications for Language Teaching 105

mentioned in Chapter 1, as a natural part of language production,

mistakes may not be worth correcting, especially in an activity in
which fluency is the aim. Yet high rates of mistakes may signal that the
learning environment is causing undue stress to the learner to the point
where s/he is not able to perform to her/his highest potential. If this is
the case, it may signal a need for the teacher to make changes to the
classroom environment to provide a more supportive environment
that scaffolds the learners as they develop.

OBSERVATION 5: SECOND LANGUAGE LEARNING IS VARIABLE IN ITS

OUTCOME

Against the background of systematicity and developmental stages is

the unavoidable fact that second language learning is variable in its
outcome, a fact that is even more pronounced the older the learner is
when first exposed to a language, as pointed out in Bley-Vroman’s
Fundamental Difference Hypothesis. It is clear that not all learners
arrive at a well-developed competence, sometimes even after years and
years of effort and exposure. Discussion of this observation first needs
clarification of the distinction between ‘rate’ and ‘route’. Seminal early
research showed that beginning adult second language learners devel-
oped interlanguage proficiency faster than child L2 learners did: in
other words, at a more rapid rate – at least initially.8 But this conclusion
should not be used to make assumptions about route. Observation 4
tells us that developmental route is not so variable. The other important
finding of the research about rate is that, in the long term, the children
were able to acquire a higher level of proficiency than the adults. Second
language learning is indeed variable in its outcome.
One clear implication, then, is that a teacher should take age into
account. Adults have a more developed cognitive system and a richer
amount of experience than children. As experienced learners, adults will
have developed learning strategies that they can employ for metalinguistic
learning. They may also be more motivated if they have chosen to learn a
language, or need it for personal or professional reasons. These affective
factors can all contribute to learners exposing themselves to more input
and/or actively developing learning strategies for metalinguistic devel-
opment. Of course, we would do well to keep in mind our earlier
discussion about critical periods, which tells us that there is no specific
cut-off for language development, and that some aspects of language
seem to be affected by age more than others. Indeed, there is no cut-off for
learning new vocabulary; nor does basic clausal word order seem to be
106 Language Teaching

affected by age. Development of inflectional morphology and phonology,

on the other hand, seem to be very susceptible to differences in age.9
You may, however, be thinking that even among learners of the same
age group there can still be much variability. In addition to the affective
factors we have briefly mentioned, there is another clear factor that will
lead to variable outcomes in second language learning: the learner’s
native language. It is widely accepted that learners come to the task of
learning a second (or third) language from the starting point of the basic
structure of their existing language(s), or, in other words, that there is
Full Transfer of the first language. Schwartz and Sprouse (1996) argue
that L2 development is a matter of restructuring the L1 grammar when
that grammar cannot make sense of the L2 input.
Think about how MOGUL presents the learner as having a complex
collection of knowledge that is activated in response to any and all
linguistic signals. This means that in response to target language input,
the learner will use whatever aspects of target language knowledge
that they have, as well as all of the native and other language
knowledge that they have. The existing L1 knowledge should prove
very helpful when it comes to extra-linguistic features that do not vary
much among languages, such as tone of voice. It may even help out in
some areas of core grammar – for example, adjective + noun or noun +
adjective order – depending on whether there is a match between the
languages. But it is not likely to be of much use in all aspects of
language. The shapes of particular words, for example, are likely to be
of little use, since these differ from language to language.
This view of native language transfer may seem obvious, but it
quickly becomes more complicated. One widely accepted learnability
problem which illustrates the complexity (and importance) of the
effects of L1 transfer is to do with what has come to be called the
Subset Principle. You will remember from our discussion of the Null
Subject Parameter in Chapter 3 that in some languages subjects are
required for grammaticality while in others they can often be left out,
depending on either discourse or inflectional morphology. Thus, a
‘null subject’ language like Spanish or Japanese allows two gramma-
tical options: a sentence with a subject or the same sentence without.
Since English is not a null subject language, it only has the one option
of sentences with subjects. In linguistic terms, this is seen as a
parameter with +null subject and –null subject settings. Crucial for
understanding the Subset Principle is the relationship between the two
settings. We can characterise a –null subject-type language as a subset
of a +null subject-type language because both share the option of
 Implications for Language Teaching 107

expressing subjects (=subset), but only +null subject languages have

the second option of subjectless sentences (=superset) as well.
The Subset Principle is a universal learnability principle that applies
to all parameters that can be configured in this kind of way. It states
that, initially, learners will automatically assume the more restrictive
subset for a parameter, the one that allows only one option. In other
words, learners are initially conservative. If there is positive evidence in
the input that the other option is allowed, the developing grammar will
expand to include the other option. This principle is one of a set of
learnability principles that tries to explain how very young children can
acquire such complex linguistic systems without the aid of instruction
or negative evidence. So what does this mean for the second language
context? We have just said that L2 learners start learning based on their
L1. So, for example, if a learner has an L1 grammar with a subset setting
– in other words –null subjects in English – then when they try to learn a
+null subject language, like Spanish or Korean, they will correctly
assume that the language allows sentences with subjects; and as soon as
they encounter sentences without subjects their interlanguage grammar
will develop to include this added option.
Lydia White (1989) went further to ask what happens when a +null
subject language speaker tries to learn a –null subject language? Based on
their existing L1 grammar, a Spanish speaker, for example, will assume
that English has subject and subjectless sentences, like Spanish. The
question is, how can the learner unlearn the L1-based grammatical rule
for subjects? Logic tells us that there are no examples of something that
does not occur in the input. The answer, then, is that this is an area in
which a teacher needs to give explicit metalinguistic instruction for the
learners because exposure to input alone is not going to be enough.
Through metalinguistic awareness (in other words, explicit instruction),
learners can know what is correct and not correct in a language and with
deliberate effort can compensate for this kind of learnability problem.
In sum, second language variability can be attributed to differences
in rate, age, amounts of exposure and affective factors. In addition,
differences in the existing L1 of learners can have fundamental effects
on the development of the grammatical properties of the L2.

OBSERVATION 6: SECOND LANGUAGE LEARNING IS VARIABLE ACROSS

LINGUISTIC SUBSYSTEMS

From inter-linguistic variation, which addresses differences between

languages, we now turn to intra-linguistic variation, which refers to
108 Language Teaching

differences within a single language system. We have made a broad

distinction between core modular linguistic knowledge and non-
modular knowledge. And we have characterised modular knowledge
as distinct, meaning that it is implicit, automatic and not subject to
awareness or deliberate change. But at the same time, drawing on
MOGUL, we have said that non-modular knowledge can develop to a
level of automaticity. So is there any difference, then, that is relevant to
language teachers, aside from just being aware of the academic debate
that underlies linguistic theorising?
The answer, of course, is yes. The language module is associated
exclusively with implicit acquisition, and is, by definition, beyond the
individual’s control. As we have already suggested, to facilitate
modular development teachers should take a more ‘natural’ approach:
exposing learners to real, natural examples and encouraging them to
interact in the target language in order to foster natural, implicit
modular language development. Non-modular, general knowledge,
by contrast, can develop both through explicit, deliberate learning and
incidentally, or implicitly. So in addition to providing the learner with
a natural language environment, a good teacher will explicitly teach
those areas of language which are non-modular and ask their students
to do healthy amounts of practice.
This leads us to the question of which ‘areas’ are which. Linguists
have long debated which areas of language are ‘core’ and which are
not. Fortunately, developments in psycholinguistic and neurolinguis-
tic research have allowed for more empirically based arguments about
different aspects of language. As this book is about second language
learning and teaching, we will limit our discussion to findings in SLA,
acknowledging that this area is currently cutting-edge research, with
much work still to be done. The general findings in this line of research
are that learners seem to be able to master meaningful aspects of
language such as vocabulary and aspects of semantics, but have more
difficulty with some of the more grammatical aspects. Learners are
able to come to grips with a new language in terms of its semantics,
pragmatics and the inventory of words. In other words, learners can
readily master the more conceptual-based aspects of a language
which, in MOGUL, are part of general cognition.
This ability contrasts with the complex picture emerging from
research on acquisition of grammar, which, on the face of it, seems
contradictory. On the one hand, second language learners seem to be
able to acquire the very core properties of syntax, properties which,
within MOGUL, would certainly be considered ‘modular’. These core
 Implications for Language Teaching 109

properties of syntax include properties like subject-verb-object word

order and other grammatical structures like question formation and
negation. But at the same time, they seem to have difficulty with other
grammatical properties like subject-verb agreement, articles and pas-
sive sentence structure, to name a few examples in English. This
contradiction has caused much debate, and may be the reason why
some researchers have concluded that second language development is
fundamentally different from first.
There is a very recent proposal which may help us to understand
this problem more clearly. Slabakova (2008) has brought together a
large body of research and observed that the one thing the ‘unacquir-
able’ grammatical structures seem to have in common is functional
morphology. She hypothesises that grammatical function words/mor-
phemes cause a ‘bottleneck’ for acquisition. They are a problem for
acquisition because they are extremely complex and often there is no
casually observable connection between form, function and meaning.
If you have tried to teach articles, you will know that, even though
there are general rules that work most of the time, there are also
problematic ‘exceptions’ that seem to defy explanation. Another
common problem with functional morphemes is that one form can
have lots of functions. The -s in English, for example, is used for verbal
agreement, plurals and possessive structures; and its phonetic realisa-
tion also varies, from /s/ to /z/ to /Bs/, depending on the preceding
phonetic form. This does not mean that they cannot be explicitly
learned, but it does mean that they are not easily learned for all of their
functions and with all relevant grammatical distinctions. The Bottle-
neck Hypothesis put forward by Slabakova proposes that these bits of
language are problematic for natural, implicit acquisition because they
hold a large, complex bundle of grammatical features and functions
without any clear connection to meaningful conceptual notions. One
strength of this hypothesis is that it is based on a large and growing
body of existing data, including the well-known case studies confirm-
ing learners’ continued difficulty with functional morphology that we
discussed in Chapter 3.
The implication of the observation that different domains of lan-
guage develop differently is that we should teach the different aspects
of language differently. If learners can readily acquire both meaning
on the one hand and the very core properties of syntax on the other,
then we would do well to provide learners with plenty of rich exposure
and create a learning environment in which they are actively engaging
with language to allow for implicit development of these aspects of
110 Language Teaching

language both in the language module and in general cognition. At the

same time, we should expect problems where language depends on
function words/morphemes.
Yet we should not despair. Slabakova argues that, eventually, these
things can be acquired. A MOGUL approach suggests that, instead of
waiting for acquisition to occur eventually, these areas can be explicitly
taught. Intelligent, motivated learners can consciously learn the form–
meaning connections of function words/morphemes and practise them
until they begin to occur automatically. This is especially possible for
grammatical forms that follow more straightforward form–meaning
patterns. For example, the 3rd person -s in English is a fairly dependable
rule, with a limited set of exceptions. With ample practice, a learner
should be able to master this. The problem is that some areas of language
are very complex, with linguists still debating their exact properties.
There are several analyses of the grammatical properties of articles in
English, for example. This makes it difficult for teachers to be able to
teach these properties fully. Fortunately, this is an extreme example.
Most of the grammatical forms do follow a fairly clear set of ‘rules’ which
are generally accurate and can be explained and learned by students.
So where does this bring us? Because language learning is variable
across linguistic subsystems, we should firstly be aware of the linguis-
tic particulars of that variability. Then we should make decisions
about the type of input, to include both ‘natural’ and explicit pre-
sentation of the target language. Most important is the question of
how we teach. This observation says that you are better off using your
limited class time explaining the rules for functional morphology than
for syntactic structure. Yet MOGUL tells us that explanation is not
likely to be enough. Students will need to work hard to understand
them and practise them for the repeated activation required for
learning. Returning to the other broader generalisation, you will also
want to explicitly teach ‘meaning’, which comes from experience and
use of language. Students can readily develop a large inventory of
words; so teach vocabulary. Students can also learn how to use the
language appropriately; so teach differences in formality, registers,
genres, and even dialects. But throughout, make sure your learners get
plenty of natural input both inside and outside the classroom.
In sum, as expert teachers, we need to develop our understanding of
linguistics so that we can take different aspects of language into
account in our language classroom. Aspects of language that are
modular should be singled out from aspects that are not. Core syntax
and subtle aspects of semantics can be acquired. The more experience-
 Implications for Language Teaching 111

based aspects of non-modular language can be readily taught and

learned. The more complex aspects of function words/morphemes
need as much explanation as possible and lots of practice. They may
also require an acceptance by teachers and learners that there may be
limits on how well they can be mastered. By knowing what kind of
knowledge we are dealing with, we can then make decisions on how to
teach (or not teach) that knowledge.

OBSERVATION 7: THERE ARE LIMITS ON THE EFFECTS OF FREQUENCY

ON SLA

In this discussion of observations we have mentioned the importance

of frequency several times. And the Acquisition by Processing Theory
of MOGUL defines learning as strengthening of linguistic forms
through repeated activation. Yet while there are neurological studies
on processing, there is little research on the neurological basis of
learning itself. The existing psycholinguistic research may be more
indicative. There is a large literature on priming which shows that the
activation of one word allows for faster recognition of related words.
The explanation is that the brain is a complex network of associations;
repeated activation of a word leads to higher resting levels, which in
turn give rise to faster activation. This can be seen as ‘learning’. The
clear implication of this kind of research is that there is a strong role
for frequency in language development. But this observation singles
out the limits of frequency. If learning happens through repeated
activation, why is it not the case that extensive exposure and practice
is enough for successful language development?
To make sense of this, it is important to know that this kind of
research has mostly been limited to the level of words.10 There is ample
evidence for the storage of words in associative networks. Given the
complexity of language, it is difficult to conduct valid neurolinguistic or
psycholinguistic studies that truly isolate other aspects of language.
Unfortunately, some researchers have used the findings from word-
level studies to generalise to language in general. Yet it should be very
clear by now that language is much more than a list of words. So, one
reason why there is a limit on the effect of frequency in language
development is that some areas of language development require more
than repeated exposure. If we think about the complexity of language,
we can appreciate that there are many complex systems involved in
language processing. Aside from the complex morphosyntactic proper-
ties of any utterance, processing involves either audio or visual systems
112 Language Teaching

as well; this will by necessity implicate systems for decoding pronun-

ciation or writing systems, respectively. And all this says nothing of the
content of the message that must be decoded all in the same instant.
One conclusion, then, is that there are limits to the effects of
frequency because of the sheer complexity of the processing load
on the learner. VanPatten goes further to argue that language devel-
opment depends on processing constraints. In his Input Processing
approach, development occurs when the learner’s limited processing
abilities initially lead to a misinterpretation. If led to the correct
interpretation instead, then learning occurs.11
Another implication is that the teacher should not only provide
ample and frequent input, but make sure that particular linguistic
features in the lesson are salient as well. After all, even though a new bit
of language may be very frequent in the input, if it is not picked up on by
the learner, then frequency is irrelevant. Much has been made of the role
of ‘noticing’ in language development.12 Yet, as pointed out by Trus-
cott (1998), even though teachers and researchers alike understand this
idea of noticing intuitively, there is no theoretical or psycholinguisti-
cally based definition of what noticing is exactly. For this reason, some
have suggested that teachers should not concern themselves with what
may (or may not!) be happening in the learner’s brain. Instead, a good
teacher will give much time and attention to those things they can
control: namely, the nature of the input. Sharwood Smith (1993)
proposed Input Enhancement as a way of making desired linguistic
features more salient to the learner. Enhancement can be done in a
number of ways, from making the target feature(s) bold or underlined
in a text, to asking students to pick out the feature(s).
If we also bear in mind the psycholinguistic research on priming, we
may want to point out related or associative forms in our teaching.
Emphasising associations is clearly a good idea for teaching vocabu-
lary, and it will be effective for teaching collocations like weather
forecast. Whether it is useful for more grammatical forms is still a
matter of research. Additionally, whether association facilitates more
pragmatic or cultural aspects of language is also an open question.
Given the limits in research, you may want to explore the effectiveness
of salience and association for different aspects of language. Or, you
may want to remain faithful to the research to date and teach different
aspects of language differently. As such, teaching by association with
frequent exposure and repeated activation may be the way you teach
words and fixed phrases, with consideration of processing constraints
for other grammatical aspects of language.
 Implications for Language Teaching 113

As a final note, we must not forget that processing refers to a

meaningful activity. Frequency alone says nothing about meaningful
processing. We will discuss the importance of interaction when we
discuss Observation 9, but will simply point out here that the limits of
frequency are also explained if that frequency only occurs in mean-
ingless activities.

OBSERVATION 8: THERE ARE LIMITS ON THE EFFECT OF A LEARNER’S

FIRST LANGUAGE ON SLA

In Chapter 2, we discussed how Contrastive Analysis is intuitively

appealing, but has been shown to be problematic conceptually and
empirically for many aspects of language. In other words, research
does not show that it is necessarily easier to learn morphosyntactic
features that are similar in the first and target languages; nor is it
necessarily harder to learn those that are different. Yet we have seen
that there are transfer effects in second language development. In
Observation 5 we discussed how the native language can be the source
of a learnability problem in second language development in certain
areas, using the Subset Principle as an example. These two points tell
us that we should be aware of the first language as a factor in second
language development, but that there are limits on the effect.
The implications for language teaching reinforce an emerging
theme: there are many points of which a good teacher should have
awareness, while also recognising that it is only one factor among
many. The existence of the first language is one major difference
between native and second language development. The MOGUL
framework presents a picture in which that first language is ready
for activation, in response to any linguistic input. But we now have a
growing number of other factors which may limit the effect of the first
language. Salience, as mentioned above, is likely to affect activation.
When we are teaching the appropriate use of some language form, for
example, the target language context is likely to be the more salient
context for a learner than a native language context. This may keep
the native language from being activated, thereby limiting the effect of
the first language.
This conscious, or explicit, attention to language stands in con-
trast with a more implicit process whereby the first language may
facilitate or inhibit development. Yet, we know that even in terms
of core or modular language, there are limits on language transfer.
There is the possibility of UG-constrained modular development; the
114 Language Teaching

research shows a degree of systematic development regardless of the

native language of the learner. This suggests that more universal
principles of development may at times outweigh language transfer.
Though our approach has been exclusively limited to the properties of
language, the importance of the native language will also vary from
one teaching context to another. While it is common for language
teachers in a second language teaching setting to have students from a
range of first language backgrounds, for example, a foreign language
teacher may find all the students share a mother tongue. We will bring
in these more practical issues of classroom teaching as this book
progresses.
Between the theoretical issues of language and the practical issues of
the classroom, there are other issues related to the first language.
There are, for example, some well-known stereotypes associated with
particular groups of language learners. Many of these stereotypes are
likely to be more of a product of particular educational practices than
properties of the native language itself. For example, for adults coming
from an educational system that relies heavily on rote memorisation, it
is not surprising that learners have highly developed memorisation
skills. Similarly, for those coming from an educational system that
makes heavy use of orality or, conversely, literacy, it is not surprising
to find that these learners have disproportionately developed abilities
in speaking over writing, or vice versa. As a teacher, you can tap into
existing learning strategies to enhance language development while at
the same time actively working to develop the less-developed skills, as
appropriate. We will close this discussion by reiterating that the native
language is only one of a range of factors that are implicated in
language development. A good teacher will have an awareness of the
learners’ native language, but will not accommodate L1 transfer
blindly since the effects are limited.

OBSERVATION 9: THERE ARE LIMITS ON THE EFFECTS OF INSTRUCTION

ON SLA

We have already drawn the conclusion that there is a degree of

‘natural’ acquisition in SLA, and we have discussed incidental learn-
ing. Both of these conclusions suggest that the power of instruction is
limited. But this observation makes reference to another very impor-
tant factor in language development, the role of learner interaction.
This is captured by Michael Long’s Interaction Hypothesis, which
argues that language development depends on learner interaction
 Implications for Language Teaching 115

because it is not only the challenge of making sense of but also making
sense in a language that facilitates language development.13 Long’s
view is that second language learning does not depend on just internal
or external factors, but instead on the learner’s active engagement
with the language.
This may seem like an obvious point from the advantage of present
views of language teaching. However, it was an important contribu-
tion at the time when it was proposed as an advancement on Krashen’s
Input Hypothesis. Like Krashen, we have highlighted input; yet it
should be clear that input is only one of a number of important
factors, and, as argued by Long, certainly not sufficient on its own for
successful language development. Instead what matters is what lear-
ners do with the input to which they are exposed. Also proposed at
this time to counter the Input Hypothesis was Merrill Swain’s Output
Hypothesis.14 Swain researched Canadian immersion language pro-
grammes which showed that L1 English children who did their
education through the medium of French had very good comprehen-
sion abilities, but were less proficient in writing and speaking. This led
her to conclude that language production is not just a pedagogic by-
product of language teaching, but is necessary for language develop-
ment, especially when it comes to language production.
Long’s Interaction Hypothesis captures both input and output,
relying on the dynamic process of negotiation of meaning. As learners
struggle to understand input and as they strive to produce output, the
process itself will translate into language development.15 It is this
negotiation of meaning that makes input comprehensible, and this
that enables learners to ‘notice’ gaps in their interlanguage. Students
must actively work to make sense of the input from both the teacher
and their fellow students. And they must modify their output in order
to be understood.
This emphasis on the learner means that the effect of instruction is
limited. The implication, most broadly speaking, is communicative
language teaching. Communicative activities require interaction. A
more specific application of the Interaction Hypothesis is task-based
learning. A task is an activity between two or more learners with a
goal that is not linguistic, but instead designed to require the use of
particular linguistic features. So, for example, asking one student to
refer to a map to direct another student who cannot see the map has
the goal of manoeuvring to a destination, but depends on the use of
language. More specifically, this task requires the use of imperatives as
well as the prepositions and adverbs needed for giving directions.
116 Language Teaching

Task-based teaching establishes a natural platform for Focus on Form.

In carefully devising tasks, a teacher can create situations which
require the learners to use particular forms which may be new to
them. We will discuss task-based learning and teaching further in the
next chapter, but will close by saying that when we say the effect of
instruction is limited, we are especially referring to limitations of
explicit language instruction, such as traditional Focus on FormS
techniques.

OBSERVATION 10: THERE ARE LIMITS ON THE EFFECTS OF OUTPUT

(LEARNER PRODUCTION) ON LANGUAGE ACQUISITION

This observation clarifies one of the points from the preceding ob-
servation. As we just saw, output is an important feature of language
development. Negotiation of meaning requires modification of out-
put, which can, in turn, challenge learners to use their full range of
linguistic resources. Moreover, learners might need to be pushed to
produce forms that challenge their current level in their output.
Otherwise, there is a tendency for learners to avoid difficult language
structures. Research in avoidance shows that high accuracy levels in
language production among some learners may disguise actual ability,
as they may be using a strategy to avoid more difficult language
constructions.16 A learner who is pushed to produce difficult language
will not be able to engage readily in avoidance.
The observation that production is limited is given because there is
no research that directly shows that the act of producing language
leads to higher or better language ability. Even though research in
immersion classrooms shows that learners who do not regularly
produce language had limitations in their production, this does not
mean that those who do produce language will necessarily be more
accurate or have higher levels of fluency. Production can be seen as a
reflection of the learner’s ability at any given point in time; it has not
been shown to be an agent for change itself, at least not change to the
learner’s implicit interlanguage grammar.
There may be differences between the effect of output on explicit vs
implicit knowledge. It may be that output has more of an effect on the
status of explicit metalinguistic knowledge. Through practice, parti-
cular phrases may be memorised, leading to increased speed of
production, or automatisation. While this may be useful for single
words or set phrases, reliance on rote repetition will be limited for real
language in real time, since language is creative and not just a list of set
 Implications for Language Teaching 117

phrases. It is also the case that output is generally understood to refer

to meaningful production and not rote repetition. Thus, practice in
these terms would refer to a kind of self-rehearsal, more than rote
repetition. Rehearsal could very well prepare a learner for a specific
interaction, as the learner could draw on metalinguistic knowledge to
prepare relevant vocabulary or phrases. Additionally, a learner could
think through specific grammatical rules underlying the sentences they
plan to use. This is the logic behind Krashen’s famous Monitor Model.
Learners are able to monitor their production, drawing on the current
state of their acquired knowledge and modifying their output based
on their explicit metalinguistic knowledge. Of course, the real-time
pressures of spontaneous speech may limit the effectivity of monitor-
ing, but adept development of strategies coupled with memorisation
of phrases may be enough to allow for a seemingly native-like
proficiency.
The conclusion is not that output is not useful for the learner; it is
only the observation that the effect of output on acquisition is limited.
The implication for teaching is that for language activities that are not
constrained by time, monitoring may be very helpful. And though
learner rehearsal may not actually lead to language acquisition, it may
lead to more successful interactions, which, in turn could lead to
increased levels of automaticity and increased levels of confidence.
Another implication for teaching is to do with error correction. If
output reflects the current level of language, an error in output may
reveal an underlying rule which the teacher could draw attention to,
working with the learner to develop specific strategies to overcome it.
With increased awareness, explicit instruction may help the learner to
self-correct for more effective fluency.

CONCLUSION

In this chapter we have discussed VanPatten and Williams’s ten

generalisations about second language development, looking for im-
plications for the classroom. One important point to note is that the
implications presented here are logical conclusions; not all are em-
pirically based. Applied generative SLA in particular is one area which
is in much need of empirical research.
One theme that has clearly emerged from our discussion of these
observations is complexity; language itself is complex, and the process
of language development is complex. Knowledgeable teachers need to
take on the complexity of language and language development – in
118 Language Teaching

addition to the many other practical factors and constraints that

accompany any language teaching context. A second theme also
emerges: there is a clear role for teachers and teaching in successful
language development.
Language development is not just a matter of language exposure;
teachers must decide what the input will be and how to expose the
learner to it. Language development is not just a matter of output;
teachers must make that output meaningful and it should require use
of language that challenges learners, whatever their current level.
MOGUL can help a teacher to sort through the balance between
‘natural’ and complex implicit input and explicit instructed input.
Research to date suggests that learners can acquire aspects of meaning,
including pragmatic knowledge, the meaning of words and more
subtle aspects of semantics which even linguists have difficulty ex-
plaining. Core properties of syntax such as word order and sentence
formation can also be acquired. This suggests that these are areas
which will develop in response to rich, authentic input. Other areas of
language, such as native-like pronunciation and functional morphol-
ogy, seem beyond the reach of natural acquisition except in excep-
tional cases. These are areas that will need more explicit instruction.
Given the limited time constraints of most language classrooms,
teachers would do well to decide selectively what to use their time
teaching on explicitly. Moreover, while there is much variability
among learners, there seem to be constraints on the extent to which
older learners can acquire language to native-speaker level. MOGUL
reminds us that explicit language knowledge can compensate for
implicit language knowledge, however. Learners can develop meta-
linguistic knowledge about areas of language that have the potential
for implicit development. With practice, this metalinguistic knowledge
can become automatised so that it allows for fluency. With this
understanding of language and language development, we now turn
to more explicit discussion of the classroom as we continue to apply
these ideas to classroom teaching.

FOR DISCUSSION

1. Think of your own experience with teaching (or learning) a

language. Can you come up with examples from your personal
experience that support each of the observations?
2. Again based on your experience, can you add any observations
about language development to those discussed in this chapter?
 Implications for Language Teaching 119

3. The ten observations have been discussed with little regard to

particular contexts or types of learners. Consider each observation
from the point of view of each the following variables to discuss the
extent to which they are relevant.
a. age
b. proficiency
c. native language
d. number of languages learned
e. learning environments: for example, naturalistic vs instructed
f. reasons for learning language: for instance, motivation.
4. Can you think of any other factors that might affect the develop-
ment of language?
5. Thinking of your own experience as a teacher (or learner), what
have you done in the past to address (some of) the observations for
more effective language development?
6. Consider each observation from the point of view of a language
teacher. What could you do to take each into account? Are some
easier to address than others?

