978-3-031-53414-0

Usage-Based
Second Language
Instruction
A Context-Driven Multimedia
Learning Approach
Ian Pemberton
Usage-Based Second Language Instruction
Ian Pemberton
Usage-Based Second
Language Instruction
A Context-Driven Multimedia
Learning Approach
Ian Pemberton
Warwick Foundation Studies
University of Warwick
Coventry, UK
ISBN 978-3-031-53413-3 ISBN 978-3-031-53414-0 (eBook)

https://doi.org/10.1007/978-3-031-53414-0
© The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature
Switzerland AG 2024
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher,
whether the whole or part of the material is concerned, specifically the rights of translation,
reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other
physical way, and transmission or information storage and retrieval, electronic adaptation, computer
software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher
nor the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Cover image: Peerapat Tandavanitj
This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland
AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Paper in this product is recyclable.

Contents
1 Introduction 1
1.1 Rationale 1
1.2 Outline 3
References 4
Part I Making the Case for Change

2 Historical Teaching Approaches 7
2.1 The Classical Period 8
2.2 The Reform Period 9
2.3 Summary 13
2.4 Implications 14
References 15
3 The Scientific Period 17
3.1 The Structural Syllabus 17
3.2 The Situational Syllabus 19
3.3 The Audiolingual Method 21
3.4 Universal Grammar 23
v
vi Contents
3.5 Communicative Competence 24

3.6 Summary 26
3.7 Implications 28
References 29
4 The Communicative Period 31
4.1 The Notional-Functional Syllabus 32
4.2 The Natural Approach 35
4.3 Summary 43
References 45
5 Communicative Language Teaching 47
5.1 Strong and Weak Versions 48
5.2 Classical and Current Versions 49
5.3 Communicative Principles 49
5.4 Communicative Activities 59
5.5 Criticisms 62
5.6 Summary 64
References 66
6 Other Communicative Approaches 69
6.1 Content-Based Instruction 69
6.2 Task-Based Instruction 72
6.3 Summary 81
References 82
7 The Lexical Approach 85
7.1 Lexical Categories 86
7.2 Theory of Language 90
7.3 Learning Theory 91
7.4 Methodology 92
7.5 Notebooks 97
7.6 Criticisms 100
7.7 Summary 101
References 102
Contents vii
Part II Establishing a Theoretical Basis

8 First Language Acquisition 105
8.1 Evolution 106
8.2 Primary Biological Skills 107
8.3 Constructivism 109
8.4 Sociocultural Theory 112
8.5 Social-Pragmatic Theory 115
8.6 Summary 119
8.7 Implications 121
References 122
9 Foundations 125
9.1 Dynamic Systems Theory 126
9.2 Construction Grammars 136
9.3 Summary 139
References 142
10 Usage-Based Linguistics 145
10.1 Fundamentals 146
10.2 Entrenchment 148
10.3 Association 150
10.4 Categorization 153
10.5 Analogy 153
10.6 Schematization 154
10.7 Summary 154
10.8 Implications 156
References 157
11 Empirical Research 159
11.1 Case studies 160
11.2 Cross-Sectional Studies 164
11.3 Corpus Analyses 172
11.4 Variation and Variability Studies 175
11.5 Multimedia Studies 179
11.6 Autonomy Studies 182
11.7 Summary 186
References 190
viii Contents
Part III Designing a Blueprint for Change

12 Principles 195
12.1 Environmental Principles 196
12.2 Behavioural Principles 207
12.3 Summary 214
References 217
13 Materials Design 221
13.1 The Role of the Teacher 224
13.2 Multimedia Materials 224
13.3 Ideational Content 227
13.4 Constructions 229
13.5 Pre-Watching Activities 231
13.6 While Watching Activities 232
13.7 Post-Watching Activities 235
13.8 Limitations 241
References 243
14 Conclusion 245
14.1 Making the Case for Change 245
14.2 Imagining a Contextual Approach 247
14.3 Designing a Contextual Approach 250
14.4 Implementing a Contextual Approach 251
14.5 Conclusion 253
Index 255
1
Introduction
1.1 Rationale
As a reaction to the shortcomings of earlier pedagogies, Communica-
tive Language Teaching has come to dominate language education. In
fact, communicative methods are so well-established amongst language
educators and in teacher training programmes that it is difficult to voice
criticism of them. To teach communicatively is to be a progressive and
enlightened educator. Brown (2001: 44) asserts that ‘no one these days
would admit to a disbelief in principles of CLT; they would be marked as
a heretic.’ In fact, to question the validity of Communicative Language
Teaching can raise concerns about an educator’s ability to teach. Such
an ill-advised admission has the potential invoke capability procedures,
re-education, and, in the worst-case scenario, even redundancy.
The word ‘communicative’ is a strongly emotive concept in a similar
way to concepts such as freedom, justice, and truth. It embodies sharing,
mutual understanding, and tolerance. So much so that when applied to
language teaching, it is difficult to imagine that there could be any other
way. Harmer (1982: 164) notes that ‘no self-respecting teacher, materials
designer, or applied linguist would think of teaching English as anything
else.’
© The Author(s), under exclusive license to Springer Nature 1
Switzerland AG 2024
I. Pemberton, Usage-Based Second Language Instruction,
https://doi.org/10.1007/978-3-031-53414-0_1
2 I. Pemberton
It is often said that with the advent of Communicative Language

Teaching, we now live in a post-method era. Hall (2018: 66) notes that
‘this idea is now regularly discussed and examined in teacher training and
development texts.’ To live in a post-method era means not to adhere to
any particular method. But what about the Communicative Method?
Communicative Language Teaching proponents side-step the issue by
pointing out that CLT is an approach, not a method. Even in a post-
method era, this allows them to continue to subscribe to its doctrines.
Brown (2001: 42) tells us that ‘you can with some assurance latch on
to the CLT label and, like a member of a club, aver that you “believe
in CLT” and be allowed to step inside the gates.’ But such evangelical
fervour has a habit of dissuading educators from critically evaluating
their practice. Prabhu (1990: 164) argues that statements about living
in a post-method era ‘succeed in preserving the conversational peace but
cause a loss of the productive potential of professional debate.’ This begs
the question: Is Communicative Language Teaching the answer to our
pedagogical prayers? Or is it a vain faith that promises more than it can
deliver?
The first question is not so difficult to answer. In the twenty-first
century, second language learning is still far less successful than first
language learning. This implies that Communicative Language Teaching
is not the lost horizon we have been looking for, but a way-stage on
a journey. If the journey has a destination, it is still far off in the
distance. The second question is more difficult to answer. Communica-
tive Language Teaching is a phenomenon which covers a broad spectrum
of methods that differ from each other but share common features. This
makes it difficult to evaluate its efficacy. Any serious attempt to do so
has to identify the common features of its proliferation of methods. And
any evaluation of Communicative Language Teaching leads to a further
question. If CLT is not the acme of language teaching pedagogy, then
what, if anything, is?
1 Introduction 3
1.2 Outline
The main aims of this book are to build a case for change in the field of
language teaching and to explore in-depth how people learn languages
to better understand what that change might be.
The book begins by building the case for methodological change. It
examines in detail the history of English Language Teaching to firmly
ground its case in the accumulated knowledge and wisdom of those
scholars, educators, and innovators who have searched for answers to the
same questions in the past. It looks at what we can learn from the histor-
ical language teaching methods of the Classical, Reform, and Scientific
Periods, then evaluates the methods of the Communicative Period from
1970 up to the present. This analysis of the history of ELT considers
the benefits and drawbacks of each of the major methods and identifies
important elements of a language teaching approach.
At the same time, it identifies critical weaknesses in the current ortho-
doxy, Communicative Language Teaching, which cast doubt on whether
its methodology actually leads to fluency. Having built a case for change,
the book then considers the question: If CLT is not the hallowed ground
that language educators have been searching for, then what is?
Throughout history, innovative educators have looked to first language
acquisition to answer the question of how to teach second and addi-
tional languages. Thus, a natural starting place in a search for an answer
is to investigate theories of first language acquisition. The book exam-
ines evolutionary biology, making a distinction between primary and
secondary biological skills. It also looks at the classical first language
development theories of Constructivism and Sociocultural Theory, and
their influence on second language pedagogy. Then, it introduces Social-
Pragmatic Theory, a usage-based theory, which offers a cogent explana-
tion of first language acquisition.
Advocates of usage-based theories assert that they apply equally to
second language acquisition. To gain a more detailed understanding of
usage-based linguistics, the book examines its theoretical foundations.
First, it explores a general complexity theory known as Dynamic Systems
Theory. Then, it looks at constructionist grammars. After examining
4 I. Pemberton
these foundational theories, the book explores the usage-based account

of second language development and then surveys its empirical research.
The convincing account of second language development provided by
usage-based linguistics begs the question of how to realize the theory
in practice. This leads to the introduction of a contextual approach,
a twenty-first-century multimedia language learning pedagogy which
translates usage-based linguistics into classroom practice. A contextual
approach consists of six core principles which set out the preconditions
and processes necessary for second language acquisition. In the final part
of the book, the six principles are applied to the design of materials and
activities to realize the approach in the classroom.
A contextual approach represents a paradigm shift in language
teaching and learning. Given the power of the status quo and basic
human fear of the unknown, it may be unrealistic to think that this
shift will occur any time soon. But for those language educators brave
enough to acknowledge that their teaching isn’t as effective as they would
wish, and those curious enough and able to innovate, the book provides
a rationale and guidance on how to do so.
References
Brown, H. D. (2001). Teaching by principles: An interactive approach to language
pedagogy (2nd ed.). Prentice Hall Regents.
Hall, G. (2018). Exploring English language teaching: Language in action (2nd
ed.). Routledge/Taylor & Francis Group.
Harmer, J. (1982). What is communicative? ELT Journal, 36 (3), 164–168.
Prabhu, N. S. (1990). There is no best method—Why? TESOL Quarterly,
24 (2), 161–176.
Part I
Making the Case for Change
2
Historical Teaching Approaches
It may be tempting to think that as ELT moves from method to post-

method, teacher trainees no longer need to learn about the history of
teaching methods. However, as Kumaravadivelu (2006: xvii) observes
‘those who do not study history are condemned to repeat it.’ Historical
knowledge confers a number of benefits on prospective teachers. One
is that they can use methods as a source of ideas to inform their own
practice. According to Hunter (2009: 5), ‘the investigation of past events
within the ELT profession can recover similarly useful insights, and offer
a comparable “distillation” of experiences, which can be of real practical
value to present day ELT practitioners.’ Similarly, Hall (2018: 66) notes
that ‘they offer a source of options and practical classroom interventions.’
Richards and Rodgers (2001: 16) comment that experience of different
teaching methods provides teachers with a range of basic skills which
they can supplement as they gain experience. While historical methods
can represent a source of ideas for teachers, Pennycook (1989: 590)
cautions that ‘we must view critically all of the standard orthodoxies of
TESOL and investigate the interests served by such orthodoxies.’
In the belief that historical knowledge is beneficial for ELT practi-
tioners, this book begins with a brief history of language teaching which

Switzerland AG 2024
8 I. Pemberton
examines some of the antecedents of modern-day teaching methods

before investigating more contemporary methods in subsequent chap-
ters. The overview makes use of Howatt and Smith’s (2014) four
historical periods, but as it is not intended as a comprehensive history
of language teaching, it is selective in choosing key methods from each
era which have influenced the development of language teaching.
Howatt and Smith (2014: 78) divide the history of modern language
teaching into four periods:
1. 1750–1880: The Classical Period

2. 1880–1920: The Reform Period
3. 1920–1970: The Scientific Period
4. 1970–2000 +: The Communicative Period
2.1 The Classical Period

The first epoch in modern language teaching is referred to as the classical
period due to its adoption of methods for teaching the classical languages
of Latin and Greek (Howatt and Smith (2014: 79).
2.1.1 The Grammar-Translation Method
The major teaching pedagogy associated with the Classical Period has
been known by several different names. According to Brown (2001: 18),
until the nineteenth century, it was referred to as the Classical Method.
More recently, Cook (2008: 239) refers to it as the Academic Teaching
Style.
The Grammar-Translation Method has its roots in the study of
the Classics which were written in Latin and Ancient Greek, and the
methods used to study these texts were subsequently applied to the
teaching of foreign languages. According to Cook (2008: 239), its
syllabus consists of grammatical rules and vocabulary items. Instruction
takes place in the students’ mother tongue. In a typical class, the teacher
will go through a written text line-by-line with students, explaining
difficult vocabulary, problematic grammar, and providing cultural back-

ground where necessary. Students may complete some grammar-based
exercises before going on to translate the text into their own language.
They can then be assessed on their translations.
The Grammar-Translation Method has attracted criticism because it is
limited to written material and contains no real communication. There-
fore, by today’s standards, it is highly unlikely to produce students with
‘communicative competence.’ Richards and Rodgers (2001: 7) are partic-
ularly critical of the Grammar-Translation Method and state that ‘it is a
method for which there is no theory. There is no literature that offers
a rationale or justification for it or that attempts to relate it to issues in
linguistics, psychology, or educational theory.’
While some are dismissive of Grammar-Translation, several arguments
are made in its favour. Lewis (1997: 60) points out that one’s own
language is not something to be avoided at all costs but a resource and
translation into it is a useful learning strategy. Lopez-Sanchez (2009: 39)
points out that written texts place fewer cognitive demands on learners
than spoken ones and allow learners to focus on details. Similarly, Brown
(2001: 19) notes that it makes few demands on teachers, is easy to assess,
and develops reading skills in a second language. According to Cook
(2008: 240), the study of literature provides a basis for appreciating
a country’s culture, rather than a non-intellectual focus on practising
holiday English. A further point is that the use of continuous discourse
or whole texts contextualizes the language for learners which helps them
to both understand and remember it.
2.2 The Reform Period

The Reform Period is thus called due to the efforts of more progres-
sive language educators to teach modern languages such as French and
German as living languages (Bayley, 1998: 39). Hill (1967: 85) explains
that ‘when it became clear to certain people that it was very diffi-
cult to learn a language by working through the grammar, learning
rules and lists of exceptions, and doing translation exercises, a reac-
tion set in.’ Those intent on reforming language education argued that
10 I. Pemberton
modern languages should be taught as written and spoken communica-

tion. Howatt and Smith (2014: 82) explain that there were two similar
but separate methodological developments here: one in Europe where
innovative pedagogies were known as direct methods, and the other
in the United States where they were known as natural methods. The
defining feature of both direct and natural methods was that languages
were taught exclusively in the target language. In Europe, noteworthy
direct methods included Claude Marcel’s Rational Method, Thomas
Prendergast’s Mastery System, and Francois Gouin’s Series Method.
2.2.1 The Rational Method
While Claude Marcel is less well-known than Prendergast and Gouin, he

demands some attention due to the avant-garde nature of his method.
Strictly speaking, he proposed and practised his rational method prior to
Howatt and Smith’s Reform Period (he died in 1867), but his method
has more in common with the reformists than the methods of the Clas-
sical Period and is therefore included in this section. Howatt (1984: 153)
explains that he divided language skills into impression (listening and
reading) and expression (writing and speaking) or what we now term
receptive and productive skills. Marcel argued that impression psycho-
logically precedes expression, so should be taught first. He made a second
subdivision between spoken and written language which gave rise to four
branches of study or what has come to be known today as a four skills
approach: hearing, speaking, reading, and writing.
While Marcel prioritized impression overexpression, he advocated
reading before listening because the primary need of students of his day
was for reading skills. However, this tends to result in him being excluded
from consideration together with other reformists who prioritized spoken
language as a starting point for instruction.
Another innovative aspect of his method was the classification of
teaching methods into synthetic and analytic methods. A synthetic
method is one where students are taught the language separately in
parts (e.g., grammar, vocabulary, and pronunciation) and they have
to reassemble the parts. An analytic method is one where the learner
starts with the whole language which acts as a basis for imitation
and later analysis. This analysis was particularly ahead of its time,
and the terms re-emerged over a hundred years later in Dave Wilkins’
Notional-Functional Syllabus.
2.2.2 The Mastery System
In some respects, Prendergast was also ahead of his time. His Mastery
System was founded on his observations of young children. According
to Richards and Rodgers (2001: 7), he noticed ‘children use contex-
tual and situational cues to interpret utterances.’ Howatt (1984: 157)
explains that these cues included ‘what people do, how they look, their
gestures and facial expressions, and so on.’ Furthermore, he noticed
that children ‘learn ready-made “chunks” of language, “pre-fabs” as they
have been called in recent times, and weave them into their utterances.’
Prendergast mistook these formulaic sequences for sentences, and this
led him to conclude that ‘an efficient foreign language teaching system
would consist entirely of memorized sentences, practised to the point of
instant recall’ (Howatt, 1984: 158). This led to the development of ‘the
first “structural syllabus,” advocating that learners be taught the most
basic structural patterns occurring in the language’ (Richards & Rodgers,
2001: 7). When viewed retrospectively, we can see that his observa-
tions prefigured developments over a hundred years later in language
acquisition theory and teaching pedagogy.
2.2.3 The Series Method
A French language teacher, Francois Gouin, also made some percep-

tive observations which foresaw developments over a century later in
language teaching pedagogy. Like Prendergast, he too gained these
insights while observing native-speaking children (Richards & Rodgers,
2001: 7). According to Brown (2001: 20), Gouin came to the remark-
able insight that ‘language learning is primarily a matter of transforming
perceptions into conceptions. Children use language to represent their
12 I. Pemberton
conceptions. Language is a means of thinking, of representing the world

to oneself.’
The Series Method is thus called because it involves describing the
series of steps of an activity. For example, a learner might describe
washing a car with a series of statements such as: ‘First, we get a bucket,
then we fill it with water and soap. We must have either a sponge or a
rag. We use the soap and water to scrub the entire car’ (in Krashen &
Terrell, 1983: 10).
While the series method was overshadowed by its contemporaries, it
is given a high evaluation by Brown (2001: 20) who notes that it ‘taught
learners directly (without translation) and conceptually (without gram-
matical rules and explanations) a series of connected sentences that are
easy to perceive’ and it was a success ‘because the language was so easily
understood, stored, recalled and related to reality.’
2.2.4 The Direct Method
While individual educators were testing out their theories in Europe,

others were also trying out what were known as ‘natural methods’ in the
United States. The most important natural method, which was intro-
duced by Lambert Sauveur and Maximilian Berlitz, was known as the
Direct Method. Brown (2001: 21) explains that the central idea of this
method was that second language learning should resemble first language
learning. There should be a lot of unplanned speaking, no translation,
and no explicit teaching of grammar. However, Berlitz never referred to
the method used in his schools as the Direct Method, choosing instead
to call it the Berlitz Method.
The Direct Method has had its share of critics. Richards and Rodgers
(2001: 12) point out that it requires a strong command of the target
language, and so, defacto, it limits the role of teacher to native speakers.
A particularly vocal critic is Cook (2003: 34), who charges that in its
strong form, as practised in the Berlitz chain of language schools, a ban
on the use of mother tongue ‘emulated the most repressive monolingual
nations.’ The Direct Method continues to influence language pedagogy
today, although its repressive character takes a more subtle form. For
example, Brown (2001: 377) advises trainee teachers to prevent learners

from using their bilingual dictionaries. His rationale is ‘such practices
rarely help students to internalize the word for later recall and use.’ It
hardly needs to be pointed out that not checking a meaning is even less
likely to promote internalization.
2.3 Summary
A knowledge of the history of ELT is important for teachers. It provides
teachers with information about procedures and techniques that they can
use in their own practice. In addition, knowing the history of a field helps
to avoid repeating its mistakes. The history of ELT is generally divided
into four periods. The first period, the Classical Period, was dominated
by a single approach known as the Grammar-Translation Method. In
the Grammar-Translation Method, learners were taught the grammar of
a language in their mother tongue, and they were expected to translate
the target language into their first language. It has received strong criti-
cism from modern-day advocates of Communicative Language Teaching
because it does not teach ‘communicative competence’. However, its
benefits include lower cognitive demands for beginning students, the
ability to develop reading skill, and the ability to develop an appreciation
of foreign cultures. In addition, critics overlook the fact that a continuous
text creates an authentic context for the language it contains.
The second period, the Reform Period, was a very active historical era.
It was so-called due to attempts by more innovative educators to reform
the way languages were taught. There are four main methods associated
with this period and they are the Rational Method, the Mastery System,
the Series Method, and the Direct Method.
To begin with, Claude Marcel’s Rational Method proposed the idea
of four language skills, the division of language skills into impression
(reception) and expression (production), and that receptive skills should
precede productive skills. Thomas Prendergast’s Mastery System was
based on his observation that children watched what people did and their
facial expressions to interpret language. He was significantly ahead of
his time in noticing that children learnt ready-made chunks of language
14 I. Pemberton
and combined these to communicate messages. Francis Gouin’s Series

Method was also based on observations of native-speaking children. The
Series Method was so-called because it was based on sequences of related
actions which facilitated understanding without grammatical rules or
explanations.
The Direct Method was developed by language educators in the
United States. Like their European counterparts, proponents believed
that second language learning should be more like first language learning.
Thus, all instruction took place in the target language and without the
use of translation or grammatical explanation. The Direct Method has
been the target of several criticisms, the first being that a teacher needs a
strong command of the target language. In addition, in its worst incarna-
tions, it has been described as repressive, with teachers doing such things
as preventing students from using their dictionaries in the misguided
belief that this promotes better language learning.
2.4 Implications
A number of points can be distilled from historical teaching methods.
The drawbacks and benefits of each method give an indication of
what language educators might include in an approach, and what they
might avoid. The innovators of the Reform Period observed that the
Grammar-Translation Method did not imbue learners with the ability to
communicate. They recognized that first language acquisition was vastly
more successful than second language learning and developed innovative
methods which drew upon features of first language learning. In their
view, an effective language teaching method would:
• emphasize listening
• map language onto real-world events
• teach prefabricated chunks
• focus on use
All of these features were precursors to later “innovations” in language

teaching.
References
Bayley, S. N. (1998). The Direct method and modern language teaching in
England 1880–1918. History of Education, 27 (1), 39–57.
Cook, G. (2003). Applied linguistics. Oxford University Press.
Cook, V. (2008). Second language learning and language teaching (4th ed.).
Hodder Education.
ed.). Routledge/Taylor & Francis Group.
Hill, L. A. (1967). Selected articles on the teaching of English as a foreign language.
OUP.
Howatt, A. P. R. (1984). A history of English language teaching. Oxford
University Press.
Howatt, A. P. R., & Smith, R. (2014). The history of teaching english as
a foreign language, from a British and European perspective. Language &
History, 57 (1), 75–95.
Hunter, D. (2009). Communicative language teaching and the ELT Journal: a
corpus-based approach to the history of a discourse. Ph.D. thesis, University of
Warwick.
Krashen, S. D., & Terrell, T. D. (1983). The natural approach: Language
acquisition in the classroom. Pergamon.
Kumaravadivelu, B. (2006). Understanding language teaching: From method to
post-method . Lawrence Erlbaum Associates.
Lewis, M. (1997). Implementing the lexical approach. Thomson Heinle.
Lopez-Sanchez, A. (2009). Re-writing the goals of foreign language teaching:
The achievement of multiple literacies and symbolic competence. Interna-
tional Journal of Learning, 16 (10), 29–38.
Pennycook, A. (1989). The concept of method, interested knowledge, and the
politics of language teaching. TESOL Quarterly, 23(4), 589–618.
Richards, J. C., & Rodgers, T. S. (2001). Approaches and methods in language
teaching (2nd ed.). Cambridge University Press.
3
The Scientific Period
The Scientific Period is thus called because it was an era during which
there was a trend for linguists to justify their methods in terms of
‘the new social sciences: particularly linguistics but also, increasingly,
learning theory derived from psychology’ (Howatt & Smith, 2014: 85).
The major method that emerged during this era was the Audio-Lingual
Method. While not strictly a teaching method, an analysis of Noam
Chomsky’s Universal Grammar is included here as it is generally consid-
ered to have had a direct influence on the emergence of Communicative
Language Teaching.
3.1 The Structural Syllabus

Structural syllabi emerged around the middle of the twentieth century.
Richards (2017: 41) comments that there were attempts to draft struc-
tural syllabi from the 1920s onwards, and by the 1940s, beginners’
courses started to appear which were based on carefully sequenced gram-
matical structures (ibid.: 44). Wilkins (1976: 56) explains that the desire
to produce a structural syllabus was driven by the view that the most

Switzerland AG 2024
18 I. Pemberton
useful patterns in a language could be identified and taught, and then

learners would be able to apply them widely in different situations.
Over time, the structural syllabus became the dominant way to organize
language teaching materials. In 1967, Hill remarks that ‘the structural
syllabus is now familiar in many parts of the world’ (p. 115). In 1976,
Wilkins notes that the majority of commercially produced courses are
based on a structural syllabus (p. 7). In 1983, Yalden comments that
the structural syllabus is very familiar to teachers and has been used to
instruct generations of learners (p. 26).
A structural syllabus is an inventory of syntactic structures. Yalden
(1983: 19) notes that a structural syllabus may also be accompanied by a
lexicon (a list of vocabulary). The structures are arranged in a sequence.
Hill (1967: 115) comments that structures ‘are graded according to
what the authors of the syllabus consider to be the relative ease, utility
and teachability.’ Once items are sequenced, they can represent levels:
beginner, intermediate, advanced, and so on, with each level being taught
different structures (Yalden, 1983: 19). The syntactic structures of a
structural syllabus can take a variety of forms. They can be sentence
patterns, or they can be grammar or function words such as ‘and’ or
‘but’ (Hill, 1967: 115). Hall (2018: 221) adds that structures may also
be grammatical features such as tense, modal verbs, or relative clauses.
Structural linguists theorize that a language is a system of hierarchi-
cally related structures which encode meaning. Kumaravadivelu (2006:
99) comments that this system consists of related elements at differing
levels of granularity: phonemes, phrases, clauses, and sentences. These
elements are connected in a linear manner and governed by rules.
Richards and Rodgers (2001: 20) add that the system also includes gram-
matical operations such as ‘adding, shifting, joining, or transforming
elements.’ Structural linguists also theorize that you learn a language
through mastery of its grammar. Wilkins (1976: 18) states that ‘the
grammatical syllabus seeks to teach the language by taking the learner
progressively through the forms of the target language.’ Similarly, Yalden
(1983: 20) comments that the assumption was that it was necessary to
master the grammatical system of a language to learn it. Hill (1967: 117)
praises the structural syllabus, noting that it ‘has been of inestimable
value in showing us the great importance of patterning in language
learning. It freed us from the obsession with words by showing us that

even more important than the words themselves are the ways in which
they are arranged together into larger units.’
However, a major issue with structural syllabi is thought to be their
lack of focus on meaning. Hall (2018: 221) makes the point that major
developments in the field on ELT such as Communicative Language
Teaching and Task-Based Instruction have resulted in recognition of the
importance of meaning, but this has had little impact on ELT course-
books which still use structures as an organizing principle. Wilkins
(1976: 10) questions the adequacy of structural syllabi, noting that even
after teachers have explained the grammatical meaning of a sentence,
they have not taught learners how to use it. Yalden (1983: 28) expands
on this point, explaining that when we teach the grammatical meaning
of a sentence, what has been taught is the relations that exist between
elements in the sentence but not their relationship to other elements
outside of the sentence in other parts of the discourse. Another issue
is the way structures are chosen for a syllabus. Richards (2017: 41)
makes the point that creators of vocabulary lists analysed texts and made
frequency counts to determine the most important words. However,
creators of lists of structures tended to rely on their intuitions to choose
structures based on criteria such as teachability and learnability. A further
issue, highlighted by Hall (2018: 221), is that structural syllabi are not
consistent with what is known about learners’ internal syllabi and order
of acquisition of grammatical structures. That is to say, some structures
are taught early but are actually acquired much later.
3.2 The Situational Syllabus

The situational syllabus was developed in the first half of the twen-
tieth century. However, Yalden (1983: 34) comments that ‘it appears to
run through the history of second language teaching, turning up in the
Renaissance, in the work of Erasmus and of Comenius.’ Howatt (1984:
272) explains that a situational syllabus is founded on the premise that
‘the meaning of an utterance is a function of the cultural and situational
context in which it occurs.’
20 I. Pemberton
A situational syllabus is an inventory of situations in which learners

will use the target language. Wilkins (1976: 18) explains that the syllabus
designer’s job is to recreate the situations in which native speakers use the
language. He provides typical examples such as visiting the post office,
purchasing a ticket for the theatre, and asking for directions. Yalden
(1983: 35) adds purchasing an airplane ticket and taking part in a job
interview. Wilkins (1976: 16) recommends a number of parameters for
defining situations. These features include:
• the physical context of the language event