NOTES

1. Stephen Krashen has devoted much of his more recent career to re-
searching the effectiveness of ‘sustained silent reading’, which, he argues,
has enormous potential for bolstering learners’ second language ability.
See Krashen (2004). There is also research showing the role that reading
can play in the development of vocabulary. See Nation (2001).
2. See Schwartz (1993), who defends the position that learned and acquired
knowledge are and remain distinct.
3. See the edited volume by Doughty and Williams (1998) for a discussion
of Focus on Form and its relationship with Focus on Meaning and Focus
on FormS.
4. The work of Dekydtspotter and colleagues (see note 11 of Chapter 3)
is all within the realm of semantic interpretation, as is the work of
Marsden.
5. Early research on the development of negation in L2 English focused on
child L2. See Wode (1977) for research on German native-speaker
children, and Schumann (1979) for research on Spanish native-speaker
children, both acquiring English negation. Hyltenstam (1977) also finds
stages of development in acquisition of negation for L2 Norwegian
learners. Later work researched negation as one property of a larger
parameter cluster. For example, see White (1992) for research in L2
English, and Yuan (2001) for L2 Chinese research.
120 Language Teaching

6. Stages of development constitute an area of L2 research that is in need of

more study. The most developed line of research in this direction at the
moment is Pienemann’s Processability Theory and associated work in
Teachability. See Pienemann (1998, 2005).
7. The best-known research in this area is based on the Accessibility
Hierarchy for relative clauses (Keenan and Comrie 1977). The original
study showing that learners who are taught a more difficult, or marked
form, is Gass (1982). Since then numerous studies have been done, some
confirming and others disconfirming this result. For a more recent study,
see Ammar and Lightbown (2003). Not surprisingly, additional research
has led to the realisation that there is much complexity in terms of the
language facts, which may affect overall results.
8. Research by Snow and Hoefnagel-Höhle (1978) included English speak-
ers learning Dutch in an immersion setting. The adults were all middle-
class, educated professionals. Despite the possibility that these learners
were unusually diligent and motivated, the larger point still holds: rate of
development is distinct from route of development. See also Wode
(1978).
9. Flege, who researches L2 phonology in general, has also done work on
age and phonology; see Flege (1999, 2009) in particular.
10. See Paradis (2004) for a discussion of how most of the research in
neuroimaging is limited to words.
11. See VanPatten (1996, 2004) for more detailed discussion.
12. The Noticing Hypothesis was proposed by Schmidt (1990, 1993) and
was subsequently adopted by many notable researchers. See, for exam-
ple, Ellis (1994), Long (1991) or Robinson (1995).
13. See Long (1981, 1983) for the original formulation of the Interaction
Hypothesis and Long (1996) for an update.
14. See Swain (1985) for the original formulation of the Output Hypotheses
and Swain (2005) for a recent discussion.
15. Another significant contribution from Michael Long has been to inspire
a research agenda that analyses interaction among native and non-native
speakers. This has led to a large body of literature on feedback tech-
niques such as repetition, clarification and expansion. While much of the
early research was conducted in artificial contexts, more recent work has
studied real classroom interaction. See Keck et al. (2006) for a meta-
analysis of research on interaction and Russell and Spada (2006) for a
meta-analysis of research on corrective feedback.
16. Schachter (1974) took a Contrastive Analysis Hypothesis approach to
argue that avoidance occurs when there are differences between the
native language and target language. While this argument has been much
debated since then (see Laufer and Eliasson 1993, for example), there
seems little doubt that avoidance of particular forms is a common feature
of L2 production.
 6
APPROACHES TO ENGLISH LANGUAGE TEACHING

INTRODUCTION

In Chapter 2, we took a historical perspective, exploring how differing

views about language and language development have translated into
approaches to language teaching in Europe over the ages. We saw that
the influence of classical language in the early years of language
teaching meant that the method of teaching Latin was adopted for
teaching other European languages. This Grammar Translation method
was challenged, however, by the Reform Movement, which looked
to native language development and proposed the Direct Method,
requiring exclusive use of the target language in the classroom. In
some ways, this quite radical approach was ahead of its time. The
enormous difference between Grammar Translation and the Direct
Method no doubt contributed to the latter’s failure, since change tends
to happen much more gradually.
The prevailing viewpoints of the growing academic fields of lin-
guistics and psychology from this time left a mark on the language
teaching field. A structuralist view of language provided a list of
grammatical structures to consider alongside grammarians’ language
rules. And the emerging behaviourist view of habit formation en-
couraged teachers to adopt an approach which depended heavily on
repetitive practice and drilling. The progressive contribution of the
Oral and Situational Approaches was not only to emphasise spoken
language, but also to recognise that language should be presented in
meaningful contexts. Less progressive was the behaviourist influence,
as the Oral Approach in the UK and the Situational Approach in
Australia looked very much like the Audiolingual Approach in North
America in terms of classroom practice, relying heavily on repetition,
memorisation and drilling.
This was the situation in English language teaching when the so-
called Chomskyan Revolution in linguistics took place. Generative
Linguistics discredited the behaviourist view of language and native
122 Language Teaching

language development, arguing that language is comprised of uni-

versal principles which children possess innately at birth. We ended
Chapter 2 by exploring the effects of this development in SLA. We saw
how the idea that language is a system has also been supported by SLA
research, with investigation into second language learning as a sys-
tematic process. Unfortunately, the early practices of Error Analysis
and Contrastive Analysis did not support the overly optimistic view
that merely identifying where native and target language are the same
and different would be enough to determine which areas of second
language development are easy and problematic, respectively. In the
last chapter we explored the implications of more current under-
standings of second language development for language teaching. In
this chapter we ask how these implications translate into approaches
to language teaching in the more recent period leading up to the
present.
Firstly, however, we need to clarify some terminology. There is a
widely held distinction between method and approach. An approach
to teaching refers to the conceptual principles, views or assumptions
that underpin the way one teaches. These beliefs may be explicit and
supported by research, or they may be deeply held intuitions of which
a teacher is not even explicitly aware. In terms of language teaching, it
includes the understanding of what language is and how it develops;
this influences a teacher’s fundamental decisions. One main aim of this
book is to explore some of these fundamental assumptions in order to
challenge you as a teacher to be explicitly aware of your under-
standings and beliefs. These beliefs will affect your choice of method.
Method refers to the way the language is taught, the type of activities
and how these are structured and presented.1
The following exemplifies these two terms: if I held a structuralist
view of language, this would affect my overall approach, as I would
view my task as being to teach sets of language structures, emphasising
grammatical rules and correct forms. And if I had a behaviourist view
of language development, I would want to appeal to memorisation
and repetition. Thus, the method I would choose would be Audio-
lingualism. In this chapter we will be looking at both approaches and
methods in language teaching. At times, we will also make reference to
more specific techniques of classroom teaching, like lesson planning
and error correction, for example. However, these more specific
aspects of teaching are more the subject of the last couple of chapters
of this book.
 Approaches to English Language Teaching 123

‘DESIGNER’ APPROACHES

In this section we will briefly survey what I will call ‘designer’

approaches, exploring a range of language teaching practices that have
been devised in order to translate specific beliefs about language or
language development into classroom practice. And, as there are limits
to what one book can do, we will limit our discussion to English
language teaching. The period from the 1960s to the 1990s saw much
experimentation as new developments in language study inspired new
and creative approaches to language teaching. Unsurprisingly, new
ideas tended to influence the profession locally, at least in the first
instance. Thus, in the early part of the ‘modern’ period there were
differences in North America on the one hand and the UK and Australia
on the other. As we will see at the end of this section, however, by the
1990s the differences had all contributed in their own way to a more
general trend among English language teaching professionals globally.
We start this discussion of designer approaches in North America.
The Natural Approach of Krashen and Terrell (1983) was based on
what has come to be known as Krashen’s Monitor Model.2 An
attempt to apply the new ideas coming out of Generative Linguistics
to language teaching, the Monitor Model included five hypotheses.
The first, the Acquisition-Learning Hypothesis, made the very useful
distinction between implicit and explicit learning, with acquisition as
subconscious, implicit development of language and learning as the
conscious, explicit process. A speaker draws on acquired knowledge
without thinking about it, while learned knowledge is explicit meta-
linguistic language knowledge. Additionally, acquisition is associated
with the language module and learning with general cognition. For
Krashen, the two are distinct processes; learning cannot become or
affect acquisition. The acquisition-learning distinction was appealing
to teachers since there is a well-known tendency for learners to ‘know’
a form, but still make mistakes. With Krashen’s distinction, we can say
that they have learned the form, but they have not acquired it.
The second hypothesis is the Monitor Hypothesis, whereby learner
output, which originates from acquired knowledge, can be con-
sciously ‘monitored’, making use of learned knowledge. In order
for a learner to use the Monitor, however, there are some require-
ments. There must be time for reflection during output; the learner
must pay attention to the form of the output; and the learner must
have enough explicit knowledge of the relevant language rules in order
to self-correct.
124 Language Teaching

The next two hypotheses are closely related. The Natural Order
Hypothesis specified that rules of a language are acquired in a
‘natural’ order. For this part of the Monitor Model there was little
discussion of differences between native child language development
and adult second language development. Krashen put much stock in
the research he did on morpheme orders, which showed very little
differences between in language development in terms of age or native
language – at least for the development of English functional mor-
phemes.3 For Krashen and the Natural Approach, then, language
development followed a natural order depending more on an internal
syllabus than on instruction. Because there is a natural order, acquisi-
tion must proceed from stage to stage. The Input Hypothesis asserted
that, instead of explicit teaching, what is really needed is lots of input
‘appropriate’ to the stage of the learner. More specifically, learners
need comprehensible input, also known as i + 1, where i is the current
level of the learner and 1 is the stage just one step beyond the current
level. The role of the teacher in the Natural Approach, then, is to
provide sufficient amounts of input at the right level of development.
The final piece of the Monitor Model puzzle is the Affective Filter
Hypothesis. Recognising that adult language development is different
from child development, Krashen suggested that adults have a ‘filter’
which determines how much of the comprehensible input is ‘allowed’
into the language module. In Krashen’s terms, the filter would be
raised if a learner had little interest and motivation or if other affective
factors led to a less than ideal learning environment. In addition to
child–adult differences, this filter can explain individual variation
among adult learners since learners differ widely in terms of affective
factors. Like many aspects of Krashen’s ideas, this affective filter idea
may be intuitively appealing, as it seems to capture what we as
teachers experience with our learners. The problem, however, is that
there was and is no principled basis for it – there is no theory in
psychology or linguistics which includes any kind of filter like this, nor
is there any empirical way to prove that such a filter might exist.
Based on these basic tenets, Krashen and Terrell developed the
Natural Approach, which emphasised comprehension first, putting
much stock on input from the teacher. While this input was expected
to be graded, following a natural order, in reality the approach was
not able to specify what that order might be. Instead, the emphasis was
on sufficient input with ample time for learners to acquire before
producing language. At low levels, learners were allowed to go
through a silent period, much as native language learners do. With
 Approaches to English Language Teaching 125

early production, errors were to be ignored in order to encourage

fluency. The emphasis on fluency meant classroom activities that were
communicative in nature. Also essential in this approach was a
commitment to a very supportive atmosphere in order for affective
filters to be lowered as much as possible.
While Krashen is to be respected for trying to relate work in
theoretical linguistics to language teaching, he went beyond what
the theory could actually support, both empirically and conceptually.
And the Natural Approach itself was not quite the success in the
classroom for which he might have hoped. Second language devel-
opment in the classroom cannot be left entirely to ‘natural’ tendencies.
Unfortunately, the Chomskyan ‘revolution’ in linguistics did not have
the revolutionary effect on classroom teaching for which some applied
linguists may have hoped. In fact, the failings of the Natural Approach
brought to mind the much earlier failings of the Direct Method. The
problem is that second language learning is different to first. Adult
language learners, for example, do not necessarily go through a silent
period as young children do. Yet, despite the failings of his Monitor
Model and Natural Approach, Krashen made an important contribu-
tion and his ideas remain influential. As has been made evident in this
book, generative research does have an important role to play in
informing classroom language teaching. But as we have seen, this
requires a much more sophisticated understanding of what is happen-
ing in language development than the picture painted by the Monitor
Model.
Krashen’s ideas contrast with approaches that had developed in
other parts of the world. The predominant view of language in
Australia and the UK in the 1970s and 1980s was the functionalist
view of Halliday. Building on the tradition of the Oral/Situational
Approach, David Wilkins (1976) proposed the Notional Syllabus. As
reflected in the title, language in this view is seen as a set of conceptual
notions. Examples of notional categories are time, quantity or loca-
tion. However, as a syllabus based only on notions is a bit limited,
connections were made between notions with specific language func-
tions, such as requests, denials, offers, apologies, and so on. Devising a
syllabus in terms of notion and function provided a concrete step away
from the traditional reliance on language structure as an organising
approach to teaching. This allowed for a genuine shift to teaching
language in context. After all, a lesson devised around a notion such as
location and a function such as requests leads very naturally to
activities that embody language in use. This gave rise to the type of
126 Language Teaching

textbook that is still popular as a format for modern textbooks, with

lesson titles like ‘At the train station’ or ‘Asking or directions’.
The move to a notional syllabus was well received and reasonably
successful, as teachers and learners alike recognised the importance of
meaningful language learning. All of the approaches of the post-war
period were trying, in their own way, to move away from the
traditional tendency to teach language as a static set of grammar
rules. This tendency was (and still is!) seen in the PPP approach to
lesson planning, which lends itself to teaching specifically identified
language points or rules and then asking learners to practise and
produce them. While PPP can be extended to more functional features
of language, it is, in effect, inherently structuralist, as it breaks
language down into structures for presentation.
In addition to approaches to teaching based on particular views of
language, there have been designer approaches that develop other
pedagogic principles. While we will not discuss any of them in great
detail, we will try to give a flavour of some of the so-called ‘alternative’
approaches that have been proposed.4 Perhaps the best known of the
designer approaches is the Silent Way, devised by Caleb Gattegno.
The main idea of this approach is that learners should produce as
much language as possible, and teachers should be as silent as
possible. This approach to teaching language adopts the widely
accepted pedagogic principle that learning occurs through problem-
solving and discovery, and not through the teacher simply telling
students what they ought to know. The Silent Way does recognise that
some spoken input is needed. At times, teachers do model new
language forms, but the majority of the lesson requires learners to
work out language rules, to repeat new target forms and both to self-
correct and peer-correct. With limited oral input, the Silent Way
makes much use of physical objects, body language and gestures.
Obviously, this view sees classroom second language learning as very
different from native FLA. Proponents of the Silent Way insist that this
difference should be explicitly recognised.5
The Silent Way stands in stark contrast to Comprehension Ap-
proaches like Krashen’s Natural Approach, which emphasise the
understanding of language over production, at least initially. The
Learnables, devised by Harris Winitz, also de-emphasises learner
production. This approach depends exclusively on pictures accom-
panied by sound. Mimicking the child experience, learners are ex-
pected to learn by looking and listening.6 Another comprehension
approach, James Asher’s Total Physical Response (TPR), highlights
 Approaches to English Language Teaching 127

the importance of kinetics; learners learn language by reacting phy-

sically to language input. TPR develops the ideas of educational
psychology in which learning is rooted in stimulus–response cycles.
By asking learners to respond physically to language commands, the
right brain is engaged for motor functioning and the left brain for
reasoning.7 As in the Natural Approach, learners are not expected to
speak until they are ready. For advocates of comprehension ap-
proaches like the Learnables, TPR and the Natural Approach, the
absence of pressure to produce language will reduce stress and
facilitate learning.
The idea of reducing stress is taken even further by Suggestopedia,
developed by Georgi Lozanov, a Bulgarian psychologist with an
interest in yoga. Relaxation to allow the learner to be primed for
learning is essential in this approach. A Suggestopedia classroom is
immediately identifiable by its comfortable chairs, candles and music,
used to create a relaxed atmosphere. Coupled with this idea of ‘psycho-
relaxation’, however, is the importance of the teacher as an authority
figure. The idea is that learners will be more susceptible to learning if
they enter a more child-like state, which will happen if the teacher
carries authority. Another child language-inspired requirement is that
the teacher exaggerates intonation when speaking, emphasising the
rhythm of the language. In terms of the actual lesson, the procedure is to
use what is referred to as a Séance or Concert Session, in which the
teacher ‘performs’ a dialogue which students follow, looking at a script.
The script may have notes or even native language translations, and
particular vocabulary or difficult grammar will be identified and
taught. The dialogue is then delivered again, but this time the students
listen without the aid of a text. At the end, students leave the class in
silence, absorbing the day’s lesson for themselves.8
Another approach based on ideas from psychology and, more
specifically, counselling is the Community Language Learning of
Charles Curran. In this approach, the teacher is a counsellor and
learners are clients. Language is a social process of collaboration and
community development. Learning will occur if the whole person is
engaged, not just their intellect, but their emotions and feelings.
Modelling the stages of development from childhood to adulthood,
learners are expected to feel insecure in early stages before being able
to assert a degree of independence. In doing so, they will begin to
establish an identity in the new language. It is only at more proficient
stages that the learners will be able to accept criticism, and so error
correction should be avoided until then.9
128 Language Teaching

The final designer approach that we will consider here, Rassias, is

named after its founder, John Rassias. For Rassias, language is a
cultural artefact which must be learned from within the social context
of the target language. Since there is no such environment in the usual
foreign language teaching setting, it is important for the teacher to
create an immersion environment which mimics the target language
environment as far as possible. This aspect of the approach requires
skills in drama, as there is a degree of ‘acting’ involved. It also requires
much enthusiasm and energy from the teacher, who is expected not
only to create an authentic environment, but also to eliminate the
inhibitions of the learners so that they can take part in their new
environment. In practical terms, there is also use of repetition and
drills, but in such a way that the emphasis is on positive reinforcement,
not rote memory.10
This idea of creating a target language environment is an attempt to
create an immersion situation when immersion is otherwise not
possible. Total immersion has been advocated as the most effective
way to promote language learning. In an immersion programme,
content classes such as geography and science are taught through
the medium of the target language while the target language itself does
not have explicitly dedicated class time. Because immersion is the
closest setting to a native environment, it might be expected that this
would be the most successful of language approaches. However, as we
saw in the last chapter, it has been found that such programmes do not
necessarily result in higher language ability than tradition language
classes, especially in terms of learners’ ability to produce language.
Nevertheless, it remains a viable option and should be considered as
an option if the circumstances allow it.
Immersion programmes are normally associated with primary level
schooling. When implemented at secondary levels or above they tend
to take the form of Content-Based Instruction (CBI). Again, the idea is
that the object of the lesson is some meaningful content, like biology,
for example. But this content is taught through the medium of the
target language.11 Because the nature of the content for adults is likely
to be relatively advanced, CBI is often associated with advanced levels
of language learning as well. Thus CBI is a type of English for Specific
Purposes (ESP), in which English classes are tailored to the specific
needs of the particular set of learners.12
While all of these approaches are based on pedagogically relevant
principles, they vary in terms of their emphasis and understanding of
what language is, how language develops and what best fosters this
 Approaches to English Language Teaching 129

development. Unfortunately, one thing they all have in common is that

no single approach or method is appropriate for the range of language
classroom settings that exist. And even within the particular context
for which it may be designed, no approach has shown itself to be
foolproof. Against this somewhat pessimistic conclusion, however, is
the more optimistic development of a general trend within modern
language teaching circles. This trend is for a Communicative Lan-
guage Teaching approach to language teaching.

CURRENT TREND: COMMUNICATIVE LANGUAGE TEACHING

The development of Communicative Language Teaching (CLT) is

often seen as a natural outgrowth of the functionalist view of lan-
guage. Notably, however, even though there is a fundamental differ-
ence between the generative view that developed initially in North
America, and the functionalist view originating in Australia and the
UK, both have reasons for agreeing that CLT is an appropriate
approach to teaching. For generativists, language is ‘natural’ and
language acquisition occurs in response to ‘real’ or positive input.
For functionalists, language is a tool for making meaning and lan-
guage will develop through interaction and use. As we will see,
authentic input and interaction are two basic features of CLT.
In fact, CLT has not developed only out of these two linguistic
approaches. Many of the features of CLT are a natural outcome of a
broader trend in education as well. Progressivism in mainstream
education is a twentieth-century trend which emphasises the needs
of students as individuals; it promotes the idea that active learning
through doing and discovery is more effective than the passive absorp-
tion of bodies of knowledge.13 Another reason for converging on CLT
is simply the passage of time. While language specialists have been
calling for a shift away from the teaching of language structures for a
very long time, in practice, fundamental change is often slow. The call
to teach notions and functions, for example, looked progressive, but in
practice it tended to result in lists of phrases that were often drilled and
memorised. In fact, the early attempts to make language teaching more
communicative also tended, in reality, to atomise language and teach
lists and rules. In time, however, the consensus has led to an emergence
of genuinely communicative practices.
CLT is an approach, and not a method. It adheres to a range of
principles, which in turn give rise to particular teaching methods. We
will discuss the main basic principles underlying CLT. The first basic
130 Language Teaching

premise is that language is communication which develops through

use. Because of the importance of language use, it may be more
appropriate to refer to students as language users than language
learners. Accordingly, fluency is emphasised over the more traditional
focus on accuracy in terms of grammar, pronunciation and word
choice, for example. CLT is often willing to overlook accuracy at the
level of detail for the sake of promoting fluent use. Learners are
encouraged to speak without worrying too much about ‘correct’
forms, as long as communication is successful. This is not to say that
errors are entirely ignored, but they are often seen as secondary to the
more important aim of maximising language production. Fluency is
generally associated with production; yet the ability to understand
spoken speech in real time as well as target-like reading speed can be
seen as other aspects of fluency which should be promoted and
developed. Additionally, in practice the emphasis on fluency over
accuracy tends to be true more for spoken than for written produc-
tion, as many CLT practitioners will overlook inaccuracies in speech
which they would correct in writing.
This accuracy vs fluency issue is closely related to the question of
process vs product. Since communication is basic to CLT, there is an
emphasis on process over product. In other words, value is given to the
act of making meaning rather than to the particular choice of output in
terms of form. This is because of the importance of interaction in CLT.
As we discussed in Chapter 5, Long’s Interaction Hypothesis calls for
learners to negotiate meaning. By engaging in actual communication,
language-learning users will have to work through gaps in their
comprehension and production. This hard work requires strategies
for overcoming communication breakdown, which in turn entails a
degree of self-awareness and self-reflection. In other words, learners
will benefit from a sophisticated understanding of and control over the
processes involved in language. Or, to use a term recently coined by
Swain (2006), learners develop competent skills in the process of
languaging. This hard work of making meaning in interaction is the
basis for language development.
Another basic characteristic of CLT, which we have already
touched on, is the importance of language meaning instead of the
more traditional focus on grammar rules. Language lessons should
revolve around meaningful activities. While involved in meaningful
activities, learners will be receiving input which can lead to incidental
learning and natural acquisition. And we know from empirical
research that learners can acquire both core aspects of syntax and
 Approaches to English Language Teaching 131

very subtle properties of semantic interpretation. In order to make

sure tasks will be seen as meaningful, a CLT practitioner will need to
be aware of the learners’ language learning needs. Needs analysis is an
accepted practice in CLT at many levels. This could include formal
assessment that takes place before a course is designed, but in most
CLT settings, it happens more informally, as teachers are expected to
gauge the needs of their learners from day to day, and plan lessons
accordingly.14
While the emphasis is on meaning, there is some room in CLT for the
more structural aspects of language. Any explicit teaching of language
rules, however, should occur only in the context of use. In other words,
there should be Focus on Form, as we discussed in Chapter 5. This
means that any problems with or questions about language forms or
structures should be addressed as they arise. If they are explicitly
introduced by the teacher, this should not be done outside of mean-
ingful language tasks. And when teachers are explicitly teaching them,
language points should be taught in terms of their meaning and use. To
illustrate, if the passive comes up in an English class as an area of
difficulty, the teacher should highlight the usefulness of omitting or
sidelining the subject of a sentence in order to achieve an effect in the
message; the passive should not be taught purely as a grammatical rule
that applies to active sentences. In a classroom that is strictly CLT,
attention to form will occur only if there is breakdown in communica-
tion or if a learner is unclear about some point of language. In CLT,
there should never be Focus on FormS.
It is easy to understand why the emphasis on meaning and com-
munication in CLT leads to an uneasy feeling about error correction,
or what I will refer to as selective error correction. In the interests of
promoting fluency, it is not a priority to correct every non-targetlike
feature of learner production. Moreover, a proponent of CLT does not
want to distract the student continuously from the communicative
message in an interaction, whether between learners or between the
teacher and the student. There is also a sense in which too much
correction may easily discourage and thereby de-motivate the student.
For these reasons, error correction is often restricted to Focus on
Form-type situations – in other words, only when the error leads to a
breakdown in communication.
This is especially true in terms of spoken language production, but
can also apply for written forms. In response to a written piece of
work, CLT teachers should comment on the ideas and/or overall
message of writing. The extent to which they also comment on the
132 Language Teaching

grammar, vocabulary and spelling is likely to vary, unless these errors

lead to an inability to make sense of the written message. While
hesitation to correct errors is compatible with other principles of CLT,
it is also supported by a view which sees language development as
occurring in natural stages, as it may be more useful to overlook errors
which signal a level of development yet to be achieved.
Another feature of CLT is the use of ‘real’ or authentic materials. In
their true form, authentic materials will be spoken and written texts
taken from non-pedagogic sources: in other words, realia such as
magazines, brochures and ‘real’ video or audio broadcasts. In practice,
mainstream language textbook writers usually create texts that are
‘typical’ examples, but written specifically for learners. This enables
closer control of specific vocabulary and grammatical forms. CLT
materials are also devised for authentic language tasks, generally
known as ‘gap’ tasks. Information gap tasks require language-learning
users to gather information. Materials such as role play cards can be
written to simulate ‘real’ situations to form the basis for gathering
information. Language teachers dedicated to unfabricated authentic
materials could bring in cultural artefacts. For example, with maps
and train timetables, students could then be asked to create a package
holiday brochure. This type of ‘real-life’ language activity is commu-
nicative and will be more meaningful to the language-learning users
than traditional exercises.
Because learners are engaged in ‘real’ activities, using authentic
sources, a CLT approach is also characterised by an integrated skills
approach. In other words, a single activity or set of activities in a CLT
classroom is likely to make use of all four skills: speaking, listening,
reading and writing. The more traditional practice of teaching each
skill separately does not sit well with an approach which mimics real-
life interaction. After all, speaking without listening is very odd, and
rarely occurs in natural settings. Though you may question the
interdependence of reading and writing, good writers are very much
aware of their readership, just as sophisticated readers are aware of
the potential biases of the writer. For CLT, the emphasis on genuine
communication means structuring classes in terms of activities that
require a range of skills, instead of singling out a particular skill as the
only aim of the lesson.
Communicative tasks also require active learning, not passive
reception of knowledge. Gap tasks, role plays and debates are ex-
amples of active learning that require interaction among language-
learning users. This will result in higher levels of engagement with
 Approaches to English Language Teaching 133

language by the learner, which in turn will lead to more active

processing. As made clear in MOGUL, this is what is needed for
language development, whether implicit or explicit. The emphasis on
learning from others through interaction makes the use of groupwork
very common in CLT classrooms. Cooperative language learning
requires learners to work together for greater improvement in in-
dividual learning. In addition to offering increased opportunities to
use language, cooperative learning allows students to learn from each
other. If properly conducted, groupwork can also create a supportive
atmosphere for learners, thereby nurturing learners and encouraging
them to challenge themselves to their fullest potential.
CLT is also compatible with a sociocultural view which sees
language as tied to cultural practices and contexts. In order to
promote a connection between language and culture, a CLT class-
room can try to create the atmosphere of the culture where the target
language is spoken. This can be done physically with cultural artefacts
such as pictures, photos, magazines, books, and music. Many CLT
practitioners will also try to foster the development of a second
language identity by giving language-learning users a name in the
target language.15 Of course, any attempt to create a genuine target
language atmosphere requires the (exclusive) use of the target lan-
guage by the teacher, not only during formal instruction but also in the
informal interactions that take place, as well as an insistence that
learners use only the target language while in class. This will maximise
the amount of natural input that learners receive in class.
Another way to maximise quantities of input is to promote learner
autonomy. While the concept of autonomy can mean different things
to different people, the general idea of learner autonomy is that
language-learning users should take ownership of their language
development instead of relying heavily on the teacher and/or the
classroom materials. An autonomous learner will take responsibility
for their own learning by increasing opportunities to engage with
language and taking steps to improve in the particular areas in which
they need more development. After all, the time constraints on most
classroom settings means that learners need to devote time to the
language outside class as well.16 Learner autonomy is compatible with
the idea of reflective learning. A reflective learner will take time to
monitor their strengths and weaknesses in order to decide on the areas
to which they will devote more time and attention. A CLT practitioner
will train learners to be reflective in order to promote autonomous
learning. Once this is achieved, the teacher can act more as a facil-
134 Language Teaching

itator, enabling learners to progress in ways that are most appropriate

to their individual needs. The most common technique associated with
reflective learning is the learner diary, journal or blog, in which
learners are asked to record their thoughts and observations about
the process of learning as they experience it.
The overall view of the learner in CLT is very much in the tradition
of progressive education. In other words, it takes a humanistic
approach with an acute awareness of the needs of the individual
learners in the classroom. This type of learner-centeredness is a part of
the larger trend in education that views learners as people with
emotional needs that affect learning as much as, or perhaps even
more than, their intellectual capacities. A view of a learner as a whole
person with beliefs and desires rooted in personal experience, instead
of an empty vessel to be filled with knowledge, fits comfortably with
an approach which sees language as a tool for making meaning and a
means for participating in a social context. Similarly, it is proposed
that learning should be seen in terms of participation in the act of
learning instead of acquisition of a body of knowledge.17 The huma-
nistic emphasis on fostering personal development supports coopera-
tion over competition in the classroom and moves away from pre-
determined goals which all language-learning users are expected to
achieve in unison.18 After all, we know that there is considerable
variability in second language development.
The recognition of individual differences and needs may be one
reason why one specific teaching method for CLT does not exist. As
should have become clear from this discussion, CLT is an approach to
teaching, not a method. It embodies a range of beliefs and under-
standings about language, language development and learning in
general that have developed over time. One reason why it has been
widely endorsed is because of its broad appeal. Yet, the fact of being
an umbrella term is also one of its greatest weaknesses. With CLT as
an approach, a teacher must look elsewhere for specific methods for
teaching. The combination of a consensus supporting CLT with an
absence of a single specific teaching method has led to what is now
generally accepted to be the ‘Post-Methods Era’.

POST-METHODS ERA

The emergence of the Post-Methods Era is not surprising, given the

complexity inherent to language teaching. From a linguistic point of
view, strict adherence to a single aspect or view of language is going to
 Approaches to English Language Teaching 135

result in limitations, as any one view leaves out other crucial aspects of
language. Add to this the complexity of pedagogic issues and the
demands of particular educational agendas, and it is no wonder that
there is no agreed method for language teaching. Since there has been
no single, universally appropriate, foolproof method for language
teaching, a good language teacher in the Post-Methods Era will adhere
to the general principles of CLT and pick and choose from a range of
methods, mixing them up as appropriate to the particular context and
to keep the class interested and engaged.
Though there is no single accepted method, the one very useful and
flexible method that we will adopt for the rest of our discussion is
Task-Based Learning and Teaching (TBLT), briefly introduced in the
last chapter. TBLT has been endorsed by a range of applied linguists,
from those taking classroom practice as a starting point for thinking
about teaching to those preferring to start from questions of theory,
asking how findings from research in SLA should manifest themselves
in the classroom.19 TBLT sits well not only with the basic tenets of
CLT but with the generalisations that have emerged from research in
SLA as well. A task is a meaningful activity which requires active
engagement with language. Though the task may lend itself to
particular language forms or sets of vocabulary, these formal aspects
of language will support the task instead of taking centre-stage in the
lesson. Thus, for example, a task will be designed with an objective
such as ‘to design a tourist brochure’, and not ‘to learn the grammar of
descriptive adjectives’. In other words, TBLT is about purposeful
activities that emphasise communication because learning happens
through interaction and use of language and acquisition occur in
response to rich input. TBLT activities are usually authentic, though
they may be modified to facilitate communication or to highlight
particular features of language. And, crucially, tasks are things that
learner do; learners as active participants in a TBLT classroom are
engaged so that there is a maximal level of processing, as required for
language development.
One notable feature of TBLT, then, is the roles of the teacher and
the student, the former as facilitator and the latter as active partici-
pant. In observing a TBLT classroom, it may seem as though the
learners are doing all the work while the teacher is sitting by. But, in
fact, good TBLT requires a considerable amount of work from the
teacher from preparing to implementing and following up the task.
Ideally, a teacher will have a predetermined syllabus with clearly
defined tasks. A principled task-based course will include a series of
136 Language Teaching

tasks with intentionally targeted aspects of language which will be

sequenced, based on findings in SLA research that show develop-
mental paths in second language development. Of course, the tasks
will consider the interests of the students and the appropriateness of
the content of each task for the particular pedagogical context as well.
But we will limit the extent to which we become distracted by these
other pedagogical considerations.
Looking more closely at specific linguistic aims for illustration,
tasks requiring simple declarative sentences should precede tasks that
emphasise question formation. Yet at the same time, because devel-
opment does not happen in lock-step fashion, target items introduced,
or seeded, in one lesson will be included again, or recycled, in later
lessons for reiteration and reinforcement. Additionally, a principled
TBLT syllabus will recognise that language develops differently across
linguistic subsystems. Though this will affect syllabus design and task-
ordering, this point will inform decision-making by the class teacher as
well. For example, since word order facts seem to be acquired
implicitly, there is little point in correcting word order errors at early
stages of development. By contrast, the meaning and use of words, for
example, can and should be taught through explicit instruction. These
are prime candidates, then, for pre-task teaching.
At the same time, we know that a fair amount of SLA happens
incidentally. In TBLT learners are exposed to authentic input contain-
ing plenty of examples of linguistic points that push the learners’
development. Since the prime emphasis is on meaning, the particular
language points are often not addressed explicitly until post-task
activities, once the task is finished. In other words, TBLT lends itself
to a Focus on Form approach as well. This makes good sense,
considering the observation that there are limits to the effects of
frequency in the input. In TBLT, teachers expose the learners to
generous amounts of rich input, followed by explicit instruction.
Whether linguistic forms are also introduced in pre-task activities,
and the way any such forms might be introduced, will vary, depending
on the needs of the students and the nature of the ‘forms’ in question.
One danger inherent to the pre-task, task, post-task structure of a
TBLT lesson is that the first and third stages can very easily turn into
old-fashioned traditional teaching. Yet, if the true aim of the lesson is
the successful completion of task itself, then the pre- and post-task
activities will be subordinate to that task. And learners will see the
importance and value of them as a necessary component for complet-
ing the task.
 Approaches to English Language Teaching 137

Of course, by definition, the task itself must be meaningful and

engaging. This core feature of TBLT recognises that exposure to input
is necessary for SLA and that there are limits on explicit instruction.
The aim of TBLT is to promote interaction and facilitate active
engagement with language. But the hard work on the part of the
learners does not mean that teachers can take a break and relax during
the task. Instead, the teacher must remain engaged, continuously
monitoring the learners and noting difficulties as they complete the
task. During this phase a teacher must decide how to address language
problems or errors that arise. There are times when it may be
appropriate to give individual learners specific feedback or correction
in real time, as they engage in the task. But if such error correction is
likely to have a detrimental effect on the communicative act, then it
may be more appropriate to note errors for later discussion. Waiting
to address common errors being made by a number of students instead
of pinpointing particular students will be less likely to isolate and
perhaps alienate individual learners as well. This kind of feedback or
error correction can form part of the post-task lesson.
In summary, TBLT follows a three-step structure of pre-task, task
and post-task. Adhering to current understandings of both language
and language development, within TBLT the full complexity of
language can be addressed. In preparing students to do the task,
particular language points can be identified and highlighted by the
teacher, whether vocabulary, grammar, pronunciation, or pragmatic
or cultural points. During the task, the teacher manages the interac-
tion among students, pushing learners to challenge themselves and
intervening when communication breakdown cannot be resolved by
the learners. The post-task phase will involve some form of reporting
on or responding to the completed task by the teacher. In doing so, a
skilled teacher will highlight the language points s/he has identified for
this task and reinforce new vocabulary as well as relevant cultural
points.
The importance of communication is clearly central to Task Based
Learning. Despite this, however, there have been critics who caution
against an over-emphasis on production. Michael Lewis (1993, 2000)
and proponents of the Lexical Approach point out that learners need
to do more than practise using their current stores of knowledge; they
need constantly to expand that store in order to develop. Specifically,
teachers should teach words, but not as vocabulary items in isolation.
Instead, all words should be taught in terms of how they are used and,
crucially, in the context of the other words with which they tend to
138 Language Teaching

appear. This approach echoes the cognitive emphasis on semantic

networks. Words and phrases pattern together as collocations and
colligations. Therefore, words should be taught as patterns, from two-
word collocations like inspire confidence to clausal colligations like
laugh all the way to the bank. For the Lexical Approach, the store of
knowledge which learners need to increase is their store of colloca-
tions/colligations.
In this approach words and the patterns that they form are the basic
units of language. Strong proponents, such as Lewis, take the extreme
position that language is exclusively made up of sets of words and
there are no grammar ‘rules’ or constraints. Language is a collection of
words which language users tend to express following the patterns
they hear. For example, though the sentences in (1) are ‘grammati-
cally’ correct in English, in fact chances are that a native speaker
would actually use the variants in (2).

1. a. I comprehensively agree.
b. I wonder what his different motive is.
c. The gangster shot the man repeatedly until he died.
d. That was decidedly five years ago.
e. I don’t agree, but that’s next to the point.
2. a. I totally/completely agree.
b. I wonder what his ulterior motive is.
c. The gangster shot the man dead.
d. That was a good five years ago.
e. I don’t agree, but that’s beside the point.

As illustrated here, collocations occur at the level of individual words

(2a, b), or at the level of the clause (2c). Moreover, they can be fixed
phrases (2d, e) which have no grammatically equivalent alternative, or
they can have a degree of flexibility as in (2a).
The Lexical Approach is supported by corpus linguistics with
enormous databases that allow linguists to study examples of ‘real’
language actually produced by speakers instead of an idealised as-
sumption of what constitutes language.20 Corpus linguists argue that
collocations form the basis of a much more significant percentage of
language that speakers actually use than perhaps has been acknowl-
edged by mainstream linguists of the modern period. The Lexical
Approach also finds support from acquisition research on the role of
formulaic language in language acquisition, which shows that learners
take in and produce unanalysed chunks for a communicative effect,
 Approaches to English Language Teaching 139

without necessarily knowing each item in the ‘chunk’.21 This happens

in beginning English classes when learners use phrases such as I’m a
student without realising that the sentence includes four distinct
words.
In many ways, TBLT and the Lexical Approach make a natural
combination. The Lexical Approach provides lists of phrases for the
teacher to teach. This can be done through both explicit instruction
and implicit exposure, and can then be practised by learners in a task-
based communicative way. Because knowledge of words lies at the
heart of being able to communicate, strong proponents of the Lexical
Approach will be content to teach collocations without teaching
‘grammar’. Yet, as the discussions of language in this book indicate,
most linguists agree that grammar is a real feature of language which
must also be taken into account in language teaching, and as a result
teachers are urged to take a Focus on Form approach within the
broader context of CLT. In the next chapter, we will bring together all
of the strands of research discussed in the first half of this book,
together with the range of approaches to language teaching surveyed
in this chapter to construct a language lesson for illustrative purposes,
which we will dissect with questions of theory in mind.

FOR DISCUSSION

1. Consider each of the four main views of language – generative,

cognitive, functionalist and sociocultural – in terms of approach to
language teaching, identifying methods that would be appropriate
or compatible with each approach.
2. Make a list of all the approaches and methods discussed in this
chapter. Then organise them in terms of those which do and those
which do not have theoretical foundations in linguistics. Discuss
the extent to which a theoretical foundation in linguistics is
important for language teaching.
3. Consider each of the ten observations about second language
development discussed in Chapter 5. For each one, go through
the section on Communicative Language Teaching, looking for
connections between theory and practice.
4. Choose any language teaching textbook that you have access to
(for any language). Can you identify any elements of the ap-
proaches and methods identified in this chapter? Where do you
see different aspects of communicative language teaching in the
textbook?
140 Language Teaching

5. Are some of these approaches and methods more suitable for

different kinds of learners? Discuss the options in terms of differ-
ences in age, cultural context, type of institution and proficiency.
6. Are some of these approaches and methods more suitable for some
languages than others? Discuss the reasons why or why not.

NOTES

1. In defining these terms this way, I follow the tradition in applied

linguistics that is attributed to Edward Anthony in 1963 and continues
today in widely used textbooks such as Richards and Rodgers (2001).
2. Krashen developed his Monitor Model over a series of publications. See
Krashen (1977, 1981, 1982, 1985).
3. See Chapter 3 for discussion of these studies.
4. Richards and Rodgers (2001) use the term ‘alternative’ for approaches
that do not have a basis in linguistic theory. While a number of these
‘designer’ approaches are alternative in Richards and Rodgers’s sense,
others are not. Instead, the term ‘alternative’ is used here to contrast with
current mainstream approaches in language teaching. See Richards and
Rodgers (2001) for more detailed descriptions of: Total Physical Re-
sponse, The Silent Way, Community Language Learning and Sugges-
topedia.
5. For more information on the Silent Way, see Gattegno (1972, 1976). For
more information on the notion of learning through discovery, see the
work of educationalist Jerome Bruner; for example, Bruner (1961,
1967).
6. See Winitz (1981) for more information on the Learnables.
7. For a teacher’s guide to TPR, see Asher (1977). For more on the premises
of TPR, see Asher (1966).
8. ‘Suggestology’ is outlined in Lozanov (1978). See also Scovel (1979).
9. See Curran (1976).
10. For more information about John Rassias and the Rassias Method, see
the edited volume by Yoken (2007).
11. See Chapter 17 of Richards and Rodgers (2001) for a clear overview of
content-based instruction and immersion programmes, and references
for further reading.
12. For early work in ESP, see Hutchinson and Waters (1987). For current
work, consult the regularly issued journal, English for Specific Purposes,
formerly known as The ESP Journal.
13. The notion of progressive education is generally associated with philo-
sopher and educational reformer, John Dewey. See Dewey (1938, 1944).
For a discussion of progressivism in the current context, see Hayes
(2006).
 Approaches to English Language Teaching 141

14. Needs Analysis as a formal tool is generally associated with English for
Specific Purposes. See, for example, Hutchinson and Waters (1987).
15. For a historical overview of the question of identity in language learning,
see Block (2007). For a study of immigrant language learners and issues
of identity, see Norton (2000).
16. For a useful discussion and review of literature related to learner
autonomy, see Benson (2001). See also Dam (1995) and the volume
edited by Little, Ridley and Ushioda (2003).
17. Sfard (1998) has usefully presented this distinction between so-called
Participation and Acquisition Metaphors for learning. While she writes
in the context of teaching mathematics, the distinction has been readily
taken up for language, as teachers are encouraged to see learners doing
language instead of having language.
18. For seminal work on humanism and language teaching, see Stevick
(1990).
19. Early well-known examples of TBLT were the Bangalore Project (Prabhu
1987, Beretta 1990) and the Malaysian English Language Syllabus
(Kementerian Pelajaran Malaysia 1975). Much has been written about
Task-Based Learning. For an early discussion of the principles under-
lying a TBLT syllabus, see Long and Crookes (1992). For more recent
work on TBLT, see Willis and Willis (1996, 2007), Ellis (2003), Nunan
(2004), and the edited volume by Van den Branden (2006).
20. For examples of work in Corpus Linguistics see Biber, Conrad and
Reppen (1998), and the edited volume by Cowie (1998).
21. See the work of Nattinger and DeCarrico (1992); Myles, Hooper and
Mitchell (1998); and Wray (2002). For a review of literature on for-
mulaic language, see Weinert (1995).
 7
PUTTING THEORY INTO PRACTICE

INTRODUCTION AND CONTEXT

This book began with abstract concepts based on linguistic theory,

before moving to questions of relevance to language teaching. Though
we have surveyed a range of linguistic approaches to language, we
have emphasised mentalist points of view, looking for areas in which
there is general agreement, and giving most of our attention to the
most formal of the linguistic approaches – the Generative Approach.
The ambitious aim of this book is to show the research into abstract
questions of language and the cognitive processes underlying language
development are relevant to the language teacher.1
These last two chapters move into the classroom itself. In this
chapter, we will consider an actual lesson plan which makes use of
current approaches to language teaching, putting into practice the
methods endorsed in the last chapter. We will then discuss the salient
features of the lesson plan, making reference to the themes that have
emerged from the discussions of linguistic theory and language
development. In order to create a lesson, however, we must first
specify the profile and context of the class. For the purposes of our
discussion, we will try to define a ‘typical’ language-teaching context
in which to situate our lesson plan. In the next and final chapter of this
book we will then modify the context in order to highlight some
specific concerns that are particular to different groups of learners and
that draw in particular areas of research.
Of course, there is no such thing as a ‘typical’ language-teaching
context. Every context has its particulars in terms of aims, funding,
personalities, culture and political climate, to name a few. Indeed,
every class is unique. For the purposes of illustration, however, we
need to specify context for our lesson. Accordingly, I propose an
English as a Foreign Language (EFL) context because it has the
broadest appeal. The theoretical points made in this book are
relevant for the teaching of any language. It seems safe to assume
 Putting Theory into Practice 143

that, globally, most language teaching takes place in a foreign

context, not a second language context. For this reason, the lesson
discussed in this chapter is intended to be taught in a country where
English is not the dominant or official language, in contrast with an
English as a Second Language (ESL) context in which the target
language is also an official language of the country where it is being
taught.
Again, appealing to the broadest context possible, our EFL class is
envisioned as a core language class of highest/oldest level in a state-
funded secondary school, with a class size of thirty-six, all approxi-
mately aged seventeen and evenly split between girls and boys.
Assuming a national curriculum in which English is introduced at
least in a limited way at primary levels, these students are assumed to
be at a ‘lower intermediate’ level of proficiency. In order to specify
what is meant by lower intermediate, I will refer to the Council of
Europe’s Common European Framework of Reference for Lan-
guages (CEFR), as it provides a useful articulation of language levels
with a degree of detail. The learners in our context are at the lower
end of the CEFR’s ‘Independent User’ category, or B1. Quoting the
CEFR document, this means that their global ability in English is as
follows:

Can understand the main points of clear standard input on

familiar matters regularly encountered in work, school, leisure,
etc. Can deal with most situations likely to arise whilst travelling in
an area where the language is spoken. Can produce simple con-
nected text on topics, which are familiar, or of personal interest.
Can describe experiences and events, dreams, hopes & ambitions
and briefly give reasons and explanations for opinions and plans.
(CEFR, p. 5)

In order to provide a bit more detail, Table 1, again quoting the CEFR
(p. 6), shows the Council of Europe’s ‘self-assessment grid’ for a B1-
level learner in terms of Reception, Interaction and Production.
In terms of linguistic ability, the CEFR describes a B1-level learner
as having ‘enough language to get by, with sufficient vocabulary to
express him/herself with some hesitation and circumlocutions on
topics such as family, hobbies and interests, work, travel, and current
events, but lexical limitations cause repetition and even difficulty with
formulation at times’ (p. 27).
A summary of the context is given below.
144 Language Teaching

Summary of class profile and context

. National context: English as a Foreign Language (EFL)
. Type of institution: State-funded secondary school
. Number of students: 36
. Gender: evenly mixed
. Approximate age: 17 years old
. Proficiency level: Lower intermediate/CEFR B1
. Class time: 50 minutes daily

Reception Interaction Production

Listening Reading Spoken Written Spoken Written

interaction interaction production production

I can understand I can I can deal with I can write I can connect I can write
the main points understand most situations personal letters phrases in a straightforward
of clear standard texts that likely to arise describing simple way in connected text
speech on consist mainly whilst travelling experiences and order to on topics which
familiar matters of high- in an area where impressions. describe are familiar, or
regularly frequency the language is experiences and of personal
encountered in everyday or job- spoken. I can events, my interest.
work, school, related enter dreams, hopes
leisure, etc. I can language. I can unprepared into & ambitions. I
understand the understand the conversation on can briefly give
main point of description of topics that are reasons and
many radio or events, feelings familiar, of explanations for
TV programmes and wishes in personal opinions and
on current personal letters. interest or plans. I can
affairs or topics pertinent to narrate a story
of personal or everyday life or relate the plot
professional (e.g. family, of a book or
interest when the hobbies, work, film and
delivery is travel and describe my
relatively slow current events). reactions.
and clear.

Table 7.1 CEFR Self-assessment grid for B1 level proficiency

With the class profile and context now defined, we need to provide
some context for the lesson plan that follows because no lesson occurs
in isolation, but instead follows the lesson that came before, and
points to the lessons to follow. In any pedagogically sound context, it
will be situated within a syllabus following a curriculum. Presenting a
curriculum and detailing an entire syllabus would take us far beyond
the scope of this book. However, we need a degree of context for our
 Putting Theory into Practice 145

lesson plan. Thus in Table 7.2 we have included a rough sketch of a

syllabus, also including a very general idea of linguistic aims, for
illustration.

Unit Topic Target competence Grammar Pronunciation Vocabulary

Identity Fluency Simple vs complex Consonant Connecting words/

sentence structure contrasts phrases

Interviews Strategic Questions; Intonation Phrases for

auxiliary verbs clarification

Newspapers Discourse (written) Sentence structure: Consonant clusters Phrases for

punctuation description

Meeting new Pragmatic Verb tense Long and short Phrases for starting,
people vowels ending and
changing
conversations

Travelling abroad Strategic Negation Word stress Synonymous and

antonymous
phrases

Tourist attractions Discourse (spoken) Complex noun Phrasal stress Phrases for location
phrases

Animal rights Pragmatic Modal verbs Using stress for Phrases for
emphasis interrupting

English in the world Fluency Demonstratives Sentence stress Phrases for

categorising

Table 7.2 Overview of syllabus

The fundamental organising principle of the syllabus sketch in

Table 7.2 is form and function. With function in mind, there are
eight topical units, and in terms of form, specific points of linguistic
competence are listed, including grammar, pronunciation and voca-
bulary. As shown in the second column, each unit emphasises one
other aspect of communicative competence beyond linguistic compe-
tence (as discussed in Chapter 4). Attempts have been made to ensure
the linguistic points are related to the topic and the chosen commu-
nicative competence.
146 Language Teaching

Consider the first unit of the syllabus. Because it is the first lesson, it
is based on the theme of identity, allowing the class become familiar
with each other. Linguistically, the focus is on basic sentence structure,
with an emphasis on connecting phrases to show how simple sentences
might be combined to make complex sentences. The expectation is
that at this level, students should be able to create grammatically
accurate sentences to describe themselves. However, in the interests of
promoting communicative competence in general, the lessons at the
start of this unit are designed to promote fluency in an attempt to set
an early tone of active production in the classroom. To facilitate
production, the associated pronunciation focus is also very basic.
Though the focus is listed as ‘consonant contrasts’, the assumption is
that the teacher would focus specifically on consonants of particular
difficulty to the students in his/her particular class, anticipating native
language transfer. In sum, this introductory unit is designed to build
confidence and promote communicative competence, as a starting
point for the course to come.
Within each unit there are specific lessons with more specific aims.
The lessons included in the sixth unit from the above overview are
given in Table 7.3, in order to provide more context for the single
lesson plan that will be considered in detail in this chapter.

Unit Lesson topic Language focus Example Main task

Tourist attractions: Animals and zoos New vocabulary; the giraffe eating Catalogue of zoo
the zoo complex noun leaves animals
phrases

Animal habits Habitual usually sleeps, is Showcase an

expressions and sleeping now animal
expressions for
current activity

The zoo: a map Expressions for near here, far away Design a layout for
location; phrasal the zoo
stress

Animals at risk Simple imperative visit our zoo, learn Make a brochure
phrases about animals to promote
awareness

Table 7.3 Example of lessons within one unit

 Putting Theory into Practice 147

The theme of this unit is ‘Tourist attractions’, with particular focus

on ‘The zoo’. Learners at this level should already know general
vocabulary for animals to be found in a zoo, such as tigers and bears.
The linguistic focus in the first lesson is complex noun phrases,
starting with pre-modified phrases such as the hungry tiger and
introducing post-modification, such as the tiger sitting under the tree.
The task is for the class to put together a catalogue of zoo animals,
with basic descriptions. This forms the basis of the second lesson, in
which learners are asked to choose an animal to showcase in a bit
more depth.
Building on post-modified noun phrases, the second lesson of this
unit reinforces the differences between habitual verbal expressions like
usually eats leaves and progressive verbal expressions like is eating a
leaf. Throughout this unit, the attention to pronunciation goes beyond
the segment level to focus on phrasal level stress. The emphasis on
complex noun phrases and verbal clusters allows for this quite
naturally. Because the communicative competence for this unit is
spoken discourse, attention is paid to the role that word stress within
the phrase plays in conveying emphasis in spoken interaction. To get a
real sense of lessons sketched, much more would need to be said. The
aim here, however, is to provide a backdrop for the lesson to be
considered in the next section. Thus, we will cut short this description
of the first two lessons of this unit and move on to discuss the third
lesson in more detail.

A POST-METHODS LESSON PLAN

This lesson is task-based with pre- and post-task activities designed to

draw attention to the language-particular aims of the lesson. While the
communicative competence theme for this unit in general is spoken
discourse, the particular aims of this lesson are informal phrases for
seeking clarification from peers and the role of word stress for
clarification in spoken interaction. The lexical expressions highlighted
in this lesson are adverbial phrases. This makes a natural next step
after discussions of complex noun phrases and present simple vs
present progressive in the previous two lessons. The main task of
the lesson is to design a layout for a zoo. This makes use of the output
from preceding lessons, the catalogue of zoo animals and the indivi-
dual animal showcases, and offers an opportunity to reinforce recently
learned vocabulary.
The lesson plan itself follows a simple standard format appropriate
148 Language Teaching

for most teaching contexts. The procedures are listed with the inter-
action type and timing specified for each step of the lesson. Also listed
are the specific aims of this lesson, the task itself, the materials that will
be needed during the lesson, and the aspects of language that will be
attended to by the teacher for corrective feedback.

LESSON PLAN

Lesson 3: A map
Unit 6: The zoo

Aims
Linguistic aims
Lexical
. Students will be able to use location adverbs correctly
e.g. near, here, away from, separate, next to, etc.
Pronunciation
. Students will be able to use phrasal stress patterns
e.g. near the entrance vs near the entrance

Spoken discourse aims

. Students will be able to use stress patterns for communicative
purposes
e.g. near the entrance vs near the entrance
. Students will be able to ask for clarification during peer inter-
action
e.g. what did you say? huh?

Task aim
. Design a layout for a zoo

Areas for corrective feedback

. Use of location adverbials
. Use of vocabulary taught today and in recent lessons
. Use of phrasal stress for natural stress patterns and for emphasis
. Informal register, appropriate for peer interaction

Materials
. Cards or slips of paper each with a zoo animal written on it
. Large pieces of paper
. Glue or tape
 Putting Theory into Practice 149

Procedure Interaction type Timing

Before the lesson

Put location phrases on the board with simple
symbols for illustration.

e.g. the O is near the X: O X

far away from: O X
together: OX
separated from each other: O YZ X
etc.

Pre-task
Animal placement game
In pairs, each with set of zoo animals cards Teacher 1 Class 5 minutes
Listen to the instructions and place each animal Listening
accordingly. comprehension, with
(e.g. Put the tiger in the middle of your desk. peer support
Put the zebra near the tiger.
Put the elephant far away from the zebra,
etc.)

Without making any changes, discuss the layout Student 1 Student 5 minutes
of animals with your partner. Would it be a (Pair discussion)
suitable layout for a zoo? Why or why not?

Focus on meaning
Summarise the most useful points overheard Teacher 3 minutes
from group discussions (e.g. outdoor vs indoor
animals; animals needing big cages; group vs
solitary animals, etc.)