• whether the language is spoken or written
• whether it is productive or receptive
• who the participants are, and what their relationships are
• the field of activity in which the event occurs
There are a number of arguments in favour of using a situational

syllabus. Wilkins (1976: 16) explains that:
• language cannot be fully understood without reference to its social

context
• language choices may be restricted by the context, for example, the use
of formal or informal language
• learners need to use specific language in specific situations
• a situational approach is more efficient because it includes only that
language which is relevant to the learner
• the utility of the language presented and practised increases learner
motivation
• the use of a situational approach means that the distinction between
language for learning and language for use disappears
Wilkins (1976: 16) points out that it is not easy to define all situa-
tions in which learners will need to use language or their language forms.
He (ibid.: 17) makes the point that while learners have agency over the
language that they produce, they have no control over what the other
participants will say. While he concedes that a situational approach might
be adequate for the limited needs of a tourist, waiter, or telephonist, it
is not adequate for a general language learner. Mackey (1977) cited in

Yalden (1983: 38) makes the point that over-contextualization means
that the language learnt cannot necessarily be generalized to other situa-
tions. Yalden (1983: 39) expands on this point, explaining that ‘making
complaints, requests, seeking information, agreeing and disagreeing, are
not situationally determined language functions and may be encountered
in any context.’ She argues that the situational approach therefore cannot
meet its claim to prepare learners for real life.
3.3 The Audiolingual Method

The first half of the twentieth century saw two world wars with the
participation in the Second World War of the American army. This
created a need for translators and, in the 1950s, the US military
asked language specialists to develop an intensive language learning
programme that came to be known as the Army Specialized Training
Programme (ASTP) or more commonly, the ‘Army Method.’ The Army
Method created a substantial amount of interest amongst language
educators which led subsequently to the development of Audiolin-
gualism. Richards and Rodgers (2001: 57) explain that Audiolingualism
drew upon the principles of the American psychologist B.F. Skinner’s
theory of Behaviourism. This theory considers that language learning is
essentially a process of mechanical habit formation. According to Kumar-
avadivelu (2006: 100), behaviourists regard language as a response to a
stimulus which, through reinforcement, becomes a routine.
Audiolingualism considers that the main purpose of language teaching
is to help students learn structural patterns. Thus, the underlying syllabus
was a structural syllabus. A typical audiolingual lesson followed the
Present-Practice-Produce paradigm. Activities occurred in a set order.
Larsen-Freeman (2000: 46) explains that ‘the natural order of skills
presentation is adhered to: listening, speaking, reading, and writing. The
oral/aural skills receive most of the attention.’
In the presentation phase, learners hear a tape recording or reading of
a spoken dialogue, and this provides a context for a target structure. The
teacher then asks learners to repeat each line of the dialogue in order to
22 I. Pemberton
memorize it, and then, they may act it out (Kumaravadivelu, 2006: 106).
After the presentation phase, the class moves on to the practice phase. In
Audiolingualism, language learning is a process of habit formation: repe-
tition leads to greater learning. Therefore, a basic feature of the controlled
practice phase is the extensive use of drills. Kumaravadivelu (2006: 102)
citing Hockett (1959) comments that learners ‘require drill, drill, and
more drill, and only enough vocabulary to make such drills possible.’
These drills can include ‘repetition, backward build-up, chain, substitu-
tion, transformation, and question-and-answer’ (Larsen-Freeman, 2000:
44). A key feature of Audiolingualism is zero-tolerance of errors. As errors
are thought to lead to the formation of bad habits, the teacher seeks to
eliminate them when they occur. Additionally, a class may use a language
laboratory for practice in rhythm, stress, and intonation. Work in the
language lab is then followed by reading and writing practice. Finally, in
the production stage, learners roleplay similar dialogues. Kumaravadivelu
(2006: 106) comments that ‘they are supposed to modify the language
they have memorized in order to vary their production.’
Audiolingualism enjoyed some popularity with both teachers and
learners because it provided a well-structured learning environment.
Kumaravadivelu (2006: 113) notes that it was ‘a theory-driven, system-
atically organized, and teacher-friendly language-centered pedagogy.’
According to Cook (2008: 246), teachers were clear about what they
had to do, and learners were able to relax as they took part in familiar
classroom activities and knew what they were expected to do.
While its goals were practical and communicative, its results were
disappointing. Cook (2008: 244) asserts that its main objective was to
get students to communicate in the real-life situations they were going to
experience such as interacting at the railway station or in a supermarket.
However, it failed to achieve its objectives. Richards and Rodgers (2001:
65) point out that ‘students were often found to be unable to transfer
skills acquired through Audiolingualism to real communication outside
the classroom.’ Krashen and Terrell (1983: 13) explain that ‘methods in
which students never engage in real communication cannot be expected
to produce students able to communicate using the language they study.’
Another criticism is that the nature of its activities failed to engage
learners. Krashen and Terrell (1983: 14) state that teachers and learners
regarded its activities as ‘excruciatingly boring.’ Similarly, Richards and

Rodgers (2001: 65) report that ‘many found the experience of studying
through audiolingual procedures to be boring and unsatisfying.’ Ellis
(1990) cited in Kumaravadivelu (2006: 111) makes the point that ‘even
learners who were “motivated” to persevere found that memorizing
patterns did not lead to fluent and effective communication in real-life
situations.’
A further problem for Audiolingualism was that its language and
learning theories were eventually overtaken by new developments in
Psychology and Linguistics. Kumaravadivelu (2006: 109) points out that,
in contrast to grammar-translation, it ‘was based on well-articulated and
well-coordinated theories of language, language learning, and language
teaching, prompting its proponents to call it a “scientific” method.’
However, Richards and Rodgers (2001.65) comment that the language
and learning theories of Audiolingualism were pedagogically unsound.
While its methods, pattern practice, drills, and memorization led to
language-like behaviours, they did not result in competence.
3.4 Universal Grammar

While neither an approach nor a method, Universal Grammar had
a considerable influence on language education in the 1960s. Its
author, Noam Chomsky, first came to prominence after his critique
of B.F. Skinner’s behaviourist model of language learning as a stim-
ulus–response-reinforcement mechanism. Chomsky (1959) argued that
this was far too simplistic to describe the creative nature of language.
Chomsky argued that languages are so complex that if grammar were
not an innate property of the human mind, then learning one would
be impossible. He hypothesized that during human evolution, we have
developed ‘a genetically-based universal grammar’ (Tomasello, 2003: 9),
and all languages have this common underlying structure. A Universal
Grammar consists of general principles which are true of all languages
and parameters which vary within narrow limits (Cook & Newsom,
1996: 2). A minimal amount of language input to a child causes the
parameters to be set one way or another (Ellis, 1997: 32).
24 I. Pemberton
Notwithstanding, the abstruse nature of his theory, Howatt (1984:

271) argues that Chomsky’s impact on the language teaching field was
significant as it re-emphasized the importance of cognition in language
learning. However, he admits that ‘the influence of transformational-
generative grammar on language teaching, for example, in the prepa-
ration of teaching materials, was limited.’ Noting that none of the
principles or parameters of Universal Grammar have been found to be
common to all languages, American developmental and comparative
psychologist, Michael Tomasello (2003: 304) is dismissive in his assess-
ment of its significance, wondering: ‘Why do we need the phlogiston/
ether of universal grammar … at all? What is it doing anyway? Why not
just chuck it?’.
3.5 Communicative Competence

In accounts of the history of language teaching, the literature usually cites
Chomsky as a stimulus for the appearance of communicative method-
ology. In his Universal Grammar, Chomsky made a distinction between
linguistic competence and performance. He defined linguistic compe-
tence as the knowledge of the grammar of a language that a person
has, and performance as actual language use (Cook & Newsom, 1996:
22). In his writings, Chomsky (1965: 2) stresses the importance of
understanding linguistic competence, and at the same time trivializes
performance.
Taking issue with Chomsky’s concept of linguistic competence, Amer-
ican linguist Dell Hymes introduced the concept of ‘communicative
competence’ as a worthy goal of foreign language education in schools.
In his words, he stated that what he believed was crucial was ‘not so
much a better understanding of how language is structured, but a better
understanding of how language is used’ (cited in Howatt, 1984: 271).
According to Canale and Swain (1980: 16), Hymes’ (1972) notion of
communicative competence consisted of four types of knowledge:
1. whether (and to what degree) something is formally possible

2. whether (and to what degree) something is feasible in virtue of the
means of implementation available
3. whether (and to what degree) something is appropriate (adequate,
happy, successful) in relation to a context in which it is used and
evaluated
4. whether (and to what degree) something is in fact done, actually
performed, and what its doing entails
In other words, communicative competence is the outcome of the

grammatical, psycholinguistic, sociocultural, and probabilistic systems
of competence. This prompted Canale and Swain (1980: 29) to
propose their own more detailed communicative competence framework.
Initially, it had three competencies but, according to Skehan (1998: 157),
Canale later extended it to four:
1. Grammatical competence which includes morphological, syntactic,

grammatical, semantic and phonological rules, and lexical knowledge
(Canale & Swain, 1980: 29).
2. Sociolinguistic competence which is knowledge of the use of utter-
ances in different contexts as it relates to factors such as participants
status, interaction purpose, and interaction norms (Canale, 1983,
cited in Kumaravadivelu, 2006: 17)
3. Discourse competence which is how a series of sentences or utterances
coheres to form a spoken or written text (Kumaravadivelu, 2006: 17).
In contrast to grammatical and sociocultural rules, discourse rules
refer to concepts such as topic and comment (Canale & Swain, 1980:
30)
4. Strategic competence is a compensatory system which is drawn upon
when a learner’s interlanguage lacks a word or structure (Skehan,
1998: 158).
It is often said that the significance of Hymes’ proposals for commu-

nicative competence cannot be understated. Savignon (2017: 1) states
26 I. Pemberton
that ‘the introduction of communicative competence as a guide for

the teaching and evaluation of learners proved nothing short of revo-
lutionary.’ The advent of communicative competence offered new hope
that the shortcomings of grammatical and structural teaching approaches
could be overcome. According to Widdowson (1972, cited in Howatt,
1984: 287), learners in developing countries remained unable to under-
stand or use English even after years of formal instruction. Howatt
(1984: 287) states that ‘the original motivation for adopting a commu-
nicative approach in the early seventies was remedial, an attempt to
overcome the inadequacies of existing structural syllabuses, materials,
and methods.’ Similarly, Cook (2003: 47) comments that ‘inspired by
Hymes, the communicative approach … aimed to develop learners’
capacity to use the language effectively.’
However, a word of caution is in order. Ellidokuzoglu (1996) points
out that learners’ main problems in acquiring a target language may not
be remedied by teaching sociological and discourse competencies as such
competencies may have already been learnt during first language acqui-
sition. A learner may already be aware that the same language can have
different meanings in different circumstances and be able to organize
texts in order to communicate. In which case these things don’t need to
be taught. In relation to this, the communicative competence framework
gives each competence a similar status. However, an observant educator
will notice that lexical knowledge, something much more fundamental
than sociolinguistic, discourse, or strategic competence, is included only
as a component of grammatical competence.
3.6 Summary
The third historical epoch in English language teaching is known as the
Scientific Period. It was so-called because of efforts by educators to link
language teaching methodology to scientific theory. A major milestone
in the Scientific Period was the development of the structural syllabus.
A structural syllabus is an inventory of syntactic structures chosen on
the basis of learnability, utility, and teachability. Structural items can be
such things as sentence patterns, grammar words, tenses, modal verbs,
or relative clauses. These can be arranged into levels and taught in

sequence. Mastery of the structures at one level results in progression
to the next level. The structural syllabus has been described as being of
inestimable importance to language educators in focusing on patterns in
language. However, a major criticism of it has been made that it focuses
on the grammatical meaning of sentences but not how to use them for
communicative purposes. Nevertheless, it is still widely employed as the
organizing principle of a majority of ELT coursebooks. Another criti-
cism is that, unlike vocabulary lists which are based on empirical analysis,
structural syllabi tend to be based on the intuitions of the writers.
As a response to these criticisms, educators developed a different
type of syllabus known as a situational syllabus. As its name suggests,
this type of syllabus was based on an inventory of situations in which
language is used. Situations were chosen on the basis of such things
as physical context, modality, and the participants. A number of argu-
ments were made in favour of situational syllabi, for example, language
can’t be understood without relating it to its social context, situations
restrict language use, and the distinction disappears between language
for learning and language for use. A variety of criticisms have been made
of situational syllabi. These include situations not being easy to define,
and while the approach gives the learners control over what they say,
they have no similar control over what the other participants in a situa-
tion might say to them. In that sense, the approach doesn’t prepare them
for real life.
The 1960s saw the arrival of another major language teaching mile-
stone, the Audiolingual Method. This was originally developed by the
US Army after the Second World War to provide military personnel with
translation skills. It was based on B.F. Skinner’s theory of Behaviourism,
and considered language learning to be the formation of habits through
the process of stimulus, response, and reinforcement. Its main objective
was to teach structural patterns, so it was based on a structural syllabus.
Furthermore, it adopted the present-practice-produce paradigm. Some
of the benefits of the Audiolingual Method were that it was highly struc-
tured, and so teachers and learners always knew what they were supposed
to be doing. Importantly, it was underpinned by theory. However, its
results were disappointing, and it has received some severe criticism.
28 I. Pemberton
Specifically, even for motivated learners its drills could be boring, and
there was no engagement in real communication, so it did not prepare
learners for real life.
The 1960s also saw the arrival of Noam Chomsky’s Universal
Grammar. Chomsky came to prominence when he criticized the
behaviouristic view of language learning for being too simplistic. He
argued that the difficulty of learning a language was so great that it
could only be achieved if grammar were an innate property of the
human mind. He theorized that evolution had resulted in a genetically-
based grammar which contained principles which were common to all
languages and parameters which varied within a narrow range. Only
minimal language input was necessary to set the parameters. Chomsky’s
Universal Grammar was praised for re-focusing attention on human
cognition, but it has had a negligible impact on language teaching. It
has also been noted that no principle or parameter is common to all
languages.
In his Universal Grammar, Chomsky made a distinction between
linguistic competence and performance, valorizing competence and
dismissing performance. As a reaction to this, American linguist
Dell Hymes argued that communicative competence, not linguistic
competence should be the objective of school language education.
He considered competence to be based not only on grammatical,
but also psycholinguistic, sociocultural, and probabilistic competen-
cies. His concept of communicative competence was developed into a
more detailed framework by Canale and Swain with four categories,
namely: grammatical, sociolinguistic, discourse, and strategic compe-
tence. However, teaching intercultural competencies may be unnecessary
and may take up time which could be used to learn language.
3.7 Implications
The strengths and weaknesses of the methods of the Scientific Period
further indicate what language educators might embrace in a method and
what they might eschew. The structural syllabus was praised for focusing
attention on language patterns but criticized for not dealing with
meaning. The Situational Syllabus contextualized language but didn’t

provide learners with enough opportunities for processing language
receptively. The Audiolingual Method was praised for linking theory
and practice but criticized for its mindless repetition and its lack of
real communication. And while Communicative Competence focused
attention on communication, it deemphasized the importance of lexical
development which is a foundational element of language proficiency.
The implications of this analysis of the methods of the Scientific Period
are similar to those of the Reform Period. Important elements of a
method are likely to include:
• linking language to its situations of use

• affording opportunities to process language
• drawing attention to language patterns
• emphasizing vocabulary development
• providing opportunities for productive use
References
Canale, M., & Swain, M. (1980). Theoretical bases of communicative
approaches to second language teaching and testing. Applied Linguistics, 1,
1–47.
Chomsky, N. (1959). Review of B. F. Skinner’s verbal behavior. Language, 35,
26–58.
Chomsky, N. (1965). Aspects of the theory of syntax. M.I.T. Press.
Hodder Education.
Cook, V., & Newsom, M. (1996). Chomsky’s universal grammar: An introduction
(2nd ed.). Blackwell.
Ellidokuzoglu, H. (1996). Decline and fall of communicative approach. KHO
International ELT Conference: From diversity to synergy, Ankara, Turkey.
Ellis, R. (1997). Second Language Acquisition. Oxford University Press.
30 I. Pemberton
Hall, G. (2018) Exploring English language teaching: language in action (2nd

ed.). Routledge/Taylor & Francis.
Oxford University Press.
Howatt, A. P. R. (1984) A history of English language teaching. Oxford
University Press.
Howatt, A. P. R., & Smith, R. (2014). The history of teaching English as
History, 57 (1), 75–95.
Kumaravadivelu, B. (2006). Understanding language teaching: From method to
post-method . Lawrence Erlbaum Associates.
Larsen-Freeman, D. (2000). Techniques and principles in language teaching (2nd
ed.). Oxford University Press.
Richards, J. C. (2017). Curriculum development in language teaching (2nd ed.).
Cambridge University Press.
Savignon, S. (2017) Communicative competence. In J. I. Liontas (Ed), The
TESOL encyclopedia of English language teaching. Wiley.
Skehan, P. (1998). A cognitive approach to language learning. Oxford University
Press.
Tomasello, M. (2003). Constructing a language: A usage-based theory of language
acquisition. Harvard University Press.
Wilkins, D. A. (1976). Notional syllabuses: A taxonomy and its relevance to
foreign language curriculum development. Oxford University Press.
Yalden, J. (1983). The communicative syllabus: Evolution, design and implemen-
tation. Pergamon.
4
The Communicative Period
Following on from the Scientific Period, there was a marked shift in how
educators regarded language—from a system of rules which could be
systematically introduced and taught to the perspective that the main
function of language was communication, and therefore this should be
the focus of classroom activity. Thus, Howatt and Smith (2014) refer to
the era from 1970 to the present as ‘the Communicative Period.’
Unlike most historical descriptions, this chapter does not deal with
the designer methods of the 1970s (Total Physical Response, The Silent
Way, Community Language Learning, and Suggestopedia). Although
interesting, these methods were limited in scope and while elements of
each method can be used as classroom techniques, they do not represent
complete solutions to the language learning conundrum.
The following chapters are not simply intended as a description of the
history of modern ELT methods as a number of these already exist (e.g.,
Cook, 2008; Howatt & Smith, 2014; Howatt, 1984; Larsen-Freeman &
Anderson, 2011; Richards & Rodgers, 2001). Their purpose is to analyse
modern ELT methods to assess their efficacy and identify important
elements of an effective language teaching approach.

Switzerland AG 2024
32 I. Pemberton
Along the way, it also raises important questions about each of the
major methods. These include:
• While the teaching of functions is a key feature of communicative

teaching, what weakness of his Notional-Functional Syllabus did Dave
Wilkin’s highlight, but CLT proponents overlook?
• If some CLT advocates consider Steven Krashen’s Natural Method
to be the antithesis of communicative teaching, why do historians
include it in the Communicative Period. And is it communicative or
not?
• After a half-century of Communicative Language Teaching is it the
acme of language education? If not, how has it maintained its domi-
nance in the field?
• While some regard Content-Based Instruction as an unalloyed good
thing, Merrill Swain is well-known for drawing attention to its short-
comings. But was her solution the right response?
• If Task-Based Instruction has been around for around four decades,
how can it still be innovative?
• If the principles of Michael Lewis’s Lexical Approach are widely
accepted by language educators, why has it not become a mainstream
approach?
4.1 The Notional-Functional Syllabus

The 1970s proved to be a very active period of innovation for English
Language Teaching. An important development which had a major influ-
ence on other emerging pedagogies was the notional-functional approach
to syllabus design. This was introduced to the language teaching commu-
nity in 1976 by British academic David Wilkins in his seminal work
Notional Syllabuses. The book rediscovers the distinction made more
than 100 years earlier by Claude Marcel between synthetic and analytic
syllabi. A synthetic approach is one which teaches different language
elements separately in a step-by-step manner. A learner gradually puts
together the whole language structure (Wilkins, 1976: 2). Yalden (1983:
22) comments that a synthetic approach results in a structural syllabus.
On the other hand, an analytic approach can be taken to syllabus
design. This is a more holistic approach where a language educator
identifies the circumstances in which learners will use a language and
then considers the linguistic forms that they need. It produces either a
situational or a notional syllabus. According to Wilkins (1976: 16), a
structural syllabus considers language to be a general system, but in fact,
language choices are always linked to social context and those choices
are likely specific to the situation, so a situational syllabus is a more
profitable approach. As discussed earlier, this involves identifying the
situations where native speakers use language which results in a more
relevant learner-centred syllabus.
Wilkins (1976: 16) explains the shortcomings of a situational syllabus,
noting that it is limited to physical situations with easily identifiable
language interactions and associated language forms. While these might
be adequate for a person with limited aims such as a tourist, waiter, or
switchboard operator, a situational syllabus takes away the agency from
the language learner and what they want to communicate, and this might
not be language that is readily identifiable with the particular situation.
Similarly, the reactions of other participants in the situation cannot be
perfectly predicted, and the situational syllabus does not provide the
learner with a means of coping with such situations.
In place of either a structural syllabus or a situational syllabus, Wilkins
(1976: 18) proposes a notional syllabus. According to Wilkins, a notional
syllabus focuses not on how learners communicate nor where they
communicate but on what they communicate. In other words, it is a
content-based syllabus. However, the content is general and thus appli-
cable to different situations. Hall (2018: 221) describes this as a syllabus
design with a meaning focus. Wilkins (1976: 21) sets out three cate-
gories of meaning which he labels: semantico-grammatical, modal, and
communicative functions.
34 I. Pemberton
4.1.1 Semantico-grammar
This category includes common semantic concepts with a high utility

such as time, quantity, space, relational meaning, and deixis. These are
likely universals that are common to different cultures and languages.
Category Concepts
Time Points in time, duration, time relations (e.g., past, present,
and future), sequence, and frequency
Quantity Countable and uncountable items, numerals, and
mathematical operations
Space Dimensions, locations, and motion
Relational 1. Semantic roles such as agent, initiator, object,
meaning beneficiary, and instrument
2. Prediction and attribution such as John is a pilot. A man
with a long face
Deixis The time, place, and person involved in the communicative
act
4.1.2 Modal Meaning
This category represents interpersonal meaning. That is to say, a speaker’s

beliefs or feelings about something. Wilkins (1976: 38) divides it into
two different scales: language for expressing certainty and language for
expressing commitment.
The scale of certainty can be personalized or impersonalized. To be
personalized means that the individual is expressing their own view on
the likelihood of something happening and this includes language for
expressing conviction, conjecture, doubt, and disbelief. To be imperson-
alized means that individual is not expressing their own attitude, and so
this requires different language forms.
The scale of commitment can be divided into intention and obli-
gation. Intention refers to such things as wants, wishes, desires, plans,
promises, and assurances. Obligation refers to duties, liabilities, and
responsibilities.
4.1.3 Communicative Functions
This is a large category which reflects the huge diversity of communica-

tive functions. It includes judgement and evaluation, suasion, argument,
rational enquiry and exposition, personal emotions, and emotional rela-
tions.
Category Functions
Judgement and evaluation valuation, veridiction, approval, and
disapproval
Suasion inducement, compulsion, prediction,
tolerance
Argument assertion, seeking information, denial,
agreement, disagreement, concession
Emotional relations greeting, expressing sympathy,
gratitude, flattery, hostility
A number of criticisms have been made of a notional-functional

approach. Hall (2018: 221) points out that nowadays, pure notional-
functional syllabuses are not common because of the difficulty involved
in identifying and teaching the numerous functions of English. After
arguing eloquently for their adoption, Wilkins (1976: 75) himself points
out that in a notional syllabus, lexical items play an incidental role, and
that it can be difficult to specify the precise lexical content. Importantly,
he (ibid:78) also points out that notional syllabuses work on productive
skills but neglect receptive skills: ‘whereas the individual is the master of
what he himself chooses to say, he can exercise no comparable control
over the language he hears.’ His partial solution to this problem was to
recommend the use of authentic materials (ibid:79).
4.2 The Natural Approach

One highly influential second language acquisition theory to emerge
during the 1970s was Steven Krashen’s Natural Approach (Randall,
2007: 151). It is important to note that this approach was not some-
thing totally new but revisited a central idea of the Direct Method—that
36 I. Pemberton
we could teach a second language by drawing upon the processes that

learners use for their first language (Cook, 2003: 35).
The Natural Approach is based upon five hypotheses known collec-
tively as the ‘Input Hypothesis’ (Krashen, 1985: 1). These hypotheses
are as follows:
1. the Acquisition-Learning Hypothesis

2. the Natural Order Hypothesis
3. the Monitor Hypothesis
4. the Input Hypothesis
5. the Affective Filter Hypothesis
4.2.1 The Acquisition-Learning Hypothesis
The Acquisition-Learning Hypothesis states that: “there are two inde-

pendent ways of developing ability in second languages. ‘Acquisition’ is
the subconscious process identical in all important ways to the process
children utilize in acquiring their first language, while ‘learning’ is a
conscious process that results in ‘knowing about’ the language (Krashen,
1985: 3)”.
This distinction has been one of the most contentious aspects of
Krashen’s theory. While most people identify the distinction between
‘acquisition’ and ‘learning’ with Krashen, it was made by Pit Corder
about a decade earlier. In his book Introducing Applied Linguistics, Corder
(1973: 107) defines language acquisition as a process which ‘takes place
in the infant and the young child at a time when he is acquiring many
other skills and much other knowledge about the world.’ He defines
language learning as the process of learning a second language which
‘normally starts at a later stage, when language performance has actually
become established and when many other physical and mental processes
of maturation are complete or nearing completion.’
4.2.2 The Natural Order Hypothesis
The Natural Order Hypothesis argues that people learn grammar rules
in a particular order, irrespective of the order in which items are taught.
Some simple items, such as third person singular and plurals, emerge late,
which means that learners only master them after years of study even if
they are taught early on.
4.2.3 The Monitor Hypothesis
This hypothesis states that knowledge learnt in the classroom such as

explicit knowledge of grammatical rules cannot become acquired knowl-
edge. This means that it cannot be proceduralized and used automatically
without conscious thought. Krashen (1985: 2) theorized that learners
could only use the monitor when they had enough time to reflect on
their production, such as when writing. He proposed that other two
conditions had to be met: the performer knows the rule and is concerned
with accuracy.
4.2.4 The Input Hypothesis
The central component of the theory, the Input Hypothesis, claims that
language is learnt only by understanding comprehensible input. The
hypothesis argues that the input should be at the level i + 1, which
means at a level slightly higher than a learner’s existing language level.
Krashen (1985: 2) explains that ‘we are able to understand language
containing unacquired grammar with the help of context, which includes
extra-linguistic information, our knowledge of the world, and previously
acquired linguistic competence.’
38 I. Pemberton
4.2.5 The Affective Filter Hypothesis
The Affective Filter Hypothesis argues that learners need to be open

to comprehensible input to acquire it. If they have sufficient motiva-
tion and are not concerned about making mistakes, they are likely to
acquire language. On the other hand, if they are unmotivated, lack
self-confidence, or anxious, they are unlikely to acquire.
Krashen’s theory appealed strongly to the intuitions of many language
educators, particularly those who had taught grammar syllabi but seen
little improvement in their students. However, he has come under
sustained criticism from fellow academics. A particular target for criti-
cism was his Acquisition-Learning Hypothesis. It was pointed out that
the difference between the two concepts was vague. On the other hand,
the Natural Order Hypothesis, which has been demonstrated by research
and fits with language educators’ experience, is generally accepted.
Another target for criticism was the Monitor Hypothesis. His non-
interface position that learning could not become acquisition meant
that even where grammar rules were formally taught, learners still had
to acquire them through exposure to subsequent comprehensible input.
This implied that the work of many language teaching professionals was
a largely futile exercise.
The Input Hypothesis—the idea that a large amount of input is neces-
sary for language acquisition—is widely accepted. This is certainly the
experience and intuition of language learners worldwide. However, there
has been some debate over the importance of output. Krashen argued
that listening and reading were of primary importance in a language
programme, and that ‘speaking is not absolutely essential for language
acquisition’ (Krashen & Terrell, 1983: 56). Some educators expressed
doubts that speaking was the result of acquisition and not its cause.
In her Output Hypothesis, Swain (1995: 125) claims that ‘producing
language serves second language acquisition in several ways.’ Similarly,
Skehan (1998: 11) comments that it is counter-intuitive to claim that
‘one learns to speak by listening.’ A further important criticism of the
Input Hypothesis was made by Howatt (1984: 286) who notes that it
only dealt with grammar and had little to say about vocabulary.
The Affective Filter Hypothesis is generally accepted. This is the

uncontroversial idea that language learners learn more effectively when
they are relaxed and receptive to learning.
4.2.6 Teaching Procedures
Methodology for Beginners