Focus on form
If you have a disagreement in a discussion, what Teacher 1 Class 3 minutes
expressions do you use?
Elicit: I disagree. How about . . . ., etc.
There are other methods for making a point.
Model: How about next to HERE?
How about BESIDE the exit?
150 Language Teaching
Task 1
Instructions
Today’s task is to design a layout for a zoo. Use Student 1 Student 10 minutes
the arrangement of the animals on your desk as (Pair interaction)
a starting point. As you discuss the best place to
put each animal, pay attention to your use of
word stress.
[Distribute large blank piece of paper to each
pair.]

Post-task
Focus on form
Praise the students for attending to stress. Teacher 1 Class 3 minutes
Ask: What’s the difference?
near the ENTRANCE vs NEAR the
entrance
Highlight the role of stress for clarification.

What do you say if you don’t understand your Teacher 1 Class 4 minutes
classmate?
Expect: Excuse me, could you repeat . . .?
What do we actually say to friends?
Model: What did you say? Huh?
Reinforce: Formal vs informal spoken
discourse

Task 2
[Form groups of four, combining two pairs.]
Instructions
Consider the two layouts. Put them together to Student 1 Student 15 minutes
make one agreed layout. Glue the cards to the (Group interaction)
paper and draw paths, cages, fences, etc.
As you work together, consider the strategies you use
for clarification: choice of phrase and phrasal stress
[Hang the maps on the wall for tomorrow’s
lesson.]

Focus on form
Highlight good examples of clarification during Teacher 2 minutes
group interaction.
Tomorrow we are going to bring together the
animal showcases and maps to create a brochure
for our zoo which highlights animals at risk.

Homework
Read anything of your choosing for 30 minutes. If Student
you do not know what to read, search online for
websites relevant to zoos and about zoo animals.
 Putting Theory into Practice 151

DISCUSSION OF THE POST-METHODS LESSON

The lesson has been labelled a ‘Post-Methods’ lesson because it is,

broadly, communicative in its approach, but also emphasises aspects
that are associated with more traditional language teaching. And, as
will be discussed in this section, it tries to take into account a range of
principles from linguistic theory and language development. If we
were to consider all of the lessons in the course or even a whole unit,
we would see greater variation in method, as different approaches and
methods are appropriate for different aims from day to day. In other
words, instead of adhering to one single method, we would take an
eclectic Post-Methods approach, selecting from the range of options in
order to best suit the needs of the learners. This particular lesson
makes use of the usefully flexible Task-Based Approach, which allows
for Focus on Form and works in elements of a Lexical Approach. This
combination allows us to translate into classroom teaching our
theoretical understanding of what language is and how language
develops.

Form and Function

The most basic principle underlying the above lesson is a balance
between form and function. Contrary to a ‘meaning only’ view,
language forms are taught, but in such a way that they are grounded
in meaning. So, for example, the teaching of adverbial phrases in the
pre-task game requires that students understand the meaning of the
phrases. This kind of meaning-based focus on form occurs throughout
the lesson. As another example, phrasal stress is a feature of pro-
nunciation – in other words, form – but it is taught in terms of the
meaning that it conveys. These points of language form are all situated
in the larger class aim of designing a layout for a zoo. True to the
definition of a task, this aim is not actually linguistic; however, it is
designed to require the use of language and, crucially, to require
interaction among the learners. And it is structured in such a way that
the particular features of language that have been identified for the
day’s lesson can be taught. As such, this is not strictly Focus on Form
as defined by Long because the focus does not arise incidentally.
Instead, the lesson plan is designed to force the pre-identified points to
arise – so that the teacher can teach particular forms in this lesson.
One main reason to endorse a pre-determined, or deliberate, Focus
on Form approach has nothing to do with abstract theory and
152 Language Teaching

everything to do with real-world realities. In many language-teaching

contexts, there is real pressure to teach specific forms. This often
shows itself in the syllabus, or in the course materials that teachers are
expected to use. Even in places where there is a degree of freedom for
teachers to choose their own classroom materials, the formal assess-
ments for very many courses are, in fact, very ‘form-centric’. A final,
related reason is to do with current language teaching practices to be
found in much of the world’s foreign language-teaching contexts. In
many language classrooms around the world, the norm is some variety
of the traditional Focus on FormS approach. Moving to deliberate
Focus on Form may be a more achievable and appropriate develop-
ment than incidental Focus on Form. This contrasts with those second
language-teaching classrooms which have embraced a strong Focus on
Meaning approach. If these find that they need to shift back to include
some teaching of forms, an incidental Focus on Form may be a more
natural progression.
But deliberately focusing on form is not endorsed only for practical
reasons. Both psycholinguistic research and interaction research tell us
that intelligent and motivated learners ‘notice’ patterns and rules. And
our theoretical model of language development holds a place for
explicit, conscious learning alongside implicit acquisition. Like most
processing models, MOGUL tells us that repeated use of forms will
raise activation levels leading to language development. Of course,
frequency for its own sake, or rote memorisation, is not what we have
in mind, but instead repeated use in meaningful contexts. This lesson
requires that learners make use of the adverbial forms throughout the
lesson, repeating the forms for different (meaningful) purposes.
At the same time, the lesson is ‘function’-oriented. For example,
when attention is drawn to the phrases students use during discussions
within their groups, emphasis is put on using the appropriate register
for peer interaction. This is needed to contrast with the formal phrases
that are often to be found in published language course books. One
challenge facing language textbook writers is the fact that textbooks
are, by definition, a written medium. It is natural for a textbook writer
to include relatively formal phrases because the medium of a textbook
is somewhat formal. Even if a textbook writer presents informal
registers, the result is often somewhat problematic. When ‘real’
language is captured in a clear enough way to include it in a textbook,
it can often seem quite stilted and unnatural due to the mismatch
between the spoken and written medium. Yet real language use
requires the ability to speak from a range of levels from formal to
 Putting Theory into Practice 153

informal. A lesson like this one, with a focus on spoken discourse,

requires real-life spoken interaction among peers. Tasks like those in
this lesson provide an excellent opportunity to teach features of
spoken discourse such as the informal expressions that we actually
use with peers when seeking clarification.
At a more general level, on the function side of the form–function
equation is the topic-based underlying organisation of this lesson, and
all of the lessons. This lesson fits into a larger unit about zoos and
animals, a topic which is largely neutral in terms of cross-cultural
appropriateness. It also sets the scene for the less neutral unit to come,
a unit based on animal rights. While animal rights may not be a
concern to some people in some cultures, it is very much a concern
among most English-speaking cultures. Keeping in mind that language
cannot be divorced from culture and society, the next unit explicitly
raises a culturally grounded issue. The way that a teacher approaches
our current unit, ‘The zoo’, could also touch on cultural values to do
with the treatment of animals, but it does not have to. In good Post-
Methods spirit, some lessons will include more explicit reference to
cultural difference than others. In the next section we will deconstruct
our lesson further to make more connections with the more abstract
theoretical points which we encountered earlier in this book.

Providing Input and Fostering Output

Throughout this book we have focused on points of consensus
emerging from subfields of linguistics and applied linguistics which,
in the world of academic debate, are very much opposed to each other.
By understanding the conflicting views, we as teachers can gain a more
sophisticated understanding of our subject matter: language. Yet at
the same time, we need to step away from the debate and determine
where there is consensus and where there are points that can inform
the practical concerns of language teaching. In this book, the emphasis
has been primarily on mentalist approaches to language, exploring
formal, abstract approaches in particular. In this section we will use
the above lesson plan to show how formal views of language can
emerge in an activity as practical as a language lesson for a class of
intermediate learners. In doing so, we will highlight some of the
observations about second language development that we explored
in Chapter 5, showing how these can provide a principled foundation
for decision-making in preparation for the lesson and during the
lesson itself.
154 Language Teaching

In an attempt to ground ourselves in a single framework which

unifies a range of approaches, we concluded our survey of approaches
to language in Chapter 4 with a description of MOGUL, Modular On-
line Growth and Use of Language. In concert with most mentalist
approaches to second language development, MOGUL considers
input as central to the process. The centrality of input is also why
it is the first of the ten observations originally given by VanPatten and
Williams (2007): Exposure to input is necessary. The unspoken
assumption in the above lesson plan is that the entire lesson is
conducted in the target language. Arguably, target language-only
instruction is possible, even from the very beginning of language
learning. At the intermediate level of our hypothetical learners, there
is no reason why the lesson cannot be done completely in English. This
maximises the amount of input that learners receive. That way, a large
amount of the input that comes in the lesson – from instructions on the
tasks to discussions of animals and zoos – is authentic, natural input.
In order to try to maximise input, the lesson also includes home-
work, requiring reading for pleasure. Emphasising personal reading
comes from research showing that silent reading is a valuable source
of natural input for learners. It will also foster learner autonomy, as
students are encouraged to choose their own reading. The importance
of natural input is also a direct implication of the second observation:
A good deal of SLA happens incidentally. We know from research
coming out of both generative and cognitive second language research
that learners come to know features of language that have not been
explicitly taught. So it only makes sense to facilitate this feature of
language development by maximising the students’ learning potential
through maximal amounts of input, both inside and outside the
classroom.
In addition to natural linguistic input, this lesson includes explicit
metalinguistic input. By MOGUL there is natural development of
linguistic knowledge as a result of natural, authentic input and there is
the equally natural development of metalinguistic knowledge through
explicit input. For MOGUL, both types of knowledge are ‘natural’ in
the sense that they are both features of the human mental system,
depending, as they do, on internal, cognitive mechanisms in the brain.
This extends a long-held conviction of linguists working in a gen-
erative framework: the language ‘module’ is associated with implicit
linguistic knowledge while the mental component that includes ‘gen-
eral intelligence’ is associated with metalinguistic knowledge, or
knowledge about language. Importantly, however, both are used
 Putting Theory into Practice 155

by speakers of language (whether native or non-native) when they

comprehend and produce language. And the degree of development of
both will contribute to the overall level of proficiency and fluency in a
language. The dividing line and the exact relationship between the two
is a matter of debate. But the existence of two types of knowledge is
widely accepted, even among other mentalist views of language.
Cognitive linguists acknowledge a distinction between deliberate,
conscious, declarative knowledge and automatic, subconscious, pro-
cedural knowledge. And though strict proponents of theory will
disagree, this implicit, automatic knowledge can be likened to the
notion of ‘internalised’ knowledge found within Sociocultural Ap-
proaches.
In sum, all approaches agree that there is an extent to which learners
can come to know more than they have been made explicitly aware of
in their input, which is the third observation: Learners come to know
more than what they have been exposed to in the input. Accordingly,
in our lesson plan there is a healthy balance between natural language
input and explicit language instruction. This facilitates both natural
acquisition and controlled learning. As widely acknowledged, the
success of learning is often tied up in extra-linguistic factors such
as intelligence, diligence, interest and motivation. This is not surpris-
ing if an element of second language development implicates inten-
tional, deliberate learning. Simply put, those who try harder are likely
to achieve more. Yet, these affective factors could also indirectly affect
the natural acquisition potential as well. Learners who allow them-
selves to be exposed to more (or less) input, depending on levels of
interest and diligence, will affect the extent to which incidental
acquisition is even possible. By embedding classroom activity in tasks
that require the involvement of individual learners and the interaction
of learners among themselves, it is hoped that interest will be piqued
and motivation will be enhanced, leading to increased levels of all
types of input.
A further point crucially related to input is the role of output, as
captured in Swain’s Output Hypothesis and included in Long’s
Interaction Hypothesis. Output is important as one part of the
processes involved in successful and meaningful language production.
The research shows us that this is particularly true if learners are
pushed to produce language which is challenging to their current level
of proficiency. Accordingly, this lesson plan revolves around tasks
that not only require the learner to produce language in group
discussion, but which also require the specific production of pre-
156 Language Teaching

determined aspects of language, challenging the learners to use the

new adverbial forms. Moreover, focus on these linguistic aims comes
in both in the group discussions and in the ultimate creation of the
layout of the zoo.
However, as captured in Observation 10, there are limits on the
effects of output (learner production) on language acquisition. In
addition to input and output, the Interaction Hypothesis includes
the notion of negotiation of meaning. When learners encounter
difficulty in communication, they will have to attend to the gap in
knowledge that caused the breakdown in order to repair it. This
process facilitates learning, as it forces attention to language and
pushes development as the learner strives to correct the problem. In
our lesson plan, the second task includes a focus on phrases for
clarification during group discussions in order to encourage the
learners to seek repair where there is misunderstanding or lack of
clarity. The intention is that focus on phrases for clarification will
foster negotiation of meaning, which will, in turn, promote language
development.
In addition to type of input, we have discussed the question of
amount of input, concluding that the more, the better. While we have
suggested that a large amount of input is important in general, we are
also concerned with amount of input in terms of specific linguistic
forms or constructions as well. Research on the importance of
frequency in the input refers to the repetition of specific forms in
order to foster learning. As we noted in Chapter 5, most of the
research on frequency has been limited to the level of vocabulary.
It is this research that underlies the so-called seeding and recycling of
adverbial phrases throughout this lesson. Adverbial phrases are taught
in the opening animal placement game and then recycled throughout
the lesson. In addition to repeated frequency, the lesson attempts to
increase the salience of the target forms in several ways. Firstly,
learners are required to comprehend the new forms in the opening
activity, before then having to produce them in discussion. Secondly,
explicit attention is brought to the adverbial forms again by the
teacher, but this time with a focus on the stress patterns attached
to the phrases. This repeated and varied attention to the linguistic
form of the lesson is designed to increase both saliency and frequency.
The teaching of these forms also takes a lexical approach, teaching
phrases instead of lists of words. This lesson could have started by
teaching the words near, next, separate, away, and so on. But instead,
the prepositions are taught as collocational phrases. After all, the
 Putting Theory into Practice 157

locational use of next always collocates with the particle to. So next to
should be taught as a fixed phrase from the beginning. As an aside, it
would be appropriate to follow the teaching of next to with the
teaching of the temporal use of the word next, as in: Next, I will read
you a story. This is based on research that shows learners are better
able to learn abstract extensions of word meaning if they have been
taught the literal, physical meaning first.2 Importantly, the phrases are
taught in context and in association. Because psycholinguistic research
supports the view that lexical knowledge is stored in networks
connected by association of meaning, all new vocabulary items should
be taught in association with related items.
Also drawing from the cognitive view of knowledge is the way that
the phrases in this lesson are taught. When teaching locational mean-
ing, it is natural to illustrate location visually as we do in this lesson.
From a cognitive point of view, doing so appeals to two modes of
processing, the visual mode and the aural mode. This multi-modal
approach may in and of itself lead to more successful learning.
Whether it does or not, it certainly serves to increase the salience
of the forms in question. Moreover, asking the students to manipulate
the objects to reinforce the meanings of these phrases further is an
application of the idea that there is a positive connection between
kinetics, or physical movement, and learning.
Returning to our discussion of frequency, being limited to just one
lesson gives us only a snapshot of teaching in this classroom. Yet the
brief outline of the unit in which this lesson occurs, along with the
sketch of all of the units preceding and following this one, shows
continuous seeding and recycling that will be inherent to principled
language teaching. Moreover, there is a full range of seeding and
recycling to include reinforcement of language at the lexical level, at
the grammatical level, for pronunciation, and for areas of language
use as well.

Accounting for Interlanguage Development

Another general piece of the puzzle, as important for teaching as
input, is the nature of second language development. Three of the
observations by VanPatten and Williams (2007) address this specifi-
cally. First is the generalisation that language development is systema-
tic. This is the basic insight underlying the fourth observation:
Learners’ output (speech) often follows predictable paths with pre-
dictable stages in the acquisition of a given structure. While this is not
158 Language Teaching

readily seen in our single lesson plan, the sketch of the syllabus shows
language focus with increasing complexity as the course progresses.
We can see this, for example, in the focus on pronunciation which
starts at the level of individual sounds in Unit 1: single consonants and
then consonant clusters, stress within a word and then stress at the
level of the phrase, and so on. In terms of grammatical forms, the
emphasis is on noun phrases and then more complex noun phrases.
Then there is (recycled) emphasis on simple verb phrases, before the
focus turns to adverbial phrases in our zoo lesson. Of course, none of
these grammatical forms is likely to be completely new to all students.
They will have encountered these in their natural input in past lessons
and explicitly at more junior levels of instruction.
Yet predictable paths and stages can help us with decision-making
for language teaching in different ways. If you are a syllabus designer,
an understanding of these stages of development is crucial. Yet even
the classroom teacher following a set syllabus would do well to be
aware of paths of language development in order to make decisions
about everyday questions including what to emphasise within the
lesson, what to teach explicitly, what to recycle and what requires
corrective feedback. This last point is the area over which teachers in
any teaching context have the most control and that they should
therefore think about carefully. It is a fact of life that there is a limit to
what one can do in a single lesson on a single day. Teachers have
constantly to make decisions about how to use the limited amount of
time that they have with their students. It is with this in mind that our
lesson plan includes the identification of areas in which the teacher
will engage in corrective feedback.
Constant correction is both unrealistic for the teacher and demoti-
vating for the student. The commitment to humanism inherent to
communicative language teaching can lead some teachers to avoid
correction at all, especially correction in verbal contexts. The approach
taken here is that feedback should be restricted to those areas identified
as achievable at the learners’ current level of development. These areas
include those that have been explicitly taught in the lesson, which in
turn are presented at the correct stage in language development. In the
case of our lesson, ‘The zoo’, they are the correct use of location
adverbs, recently taught vocabulary, phrasal stress and appropriate
register. This is supported by research on corrective feedback, which
indicates increased usefulness if the learner is ‘ready’. In other words, if
learners are at the appropriate stage in the developmental path, then
corrective feedback can have positive effects for learning.
 Putting Theory into Practice 159

Stage-wise development is also one reason why the new linguistic

information in our lesson focuses on the fine-grained distinctions
between adverbial phrases such as near, next to and separate. At this
intermediate level, the learners should already have a basic store of
adverbials. This lesson may add some completely new expressions,
although there will be some individual variation as some learners will
have larger or more robust lexical stores than others. Yet all learners
will benefit from the recycled emphasis on adverbial phrases, espe-
cially as it builds on the verb phrases that came in the lesson the day
before. As shown in Table 7.2, the previous lesson reinforced the
distinction between habitual expressions using the simple present, and
expressions for current activity using the present progressive. The idea
of adverb phrases after verb phrases (and verb phrases after noun
phrases) is endorsed by Pienemann’s Processability Theory, which
provides a clear order of language development in terms of complex-
ity. Within this lesson we can also see growing complexity as learners
first comprehend and then produce the forms, followed by production
with correct phrasal stress.
Also true of second language development is the fifth observation:
Second language learning is variable in its outcome. This is another
reason why continuous recycling is needed. Some learners will acquire
and/or learn some aspects of language before or quicker than others.
This is the great mystery of second language development. While
researchers continue to try to find answers to why this happens, we
teachers would do well to accept it as reality and do our best to
accommodate it in our teaching. What does seem to be emerging are
some generalities about those areas of language which seem to develop
and those that are more prone to fossilisation, a point relevant to the
observation about difference in development across the linguistic
system.
Returning to variability in general outcome, we might consider
what this means for questions of assessment. From an SLA point of
view, assessment may be useful for monitoring learners in order to
determine what point they have reached along the path of develop-
ment. Knowing that development is step-wise, it would be unfair to
expect learners to reach a stage far beyond the stage in which the
course began. Equally, it is unrealistic to expect all learners to achieve
the same level in the same amount of time. Thus, it would not be
reasonable to expect all learners to achieve the same level in assess-
ment. Of course, there are legitimate factors involved in assessment,
such as institutional pressures to report on progress in a formal and
160 Language Teaching

standard way such that individual learners can be ranked and mea-
sured. From the point of view of language development, this will
document the variability that we have come to expect. Thinking of our
discussion of input in the last section, perhaps formal assessment
should also include a measure of how much input an individual
learner has exposed themselves to outside of the class. An ‘effort’
mark such as this would reflect the learner’s potential development
while allowing for variation in terms of actual development in the time
constraints of the course. Additionally, the washback effect might
encourage learners to seek as much input as possible in order to
receive the reward of higher marks, all of which will hopefully
improve their language ability.
One alternative implication for assessment is that there is a prin-
cipled reason to test precisely what was explicitly taught. If implicit
acquisition is beyond the control of the teacher or learner, and if we
are not entirely aware of what aspects of input become intake for the
learner, then perhaps it is unfair to test anything connected with
implicit acquisition. Explicit, metalinguistic knowledge, on the other
hand, is by definition identifiable. If metalinguistic knowledge is
knowledge that can be learned through deliberate attention and hard
work, then the logical conclusion is that testing such knowledge is
valid and fair. Of course, it must be recognised that such assessment
can only provide a fraction of any learner’s total language knowledge
and ability. But one logical conclusion of the view we have presented is
that there could be a place for using traditional testing practices.
Beyond variability in general outcome, we have seen that Second
language learning is variable across linguistic subsystems. As already
mentioned, some aspects of language seem to develop more readily
than others. In general, those aspects of language connected with
meaning seem to be acquirable, as are core properties of syntax. It is
the area of functional morphology that seems particularly trouble-
some for acquisition. This is why returning to a relatively basic
distinction between simple present and progressive present in the
previous lesson is not unwarranted. Verbal morphology is one area
which needs continuous practice and rehearsal before it can be
mastered. Notice, by contrast, that there is no explicit teaching of
word order within the adverbial and prepositional phrases in this
lesson. Teaching the phrases in terms of their meaning and having the
learners use them in appropriate contexts should be enough to allow
for the relevant properties of word order to develop implicitly (unless
the native language causes them to use an incorrect order, in which
 Putting Theory into Practice 161

case explicit instruction is advised). Explicit emphasis in this and all of

the lessons on aspects of pronunciation, by contrast, is needed because
of the difficulty learners have in the acquisition of phonology. Intra-
linguistic variability is also useful when thinking about corrective
feedback. Those areas which seem open to explicit learning may
benefit more from corrective feedback. Whether a teacher corrects
in areas that have been found impervious to deliberate learning may
depend on external language teaching pressures.

Deciding What to Teach (and What Not to Teach)

In addition to helping us think about how to teach, an understanding
of research and theory can help us to determine what to teach. It may
also inform us about what not to teach. One point which we have only
touched on in this discussion so far is the issue of native language
transfer. While we know that there are effects from native language
transfer, for this lesson, we have not specified the native language of
the learners. That we have specified a foreign language context allows
us to assume the learners share a mother tongue. Regardless of the
native language, any lesson can appeal to those aspects of commu-
nication that are largely cross-linguistic. In this lesson, the emphasis
on word stress in the phrase is, for the most part, one that holds across
languages. So, you may ask, what is the value of teaching something
that the learners are already likely to know? There are several reasons
for this. Firstly, there is no reason to expect that the learners will know
which aspects of communication are universal, and which are not.
Secondly, by drawing on existing L1-based knowledge, learners
should be able to transfer that knowledge explicitly for successful
application in the second language. The way that this lesson is
structured allows for renewed attention to the forms which are the
linguistic aim of the lesson, but with a different emphasis, and one
which is likely to be easily picked up on. All of this will help learners to
feel good about what they are learning and thus remain more engaged.
Even though there is clear evidence of native language effects in
second language development, we also know that there are limits on
the effect of a learner’s first language. This is because some of these
effects can be overcome through implicit development and/or explicit
learning. With the latter in mind, corrective feedback may be of
particular use in overcoming native language transfer effects. In fact,
research in this exact area has shown that explicit instruction helped
French learners overcome the subset problem which occurs with
162 Language Teaching

adverb placement, since French includes the superset option of verbs

before and after the main verb, while English allows only the subset
pre-verb option. This is why the caveat was given in the above section
urging the explicit teaching of word order in adverbial phrases if
needed for reasons of L1 transfer. Of course, gains have only been
shown to hold in the short term. Yet as this is an area of research that
has very little data, it remains an empirical question whether and the
degree to which explicit learning can result in long-term gains.
All of these points about second language development connect to the
only observation we have yet to address, the ninth in our original list:
There are limits on the effects of instruction on SLA. This may not be a
notion that we like to think about as teachers. Yet it is a generalisation
that any teacher would surely agree with, and one that is borne out in
empirical research as well. This, again, has direct implications for
deciding what to focus on with corrective feedback. While you can
insist that your students include the agreement -s morpheme on every
regular 3rd person singular verb, the robust result that this is an area that
learners seldom master suggests that you might divert your energies into
something more useful. This is further supported by the more functional
observation that absence of agreement does not cause problems for
understanding in spoken language. From a communicative language
point of view, then, it may not be worth insisting on it.

CONCLUSION

In this chapter, we have explored one lesson plan in an attempt to

illustrate how knowledge of abstract principles of linguistics and
research in second language development manifests itself in the
language classroom. There was no attempt to propose any new
super-method. In this post-methods era, it is widely accepted that
no such method exists. Instead we found ourselves with a mix of
contemporary and traditional approaches to language teaching, trying
to take a range of factors into account working within a degree of
practical, real-life constraints as regards questions of theory and
research. The main aim was to uncover some of the theoretical points
which are foundational to sound language-teaching practices. While
some teachers may do these things instinctively, well-qualified tea-
chers will also be able to justify them in terms of theory and research.
Because we defined a ‘generic’ teaching context in this chapter, a
number of important issues have not been discussed. In particular,
differences in age of acquisition are irrelevant for this discussion of one
 Putting Theory into Practice 163

single class of teenagers. Additionally, we have not addressed differ-

ences in levels of proficiency. Do all of these points still hold at
beginner and advanced levels? These and other current questions in
language development and language teaching form the basis of the
next and final chapter of this book.

FOR DISCUSSION

1. Develop a lesson plan for the class directly before or after the ‘Zoo’
lesson presented in this chapter. Be ready to justify the different
parts of your lesson in terms of what you know about language
and/or second language development.
2. Develop a lesson plan and exchange it with a classmate. Identify
the beliefs about language and second language development that
you can see in your classmate’s lesson plan.
3. Sketch a lesson plan that adheres to a strictly structural or form-
based view of language. Then sketch one that is exclusively func-
tion-oriented. Discuss the merits and shortcomings of each.
4. Discuss the lesson plan given in this chapter in terms of cultural
appropriateness. (Think about culture in terms of traditions in
society as well as educational culture.)
5. Discuss how you would alter this lesson if the context of the class
differed in terms of
a. proficiency: beginners vs advanced
b. age: young children vs adults
c. class size: 50 vs 5 students.
6. Develop supplementary activities that you could do with this lesson
if the class were twice as long or as activities to be done outside of
class.
7. Discuss the extent to which the lesson presented in the chapter is
appropriate for teaching languages other than English.

NOTES

1. As mentioned in Chapter 1, the majority of existing textbooks on

language teaching take a more ‘external’ approach.
2. See Boers (2000) and (2001) for research on words and idioms, respec-
tively. For papers that explicitly apply cognitive linguistic research to
classroom teaching, see the (2008) edited volume by Frank Boers and
Seth Lindstromberg.
 8
PRACTICE AND PRACTICES – RESPONDING TO
STUDENT NEEDS

INTRODUCTION

The last chapter presented a lesson plan for deconstruction in terms of its
underlying beliefs. The aim was to draw connections between linguistic
theory and agreed generalisations coming out of research in SLA on the
one hand, and the language classroom on the other. We did so, assuming
the most widespread teaching context possible. In this chapter, we
address questions of language development and language teaching
which could not be discussed in the very general context defined in
Chapter 7. We do so by broadly assuming the same syllabus and unit
topics, but adapting them to suit different types of learners. You will
notice that, while many (though not all) of our points are valid for the
teaching of any language, we have again limited our discussion to
English language teaching contexts. Firstly, we address questions rele-
vant to adult language learning at low levels of language proficiency. We
then go to the other end of the spectrum to discuss the teaching of
language at very high levels of language competence. Because all of our
discussions about teaching have assumed adult learners, we then address
the question of age, exploring the lesson in a child second language
development context. We end this chapter by exploring a current
controversial topic which has arisen out of the unique position that
English holds in the world when we consider our lesson from the point of
view of the English as a Lingua Franca (ELF) agenda.

LOW-LEVEL ADULT ESL LEARNERS

Adults relocating to new countries are often faced with not only a new
environment and culture, but also the difficulty of learning a new
language. Thus, in this section we consider adults newly arrived in
country, whether for work or accompanying a partner who has
 Practice and Practices – Responding to Student Needs 165

immigrated for work. We envision a range of professional back-

grounds and ages as well as a mixture of male and female students.
Additionally, we are assuming learners who are literate with a
secondary level of education – though many of the points we will
make could be relevant to secondary school learners as well.1 The
teaching context assumed is some kind of further education institute
or ‘night class’. We also assume an English-speaking country, and
assume at least a rudimentary knowledge of English given the perva-
sive nature of English in the world today. Thus, in terms of level, these
are not absolute beginners. Instead, we are defining these learners’
‘low’ proficiency as the lowest level in the CEFR proficiency scheme,
the A1 category. As characterised there, A1-level learners:

Can understand and use familiar everyday expressions and very

basic phrases aimed at the satisfaction of needs of a concrete type.
Can introduce him/herself and others and can ask and answer
questions about personal details such as where he/she lives, people
he/she knows and things he/she has. Can interact in a simple way
provided the other person talks slowly and clearly and is prepared
to help. (CEFR, p. 5)

While ‘The zoo’ may seem like a juvenile topic for adults, it does not
have to be. With consideration, most topics can be modified so they are
appropriate to the needs of the group in question. In order to illustrate,
we will keep the topic, but modify it. As presented in Chapter 7, the
lesson occurs within a unit about zoos as tourist attractions, with the
day’s lesson entitled: ‘The zoo: a map’. The changes to the lesson plan
are to teach (rather than recycle) basic vocabulary and for the main
language focus to be asking for directions.
The first task remains essentially the same, asking learners to position
animals in response to verbal input, creating a layout for a zoo. The
difference is a much more basic set of vocabulary, including common
animal names and simple prepositional phrases for location. Given the
needs of this group, instead of asking learners to debate the merits of
particular choices in positioning animals, Task 2 would ask them to use
the map they have created to practise asking and giving directions. This
task would come with a pre-task review of vocabulary and phrases for
asking directions. For homework, it is still important for learners to
expose themselves to as much input as possible, but instead of asking
learners to read for 30 minutes, they would be asked to look for articles
on-line or in print that have something to do with animals and to come
166 Language Teaching

to the next class having noted down as many headlines of articles to do

with animals as they can find. This kind of task requires engagement
with language without the burden of having to read and understand
text fully, which may be too challenging at this level.
The most important point in teaching ‘low-level’ learners is the
recognition that the label ‘low’ refers exclusively and only to language
ability. Fundamental to teaching in this context is genuine respect for
the learners, both in terms of their rich life experiences and their
existing abilities in other areas, whether academic or not. There is no
justification for equating difficulty in communicating in a second (or
third) language with low levels of general intelligence or a lack of rich
knowledge. Beyond level of formal education, which admittedly may
vary, as adults these learners have a fully developed set of cognitive
resources and deserve the dignity and respect that is accorded adults in
any setting. Moreover, as these learners are likely to be taking your
English class for very specific reasons, it is crucial for you as a teacher
to identify their individual language-learning needs, and to try to
address these as much as you can, while also keeping the needs of the
whole class in mind.
What, then, can we take from linguistic research when we consider
teaching in a low-level context? It seems that adult learners readily
acquire core language structure such as the basic word order within
sentences and most phrases.2 Perhaps surprisingly, in time, adult
learners can even acquire complex structural forms such as questions
and relative clauses. Thus, given the limits that exist in any teaching
context, it makes sense to provide as much natural input as possible to
foster acquisition of core syntactic properties and to use precious class
time for explicit instruction on other aspects of language. Specifically,
at early stages, learners need to learn vocabulary explicitly to increase
their range of resources for expression. In addition to amount, the type
of vocabulary is important. Based on research in cognitive psychology,
at this level it makes sense to teach generic instead of very specific
vocabulary items. For example, in our lesson it is the most common
animal names that should be taught, not obscure ones. Additionally,
at this level you should teach a full set of basic location words like next
to, behind and in front of, which capture the basic positional options
instead of teaching multiple ways to express the same basic notion. In
other words, avoid bombarding the learners unnecessarily with var-
iants on next to, such as beside, near or to the right/left of, unless
explicitly needed for communication.
Of course, the demands of a particular class or teaching situation
 Practice and Practices – Responding to Student Needs 167

may not allow for the luxury of providing rich authentic input to
facilitate (the eventual) acquisition of syntactic forms. Thus, it may be
that you explicitly teach some aspects of grammar, appealing to your
adult learners’ general intelligence to learn rules and patterns explicitly,
in order to consciously apply them in their language production as a
compensatory measure. This approach is certainly appropriate for
grammatical points which include functional morphology, such as
verbal agreement and plural forms. The somewhat depressing research
showing that even very advanced learners can often continue to have
difficulty with functional morphology in spontaneous production is
tempered by the fact that learners can come to know the rules explicitly
and can apply them if careful attention is given to production.
Another implication from research is that there is value in teaching
formulaic expressions. As we saw in Chapter 6, research on very
beginning-level learners shows that phrases that include grammar
beyond a learner’s level can be taught as fixed phrases, or unanalysed
chunks. In a very first lesson, for example, the incomplete phrase I’m a
. . . can be taught as a fixed opener in which to add nouns such as
student, man or nurse. Only later will learners learn that this useful
chunk comprises a 1st person singular pronoun, a 1st person singular
form of the verb to be and, perhaps later still, an article. There is no
reason why the use of formulaic chunks should not be extended for use
in a lesson like ours to facilitate communication and interaction. For
example, your lesson could include the teaching of a flexible general
expression for asking for a specific location such as: Could you tell me
where the nearest [noun] is? This might even be repeatedly practised in
to order to achieve automatisation without concern for the particular
linguistic features it contains. Looking ahead, when learners are ready
to handle more complex language they may be able to make use of this
stored chunk to develop the specific features of grammar and pronun-
ciation that it entails. The exact properties of modals, for example, are
beyond the A1 level of proficiency of these learners.
Another important point about these learners is that, as adults, they
will have a fully developed, albeit implicit, knowledge of the semantic
interpretation that underlies all of language by virtue of speaking
whatever language(s) they already speak. Part of learning a language is
learning how to map existing knowledge of meaning on to the specific
forms in the new language. For this reason, in addition to the usual
teaching of new language points, you should constantly rephrase what
you say to learners in order to maximise their ability to make sense of
the input. In addition, you could devote time in your teaching to
168 Language Teaching

developing their own ability to rephrase as a strategy for making

themselves better understood. As there is likely to be frequent com-
munication breakdown for these learners, emphasis on strategic
competence is easily justified. In sum, when teaching adults with
low levels of language proficiency, the advice here is to maximise
their existing stores of knowledge, focus the explicit teaching on
vocabulary as the building blocks of language, and carefully choose
which aspects of grammar you teach explicitly.

ADVANCED-LEVEL ADULT EAP LEARNERS

Presenting a different set of challenges are learners at the other end of

the proficiency spectrum. As with the previous section, the label
‘advanced’ refers to language ability; however, for the context as-
sumed here the learners are relatively advanced in terms of educational
level as well. For this discussion I will assume a typical context which
calls for the teaching of advanced learners: the university. Many
students who go abroad for a higher degree need language support
before they begin their degree course. The context for our discussion,
therefore, is an institute of higher education on a preparation course
for students already accepted to do postgraduate level study in a
particular discipline, but needing to improve their English. So pre-
valent is this area of teaching that it has developed as a distinct
subfield known as English for Academic Purposes, or EAP.3 While
some EAP programmes are large and resourced enough to provide
discipline-specific classes, most programmes teach a more general
form of academic English and so we will assume a general EAP course
for our learners. We can map our learners’ ability on to the CEFR,
putting them in the C1 category. Accordingly, a C1 learner:
Can understand a wide range of demanding, longer texts, and
recognise implicit meaning. Can express him/herself fluently and
spontaneously without much obvious searching for expressions.
Can use language flexibly and effectively for social, academic and
professional purposes. Can produce clear, well-structured, detailed
text on complex subjects, showing controlled use of organisational
patterns, connectors and cohesive devices. (CEFR, p. 5)

Of course, the idea of asking these learners to make a map of a zoo

is completely inappropriate. Instead, we will retain the animal theme,
but instead discuss biodiversity and the effect of climate change on
animal life, in preparation for the unit on animal rights to come. In
 Practice and Practices – Responding to Student Needs 169

terms of lesson plan, this class will start by making use of the
homework that students will have done before class. For homework,
each student will have been asked to find a primary source in the
university’s library relevant to the question of animals and climate
change, to skim it and to be prepared to give the class a general idea of
what it is about. Additionally, students will be asked to bring a copy of
the reading to class. The lesson itself begins with a brief discussion of
the reading that the students have done. The main language focus of
this class, however, will be one central to EAP: language for argu-
mentation. In class, students are asked to identify specific sentences/
phrases in their chosen source which indicate development of an
argument. This could range from common short phrases, such as
Author (date) claims that . . . to more complex phrases, such as While
some insist that x, a more reasonable view is y. The task is for each
student to go through the source picking out phrases for argumenta-
tion to record in a list. The aim of the day’s lesson will be to create a
single class list of phrases for argumentation from the individual
contributions of the learners.4 Homework will then be for the learners
to write a paragraph about animals and climate change which uses as
many of the phrases from the list as possible.
At this level, learners have a well-developed knowledge of grammar
rules and a wide range of dictionary knowledge. What they often lack is
native-like grammatical accuracy and pronunciation, as well as the
more subtle knowledge of how particular words or structures are used.
Because these learners need to develop the language and conventions
used within the academic community, it is natural that most EAP
programmes rely heavily on a Genre Analysis approach to teaching.5 In
this context, a genre is a communicative event that is associated with a
specific group who have a shared purpose. A text within a particular
genre, whether written or spoken, can be identified by specific struc-
ture, linguistic features, terminology and register.6 There are, in other
words, features that clearly identify an essay as different from a
newspaper article or an email. Because advanced learners already have
a well-developed level of general English knowledge, situating their
language learning directly in the genre of academic English provides a
very useful way for improving not only their language level, but also the
specific language knowledge they will need in order to succeed on their
degree course. Our lesson exemplifies a Genre Analysis approach. The
basis of the lesson is an example from the genre of academic writing
which the learners have chosen, and the focus is on one distinctive
feature of the genre: ‘language for argumentation’.
170 Language Teaching

Connecting this lesson back to the discussions of linguistics and

SLA, the use of ‘authentic’ materials is a principled choice and can be
justified when thinking of the research showing that learners are able
to acquire some very subtle, so-called poverty of the stimulus phe-
nomena. While explicit attention is being paid to the features of
academic text in the lesson, learners are also receiving natural input
which will allow for implicit acquisition as an added bonus. The use of
authentic materials also allows for attention to subtle differences in the
way language is actually used. The following verbs, for example, are
broadly synonymous: argue, claim, assert. However, a sophisticated
writer will know that assert, in contrast with argue, suggests the
argument is not well supported. One useful tool for both teachers and
learners at this level is the use of concordancing. One of the many
benefits of the Internet is access to very large databases of language.
There are free concordancing websites that allow anyone to search
databases of text for examples of specific words or phrases. A careful
analysis of examples of assert and argue from a concordancer could
show the difference between the use of these two words.7
You will have noticed that this lesson requires quite a high level of
analysis on the part of the learner. As with all teaching, the strengths of
the particular learners should be recognised and utilised in the class-
room. For this group, because they have been accepted for postgrad-
uate study, it is safe to assume a high level of general intelligence and
educational training. Moreover, when considering the needs of this
group, it is appropriate to develop their analytical skills to equip them
better for postgraduate study. Thus, it is not inappropriate to teach
them to engage in some basic discourse analysis. With this kind of
training they will be able to read not just for academic content, but to
develop their own abilities to write as well.
As illustrated here, language teaching at this level and for this
context is heavily skills-based, focusing on the academic skills of
research, reading and writing. Yet this does not mean that the more
traditional areas of language instruction are ignored. Again appealing
to existing research, we know that very advanced speakers often
fossilise in the areas of functional morphology such as verbal agree-
ment and grammatical words such as articles – at least in terms of
implicit, automised language production. Keeping in mind that much
of the language production burden for this group will fall on their
ability to write, emphasis should be given to editing skills so that these
minor errors are caught and fixed as part of the writing process. For
speaking, by contrast, the needs are different. In an academic setting it
 Practice and Practices – Responding to Student Needs 171

is the ideas that matter most in speaking, whether inside or outside the
class. Thus, emphasis should be placed on areas of expression and
interpretability. In the final section of this chapter, we will take this
idea further when we explore one radical idea which clearly delimits
which features of language are important for interpretability. For
now, we will leave this discussion and move on to another context
which highlights different points about language development.

YOUNG LEARNERS

Almost all of the discussion in this book has been about adult learners.
In this section we will make use of our lesson to discuss briefly some of
the issues relevant to child second language development. The first
tricky question is what age constitutes child second language learning,
as opposed to bilingualism on the one hand and ‘adult’ learning on the
other. As mentioned in Chapter 1, if a child is exposed to two languages
simultaneously from birth, this is generally understood to result in two
native languages, or a bilingual child.8 Following the practice that has
become accepted in the field, we use the term child L2 to refer to the
situation in which a child already has a native language (or two) when
they begin to develop an additional language. This is also sometimes
referred to as successive child bilingualism. While an exact age for child
L2 is not agreed, the age by which a child will have developed the core
structure of their native language in terms of sentence structure and
pronunciation is usually accepted to be 5 years.9 For the upper
boundary, the long tradition of viewing puberty as the dividing line
between child and adult learning is broadly supported by empirical
research. Setting aside the complications surrounding the critical
period, therefore, we will define child L2 acquisition as second lan-
guage development between the ages of 6 and 13.10
For this particular lesson, we will assume 9-year-old learners, an age
in the middle of the child L2 range. The class size is 24 and the setting a
primary school in one of the many countries which have introduced
English at the primary level. In other words, we assume an EFL, not an
ESL context.11 In terms of the CEFR, these children are nearing the A1
level, having had three years of non-intensive English lessons as part of
the weekly curriculum. A lesson about animals in a zoo is very
accessible for children. For the young learners, we would include
the same level of vocabulary as for the A1 adult learner: namely, basic
animal names and single preposition/adverb phrases for location. Yet
we would limit the number of animals to only five or six. We would,
172 Language Teaching

however, retain the original B2-level linguistic aim of phrasal stress,

though, as we will see, the way of teaching it would be different.
For the lesson plan, we would replace the pre-task with explicit
instruction on zoo animal names using pictures. The pre-task from the
original lesson in which the learners place animals on their desk in
response to instruction from the teacher would then become the main
task. Instead of cards with the animal names written on them,
however, both teacher and learners will have pictures of the animals.
A second difference is that the teacher would model the placement for
the children to follow as s/he gives instructions. Once the animals are
laid out on each desk, the children will be asked to practise the animal
names and phrases of location by repeating the same placement
exercise with variations in placement, and without modelling from
the teacher. The next modification to the ‘game’ is for the teacher to
hand out each of the teacher’s animal pictures to a different child to
call out the placement instruction for the rest of the class to follow. As
the children are doing this, the teacher will recast the children’s
phrases, paying particular attention to pronunciation and phrasal
stress. While this game could be repeated as many times as time and
interest allow, the next task is for the children to glue their animals on
to a large piece of paper to make a map. The last step in the lesson is to
begin teaching the spelling of each of the animal names. This might
begin in this lesson but be completed in the next. Hopefully there will
also be time for the children to draw cages, trees and flowers on their
maps. Homework, if appropriate in the institutional context, would
be to practise writing a couple of the new animal names taught in this
lesson.
The lesson is designed to take into account the cognitive level of 9-
year-old children in relation to type of task, expectations, response
and engagement. In addition, the lesson recognises the child–adult
relationship with clear instructions and ample modelling from the
teacher, so that the children feel well supported as they learn new
things. After doing the task with clear and structured support, it is
appropriate to encourage children of this age to become more in-
dependent by asking them to do the task themselves with support
readily available, and then involving them in the role of giving
instructions. This lesson also recognises that the children will already
have the concepts associated with the task; they will be familiar with
zoos, zoo animals and the concept of location and relative placement.
Practising the unfamiliar language associated with these known con-
cepts is facilitated by reliance on the ability children have to be creative
 Practice and Practices – Responding to Student Needs 173

– not only to engage their interest, but also to make up for linguistic
deficiencies.
Notice that the lesson requires much more listening than speaking
from the children. This is appropriate given the importance of input in
language acquisition. The assumption here is that the teacher would
be doing his/her best to speak exclusively in English throughout the
lesson. It is not unreasonable to expect a silent period as part of early
child second language development. While the children at this age and
with a couple of years of classroom exposure should hopefully be
beyond a silent period, the variability expected among learners means
that some may not. An attentive teacher will choose those children
who are ready and comfortable with the demands of speaking for the
task that asks them to deliver instruction to the class. The emphasis on
oral instead of visual input is also deliberate. There is an extent to
which heavy dependence on literacy might adversely affect some
aspects of language development. If new words or phrases are pre-
sented in writing first, child learners could make a connection between
form and pronunciation which could be unhelpful, given that many
words do not correspond exactly to their spelling. By using pictures
and oral presentation first, the children will rely on the sounds of the
words and phrases before learning to spell them.
From a linguistic point of view, the main difference between this
lesson and the lesson for adults is the balance of implicit and explicit
instruction. Because the ability for metalinguistic knowledge develops
with age, certain types of explicit instruction would be particularly
unhelpful. We mentioned that phrasal stress would remain an aim in
this lesson. However, the lesson does not contain any explicit instruc-
tion on phrasal stress. Instead, when giving instructions, the teacher
would be expected to model phrasal stress, perhaps in a somewhat
exaggerated fashion.12 Additionally, at every opportunity the teacher
is to recast not only the pronunciation, but also the correct phrasal
stress. This is because of both the cognitive limitations of children with
regard to metalinguistic knowledge and the research which supports a
potential among child learners for acquisition within the domain of
phonology. Similarly, the explicit teaching of words is limited to
animal names. The focus on location phrases is integrated into the
placement task, but without explicit teaching before the task, nor with
explicit practice in writing after. This is because of the child’s ability to
acquire grammatical structures. Vocabulary building, by contrast,
benefits from explicit instruction, regardless of the age of the learner.
In sum, despite the research showing that that there is no specific
174 Language Teaching

age for a critical period for language acquisition, we know that for
successful language development, younger is better. For both general
cognitive and specific linguistic reasons, children should receive less
explicit instruction and more natural unexplained engagement with
language than their adult counterparts. Since research suggests that
development of inflectional morphology and phonology seems to
mirror the native language situation for young learners, it is especially
important that these areas are ‘taught’ using lots of rich authentic
input with ample time for comprehension before expecting accurate
production from children. Yet production is important as well, so
children should be given opportunities to use the language they are
learning.13 Finally, we know that some aspects of language rely on
explicit instruction and experience, regardless of age. New words can
be learned with the aid of instruction and opportunities for practice.
And, considering the cognitive levels of child learners, repetition of
activities and mimicry of the teacher are much more appropriate then
for adults. Thus, while there are some similarities for teaching adults
and children, there are also many differences.

THE GLOBAL CONTEXT: ENGLISH AS A LINGUA FRANCA

While most of this book has been concerned with issues that are
relevant to the teaching of any language, these last two chapters have
illustrated the main points based on specific English language teaching
contexts. The reason to include a section on ELF is that any discussion
of English language teaching needs to acknowledge the unique posi-
tion that English holds as a global language. It is widely accepted that
there are more non-native speakers of English than there are native
speakers. Moreover, the most prevalent use of English among non-
native users is with other non-native speakers, not with native speak-
ers.14 It is in this context that the term English as a Lingua Franca, or
ELF, has come into being. The argument is that ELF should be seen as
a legitimate form of English alongside the many other varieties that
exist, from ‘inner circle’ dialects such as General American and
Received Pronunciation to ‘outer circle’ varieties such as Singapore
English or Indian English.15
While ELF is argued to be a legitimate form of English, it is not
being promoted as yet another dialect of English, just as ESL and EFL
are not varieties of English. The main point that distinguishes ELF
from other contexts is its intended use. Both EFL and ESL view the
ultimate aim of the learner to be native-like production of the language
 Practice and Practices – Responding to Student Needs 175

of a particular (usually inner circle) English-speaking group, whether

Australian, American or British, for example. ELF, by contrast, views
the ultimate aim as being able to communicate successfully with other
speakers, whether native or, more probably, non-native. What matters
most for ELF is successful interaction, not the degree to which features
of English approximate any one particular dialect. That the majority
of non-native interactions are with other non-native speakers is the
reason why the label lingua franca has been applied.
This view of English as an international language is very much in
concert with a view of language as dynamic and constantly changing.
It is uncontroversial to say that language is always changing, often as a
result of contact with other languages. What makes ELF different is
the recognition that changes to the English used in international
interaction are predominantly a product of non-native speaker influ-
ence, not native speaker innovations. The idea is that ELF speakers use
a particular set of English features which are needed in order to
communicate, whether they conform to ‘standard’ dialect features or
are influenced by the speaker’s native language. Central to the ELF
research agenda is to describe what non-native speakers of English
actually do when they speak, not to prescribe what they should do.
The methodology of ELF research is to record real interactions and
to note which features of English seem to be required for interpret-
ability and which lead to breakdown in communication. For example,
in her work on English phonology, Jenkins (2000) shows that features
of pronunciation often taught in English language classrooms, includ-
ing word stress, connected speech, weak forms and stress-timed
rhythm, are unnecessary at best and can be, at worst, unhelpful for
intelligibility. Similarly, the ability for a learner to produce either the
voiced /ð/ of this or the unvoiced /y/ thank is not necessary for
interpretability as these sounds can be replaced with /d/ and /t/
respectively, and interpretability is maintained.
If one truly accepts the idea of ELF, the logical conclusion to the
language classroom can be rather radical. If the genuine aim for
teaching is to lead to learners with the skills needed for successful
communication, and if that communication is likely to be largely with
other non-native speakers, then it is no longer the specific features of a
specified or accepted dialect that need teaching, but instead the
features of ELF. In this way, ELF challenges the native speaker norm
which predominates in English language teaching. We begin our
discussion of ELF in the classroom by asking in what context it might
be appropriate to take an ELF perspective. In our discussions of
176 Language Teaching

English language teaching so far we have considered four contexts.

Both the original lesson plan and the child L2 lesson were devised with
an EFL context in mind. The advanced and low-level learner contexts,
by contrast, were defined as ESL contexts. Since ESL refers to teaching
within a country in which English is a native language, successful
communication with native English speakers is an appropriate aim
and thus ELF would not be as appropriate as the main guiding
principle – though it may be useful for learners to be aware of the
features of ELF.
In an EFL classroom, it is more likely that learners who become
users of English will primarily interact with other non-native speakers
and thus ELF is more appropriate for an EFL context. Of our two EFL
contexts, however, we will rule out the child L2 context as an
appropriate context for applying ELF for several reasons. Firstly,
language teaching at this level and age is foundational in nature, with
attention to teaching core vocabulary and fostering as much natural
acquisition as possible. The ability for children to acquire language
naturally means that some of the difficult features of English, such as
certain aspects of pronunciation and grammatical morphology, are
candidates for native speaker levels of knowledge. Thus, we would
support the more traditional EFL approach to teaching children in
order to maximise the child’s ability to acquire language, with any of
the standard dialects as the appropriate stated aim.
The adult, secondary language teaching context, by contrast, may
be a more appropriate place to consider ELF principles because those
learners who, upon completing their studies, find that they need to use
English are likely to use it with other non-native speakers. However,
this idealistic position completely disregards the pragmatic reality that
secondary schooling includes formal assessment and tends to be
reliant on mass-produced mainstream textbooks. While it may be
that the ELF agenda is so persuasive that one day it is accepted in EFL
contexts, the current situation is far from this point. This causes me to
hesitate to promote an ELF-based curriculum, but instead, to endorse
ELF as a valuable part of teacher training courses so that teachers are
aware of the concept and familiar with the linguistic features of ELF.
This knowledge is likely to be very useful at higher levels of secondary
schooling when deciding which features of language to insist that
learners master, and which to overlook, especially when teaching
speaking and when managing interaction in the classroom.
While an ELF-based syllabus does not seem entirely appropriate for
the contexts discussed so far, there is a context in which ELF should
 Practice and Practices – Responding to Student Needs 177

inform the basic principles of teaching: any professional development

context in which learners are developing their English for use in
workplace interaction with other non-native speakers. Thus, the
context we will consider here is one of adult workers in a multi-
national corporation or a not-for-profit organisation who will reg-
ularly come into contact with people from a range of linguistic
backgrounds. We will place our learners at the same B2 level as
our original lesson plan, but the overall aims of the course will be quite
different. There is no formal assessment, nor any expectation that
these learners might move to an English-speaking country. Instead,
these learners know that they are going to need to use English in order
to be effective at their job.
Retaining our zoo theme, we would alter the lesson to suit this
group of learners. Recognising the need for conflict resolution that
comes up at personal and institutional levels in most professional
settings, we would present the learners with a scenario in which a zoo
is in deficit and deciding how it might cope with the news that it must
sell half of its land and get rid of half of its animals. After putting the
class into groups and handing out cards outlining details of the
scenario, the first task of this lesson would be to ask each group to
come up with a five-year business plan for the zoo. The second task
would be for each group to present their plan to the rest the class, who
are evenly divided between zoo employees and zoo shareholders.
In this lesson, primary emphasis is placed on discussion and
negotiation because this is the kind of language skill these learners
are expected to need in their professional settings. The influence of
ELF underpins the lesson, which revolves around verbal interaction.
Taking the concept of ELF one step further, however, this lesson will
also include, as a regular feature on this language programme, one
‘language monitor’ per group. Taking it in turns so that all students
act at some point as ‘language monitor’, this role would rotate from
person to person from lesson to lesson. Every time there is groupwork,
then, one member of the group would be designated a ‘language
monitor’. The task of the language monitor is to observe the group
interaction silently, making note of all instances in which there was
any difficulty in understanding on the part of any member of the
group. After the first group discussion task in our lesson, the monitors
would present their notes to the class. Then, with the aid of the
teacher, the class would analyse and discuss the causes of breakdown
in communication. So as not to disrupt the main task too much, the
teacher would plan to address the highlighted points more explicitly in
178 Language Teaching

a follow-up class. The monitors would be asked to continue their note-

taking during the second whole-class discussion.
The rationale behind the regular use of language monitors is the
idea that learners will be better users of language if they are aware of
which features of their language tend to lead to breakdown of
communication. With this awareness, they can work to develop these
areas in their own speech, and to look to them in the speech of others if
they find themselves on the receiving end of communication break-
down. Notice that the application of ELF here is not to teach any
specific features explicitly as correct or incorrect, but instead to apply
the rationale of ELF for determining which features of English are
worthy of concern. One reason for this is the fact that language is
dynamic. Despite the regularities that are emerging in research on
ELF, from situation to situation, the exact causes of communication
breakdown are likely to alter. Moreover, this approach avoids any
potentially patronising attitude whereby learners are made to feel as
though they are being taught some simplified form of English on a
‘good-enough’ basis. In a sense, this approach is one which places
strategic competence at its core, as learners are taught to hone in on
potentially problematic areas of both their own language and the
language of those they are likely to encounter when using English.
Thus, like all of the approaches discussed in these last two chapters,
this approach takes as a starting point for the specific needs of the
particular set of learners in each given context, drawing on under-
standing of both conceptions of language and research in SLA to
inform the particular parameters of the lesson.

CONCLUSION

We have covered much ground in this book, from questions of

theoretical linguistics to the practice of teaching. We started by asking
the very basic question of what language is, looking at a range of
answers. We have ended this book by taking the question to a specific
extreme by asking what the English language is. While many per-
spectives and many answers have been given throughout, hopefully
you will hesitate to assume that any one answer is wholly acceptable
and applicable for all situations. In fact, in many ways this book raises
more questions than it answers. One danger of this is that teachers
may feel as though the lack of definitive, agreed answers among
researchers means that no one has any answers. Yet hopefully you will
instead have developed a deeper understanding of the complexity of
 Practice and Practices – Responding to Student Needs 179

your subject area: language. And you will now have a sense of some of
the agreed generalisations which have emerged from SLA research
coming out of a range of linguistic perspectives. In this way, the
decisions that you make in your classroom can be informed not only
by intuition, your experience and the pressures of your particular
context, but by research and understanding of both language and how
language develops.