In The Natural Approach (1983) Steven Krashen collaborated with
teacher-trainer Tracy Terrell to translate his language learning theory
into classroom practice. According to Krashen and Terrell (1983:67),
the overall aim of the Natural Approach is ‘to develop basic personal
communication skills—both oral and written.’ There is a clear division
in their work between the approach taken with beginners and that with
more experienced learners and this varies along two different dimensions:
input and output.
Inputs
Input to beginners is divided into three stages. The first stage is
the personal identification stage. In this stage, beginners learn to talk
about themselves, their families, and friends. The second stage consists
of comprehensible input about experiences and beginners learn to talk
about their own experiences. The third stage involves comprehensible
input about opinions and according to Krashen and Terrell (1983: 73),
beginners will discuss ‘political issues, civil rights, marriage, family and so
forth, and gain the competence to express their views.’ While the first two
stages seem achievable for beginners, even advanced learners may struggle
to discuss topics such as politics and civil rights, and it is doubtful
whether Krashen and Terrell fully grasped the difficulty beginners would
have in doing so.
Outputs
In terms of output, expectations of beginners are also divided
into three phases: Prespeech (Preproduction), Early Production, and
Extending Production. In the Prespeech phase, the students remain silent
while the teacher provides comprehensible input through teacher talk.
Krashen and Terrell (1983: 76) suggest using James Asher’s Total Physical
40 I. Pemberton
Response method. After demonstration of several actions by the teacher,

the learners respond to commands to perform them. There is no pressure
on them to produce the target language. Note that while this may be suit-
able for younger learners, older learners may be resistant to getting out
of their chairs to hop, skip, and jump, and the arrangement of classroom
furniture may also prevent students from performing actions. In addition
to using TPR, teachers may describe the physical attributes of people in
the room, or the contents of pictures. Students can be asked to point
to the people, pictures, or elements within the pictures. Krashen and
Terrell (1983: 77) explain that a preproduction stage allows the students
to get acquainted with the teacher, classroom procedures, and each other.
It allows them to focus on developing comprehension skills and gives
them the confidence that they can successfully learn a new language.
Furthermore, it also kickstarts the language acquisition process.
The next phase is the early production phase. In this phase, the teacher
provides comprehensible input, and the students begin to produce
language. However, this will be limited to single-word utterances and
short phrases; in fact, the earliest responses are likely to be yes and no.
Once these are established, the teacher moves on to either-or questions.
The following step is to ask students to identify people and items by
asking basic questions such as What is it? What colour is it? and What
is s/he doing? In the early production phase, Krashen and Terrell (1983:
83) also recommend the use of realia which display things in visual and
numerical form such as charts, timetables, forms, and advertisements.
After the early production phase comes the extending production
phase. Krashen and Terrell (1983: 84) suggest two specific exercises to
extend student production, the open-ended sentence, and the open-
ended dialogue. An open-ended sentence is a fixed pattern which
includes an open slot, for example: In this room, there is a …, I am
wearing a …., and I want to ….
An open-ended dialogue is a fixed conventional dialogue with two
open slots which allow students to be creative in how they answer. For
example: Where are you going? To …. What for? To …
Krashen and Terrell (1983: 85) comment that pictures, charts, and
advertisements can continue to be used but that the instructor asks
questions that need to be answered with short phrases or sentences.
Methodology for Post-beginners

In their recommendations for post-beginners, there is a noticeable
shift from the use of teacher talk to provide comprehensible input to
activities which rely on student interaction. Krashen and Terrell (1983:
97) explain that students find it ‘affectively satisfying’ when they commu-
nicate in the target language and that student interlanguage provides
comprehensible input for their classmates. They (1983: 95) recommend
four types of activity for classroom use: Affective-Humanistic Activities,
Problem-Solving Activities, Games, and Content Activities. In order to
promote acquisition, topics must be interesting and meaningful to focus
attention on content not form (ibid:97).
Affective-Humanistic Activities
Affective activities evoke feelings, opinions, desires, reactions, ideas,
and experiences (Krashen & Terrell, 1983: 100). The teacher can
continue to use dialogues, charts, and tables but with more sophisticated
questions and longer replies. In addition to these, Krashen and Terrell
(1983: 101) recommend interviews; for example, students work in pairs
to ask a series of questions about childhood. They also suggest preference
ranking, where students rank several items in order of preference such as
leisure activities before having a follow-up discussion.
Problem-Solving Activities
The most important aspect of these activities is that learners focus on
solving a question, problem, or situation (Krashen & Terrell, 1983: 108).
In a typical problem-solving activity, the teacher will provide compre-
hensible input while explaining the problem, and then the students
discuss and solve the problem. During this process, student interlanguage
provides further comprehensible input.
One type of problem-solving activity is known as a task. Krashen
and Terrell (1983: 108) explain that the main purpose of the task is to
describe a sequential activity; for example, how to wash a car. There are
three stages in the task. In the first stage, the teacher elicits vocabulary
from the students. In the second stage, the teacher and students work
together to describe the activity. In the final stage, the teacher asks ques-
tions related to whether students perform the activity, e.g., How often do
you wash your car? When? Where? Do you enjoy it?
42 I. Pemberton
A similar activity is known as a series. The teacher will show a

sequence of pictures or photographs which tell a story. Students will go
through a similar three-step process to recount the story. Other problem-
solving activities can involve realia such as charts, graphs, and maps. For
example, students might have similar maps with different marked and
unmarked locations. They will practise giving directions to guide each
other to the locations.
For more advanced students, the concept of incomplete dialogues can
be expanded to whole situations by giving students a prompt such as You
are in a restaurant full of people. You approach the hostess, and you say to
her… A related activity is the situation reaction. The teacher describes a
situation for the students to react to. For example, Your washing machine
is broken. You called the repair service two days ago and they made an
appointment with you for today at 11 am. You have waited all morning,
and no one has shown up. What will you do?
Games
The purpose of playing games is not for their entertainment value but
to foster acquisition, which they promote because players are focused on
the outcome of the game not the language forms (Krashen and Terrell,
1983:121). These include action games such as the classic icebreaker, find
someone who …, races against the clock, and games based on television
shows, e.g., charades.
Content Activities
A parallel is drawn between the traditional use of content activities
in language teaching to learn a subject such as science, social studies,
or art in the target language, and their use in the Natural Approach,
which includes input activities such as films, television reports, and news
broadcasts, together with output activities such as individual reports,
presentations, and show and tell activities. To highlight the effectiveness
of content activities, Krashen and Terrell (1983: 124) refer to immer-
sion programmes in Canada and the United States. They comment that
students ‘acquire impressive amounts of the second language as well as
the subject matter.’ Krashen and Terrell theorize that the mechanisms by
which students learn are common to both the immersion programmes
and the Natural Approach. That is to say, they are exposed to a large
amount of comprehensible input, and they are focused on the message.

However, they overlook the fact that learners in immersion programmes
are focused on understanding and remembering sets of logically related
concepts and ideas, whereas, in the Natural Approach, they are not.
4.3 Summary
The Notional-Functional Syllabus represented a move away from rule-
based and structure-based approaches to linking language with the
contexts in which it is used. While this might suggest a situational
syllabus, Wilkins was critical of this as being too specific and limiting
to be of widespread application. Instead, he proposed a content-based
approach using general concepts and functions which would apply across
different situations with similar features. He proposed three general cate-
gories: semantico-grammatical, modal, and communicative. Semantico-
grammatical items described common concepts, modal items expressed
beliefs and feelings, and communicative items expressed functions. A
number of criticisms have been made of notional-functional syllabi: there
are too many functions to specify, the approach downplays lexis, and it
neglects receptive skills.
The late 1970s saw the appearance of Steven Krashen’s Natural
Approach. It is no understatement to say that the Natural Approach
has been one of the most controversial methods of all time. While it is
largely eschewed by academics and has been for some decades, it is still
being discovered and debated by practitioners even today. The Natural
Method was based on a number of hypotheses, collectively referred to as
the Input Hypothesis. These hypotheses were the Acquisition-Learning
Hypothesis, the Natural Order Hypothesis, the Monitor Hypothesis, the
Input Hypothesis, and the Affective Filter Hypothesis. The Acquisition-
Learning Hypothesis states that language is acquired not learnt. The
Natural Order Hypothesis states that grammar items are learnt in a set
order that cannot be influenced by instruction. The Monitor Hypothesis
states that learners can use conscious learning as a monitor when they
have enough time to do so, such as when writing. The Input Hypoth-
esis states that comprehensible input is necessary for acquisition, and
44 I. Pemberton
the Affective Filter Hypothesis states that acquisition can only take place
when learners have a low affective filter, in other words, when they are
relaxed.
To operationalize the Natural Approach Krashen and Terrell describe a
detailed set of classroom procedures. These are divided into activities for
beginners and post-beginners. The main source of input for beginners
is teacher talk. They are expected to go through a pre-speech or silent
phase before progressing to an early production phase where they ask and
answer questions. Next is an extending production phase, where a teacher
uses realia and asks more complex questions. In the post-beginners’
phase, comprehensible input is increasingly generated by student inter-
actions. Students work together to complete information gap activities,
information gathering activities, role plays and simulations, and discus-
sions. This being the case, a Natural Approach classroom in the early
1980s must have looked very similar to a modern-day CLT classroom.
A number of criticisms were made of the five hypotheses of the
Natural Approach. While the Natural Order, Input, and Affective filter
hypotheses are generally accepted, the Acquisition-Learning and Monitor
hypotheses have come in for criticism. Taken together, Krashen’s posi-
tion has been that explicit instruction in grammar (learning) cannot
become implicit acquisition. When viewed through a usage-based lens,
all language patterns, whether grammatical or lexical in nature, exist
alongside each other in cognition. The teaching of grammatical rules
imparts a knowledge of grammatical rules. If acquisition is taken to mean
automatization, it doesn’t result in acquisition. That requires a further
process of entrenchment—repeated processing and production of actual
instances of language. However, a knowledge of grammatical structures
can certainly act as a sociocultural tool which can mediate understanding
and lead ultimately to acquisition.
References
Hodder Education.
Corder, S. P. (1973). Introducing applied linguistics. Penguin.
Hall, G. (2018). Exploring English language teaching: language in action (2nd
edn). Routledge/ Taylor & Francis Group.
University Press.
Howatt, A. P. R., & Smith, R. (2014). The history of teaching english as
History, 57 (1), 75–95.
Krashen, S. D. (1985). The input hypothesis: Issues and implications. Longman.
Larsen-Freeman, D. & Anderson, M. (2011). Techniques and principles in
language teaching (3rd edn.). Oxford University Press.
Randall, M. (2007). Memory, psychology and second language learning. John
Benjamins Publishing Company.
Press.
Swain, M. (1995) Three functions of output in second language learning.
In Cook, G. and Seidelhofer, B. (eds.), Principles and practice in applied
linguistics. Oxford University Press.
tation. Pergamon.
5
Communicative Language Teaching
The 1970s saw the emergence of Communicative Language Teaching.

Howatt (1984: 192) explains that the Communicative Approach was not
new but that it shared the same underlying philosophy as earlier methods
such as the Natural Method, the Conversation Method, and the Direct
Method.
Communicative Language Teaching has had a powerful and enduring
impact on language education. Cook (2003: 35) points out this has
proven to be a significantly more durable phenomenon than Krashen’s
Natural Approach. Richards (1985) notes that its wide acceptance by the
British language teaching establishment led to it rapidly becoming the
dominant language teaching orthodoxy worldwide (cited in Richards &
Rodgers, 2001). Harmer (2007: 71) comments that ‘the Communicative
Approach has left an indelible mark on teaching and learning, resulting
in the use of communicative activities in classrooms all over the world.’
Despite its widespread acceptance, there is a lack of clarity about what
Communicative Language Teaching is. In part, this is because there has
never been agreement on the principles of the approach. Ellis (1982: 73)
comments that ‘the label “communicative” has become a catchphrase in

Switzerland AG 2024
48 I. Pemberton
language teaching’ which ‘suggests that the term has no clearly under-
stood and received meaning when it is applied to language teaching.’
Richards and Rodgers (2001: 155) assert that ‘there is no single text
or authority on it, nor any single model that is universally accepted as
authoritative.’ Harmer (2007: 69) makes the point that ‘the real problem
when attempting to define CLT (or the Communicative Approach as
originally called) is that it means different things to different people.’
The combination of enthusiasm for the Communicative Approach and
a lack of clarity about what it represented led to the emergence of both
strong and weak versions of the approach.
5.1 Strong and Weak Versions

Strong versions of the Communicative Approach assert that it isn’t neces-
sary to teach grammar to learners as they naturally acquire it during
the process of communication (Ellis, 1997: 79). Randall (2007: 150)
explains that ‘the desire to communicate will drive the learner to use
the language and to acquire the correct forms in order to commu-
nicate better.’ Allwright and Hanks (2009: 47) advocate ‘keeping the
learners busy,’ and ‘constantly engaged in “productive” tasks.’ Harmer
(1982: 166; 2007: 70) comments that the materials should not restrict
the language learners use, leaving them free to choose their own language.
In addition, the teacher should not intervene while the activity is in
progress. The language input comes from the learners themselves. As
learners work together on communicative tasks, they naturally share
information which leads to greater improvement than that which can
be achieved individually.
Notwithstanding the enthusiasm of some academics for strong
versions of the Communicative Approach, weak versions came to form
the mainstream of Communicative Language Teaching. Allwright and
Hanks (2009: 47) remark that the strong view was too radical to be a
commercial success. Howatt (1984: 279) notes that within ten years of
its appearance, the weak version of CLT became standard practice. In
Thornbury’s (1998: 110) opinion, ‘from a communicative perspective,
CLT is not only weak but very weak.’
5.2 Classical and Current Versions

There is one other factor which makes it difficult to determine what
CLT is. In Communicative Language Teaching Today, Richards (2006: 8)
points out that the Communicative Approach has continued to evolve in
the decades since its inception. He draws a distinction between Classical
Communicative Language Teaching from 1970 to 1990, and Current
Communicative Language Teaching from the late 1990s to the present.
According to Randall (2007: 160), ‘present day CLT has moved away
from a primary concern with meaning to one that incorporates more
attention to language form.’ While he differentiates between the two
phases, Richards (2006: 45) concludes that ‘today CLT continues in its
classic form as seen in the huge range of course books and other teaching
resources that cite CLT as the source of their methodology.’
5.3 Communicative Principles

Defining Communicative Language Teaching involves not only exam-
ining different methods which claim to be communicative but also
looking for differences which may have emerged across different time
frames. One approach to definition is to think of Communicative
Language Teaching as an ‘umbrella’ term for a range of methods which
share some common characteristics. While there is a lack of agreed prin-
ciples, an array of taxonomies can be found in the literature (e.g., Brandl,
2008; Brown, 2001; Richards, 2006; Richards & Rogers, 2001). These
can be analysed thematically to identify common principles.
5.3.1 Aim
One distinguishing feature that unites all communicative approaches is

that they aim to develop communicative competence. This means that
‘the goal is to enable students to communicate in the target language’
(Larsen-Freeman, 2000: 128). In other words, Communicative Language
Teaching aspires to give learners a practical command of the language,
50 I. Pemberton
rather than an understanding of its grammar or the ability to read its

literature. According to Richards (2006: 3), a learner can master the rules
of grammar and still not be successful at using the target language for
communication. Therefore, the main aim is to teach learners to use the
target language to interact with other speakers of the language.
5.3.2 Theory of Learning
Another distinguishing feature which unites different versions of CLT

is a firm belief that learners learn a language by communicating in it.
To an adherent of communicative methodology, this is axiomatic and
requires no explanation or evidence. According to Richards (2002: 22)
proponents of CLT ‘never felt compelled to produce any evidence
to demonstrate that learning was more successful if “communicative”
teaching methods and materials were adopted.’ Richards (2006: 22)
explains that a core assumption of CLT is that ‘second language learning
is facilitated when learners are engaged in interaction and meaningful
communication.’ Harmer (2007: 69) remarks that for many practitioners
an essential belief is that when learners participate in a communicative
task with a focus on meaning, the ‘language learning will take care of
itself.’ Skehan (1998: 16) states that ‘to learn to speak, we actually have
to speak.’
The corollary of this strong belief in the importance of speaking is
the weakness of CLT’s explanation of language learning. Nevertheless,
a number of attempts at explanation have been made. Allwright and
Hanks (2009: 46) claim that learners develop ‘by deploying their existing
linguistic resources, at whatever level, to solve their immediate commu-
nication problems.’ According to Randall (2007: 150), CLT proponents
consider that ‘the desire to communicate will drive the learner to use
the language and to acquire the correct forms in order to communicate
better.’ Finocchiaro and Brumfit (1983) theorize that ‘the target linguistic
system will be learned best through the process of struggling to commu-
nicate’ (in Richards & Rodgers, 2001: 156). Ellis (1982) cited in Harmer
(2007: 55) speculates that ‘it may be that in making efforts to retrieve
and then use all and any of the language, we flick a ‘switch’ that takes
language from the learnt to the acquired store.’ In summary, in desiring

to solve their communication problems, and struggling to do so, learners
flick a switch to acquire the language. While no language educator would
deny the importance of motivation to use a language, the explanation of
the psychological processes by which this occurs is rudimentary at best.
This lack of a theoretical foundation for Communicative Language
Teaching has not gone completely unnoticed. Randall (2007: 174)
comments that given their overriding focus on communication of
meaning, ‘the interests of (the communicative) movement were not on
the cognitive structures involved or, in any detail, on the processes
involved in learning rules.’ Cook (2008: 250) points out that CLT
‘assumes little about the learning process, apart from claiming that, if the
right circumstances are provided to them, something will happen inside
the students’ minds.’ Similarly, Dornyei (2009: 34) compares CLT to
Audiolingualism noting that even this method was based upon a specific
learning theory (Behaviourism). He adds that ‘the only learning-specific
principle that was available for CLT materials developers and practi-
tioners was the broad tenet of “learning through doing”, coupled with
the only marginally less ambiguous guideline of developing the learners’
communicative competence through their active participation in seeking
situational meaning.’
In Approaches and Methods in Language Teaching, Richards and
Rodgers (2001: 161) also recognize CLT’s lack of theoretical under-
pinnings, commenting that in contrast to its well-defined theory of
language, little has been said about its theory of learning. They suggest
that elements of learning theory can be derived from the principles of
CLT. One principle is the communication principle which states that real
communication leads to language learning. A second principle is the task
principle which states that learning is enhanced when language is used to
complete a task. A third principle is the meaningfulness principle which
states that language that is meaningful for a learner promotes learning.
They concede that these principles specify the conditions for learning
rather than explain the process of learning.
Some proponents cite Michael Long’s Interaction Hypothesis as a
theoretical basis for language learning in the communicative classroom.
For example, Lopez-Sanchez (2009: 31) states that ‘CLT does not claim
52 I. Pemberton
any particular theory of language as its basis, but its emphasis on

speaking (communication) suggests it accepts the premises of the Inter-
action Hypothesis.’ Randall (2007: 155) remarks that CLT proponents
consider interaction to be ‘the main mechanism by which languages are
learnt.’ Schaefer (2015: 173) considers the Interaction Hypothesis to
be ‘commonsensical’ and Whong (2013: 122) describes it as influential,
noting that it involves ‘input, output, and negotiation of meaning.’
According to Ellis (1991: 4), the Interaction Hypothesis makes two
major claims about the influence of interaction on second language
acquisition.
1. Comprehensible input is necessary for L2 acquisition.

2. Modifications to the interactional structure of conversations which
take place in the process of negotiating a communication problem
help to make input comprehensible to an L2 learner.
An interaction can take place between an L2 learner and an L1

speaker, or between more and less proficient L2 learners. Modifica-
tions can be made by either the speaker or listener. According to Hall
(2018: 123), modifications can be initiated by the interlocutor, for
example, comprehension checks such as ‘Do you understand me?’, or
complete repetition of a word or phrase. On the other hand, learners
might make a confirmation check such as ‘so, what you mean is …’ or a
request for clarification such as ‘could you repeat that, please?’.
While there is some support for the Interaction Hypothesis, it has
several weaknesses as an explanation of second language acquisition. One
weakness is that listeners process input for meaning and do not attend to
form unless there is a communication breakdown. Thus, many unknown
language items will not be attended to. This led Ellis (1991: 30) to
propose a three-stage process of noticing, comparison, and integration.
The first two stages take place in working memory. In the noticing phase,
the learner recognizes that their conversational partner is adapting their
speech and notices the form they are using. In the comparison phase, the
learner compares the input to their existing cognitive representation, and
in the integration phase, the learner consigns the new form to long-term
memory. Thus, (Ellis, 1991: 36) modifies the hypothesis as follows:
1. Comprehensible input facilitates L2 acquisition but is neither neces-

sary nor sufficient.
2. Modifications to input, especially those which take place in the
process of negotiating a communication problem make acquisition
possible providing that the learners:
(a) comprehend the input
(b) notice new features in it and compare what is noticed with their
own output.
3. Interaction that requires learners to modify their initial output facili-
tates the process of integration.
However, even with Ellis’s modifications, the hypothesis still faces two
serious criticisms. The first is the limitations of working memory. While
processing input for meaning, a listener needs to recognize that a speaker
is adapting their language, hold the adapted version in working memory,
recall their current cognitive representation, compare the two, and send
the new form for storage in long-term memory. This represents a signif-
icant cognitive overload of working memory. The second, and perhaps
more serious criticism, is that learners have to learn a huge number of
language items, and it is infeasible that they can negotiate them all.
According to Ellis (1991: 4), in his later publications even Long concedes
that the hypothesis can’t account for the bulk of acquisition.
To sum up, a major weakness of Communicative Language Teaching
is that its theory of learning is vague at best. While adherents believe
that the pressure to communicate motivates learners to use their existing
language knowledge resulting in cognitive changes taking place, there
has been no explanation of how this happens. Richards & Rodgers have
proposed that its language learning theory can be inferred from commu-
nicative principles, but these represent nothing more than a specification
of the conditions under which acquisition can occur, not the process of
acquisition. Michael Long’s Interaction Hypothesis has been suggested as
an account of this process, but it is too rudimentary to account for the
bulk of acquisition that takes place.
54 I. Pemberton
5.3.3 Inputs
While the heavy emphasis on output activities in the literature can make
it a challenge to identify what the inputs might be to learners in the
communicative paradigm, one way to consider the types of input is to
look at the proposals for communicative syllabi. The two main syllabus
types that have been suggested for communicative courses are structural
syllabuses and functional syllabuses.
Several well-known educators have suggested that communicative
courses should retain a structural core. According to Littlewood
(1981: 77), taking a communicative approach is not a reason for aban-
doning a structural syllabus. He asserts that a mastery of structural
features is still necessary for communication. His only proviso being that
language examples are meaningful to learners. In the early 1980s, Sandra
Savignon also underscored the importance of grammar in the commu-
nicative paradigm. She analysed communicative curricula into five
components: language arts, language for a purpose, personal language
use, theatre arts, and beyond the classroom. Savignon (2001: 237–238)
defined the first component, language arts, as follows: this ‘represents
what language teachers do best–most often because it is all they have
been taught to do. Language arts focuses on rules of usage and provides
explanations of how language works.’ In her book, The Communicative
Syllabus, Yalden (1983: 110) argues that learners should be taught the
structural elements of a language before moving on to functions. Celce-
Murcia et al. (1997: 141) comment that communicative language classes
apply instruction in grammar to the teaching of conversation.
The retention of a structural syllabus has proven unpopular with
some CLT advocates. According to Allwright and Hanks (2009: 47),
Littlewood’s publication Communicative Language Teaching ‘solved the
commodity problem by offering much less challenging ideas and
returning to carefully chosen syllabus content.’ Richards and Rodgers
(2001: 155) also take a critical stance, commenting that some consider
that Communicative Language Teaching represents little more than
integrating grammar and functions.
While some educators recommended maintaining a structural
syllabus, a hallmark of CLT is its adoption of a functional syllabus. Berns
(1983: 5) points out that ‘the functional approach to language teaching

is intimately related to the communicative approach and in recent years
has enjoyed widespread popularity.’ Littlewood (1981: 1) states that ‘one
of the most characteristic features of communicative language teaching
is that it pays systematic attention to functional as well as structural
aspects of language, combining these into a more fully communicative
view.’ Similarly, Harmer (2007: 69) comments that the ‘concern was
with spoken functions as much as with written grammar, and notions of
when and how it was appropriate to say certain things were of primary
importance.’
Even where functions are used as an organizing principle, a struc-
tural emphasis may be maintained. Littlewood (1981: 79) suggests that
syllabus writers can begin by pairing functions with simple language
forms and then gradually introduce more complex ones which allows
for the retention of a structural approach. He provides an example of a
course starting with basic functions such as: ‘asking somebody’s name,’
‘asking and saying where places are,’ ‘saying what somebody’s job is,’
and ‘talking about nationality.’ All of these functions can be expressed
with simple structural patterns using the ‘be’ verb, e.g., ‘What’s your
name?’, ‘Where’s Kent Road?’, ‘Is Sally a journalist?’—‘Yes, she is,’ ‘Are
they English?’—‘No, they aren’t.’
There are also suggestions that a communicative syllabus should
progressively evolve according to proficiency. According to Littlewood
(1981: 79), a basic language course would adopt a structural syllabus.
Similarly, Yalden (1983: 113) notes that one approach might be to ‘move
increasingly towards a functional emphasis, while retaining a structural
progression.’ She suggests a formal-rhetorical-instrumental approach. In
this approach, inputs would vary according to level. After learning struc-
tures at a beginning level, learners would progress to functions before
moving on to learning school subjects at an advanced level. For general
learners, Littlewood (1981: 81) suggests topic-based organization. The
teacher can use common topics of relevance to the learners, e.g., food,
houses, holidays, sports, and music, as a basis for reading, listening,
speaking, and writing activities.
56 I. Pemberton
5.3.4 Authentic Materials
In general, CLT practitioners regard authentic input as a key element of

its methodology. After describing the rich input available to children in
first language acquisition, Brandl (2008: 12) argues that the rich input
cannot be replicated in the classroom to develop fluency. However, he
(ibid: 17) suggests that teachers should try to make the input as rich as
possible and one way that they can do this is through the use of authentic
materials. A number of examples of authentic materials are provided
by Richards and Rodgers (2001: 170) including ‘language-based realia,
such as signs, magazines, advertisements and newspapers or graphic and
visual sources around which communicative activities can be built, such
as maps, pictures, symbols, graphs, and charts.’
Authentic materials provide a number of benefits to learners. Richards
(2006: 24) explains that they promote interest and supply accurate
language models. Brandl (2008: 13) points out that they mirror the situ-
ations and demands of real life. Larsen-Freeman (2000: 129) adds that
authentic materials provide learners with the chance to practice strategies
for understanding real language.
5.3.5 Active Learning
A common criticism of teacher-centred activities is that they cast

students in the role of passive recipients of knowledge which results in
superficial learning. This has resulted in pedagogical trends in general
education such as active learning and cooperative learning which have in
turn strongly influenced Communicative Language Teaching. Richards
and Rodgers (2001: 158) state that communicative pedagogy represents
the language learning version of ‘learning by doing.’ Brandl (2008: 12)
notes that learning by doing is not a new way of learning but has been
recognized as a fundamental educational principle throughout history.
As an example, Howatt (1984: 275) describes how the EAL publication
‘Scope’ brought together ‘the EFL tradition of the linguistically orga-
nized syllabus (structural patterns, controlled vocabulary, etc.) and the
primary school tradition of activity methods which require children to
use the new language co-operatively to make puppets, charts, models of

various kinds, and so on.’
5.3.6 Interaction
As a realization of the principle of active learning, CLT advocates cite

the importance of interaction. Brown (2001: 48) states that communica-
tive interaction lies at the core of communicative competence. Similarly,
Richards (2006: 25) comments that learning is a social not individual
activity which depends on interaction. In the classroom, this is realized
through pair- and groupwork activities (ibid: 20). According to Larsen-
Freeman (2000: 130), students can interact with each other in pair-,
triad-, small group-, and whole-group configurations.
The rationale for interactive activities has both practical and cognitive
dimensions. To begin with practical arguments for the use of CLT activ-
ities, Larsen-Freeman (2000: 130) explains that small group interactions
maximize the time each student spends in communication. Richards
(2006: 20) comments that students learn from listening to other group
members and produce more language than in a teacher-centric class.
5.3.7 Meaningfulness
The interactive activities of CLT emerged as a reaction to earlier method-

ologies. In his evaluation of the Communicative Approach, grammarian
Michael Swan (1985: 3) notes that a doctrine of the approach is that
other approaches did not have enough focus on meaning. Similarly,
Brandl (2008: 16) asserts that it emerged as a counter-reaction to the
repetitive drills of audiolingual methodology which did not require
learners to process the language for meaning. Richards (2006: 4) notes
that CLT started a move away from traditional teaching which focused
on grammar and controlled practice such as drills and rote memorization
of dialogues towards pairwork, role plays, groupwork, and project work.
To further illustrate this point, Larsen-Freeman (2000: 129) provides the
example of a well-known communicative activity—the information gap.
58 I. Pemberton
She makes the point that each learner knows something the other does
not.
In favouring meaningful activities over meaningless drills, commu-
nicative educators argue that they facilitate language learning. Johnson
(1982) states that language use in meaningful tasks promotes learning
(cited in Richards & Rodgers, 2001: 161). Richards (2006: 22) asserts
that interaction involving meaningful communication facilitates second
language learning adding that the driving force of learning is meaning
(ibid: 25). Harmer (2007: 69) notes that an essential belief of CLT is
that if learners do communicative tasks with a focus on meaning, then
‘language learning will take care of itself.’
5.3.8 Treatment of Errors
Beliefs about the treatment of errors within the communicative paradigm

have changed over the past fifty years. In a reaction to the zero-tolerance
of errors in Audiolingualism, some early adopters of communicative
methodology, for example, Revell (1979: 8) argued that making errors
was a necessary step in progress towards fluency. Similarly, Larsen-
Freeman (2000: 132) noted that errors were a natural outcome of
fluency development. However, when researchers of Canadian immer-
sion programmes found that learners did not achieve the grammatical
accuracy of native speakers, the position shifted. In describing current
communicative teaching, Brandl (2008: 20) concedes the necessity of
error correction, remarking that it is a long-term strategy which has
a positive effect on individual learners. In summary, the communica-
tive field has moved from a position of believing that errors would be
corrected by subsequent input to a belief that some kind of intervention
is necessary to support language development.
5.4 Communicative Activities