FOR DISCUSSION

1. For each of the four contexts given here, map out the lesson plan as
described, using the format for the lesson plan given in Chapter 7.
Then compare and contrast the interaction patterns for each lesson.
What connection is there between interaction patterns and the
particular context?
2. Find a lesson plan designed for secondary school teaching. Modify
it to suit low-level learners, advanced-level learners or child L2
learners.
3. Go back, once again, to the ten observations about second lan-
guage development. To what extent are they equally relevant for
the different contexts we have considered?
4. All of the lessons made use of task-based learning. Modify one to
reflect a PPP structure. Then try modifying it to adhere to one of the
‘designer’ approaches presented in Chapter 6.
5. Discuss the merits and limitations of ELF. To what extent is the
ELF concept useful for your English language teaching context? Is
the ELF concept applicable to any other languages that you are
familiar with other than English?
6. What do you consider the most important points made in this
book? To what extent will they affect your views on teaching?
What, specifically, could you plan to do differently as a teacher that
you did not do before (as a language teacher or learner)?

NOTES

1. Research on illiterate adults learning a second language (or students with

interrupted formal education) is a small but growing area of interest. See
Kurvers, van de Craats and Young-Scholten (2006), and Tarone and
Hansen (to appear). Details about a professional organisation dedicated
to Low Education Second Language and Literacy Acquisition (LESLLA)
can be found at http://www.leslla.org/.
180 Language Teaching

2. While learners seem to have very little difficulty with word order within
noun, verb and prepositional phrases, there can be difficulty of ordering
of adjectives within noun phrases for many learners.
3. Not only is this an area of professional expertise, but one of academic
interest as well. See the regularly issued Journal of EAP, published by
Elsevier, for academic papers devoted to research in EAP.
4. John Morley at Manchester University has developed a very useful and
extensive collection of phrases such as these, organised by different
types, all relevant to academic English. His Academic Phrasebank can be
found at: http://www.phrasebank.manchester.ac.uk/
5. John Swales is widely regarded for both academic and pedagogical
contributions to the fields of genre analysis and EAP. See Swales
(1990), and for a very useful pedagogical work, Swales and Feak (1994).
6. Genre analysis has developed out of systemic functionalist approaches to
language. The earliest use of the term genre in English language teaching
is attributed to Tarone et al. (1981). See Paltridge (2001) for a very clear
discussion of genre and language teaching.
7. One particularly good concordancer is provided within the University of
Quebec’s Compleat Lexical Tutor site at: http://www.lextutor.ca/con-
cordancers/concord_e.html. It has a wide range of database collections
to search from, distinguishing between spoken and written collections of
text. Additionally, at the time of writing, it includes databases in both
French and German in addition to English.
8. For a review of research in simultaneous bilingualism, see Meisel (2004).
9. For discussion of the age at which children are said to have a fully
developed grammar, see Guasti (2002).
10. See Philp, Oliver and Mackey (2008) and Snyder (2007) for discussions
of age and other issues relevant to child language learning. For a
collection of empirical studies on child L2 from a generative perspective,
see Haznedar and Gavruseva (2008).
11. Interestingly, most of the child L2 research has been done in either
immersion settings or naturalistic settings, in other words in the ESL
context. Very little research has been done in the child EFL classroom.
12. Even in native language acquisition, intonation is known to develop at a
relatively late age, certainly after the age of 5, but before our cut-off age
of 13.
13. Recall from Chapter 5 that research in immersion classrooms shows that
input alone is not sufficient.
14. While the exact figures behind these claims are hotly disputed, the general
claim as put forward by Graddol (1997, 2006) is largely accepted.
15. Two leading proponents of ELF are Jennifer Jenkins and Barbara
Seidlhofer. See Jenkins (2007) for an overview of ELF and for references
to other proponents of the ELF agenda. The Inner vs Outer Circle
concept is from Krachu (1985).
 GLOSSARY

Access to UG The phrase used to describe the extent to which second

language development is constrained by Universal Grammar. Because it
suggests an active, deliberate process, it has mostly been abandoned in
favour of the term UG-constrained development instead.
Accuracy Usually mentioned in contrast with fluency, this refers to the extent
to which a language learner is able to produce (and comprehend) language
without errors or mistakes.
Acquisition Implicit development which occurs without conscious or delib-
erate effort.
Acquisition by Processing Theory (APT) A theory proposed by John Truscott
and Mike Sharwood Smith as part of MOGUL, in which learning occurs
through the exposure to and reinforcement of input.
Acquisition-Learning Hypothesis The hypothesis proposed by Krashen to
distinguish between implicit and explicit development of knowledge. For
Krashen, these types of language knowledge remain distinct.
Activation level The extent to which a particular representation of knowl-
edge in the mind is primed, or ready to be called on in response to some
signal or input.
Active learning A basic tenet of communicative language teaching, this refers
to the participation of learners in discovering knowledge and developing
abilities themselves, instead of passively receiving input from the teacher or
course materials.
Affective factors Personal factors that may affect linguistic development,
such as motivation, personality, attitude and other emotional factors.
Affective Filter Hypothesis Within Krashen’s Monitor Model, the Affective
Filter refers to a psychological hindrance to language development that can
result from a particular learner’s levels of motivation, interest and other
emotional factors.
Approach The underlying conceptual basis for making decisions. With
regard to teaching, this may include conclusions based on empirical
results, more intuitive-based inclinations or ideas that have developed
based on personal experience.
Associative learning A view of learning held by cognitivists whereby learning
involves making connections between new knowledge and existing knowl-
edge.
182 Language Teaching

Audiolingual Approach An approach to teaching language based on the

ideas of Behaviourism, in which repeated input and rapid-fire drilling are
core to developing language.
Authentic materials Resources used for language teaching which were not
specifically designed for the classroom nor for language teaching. Typical
examples include material found on the Internet, or printed in newspapers
or magazines.
Automatisation The point at which knowledge no longer makes use of
explicit attention but instead exists without explicit awareness.
Avoidance The tendency for language learners to refrain systematically from
using a particular linguistic form in favour of an easier alternative form
that expresses the same meaning.
Behaviourism This theory from psychology, which reached prominence in
the 1950s, was based on experimental work with animals and posited a
view of human behaviour which depended on experience and conditioning
for the formation of habits.
Bilingual This term contrasts with child L2 development, referring to the
situation in which a speaker has native-level knowledge and ability in two
(or more) languages as a result of full exposure and development of the
languages from infancy.
Bottleneck Hypothesis A hypothesis proposed by Slabakova (2008) which
argues that some aspects of language, such as functional morphology,
create a ‘bottleneck’, causing difficulty for language development.
Brain The physical organ encased in the skull; for humans it consists of two
hemispheres and is made up of neurons, axons, etc.
Child L2 development This refers to the development of a second (or
additional) language through exposure to that language after native
language development, but before the so-called critical period has passed,
approximately between the ages of six and thirteen.
Co-construction The idea that development is not uni-directional; for lan-
guage teaching this refers to language knowledge developed by learners
and the effect on those imparting that knowledge, as well as the properties
of language itself.
Cognate Words in different languages that are related to each other etymo-
logically and thus share a common root. For example, luminous in English
and luminoso in Spanish are both derived from the Latin root lumin-,
which means light.
Cognitive Linguistics A branch of linguistics that is closely associated with
research in psychology in terms of theoretical orientation and methodology.
Colligation Phrases that are regularly used together to such an extent that
they take on the feeling of a fixed phrase.
Collocation Words that are regularly used together to such an extent that
they take on the feeling of a fixed phrase.
Communicative competence The notion proposed by Dell Hymes to counter
 Glossary 183

Chomsky’s competence/performance distinction; it refers to the knowledge

and abilities required for a speaker to be able to participate successfully in
a speech community.
Communicative Language Teaching (CLT) A very broad term which en-
compasses any approach to language teaching which takes communication
as its most fundamental aim.
Communities of practice Within Situated Learning Theory, identifiable groups
whose members have shared aims and norms with specified functions.
Community Language Learning Drawing on ideas from counselling, Charles
Curran’s approach to language teaching emphasises collaboration and
community development as a social process.
Competence The term proposed by Chomsky to refer to the core linguistic
knowledge that all native speakers of a language develop, whether they
have explicit knowledge of the properties of language or not. It contrasts
with ‘performance’ and exists as an idealised version of language.
Complexity Theory As recently applied to language, a theory that sees
language as in a perpetual state of change as a result of internal and
external pressures.
Comprehensible input Part of Krashen’s Input Hypothesis, this describes the
optimal type of input for learners – that which is one step ahead of the
learner’s current language level, also referred to as i + 1.
Comprehension approaches This refers to any method of language teaching
which considers comprehension as primary to learning and more impor-
tant to the learning process than production.
Concordancing The use of a concordancer, which is a search engine con-
nected to a large database of texts, in order to discover how specific words
or phrases have been used in existing texts.
Connectionism In psychology, this view sees cognitive knowledge, including
language, as one intricate system of connections, and not inventories of
symbols and abstract rules.
Construction A mapping between form and meaning which follows an
identifiable linguistic pattern and is associated with particular meaning.
Content-Based Instruction (CBI) An approach to language instruction in
which a subject area, such as geography or science, is taught through the
medium of the target language instead of the explicit teaching of language
as a subject.
Contrastive Analysis An approach to second language development begun in
the 1940s by Charles Fries, based on comparisons between linguistic
structures in the target and native languages.
Cooperative language learning This refers to the practice of collaborative
learning in groups as a way of improving individual learners’ abilities.
Corpus Linguistics A branch of linguistics based on a methodology which
searches databases of existing language in order to make claims about the
nature of language and/or the use of language.
184 Language Teaching

Creative Construction Hypothesis Also called the L2=L1 hypothesis, this

hypothesis argued that second language development makes use of the
same innate processes as native language development.
Critical period The window of time in which a biologically constrained
ability such as language can develop.
Critical Theory As an approach to language and language learning, this
theory recognises the importance of power relations within particular
communities.
Cues The features of a signal that help to determine the most probable match
between the signal and existing knowledge.
Deductive Learning by making sense of rules that explicitly state the proper-
ties of the phenomenon in question.
Descriptive grammar The grammar of a language as defined by the attested
use of the speakers of that language.
Direct Method An approach supported by the Reform Movement in the early
1900s in which all language teaching was done exclusively through the
medium of the target language.
Discourse competence The term used by Canale and Swain (1980) to refer to
the features of language which determine how texts, whether written or
spoken, are interconnected.
English as a Foreign Language (EFL) English taught in a context in which
English is not the language widely spoken in the country in which it is
being taught.
English as a Lingua Franca (ELF) A recently proposed variation of English
which is made up of those features of the language which are necessary for
successful communication, whether they conform to one of the accepted
‘standard’ dialects of English or not.
English as a Second Language (ESL) English taught in a context in which
English is widely spoken, usually as the official language of the country in
which it is being taught.
English for Academic Purposes (EAP) A type of ESP which teaches the
English as needed specifically for use in academic settings.
English for Specific Purposes (ESP) An approach to teaching English in which
the specific needs of the learners are determined in order to inform the
content of the course. These needs are more commonly professional, such
that there are ESP courses and textbooks with titles like English for
Nursing, or English for Bankers, and so on.
Entrenchment The process by which knowledge becomes strengthened, such
that it is more stable and more robust in the mind.
Error Non-targetlike use of language which reflects a consistent rule or
linguistic constraint.
Error Analysis The methodology used to analyse second language learner
production in terms of effects from the native language.
Exemplar-driven learning This captures the notion that language develop-
 Glossary 185

ment occurs through repeated exposure to examples of the new language

point in question.
Explicit input Information about or explanation of language, such as rules,
definitions or norms of use.
False friends Words that are similar in two different languages, usually
because of a shared etymological root, but which have evolved to have
divergent meanings. They are false friends as they may in fact mislead the
speaker into assuming they know the meaning of the word.
Field In Systemic Functional Grammar, this refers to the nature of the activity
which forms the context for a particular text.
Fluency The ability to produce (and comprehend) language without pauses
or errors which may lead to a breakdown in communication.
Focus on Form An approach to teaching language which attends to meaning
first, addressing difficult forms that cause problems for learners as they
arise.
Focus on FormS An approach to teaching language which attends exclusively
to the forms or structures of language. In its extreme form, it includes the
repeated drilling of decontextualised language patterns.
Focus on Meaning An approach to teaching language which attends exclu-
sively to the meaning of language without any explicit instruction on the
form of language.
Foreigner-talk The modification of language production by a speaker to
adapt to the language ability of the listener. This might involve the use of
simple structures or even non-standard forms.
Formulaic language Set phrases or clauses which are frequently used in
language production without modification.
Fossilise, Fossilisation These terms characterise the stage in which a learner’s
knowledge is no longer developing. While they are often applied generally,
it is usually more accurate to refer to the fossilisation of specific language
features for any particular speaker.
Frequency An important requirement for learning within associative learning
processes, as repeated examples of input are needed for cognitive devel-
opment.
Full Transfer A theoretical stance in SLA in which the starting point of
second language development is assumed to be the grammar of the existing
first language in its totality.
Functional Linguistics A branch of linguistics that approaches language in
terms of its function or use, and bases its analysis on the meaning of
language in context.
Fundamental Difference Hypothesis Proposed by Robert Bley-Vroman
(1990), this hypothesis assumes a qualitative difference between child
first and adult second language development.
Fuzziness Associated with Prototype Theory, this refers to the indistinct
nature of categories, especially at the boundaries.
186 Language Teaching

General knowledge/cognition The encyclopedic knowledge that a person

develops from experience over time. This ranges from knowledge of
specific facts to knowledge underlying skills such as driving a car.
Generative Linguistics A branch of linguistics that views language as the
product of a universal set of constraints which, once acquired, allow for
creative generation of language within those constraints.
Genre Analysis As an approach to teaching, exemplary texts associated with
a particular speech community are used to teach the specific properties of
that text type as well as practices of the speech community in question.
Grammar Translation The approach to language teaching which grew out of
the way that Classical languages like Latin and Ancient Greek were taught.
In this approach literary texts are translated with a focus on grammar, as
well as on the cultural values embedded in the text.
Grammatical/Linguistic competence The term used by Canale and Swain
(1980) to characterise grammatical knowledge as one type of knowledge
required for communicative competence.
Grammatical Metaphor This term, as used by Halliday, refers to the way in
which particular grammatical structures are extended in order to serve a
particular communicative function.
Head Parameter A universal parameter which regulates whether the main word
in a phrase precedes or follows the dependent words within the phrase.
Humanism An approach to teaching that comes out of the Progressive
movement in education whereby the needs of learners as individuals
and as people are seen as primary.
Immersion approaches An approach to foreign or second language instruc-
tion in which all classes are taught in the target language and not through
the medium of the language of the wider community.
Implicit input The target language input to which a language learner is
exposed. This means that explicit input is also implicit if it is delivered
exclusively using the language.
Inductive Learning by making sense of examples that give rise to regularities
that illustrate the properties of the phenomenon in question.
Inflectional morphology The functional elements in language required for
grammaticality, but which may or may not hold much semantic meaning.
Examples from English include 3rd person singular -s, progressive -ing and
articles a(n)/the.
Information gap A frequently used technique in communicative language
teaching, this refers to any task which requires learners to gather informa-
tion, whether from classmates or other sources. Such a task is assumed to
be motivating, as it requires the learner to discover new information
actively.
Input Enhancement Promoted by Sharwood Smith (1993), this refers to the
practice of drawing attention to particular language forms in attempt to
increase the salience of the forms in question.
 Glossary 187

Input Hypothesis Krashen’s hypothesis which says that the most effective
input for the learner is that which is one step ahead of the learner’s current
interlanguage.
Input Processing An approach proposed by VanPatten whereby language
development occurs when the learner’s processing abilities initially lead to
a misinterpretation because of limitations at particular stages of learning.
Integrated skills This refers to the idea that language should be taught in such
a way as to make use of speaking, listening, reading and writing together,
instead of teaching them as separate skills.
Interaction Hypothesis Proposed by Long (1981, 1983), this hypothesis
holds that language development requires learners to make sense of
and make sense in the target language.
Interlanguage (IL) A term coined by Larry Selinker (1972) to describe the
grammar of the learner acquiring another language. This grammar is
understood to be systematic, rule-based and dynamic.
Inter-linguistic variation Differences that exist between different languages.
Internalisation The development of psychological functions as a result of
interaction between the individual and the social environment.
Intra-linguistic variation Differences that exist within a single language
system.
L1/language transfer The influence that the existing language(s) has when
learning or using a second, or additional, language.
Language Acquisition Device (LAD) Coined by early generativists to refer to
the part of the mind implicated in native language development, this term
has been largely abandoned because of the mechanistic image it suggests.
Languaging Coined by Swain (2006), this term refers to the process required
of learners to do the hard work of making meaning through interaction in
order for language development to occur.
Langue A term used by Ferdinand de Saussure to refer to the properties of the
language system itself, in contrast with parole.
The Learnables This teaching method, developed by Harris Winitz, centres
on pictures and spoken input, graded in such a way as to lead to effective
language development.
Learner autonomy The idea that language learners should take responsibility
for their language development instead of relying heavily on the teacher
and/or the classroom materials.
Learner diary A technique for facilitating language teaching whereby learners
are asked to record their learning experiences on a regular basis, usually in
the form of a journal entry or a blog.
Learning Explicit development which occurs through conscious or deliberate
effort.
Lexical Approach Associated with Michael Lewis, this approach teaches
words and word combinations, as language is viewed solely as words and
not a system of rules based on structural principles.
188 Language Teaching

Lexical Metaphor The extension of a word or phrase to a meaning beyond

the literal.
Linguistic Determinism This strong version of the Sapir–Whorf Hypothesis
holds that language determines the way people think.
Linguistic general knowledge Explicit knowledge of the properties of lan-
guage, also known as metalinguistic knowledge.
Linguistic knowledge Implicit knowledge of language made use of in lan-
guage processing and production without explicit awareness or under-
standing.
Linguistic Relativism This weaker version of the Sapir–Whorf Hypothesis
holds that language has an effect on the way people think.
Literacy The ability to make sense of the written representation of a
language, usually in the form of a writing system, such as an alphabet.
Markedness This refers to the extent to which some linguistic form is more
common, usual or simple in comparison with other forms. A subject
relative clause, such as the girl who ran away, for example, is considered
less marked than a possessive relative clause, such as the girl whose puppy
ran away.
Mediation Within a sociocultural approach, the function that language
serves to connect an individual and the social environment.
Metalinguistic knowledge Explicit knowledge of language which allows for
thoughts about and discussion of language in terms of rules, patterns and
generalisations.
Method The way in which lessons are structured and delivered. This is
usually based on one’s approach to language in addition to more practical
constraints such as type of learners or institutional setting, as well as
questions of general pedagogy.
Mind The cognitive functions and representations of cognitive functions
associated with the brain.
Minimalist Theory The most recent version of Chomsky’s theory, this
maintains the essential concepts of generativism, but reduces the mechan-
isms involved in the theory to a minimum.
Mistake A one-off blunder which can be attributed to factors outside the core
grammatical system, such as fatigue, anxiety or distraction.
Mode In Systemic Functional Grammar, this refers to the medium through
which a particular text is delivered.
Modular, Modularity of Mind A view of how language is represented in the
mind proposed by Fodor (1983) which posits language as a distinct type of
knowledge akin to the senses and unlike general knowledge.
Modular Linguistic Knowledge Core knowledge of language, such as syntax
and phonology, which all native speakers have despite not being explicitly
aware of that knowledge.
Modular On-line Growth and Use of Language (MOGUL) A framework
developed by John Truscott and Mike Sharwood Smith which draws from
 Glossary 189

a range of linguistic approaches to make sense of the complex process of

language development.
Monitor Hypothesis Proposed by Krashen, this hypothesis captures the
ability that cognitively mature adults have to gauge and alter their
language output explicitly.
Monitor Model The label given to Krashen’s ideas about how adult second
language development happens. It includes five hypotheses: Acquisition-
Learning Hypothesis, Affective Filter Hypothesis, Input Hypothesis,
Monitor Hypothesis, and Natural Order Hypothesis.
Monolingual Having a fully developed linguistic system for only one lan-
guage which has developed as a result of exposure from infancy.
Multiple sensitive periods This term captures the idea that different aspects of
development of a complex system such as language are likely to have
different time periods in which natural, full development is possible.
Nativism The theory that language is a biological property with which
humans are endowed, which includes a predisposition for language
development. Special nativism further posits language as a distinct or
privileged type of cognitive ability, distinct from other types of cognition.
Natural Approach An approach to language promoted by Krashen and
Terrell (1983), which took a native language approach to second language
teaching.
Natural Order Hypothesis The hypothesis proposed by Krashen based on
research on the development of English functional morphemes; it claims
that second language development follows a ‘natural’ path, regardless of
other factors such as native language, age, type of input and so on.
Needs analysis The practice of formally assessing the needs of the learners in
order to deliver language classes effectively. This may take the form of a
formal battery of questions to design a whole course or of a more day-to-
day monitoring of students’ needs in order to deliver lessons that are as
relevant to the students as possible.
Negative evidence Explanations about what is unacceptable, ungrammatical
or not possible in the grammar of a language.
Negotiation of meaning Within the Interaction Hypothesis, this is the process
by which learners develop language ability as they work through chal-
lenges in their ability to communicate.
Neo-Vygotskyan The term is used to refer to sociocultural ideas that have
recently developed from the ideas proposed by L. S. Vygotsky in the early
1900s.
Nominalisation An example of grammatical metaphor, this refers to a
process in which an event whose basic grammatical form is verbal is
presented in a nominal form instead.
Noticing Having wide intuitive appeal but little theoretical underpinning,
this concept was proposed by Schmidt (1990, 1993) to refer to the idea
that input has to be taken up by learners in order for learning to take place.
190 Language Teaching

Notional Syllabus The term used by David Wilkins to describe a language

syllabus which takes notions and functions as its basic organising prin-
ciple, in contrast with a more structural or grammatical syllabus.
Null Subject Parameter A universal parameter which regulates whether a
language allows sentences without overt subjects or not. Spanish is an
example of a grammatically based null subject language while Japanese is a
discourse-based null subject language. English, by contrast, requires overt
subjects.
Object Regulation The first stage of development as understood by Vygots-
ky, in which learners rely on objects to make sense of what they are
learning.
Oral Approach An approach to teaching proposed in the early 1900s in
England, which advocated the teaching of language in increasingly com-
plex sequences, with a focus on speaking. It came to be associated with the
Situational Approach.
Other Regulation In this early stage of development, as understood by
Vygotsky, learners rely on others for assistance and direction as they learn.
Output Hypothesis Proposed by Swain (1985), this hypothesis challenged
Krashen’s Input Hypothesis, arguing that language production is also
necessary for language development.
Parole A term used by Ferdinand de Saussure to refer to the properties of
language as evidenced through the use of language, in contrast with
langue.
Performance The term proposed by Chomsky to characterise the language
that is produced by a speaker, which may or may not mirror the underlying
competence of the speaker, as it is prone to deficiencies due to distraction
or other production pressures.
Plato’s Problem This refers to a general learning problem articulated by Plato
which asks how a person can come to know so much more than the limited
set of data to which he/she is exposed. Generativists have applied this
question to language to support a nativist stance.
Post-Methods Era This is the term that has been used to describe the current
era in language teaching, as there is general agreement that there is no one
foolproof method which is effective for all language learners in all
language-learning contexts.
Poverty of the stimulus The observation that the input, or stimulus, to which
any child is exposed is impoverished in relation to the complexity of
knowledge that the child comes to know as a native speaker of a language.
PPP: Present, Practise, Produce This is a well-known structure for lesson
planning, widely adopted since the 1970s for language teaching.
Prescriptive grammar The grammar of a language as set out by those who
want to specify the rules of a language in terms of correct forms. This
grammar may or may not completely correspond to what is attested
among speakers of the language.
 Glossary 191

Priming The process whereby the mind has been prepared for a particular
signal by activating some associated concept first, thereby leading to
quicker activation in response to the signal.
Principles and Parameters The name given to the generative theory of the
1980s which captured the properties of language into universal principles
with parameters accounting for cross-linguistic differences.
Private speech From a Vygotskyan approach, inner speech directed to oneself
to mediate between thinking and behaviour.
Proceduralisation The process whereby a skill becomes automatic, allowing
for action without any conscious thought or attention to that action.
Process vs product In the context of language teaching, this opposition refers
to the ability to do some language-related activity, process, as opposed to
the actual language output, product.
Progressivism A movement in mainstream education in the twentieth century
which emphasises and values the needs of individuals, especially in terms
of how they might take responsibility for their own learning by becoming
more active learners.
Property Theory In second language acquisition, this describes a theory
which offers an explanation for the particular properties of any particular
stage of learner development at a moment in time.
Prototype Theory A theory which says that the ability to classify concepts
into categories is basic to human cognition. Prototypes are best examples
of a category, as they have a large number of core attributes of the category
in question.
Rassias Named after its founder, John Rassias, this teaching method sees
language as a cultural artefact which must be learned from within its social
context.
Realia Artefacts used in class that reflect the culture or practices of the target
language community, including leaflets, pictures, radio broadcasts, tele-
vised programmes, and so on.
Reflective learning The idea that learning will be more effective if learners are
aware of and able to reflect on their learning experience.
Reform Movement In response to Grammar Translation in the 1880s, this
approach to language teaching advocated teaching in terms of the four
skills.
Resting level The level of a particular representation of knowledge when it is
not active.
Rheme A term proposed by the Prague School to contrast with theme, rheme
refers to idea that the speaker moves to, or the new information in a
sentence or utterance.
Salience The extent to which something is clear, prominent or noticeable. In
language learning it often refers to more prominent features of language,
such a stressed syllables, in contrast with weak forms such as unstressed
syllables.
192 Language Teaching

Sapir–Whorf Hypothesis This refers to the controversial claim that language

affects or even determines thought, named after Edward Sapir and
Benjamin Whorf.
Scaffolding Proposed by educational psychologist Jerome Bruner, this me-
taphor refers to the support provided by teachers in order for learners to
develop knowledge and skills.
Schema A framework or blueprint for a concept developed in an individual’s
mind, based on both personal experience and human processes of cogni-
tion.
Selective error correction The practice of choosing which learner errors to
correct, based on a principled reasoning and knowledge of interlanguage
in addition to the more pedagogically motivated considerations of motiva-
tion and learner needs, for example.
Self-regulation A developmental stage beyond other regulation in which a
learner is able to rely on themselves for particular knowledge and skills.
Semiotics The study of the relationship between signs and meaning, semiotics
has had a large influence on the development of functional approaches to
language.
Sensitive Period A more accurate way to characterise the time period for
development of a biologically constrained property such a language. While
development may still occur outside the ‘sensitive period’, it is not likely to
develop naturally and fully.
Silent period A time period associated with native first language acquisition
in which the child understands language that they are not yet able to
produce. It has been suggested that adult language learners may also have
a silent period.
Silent Way Devised by Caleb Gattengo, this teaching method de-emphasises
input and forefronts the active learner instead. Teachers are to remain as
silent as possible so as to allow learners to work out properties of the
language for themselves.
Situated Learning Theory A theory which views learning as inextricably tied
to context and culture, and to the authentic activities within particular
communities.
Situational Approach An approach to teaching that developed in the early
1900s in Australia, which emphasised the teaching of language in context.
Sociocultural theory An approach which sees language as a product of
society instead of a biological feature of being human.
Sociolinguistic/Pragmatic competence The term used by Canale and Swain
(1980) to characterise knowledge required for appropriate use of a
language in terms of the norms and conventions of the speakers of the
language.
Speech Act Theory Proposed by John Searle, this theory captures the
meaning of linguistic expressions by analysing the locutionary, illocu-
tionary and perlocutionary acts within an utterance.
 Glossary 193