While it can be difficult to identify common features of inputs to learners
in the communicative paradigm, there is a huge volume of literature
which discusses outputs. Harmer (1982: 164) comments that teachers
might feel confused by the sheer number of communicative activity types
which range from drills and dialogues to simulations and communication
games. Aspects common to different communicative methodologies can
relate to general principles, materials, specific activity types, and the roles
of classroom participants. Some of the most common ones are detailed
below.
5.4.1 Information Gap
One of the most widely described communicative activity types used

for practising functions is information gap activities. In an informa-
tion gap activity, one learner has knowledge of something that the
other learner does not know (Larsen-Freeman, 2000: 129). Similarly,
Richards (2006: 18) comments that people communicate in a real situ-
ation to obtain information that they do not have. Harmer (1982: 166)
explains that this is important because without a gap there is no reason
for communication to take place. Revell (1979: 6) points out that in
an information gap, students have a choice of what language to use,
ask questions, and have a reason to listen to each another. Littlewood
(1981: 26) provides the following example: learners have the same table
containing information about various towns but with different elements
deleted. They have to ask questions to obtain the missing information.
Similarly, learners may have the same map but with different landmarks
deleted. They have to guide each other to the deleted landmarks (Little-
wood, ibid: 32). While Littlewood (ibid: 20) comments that learners can
use whatever language they know to complete the activity and that it
doesn’t matter if the language they use is correct, Harmer (1982: 167)
notes that activities such as completing tables are not entirely commu-
nicative because the range of language choices a speaker can use is still
60 I. Pemberton
limited. Another point is that if the language they use is not correct, it is
not clear how they will develop accurate representations of language.
5.4.2 Jigsaw
A jigsaw activity is a kind of information gap but usually takes place

on a group level rather than as a pairwork activity. Richards (2006: 19)
explains that the teacher divides the class into groups and gives each
group different information. The groups have to put their informa-
tion together in order to complete the task. Revell (1979: 6) points out
that the activity can use both written and spoken materials. Littlewood
(1981: 73) notes that the texts might provide information to solve a
problem or complete an account of an event.
5.4.3 Information Transfer
In an information transfer activity, the learners convert information from

one form to another. Richards (2006: 19) explains that learners might
read instructions on how to get from one place to another and then draw
a map, or they might read information and draw a graph. In addition to
converting written information to graphic form, Littlewood (1981: 71)
suggests that the activity can be used to practise listening, for example,
learners hear descriptions of people and fill in a table. A more challenging
activity might be to listen to a description or discussion and draw a scene
or a plan. Harmer (1982: 167) comments that he considers this describe
and draw activity to be communicative. It might be noted, however, that
it doesn’t involve any language production.
5.4.4 Information Gathering
A frequently used communicative activity is to ask learners to gather

information from their classmates. Richards (2006: 19) comments that
in-class information-gathering activities can involve surveys, interviews,
and searches. These can also involve individuals outside of the classroom.
Savignon (2001: 240) suggests that in situations where the learners are
living in the area where the language is spoken, they can source adver-
tisements from local newspapers to find out where they can buy products
cheaply before actually going out into the community to ask about the
prices of such products as second-hand cars, watches, and cameras, and
then report back to the class.
5.4.5 Role Play and Simulation
Richards (2006: 20) defines role plays as activities where the teacher
assigns students roles and asks them to improvise a scene or provides
cues and information as the basis for an exchange. Harmer (2007: 69)
comments that they are useful activities as the communicative classroom
places emphasis on real or, at least, realistic communication where task
completion is as important as accurate language use. On a basic level,
Littlewood (1981: 70) suggests that the teacher instructs learners to
perform actions or mime them. Savignon (2001: 239) notes that fantasy
and play-acting are a natural part of growing up which provide opportu-
nities for language use. Thus, theatre arts are an important component
of second language learning. However, she cautions that they need to
be set up carefully to be useful. While the criticism could be made that
learning lines is not real language use, Littlewood (1981: 60) suggests
that learners improvise. They can be presented with a stimulus and make
their own language choices. In addition, learners can be asked to adopt
certain identities or act out certain personality types.
5.4.6 Discussion
A natural addition to the canon of communicative activities which

provide ample opportunities for language practice is discussions. They
are particularly useful to promote engagement amongst higher profi-
ciency learners. The learners can be asked to compare opinions, beliefs,
and values. For example, they can be given a debatable proposition to
discuss. Another discussion activity, suggested by Littlewood (1981: 36),
is to ask learners to discuss solutions to a problem: for example, they
62 I. Pemberton
are going on a three-day hike in the mountains and have a list of gear,
but they can only carry 11 kilograms each. They have to decide what to
take. A less challenging activity might be to rank important qualities in a
partner (Richards, 2006: 19). As opposed to just being a free exchange of
opinions, Littlewood (1981: 57) also explains that learners can be given
roles in a discussion. While the discussion can focus on either a real or
simulated issue, the use of roles ensures that the participants introduce
different arguments which leads to a lively exchange of views. At the end
of the discussion, the participants have to come to a consensus or report
back to the class.
5.5 Criticisms
While it is widely accepted, Communicative Language Teaching has its
critics. One common criticism is that it provides insufficient input for
language learning to occur. According to Field (2008: 2) the second half
of the twentieth century has seen instructors place great emphasis on
speaking in the belief that it is the most important skill for language
learners. He contends that the overemphasis on speaking is the result of
‘muddled thinking’ and makes the point that communication is a two-
way process in which it is as important to listen as it is to speak. Howatt
(1984: 287) observes that CLT rests on an unstated assumption that
learners already know English and the instructor’s job is to make sure that
they use their knowledge to communicate. Similarly, Field (2007: 33)
notes that what learners are actually doing is using what they already
know rather than extending their language knowledge. Ellidokuzoglu
(1996: 4) agrees, pointing out that while proponents of Communicative
Language Teaching do not assert that input is not important for second
language acquisition, CLT coursebooks underemphasize input, focusing
instead on output. Even CLT advocate H. Douglas Brown (2001: 377)
concedes that ‘in the hustle and bustle of our interactive classrooms,
sometimes we get so caught up in lively groupwork and meaningful
communication that we don’t pause to devote some attention to words.’
Finally, Skehan (1998: 40) notes research shows that ‘second language
use, in itself, does not reliably lead to second language change.’
In addition to the amount of input to learners, there is also a ques-

tion about its quality. Randall (2007: 153) notes that author of the
Interaction Hypothesis, Michael Long, and others eventually recognized
that different qualities of input had different impacts on learning. Swan
(2005: 389) openly questions the efficacy of elementary students learning
from others of the same proficiency, stating that while L2 learners might
learn vocabulary from each other and correct one another’s mistakes,
what they learn will be limited and not always accurate. In a commentary
on input differences between naturalistic learners and classroom learners,
Rothman and Guijarro (2010: 301) note that there is a difference in the
grammatical quality (morphology and syntax) of the input they receive.
They comment that L2-L2 interactions generate competing language
forms not experienced by naturalistic learners, and these make the task
of identifying features of the target language more difficult.
A serious weakness of Communicative Language Teaching is its failure
to clearly explain how language development takes place. The over-
riding focus of the early communicative movement was on its theory
of language—the communication of meaning—rather than its theory
of learning. A number of linguists have commented on this weakness
of CLT, and the suggestion has been made that its learning theory can
be derived from its principles, but these do nothing more than state
the conditions for language acquisition rather than explain the process.
Michael Long’s Interaction Hypothesis has also been suggested as an
explanation for language learning. This states that when a miscommu-
nication takes place both the speaker and listener modify their language
to make the input comprehensible which promotes learning. However,
the hypothesis has a number of flaws: listeners process input for meaning
not form, working memory is insufficient to process the modifications,
and there is simply too much learning for it all to take place through
negotiation.
Finally, the ubiquity of pair and groupwork within communicative
language teaching seems to suggest that such procedures are essential for
effective learning but Savignon (2017: 6) makes the important obser-
vation that ‘CLT is not concerned exclusively with face-to-face oral
communication’ and ‘group or pair work should not be considered an
essential feature and may well be inappropriate in some contexts.’
64 I. Pemberton
5.6 Summary
This chapter has considered Communicative Language Teaching which
has been the orthodox teaching approach for more than four decades.
It is widely considered to have appeared as a reaction to mechanistic
approaches to language teaching such as Audiolingualism and Struc-
turalism. The 1970s saw both strong and weak versions of the approach
emerge. Strong versions were favoured by some academics, but commer-
cial considerations and the practicalities of the classroom eventually saw
weaker versions become mainstream.
As Communicative Language Teaching acts as an umbrella term for
a wide variety of different methods, it is necessary to examine their
common features to ascertain its nature and scope. To begin with,
communicative methods share common principles. The main aim of
communicative methods is to teach communicative competence, in
other words, to teach learners how to communicate. All communica-
tive methods share a common theory of learning which is that people
learn language by communicating with others. A majority of main-
stream communicative methods use structural and functional syllabi
or some combination of these to specify inputs. A fourth principle
shared by communicative methods is the use of authentic materials such
as signs, magazines, advertisements, and maps. These are thought to
promote engagement and to provide accurate language models. A fifth
principle is active learning or learning by doing. This is thought to
be more effective than didactic teacher-centric instruction. In order to
realize active learning in the classroom, a communicative teacher sets
up interactions in dyads, triads, and small groups. To promote language
acquisition, the interactive activities must be meaningful. In contrast to
the drilling and rote learning of earlier approaches, they concentrate on
the exchanging meaning. A final common feature is the communica-
tive approach to error correction. In the early days, it was thought to
be unnecessary to correct mistakes, but in contemporary communicative
language teaching, there has been a renaissance in error correction.
In addition to common principles, communicative methods share
common activities. The archetypal communicative activity is the infor-
mation gap. This is a pairwork activity where each person has the same
resource (e.g., a train timetable) but with different information missing.

In this case, learners have to ask questions to obtain the missing infor-
mation. This is considered to be a meaningful activity. A similar activity
type, which may be done on a whole class level, is a jigsaw activity. Small
groups of learners are given different parts of a whole text, for example,
a story, and they have to put these together in the correct order. A third
activity type is information transfer, where a learner converts one form of
information, for example, a radio report, into another such as a table or
map. Another common activity type is information gathering whereby
a learner asks questions to classmates and then analyses the answers to
produce something such as a graph. A fifth activity type is role plays
where learners are asked to work in pairs and given prompts as the basis
for an exchange, or they work in small groups, are given roles, and are
asked to improvise a scene. A final canonical activity type is discussions
where learners may be given a debatable proposition to discuss or set a
problem to solve. In some cases, they may be given roles which require
them to provide different perspectives.
Despite its widespread acceptance, CLT has been subject to a number
of criticisms. First of all, it valorizes output which limits attention to
input. In the absence of sufficient input, research shows that practice
does not necessarily lead to accuracy. There is also a question about
whether the input is of sufficiently high quality to promote change in
learner interlanguage. While advocates of communicative methodology
believe that learners learn through participation in communicative activ-
ities, they offer no clear theoretical explanation of how this might occur.
Some pundits have suggested that CLT’s theory of learning is implicit in
its principles but concede that these do no more than specify the condi-
tions for language learning (e.g., use meaningful materials) rather than
explaining how learning (on a cognitive level) occurs. Michael Long’s
Interaction Hypothesis, including the idea of negotiation of meaning, is
sometimes co-opted as an explanation of language learning in the CLT
paradigm, but it is considered to be insufficient to account for the bulk
of language learning. Additional criticisms of CLT are that the receptive
dimension of communication and the potential for individual learning
are often overlooked in favour of productive pair and group activities.
66 I. Pemberton
References
Allwright, D., & Hanks, J. (2009). The developing language learner: An
introduction to exploratory practice. Palgrave Macmillan.
Berns, M. S. (1983). Functional approaches to language and language teaching:
Another look. Studies in Language Learning, 4 (2), 4–22.
Brandl, K. (2008). Communicative language teaching in action: Putting principles
to work. Pearson Prentice Hall.
Celce-Murcia, M., Dornyei, Z., & Thurrell, S. (1997). Direct approaches in L2
Instruction: A turning point in communicative language teaching? TESOL
Quarterly, 31(1), 141–152.
Hodder Education.
Dornyei, Z. (2009). Communicative language teaching in the 21st century:
The ‘principled communicative approach.’ Perspectives, 36 (2), 33–43.
International ELT Conference: From diversity to synergy. Ankara, Turkey.
Ellis, R. (1982). Informal and formal approaches to communicative language
teaching. ELT Journal, 36 (2), 73–81.
Ellis, R. (1991). The interaction hypothesis: A critical evaluation. Paper presented
at the Regional Language Centre Seminar (Singapore, April 22–28, 1991).
Ellis, R. (1997). Second language acquisition. Oxford University Press.
Field, J. (2007). Looking outwards, not inwards. ELT Journal, 61(1), 30–38.
Field, J. (2008). Listening in the language classroom. Cambridge University Press.
ed.). Routledge/ Taylor & Francis Group.
Harmer, J. (1982). What is communicative? ELT Journal, 36 (3), 164–168.
Harmer, J. (2007). The practice of English language teaching (4th ed.). Pearson
Longman.
University Press.
Larsen-Freeman, D. (2000). Techniques and principles in language teaching (2nd
ed.). Oxford University Press.
Littlewood, W. (1981). Communicative language teaching: An introduction.

Randall, M. (2007). Memory, psychology and second Language learning. John
Revell, J. (1979). Teaching techniques for communicative English. Macmillan.
Richards, J. C. (2002). Theories of teaching in language teaching. In J. C.
Richards & W. A. Renandya (Eds.), Methodology in language teaching: An
anthology of current practice. Cambridge University Press.
Richards, J. C. (2006). Communicative language teaching today. Cambridge
University Press.
Rothman, J., & Guijarro-Fuentes, P. (2010). Input quality matters: Some
comments on input type and age-effects in adult sla. Applied Linguistics,
31(2), 301–306.
Savignon, S. (2001). Communicative language teaching. Theory into Practice,
26 (4), 235–242.
Savignon, S. J. (2017). Communicative competence. In J. I. Liontas (Ed), The
TESOL encyclopedia of English language teaching. Wiley.
Schaefer, M. (2015). Conversational shadowing. New Directions in Teaching
and Learning English Discussion, 3, 173–181.
Press.
Swan, M. (1985) A critical look at the Communicative Approach (1). ELT
journal, 39 (1), 2–12.
Swan, M. (2005). Legislation by hypothesis: The case of task-based instruction.
Applied Linguistics, 26 (3), 376–401.
Thornbury, S. (1998). Comments on Marianne Celce-Murcia, Zoltan Dornyei,
and Sarah Thurrell’s “Direct approaches in L2 instruction: A turning point
in communicative language teaching?”: A Reader Reacts. TESOL Quarterly,
32(1), 109–116.
Whong, M. (2013). A linguistic perspective on communicative language
teaching. Language Learning Journal, 41(1), 115–128.
tation. Pergamon Press.
6
Other Communicative Approaches
6.1 Content-Based Instruction

An important pedagogy which comes under the umbrella of Commu-
nicative Language Teaching is Content-Based Instruction. In their
seminal publication, Content-Based Second Language Instruction, Brinton
et al. (2003: ix) describe Content-Based Instruction as ‘the integration
of content learning with language teaching aims.’ They compare three
different models of instruction: theme-based language instruction, shel-
tered instruction, and adjunct language instruction. Each of these models
is found in different settings. Theme-based instruction can be applied to
general EFL classes; sheltered instruction is used in immersion education
settings, and the adjunct model is most commonly found in the teaching
of English for Academic Purposes.
There are a number of compelling arguments in favour of Content-
Based Instruction for second language learning. Brinton et al. (2003: 3)
report that second language acquisition research indicates that ‘a neces-
sary condition for successful language acquisition is that the “input” in
the target language must be understood by the learner.’ They add that
‘rich second language input in relevant contexts is the key, where the
attention of the learner is focused mostly on the meaning rather than
Switzerland AG 2024
70 I. Pemberton
on the language.’ And this ‘enables learners to map new language onto
meaning and thought’ (Snow & Brinton, 2017: 4).
In its purest form, Content-Based Instruction is realized in immer-
sion language programmes that occur in secondary and tertiary settings.
Krashen (1985: 16) describes these as public-school programmes where
majority language students study in a minority language (e.g., English-
speaking students studying in Spanish in the USA).
An important aspect is the amount of second language input that
learners receive. Brinton et al. (2003: 7) explain that early successes with
a 1965 French immersion project led to the emergence of ‘a mass educa-
tional movement’ where Canadian school children received ‘much or all
of their school instruction through a second language.’ They also describe
bilingual secondary programmes where students took forty to fifty per
cent of their classes in French.
Immersion programmes are considered to be highly successful.
Brinton et al. (2003: 9) report that ‘immersion students achieve a high
level of functional ability in the second language with near-native profi-
ciency in receptive skills by the time they graduate.’ Skehan (1998: 12)
notes that ‘immersion educated children reach much higher levels of
achievement than do children educated by traditional “core” methods.’
Krashen (1985: 16) comments that immersion programmes have shown
teaching subject-matter is language teaching when it is made compre-
hensible.
However, some educators point out that they are not an unqualified
success. Lyster (2019: 12) draws attention to Swain’s concern ‘about the
L2 development of French immersion students, stating that “some of the
errors do not disappear” even after many years in immersion.’ Skehan
(1998: 12) notes that ‘although the children concerned perform at levels
of comprehension close to native speakers, the same cannot be said of
their productive skills.’ In investigating why this may be the case, Swain
and Lapkin (1986, cited in Lyster, 2019: 14) observed that there was
little information exchange in immersion classes and there was not much
discussion of ideas. In addition, Brinton et al. (2003: 48) point out that
‘students with either too low or too advanced second language profi-
ciency do not benefit fully from sheltered courses. If their skills are not
adequate for understanding the lectures and readings with reasonable
efficiency, they will fail at the subject matter and not learn very much
language either.’
The Comprehensible Output Hypothesis

On the basis of her observations that learners developed high levels of
proficiency but did not achieve syntactical accuracy at the same level as
first language users, Swain (1995: 125) drew a deliberate contrast with
Krashen’s concept of Comprehensible Input and proposed a Compre-
hensible Output Hypothesis, or as Skehan (1998: 16) puts it ‘to learn to
speak, we actually have to speak!’
While the Output Hypothesis proposes a number of ways that
speaking could improve performance, Swain (1995: 141) stresses three
core functions which might have a positive effect on accuracy. The first of
these is what Swain refers to as noticing. During speaking a learner might
notice a gap between what they want to say and what they can say. Swain
(1995: 126) speculates that this ‘may trigger cognitive processes which
might generate linguistic knowledge that is new for the learners, or which
consolidate their existing knowledge.’ However, she offers no explanation
of how cognitive processes might generate new linguistic knowledge. The
most likely mechanism is that noticing a gap primes the learner to iden-
tify the correct form in subsequent input, but this relies on the form
subsequently occurring, and the learner remembering to notice it.
A second function of output is hypothesis testing. In this case, a
learner tries out something to see if it is correct. The success of this
language learning strategy is dependent on the reaction of the listener,
who might look confused, tell them that they are wrong and correct
them, or not react visibly. While only the second type of response
would teach learners how to say something correctly, Skehan (1998: 16)
points out that output can act ‘as a signalling device to negotiate better
input’—presumably for more proactive learners.
A third function of output is what Swain calls ‘meta-talk.’ She
contrasts this with negotiation of meaning and refers to it as ‘negoti-
ation of form.’ Specifically, a pair (or group) of learners might discuss
the grammar of what one of them is trying to say. However, there is no
guarantee that learners will arrive at the correct target form, and her own
72 I. Pemberton
research (1995: 135) indicates that, through this process, a proportion of

learners will go on to internalize inaccurate form.
6.2 Task-Based Instruction

Another key pedagogy that has emerged from Communicative Language
Teaching is Task-Based Instruction. Randall (2007: 158) comments that
task-based approaches “are often seen as being at the ‘cutting edge’ of
language learning approaches.” In 2019, three of the five nominations for
the prestigious British Council ELTons award for Excellence in Course
Innovation were task-based approaches. However, it is debatable whether
Task-Based Instruction can be innovative, given its emergence some
thirty years earlier. Brown (2001: 50) explains that ‘task-based instruc-
tion is not a new method’ but simply one that ‘puts task at the center of
one’s methodological focus.’
The unit of analysis in task-based learning is the task. Long and
Crookes (1992: 27) make a distinction between two syllabus types:
synthetic and analytic—the distinction originally made by Marcel over
one hundred years earlier and drawn upon by Wilkins as a basis for his
Notional-Functional syllabus. A synthetic syllabus is one which focuses
on teaching different components of a language separately, for example,
structures, lexis, functions, or notions. This type of approach relies on a
learner’s ability to recombine the different components for communica-
tive purposes. On the other hand, the task-based syllabus is analytic. This
means that learners are presented with whole dialogues which haven’t
been analysed by a teacher. In this case, it is the learner’s task to parse
the discourse into its component parts.
6.2.1 Defining Task
One early attempt to define tasks often cited in the literature is Breen:
Task is therefore assumed to refer to a range of work-plans which have

the overall purpose of facilitating language learning – from the brief and
simple exercise type to more complex and lengthy activities such as group
problem solving or simulations and decision making. (Breen, 1987, in
Willis & Willis, 2007: 12)
Willis and Willis (2007: 12) make the point that this definition is
of limited usefulness as it could cover anything that takes place in the
classroom. A number of linguists offer more focused definitions. Cook
(2008: 257), citing Bygate (2001), explains that ‘a task is an activity
which requires learners to use language, with emphasis on meaning,
to attain a goal.’ Jane Willis (1996) provides a very similar defini-
tion: ‘a goal-oriented activity in which learners use language to achieve
a real outcome.’ These definitions are still considered to be vague,
so consideration has also been given to what is NOT a task. Willis
(1996) explains that any activity which has a primary focus on practising
specific language forms such as a pre-scripted dialogue or an information-
gathering activity intended to practise a single structure (e.g., Do you
like -ing?) would not count as a task. Skehan (1998: 96) further clar-
ifies that form-focused ELT practices such as transformation exercises
and students discovering the rules of grammar inductively would also
not count as tasks. Richards (2017: 175) adds drills, cloze activities, and
controlled writing activities to the list. He states that these activities are
better described as exercises. Usefully, Willis and Willis (2007: 13) set
out a number of questions to help teachers evaluate whether an activity
counts as a task or not:
1. Does the activity engage learners’ interest?

2. Is the primary focus on meaning?
3. Is there an outcome?
4. Is success judged in terms of outcome?
5. Is completion a priority?
6. Does the activity relate to real-world activities?
Finally, Skehan (1998: 96) notes that even with a set of questions,
judgement may sometimes be subjective and that activities may be better
thought of as points on a cline from less task-like to more task-like in
character.
74 I. Pemberton
6.2.2 Task Types
Tasks can be based on texts, or they can be topic-based. A topic can act
as an umbrella for different types of tasks (Willis, 1998). Nunan (2004:
131) expresses a preference for a topic-based approach, commenting that
‘it affords maximum flexibility and allows me to bring in a wide variety
of content that can be tailored to learner needs.’ To further clarify, Willis
and Willis (2007: 66) provide a typology of tasks which are based on
different cognitive processes.
• Listing: Examples of listing include brainstorming the qualities of a

world leader or criteria for choosing a holiday destination. Learners
might also use leaflets or the web for fact finding; for example, they
could list up five facts about a volcano or, for more advanced students,
a list of principles for ecotourism.
• Ordering, Sorting, Classifying: These tasks are more challenging than
listing tasks. Learners can sequence pictures to make a story or steps
in a recipe. They can sort the qualities of world leaders or holiday
destinations according to criteria such as importance or popularity.
Learners can be given classes and sort things into them e.g., healthy
and unhealthy foods or more proficient learners can create their own;
for example, categories for groups of objects such as junk we carry
around in our bags.
• Matching: Lower-level learners can do listen and do activities, like
performing actions or matching descriptions to pictures. Similarly,
they can read and match short texts to pictures.
• Comparing: Advanced learners could compare the strategies that they
use for language learning. They could also compare different news-
paper articles on the same topic or advertisements screened in different
countries but for the same product.
• Problem Solving: Learners can be asked to discuss solutions to envi-
ronmental (e.g., pollution), social (e.g., teenage issues), or work-
related (e.g., inequality) issues. Problem solving can be combined with
tasks which generate other cognitive processes such as listing and
ranking solutions.
• Creative Tasks, Project Work: These are more cognitively challenging

tasks which are normally done in groups over a longer period of
time. Learners could create a newspaper, make a radio programme,
or research and deliver a presentation.
• Sharing Personal Experiences: Willis and Willis (2007: 105) assert that
personal anecdotes are a major component of expert speaker language
use. They also recommend task repetition to enhance fluency. Learners
can relate experiences such as their most memorable childhood expe-
rience or the funniest person that they know.
6.2.3 Terminology
Tasks are described in several different ways by different writers. Willis

and Willis (2007: 21) make the point that a task-based lesson does not
focus on a single task but will include a sequence of interconnected tasks.
Broadly speaking, these tasks can be divided into two types: facilitating
and target tasks. Facilitating tasks are those tasks which help to prime
or prepare learners for the main or ‘target’ task. Facilitating tasks will be
used during the pre-task or preparatory phase.
Willis and Willis (2007: 64) make a further distinction between peda-
gogic and real-world tasks. Pedagogic tasks have a linguistic outcome,
for example, a dictogloss activity. On the other hand, real-world tasks
resemble things that a learner may do in the outside world, for example,
take part in an interview, book a holiday, or peruse an earthquake safety
procedures leaflet. Long (1985), self-citing in Long and Crookes (1992:
43) adds everyday events such as buying a train ticket, making an airline
reservation, or taking a driving test to the list of real-world tasks. Peda-
gogic tasks may be used as pre- and post-task activities, whereas target
tasks will (ideally) reflect real-world tasks.
Nunan (2004: 19) introduces two further technical terms: rehearsal
tasks and activation tasks. Both types of task have a non-linguistic
outcome: a rehearsal task is a real-world task but an activation task
does not relate to the real world. For example, learners could be given
a scenario such as being shipwrecked and deciding what items they can
carry to a desert island.
76 I. Pemberton
6.2.4 Learning Theory
Advocates of task-based learning believe that participation in tasks results

in language acquisition. According to Richards (2006: 30), they claim
that the interactional processes required by tasks result in language
learning. Nunan (2004: 36) asserts more strongly that ‘learners learn
best by actively using the language they are learning.’ He speculates that
creative language use in tasks integrates language skills and resources
which leads to maximum acquisition (ibid.: 20). Skehan (1998: 4) also
asserts that performance in tasks leads to interlanguage development.
6.2.5 Willis’s Task Cycle
Inputs and outputs are interconnected but can be considered by exam-

ination of task cycle. Proponents of TBI agree that a lesson should be
divided into distinct phases. However, they differ in how many activities
there should be, and in what order different activities should take place.
A highly influential model is Willis’s three-stage task cycle.
The Pre-Task Stage: In the first stage in the task cycle, the pre-task,
Willis (1996) explains that the teacher can set the objectives of the
lesson, and brainstorm ideas and vocabulary, but not pre-teach struc-
tures. Skehan (1998: 137) comments that the introduction of new
language is important because ‘the impetus for change has to come
from somewhere, and the pre-task stage can be useful as one means of
introducing new elements into the interlanguage system.’
Willis (1996) suggests that after the introduction of useful vocabulary,
learners can then read something about the topic or listen to a recording
of a task which is similar to the one that they will perform, and that this
will also provide useful language which can be deployed during the task.
The Task Stage: Willis divides this into three steps.
1. The Target Task

2. Preparation
3. Reporting
The first step is the task itself. During this stage, commentators agree
that it is important that learners are free to make their own language
choices. Richards (2006: 33) notes that learners work in groups or pairs,
and it provides them with the opportunity to use whatever language they
wish to fulfil the task. Similarly, Nunan comments that learners ‘are free
to use any linguistic means at their disposal to complete the task.’ Skehan
(1998: 127) comments that the target task has the potential to develop
learners’ fluency, accuracy, and complexity as they use language and test
their hypotheses about it.
After completing the main task, Willis (1996) suggests a planning
phase and then a reporting phase. In the planning phase, learners
consider how they are going to report the outcome of the task. When
planning is complete, at least some of the learners will report back
on their task performances to the whole class. Skehan (1998: 148)
comments that the reporting phase might actually be more appropriately
grouped together with post-task activities.
Willis (1996) suggests one further activity after the reporting has taken
place. Whereas learners listened to a recording of native speakers doing
a parallel task before they attempted their own task, they now listen to
a recording of native speakers doing the same task and compare it with
their own performance.
The Post-Task Stage: In the final phase in the three-stage task cycle,
learners focus on form. If students have used written or spoken texts
earlier in the cycle, these are re-introduced for language analysis. Willis
(1996) suggests various activities including choral drilling of phrases,
work on recalling them, or matching verbs with their subjects and
objects.
A further reason relates to learnability. Long and Crookes (1992: 31)
point out that ‘where syntax is concerned, research has demonstrated that
learners rarely, if ever, move from zero to target-like mastery of new items
in one step.’ Thus, Willis and Willis (2007: 18) argue that when struc-
tures are introduced at the beginning of a lesson, learners will end up
using them inaccurately leading to a sense of failure.
78 I. Pemberton
6.2.6 Nunan’s Task Cycle
Task-Based Instruction advocate David Nunan recommends a six-step

task cycle.
Step 1: Nunan (2004: 31) describes this step as a schema building

phase which not only creates a context for the task, but also highlights
key language that the students will need to complete the task.
Step 2: Nunan (2004: 32) recommends controlled practice. He
suggests that teachers play a model conversation while learners read
it. They then rehearse in pairs, perform with variations, and act
it out without reading. Nunan concedes that this may appear very
similar to Audiolingualism, which it does. However, he asserts that ‘the
learners have been introduced to the dialogue within a communicative
context.’
Step 3: Nunan next suggests further listening exercise, such as
listening to multiple conversations and matching them to newspaper
advertisements, which extends the vocabulary introduced in the model
conversation.
Step 4: The fourth step is to get learners to listen to the conver-
sation again but with a specific focus on language elements, for
example, intonation. He contrasts this with what he calls ‘a more tradi-
tional synthetic approach’ where he asserts that the focus on language
elements would occur as Step 1. However, in his task-based approach,
he points out that prior to analysing the language system, learners have
‘seen, heard and spoken the target language within a communicative
context.’
Step 5: In describing Steps 1–4 as reproductive practice, learners now
move to freer practice, and this might consist of something such as an
information gap role play. Nunan points out that while some students
will ‘stick to the script,’ others will be more innovative. Students’ inter-
language will initially be idiosyncratic, but he believes that it will grow
closer to native speaker norms over time.
Step 6: The final stage of the task cycle is the main or target task.
Having already analysed the language system, this will be the final
phase for Nunan’s learners.
6.2.7 Focus on Form
The placement of the form-focused practice exercises is one of the main

ways that Task-Based Instruction differs from mainstream Commu-
nicative Language Teaching. The major difference is that the language
analysis comes at the end of the sequence of activities. Willis (1996)
refers to this as PPP upside down.
Proponents make a number of arguments for placing language analysis
after practice. Skehan (1998: 128) asserts that focusing on form at the
end ensures language is both relevant and communicative, not the next
point in a syllabus. Willis and Willis (2007: 17) argue that when students
focus on pre-taught forms during the task, they are less likely to integrate
new language into their existing interlanguage and lose the opportunity
to make fluency and confidence gains.
However, not all proponents agree that focus on form should come at
the end of the task cycle. Nunan believes that it should come later than
traditional treatments but before the main task, and this allows learners
to practice the forms during the task itself. In other words, Nunan’s
approach retains the Present-Practice-Produce paradigm.
6.2.8 Negotiation of Meaning
A key activity in a task-based lesson is ‘negotiation of meaning.’ While

learners are interacting with each other in the target language, they will
encounter language that they do not understand. When this happens,
they have to negotiate with each other to determine the meaning using
strategies such as asking questions. In speculating how this might lead
to language development, Skehan (1998: 17) explains that ‘engaging in
meaningful negotiation … evidences efficient signalling of miscompre-
hension and the clear engagement of a malleable interlanguage system
which is more likely, as a result, to develop productively.’
80 I. Pemberton
6.2.9 Criticisms
Task-Based Instruction has its share of critics. A common criticism is

that there is a lack of evidence that it works. In a wide-ranging anal-
ysis of task-based transcripts, Seedhouse (1999: 150) comments that he
was unable to find any studies demonstrating the benefits of task-based
instruction. Similarly, Swan (2005: 396) notes that empirical evidence
for the claims made by proponents is sparse. When comparing with a
standard P-P-P approach, Richards (2006: 35) remarks that there is not
much evidence that task-based instruction is any more effective. Skehan
(1998: 4) comments that there is a risk that engaging in meaningful
communication de-emphasizes form.
An important criticism is that tasks do not provide enough new
language input. As early advocates wanted students to draw upon their
existing interlanguage, they actively avoided pre-empting the language
that students would use. While more recent task-based approaches
include a pre-task phase, Swan (2005: 391) argues that this is mainly for
orientation purposes. He (ibid.: 392) concedes that for more advanced
learners the opportunity for practice may be welcome but argues that
tasks do not provide enough input for less proficient learners.
A number of criticisms have been made of learner outputs. One crit-
icism is that they are not of a high enough quality to promote mutual
interlanguage development. In an analysis of over three hundred tran-
scripts of task-based interactions, Seedhouse (1999: 152) observes that
learner outputs demonstrate consistent irregularities and these may lead
to grammatical simplification and limit vocabulary development. While
recognizing that some positive transfer can occur between learners, Swan
(2005: 389) notes that negative transfer will also occur. Furthermore,
learner outputs tend to be minimal. Seedhouse (1999: 153) found that
the learners use only enough language to complete the task. Finally,
learner outputs are inexplicit as meaning tends to be facilitated by the
common context (e.g., deictic elements).
A number of criticisms have also been made of the use of negotia-
tion of meaning as a key second language acquisition strategy. To begin
with, Krashen (1985: 35) comments that it cannot account for acqui-
sition which occurs with little or no production. While not denying its
occurrence, Allwright and Hanks (2009: 98) point out that it is diffi-
cult to establish its impact on acquisition. Aston (1986) cited in Skehan
(1998: 20) considers that such negotiations can be irritating for students,
and not representative of what happens in normal spoken interaction.
Its importance is brought into question by Nation (2001: 124) who
reports that as the main goal is to complete a task, this necessarily limits
the amount of language learners negotiate. In fact, Skehan (1998: 20)
cites research by Foster (1998) who found that ‘most individual students,
whatever the task or participation pattern, engage in the same amount
of negotiation of meaning – nil.’
6.3 Summary
This chapter has looked at two process-based communicative teaching
approaches: Content-Based Instruction and Task-Based Instruction.
Content-Based Instruction has been divided into three broad categories:
theme-based instruction, sheltered instruction, and adjunct instruction
which are appropriate for EFL, EMI (English as a Medium of Instruc-
tion), and EAP, respectively. Content-Based Instruction promotes high
levels of language proficiency, but it does not result in first language
levels of proficiency in the productive skills of writing and speaking. This
led Canadian linguist Merrill Swain to formulate her Output Hypoth-
esis which argued that output is a necessary condition for acquisition.
Swain suggested three roles for output: noticing a gap, hypothesis testing,
and discussing grammatical form. However, none of these three func-
tions guarantee the presentation of correct forms to the learner, and this
seriously weakens the ability of the Output Hypothesis to explain how
acquisition takes place.
Task-Based Instruction means different things to different teachers.
However, advocates agree on some points. They are agreed that a task is
a meaning-focused activity in which learners are free to use any language
that they know to complete the task. Additionally, they agree that a
task should have a clear goal and, ideally, a relationship to the real
world. While tasks can be classified in different ways, an influential
typology is Willis’s classification according to the cognitive processes
82 I. Pemberton
they employ. This allows them to be matched to different levels of

learners. In order of increasing complexity these are: listing, ordering
and sorting, matching, comparing, problem solving, doing projects, and
sharing personal experiences.
While proponents agree that a macro-task should be divided into iden-
tifiable steps, they are not in agreement about how many or what those
steps should be. However, there is agreement that a task cycle should
start with schema building by eliciting background knowledge followed
by a listening activity to extend relevant vocabulary knowledge. There is
significant disagreement on where to place focus on form. While Nunan
places it before the main task, Willis places it afterwards. Nunan thinks
that considering forms in advance promotes the development of accuracy
during use, whereas Willis thinks that focus on form during performance
reduces the likelihood that new forms will be integrated into a learner’s
interlanguage.
There have been a range of criticisms of Task-Based Instruction, not
only from its detractors, but also from its proponents. The main criti-
cisms are that there is a lack of research that shows that it works, it doesn’t
provide enough input to promote language acquisition, the language that
learners produce and exchange does not possess the qualities which will
result in acquisition, and focusing on a task de-emphasizes focus on form.
References
Brinton, D., Snow, M. A., & Wesche, M. B. (2003). Content-based second
language instruction (Michigan classics ed.). University of Michigan Press.
Hodder Education.
Long, M. H., & Crookes, G. (1992). Three approaches to task-based syllabus
design. TESOL Quarterly, 26 (1), 27–56.
Lyster, R. (2019). Pushing immersion forward. In M. Haneda & H. Nassaji

(Eds.), Perspectives on language as action (pp. 11–26). Multilingual Matters.
Nation, I. S. P. (2001). Learning vocabulary in another language. Cambridge
University Press.
Nunan, D. (2004). Task-based language teaching. Cambridge University Press.
Richards, J. C. (2006). Communicative language teaching today. Cambridge
University Press.
Richards, J. C. (2017). Curriculum development in language teaching (2nd ed.).
Seedhouse, P. (1999). Task-based interaction. ELT Journal, 53(3), 149–156.
Press.
Snow, M. A., & Brinton, D. M. (2017). The content-based classroom: New
perspectives on integrating language and content (2nd ed.). University of
Michigan Press.
Swain, M. (1995). Three functions of output in second language learning. In
G. Cook & B. Seidlhofer (Eds.), Principle & practice in applied linguistics:
Studies in honour of H. G. Widdowson. Oxford University Press.
Swan, M. (2005). Legislation by hypothesis: The case of task-based instruction.
Willis, D., & Willis, J. (2007). Doing task-based teaching. Oxford University
Press.
Willis, J. (1996). A flexible framework for task-based learning. In J. Willis &
D. Willis (Eds.), Challenge and change in language teaching. Heinemann.
Willis, J. (1998). Task-based learning. What kind of adventure? The Language
Teacher, 22(7).
7
The Lexical Approach
One of the few approaches that has emerged since the arrival of the
Communicative Approach is the Lexical Approach. While it has not been
widely adopted, there is almost universal agreement on its significance
for language teaching and learning. In contrast to structural approaches,
a Lexical Approach considers lexis as the main organizing principle
of language teaching and learning. Richards and Rodgers (2001: 132)
state that ‘lexical approaches in language teaching reflect a belief in the
centrality of the lexicon to language structure, second language learning,
and language use.’ Advocates of the Lexical Approach do not draw a
distinction between grammar and lexis, rather they see language as occu-
pying a continuum from concrete items at one end to fully abstract
expressions at the other. Author of the Lexical Approach, Michael Lewis
(1997: 3), argues that the components of language are not grammar and
vocabulary but prefabricated multi-word chunks.
In the early 1990s, evidence of the existence of these multi-word
chunks emerged from the analysis of language corpora in the field of
corpus linguistics. This led John Sinclair (1991: 110) to formulate his
idiom principle which states that ‘a language user has available to him or
her a large number of semi-preconstructed phrases that constitute single

Switzerland AG 2024
86 I. Pemberton
choices, even though they might appear analyzable into segments.’ The
existence of preconstructed phrases highlights the scale of the task faced
by language learners. Lewis (1993: 92) asserts that a language consists
of hundreds of thousands of multi-word chunks. Harmer (2007: 74)
provides some familiar examples:
How are you? See you later. You must be joking. I’ll give it my best shot.
Changing the subject slightly. Might as well …. If it’ll help ….
Harmer (ibid.) notes that the emergence of the Lexical Approach

played an important role in encouraging debate about what language
learners should learn. On this point, Lewis (1993: 17) is unequivocal,
arguing that the Lexical Approach stipulates a much stronger focus
on building a bigger vocabulary at a faster rate than any conventional
syllabus.
7.1 Lexical Categories

Proponents of the Lexical Approach analyse lexis into a number of
different categories although the distinction between categories is not
always clear cut. While there is some variation in the categories and
terminology used, common categories are:
1. Words
2. Collocations
3. Fixed or institutionalized expressions
a. Fully fixed expressions
b. Semi-fixed expressions
7.1.1 Words
Lewis (1997: 8) remarks that the first category will be familiar to teachers
as vocabulary but it also includes contractions, compound nouns, and
polywords. Lewis (1997: 22) holds the view that contractions should
be included in the category, arguing that “can’t” and “don’t” represent

independent lexical items or single words. The category also includes
polywords. According to Nattinger and DeCarrico (1992: 38), poly-
words are short invariable phrases performing a similar function to
individual words. Lewis (1997: 22) characterizes them as short two- or
three-word phrases which are obvious units. They include compound
nouns, which are two-word units that are so closely related that dictio-
naries list them as single entries; for example, taxi rank, put off, and of
course. They also include three-word phrases such as: all in all , by the
way, by and large, and up to now. Lewis calls these ‘words-with-spaces-
in-them.’ Nattinger and DeCarrico (1992: 39) note however, that in
some cases the spaces have disappeared as a result of usage; for example:
moreover, however, nevertheless and notwithstanding.
7.1.2 Collocations
The next category, which will probably be the most familiar to language
teachers are collocations. Richards and Rodgers (2001: 133) explain that
collocations are words which regularly occur together. Lewis (1997:25)
adds that collocations are words which occur together at a frequency
greater than random. Sinclair (1991: 112) points out that while collo-
cations vary in strength, there are many strong collocations such as hard
luck, hard work, hard evidence, hard facts.
However, collocation is an arbitrary phenomenon and different
languages will collocate words in different ways; for example, an English
speaker will talk about a strong wind , heavy rain and heavy or deep snow.
On the other hand, a Japanese speaker will talk about a strong wind but
strong rain and deep but not heavy snow. Whereas strong rain is marginally
possible in English, we don’t say strong snow. The fact that we generalize
some adjectives to particular nouns but not others clearly indicates that
we experience and store language as lexical chunks but that these do not
necessarily translate into other languages.
Lewis (1997: 30) continues that while collocation is a powerful way
to learn language, learners require assistance to identify them in a text.
Also, the fact that two words may be adjacent in a text does not guarantee
88 I. Pemberton
that they collocate. On the other hand, many collocations are spatially
separate.
7.1.3 Fixed (Institutionalized) Expressions
This category can be divided into those expressions which are fully fixed,
and others which are semi-fixed ‘frames’ with ‘slots’ which may be filled
in a limited number of ways (Lewis, 1997: 9)
Fully fixed expressions:
There is general agreement amongst Lexical Approach advocates
about the nature of fully fixed expressions. Early advocates of a Lexical
Approach, Pawley and Syder (1983: 191), refer to fully fixed multi-
word expressions as ‘institutionalized’ expressions. While their language
is somewhat rigidly formal, their intention here is to indicate that they
are conventionalized through use and that the same or very similar
expressions are employed by almost all other speakers of a language.
Lewis (1993: 95) picks up on their terminology and describes fully fixed
expressions as complete sentences with a clear pragmatic meaning and
easily identifiable as such. Nattinger and DeCarrico (1992: 39) agree
with Lewis that institutionalized expressions are of sentence length and
add that they are invariable and mostly continuous. They (1992: 40)
agree with Pawley and Syder that some may be phrases in general use
in a speech community (e.g., How do you do? How are you? ) but point
out that some might be idiosyncratic phrases that individuals have found
effective in getting their meaning across (e.g., give me a break, have a nice
day). By way of illustration, Pawley and Syder (1983: 206–7) provide
the following examples:
Can I come in? Can I help? Need any help? Need a hand? Are you all
right? Is everything OK? Are you ready? What did you say? What’s for
dinner? … I see what you mean. I’ll believe it when I see it. That’s easier
said than done. I thought you’d never ask. He never has a bad word to
say about anyone. If I’d known then what I know now. He’s not the man
he used to be.
Semi-fixed expressions
Advocates agree that by far the more interesting category is semi-fixed
expressions. Lewis (1997: 43) considers that this category is important
for understanding how language is learnt and its representation in the
lexicon. Pawley and Syder (1983: 205) provide a ballpark estimate of
how many of these expressions there may be. They estimate that while
an adult English speaker knows many thousands of complete clauses and
sentences, phraseological expressions which are not completely specified
clauses are much more numerous.
Lewis (1997: 34) explains that ‘in general, semi-fixed expressions
consist of a pragmatic (or functional) frame, which is completed by a
referential slot filler.’ Pawley and Syder (1983: 205) describe these expres-
sions as ‘sequences which contain a nucleus of fixed lexical items standing
in construction with one or more variable elements (often a grammatical
inflection), the specification of the variables being necessary to complete
the clause.’
The category can be broken down further. Nattinger and DeCarrico
(1992: 41) analyse semi-fixed expressions into phrasal constraints and
sentence builders. They explain that phrasal constraints are short- and
medium-length phrases which allow for variation of both lexical and
phrasal (grammatical) categories. Examples include:
A _____ ago. To _____ this up. As I was _____. See you _____. _____
as well as _____. As far as I _____. The _____er the_____ er. What with
_____ (and all).
Their second category, sentence builders, are lexical phrases which

create a frame for complete sentences. These frames contain slots for
parameters (arguments) to express a complete idea. They allow consider-
able variation of phrasal (grammatical) and clausal (semantic) elements.
Nattinger and DeCarrico (1992: 43–44) provide the following examples:
I think that X. Not only X but also Y; my point is that X; I’m a great
believer in X; The _____er X, the _____er Y.
90 I. Pemberton
Lewis (1993: 94) adds to these what he refers to as sentence heads or

frames. These include the first words of utterances and serve a mostly
pragmatic purpose; e.g., sorry to interrupt but can I just say … That’s all
very well, but … I see what you mean but … I wonder if it wouldn’t be
better to ….’
7.2 Theory of Language

According to proponents of the Lexical Approach, multi-word chunks
are crucial to explain language use and language understanding. It might
be useful to start here with a well-known quotation from Michael Lewis
(1993: 89):
Language consists of grammaticalized lexis, not lexicalized grammar.
What Lewis means by this is that rather than creating sentences

from scratch each time, that is, bringing grammatical structures into
working memory and populating their abstract functional categories
with words, language users draw upon prefabricated chunks held in long-
term memory, grammaticalize (inflect) them and string them together to
suit communicative intent.
In other words, there are potentially two ways of making a message.
Lewis (1993: 96) cites Cowie (1988) who asserts that language speakers
can produce the same language in one of two different ways. They can
use their acquired competence to generate it, or they can recall whole
sentences that they have previously learnt. Lewis (1997: 22) asserts his
own view that rather than constructing multi-word units, speakers recall
them directly from memory. Other linguists are in general agreement.
Pawley and Syder (1983: 205) contend that only a small number of
utterances feature a pattern of lexical items which is completely original
and sequences stored in memory form a high proportion of everyday
conversation (ibid.: 208).
Given that there are two ways of making the same message, there are
a number of compelling reasons why language users might choose to use
formulaic sequences rather than construct messages from scratch. The
first reason is that formulaic expressions reduce cognitive load during

processing and production. Lewis (1993: 90) explains that ‘speech may
be processed more rapidly—both receptively and productively—if units
are perceived as single unanalyzed wholes.’ Wray and Perkins (2000: 15)
state that ‘it makes little sense to produce from scratch those word strings
which we use many times, and we appear to use formulaic sequences
to reduce the amount of new processing to only that which has to be
new.’ Pawley and Syder (1983: 192) explain that in recalling expres-
sions whole from long-term memory, a speaker reduces clause-internal
encoding work and frees themselves to attend to other tasks such as the
forward planning of subsequent messages.
In addition to reducing the cognitive load of the speaker, there is
a second important reason why speakers preferentially choose to use
formulaic expressions: the use of formulaic language also facilitates
understanding for the hearer. Wray and Perkins (2000: 18) explain that
the main driver of socio-interactional formulas is to ensure that commu-
nication is successful and formulaic sequences are better for this purpose
than original language as hearers more likely to comprehend speech
which is similar to what they have heard previously and can understand
without a full analysis.
In summary, formulaic language bestows two important advantages on
the participants in a conversation. It reduces the amount of encoding that
needs to be done in working memory which helps a speaker to construct
a message more fluently, and the familiarity of multi-word expressions to
a hearer makes it more likely that they will be understood.
7.3 Learning Theory

In setting out the different categories of multi-word expression, Lewis
alludes to the importance of such expressions for language acquisition. To
recap, he (1997: 43) states that ‘there is an important category of semi-
fixed items, frames with slots, which fit somewhere between traditional
fixed words and generative grammar. This category may be central to
understanding how language is acquired and stored in the lexicon.’
92 I. Pemberton
Lewis (1993: 95) explains that ‘it now seems plausible that an impor-
tant part of language acquisition is the ability to produce lexical phrases
as unanalyzed wholes or “chunks”, and that these chunks become the raw
data by which the learner begins to perceive patterns, morphology, and
other features of language traditionally thought of as grammar.’ While
Lewis is not particularly explicit, the basic proposal here is that learners
create an inventory of concrete linguistic items and then analyse these
concrete examples of language to identify grammatical categories and
create more abstract syntactic patterns.
7.4 Methodology
Lewis (1993: 104) argues that, historically, we have analysed language
into words and grammatical structures but that this analysis is incorrect,
or at best, unhelpful. A language is more usefully analysed into lexis; that
is to say, meaningful phrases of, generally, more than one word.
He (1997: 86) asserts that languages are best learned by doing a
large amount of listening to acquire lexis, and then complementing that
with an equally large amount of reading. This sounds very similar to
Krashen’s Natural Approach, but in contrast to Krashen, Lewis asserts
that comprehensible input alone is not enough, and that learners benefit
from direct instruction. According to Lewis (1993: 104; 1997: 117),
effective instruction should focus on three things:
1. how to identify lexis

2. how to record it in helpful ways
3. practice to move it from short-term to long-term memory
According to Lewis (1997: 85), after noticing lexis in texts, learners

should record it in lexical notebooks. In addition to encouraging
noticing, he proposes teachers use activities and exercises. Activities are
cooperative procedures done together with other learners in the class-

room, whereas exercises are individual reflective tasks which may be done
either in the classroom or at home (Lewis, 1997: 86).
Lewis (1997: 89) proposes six different categories of exercise:
1. identifying chunks
2. matching
3. completing
4. categorising
5. sequencing
6. deleting
Activities include drills, discussions, and production.
7.4.1 Identifying Chunks
Lewis (1993: 122; 1997: 70) explains that the identification of lexical
chunks (including but not limited to collocations) is central to the
Lexical Approach. He suggests a range of pedagogical activities. The first
of these is underlining chunks as they occur in texts (Lewis, 1997: 108).
This assumes that learners are able to do so but Lewis provides a caveat
that even where they are unable to, it will give the teacher an idea of what
learners regard as chunks. A second activity offers more guidance. In this
case, teachers can provide a list of words which collocate in a text (e.g.,
verbs or headword nouns) and ask learners to scan the text to identify
the collocates. A third activity might involve asking students to pick out
all the nouns in a text and then review the text to find their collocates.
A fourth activity is based on the green cross code:
When you see a word, even if it is a word with which you are familiar,
STOP, LOOK LEFT, LOOK RIGHT, LOOK LEFT AGAIN, AND,
WHEN SATISFIED, PROCEED. (Lewis, 1997: 111)
94 I. Pemberton
Further activities involve analysing corpus-generated concordance

lines to pick out collocates of a particular word. (Lewis, 1997: 112)
In summary, identifying chunks involves the analysis of text using
both free and guided activities, and decontextualized activities involving
dictionaries and corpus data. Lewis’s proposals for activities and exercises
supplement identification of contextualized lexis with procedures and
games that present lexis in a decontextualized manner but are intended
to encourage recall and thus help to move lexis from short-term to long-
term memory. While they will be familiar to most language teachers, they
are given a lexical slant.
7.4.2 Matching
Lewis (1997: 114–117) suggests a variety of matching exercises of

differing levels of granularity. Learners can match paragraph headings to
paragraphs, sentence heads to sentence endings, or words which collocate
and their collocations. In addition, they can play traditional language
teaching games such as word dominoes or the well-known card game,
Happy Families. In all cases, these activities and exercises encourage
recognition of polywords and collocations.
7.4.3 Completing
Lewis (1993: 126–7) draws upon traditional cloze procedures to create

activities and exercises. Learners can complete lists of polywords or collo-
cations or continuous texts that are given a lexical focus by deleting words
which form part of a collocation or a fixed expression. He makes the
point that these cloze exercises are not intended to teach lexis but to
encourage its recall.
In addition to these basic exercises, Lewis suggests two more chal-
lenging activities. The first is to provide a short talk together with a
gapped transcript (Lewis, 1997: 136), and the second is to ask students
to watch a short extract from a soap opera while completing a gapped
transcript or task-sheet. Both of these activities direct their attention to
useful collocations.
7.4.4 Categorizing
To help learners to identify patterns in lexis, they can be asked to sort

collocations, expressions, or multi-word adverbials into different cate-
gories. They can either decide the categories themselves or work with
categories provided by the teacher. For example, ask learners to sort the
following multi-word adverbials into groups according to whether they
answer the questions When? Where? or Why?
At the side of the road, on the way to work, over the next few days, by the
end of the week, for a change, for some time, in the 60s (Lewis, 1997: 122).
7.4.5 Sequencing
Sequences, such as a series of connected events or stories, are some of

the earliest patterns that we learn to identify in language input, and
they reflect our experience of the world which helps us understand
them. Thus, Lewis (1997: 90) proposes that sequences can be effectively
employed to help learners remember sets of related lexis. One example
given is that of the active learning jigsaw activity where learners, each
with a different part of a connected dialogue, either walk around the
classroom assembling it, or they assemble the dialogue on their table
(Lewis, 1997: 136). Additionally, learners can rearrange the words in
sentences which have been mixed up.
7.4.6 Deleting
Lewis (1997: 91) suggests that deletion can be used as a form of negative
evidence to prevent overgeneralization errors. For example, learners may
be asked to decide which of a group of collocates is the odd one out, i.e.,
does not collocate with a particular word (Lewis, 1997: 94), or they may
be asked which of a number of words that fit in a slot is different from
the other words, e.g.,
You look a bit ……………….
worried under pressure stressed anxious off-colour

96 I. Pemberton
7.4.7 Drills
While language drills fell out of favour due to their association with the
behaviouristic approach of Audiolingualism, Lewis (1997: 127) advo-
cates asking students to practise repeating longer expressions to master
their phonological features.
7.4.8 Discussion
Although Lewis favours input activities to build learners’ lexicons, he

(1997: 123–4) suggests several discussion activities. These will most
likely work with more advanced learners who already have a working
understanding of the lexis. They are intended to impart pragmatic
understanding. For example, he asks learners to discuss the following
expressions which have the same function, and to decide which ones
they would use themselves.
How are you? How’s things? How’s tricks? What have you been up to
recently? What’s new with you? How’s life been treating you?
Another discussion activity involves deciding by whom, where, and why

the following expressions were used.
Don’t worry – I’ll pick you up / It’s the sort of thing you think can never
happen to you / Can I give you a hand? Do we have to ask your mother?
Not tonight. I’ve got a headache.
7.4.9 Production
Finally, Lewis suggests several output activities which provide learners

with the opportunity to practise the language. One of these (Lewis, 1997:
121), a familiar language teaching activity, is to ask learners to name loca-
tions on a picture using multi-word prepositional expressions. A second
is to play the well-known ELT activity, Kim’s game, showing learners a
picture and then asking them to not only remember objects, but also use
multi-word prepositional expressions to say where they are in a picture