Standard Theory, Extended Standard Theory and Revised Extended Stan-

dard Theory The subsequent names given to early revisions of generative
theory proposed by Noam Chomsky in the 1950s, 1960s and 1970s.
Strategic competence The term used by Canale and Swain (1980) to refer to
knowledge needed to repair breakdown in communication.
Structuralism This mainstream approach to linguistics in the late nineteenth
century analysed language in terms of its structures, with particular
emphasis at the level of phoneme and morpheme.
Structure dependency/preservation This refers to the universal property of all
languages whereby language rules or constraints respect the structure of
the language and not, for example, linear rules, even if such rules seem
more ‘logical’.
Subset Principle A principle that helps to explain how complex knowledge
such as language is acquired. By this principle, a learner will have a
predisposition for a subset of knowledge and only adopt the larger relevant
superset based on evidence in the input.
Suggestopedia Developed by Georgi Lozanov, this approach to teaching
combines ‘psycho-relaxation’ techniques with a firm authority figure to
promote language development.
Systemic Functional Grammar (SFG) The functionalist theory proposed by
Michael Halliday which posits language as communication; this approach
is especially known for its analysis of discourse.
Target language The language that the learner is trying to acquire.
Task A language learning activity with a goal that is not specifically
linguistic, but which is designed to require the use of particular linguistic
features, forms, knowledge and/or ability.
Task-based learning A communicative approach to language teaching which
centres on carefully constructed tasks.
Teacher-talk The modification of language by a teacher in order to adapt to
the language proficiency of the learners, usually by using simple structures
and well-known vocabulary.
Techniques The specific way of carrying out teaching within a teaching
method and adhering to approach: for example, the actual lesson plan
template or the way in which a teacher checks for understanding.
Tenor In Systemic Functional Grammar, this refers to the relationships
between the producer and receiver(s) of a particular text.
Text Either written or spoken, a text is a sample of discourse produced
within a specific context for a particular function.
Theme A term proposed by the Prague School to contrast with rheme, theme
refers to the starting point or already known information in a sentence or
utterance.
Total Physical Response (TPR) Developed by James Asher, this approach to
teaching places much stock in kinetics and the extent to which learning is
seen as a product of stimulus and response.
194 Language Teaching

Transformations and Transformational Generative Grammar The basis of

early generative theory, this refers to the relationship between grammatical
constructions. For example, an interrogative sentence in English is a
transformation of declarative sentence that involves the movement of a
wh- word to the beginning of the sentence.
Transition theory In second language acquisition, this describes a theory
which offers an explanation for how a learner moves from one stage of
development to another.
UG-constrained development The more current term used to refer to the
extent to which second language development is constrained by principles
of Universal Grammar.
Unanalysed chunk A phrase which is taught and learned as one single unit
corresponding to a specific meaning and/or functional intent without
attention given to the meaning of the smaller units which make up the
phrase, nor their grammatical properties.
Universal Grammar (UG) The set of universal constraints by which all
natural languages abide.
Usage-based Usage-based approaches to language emphasise the role that
using language plays in the development of language.
Vernacular The local form of a language which is widely used by a com-
munity of speakers; this form may or may not correspond to the standard,
and thus usually more prestigious, form of the same language.
Zone of proximal development (ZPD) As defined by Vygotsky, the potential
range of development for any given learner, as determined in relation to
the learner’s current abilities.
 REFERENCES

Aarts, B. (2007), Syntactic Gradience: The Nature of Grammatical Inde-

terminacy, Oxford: Oxford University Press.
Adger, D. (2003), Core Syntax: A Minimalist Approach, Oxford: Oxford
University Press.
Aitchison, J. (1998), The Articulate Mammal: An Introduction to Psycho-
linguistics, 4th edn, London: Routledge.
Ammar, A. and P. Lightbown (2003), ‘Teaching marked linguistic structures
– more about the acquisition of relative clauses by Arab learners of
English’, in A. Housen and M. Pierrard (eds), Investigations in Instructed
Second Language Learning, Berlin: Mouton de Gruyter, pp. 167–98.
Anthony, E. (1963), ‘Approach, method and technique’, English Language
Teaching, 17: 63–7.
Asher, J. (1966), ‘The learning strategy of the total physical response: A
review’, Modern Language Journal, 50: 79–84.
Asher, J. (1977), Learning Another Language through Actions: The Com-
plete Teacher’s Guide Book, Los Gatos, CA: Sky Oaks.
Atkinson, D. (2002), ‘Toward a sociocognitive approach to second language
acquisition’, Modern Language Learning, 86: 525–45.
Austin, J. L. (1962), How to do Things with Words, Oxford: Oxford
University Press.
Bailey, N., C. Madden and S. Krashen (1974), ‘Is there a ‘‘natural sequence’’
in adult second language learning?’, Language Learning, 24: 235–43.
Benson, P. (2001), Teaching and Researching Autonomy in Language
Learning, Harlow: Pearson Education.
Beretta, A. (1990), ‘Implementation of the Bangalore Project’, Applied
Linguistics, 11: 321–37.
Bialystok, E. and K. Hakuta (1994), In Other Words: The Science and
Psychology of Second-Language Acquisition, New York: Basic.
Biber, D., S. Conrad and R. Reppen (1998), Corpus Linguistics: Investigat-
ing Language Structure and Use, Cambridge: Cambridge University Press.
Birdsong, D. (1992), ‘Ultimate attainment in second language acquisition’,
Language, 68: 706–55.
Birdsong, D. (ed.) (1999), Second Language Acquisition and the Critical
Period Hypothesis, Mahwah, NJ: Lawrence Erlbaum.
Bley-Vroman, R. (1990), ‘The logical problem of foreign language learning’,
Linguistic Analysis, 20: 3–49.
196 Language Teaching

Block, D. (2007), Second Language Identities, London: Continuum.

Bloomfield, L. (1933), Language, New York, NY: Holdt.
Boers, F. (2000), ‘Metaphor awareness and vocabulary retention’, Applied
Linguistics, 21: 553–71.
Boers, F. (2001), ‘Remembering figurative idioms by hypothesising about
their origin’, Prospect, 16: 35–43.
Boers, F. and S. Lindstromberg (eds) (2008), Cognitive Linguistic Ap-
proaches to Teaching Vocabulary and Phraseology, Berlin: Mouton de
Gruyter.
Bongaerts, T. (1999), ‘Ultimate attainment in foreign language pronuncia-
tion: The case of very advanced foreign language learners’, in D. Birdsong
(ed.), Second Language Acquisition and the Critical Period Hypothesis,
Mahwah, NJ: Lawrence Erlbaum, pp. 133–59.
Bongaerts, T., B. Planken and E. Schils (1995), ‘Can late learners attain a
native accent in a foreign language? A test of the Critical Period Hypoth-
esis’, in D. Singleton and Z. Lengyel (eds), The Age Factor in Second
Language Acquisition, Clevedon: Multilingual Matters, pp. 30–50.
Brown, R. (1973), A First Language: The Early Stages, London: George
Allen & Unwin.
Bruner, J. (1961), ‘The act of discovery’, Harvard Educational Review, 31:
21–32.
Bruner, J. (1967), On Knowing: Essays for the Left Hand, Cambridge, MA:
Harvard University Press.
Butler, C. (2003), Structure and Function: A Guide to Three Major Struc-
tural-Functional Theories. Part 1, Approaches to the Simplex Clause,
Amsterdam: John Benjamins.
Canale, M. and M. Swain (1980), ‘Theoretical bases of communicative
approaches to second language teaching and testing’, Applied Linguistics,
1: 1–47.
Carroll, S. (2001), Input and Evidence: The Raw Material of Second
Language Acquisition, Amsterdam: John Benjamins.
Carter, R. (1998/2000), Mapping the Mind, London: Phoenix.
Chomsky, N. (1965), Aspects of the Theory of Syntax, Cambridge, MA:
MIT Press.
Chomsky, N. (1970), ‘Linguistic theory’, in M. Lester (ed.), Readings in
Applied Transformational Grammar, New York, NY: Holt, Reinhart &
Winston.
Chomsky, N. (1981), Lectures on Government and Binding: The Pisa
Lectures, Dordrecht: Foris.
Chomsky, N. (1986), Barriers, Cambridge, MA: MIT Press.
Chomsky, N. (1995), The Minimalist Program, Cambridge, MA: MIT Press.
Clahsen, H. and U. Hong (1995), ‘Agreement and null subjects in German L2
development: New evidence from reaction-time experiments’, Second
Language Research, 11: 57–87.
 References 197

Corder, P. (1967), ‘The significance of learners’ errors’, International Review

of Applied Linguistics, 5: 161–9.
Corder, P. (1971), ‘Idiosyncratic errors and error analysis’, International
Review of Applied Linguistics, 9: 147–59.
Corder, P. (1981), Error Analysis and Interlanguage, Oxford: Oxford
University Press.
Coulmas, F. (1989), The Writing Systems of the World, Oxford: Blackwell.
Council of Europe [accessed November 2009], The Common European
Framework of Reference for Languages, http://www.coe.int/t/dg4/linguis-
tic/CADRE_EN.asp.
Cowie, A. P. (ed.) (1998), Phraseology, Theory Analysis and Applications,
Oxford: Oxford University Press.
Croft, W. (2001), Radical Construction Grammar, Oxford: Oxford Uni-
versity Press.
Croft, W. and A. Cruise (2004), Cognitive Linguistics, Cambridge: Cam-
bridge University Press.
Curran, C. (1976), Counseling-Learning in Second Languages, Apple River,
IL: Apple River.
Dam, L. (1995), Autonomy from Theory to Classroom Practice, Dublin:
Authentik.
Danesi, M. (1995), ‘Learning and teaching languages: The role of ‘‘con-
ceptual fluency’’ ’, International Journal of Applied Linguistics, 5: 3–20.
Daniels, P. and W. Bright (eds) (1996), The World’s Writing Systems,
Oxford: Oxford University Press.
de Bot, K., W. Lowie and M. Vespoor (2007), ‘A dynamic systems theory
approach to second language acquisition’, Bilingualism: Language and
Cognition, 10: 7–21 and 51–5.
Dekydtspotter, L. (2001), ‘The Universal Parser and interlanguage: domain-
specific mental organization in the comprehension of combien interroga-
tives in English–French interlanguage’, Second Language Research, 17:
91–143.
Dekydtspotter, L. and J. Hathorn (2005), ‘Quelque chose (. . .) de remarqu-
able in English–French acquisition: Mandatory, informationally encapsu-
lated computations in second language interpretation’, Second Language
Research, 21: 291–323.
Dekydtspotter, L., R. A. Sprouse and B. Anderson (1997), ‘The interpretive
interface in L2 acquisition: The process–result distinction in English–
French interlanguage grammar’, Language Acquisition, 6: 297–332.
Dekydtspotter, L., R. A. Sprouse, K. Swanson and R. Thyre (1999), ‘Se-
mantics, pragmatics and second language acquisition: A case of combien
extractions’, in A. Greenhill, H. Littlefield and C. Tano (eds), Proceedings
of the 23rd Annual Boston University Conference on Language Devel-
opment, Volume 1, Somerville, MA: Cascadilla, pp. 162–71.
Dekydtspotter, L., R. A. Sprouse and R. Thyre (1998), ‘Evidence for Full UG
198 Language Teaching

Access in L2 acquisition from the interpretive interface: Quantification at a

distance in English–French interlanguage’, in A. Greenhill, M. Hughes, H.
Littlefield and H. Walsh (eds), Proceedings of the 21st Annual Boston
University Conference on Child Development, Volume 1, Somerville, MA:
Cascadilla, pp. 141–52.
Dewey, J. (1938), Experience and Education, New York, NY: Kappa Delta
Pi.
Dewey, J. (1944), Democracy and Education, New York, NY: Free Press.
Dik, S. C. (1978), Functional Grammar, Amsterdam: North Holland.
Diller, K. (1971), The Language Teaching Controversy, Rowley, MA:
Newbury House.
Doughty, C. and J. Williams (1998), Focus on Form in Classroom Second
Language Acquisition, Cambridge: Cambridge University Press.
Dulay, H. and M. Burt (1973), ‘Should we teach children syntax?’, Language
Learning, 23: 245–58.
Dulay, H. and M. Burt (1974), ‘Natural sequences in child second language
acquisition’, Language Learning, 24: 37–53.
Dulay, H. and M. Burt (1975), ‘Creative construction in second language
learning and teaching’, in M. Burt and H. Dulay (eds), On TESOL ’75:
New Directions in Second Language Learning, Teaching and Bilingual
Education, Washington, DC: TESOL, pp. 21–32.
Eckman, F. (1977), ‘Markedness and the Contrastive Analysis Hypothesis’,
Language Learning, 27: 315–30.
Ellis, N. (2003), ‘Constructions, chunking and connectionism: The emer-
gence of second language structure’, in C. Doughty and M. Long (eds),
Handbook of Second Language Acquisition, Oxford: Blackwell, pp. 63–
103.
Ellis, N. (2007), ‘The associative-cognitive CREED’, in B. VanPatten and J.
Williams (eds), Theories in Second Language Acquisition: An Introduc-
tion, Mahwah, NJ: Lawrence Erlbaum, pp. 77–95.
Ellis, R. (1994), ‘A theory of instructed second language acquisition’, in
N. Ellis (ed.), Implicit and Explicit Learning of Languages, London:
Academic, pp. 79–114.
Ellis, R. (1997), ‘SLA and language pedagogy: An educational perspective’,
Studies in Second Language Acquisition, 20: 69–92.
Ellis, R. (2003), Task-based Language Learning and Teaching, Oxford:
Oxford University Press.
Evans, V. and M. Green (2006), Cognitive Linguistics: An Introduction,
Edinburgh: Edinburgh University Press.
Fawcett, R. P. (2000), A Theory of Syntax for Systemic Functional Linguis-
tics, Amsterdam: John Benjamins.
Fillmore, C. (1988), ‘The mechanisms of ‘‘Construction Grammar’’ ’, Pro-
ceedings of the Fourteenth Annual Meeting of the Berkeley Linguistics
Society, 14: 35–55.
 References 199

Finer, D. (1991), ‘Binding parameters in second language acquisition’, in

L. Eubank (ed.), Point Counterpoint: Universal Grammar in the Second
Language, Amsterdam: John Benjamins, pp. 351–74.
Finer, D. and E. Broselow (1986), ‘Second language acquisition of reflexive-
binding’, in S. Berman, J.-W. Choe and J. McDonough (eds), Proceedings
of NELS 16, University of Massachusetts at Amherst: Graduate Linguis-
tics Students Association, pp. 154–68.
Firth, A. and J. Wagner (1997), ‘On discourse, communication, and (some)
fundamental concepts in SLA research’, Modern Language Journal, 81:
285–300.
Flege, J. E. (1999), ‘Age of learning and second language speech’, in D.
Birdsong (ed.), Second Language Acquisition and the Critical Period
Hypothesis, Mahwah, NJ: Lawrence Erlbaum, pp. 101–31.
Flege, J. E. (2009), ‘Give input a chance!’, in T. Piske and M. Young-Scholten
(eds), Input Matters in SLA, Bristol: Multilingual Matters, pp. 175–90.
Fodor, J. (1983), Modularity of Mind: An Essay on Faculty Psychology,
Cambridge, MA: MIT Press.
Fries, C. (1952), The Structure of English: An Introduction to the Structure
of English Sentences, New York, NY: Harcourt, Brace.
Gass, S. (1982), ‘From theory to practice’, in M. Hines and W. Rutherford
(eds), On TESOL ’81, Washington, DC: Teachers of English to Speakers
of Other Languages, pp. 129–39.
Gass, S. and L. Selinker (2008), Second Language Acquisition: An Intro-
ductory Course, 3rd edn, London: Routledge.
Gattegno, C. (1972), Teaching Foreign Languages in Schools: The Silent
Way, 2nd edn, New York, NY: Educational Solutions.
Gattegno, C. (1976), The Common Sense of Teaching Foreign Languages,
New York, NY: Educational Solutions.
Goldberg, A. (2006), Constructions at Work: The Nature of Generalization
in Language, Oxford: Oxford University Press.
Goldschneider, J. M. and R. DeKeyser (2001), ‘Explaining the ‘‘natural order
of L2 morpheme acquisition’’ in English: A meta-analysis of multiple
determinants’, Language Learning, 51: 1–50.
Graddol, D. (1997), The Future of English? London: British Council.
Graddol, D. (2006), English Next. Why Global English May Mean the End
of ‘English as a Foreign Language’, London: British Council.
Greenberg, J. (1963), ‘Some universals of grammar with particular reference
to the order of meaningful elements’, in J. Greenberg (ed.), Universals of
Language, Cambridge, MA: MIT Press, pp. 73–113.
Guasti, M. T. (2002), The Growth of Grammar, Cambridge, MA: MIT Press.
Halliday, M. A. K. (1985), An Introduction to Functional Grammar,
London: Edward Arnold.
Halliday, M. A. K. (2004), An Introduction to Functional Grammar, 3rd
edn, London: Edward Arnold.
200 Language Teaching

Harley, B. and W. Wang (1997), ‘The critical period hypothesis: where are we
now?’, in A. de Groot and J. Kroll (eds), Tutorials in Bilingualism, Psycho-
linguistic Perspectives, Mahwah, NJ: Lawrence Erlbaum, pp. 19–51.
Haspelmath, M. (2006), ‘Against markedness (and what to replace it with)’,
Journal of Linguistics, 42: 25–70.
Hayes, W. (2006), The Progressive Education Movement: Is it Still a Factor
in Today’s Schools?, New York, NY: Rowman & Littlefield.
Haznedar, B. (2001), ‘The acquisition of the IP system in child L2 English’,
Studies in Second Language Acquisition, 23: 1–39.
Haznedar, B. and E. Gavruseva (eds) (2008), Current Trends in Child Second
Language Acquisition: A Generative Perspective, Philadelphia, PA: John
Benjamins.
Hensch, T. K. (2004), ‘Critical period regulation’, Annual Review of Neu-
roscience, 27: 549–79.
Hilles, S. (1986), ‘Interlanguage in the pro-drop parameter’, Second Lan-
guage Research, 2: 33–52.
Holme, R. (2004), Literacy: An Introduction, Edinburgh: Edinburgh Uni-
versity Press.
Howatt, A. P. R. and H. G. Widdowson (2004), A History of English
Language Teaching, 2nd edn, Oxford: Oxford University Press.
Hutchinson, T. and A. Waters (1987), ESP: A Learner Centred Approach,
Cambridge: Cambridge University Press.
Hyltenstam, K. (1977), ‘Implicational patterns in interlanguage syntax
variation’, Language Learning, 27: 383–411.
Hymes, D. (1972), ‘On Communicative Competence’, in J. B. Pride and J.
Holmes (eds), Sociolinguistics: Selected Readings, Harmondsworth: Pen-
guin, pp. 269–93.
Ingram, D. (1989), First Language Acquisition: Method, Description and
Explanation, Cambridge: Cambridge University Press.
Ingram, J. (2007), Neurolinguistics: An Introduction to Spoken Language
Processing and Disorders, Cambridge: Cambridge University Press.
Ioup, G., M. Boustagui, M. El Tigi and M. Moselle (1994), ‘Re-examining
the critical period hypothesis: A case study of successful adult SLA in a
naturalistic environment’, Studies in Second Language Acquisition, 16:
73–98.
Jenkins, J. (2000), The Phonology of English as an International Language,
Oxford: Oxford University Press.
Jenkins, J. (2007), English as a Lingua Franca: Attitude and Identity,
Oxford: Oxford University Press.
Johnson, J. and E. Newport (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: 60–99.
Kanno, K. (1997), ‘The acquisition of null and overt pronominals in Japanese
by English speakers’, Second Language Research, 13: 265–87.
 References 201

Kanno, K. (1998), ‘The stability of UG principles in second-language

acquisition: Evidence from Japanese’, Linguistics, 36: 1125–46.
Keck, C., G. Iberri-Shea, N. Tracy-Ventura and S. Wa-Mbaleka (2006),
‘Investigating the empirical link between task-based interaction and ac-
quisition: A meta-analysis’, in J. Norris and L. Ortega (eds), Synthesising
Research on Language Learning and Teaching, Amsterdam: John Benja-
mins, pp. 91–132.
Keenan, E. and B. Comrie (1977), ‘Noun phrase accessibility and Universal
Grammar’, Linguistic Inquiry, 8: 83–9.
Kementerian Pelajaran Malaysia [Malaysian Ministry of Education] (1975),
English Language Syllabus in Malaysian Schools Tingatan [Level] 4–5,
Kuala Lumpur: Kementerian Pelajaran.
Klein, W. and Perdue, C. (1997), ‘The basic variety (or: Couldn’t natural
languages be much simpler?)’, Second Language Research, 13: 301–47.
Krachu, B. (1985), ‘Standards, codification and sociolinguistic realism: the
English language in the outer circle’, in R. Quirk and H. G. Widdowson
(eds), English in the World: Teaching and Learning the Language and
Literature, Cambridge: Cambridge University Press.
Kramsch, C. (2003), ‘Second language acquisition, applied linguistics and the
teaching of foreign languages’, Language Learning Journal, 28: 66–73.
Krashen, S. (1977), ‘Some issues relating to the Monitor Model’, in H.
Brown, C. Yorio and R. Crymes, On TESOL ’77, Washington, DC:
TESOL.
Krashen, S. (1981), Second Language Acquisition and Second Language
Learning, Oxford: Pergamon.
Krashen, S. (1982), Principles and Practice in Second Language Acquisition,
Oxford: Pergamon.
Krashen, S. (1985), The Input Hypothesis: Issues and Implications, London:
Longman.
Krashen, S. D. (2004), The Power of Reading: Insights from the Research,
Portsmouth, NH: Heinemann.
Krashen, S. D. and T. D. Terrell (1983), The Natural Approach: Language
Acquisition in the Classroom, New York, NY: Pergamon.
Kurvers, J., I. van de Craats and M. Young-Scholten (2006), ‘Research on
low-educated second language and literacy acquisition’, in J. Kurvers, I.
van de Craats and M. Young-Scholten (eds), Low Educated Adult Second
Language and Literacy Acquisition, Utrecht: LOT.
Labov, W. (1973), ‘The boundaries of words and their meaning’, in C.-J. N.
Bailey and R. Shuy (eds), New Ways of Analyzing Variation in English,
Washington, DC: Georgetown University Press, pp. 340–73.
Lado, R. (1957), Linguistics Across Cultures, Ann Arbor, MI: University of
Michigan Press.
Lakoff, G. (1987), Women, Fire and Dangerous Things: What Categories
Reveal about the Mind, Chicago: University of Chicago Press.
202 Language Teaching

Langacker, R. (1987), Foundations of Cognitive Grammar, Volume I,

Theoretical Prerequisites, Stanford, CA: Stanford University Press.
Lantolf, J. (ed.) (2000), Sociocultural Theory and Second Language Learn-
ing, Oxford: Oxford University Press.
Lantolf, J. and S. L. Thorne (2006), Sociocultural Theory and the Genesis of
Second Language Development, Oxford: Oxford University Press.
Lardiere, D. (1998), ‘Dissociating syntax from morphology in a divergent L2
end-state grammar’, Second Language Research, 14: 359–75.
Larsen-Freeman, D. and L. Cameron (2008), Complex Systems and Applied
Linguistics, Oxford: Oxford University Press.
Laufer, B. and S. Eliasson (1993), ‘What causes avoidance in L2 learning:
L1–L2 difference, L1–L2 similarity, or L2 complexity?’, Studies in Second
Language Acquisition, 15: 33–48.
Lave, J. and E. Wenger (1991), Situated Learning: Legitimate Peripheral
Participation, Cambridge: Cambridge University Press.
Lenneberg, E. (1967), Biological Foundations of Language, New York, NY:
Wiley.
Lewis, M. (1993), The Lexical Approach, London: Language Teaching.
Lewis, M. (2000), ‘Language in the lexical approach’, in M. Lewis (ed.),
Teaching Collocation: Further Developments in the Lexical Approach,
Hove: Language Teaching, pp. 126–54.
Lightbown, P. (1985), ‘Great expectations: Second language acquisition
research and classroom teaching’, Applied Linguistics, 6: 173–89.
Lightbown, P. (2000), ‘Anniversary Article: Classroom SLA Research and
Second Language Teaching’, Applied Linguistics, 21: 431–62.
Lightbown, P. (2003), ‘SLA research in the classroom/SLA research for the
classroom’, Language Learning Journal, 28: 4–13.
Lightbown, P. and N. Spada (2006), How Languages are Learned, 3rd edn,
Oxford: Oxford University Press.
Little, D., J. Ridley and E. Ushioda (eds) (2003), Learner Autonomy in the
Foreign Language Classroom: Teacher, Learner, Curriculum and Assess-
ment, Dublin: Authentik.
Long, M. (1981), ‘Input, interaction, and second-language acquisition’, in H.
Winitz (ed.), Native Language and Foreign Language Acquisition, Annals
of the New York Academy of Science, Vol. 379, pp. 259–78.
Long, M. (1983), ‘Native speaker/non-native speaker conversation and the
negotiation of comprehensible input’, Applied Linguistics, 4: 126–41.
Long, M. (1991), ‘Focus on form: a design feature in language teaching
methodology’, in K. de Bot, R. B. Ginsberg and C. Kramsch (eds), Foreign
Language Research in Cross-Cultural Perspective, Amsterdam: John Ben-
jamins.
Long, M. (1996), ‘The role of the linguistic environment in second language
acquisition’, in W. Ritchie and T. Bhatia (eds), Handbook of Second
Language Acquisition, San Diego, CA: Academic Press, pp. 413–68.
 References 203

Long, M. and G. Crookes (1992), ‘Three approaches to task-based syllabus

design’, TESOL Quarterly, 26: 27–55.
Long, M. and C. Sato (1984), ‘Methodological issues in interlanguage
studies: an interactionist perspective’, in A. Davies, C. Criber and A.
Howatt (eds), Interlanguage, Edinburgh: Edinburgh University Press, pp.
253–79.
Lozanov, G. (1978), Suggestology and Outlines of Suggestopedy, New York,
NY: Gordon & Breach.
Lust, B. and C. Foley (eds) (2004), First Language Acquisition: The Essential
Readings, Oxford: Blackwell.
MacLaughlin, D. (1998), ‘The acquisition of the morphosyntax of English
reflexives by non-native speakers’, in M. L. Beck (ed.), Morphology and its
Interfaces in Second Language Knowledge, Amsterdam: John Benjamins,
pp. 195–226.
MacWhinney, B. (ed.) (1999), The Emergence of Language, Mahwah, NJ:
Lawrence Erlbaum.
Marsden, H. (2005), ‘L2 poverty of the stimulus at the syntax–semantics
interface: Quantifier scope in non-native Japanese’, in Y. Otsu (ed.),
Proceedings of the Sixth Tokyo Conference on Psycholinguistics, Tokyo:
Hituzi Syobo, pp. 217–41.
Marsden, H. (2008), ‘Pair-list readings in Korean–Japanese, Chinese–Japa-
nese, and English–Japanese interlanguage’, Second Language Research,
24: 189–226.
Marsden, H. (2009), ‘Distributive quantifier scope in English–Japanese and
Korean–Japanese interlanguage’, Language Acquisition, 16: 135–77.
Meisel, J. M. (2004), ‘The bilingual child’, in T. K. Bhatia and W. C. Ritchie
(eds), The Handbook of Bilingualism, Oxford: Blackwell, pp. 91–113.
Meisel, J. M., H. Clahsen and M. Pienemann (1981), ‘On determining
developmental stages in natural second language acquisition’, Studies in
Second Language Acquisition, 3: 109–35.
Mueller, J. (2005), ‘Electrophysiological correlates of second language
processing’, Second Language Research, 21: 152–74.
Musumeci, D. (1997), Breaking Tradition: An Exploration of the Historical
Relationship between Theory and Practice in Second Language Teaching,
Boston: McGraw Hill.
Myles, F., J. Hooper and R. Mitchell (1998), ‘Rote or rule? Exploring the
roles of formulaic language in classroom foreign language learning’,
Language Learning, 48: 323–63.
Nation, I. S. P. (2001), Learning Vocabulary in Another Language, Cam-
bridge: Cambridge University Press.
Nattinger, J. R. and J. S. DeCarrico (1992), Lexical Phrases and Language
Teaching, Oxford: Oxford University Press.
Newmeyer, F. (1998), Language Form and Language Function, Cambridge,
MA: MIT Press.
204 Language Teaching