(Lewis, 1997: 122).
In addition to spoken activities, Lewis (1997: 136) suggests a written
activity. Ask learners to read a short text such as an advertisement for
a foreign holiday with questions, comments, or reactions attached to it
and then write a letter of enquiry asking for more information about the
vacation or imagine they have been on the vacation and write a letter of
complaint.
Earlier, under completion activities, it was suggested that learners
watch part of an episode of a soap opera while completing a gapped tran-
script or task-sheet. The soap opera can be further exploited in output
activities by asking learners to summarize the episode or even act out a
scene from it.
7.5 Notebooks
A key pedagogic strategy of the Lexical Approach is to get learners to
record new vocabulary that they hear or read in a notebook. In Imple-
menting the Lexical Approach, Lewis (1997: 67–75) suggests ten different
principles for organizing the notebook. He comments that there is no
best way to organize lexis and that different ways are appropriate for
different lexical areas.
7.5.1 Organizing Principles
1. Topic: A traditional principle which groups related lexis into real-

world sets. For example,
• Food: chicken salad/rare steak/How would you like your steak?
Would you like a dessert?
• Office: plain paper/address labels/application form/Did you get
my memo/ Can you fill in this form, please?
2. Situation: Another traditional organizing principle. For example,
kitchen-related vocabulary.
98 I. Pemberton
• Where do you keep the (N: sugar)? is as much a part of kitchen

vocabulary as sieve.
3. Collocation: This category may be broken down according to collo-
cation types. For example:
• Strong collocations: have breakfast/lunch/dinner
• Adjective plus noun: bleak/daunting/dismal/exciting/vague
prospect
• Verb plus noun: dismiss/express/meet/raise/withdraw an objec-
tion
• Contextual opposites: A silly/bright idea. A cushy/challenging
job. A slight/serious hindrance.
4. Notion: A speech event such an apology may incorporate a number
of functions, e.g.,
• Apologizing: I’m sorry, I was miles away. I’m sorry about
yesterday, I was out of order. I shouldn’t have told her. I’m really
sorry.
• Making amends: I brought you a few flowers. I hope you’ll let me
pay for the damage.
• Reassuring: Don’t worry, I’ve done the same thing myself. Don’t
worry, it wasn’t your fault. You couldn’t have known.
5. Narration: Sequences of events such as stories, comic sketches,
instructions, and recipes. For example, verbs related to letter can be
organized into a narrative sequence:
• write, sign, address, put (into an envelope), affix (a stamp), post,
receive, reply to
6. Metaphor: Emotional concepts or personal relationships, e.g., Anger
is Fire.
• She was fuming. They had a blazing row. Smoke was coming out
of his ears. (Lewis, 1997: 71)
7. Person: First person helps students to talk about themselves.
• I’ll see you tomorrow, I’ve got something in my shoe. I haven’t

seen you for ages. I’m sure I’ve met her before.
8. Phonological Chunking: On the basis that it is easier to remember
a tune than a series of random notes, lexis can be categorized by
phonological features such as stress patterns.
9. Keywords: Key delexicalized verbs such as take, have, and make.
These are verbs with a core meaning but which are also used
idiomatically. For example,
• Take your time. Have a ball. Make your mind up.
10. Grammar: Lexis with similar syntactic patterns can be grouped
together allowing analysis and the emergence of more abstract
grammatical frames.
7.5.2 Formats
In addition to the ten macro-organizing principles, Lewis (1993: 124)

suggests three formats for recording lexis: collocation boxes, pattern
displays, and discourse structures.
1. Collocation boxes
These will generally illustrate adjective-noun, verb-noun, noun–verb,

and verb-adjective-noun combinations. For example:
(re) draft confidential letter
dictate sales
go through special offer
get urgent
fax promotional
Lewis (1993: 126)
2. Pattern Displays
Pattern displays are a good way to represent de-lexicalized verbs (e.g.,

get and have) and prepositions. For example, write a key word, then
100 I. Pemberton
enclose it in a circle (e.g., have). Next, group collocates into categories

and write them around the circle. For example, meals (breakfast/lunch),
food or drink (a coke, a cup of tea, an apple), free time (a day off,
a holiday, a break), or personal relationships (a friend in America, a
daughter in London). Link collocates to the encircled verb by drawing
lines.
3. Discourse Structures
These are larger stretches of text which represent formulaic interac-

tions or illustrate logical relations such as cause and effect, contrast, or
exemplification. Lewis (1993: 126) provides the following examples of
ways to greet workplace visitors:
• Good morning (Ladies and Gentlemen). On behalf of ….. may I welcome

you to ……
• It is a great pleasure to have you with us today. I hope you enjoy your visit/
meeting/the conference.
• If there is anything we can do to help, please do not hesitate to ask.
• Now, you don’t want to listen to me all day, so I’ll hand you over to my
colleague …… (who will show you…… /take you to……)
7.6 Criticisms
Lewis recognizes some of the limitations of his own approach. He admits
that it means that the amount of study learners have to do increases.
He (1997: 33) concedes that ‘the lexicon is somewhat larger than we
suspected, and the learning load is in some ways consequently greater.’
In addition, he loses the courage of his convictions somewhat, conceding
that a Lexical Approach alone is insufficient. After arguing passionately
that there is no distinction between grammar and lexis, Lewis (1997:
41) states that the Lexical Approach ‘recognizes clearly that lexis is
not enough and that courses which totally discard grammar are doing
learners a disservice.’
Richards and Rodgers (2001: 134) pick up on some of his self-doubts.

They point out that ‘if as Pawley and Syder estimate, native speakers have
hundreds of thousands of pre-packaged phrases in their lexical inven-
tory, the implications for second language learning are uncertain. How
might second language learners, lacking the language experiential base of
native speakers, approach the daunting task of internalizing this massive
inventory of lexical usage?’ Smelling blood, they continue their offen-
sive: ‘regardless of the learning route taken, a massive learning load seems
an unavoidable consequence of a Lexical Approach in second language
instruction.’
They (2001: 138) make one more criticism, this time, perhaps a fair
one: ‘Lewis and others have coined the term Lexical Approach to char-
acterize their proposals for a lexis-based approach to language teaching.
However, such proposals lack the full characterization of an approach
or method as described in this book.’ We have noted that the Lexical
Approach is somewhat vague on its theory of learning.
7.7 Summary
A Lexical Approach divides language into a number of categories, and
common categories are words, collocations, fully fixed expressions, and
semi-fixed expressions. Proponents of a Lexical Approach don’t differen-
tiate between vocabulary and grammar. They believe these things exist
on a cline, and they assert that much more focus needs to be placed
on teaching and learning lexis, as opposed to grammar. Advocates of
a Lexical Approach believe that pre-formed units of language above a
word level are essential to language understanding and use. They assist
language processing and production by reducing cognitive load for both
listeners and the speakers. Analysis of corpus data by researchers has
confirmed that people generally communicate using conventional expres-
sions, and the existence and importance of lexis are widely accepted by
both theorists and practitioners.
In his book Implementing the Lexical Approach, Michael Lewis
discusses methodology for teaching lexis, and he highlights the impor-
tance of identifying it in spoken and written texts and recording it in
102 I. Pemberton
notebooks. Once lexis has been identified, it can be practised through

exercises and activities. Exercises are individual endeavours which
strengthen mental representations of language through the recall and
application of lexis to such things as gap fills, categorizing, sequencing,
and deleting exercises. Activities are collaborative events which take place
in the classroom and can include drills (for pronunciation), describing
things, writing letters, and discussions. Despite its wide acceptance, some
important criticisms have been made of lexical approaches and the main
one is that the number of lexical units to be remembered is so vast that
it is a challenge to teach them in any systematic manner.
References
Harmer, J. (2007). The practice of English language teaching (4th ed.). Pearson
Longman.
Lewis, M. (1993). The Lexical approach: The state of ELT and a way forward .
Language Teaching Publications.
Lewis, M. (1997). Implementing the Lexical Approach. Language Teaching
Publications.
Nattinger, J. R., & DeCarrico, J. S. (1992). Lexical phrases and language
teaching. Oxford University Press.
Pawley, A. & Syder, F. (1983). Two puzzles for linguistic theory: Nativelike
selection and nativelike fluency. In J. C. Richards & R. W. Schmidt (Eds.),
Language and communication (pp. 191–226). Routledge.
Sinclair, J. (1991). Corpus, concordance, collocation. Oxford University Press.
Wray, A., & Perkins, M. R. (2000). The functions of formulaic language: An
integrated model. Language & Communication, 20 (1), 1–28.
Part II
Establishing a Theoretical Basis
After examining historical language teaching methods, Part I made

an analysis of the main methods of the Communicative Period. The
Notional-Functional Syllabus emphasised the importance of functions
but deemphasised the teaching of lexis necessary to build receptive
ability. The Natural Approach strongly emphasised input but received
strong criticism for its contention that formal learning couldn’t become
acquisition. As its name suggests, Communicative Language Teaching
provides plenty of opportunities for communication, but its lack of a
coherent theory of learning has led to an overemphasis on productive
use and an underemphasis on receptive learning opportunities. Content-
Based Instruction demonstrates the efficacy of learning language through
teaching subject matter and while its learners don’t achieve the same
level of ability as L1 users, there is no reason why they should. As a
variant of CLT, Task-Based Instruction similarly overemphasizes produc-
tive use at the expense of receptive learning opportunities. However, it
has reinstated a focus on form. The Lexical Approach set about defining
the nature and scope of vocabulary learning but had no systematic way
of tackling the scale of the task. This analysis of historical and modern
teaching methods has afforded an insight into what language educators
consider important. Key elements include:
104 Part II: Establishing a Theoretical Basis
1. providing opportunities to process substantial amounts of authentic

input
2. linking language to its situations of use
3. linking language to real world phenomena
4. drawing attention to language patterns of all levels of specificity
including words, phrases, clauses, sentences and suprasentential
elements
5. providing opportunities for productive use
There is no one approach, including the current orthodoxy, which

incorporates all of these elements, and there is no one approach that
delivers superior results for L2 learners. In the search for an alternative
pedagogy, the question is where to begin. Historically, innovative educa-
tors have focused on L1 acquisition to guide the design and development
of their methodology. However, as almost any language educator will tell
you, you can’t teach a second language like a first one. That is certainly
true. You can’t teach a second language like a first one, but you can learn
one.
8
First Language Acquisition
It is uncontroversial to say that first language acquisition is effective. Gass

(1997: 53) states that all cognitively normal learn how to speak. A child’s
vocabulary develops rapidly. Nation (2001: 96) reports that a five-year-
old native speaker has a vocabulary of approximately five thousand word
families. Shipley and McAfee (2019) state that one-year-olds know about
fifty words; three-year-olds know about a thousand words; and five-year-
olds know more than 10,000 words.
However, for the vast majority of learners, second language acqui-
sition is much less effective. Ellidokuzoglu (2017: 27) notes that
children effortlessly acquire a new language, but adults spend years
or decades studying without becoming fluent. Similarly, Ble-Vroman
(1989) remarks that apart from unusual cases, first language learning is
always successful, whereas second language learning is frequently unsuc-
cessful (cited in Skehan, 1998: 8). Allwright and Hanks (2009: 92) hint
that learning environment may be a factor in the remarkable success of
first language acquisition, and the failure of language learning in schools.
On the same theme, Ellidokuzoglu (2017: 27) notes that this failure is
most commonly observed in the classrooms of developed nations. Thus,
there is general agreement that it is more difficult to learn a second

Switzerland AG 2024
106 I. Pemberton
language than a first language, and some academics speculate that the
formal educational setting of most second language learning may have a
negative impact on its effectiveness.
Throughout the history of language education, the intuitions of many
language educators have been that the processes of first language acqui-
sition hold the key to successful second language acquisition. Howatt
(1984: 295) comments that ‘the success of informal learning, and partic-
ularly of the child acquiring its mother tongue, has always impressed
language teachers, and attempts to reproduce the same effect by creating
the same causes have been a regular feature of language teaching history.’
This was particularly evident during the Reform Period. Richards and
Rodgers (2001: 7) note that early exponent of input-based learning,
Claude Marcel, considered that L1 learning by children was a good
model for L2 teaching, and that Francois Gouin attempted to build his
methodology based on observations of children learning language (ibid.:
11). In addition, Howatt (1984: 157) reports that Thomas Prendergast
also started from the example provided by child language acquisition.
All of this suggests that an examination of the processes of first language
acquisition may prove to be a fruitful place to start in the development
of an approach to second language learning. An analysis of first language
acquisition can be performed in two ways: by considering the evolution
of language over human history and by considering how language evolves
over a person’s lifetime.
8.1 Evolution
While there is much debate and disagreement about the origins of
language and in particular, when this actually took place (estimates
range from somewhere between 3.5 million and 50,000 years ago),
there is some agreement on how language came into being. Feeney
(2018: 78) points out that traditionally the mainstream has assumed
a Darwinian account based on natural selection. MacWhinney (2015:
20) notes that Darwin demonstrated competition between organisms
for survival and reproduction leads to evolution. According to evolu-
tionary education psychologist, David C. Geary (2002: 319), natural
selection involves factors which influence the prospects of survival such

as disease, while sexual selection involves social factors which influence
the prospects of reproduction. Geary (2002: 322) argues that, relative to
other species, humans were uniquely successful in gaining control over
essential resources such as food and land and this reduced the impact of
natural selection which led to primary evolutionary pressure from social
factors influencing reproductive prospects.
As a consequence, the traits that started to have a stronger influence on
human survival and reproductive outcomes were those that allowed indi-
viduals to compete successfully against other members of their groups.
These included ‘sophisticated social competencies, such as language and
theory of mind, an accompanying increase in brain size, and other adap-
tations that facilitated the formation and maintenance of kin-based social
coalitions that compete with other coalitions for resource control’ (Geary,
2002: 323).
Broadly speaking, there are two schools of thought about the emer-
gence of language. According to Feeney (2018: 78), some theorists
believe that the emergence of language was to maintain social relations
in large groups, whereas others think that the evolution of language was
for instrumental and social purposes such as communicating important
information for survival such as how to fish or how to avoid danger.
8.2 Primary Biological Skills

The sophisticated social competencies which have emerged from natural
and sexual selection are known as biologically primary skills. According
to Australian educational psychologist John Sweller (2015: 190) humans
have evolved over innumerable generations to learn to listen and speak,
recognize faces, and learn general strategies for problem-solving. While
we acquire such primary knowledge easily and unconsciously, secondary
knowledge, which we need to function in society (e.g., reading and
writing) requires effort and explicit learning. Secondary knowledge has
to be taught in schools by teachers. Nevertheless, not all pupils succeed
in acquiring it.
108 I. Pemberton
In summary, natural selection and sexual selection over millennia have

resulted in the emergence of primary biological skills, including listening
and speaking, in order to gain access to and control resources for survival
purposes. While the primary biological skills of watching and listening
are used to learn our mother tongue, in most cases, learning a second
language in school relies on the much less effective biologically secondary
skills of reading and writing.
What should be clear from the foregoing discussion is that out of the
four language skills, listening is the most important for language acquisi-
tion. Many language educators have intuitively recognized this over the
centuries. Richards and Rodgers (2001: 55) quote Brooks (1964) who
reports that ‘many languages do not have a written form and we learn
to speak before we learn to read or write.’ Thus, on a fundamental level,
language is spoken not written. Author of The Lexical Approach, Michael
Lewis (1993: 8), notes that ‘initial L1 learning is exclusively based on
listening. All language which we produce is ultimately based on language
which we have previously met.’
Despite its importance to language learning, listening is routinely
under-emphasized in the communicative classroom. Nunan (1997)
comments that ‘listening is the Cinderella skill in second language
learning. All too often, it has been overlooked by its elder sister:
speaking.’ When listening is taught, it has become increasingly common
to teach listening strategies. However, Renandya (2012: 1) reports that
there is a lack of empirical evidence for teaching listening strategies. Simi-
larly, Sweller (2020: 4) points out that there is no reliable research which
demonstrates superior performance as a consequence of instruction in
generic skills such as listening. Renandya (2012: 1) notes that learners
may have picked up listening strategies while learning their first language.
In addition to considering language development over historical time,
an examination of language development during childhood can also
be made. Two of the most well-known theories of child language
development are Constructivism and Sociocultural Theory.
8.3 Constructivism
The constructivist theory of learning is most often associated with Swiss
psychologist Jean Piaget. In his theory of cognitive development, Piaget
set out that children’s thought processes change as they get older and that
their language develops in parallel to these changes in their thinking.
According to Piaget, as children experience new things, they develop
schema. A schema is a basic building block of memory which encodes
physical experiences, for example, with objects and actions, as well as
more abstract cognitive ones, e.g., theories.
A child’s schema can be altered in one of two ways: assimilation or
accommodation. Assimilation is the process of integrating new knowl-
edge with existing cognitive structures. In the case of assimilation, the
new information does not alter existing beliefs and understandings of
the world. Pound (2014: 48) explains that the new knowledge is stored
in parallel to existing knowledge.
Not all new information can be assimilated into existing cogni-
tive structure. In some cases, the new knowledge conflicts with
current understanding leading to accommodation. Accommodation is
the process of incorporating new knowledge into existing schematic
understanding where the new information results in a revision of the
existing understanding.
In the course of his research, Piaget identified four distinct stages that
a child progresses through in their cognitive and linguistic development.
Piaget’s stages are:
8.3.1 Sensorimotor Stage: Birth to 24 Months
In this stage, babies and toddlers obtain knowledge and understanding

via their senses: touch, sight, smell, hearing, and taste (Pound, 2014: 47).
An infant exists entirely in the present without a mental representation of
the world in memory and no sense of object permanence. In other words,
what a child cannot see does not exist. As children begin to notice that
words are used symbolically to describe objects and feelings, language
gradually appears.
110 I. Pemberton
8.3.2 Preoperational Stage: 2 to 7 Years
In this second stage, as young children repeatedly come into contact

with things, they develop the ability to represent the world mentally
using images and language. They are able to think about things symbol-
ically and use words and objects to stand for other things. In his
research into child language development, Piaget noticed that children
produced mainly ego-centric speech. They appeared to be conversing
with each other but in, actual fact, their language did not resemble
normal conversational behaviour. Instead, their speech patterns were
generally commentaries on what they were thinking or doing and were
not intended to elicit reactions from other children. Houston (2019:
140) comments that until a child reaches their seventh year, this is the
primary purpose of their speech. The fact that children chatter endlessly
does not mean they are social beings because they are thinking aloud not
communicating with others (ibid.: 142).
8.3.3 Concrete Operational Stage: 7 to 11 Years
In the next stage of cognitive development, the child begins to demon-

strate logical (operational) thinking. It is called the concrete operational
stage because manipulation of real materials or pictures helps children
to think logically. Children begin to develop categories and to classify
things according to similarity or difference (Pound, 2014: 47). In terms
of language development, children become communicatively competent
enough to talk to their listener in a way that they will understand and
respond. Piaget calls this the move from ego-centric to socialized speech.
8.3.4 Formal Operational Stage: Ages 12 and Up
The final stage of child cognitive development is known as the formal

operational stage. The difference between concrete and formal operations
is that concrete operations are performed on things, but formal opera-
tions are performed on ideas. As children increasingly develop knowledge
of their environment, they also understand that language can be used to
represent knowledge symbolically which allows them to use ego-centric

and socialized speech to think about things and communicate them to
others, and this ultimately leads to the ability to think about things in an
abstract manner. According to Pound (2014: 47), at this stage, children
can comprehend abstract ideas, they do not need concrete examples to
follow argument, and they can solve hypothetical problems.
8.3.5 Constructivism and Second Language

Acquisition
The main contribution of Constructivism to second language education

is the notion that knowledge is not something which a teacher trans-
mits to a learner, but it is something that a learner constructs based on
what they already know. This insight into learning has precipitated a
movement away from teacher-centred education where learners passively
receive input from the teacher to a style of education borrowed from
general education known as Active Learning. Active Learning requires
that learners work together to co-construct their own knowledge. Of all
the subjects, language teaching seems to embody this concept the most
with its broad range of interactive communication-based activities.
However, Mayer (2014: 21) points out that Active Learning can be
either behavioural or cognitive. Only cognitively active learning will
lead to deeper understanding. This kind of learning will involve learners
selecting information from input, organizing it into a coherent message,
and integrating it with their current knowledge (ibid.: 50). However,
cognitive activity doesn’t require partnership with another learner. It’s
something that can equally take place when a learner works indepen-
dently. Furthermore, pairwork and groupwork can negatively impact on
learning. If the input learners provide to each other lacks new informa-
tion, or if they are unable to understand it or process it for meaning and
form due to working memory constraints, then less rather than more
learning may take place.
112 I. Pemberton
8.4 Sociocultural Theory

Sociocultural Theory is associated with Russian psychologist, Lev
Vygotsky. Some of Vygotsky’s ideas were similar to Piaget’s. According
to Pound (2014:54), Vygotsky agreed that learners construct knowledge
and understanding from experience. In other words, like Piaget, he held
a constructivist view of learning. However, a major difference between
the two was that whereas Piaget regarded early learning as internal
rather than interactive, Vygotsky thought that interaction played a major
role in mental development (Garhart-Mooney, 2013). Pound (2014:
54) explains that Piaget emphasized the development of knowledge
from experience, whereas Vygotsky argued that the family, community,
and other children played an important role in the development of
knowledge.
There are a number of key ideas in Vygotsky’s theory of child devel-
opment. First of all, Vygotsky made an important distinction between
what he called spontaneous concepts and scientific concepts. According
to Fosnot and Perry (1996), spontaneous concepts were pseudo-concepts
which originated as a child reflected on their daily experience. On the
other hand, scientific concepts emerged from classroom instruction.
Vygotsky speculated about how children could learn scientific
concepts, which fell outside of the realm of their perceptual experi-
ence, and this led to the development of one of his most well-known
ideas: the Zone of Proximal Development (ZPD). The ZPD describes
the difference between what a child can learn by themselves and what
they can learn with assistance. In order to bridge the gap between a
child’s everyday concepts and scientific concepts requires a more knowl-
edgeable other (MKO). This would normally be a parent or teacher,
but it could also be a sibling or friend. The more knowledgeable other
converses, questions, explains, and negotiates with the child to extend
their understanding. Fosnot and Perry (1996) assert that psychologists
such as Jerome Bruner developed this idea into what subsequently
became known as ‘scaffolding.’ Scaffolding is support which is given
temporarily to help a learner understand something and then taken away
when they have understood it.
While some researchers assume that social interaction in the ZPD

with a more knowledgeable other is a necessary condition to transmit
knowledge to a learner, others have called for a broader interpretation
(Lantolf, 2000: 17). In fact, human interaction is not a necessary condi-
tion for development. According to Lantolf (2007: 698), any one of a
range of artefacts could be used to mediate new knowledge including
computers, calculators, or even a piece of paper and a pencil. Similarly,
Flowerdew (2015: 5) explains that the scaffolding could be provided by
a variety of cognitive tools which include texts, realia, digital media, and
motivational tasks.
8.4.1 Vygotsky’s Theory of Language Development
A major difference in emphasis in Vygotsky’s theory of child develop-

ment was his view of language learning. According to Pound (2014:
54), Vygotsky regarded language as a tool for thinking and learning
alongside numbers, notations, signs, diagrams, and plans. Whereas as
physical tools act upon an external object, a psychological tool has
bidirectionality. While language may be outwardly directed for social
communication, it may also be directed inwardly ‘to regulate and control
mental processes such as memory, attention, rational thinking and
learning’ (Lantolf & Thorne, 2006: 26).
Another area where Vygotsky differed in his thinking from Piaget was
in the role of private speech. Vygotsky’s belief was that a new-born child
initially had separate systems of language and thought, and these didn’t
merge until the child was approximately three years of age. In contrast to
Piaget, he divided speech into three types: social speech, private speech,
and inner speech. Young children use social speech up to the age of three
in response to parental speech, to express their needs and communicate
feelings and emotions. Between three and seven years, children begin to
use private speech. Children use private speech to collaborate with them-
selves in the same way that an adult collaborates with them to achieve
an objective. When a child is around seven years old, private speech
diminishes in volume and becomes quiet inner speech as the child uses
it increasingly for self-regulation.
114 I. Pemberton
8.4.2 Sociocultural Theory and Communicative

Language Teaching
Advocates of CLT draw upon a broadly interactionist model of language

acquisition which co-opts the sociocultural model of learning. To
envisage the process, dialogue takes place between the learner and a more
knowledgeable other, and through the processes of hypothesis testing
and negotiation of meaning, the more knowledgeable other imparts new
knowledge to the learner which is either assimilated or results in accom-
modation within memory. Hence, learning takes place. Thus, Socio-
cultural Theory provides a rationale for pairwork and groupwork tasks
within the communicative classroom. However, it rests on the assump-
tion that social interaction plays a primary role in first language learning
and therefore must also be necessary for second language learning. But
how important is social interaction in first language acquisition?
It is not disputed that adults play a role in helping a child to acquire
a language. Bloom (2000: 84) notes that when adults attempt to teach
children new words, it might speed up the learning process. Similarly,
Tomasello (2003: 177) concedes that it is possible that adult feedback
to children might help to constrain children’s tendency to overgener-
alize. While it appears that parents and other adults play a role in child
language acquisition, it is not a central one.
In fact, caregiver speech is not even a necessary condition. Bloom
(2000: 83) reports there are some societies where parents make no
attempt to teach their children to speak but they learn to do so anyway.
He also reports that children do not need to be jointly attending to things
with adults to learn the language to describe them. In other words, the
attempts of adults to teach words do not take the place of a child’s ability
to read the intentions of others. Tomasello (2003: 177) agrees, noting
that caretaker speech does not play a necessary or sufficient role in child
language development. According to Yule (2020: 206), few children are
taught how to speak, and it is more likely that they actively construct
language from what they see and what they hear going on around them.
Thus, it appears that children learn much of their mother tongue without
explanation from a more knowledgeable other.
To take this argument a step further, it appears that even error correc-
tion is not necessary for language learning. Firstly, explicit correction
does not seem to have a major influence. Bloom (2000: 1) reports that
children do not receive consistent feedback on the accuracy of what they
say. Krashen (1985: 35) cites research which shows that parental approval
(e.g., a parent saying ‘That’s right’) does not improve accuracy. Secondly,
implicit correction does not appear to change childish forms. Yule (2020:
207) states that even subtle correction, such as when an adult repeats
what they have said with a correction, they continue to use their own
language. Tomasello (2003: 177) comments that few linguists think that
recasts constrain children’s tendency to overgeneralize.
A final point to note is that Vygotsky envisaged the Zone of Proximal
Development as a place where secondary, not primary learning would
take place, and in fact, as has been set out above, language learning
draws upon primary not secondary biological skills. In other words,
learning language and learning conceptual knowledge, particularly tricky
threshold concepts, are qualitatively different processes.
8.5 Social-Pragmatic Theory

Notwithstanding their ubiquity in educational theory, Constructivism
and Sociocultural Theory deal mostly with the learning of concepts, as
opposed to the learning of the language used to articulate them, and as a
consequence, they are vague on the processes that a child uses to actually
learn a language. A less well-known but detailed theory of how children
learn language which offers a cogent explanation of these processes is
Social-Pragmatic Theory (Tomasello, 2000, 2003).
The Social-Pragmatic Theory of first language acquisition is a usage-
based theory. The term ‘usage-based’ appears to suggest that language
learning occurs as a result of language use, which may sound immedi-
ately appealing to CLT advocates, but an important distinction needs to
be drawn between the learner’s use of language and the use of language
by others. In actual fact, ‘usage-based’ implies mainly the latter of the
two, rather than the former. In other words, language acquisition occurs
116 I. Pemberton
mainly as a result of attending to the language of others, not practising

it oneself.
Social-Pragmatic Theory sets out that children use general cognitive
skills to learn language. These are intention-reading skills (or theory of
mind), cultural imitation, and pattern finding skills.
8.5.1 Intention-Reading Skills
To begin with, Bloom (2000: 61) states that children use naive
psychology (theory of mind) to determine what words refer to.
Tribushinina and Gillis (2017: 32) also point out that theory of mind
and language develop in parallel, as theory of mind enables children
to understand things from the perspectives of others and grasp their
communicative intent.
American developmental and comparative psychologist, Michael
Tomasello, outlines the set of key skills which children possess. These
include the abilities to:
• share attention with others,

• follow the attention and gestures of others,
• direct the attention of others using gestures
Tomasello (2003: 3).