Norton, B. (2000), Identity and Language Learning: Gender, Ethnicity, and

Educational Change, Harlow: Longman/Pearson.
Norton, B. and K. Toohey (eds) (2004), Critical Pedagogies and Language
Learning, Cambridge: Cambridge University Press.
Nunan, D. (2004), Task-Based Language Teaching, Cambridge: Cambridge
University Press.
O’Grady, W. (1999), ‘Toward a new nativism’, Studies in Second Language
Acquisition, 21: 621–33.
O’Grady, W. (2003), ‘The radical middle: Nativism without universal
grammar’, in C. J. Doughty and M. H. Long (eds), The Handbook of
Second Language Acquisition, Oxford: Blackwell, pp. 43–62.
Paltridge, B. (2001), Genre and the Language Learning Classroom, Ann
Arbor, MI: University of Michigan Press.
Paradis, M. (2004), A Neurolinguistic Theory of Bilingualism, Amsterdam:
John Benjamins.
Philp, J., R. Oliver and A. Mackey (2008), Second Language Acquisition and
the Younger Learner, Amsterdam: John Benjamins.
Pienemann, M. (1998), Language Processing and Second Language Devel-
opment: Processability Theory, Amsterdam: John Benjamins.
Pienemann, M. (ed.) (2005), Cross-linguistics Aspects of Processability
Theory, Amsterdam: John Benjamins.
Pinker, S. (1994), The Language Instinct: How the Mind Creates Language,
New York, NY: Harper Collins.
Piske, T. and M. Young-Scholten (2009), Input Matters in SLA, Bristol:
Multilingual Matters.
Prabhu, N. S. (1987), Second Language Pedagogy, Oxford: Oxford Uni-
versity Press.
Radford, A. (1997) Syntax: A Minimalist Introduction, Cambridge: Cam-
bridge University Press.
Richards, J. and T. Rodgers (2001), Approaches and Methods in Language
Teaching, Cambridge: Cambridge University Press.
Robinson, P. (1995), ‘Attention, memory and the ‘‘noticing’’ hypothesis’,
Language Learning, 45: 283–31.
Robinson, P. and N. Ellis (2008), Handbook of Cognitive Linguistics and
Second Language Acquisition, Mahwah, NJ: Lawrence Erlbaum.
Rosch, E. (1973), ‘Natural categories’, Cognitive Psychology, 4: 328–50.
Rosch, E. (1975), ‘Cognitive ReferencePoints’, CognitivePsychology, 7: 532–47.
Rothman, J. and M. Iverson (2007), ‘On L2 Clustering and Resetting the
Null Subject Parameter in L2 Spanish: Implications and Observations’,
Hispania, 90: 329–42.
Russell, J. and N. Spada (2006), ‘The effectiveness of corrective feedback for
the acquisition of L2 grammar: A meta-analysis of the research’, in J.
Norris and L. Ortega (eds), Synthesising Research on Language Learning
and Teaching, Amsterdam: John Benjamins, pp. 133–64.
 References 205

Rutherford W. (1968), Modern English: A Textbook for Foreign Students,

New York, NY: Harcourt, Brace & World.
Rutherford, W. (1987), Second Language Grammar: Learning and Teach-
ing, London: Longman.
Schachter, J. (1974), ‘An error in error analysis’, Language Learning, 24: 205–14.
Schmidt, R. W. (1990), ‘The role of consciousness in second language
learning’, Applied Linguistics, 11: 129–58.
Schmidt, R. W. (1993), ‘Awareness and second language acquisition’,
Annual Review of Applied Linguistics, 13: 206–26.
Schumann, J. (1979), ‘The acquisition of English negation by speakers of
Spanish: a review of the literature’, in R. Anderson (ed.), The Acquisition
and Use of Spanish and English as First and Second Languages, Wa-
shington, DC: TESOL, pp. 3–32.
Schwartz, B. D. (1993), ‘On explicit and negative data effecting and affecting
competence and linguistic behavior’, Studies in Second Language Acquisi-
tion, 15: 147–63.
Schwartz, B. D. and M. Gubala-Ryzak (1992), ‘Learnability and grammar
reorganization in L2A: Against negative evidence causing unlearning of
verb movement’, Second Language Research, 8: 1–38.
Schwartz, B. D. and R. A. Sprouse (1994), ‘Word order and nominative case
in non-native language acquisition: a longitudinal study of (L1 Turkish)
German interlanguage’, in T. Hoekstra and B. D. Schwartz (eds), Lan-
guage Acquisition Studies in Generative Grammar, Amsterdam: John
Benjamins, pp. 317–68.
Schwartz, B. D. and R. A. Sprouse (1996), ‘L2 cognitive states and the full
transfer/full access model’, Second Language Research, 12: 40–72.
Scovel, T. (1979), ‘Review of Suggestology and Outlines of Suggestopedy’,
TESOL Quarterly, 13: 255–66.
Searle, J. R. (1969), Speech Acts. An Essay in the Philosophy of Language,
Cambridge: Cambridge University Press.
Sebeok, T. A. (1966), Portraits of Linguists: A Biographical Source Book for
the History of Western Linguistics 1746–1964, Bloomington, IN: Indiana
University Press.
Selinker, L. (1972), ‘Interlanguage’, International Review of Applied Lin-
guistics, 10: 209–30.
Sfard, A. (1998), ‘On two metaphors for learning and the dangers of
choosing just one’, Educational Researcher, 27: 4–13.
Sharwood Smith, M. (1993), ‘Input enhancement in instructed SLA: Theo-
retical bases’, Studies in Second Language Acquisition, 15: 165–79.
Sharwood Smith, M. (1994), Second Language Learning: Theoretical Foun-
dations, London: Longman.
Sharwood Smith, M. (2004), ‘In two minds about grammar: On the inter-
action of linguistic and metalinguistic knowledge in performance’, Trans-
actions of the Philological Society, 102(2): 255–80.
206 Language Teaching

Singleton, D. and L. Ryan (2004), Language Acquisition: The Age Factor,

2nd edn, Clevedon: Multilingual Matters.
Slabakova, R. (2008), Meaning in the Second Language, Berlin: Mouton de
Gruyter.
Snow, C. and M. Hoefnagel-Höhle (1978), ‘The critical period for language
acquisition: evidence from second language learning’, Child Development,
49: 1114–28.
Snyder, W. (2007), Child Language: The Parametric Approach, New York,
NY: Oxford University Press.
Steinberg, D. and N. Sciarini (2006), An Introduction to Psycholinguistics,
2nd edn, Harlow: Pearson/Longman.
Stevick, E. (1990), Humanism in Language Teaching, New York, NY:
Oxford University Press.
Swain, M. (1985), ‘Communicative competence: Some roles of comprehen-
sible input and comprehensible output in its development’, in S. Gass and
C. Madden (eds), Input and Second Language Acquisition, Rowley, MA:
Newbury House, pp. 235–52.
Swain, M. (2005), ‘The output hypothesis: Theory and research’, in E.
Hinkel (ed.), Handbook on Research in Second Language Teaching
and Learning, Mahwah, NJ: Lawrence Erlbaum, pp. 471–84.
Swain, M. (2006), ‘Languaging, agency and collaboration in advanced second
language proficiency’, in H. Byrnes (ed.), Advanced Language Learning: the
Contribution of Halliday and Vygotsky, London: Continuum, pp. 95–108.
Swales, J. (1990), Genre Analysis: English in Academic and Research
Settings, Cambridge: Cambridge University Press.
Swales, J. and C. Feak (1994), Academic Writing for Graduate Students: A
Course for Nonnative Speakers of English, Ann Arbor, MI: University of
Michigan Press.
Tarone, E., S. Dwyer, S. Gillette and V. Icke (1981), ‘On the use of the passive
in two astrophysics journal papers’, ESP Journal, 1: 123–40.
Tarone, E. and K. Hansen (to appear), ‘Alphabetic literacy and second
language acquisition by older learners’, in J. Herschensohn and M.
Young-Scholten (eds), The Handbook of Second Language Acquisition,
Cambridge: Cambridge University Press.
Taylor, J. R. (1995), Linguistic Categorization: Prototypes in Linguistic
Theory, 2nd edn, Oxford: Clarendon.
Thomas O. (1965), Transformational Grammar and the Teacher of English,
New York, NY: Holt, Reinhart, & Winston.
Tomasello, M. (2003), Constructing a Language, Boston, MA: Harvard
University Press.
Traxler, M. J. and M. A. Gernsbacher (eds) (2006), Handbook of Psycho-
linguistics, 2nd edn, Amsterdam: Academic.
Truscott, J. (1998), ‘Noticing in second language acquisition: a critical
review’, Second Language Research, 14: 103–35.
 References 207

Truscott, J. and M. Sharwood Smith (2004a), ‘Acquisition by processing: A

modular perspective on language development’, Bilingualism: Language
and Cognition, 7: 1–20.
Truscott, J. and M. Sharwood Smith (2004b), ‘How APT is your theory:
present status and future prospects’, Bilingualism: Language and Cogni-
tion, 7: 43–7.
Vainikka, A. (1993/94), ‘Case in the development of English syntax’,
Language Acquisition, 3: 257–325.
Van den Branden, K. (ed.) (2006), Task-Based Language Education: From
Theory to Practice, Cambridge: Cambridge University Press.
VanPatten, B. (1996), Input Processing and Grammar Instruction: Theory
and Research, Norwood, NJ: Ablex.
VanPatten, B. (2004), ‘Input processing in SLA’, in B. VanPatten (ed.),
Processing Instruction: Theory, Research and Commentary, Mahwah,
NJ: Lawrence Erlbaum, pp. 1–31.
VanPatten, B. (2009), ‘Processing matters in input enhancement’, in T. Piske
and M. Young-Scholten (eds), Input Matters in SLA, Bristol: Multilingual
Matters, pp. 47–61.
VanPatten, B. and J. Williams (2007), Theories in Second Language Acqui-
sition, Mahwah, NJ: Lawrence Erlbaum.
Van Valin, R. D. and W. A. Foley (1980), ‘Role and reference grammar’, in E.
Moravcsik and J. R. Wirth (eds), Current Approaches to Syntax, New
York, NY: Academic, pp. 329–52.
Vygotsky, L. S. (1979), ‘Consciousness as a problem in the psychology of
behaviour’, Soviet Psychology, 17: 3–35.
Vygotsky, L. S. (1986), Thought and Language, translated by A. Kozulin,
Cambridge, MA: MIT Press.
Webster, J. (ed.) (2007), Collected Works of M. A. K. Halliday, London:
Continuum.
Weinert, R. (1995), ‘The role of formulaic language in second language
acquisition: A review’, Applied Linguistics, 16(2): 180–205.
Wenger, E. (1998), Communities of Practice: Learning, Meaning and Iden-
tity, Cambridge: Cambridge University Press.
White, L. (1989), ‘The principle of subjacency in second language acquisi-
tion: do L2 learners observe the subset principle?’, in S. Gass and J.
Schachter (eds), Linguistic Perspectives on Second Language Acquisition,
Cambridge: Cambridge University Press.
White, L. (1992), ‘Long and short verb movement in second language
acquisition’, Canadian Journal of Linguistics, 37: 273–86.
White, L. (2003a), ‘Fossilization in steady state L2 grammars: Persistent
problems with inflectional morphology’, Bilingualism: Language and
Cognition, 6: 129–41.
White, L. (2003b), Second Language Acquisition and Universal Grammar,
Cambridge: Cambridge University Press.
208 Language Teaching

White, L. and F. Genesee (1996), ‘How native is near-native? The issue of

ultimate attainment in adult second language acquisition’, Second Lan-
guage Research, 11: 233–65.
Wilkins, D. A. (1976), Notional Syllabuses, Oxford: Oxford University
Press.
Willis, J. and D. Willis (eds) (1996), Challenge and Change in Language
Teaching, Oxford: Heinemann.
Willis, D. and J. Willis (2007), Doing Task-Based Teaching, Oxford: Oxford
University Press.
Winitz, H. (ed.) (1981), The Comprehension Approach to Foreign Language
Instruction, Rowley, MA: Newbury House.
Wode, H. (1977), ‘On the systematicity of L1 transfer in L2 acquisition’,
Proceedings from 1977 Second Language Research Forum (SLRF), Los
Angeles, CA: University of California Press, pp. 160–9.
Wode, H. (1978), ‘Developmental sequences in naturalistic L2 acquisition’,
in E. Hatch (ed.), Readings in Second Language Acquisition, Rowley, MA:
Newbury House.
Wood, D., J. Bruner and G. Ross (1976), The role of tutoring in problem
solving, Journal of Child Psychology and Psychiatry, 17: 89–100.
Wray, A. (2002), Formulaic Language and the Lexicon, Cambridge: Cam-
bridge University Press.
Yoken, M. (ed.) (2007), Breakthrough: Essays and Vignettes in Honour of
John A. Rassias, New York, NY: Peter Lang.
Yuan, B. (2001), ‘The status of thematic verbs in the second language
acquisition of Chinese’, Second Language Research, 17: 248–72.
Zobl, H. (1980), ‘The formal and developmental selectivity of L1 influence
on L2 acquisition’, Language Learning, 30: 43–57.
Zuengler, J. and E. Miller (2006), ‘Cognitive and sociocultural perspectives:
Two parallel SLA worlds?’, TESOL Quarterly, 40: 35–58.
 INDEX

acquisition, 10–11, 30, 32, 98, 123–4, coherence, 3, 71

129, 130, 134, 135, 138, 152, Cognitive Linguistics, 5, 6, 17, 49, 68,
155, 157, 160, 161, 166–7, 170, 69, 70, 82–7, 90, 154, 155, 157,
173–4, 176, 181 166, 182
Acquisition-Learning Hypothesis, 98, colligations, 76, 138
123, 181 collocations, 76, 112, 138–9
Acquisition by Processing Theory Common European Framework of
(APT), 90, 111, 181 Reference for Languages (CEFR),
first/native language acquisition, 11, 143–4, 165, 168, 171
32, 46–8, 55, 80, 91, 126 Communicative Language Teaching
second language acquisition, 13, 34–8, (CLT), 115, 129–35, 158, 181,
41, 48–62, 63–5, 86, 95–118, 122, 183, 186
135–7, 154, 159, 162, 164, 170, Community Language Learning, 127,
178–9 183
activation level, 85, 86, 152, 181 community of practice, 80–1, 91, 183
affective factors, 97, 105–6 competence, 45, 47, 49, 58, 70–2, 105,
Affective Filter Hypothesis, 124, 181, 183
189 communicative competence, 70, 72,
approach, 122, 129, 134, 181 145–6, 147, 182
assessment, 131, 152, 159–60, 176, 177 discourse competence, 71–2, 145,
atomistic, 42, 68, 129 147, 184
Audiolingual Approach, linguistic/grammatical competence,
Audiolingualism, 32–4, 121–2, 182 70, 78, 145, 186
authentic materials, 132, 170, 182 sociolinguistic/pragmatic competence,
automatisation, 86, 95, 97, 104, 108, 71, 145, 192
110, 116–17, 167, 182 strategic competence, 72, 168, 178,
avoidance, 116, 182 145, 193
Complexity Theory, 80, 85, 183
Behaviourism, 30, 32, 37, 38, 44, 60, comprehensible input, 124, 183
62, 80, 121, 122, 182 Comprehension Approaches, 126–7,
Bilingualism, 15, 171, 182 183
Bley-Vroman, Robert, 11, 105, 185 Computed axial tomography (CAT), 4
Bottleneck Hypothesis, 109, 182 concordancing, 170, 180n, 183
brain, 5–7, 10, 17, 46, 48, 61–2, 78, Connectionism, 82, 183
80, 82, 86, 101, 111, 112, 127, construction, 73, 75–6, 82, 86, 89, 183
154, 182 Content-Based Instruction (CBI), 101,
Brown, Roger, 11, 47, 51 128, 183
Contrastive Analysis Hypothesis, 14,
child L2 development, 51, 105, 171, 33–9, 51, 60, 113, 122, 183
176, 182 Cooperative language learning, 133,
Chomsky, Noam, 3, 10, 29, 30–1, 37, 183
39, 41–2, 44, 46, 68–9, 70, 183, Corder, Pit, 13, 37
188, 190, 193 Corpus Linguistics, 76, 138, 183
210 Language Teaching
Creative Construction Hypothesis, 37, 24, 34, 49, 54, 65, 68–70, 72–7,
184 82, 84, 86–7, 91–2, 95, 125, 129,
critical period, 14–15, 19, 58–9, 105, 185; see also Systemic Functional
171, 174, 182, 184 Linguistics
Critical Theory, 94n, 184 functional magnetic resonance
imagining (fMRI), 4, 61
de Saussure, Ferdinand, 27–8, 70, 187, Fundamental Difference Hypothesis,
190 11, 105, 185
descriptive grammar, 8, 27, 175, 184
Direct Method, 26, 33, 96, 100, 121, Generative Linguistics, 3, 5, 6, 16, 29,
125 31, 41–8, 49, 50–8, 60–3, 64–5,
68–70, 72, 73, 82, 83, 86, 88,
Ellis, Nick, 94n, 98 101, 121, 123, 125, 129, 142,
English 154, 186, 191, 193
English as a Foreign Language Generative grammar, 29, 194
(EFL), 142–3, 171, 174, 176, 184 genre, 18, 110, 169
English as a Lingua Franca (ELF), Genre Analysis, 169, 186
174–8, 184 Grammar Translation, 25–6, 31, 39,
English as a Second Language (ESL), 96, 121, 186, 191
51, 143, 171, 174, 176, 184 Greenberg, Joseph, 2–3, 28–9
English for Academic Purposes
(EAP), 168–9, 184 Halliday, MAK, 3, 72–4, 125, 186,
English for Specific Purposes (ESP), 193
128, 184 humanism, 37, 134, 158, 186
error, 11–13, 18–20, 35, 37–8, 60–1, Hymes, Del, 70, 182–3
125, 130, 170, 181, 184
Error Analysis, 14, 37–8, 51, 122, immersion, 101, 115, 116, 128, 186;
184 see also Content-Based Instruction
selective error correction, 103–4, implicational, 3, 36
117, 122, 127, 131–2, 136, 137, individual variability, 13–14, 65, 97,
192 105–7, 110, 118, 134, 159–61,
173
false friends, 35, 185 information gap, 132, 186
feedback, 137, 148, 158, 161, 162 input
fluency, 11, 18–19, 37, 97, 105, explicit input, 96–8, 154, 185, 186
116–17, 125, 130–1, 146, 155, implicit input, 96, 118, 186
181, 185 Input Enhancement, 112, 186
Focus on Form, 98–9, 101, 116, 131, Input Hypothesis, 115, 124, 183,
136, 139, 151–3, 185 187, 189, 190
Focus on FormS, 98, 116, 131, 152, Input Processing, 112, 187
185 natural/authentic input, 16, 20, 48,
Focus on Meaning, 98, 152, 185 56, 97, 100–1, 110, 118, 129, 133,
foreigner-talk, 16, 185 136, 154, 158, 166–7, 170, 174
formalism, formalists, 2–3, 50, 68, 69, positive input, 19, 129
95, 103 Interaction Hypothesis, 114–15, 130,
formulaic language, 76, 138, 167, 185; 155–6, 187, 189
see also unanalysed chunks integrated skills, 132, 187
fossilise, fossilisation, 13, 19, 62, 104, interlanguage (IL), 13, 14, 19, 38, 39,
170, 159, 185 52, 54, 57, 60, 64, 65, 102, 103–
frequency, 63, 84, 99, 102, 111–13, 4, 105, 107, 115, 116, 157–61,
152, 156–7, 185 187
Full Transfer, 106, 185 inter-linguistic variation, 107, 187
Functional Linguistics, 3–4, 6–7, 16–17, intra-linguistic variation, 107, 161, 187
 Index 211

International Phonetic Association 188; see also modularity: modular

(IPA), 28, 31 linguistic knowledge
linguistic relativism, 83, 188
Johnson, Jacqueline and Elissa literacy, 8–9, 25, 32, 47, 114, 165,
Newport, 14, 58 173, 179n, 188
Long, Michael, 98, 115, 120n, 151,
kinetics, 127, 157, 193 187
Krashen, Stephen, 51, 98, 100, 115,
117, 123–5, 181, 183, 187, 189 manner of articulation, 28
markedness, 36, 103, 188
language Markedness Differential Hypothesis, 36
Language Acquisition Device (LAD), mean response time studies, 6
46, 187 metalinguistic knowledge, 7–9, 16–20,
language as form, 1–3, 20, 23, 98, 45, 88, 90, 97, 116–18, 154, 160,
101, 113, 126, 131, 135, 151–3; 173, 188
see also formalism metaphor, 74, 84, 186, 188–9, 192,
language as function, 1, 3–4, 16, 23, 196
34,125, 151–3; see also Functional method, 122, 129, 134, 188
Linguistics mind, 4–6, 39, 46, 82, 83, 84, 88–90,
language/L1 transfer, 34, 51, 60, 62, 96, 188
65, 106, 113–14, 146, 161–2, 187 mind–body dualism, 4
languaging, 130, 187 Minimalism, Minimalist Theory, 43–4,
langue, 28, 70, 187, 190 188
Lantolf, James, 79 mistake, 13, 18–20, 105, 123, 188
Latin, 8, 24–6, 27, 28, 39, 121, 186 Modular On-line Growth and Use of
The Learnables, 126–7, 187 Language (MOGUL), 62, 87–92,
learner autonomy, 133, 154, 187 95, 97, 101, 104, 106, 108, 110,
learner diary, 134, 187 111, 113, 118, 133, 152, 154, 188
learning, 17, 18, 30, 32–3, 34, 36–7, modularity, 45–6, 88–92, 95, 101–2,
49, 51, 58, 79, 82, 84–7, 90, 96, 108, 110, 113, 188
98, 102, 110, 111–12, 123, 126–7, modular linguistic knowledge, 64,
130, 132–3, 152, 155–8, 161–2, 88–92, 104, 108, 188
167, 171, 181, 187, 189 non-modular knowledge, 64, 88–92,
active learning, 18, 129, 132, 181 108, 111
associative learning, 82, 84–5, 181, Monitor Hypothesis, 123, 189
185 Monitor Model, 117, 123–5, 181, 189
deductive learning, 24, 26, 184 monolingualism, 15
exemplar-driven learning, 85, 184–5 morphology, 35, 54, 64
incidental learning, 65, 98–9, 100, derivational morphology, 98
108, 114, 130 functional morphology, 109–10,
inductive learning, 24, 26, 186 118, 160, 167, 170, 182
learning strategies, 19, 20, 72, 98, inflectional morphology, 60–2, 64,
104, 105, 114, 117, 130, 168 98, 99, 106, 174, 186
reflective learning, 133–4, 191 morpheme, 2, 75, 162, 193
see also Task-Based Learning and morpheme order study, 11, 47, 51,
Teaching 64, 102, 124
Lenneberg, Eric, 58 motivation, 97–8, 124, 155, 181, 192
Lewis, Michael, 137–8, 187
Lexical Approach, 137–9, 151, 156, Nativism, 10, 30, 47, 189
187 Natural Approach, 100, 108, 123–7,
Lightbown, Patsy, 66n 189
linguistic determinism, 83, 188 Natural Order Hypothesis, 124, 189
linguistic knowledge 7–9, 44–5, 47, 64, needs analysis, 131, 189
212 Language Teaching
negative evidence, 96, 107, 189 reductionist, 42; see also atomistic
negotiation of meaning, 115, 116, 156, Reform Movement, 26, 31, 39, 121,
189 184, 191
neurolinguistics, 108, 111 register, 98, 99, 110, 148, 152, 158,
Noticing, 112, 115, 152, 189 169
Notional Syllabus, 125–6, 190 resting level, 85, 92, 111, 191
null subjects, 53–7, 64, 106–7, 190 rheme, 73–4, 191

Oral Approach, 32, 121, 190 salience, 99, 112–13, 156–7, 186, 191
Output Hypothesis, 115, 155, 190 Sapir–Whorf Hypothesis, 83, 188, 192
Overt Pronoun Constraint, 56–7, 63, scaffolding, 79–80, 105, 192
100–1 schema, 84–8, 192
Selinker, Larry, 14, 52, 187
parole, 28, 70, 187, 190 sensitive period, 59, 63, 189, 192; see
performance, 70, 72, 95, 183, 190 also critical period
phonetics, 2, 109; see also the Sharwood Smith, Mike, 64, 88, 113,
International Phonetic Association 182, 187, 189
(IPA) silent period, 100, 124–5, 173, 192
place of articulation, 28 Silent Way, 126, 177, 193
Plato’s Problem, 5, 190 Situated Learning Theory, 81, 183,
Post-Methods Era, 134–9, 151, 153, 193
162, 190 Situational Approach, 32, 121, 125,
poverty of the stimulus, 31, 56–7, 63, 190, 192
100–1, 170, 190 Slabakova, Roumyana, 109, 110, 182
pragmatics, 7, 108 Sociocognitive Theory, 81
prescriptive grammar, 8, 27, 175, Sociocultural Theory, 49, 65, 68, 69,
190 77–81, 91, 104, 133, 155, 188,
Present, Practice, Produce (PPP), 37, 189, 192
126, 190 sociolinguistics, 7; see also competence:
priming, 85, 111, 112, 191 sociolinguistic/pragmatic
Principles and Parameters, 43, 46–7, competence
52, 191 speech act, 75, 193
process vs product, 130, 191 stages of development, 11, 12, 20, 47,
Processability Theory, 120n, 159 51, 54, 64, 79, 102–5, 124, 127,
processing, 5, 6, 17, 20, 61–2, 81, 82, 132, 136, 157–9
86, 88, 91, 95, 101–2, 111–13, Structuralism, 1–3, 16, 18, 20, 27–8,
133, 135, 152, 157; see also 29, 32, 38, 41–2, 73, 121, 122,
acquisition: Acquisition by 126, 193
Processing Theory (APT); input: structure dependency, 29, 42, 193
input processing Subset Principle, 106–7, 113, 193
proficiency, 11, 19, 20, 57, 58, 59, 61, Suggestopedia, 127, 193
62, 98, 99, 103, 105, 117, 143, Swain, Merrill, 71, 115, 131, 156,
155, 164, 165, 167, 168 188, 190
Progressivism, 37, 129, 134, 186, 191 syntax, 3, 21, 35, 36, 42, 44, 54, 59,
property theory, 86, 191 61, 65, 88, 89, 102, 109, 110,
Prototype Theory, 82–4, 85, 185, 191 111, 118, 131, 161
psycholinguistics, 5, 6, 9, 90, 95, 108, systematicity, 13, 19, 38, 51–2, 103,
111, 112, 152, 157 105
Systemic Functional Linguistics, 6–7,
Rassias, 136, 191 72, 73, 193
realia, 32, 132, 191; see also input:
natural/authentic input target language, 13, 14, 19, 35, 37, 39,
recast, 172, 173 47, 52, 54, 59, 63, 97, 98, 100,
 Index 213

101, 106, 108, 111, 113, 114, Universal Grammar (UG), 3, 12,
122, 128, 133, 143, 154, 181, 18–19, 30, 44, 46–7, 50, 51–2,
184, 187, 192 181, 194
Task-Based Learning and Teaching UG-constrained development, 52–7,
(TBLT), 115–16, 135–7, 139, 147, 63, 100, 113, 181, 194
151, 193 usage-based, 85, 194
teacher-talk, 16, 193
text, 6–7, 71, 74–5, 99, 112, 169, 170, VanPatten, Bill, 63–4, 96, 112, 117,
186, 193 154, 157, 187
theme, 73–4, 193 variability, 61, 63–5, 105–11, 159–60
Total Physical Response (TPR), 126–7, vernacular, 25, 39, 194
193 Vygotsky, L. S., 79–80, 189
Transformations, 29, 42, 194
transition theory, 86, 194 washback, 160
Truscott, John, 62, 88, 112, 181, 188 White, Lydia, 55, 57–8, 60, 63–4, 107

unanalysed chunks, 138–9, 167, 194; zone of proximal development (ZPD),

see also formulaic language 74, 91, 104, 194