Yale University Psychology professor Paul Bloom (2000: 62) explains
that nine-month-old babies follow their mother’s gaze and pointing, and
that one-year-olds will point at things and then check where an adult
is looking. In sum, these skills demonstrate a child’s theory of mind. A
child recognizes that an adult is an intentional being and that their gazing
and pointing are non-random acts which are meant to direct the child’s
attention to objects of mutual interest. Tomasello (2003: 4) argues that
intention-reading skills are domain general. Children use them for not
only to learn language but also to learn other cultural skills such as how
to use tools, pretend play, and rituals.
8.5.2 Cultural Learning
In order to internalize language, children use cultural learning. Tomasello

(2003: 3) defines cultural learning as the ability to imitatively learn the
intentional actions of others. This includes not only their physical but
also their communicative acts. It is important to note that imitation
does not simply involve behaviouristic reproduction of an adult’s action
but also incorporates their intention. Tribushinina and Gillis (2017: 16)
comment that rather than parroting parental input, children interpret
the communicative intent of their parents and repeat what they say with
the same function. Imitation is anything but simplistic. Bybee (2010:
16) makes the point that it involves a high skill level together with the
ability to segment, categorize, and recombine language.
8.5.3 Pattern Finding
Another set of skills which children use to learn language is pattern

finding skills. In his book Constructing a Language, Tomasello (2003:
4) outlines the scope and nature of pattern finding skills. In particular,
these skills involve the ability to:
• categorize similar objects and events,

• create schemas to represent similar experiences,
• segment experiences into functional elements,
• make analogies between less similar experiences.
These skills are also domain general. Dolgova and Tyler (2019: 942)
state that language learning involves general thinking skills like cate-
gorization, pattern finding, and noticing frequency. Tomasello (2003:
4) comments that children use general cognitive abilities to parse
language and store the whole utterance together with its parts. Langacker
(2000) states that children represent the form and meaning of words
in cognition using general learning skills and these include association,
categorization, and abstraction (cited in Verspoor & Nguyen, 2015:
312).
118 I. Pemberton
Of these general learning processes, categorization is the first to

emerge. This occurs between birth and three years of age (Gopnik et al.,
2001: 81). To answer the question of how young children categorize
things, Bloom (2000: 149) points out that to begin with, babies use
perceptual properties to categorizing new things such as appearance,
sound, etc. At this stage, shape is especially important (ibid.: 155).
Another key language learning process is schematization. Tomasello
(2003: 297) reports that ‘young children hear and use – on a mind-
numbingly regular basis – the same utterance repeated over and over
but with systematic variation.’ Schematization is the process of gener-
alizing from concrete examples of language to create more abstract
syntactic frames. Tribushinina and Gillis (2017: 18) explain that chil-
dren generalize concrete instances of language to create abstract syntactic
representations. They point out that a child has to build up a ‘crit-
ical mass of exemplars’ in order to be able to generalize and form
abstract rules. Eskildsen (2009: 336) describes schematization as a three-
stage process of prototype construction which progresses from formulas
(language patterns involving concrete items) to low-scope patterns (slot
and filler) to fully abstract constructions. In summary, children induc-
tively identify patterns from multiple concrete examples of language and
are able to use these to form prototypical constructions.
A further process is what Tomasello (2003: 145) calls a ‘functionally-
based distributional analysis.’ This draws upon both the processes of
categorization and schematization and allows a child to craft more
abstract language patterns. By way of explanation, a child identifies a
low-scope pattern (a slot and filler construction with some words fixed
and some words substitutable) and then attends to the kind of words that
can occupy the slots. This allows the formation of more abstract syntactic
categories such as noun and verb.
A final general learning skill mentioned frequently in the literature is
analogy. Analogy allows the learning of unfamiliar language. It is similar
to schematization but whereas schematization involves generalizing from
similar items of language to create linguistic prototypes, analogy involves
the recognition of similarities between patterns of language which super-
ficially appear different but contain similar functions, and this allows the
formation of more abstract representations.
8.6 Summary
This chapter has examined accounts of first language acquisition as a
basis for second language teaching and learning pedagogy. It looked at
language acquisition from both an evolutionary perspective and from
explanations of first language acquisition during childhood.
From an evolutionary perspective, language has arisen through the
processes of natural and sexual selection. Early humans who could use
language to understand and communicate gained greater access and
control of essential resources, improving their survival rates and those
of their kin. On the basis of the emergence of language through evolu-
tionary forces, a distinction has been proposed between primary and
secondary biological skills. Primary biological skills, specifically, the skills
of observing and listening, allow all normal human beings in all normal
environments to acquire their first language. That is to say, we are biolog-
ically adapted to learn language. Secondary biological skills are those
we use in the classroom to learn culturally important knowledge for
participation in society, for example, reading and writing. We are not
biologically adapted for this, and it is much more difficult. Second
languages taught through biologically secondary skills are much more
difficult to learn. This suggests that if we are able to employ our primary
biological learning abilities, we may be able to learn languages more
successfully.
Two well-known language learning theories are Constructivism and
Sociocultural Theory. The first of these theories, Constructivism, is
associated with Jean Piaget. This is a cognitive theory of knowledge
acquisition that states that recurrent perceptual experience results in the
creation of knowledge frameworks known as schema within memory.
A schema is a general representation of real-world phenomena that a
child experiences. Schema are used in processing to understand, and in
production to articulate the phenomena that they represent.
In terms of child language development, Piaget drew a distinction
between ego-centric speech and socialized speech. Whereas up until
around seven years of age, children may appear to be communicating,
they are essentially thinking out loud and providing a commentary on
what they are doing. They have no particular expectation that listeners
120 I. Pemberton
will understand or respond to what they are saying. Piaget’s view was that
when a child matures sufficiently, they become communicatively compe-
tent enough to talk to their listener in a way that they will understand
and respond. Piaget calls this the move from ego-centric to socialized
speech.
The second theory, Sociocultural Theory, is associated with Lev
Vygotsky. Vygotsky believed that learning took place as a result of social
interactions between a learner and a more knowledgeable other in what
he termed the Zone of Proximal Development (ZPD). Like Piaget,
Vygotsky recognized the phenomenon of private speech (ego-centric
speech) and also regarded it as thinking out loud rather than an attempt
to communicate but he felt that its disappearance at around seven years
of age was not due to the emergence of social interactional skills but
represented the movement of externally verbalized thought to internally
verbalized thinking or ‘inner speech.’
While advocates of CLT hold that language is learnt through inter-
action and draw upon Sociocultural Theory as a justification for pair
and groupwork activities, research shows that adult speech may have an
impact, but it is a minor one, and caretaker speech is not a necessary
condition for language learning anyway. Moreover, research into error
correction demonstrates that even when correction is subtle, such as
when parents use recasts, it does not change a child’s erroneous language
forms.
In contrast to classical theories of language acquisition, Social-
Pragmatic Theory provides a detailed account of first language acqui-
sition. Social-Pragmatic Theory claims that language is learnt via the
application of general learning skills to concrete examples of language.
Children employ three types of general learning abilities. The first set
are intention-reading skills which are also known as theory of mind.
These include the ability to share attention, to follow attention, and
to direct attention. Intention-reading skills allow children to interpret
the communicative intent of adults and thus understand what they
are saying. Secondly, children employ cultural imitation to internalize
language. This involves copying both the language and the intent of a
speaker. A third set of skills are general pattern finding skills including
categorization, analogy, functionally based distribution analysis, and

schematization.
8.7 Implications
We learn our mother tongue using the primary biological skills of obser-
vation and listening, but we use secondary skills to learn conceptual
knowledge for use in society. When these are applied to second language
learning, they are much less effective. This implies that if primary biolog-
ical skills can be applied to second language learning, outcomes might be
better.
The broad interpretation of Constructivism for many educators is as a
call for Active Learning. Active Learning implies learning which is inde-
pendent of the teacher, and this usually means putting learners into pairs
or groups to discuss things together. However, neither independence
from the teacher nor pair or group work are sufficient conditions for
learning. This pedagogical strategy can lead to behavioural activity, for
example, learners chatting to each other, whereas, cognitive activity is
necessary for learning, in other words, comprehending, selecting, orga-
nizing, and integrating information with existing knowledge, and this
can be done with or without a partner.
Sociocultural Theory, with its concept of a more knowledgeable other,
is also used to justify the use of pair and group work but interaction
in the Zone of Proximal Development can’t account for the bulk of
language learning. Thus, implementing activities that involve extensive
amounts of negotiation of meaning may not be a productive use of time
and energy.
The less well-known Social-Pragmatic Theory of first language acqui-
sition offers a compelling account of how children learn their first
language. That is, children learn through intention-reading, imitation,
and general pattern finding skills. This implies that learners can learn
without explicit instruction or correction from a more knowledgeable
other.
122 I. Pemberton
Taken together, these theories imply that instead of focusing on

spoken output in pairs and groups in the second language learning class-
room, it might be prudent to employ learners’ primary biological skills
of observing, listening, and imitation to acquire language.
References
Allwright, D. & Hanks, J. (2009) The developing language learner: An introduc-
tion to exploratory practice. Palgrave Macmillan.
Bloom, P. (2000). How children learn the meanings of words. MIT Press.
Bybee, J. (2010). A usage-based perspective on language. In J. Bybee (Ed.),
Language, usage and cognition (pp. 1–13). Cambridge University Press.
Dolgova, N., & Tyler, A. (2019). Applications of usage-based approaches to
language teaching. In X. Gao (Ed.), Second handbook of english language
teaching (pp. 939–961). Springer International Publishing.
Ellidokuzoglu, H. (2017). Towards a receptive paradigm in foreign language
teaching. Turkish Online Journal of English Language Teaching (TOJELT),
2(1), 20–39.
Eskildsen, S. W. (2009). Constructing another language-usage-based linguistics
in second language acquisition. Applied Linguistics, 30 (3), 335–357.
Feeney, A. (2018). 5. Language evolution and the nature of the human faculty
for language. In C. Wright, T. Piske, & M. Young-Scholten (Eds.), Mind
matters in SLA (pp. 72–90). Multilingual Matters.
Flowerdew, J. (2015). John Swales’s approach to pedagogy in Genre Analysis: A
perspective from 25 years on. Journal of English for Academic Purposes, (19),
102–112.
Fosnot, C. T. & Perry, R. S. (1996). Constructivism: A psychological theory
of learning. In Constructivism: Theory, perspectives, and practice (2nd ed.).
Teachers College Columbia University.
Garhart-Mooney, C. (2013). Theories of childhood: An introduction to Dewey,
Montessori, Erikson, Piaget, and Vygotsky (2nd ed.). Redleaf Press.
Gass, S. M. (1997). Input, interaction, and the second language learner. Lawrence
Erlbaum Associates.
Geary, D. C. (2002). Principles of evolutionary educational psychology.
Learning and Individual Differences, 12(4), 317–345.
Gopnik, A., Meltzoff, A. N., & Kuhl, P. K. (2001) How babies think: The
science of childhood . Phoenix.
Houston, S. H. (2019). A survey of psycholinguistics. De Gruyter Mouton.
University Press.
Lantolf, J. P. (2007). Sociocultural theory. In J. Cummins & C. Davison (Eds.),
International handbook of English language teaching (pp. 693–700). Springer
US.
Lantolf, J. P. (2000). 1. Introducing Sociocultural theory. In J. P. Lantolf (Ed.),
Sociocultural theory and second language learning. Oxford University Press.
Lantolf, J. P. & Thorne, S. L. (2006). Sociocultural theory and the genesis of
second language development. Oxford University Press.
Mayer, R. E. (2014). Introduction to multimedia learning. In R. E. Mayer
(Ed.), The Cambridge handbook of multimedia learning (2nd ed., pp. 1–24).
MacWhinney, B. (2015). Multidimensional SLA. In T. Cadierno & S. W.
Eskildsen (Eds.), Usage-based perspectives on second language learning. De
Gruyter, Inc.
University Press.
Nunan, D. (1997) Listening in language learning. The Language Teacher.
Available at: http://www.jalt-publications.org/tlt/files/97/sep/nunan.html.
Accessed May 11, 2021.
Pound, L. (2014). How children learn: Educational theories and approaches.
Practical Pre-School Books.
Renandya, W. A. (2012). Five reasons why listening strategy instruction might
not work with lower proficiency learners. ELTWorldOnline.com, Vol. 4.
Shipley, K. G. & McAfee, J. G. (2019) Assessment in speech-language pathology:
a resource manual (Sixth ed.). Plural Publishing, Inc.
Press.
Sweller, J. (2015). In Academe, what is learned, and how is it learned? Current
Directions in Psychological Science: A Journal of the American Psychological
Society, 24 (3), 190–194.
124 I. Pemberton
Sweller, J. (2020) Cognitive load theory and educational technology. Educa-

tional technology research and development, 68 (1): 1–16.
Sweller, J., van Merrienboer, J. J. G., & Paas, F. (2019). Cognitive architecture
and instructional design: 20 years later. Educational Psychology Review, 31(2),
261–292.
Tomasello, M. (2000). The social-pragmatic theory of word learning. Prag-
matics, 10 (4), 401–413.
Tomasello, M. (2003). Constructing a language: A usage-based theory of language
acquisition. Harvard University Press.
Tribushinina, E. & Gillis, S. (2017). Advances and lacunas in usage-based
studies of first language acquisition. In J. Evers-Vermeul & E. Tribushinina
(Eds.), Usage-based approaches to language acquisition and language teaching.
De Gruyter, Inc.
Verspoor, M. & Nguyen, T. P. H. (2015). A dynamic usage-based approach to
second language teaching. In T. Cadierno & S. W. Eskildsen (Eds.), Usage-
based perspectives on second language learning. De Gruyter, Inc.
Yule, G. (2020). The study of language (7th ed.). Cambridge University Press.
9
Foundations
Half a century ago, British linguist Pit Corder (1973: 109) observed that
‘until we have a much better idea of what (the learning processes) are
we cannot, on a systematic and principled basis, create the necessary
conditions for optimal learning; we can only do what we have largely
been doing, that is, working on a hit and miss basis.’ Corder pointed
out that sceptics doubted the study of first language acquisition had any
relevance for second language education as there were too many differ-
ences between the conditions under which first and second languages are
learnt.
In the intervening five decades, second language acquisition has
become a research field in its own right. Even so, its impact on class-
room teaching is somewhat limited. According to Lantolf (2012: 63),
SLA researchers express concerns about the applicability of their research
and most recognize the importance of making their theory and findings
more accessible and comprehensible for classroom teachers.
A main aim of this book is to bridge the research-practice gap by trans-
lating complex language acquisition theory into classroom pedagogy. The
previous chapter examined the Social-Pragmatic Theory of first language
acquisition. This theory posits that children learn their first language

Switzerland AG 2024
126 I. Pemberton
as a result of exposure to a very large number of concrete examples

of language. They understand what they hear using the skill of inten-
tion reading, and they employ general learning skills to create linguistic
prototypes from recurrent patterns which they can then use to generate
original language.
The Social-Pragmatic Theory of first language acquisition is a usage-
based theory. The proponents of usage-based theories argue that they
apply to all language learning, not just first language learning, but
also additional language learning. Ellis et al. (2016) cited in Dolgova
and Tyler (2019: 944) maintain that L1 and L2 learners use the same
cognitive skills to learn a language. This suggests that, in line with the
intuitions of the language education innovators of the Reform Period,
the key to second language acquisition lies in an understanding of
first language acquisition processes and their application to additional
languages. With this in mind, this chapter looks at the foundational
theories of usage-based linguistics. To begin with, the chapter considers
Dynamic Systems Theory (also known as Chaos/Complexity Theory)
which usage-based theories draw upon for their explanation of language
acquisition.
9.1 Dynamic Systems Theory

Dynamic Systems Theory began life as a mathematical theory. De Bot
et al. (2007: 8) explain that initially, the theory was only applied to
simple two-variable systems, for example, the swinging of a double
rod pendulum. As a result of its application to more complex systems
with multiple variables, it became a complex systems theory. Since the
1960s, Dynamic Systems Theory has been applied to a wide variety of
fields such as Physics, Biology, Meteorology and Economics, providing
powerful accounts of the development of phenomena within these
systems. The theory postulates that the outcomes of a complex system
are not predictable. This is because incremental changes in the system
over time are always based on its previous state, thus, only short-term
predictions, such as tomorrow’s weather but not next month’s weather,
can be made with any accuracy (Lowie, 2013: 1).
9 Foundations 127
According to Lowie (2013: 4), second language learning has been

traditionally thought of as a linear process with an initial state where a
learner knows nothing about the target language, and a final state where
the learner achieves native-like proficiency in the language. In 1997,
Diane Larsen-Freeman published a somewhat ambitious but what was to
become a seminal article entitled ‘Chaos/Complexity Science and Second
Language Acquisition.’ In it, she argued that languages and language
learning are in fact complex dynamic systems with multiple variables and
unpredictable outcomes. Response to this article was initially muted but
has begun to gain momentum as an increasing number of linguists have
recognized the relevance of the theory for language acquisition. De Bot
et al. (2007: 14) comment that a majority of applied linguists will likely
concur that second language acquisition is complex and that there are
many factors, for example, motivation, mother tongue, amount of input
and aptitude, which affect the learning process. Nevertheless, it is still
common to find linear explanations of language phenomena, such as the
Information Processing Model, propagated in the SLA literature.
In applying DST to second language acquisition, different writers
from different disciplinary backgrounds use different terminology, which
can be a cause of confusion. Dynamic Systems Theory might also
be referred to as complexity theory, chaos theory or chaos/complexity
theory, and the systems within Dynamic Systems Theory may be vari-
ously referred to as complex systems, complex adaptive systems, complex
dynamic systems, or complex adaptive dynamic systems.
Larsen-Freeman (2012: 74) sets out that Dynamic Systems Theory
features systems which are complex, dynamic, open, self-organizing,
adaptive, non-linear, and emergent.
9.1.1 Complexity
A defining characteristic of a complex system is that it has a large number

of components. Baicchi (2015: 9) comments that these components of
the system are autonomous but at the same time they are interdepen-
dent. In order to illustrate how complex a system can be, Larsen-Freeman
(1997: 143) gives the example of the human brain. While researchers
128 I. Pemberton
are in disagreement about the number of components a human brain

contains, it is an astronomically large number. A figure often quoted in
the literature is 100 billion nerve cells (neurons) which is a similar order
of magnitude to the number of stars found in the Milky Way (Gopnik
et al. 2001: 183). More recently, using revised methodology to perform
calculations, Herculano-Houzel (2009: 5) arrived at a figure of 86 billion
neurons. A neuron receives electrical impulses from dendrites and each
neuron can be connected to up to 10,000 dendrites. This provides some
idea of the potential scale of a complex system.
9.1.1.1 Subsystems
Complex dynamic systems are described as ‘nested’ systems. This means

that a system contains subsystems of all levels of granularity. De Bot et al.
(2007: 8) explain that all systems are a component of another system,
ranging from tiny particles of a sub-molecular level up to the vast scale of
the universe. Nevertheless, the same dynamic principles operate at every
level.
In language theory, the concept of modular subsystems has existed
for some time. De Bot et al. (2013: 207) note that the dominant view
of language has been that a language is modular and consists of sepa-
rate subsystems such as phonology, syntax, and semantics. According to
Larsen-Freeman (1997: 149), language subsystems include ‘phonology,
morphology, lexicon, syntax, semantics, pragmatics.’ A similar list is
provided by Lowie (2013: 2) who states that subsystems exist at all levels
of language processing and production including conceptualization,
syntax, lexicon, semantics, phonetics, and phonology. However, Larsen-
Freeman and Lowie draw a distinction with the traditional modular
view of language, arguing that these subsystems are not independent but
interdependent. That is to say, they are dynamic, open, and adaptive.
9.1.1.2 The Lexicon
One subsystem that deserves special attention is the lexicon. To make

the point, Milton and Donzelli (2013: 441) quote Wilkins (1972) who
9 Foundations 129
remarked that ‘without grammar very little can be conveyed, without

vocabulary nothing can be conveyed.’ Experts are in agreement that
learning the vocabulary of a language is a herculean task. According to
Lewis (1993: 9) ‘the first thing students need to do is to learn to under-
stand quite a lot of words.’ Similarly, Willis (1990: 39) observes that
‘the most difficult thing about learning a language is that there is simply
so much to learn. An educated speaker of English is likely to have a
vocabulary of some 50,000 words.’
However, experts are not in agreement about just how many words
a fluent speaker knows. Bloom (2000: 25) draws upon the work of
a number of scholars in estimating that a typical British or American
high school graduate knows around 60,000 words. Usage-based theo-
rist Joan Bybee (2010: 15) claims that expert language speakers know
tens or possibly hundreds of thousands of words. Author of Learning
Vocabulary in Another Language, Paul Nation (2001: 9), suggests that
an educated native English speaker knows approximately twenty thou-
sand word families. Brysbaert et al. (2016) calculate that an average
twenty-year-old American English speaker knows 42,000 dictionary
forms (lemmas), 11,100 word families, and 4,200 idiomatic multi-word
expressions. While these figures are inconsistent, we can see that linguists
consider that a native speaker knows, at the very least, tens of thou-
sands of words and that estimates are increasing as methods of calculating
vocabulary size become more sophisticated over time. This underscores
the importance of this all too often-overlooked element of L1 speaker
competence. Milton and Donzelli (2013: 441) citing Long and Richards
(2007) observe that vocabulary may be ‘the core component of all the
language skills.’
9.1.1.3 The Learning Process
It is not only the composition of a language which is complex, but also

second language acquisition is a complex process. Larsen-Freeman (1997:
151) sets out a large number of variables which can have an effect on the
process of acquisition. These include:
130 I. Pemberton
• the source language

• the target language
• the markedness of the L1
• the markedness of the L2
• the amount and type of input
• the amount and type of interaction
• the amount and type of feedback
• whether a language is acquired in untutored or tutored contexts.
Further factors which affect the success of the language learning

process include:
• age
• aptitude
• motivation
• attitude
• personality factors
• cognitive style
• hemisphericity
• learning strategies
• gender
• interests
9.1.2 Dynamism
A dynamic system is one which changes over time. The state the system is
in the moment before a change occurs determines how it develops. The
outcomes of interactions between its components are unpredictable as
each interaction alters the environment in which further interactions take
place. Baicchi (2015: 14) points out that this means a complex system
cannot be understood in simple mechanical ways. While parts of the
9 Foundations 131
system can be understood in isolation, the whole system can only be

understood in terms of principles or patterns.
In terms of language, either a language can change, or a learner’s
knowledge of the language can change. While they are often regarded
as fixed entities, languages are in a constant state of flux as speakers
use them. Beckner et al. (2009: 15) citing Hopper (1987) explain
that the only constant in individual idiolects and the language spoken
by a community is the ubiquity of change. Larsen-Freeman (2012)
agrees, noting that according to complexity theory, language evolution
is a dynamic process in which change is continuous. De Bot et al.
(2013: 210) report that grammaticalization studies show well-established
language patterns change because of ‘use, overuse and abuse.’
In addition to language change, speakers’ language knowledge also
changes over time. Diessel (2017: 2) states that the continuous pres-
sure of performance leads to the ongoing restructuring of a speaker’s
language knowledge. Beckner et al. (2009: 15) cite Bybee (2006) who
argues that at the individual level, every instance of language use changes
an idiolect’s internal organization. Verspoor and Nguyen (2015: 307)
expand on this point, explaining that change takes place as a result of the
continuous interaction of its various subsystems—sounds, morphemes,
lexis, syntax, semantics and pragmatics. They (ibid.: 308) remark that
a change in one subsystem causes a change in another, which causes a
change in another subsystem, leading to a process of continuous change.
Larsen-Freeman (2011: 58) adds that change is not always positive—a
learner’s interlanguage can both expand and contract. She also observes
that formal grammatical approaches such as generative grammar have
trouble describing the dynamic nature of language, and in particular, its
evolution over time (Larsen-Freeman (1997: 151).
9.1.3 Openness
Openness is a fundamental characteristic of living systems. An open

system receives energy from its environment. According to Baicchi
(2015: 16) this is essential so that the system can maintain its structure
and continue to function. In relation to this, Beckner et al. (2009: 16)
132 I. Pemberton
explain that while a closed system arrives at a steady state and does not
change, energy flows into an open system which allows it to change. A
similar point is made by Larsen-Freeman (2012) who observes that an
open system will interact with systems outside of itself, which allows it
to receive feedback as a basis for change.
All living languages are also open to and interact with their envi-
ronments. According to Verspoor and Nguyen (2015: 307), a language
is an open system which interacts with ‘cognitive, historical, pedagog-
ical, economic, social, and a number of other systems.’ According to De
Bot et al. (2007: 14), a language learner also represents an open system
which interacts with a larger social system. A learner also has their own
ecosystem which consists of cognition, intelligence, motivation, aptitude,
intentionality, mother tongue, knowledge of the target language, and so
on. This ecosystem is affected by maturity, education, the amount of
exposure to the target language, etcetera.
9.1.4 Self-Organization
Self-organization is a fundamental process of complex systems. Baicchi

(2015: 16) defines it as the spontaneous emergence within a system of
new forms of organization. Self-organization takes place continuously
according to systemic principles. To illustrate this, Lowie (2013: 3), gives
the example of ripples in a sand dune. Even when someone walks over a
sand dune and leaves footprints, the sand dune will repair itself and the
ripples will re-emerge after some time. An important point is that this
self-organization occurs without any external central control. The sub-
components of a system interact with each other at a local level but are
not aware of the wider system. However, multiple interactions can result
in changes to the whole system (Baicchi, 2015: 19).
Scholars from all theoretical camps assume that learners’ interlan-
guages self-organize. Advocates of CLT believe that some kind of cogni-
tive reorganization takes place as a result of productive use of the target
language. On the other hand, usage-based linguists believe that self-
organization takes place as learners work iteratively with input. Verspoor
and Nguyen (2015: 310) provide an example of how this might work.
9 Foundations 133
They note that it is well-established that a significant increase in recep-

tive vocabulary is necessary before a change can take place in productive
vocabulary. As a learner processes new vocabulary their interlanguage
goes through multiple iterative changes. This eventually leads to a phase
shift which results in a tangible improvement in productive vocabulary.
9.1.5 Adaptation
The openness of a complex system allows it to adapt to its environment.

According to Larsen-Freeman (2012), when a system interacts with its
environment, it receives feedback and then adapts itself to the environ-
ment. Berreby (1994) cited in Larsen-Freeman (1997: 145) explains that
systems will react to changes in their environment to turn an event to
their advantage rather than simply providing a passive response. In this
sense, a system can learn.
Adaptation in a language system can take place on a number of
different levels. First of all, individuals can influence each other. As one
learner interacts with another, their respective interlanguages change in
a process known as co-adaptation (Larsen-Freeman, 2011: 54). As an
example, Beckner et al. (2009: 16) note that the participants in a conver-
sation co-influence their respective language resources. While a speaker
may aim to be brief, which leads to phonological reduction, listeners will
want ‘perceptual salience, explicitness, and clarity, which require elabora-
tion.’ On the one hand, a speaker may accommodate listeners by slowing
down, over-enunciating and elaborating points. On the other hand,
listeners may attenuate to a speaker by first perceiving and then adopting
phonological features of their speech such as assimilation, reduction, and
omission. A second form of adaptation arises from multiple interactions
across a speech community. Beckner et al. (2009: 14) remark that there
is a bi-directional relationship between communal language and idiolect.
Interactions between speakers in a society with different idiolects both
give rise to communal language, and the communal language influences
the idiolects of different speakers.
134 I. Pemberton
9.1.6 Non-linearity
A system can be linear or non-linear. One distinguishing feature of

each type of system is the relationship between its causes and effects.
Baicchi (2015: 16) explains that a linear system is one where the causes
of the system’s effects are easily identifiable, and the effects are strictly
proportional to the causes. On the other hand, a non-linear system may
sometimes exhibit linearity, but at other times a small input might result
in a major change in the output from a system. By way of illustration,
meteorologist Edward Norton Lorenz proposed the ‘Butterfly Effect.’
This was the idea that the flutter of a butterfly’s wings in Brazil could set
off a tornado in Texas. A similar analogy is the idea of a rolling pebble
triggering a landslide. Larsen-Freeman (1997: 143) describes this as the
‘camel’s back effect.’ A simple trigger has the potential to give rise to a
chaotic state. Conversely, De Bot et al. (2007: 8) comment that ‘major
perturbations may be absorbed by the system without much change.’
Another distinguishing feature of linear and non-linear systems is
the rate at which the system changes. While the rate of change in a
linear system is gradual and incremental, Larsen-Freeman (2011: 52)
comments that the rate of change in a non-linear system may be all at
once. This sudden dramatic reorganization of the system is known as a
phase transition or a phase shift.
Traditionally, the process of language learning has been regarded as
a linear system. Richards and Rodgers (2001: 55) comment that it was
assumed language learning involved systematically teaching the building
blocks of a language and how to combine them, for example, moving
from phonemes to morphemes, then to words, phrases, and sentences.
De Bot et al. (2007: 7) explain that this belief has been reflected in a
lot of SLA research. However, they report that numerous studies have
shown that language acquisition and attrition are too complex for a linear
model to predict. According to Lowie (2013: 5), this is congruent with
Dynamic Systems Theory, which posits that language does not develop
in a pre-planned or carefully staged manner.
As the process of language learning is non-linear, learners may exhibit
considerable variability. Verspoor and Nguyen (2015: 310) note that a
learner might use a language form correctly on one day but not on the
9 Foundations 135
next. Larsen-Freeman (1997: 151) provides the well-known example of

beginners’ attempts to learn regular and irregular past tense verbs. To
begin with, a beginner may produce some irregular forms correctly such
as sat, ate and slept. However, they may overgeneralize from examples of
regular verbs and begin adding -ed to previously correct irregular verbs
forms, producing sitted, eated, and sleeped . In other cases, the learner will
use both forms at the same time before eventually acquiring the correct
forms. This variability is thought to indicate the onset of a developmental
transition. It suggests that it may be unnecessary for teachers to focus on
accuracy in the early stages of learning.
The non-linear nature of language learning also gives rise to phase
shifts. While learning can frequently appear to be incremental, at other
times there may be a dramatic change in proficiency. According to
Larsen-Freeman (2006), a critical mass of building blocks is necessary
to form the basis of a transition to a new phase (cited in Lowie, 2013:
5). Beckner et al. (2009: 16) liken this to supplying heat energy to hot
water which transforms it into steam. Verspoor and Nguyen (2015: 310)
explain that these shifts are unpredictable and therefore teachers can only
try to create the conditions and interactions necessary to promote a phase
shift rather than teach for them.
9.1.7 Emergence
A consequence of non-linear systems is the concept of emergence.

Emergence is the appearance of organizational patterns from seemingly
random inputs. The concept of emergence can be seen in a variety
of different phenomena. Beckner et al. (2009: 14) speculate that the
global patterns found in flocks of birds, schools of fish, and national
economies emerge from individuals interacting on a local level rather
than individuals coordinating on a global level.
In language acquisition theory, the concept of emergence is not a
recent phenomenon. In the Natural Approach, Krashen and Terrell (1983:
56) state that ‘the best way to teach speaking is to focus on listening
(and reading) and spoken fluency will emerge on its own.’ In relation
to this, they (ibid.: 155) assert that vocabulary knowledge is critical to
136 I. Pemberton
the acquisition of morphology and syntax. In other words, the acquisi-

tion of specific items of language underpins the emergence of grammar.
More recently, Dolgova and Tyler (2019: 940) assert that a central notion
of usage-based approaches is that ‘language is learned and, in naturalist
settings, emerges through use.’
The phenomenon of emergence can also be seen on a whole language
level. Baicchi (2015: 10) explains that a language can be thought of as
a collection of many speakers who interact and over time their inter-
actions result in the emergence of changes to different aspects of the
language. These changes ‘emerge from interrelated patterns of experience,
social interaction, and cognitive processes’ (Beckner et al., 2009, cited
in Baicchi, 2015: 21). To illustrate this process, Verspoor and Nguyen
(2015: 311) explain how different question forms used for greeting
people in English and Vietnamese have emerged from users in those
speech communities repeatedly hearing and using particular expressions.
An English speaker might greet people with the expression ‘How are
you?’, ‘How are you doing?’ or (in American English) ‘What’s up?’ While
a Vietnamese speaker might also say ‘How are you?’ , alternatively they
might choose to greet people by asking ‘Where are you going?’ or ‘What
are you doing?’ Verspoor and Nguyen (ibid.) refer to such expressions as
Conventional Ways of Saying Things (CWOSTs).
9.2 Construction Grammars

The earlier analysis of historical methods explained how Dell Hymes’
concept of communicative competence was a reaction to Noam
Chomsky’s Universal Grammar (Generative Grammar). Hymes felt
that UG’s overemphasis on linguistic competence marginalized perfor-
mance and he argued for communication to be the goal of language
teaching. At the same time, dissatisfaction with generative grammar
also led to the development of alternative grammars known collec-
tively as constructionist grammars. Two of the most well-known are
Charles Fillmore’s Construction Grammar and Ron Langacker’s Cogni-
tive Grammar. Masuda (2018: 6) explains that the main difference
9 Foundations 137
between these grammars and UG is that Chomsky conceived genera-

tive grammar as independent of semantic meaning as his well-known
example of ‘colorless green ideas sleep furiously’ makes clear. On the
other hand, construction grammars are built on a semantic framework.
According to Croft and Cruse (2004: 265), construction grammars share
three essential characteristics.
9.2.1 Constructions
In a construction grammar, language is represented in the mind as

symbolic units. These units are known as constructions. Croft and Cruse
(2004: 258) define a construction as a pairing of form and meaning.
They note that in addition to syntactic form, a construction may also
include morphological and phonological form. Similarly, in addition to
semantic meaning, a construction may also comprise of its pragmatic and
discoursal meanings. While Langacker (2008: 15) refers to a construc-
tion as a symbolic pairing of a semantic structure and a phonological
structure, he comments that his category of phonological structure also
includes orthographic representation and gesture.
9.2.2 Continuum
An advantage of construction grammars is that they allow for the

uniform representation of all forms of language. According to Langacker
(1999: 1), all language can be represented in terms of constructions.
That is to say, all forms - syntactic, morphological, and lexical - can
be described in terms of constructions (Croft, 2010: 464). In addi-
tion, both core elements (i.e., grammar) and peripheral (e.g., idioms)
can be represented in this way (Hoffmann, 2017: 311). In contrast to
generative grammar, which makes a strict division between meaningless
rules and meaningful lexis, language knowledge can be thought of as
arranged on a cline, with more general grammatical knowledge at one
end and more specific lexical knowledge at the other. Croft and Cruse
(2004: 255) point out that this continuum actually has two dimensions:
from schematic (general) to substantive (specific) and from atomistic
138 I. Pemberton
to complex. In this case, conventional language categories would be

represented in different quadrants of a matrix.
Substantive Schematic
Complex Idioms Syntactic rules
Atomistic Words Syntactic categories
9.2.3 Constructicon
In contrast to generative grammar, which separates grammar and lexis

and considers that they are stored in different areas of the brain,
construction grammarians believe that all language forms are stored in a
single constructicon. Hoffmann (2017: 315) explains that the construc-
ticon is a structured inventory of language items which are represented
in taxonomic networks. The constructicon is hierarchical and different
levels of the hierarchy emerge in a bottom-up manner. At the base of
the hierarchy are micro-constructions. These are concrete examples of
language which occur frequently and have become entrenched. When
micro-constructions have a high type frequency, this leads to the forma-
tion of meso-constructions. These will be low-scope patterns containing
fixed and flexible elements. At the top of the hierarchy are macro-
constructions. These are the most abstract representations. Hoffman
(ibid.) speculates that most language can be generated from micro- and
meso-constructions.
According to Hoffman (ibid.: 313), another important feature of hier-
archical constructicons is inheritance. This means that a superordinate
construction confers its linguistic properties on a subordinate construc-
tion. In relation to this, one of the main differences between different
construction grammars is whether the brain extracts and stores general
language patterns (the principle of economy) or whether it extracts
general language patterns but also retains the exemplars from which
generalizations are made (principle of redundancy).
Goldberg (2009: 93) argues that construction grammars offer a plau-
sible explanation of why languages have particular structures and how
9 Foundations 139
learners can use finite input in order to produce a potentially infi-

nite number of new expressions. She points out that what distinguishes
construction grammars from generative grammar is that they conceive
language knowledge as an inventory of symbols which link meaning to
form and arise from result of exposure to input.
9.3 Summary
The purpose of this chapter was to consider the foundations of usage-
based linguistics. One cornerstone of usage-based linguistics is Dynamic
Systems Theory which is drawn upon to explain language develop-
ment. The seven dimensions of DST are complexity, dynamism, open-
ness, self-organization, adaptation, non-linearity, and emergence. These
dimensions can each be related to second language acquisition.
A key feature of DST is complexity. A complex system has a large
number of components which interact with each other. In addition, a
complex system is hierarchically organized into multiple subsystems at
different levels of granularity. A language meets the criteria to be thought
of as a complex system. First, it has a large number of components
and possesses multiple subsystems which interact with each other. These
include semantics, syntax, lexis, morphology, phonology, and pragmatics.
Each one of these subsystems is complex in its own right. To take lexis as
an example, experts are agreed that the lexicon contains tens of thousands
and possibly hundreds of thousands of concrete examples of language.
In addition, the process of language learning can also be thought of
as complex. It is dependent on many interacting factors such as first
language, target language, degree and quality of input, and learner factors
such as age, gender, aptitude, personality type, motivation, and interests.
A complex system is dynamic and changes iteratively over time. Any
change that takes place is always dependent on the initial condition
of the system, which makes anything other than short-term outcomes
unpredictable. A language is also dynamic. Either the language can
change, or a person’s idiolect can change. While languages are often
thought of as static entities, they change gradually over time as people
use them to interact. In addition, a learner’s knowledge of a language
140 I. Pemberton
is constantly reorganized as the learner’s subsystems interact. A change

in one subsystem influences other subsystems, leading to continuous
change.
A complex system is open, self-organizing and adaptive. An open
system exchanges energy with its environment. Similarly, each of its
subsystems also interacts with their environments. The energy that it
receives from its environment maintains a subsystem and leads to growth
and change. By contrast, a closed system is independent of its environ-
ment, and it is static which leads to entropy over time. A language can be
thought of as an open system. It interacts with and is influenced by, its
social, cultural, and political environments. Similarly, a language speaker
can be thought of as an open system interacting with other speakers of
the same language. Likewise, the subsystems within a language learner’s
cognition are open and interact with each other.
A self-organizing system changes without any external central control.
It changes spontaneously as a result of interactions between local compo-
nents which are unaware of the behaviour of the wider system. These
interactions influence other spatially close components leading to change
spreading across the system. The property of self-organization can also be
seen in languages. As an example, multiple iterative changes take place
to receptive vocabulary before a change in productive skill.
As the environment of an adaptive system changes, the system adapts
to maintain its fit with its environment. Conversely, an environment will
adapt to fit with changes to a system in a process known as co-adaptation.
With regard to language users, individuals can adapt to each other, for
example, by speakers modifying their output for listeners, or by listeners
becoming accustomed to features of a speaker’s language. As regards a
language system, multiple individuals with different idiolects interact and
this process gives rise to a communal language. At the same time, the
communal language exerts an influence on the language use of individual
speakers.
Another key feature of a complex system is that it is non-linear. In
a linear system, a cause will result in a proportional effect. In a non-
linear system, a seemingly insignificant input may result in no change,
a small change, or a major change. A major change in a system is
known as a phase shift. This is when a critical mass of inputs builds
9 Foundations 141
up to precipitate a change in the system. In terms of language learning,

learners do not progress in lockstep, mastering one language item and
then progressing to the next. There is a well-recognized phenomenon of
learners using new forms, regressing to use older incorrect forms, and
using both correct and incorrect forms at the same time.
The non-linearity of complex systems gives rise to the phenomenon
of emergence. Emergence is the appearance of organization from seem-
ingly chaotic or random inputs. In terms of an individual language
learner, the acquisition of multiple concrete examples of language with
structural similarities leads to the emergence of regular patterns. On a
whole language level, the interactions of large numbers of speakers of a
language, and the frequency with which they use familiar and emerging
forms of language eventually give rise to conventional ways of saying
things.
Another cornerstone of usage-based linguistics is constructionist gram-
mars. The Scientific Period described how Dell Hymes’ frustration with
Chomsky’s generative grammar led to the development of communica-
tive competence and this was one of the vectors that led ultimately
to the birth of Communicative Language Teaching. At the same time,
descriptive grammarians, equally frustrated with Chomsky’s genera-
tive grammar, looked to semantics to design and develop grammatical
systems. These included Charles Fillmore’s Construction Grammar and
Ron Langacker’s Cognitive Grammar. Constructionist approaches have
three common features. The first of these is the cognitive representa-
tion of language as form-meaning pairs known as constructions. The
second is that constructionist grammars do not dichotomize between
grammar and lexis but place them on a continuum which runs from
more schematic items to more concrete ones. From this perspective,
items of grammar will reside at the schematic end of the continuum
and lexical items at the specific end of the continuum. The third is the
mental representation of language in an organized taxonomy called a
constructicon. In usage-based theories, the taxonomy is hierarchical (i.e.,
similar items are classified into categories) and superordinate construc-
tions confer their linguistic properties on subordinate constructions. A
constructicon includes both abstract schemas as well as the concrete
examples of language which instantiate them.
142 I. Pemberton
In sum, Dynamic Systems Theory and Constructionist Grammars

represent cornerstones of the foundation of usage-based linguistics,
which is discussed in the next chapter.
References
Baicchi, A. (2015). Construction learning as a complex adaptive system: Psycholin-
guistic evidence from L2 learners of English. Springer.
Beckner, C., Blythe, R., Bybee, J., Christiansen, M. H., Croft, W., Ellis, N.
C., Holland, J., Ke, J., Larsen-Freeman, D., & Schoenemann, T. (2009).
Language is a complex adaptive system: Position paper. Language Learning,
59 (s1), 1–26.
Bloom, P. (2000). How children learn the meanings of words. MIT Press.
Brysbaert, M., Stevens, M., Mandera, P., & Keuleers, E. (2016). How many
words do we know? Practical estimates of vocabulary size dependent on word
definition, the degree of language input and the participant’s age. Frontiers
in Psychology, 7 (1116).
Bybee, J. (2010). Language, usage and cognition. Cambridge University Press.
Corder, S. P. (1973). Introducing applied linguistics. Penguin.
Croft, W. (2010). Construction grammar. In D. Geeraerts & H. Cuyckens
(Eds.), The Oxford handbook of cognitive linguistics (1st ed.). Oxford Univer-
sity Press.
Croft, W., & Cruse, D. A. (2004). Cognitive linguistics. Cambridge University
Press.
De Bot, K., Lowie, W., & Verspoor, M. (2007). A Dynamic Systems Theory
approach to second language acquisition. Bilingualism, 10 (1), 7–21.
De Bot, K., Lowie, W., Thorne, S. L., & Verspoor, M. (2013) Dynamic
Systems Theory as a comprehensive theory of second language development.
In: García-Mayo, M. d. P., Gutierrez-Mangado, M. J., & Martínez-Adrián,
M., (Eds.). Contemporary Approaches to Second Language Acquisition. John
Diessel, H. (2017). Usage-based linguistics. In M. Aronoff (Ed.), Oxford
research encyclopedia of linguistics. Oxford University Press.
9 Foundations 143

language teaching. In X. Gao (Ed.), Second handbook of English language
Ellis, N. C. (2017). Cognition, corpora, and computing: Triangulating research
in usage-based language learning. Language Learning, 67 (S1), 40–65.
Goldberg, A. E. (2009). The nature of generalization in language. Cognitive
Linguistics, 20 (1), 93–127.
Gopnik, A., Meltzoff, A. N., & Kuhl, P. K. (2001). How babies think: The
science of childhood . Phoenix.
Herculano-Houzel, S. (2009). The human brain in numbers: A linearly scaled-
up primate brain. Frontiers in Human Neuroscience, 3(31), 1–11.
Hoffmann, T. (2017). Construction grammars. In B. Dancygier (Ed.), The
Cambridge handbook of cognitive linguistics (pp. 310–329). Cambridge
University Press.
Langacker, R. (1999). Grammar and conceptualization.De Gruyter Mouton.
Langacker, R. (2008). Cognitive grammar: A basic introduction. Oxford Univer-
sity Press.
Lantolf, J. P. (2012). Sociocultural theory: A dialectical approach to L2
research. In S. M. Gass & A. Mackey (Eds.), The Routledge handbook of
second language acquisition (pp. 57–72). Taylor & Francis.
Larsen-Freeman, D. (1997). Chaos/complexity science and second language
acquisition. Applied Linguistics, 18(2), 141–165.
Larsen-Freeman, D. (2011). A complexity theory approach to second language
development/ acquisition. In D. Atkinson (Ed.), Alternative approaches to
second language acquisition. Taylor & Francis.
Larsen-Freeman, D. (2012). Complexity theory. In S. M. Gass & A. Mackey
(Eds.), The Routledge handbook of second language acquisition. Taylor &
Francis.
Lowie, W. (2013). Dynamic systems theory approaches to second language
acquisition. In C. A. Chapelle (Ed.), In the Encyclopedia of applied linguistics.
Blackwell Publishing Ltd.
Masuda, K. (2018). Advancing cognitive linguistic approaches to Japanese
language learning and instruction. In K. Masuda (Ed.), Cognitive linguistics
and Japanese pedagogy (pp. 5–28). De Gruyter Mouton.
144 I. Pemberton
Milton, J., & Donzelli, G. (2013). The lexicon. In J. Herschensohn &

M. Young-Scholten (Eds.), The Cambridge handbook of second language
acquisition (pp. 441–460). Cambridge University Press.
University Press.
Verspoor, M., & Nguyen, T. P. H. (2015). A dynamic usage-based approach to
Willis, D. (1990). The lexical syllabus: A new approach to language teaching.
Collins ELT.
10
Usage-Based Linguistics
According to Goldberg (2009: 98), constructionist grammars, which

consider that language knowledge arises from attention to input, are
generally usage-based. The literature attributes the first use of the
term ‘usage-based’ to Langacker circa 1987 (Javadi & Kazemirad,
2020; Kusyk, 2017; Torrent, 2012). Usage-based linguistics draws upon
Dynamic Systems Theory. That is to say, it considers that languages are
complex systems which are non-linear, self-organizing, and emergent.
While usage-based linguistics has generally been applied to the study
of first language acquisition (e.g., Social-Pragmatic Theory), it has suffi-
cient explanatory power to account for both second and additional
language learning. Notwithstanding the explanatory power of the system,
it is not well-known, but is gaining attention (Tyler, 2010). In an homage
to the Natural Approach, it emphasizes (or reemphasizes) the importance
of input. Verspoor and Nguyen (2015: 315) note that the primary char-
acteristic of their Dynamic Usage-Based approach to second language
instruction is its focus on input. However, it goes beyond the Natural
Approach in asserting that it is not enough for the input to be compre-
hensible, or, for that matter, comprehended, but a learner also has to

Switzerland AG 2024
146 I. Pemberton
internalize its form. The foundation of usage-based linguistics is built

upon the following three key tenets from constructionist grammars.
10.1 Fundamentals
10.1.1 Constructions
The concept of constructions is central to usage-based theories. Verspoor

and Nguyen (2015: 311) note that the language system organizes itself
into form-meaning pairs on multiple levels. Eskildsen, (2009: 335)
points out that these symbolic units can vary from concrete formulas
to abstract schemata. Ellis (2019: 49) comments that constructions are
of all levels of granularity from morphemes via words and phrases to
syntactic frames.
At times in the literature, constructions are referred to as form-
function mappings (e.g., Ellis, 2019: 49; Diessel, 2017: 3). Furthermore,
Bybee (2010: 9) notes the existence of passive constructions and ditran-
sitive constructions in addition to more specific constructions. This
appears to challenge the notion that there is no distinction between
grammar and lexis. However, Ibbotson (2013: 2) notes that both gram-
matical and lexical patterns—for example, morphology, word classes,
syntactic schemas, lexical items, and idioms—are all constructions which
can be placed along a general-to-specific meaning continuum with
grammatical items being more general, and lexical items being more
specific. Taking into account the above, some linguists have developed a
more comprehensive definition. Kassenberg (2020: 272) cites a paper in
which Verspoor (2017) refers to constructions as possessing not two but
three dimensions—spoken or written form, pragmatic use, and semantic
meaning. To capture this additional dimension, she refers to them as
Form-Use-Meaning Mappings or FUMMs.
This leads to an important point that constructions are schematic in
nature. Ellis (2017: 41) comments that experience of specific exemplars
gives rise to schematic constructions. Eskildsen (2009: 336) explains that
there are three categories of schemas: formulas, low-scope patterns, and
fully abstract constructions:
• The simplest of these are formulas which are chunks learnt by rote,
for example, ‘lemme see,’ ‘I wanna do it’ and ‘I dunno.’ These are
also referred to as ‘multi-word expressions’ (MWEs) or, in the Lexical
Approach, ‘fixed expressions.’
• A low-scope pattern (known in the Lexical Approach as a ‘semi-fixed
expression’) is a structure that includes both open and fixed positions
(Bybee, 2010: 9). For example, I wanna VERB, I don’t VERB.
• Finally, Diessel (2017: 3) provides an example of a fully abstract
construction, explaining that ‘player’ is a construction consisting of
a verb ‘play’ and a suffix ‘er’ denoting an actor. This likely gives rise
to abstract patterns such as ‘action-actor’ or the even more abstract
‘verb–er.’
10.1.2 Continuum
In common with constructionist approaches, usage-based theorists make

no distinction between grammar and vocabulary. Verspoor and Nguyen
(2015: 306) state that the core of a language is neither grammar nor
syntax. Similarly, Eskildsen (2009: 335) cites Tummers et al. (2005) who
assert that a common principle uniting usage-based theories is the rejec-
tion of any division between syntax and lexis. Ellis (2007: 23) explains
that usage-based approaches consider the conventions of language to
emerge. He comments that the rules of grammar are nothing more than
descriptions of patterns in language, that is to say, they are explananda
rather than explanans.
10.1.3 Constructicon
In line with constructionist grammarians, usage-based theorists reject the

nativist model of cognitive representation, opting instead for a unified
model—a constructicon—which contains language forms of all levels
of complexity and abstraction. Bybee (2012) cited in Torrent (2012:
2) asserts that usage-based models have greater plausibility than those
which separate structure and usage as they recognize the plasticity of
148 I. Pemberton
the brain and its response to experience. De Bot et al., (2013: 209) go
further, asserting that the brain does not contain an abstract grammar. In
its place, a language user has a network of constructions, and these are
conventions that have emerged over time as a consequence of language
use.
All usage-based theories of representation subscribe to the principle
of redundancy. That is to say they consider that concrete examples exist
alongside schematic structures. Tummers et al. (2005) describe the usage-
based model of language representation as one which includes not only
abstract grammatical patterns of multiple levels of schematicity but also
their lexical instantiations (cited in Eskildsen, 2009: 335). Ellis (2017:
41) alludes to the scope of the constructicon in commenting that it
contains not thousands but tens of thousands of constructions.
10.2 Entrenchment
According to Dynamic Systems Theory, a language is a complex, open,
and adaptive system. An open system exchanges energy with its environ-
ment and an adaptive system changes in response to its environment.
At the same time, the environment changes in response to the system.
In other words, the system and environment co-adapt to each other. In
terms of a language, when the language spoken by the speech community
changes, it is known as conventionalization. When a speech community
influences an individual, it is known as entrenchment.
The process of entrenchment is the key mechanism in the establish-
ment of constructions. Schmid (2017: 9) comments that the term was
first introduced in 1987 by Ron Langacker. According to Langacker
(1999: 93), entrenchment is also referred to in the literature as routiniza-
tion, automatization, and habit formation. He points out that the delib-
erate use of entrenchment implies a gradual learning of a unit over many
instances (Langacker, 2017: 40). In relation to this, Cordes (2017: 269)
comments that entrenchment is a lifelong process of adaptation and reor-
ganization of a speaker’s language knowledge. Entrenchment occurs as a
result of the confluence of a wide variety of learner external factors and
internal mechanisms.
The major external driver of language acquisition is frequency. This

refers specifically to the frequency of words and phrases in the input
to a learner. Ellis (2019: 48) states that a central tenet of usage-based
theories is that frequency of experience influences language acquisition.
Similarly, Verspoor and Nguyen (2015: 313) comment that in usage-
based theories, frequency of input leads to second language development.
Larsen-Freeman (2012: 77) points out that frequency plays a key role in
contributing regular features to learner output.
Frequency strengthens language representations in cognitive struc-
ture. Dolgova and Tyler (2019: 941) cite Bybee (2008) who explains
that frequency reinforces connections between language forms and their
meanings. Diessel (2017: 1) adds that as frequency of occurrence
strengthens language representations in memory; it aids the processing
and production of not only words but also categories and constructions.
According to Ellis (2017: 41), psycholinguistics research demonstrates
that all levels of language structure are influenced by statistical regulari-
ties in language processing.
Researchers agree that repeated exposures are necessary to learn a
word. Schmitt (2008: 343) cites Nation (2001) who reports that while
some studies showed five meetings with words were sufficient to learn
them, other studies found that more than 20 meetings were neces-
sary. Laufer and Nation (2012: 167) deem ten exposures necessary for
subsequent word recognition.
The frequency of occurrence of lexico-grammatical patterns influences
emergence. According to MacWhinney (1999, cited in Baicchi, 2015:
28) learners notice statistical regularities in the language which they
use to make generalizations. Ellis (2017: 41) argues that memories of
specific exemplars give rise to schematic constructions. More concretely,
Eskildsen (2009: 335) explains that language learning is not only item-
based, but that learners develop the three main types of schematic
constructions in a set order: formulas, then low-scope patterns, and
finally fully abstract constructions. The learning of item-based construc-
tions includes phenomena such as collocations and has a direct influence
on acquisition (Tyler, 2010, cited in Dolgova & Tyler, 2019: 941).
Frequency is not the only factor which helps learn a new word.
Another factor of importance is salience. According to Larsen-Freeman
150 I. Pemberton
(2011: 55), learners notice frequent items in input, and language which
is perceptually salient has a better chance of gaining their attention. A
further factor which affects uptake is how clear the meaning of a language
item is. De Bot et al. (2013: 210) cite MacWhinney’s competition model
which sets out that in addition to frequency, cue availability and relia-
bility also influence acquisition. While cue availability refers to frequency
of occurrence, cue reliability refers to the transparency of a language item.
This is a function of how often it is used with the same meaning. For
example, the use of ‘is’ after he or she. This will be frequently present
with the same meaning. On the other hand, the English articles ‘a’ and
‘the,’ although frequent will be used with different meanings, reducing
cue reliability.
The notion that frequency underpins fluency begins to reveal the
scale of the task. Larsen-Freeman (2011: 56) explains that once learners
have assigned a meaning to language items, they can categorize them
and begin to form prototypes. Ellis (2019: 48) cautions that learning
a language is like developing expertise in any other skill. As a rule of
thumb, skill learning requires up to ten thousand hours of practice.
In contrast to generative grammar which posits that learners use
language specific learning skills to acquire a language, usage-based theo-
rists assert that learners use general cognitive mechanisms. Diessel (2017:
1) notes that these learning mechanisms are also employed in the fields
of vision, memory, and thought. A wide variety of cognitive mecha-
nisms are used to learn new language, such as association, categorization,
abstraction, and schematization. The major mechanisms are outlined
below.
10.3 Association
Schmid (2020: 44) argues that the fundamental mechanism of language
learning is association. He defines associations as the ability of one expe-
rience to evoke another. Associations are organized into neural networks
in the brain, which constantly shifts its attention between collections of
neurons firing either in sequence or at the same time. Schmid (2017:
25) sets out what he considers to be the four fundamental types of

association.
10.3.1 Symbolic Association
At the heart of the constructicon are symbolic associations. A symbolic

association is an association between a linguistic form, which could
be either phonological or orthographic, with a specific meaning. Thus,
the sound of a word or sight of its written form evoke the meaning
in the mind of the experiencer. The more often a learner experiences
a symbolic form in association with a specific meaning, the stronger
its mental representation becomes. Nevertheless, Schmid (2017: 440)
cautions that while there is convincing evidence for the entrenchment
of lexis through association, there is less evidence for the entrenchment
of variable schemata.
10.3.2 Syntagmatic Association
Syntagmatic associations arise from the arrangement of content and

function words in specific patterns. The more often items co-occur in a
particular order, the stronger the syntagmatic association becomes giving
rise to regular patterns in language. The existence of collocations and
collostructions (content and function words which co-occur in specific
patterns) provide evidence for syntagmatic associations. Schmid (2017:
441) comments that syntagmatic elements may occur next to each other.
Through frequent co-occurrence, strings of words may unify acquiring
a holistic meaning, e.g., ‘and so on,’ ‘the thing is,’ and ‘you know
what I mean.’ In other cases, as the syntagmatic association strengthens,
semantic and paradigmatic associations may begin to weaken leading to
increasing idiomaticity as individual words lose their original symbolism,
e.g., ‘fire engine’ and ‘firework’ (ibid.: 442). On the other hand, the
elements may be spatially separate but occur within a broader span of
discourse. This gives rise to the potential for the creation of variable slot
and filler schemata.
152 I. Pemberton
10.3.3 Paradigmatic Association
A paradigmatic association is an association between a language item and

its competitors, for example, its synonyms and antonyms. In receptive
activities, these competitors will be pre-activated, i.e., receive a higher
level of activation in the neural network, before a best choice is made
for understanding input. Similarly, in productive activities, competing
language items will receive a higher level of activation prior to selection
and use. Schmid (2017: 443) theorizes that paradigmatic associations
are important in the formation of variable schema. He comments that
the repetition of tokens alone will not lead to the formation of vari-
able schema. On the other hand, type repetition which allows learners to
identify paradigmatic associations will play a role in the development of
low-scope patterns.
10.3.4 Pragmatic Association
A pragmatic association is the link between symbolic, syntagmatic, and

paradigmatic associations to situational, social, and cultural contexts.
These contexts are highly evocative of language. Schmid (2017: 444)
gives a number of affordances of paradigmatic associations. First of all,
they allow users to select situationally appropriate language taking into
account such dimensions as style and register. Second, deictic elements
of an immediate context affect choice of language. Third, language users
can assess such things as their interlocutors’ geographical origins, educa-
tion, status, and so on based on characteristics such as accent, diction,
and language use. Schmid (2017: 444) comments that pragmatic associ-
ations indicate that information about situations is stored together with
the language necessary to articulate them.
In addition to associative learning, usage-based theorists posit a wide
variety of other domain-general learning skills that children and adults
use to learn language. Among the main cognitive operations are catego-
rization, analogy, abstraction, and schematization.
10.4 Categorization
Categorization is a fundamental cognitive skill which we employ from
early childhood to understand our surroundings. It involves grouping
items and gives rise to hypernyms (group names) and hyponyms (group
members). The basis for categorization is perceptual similarities. For
very young children, this will mean characteristics such as size, shape,
and colour. Categorization is a general ability which allows us to make
predictions about unknown entities based on their similarity to known
entities. Historically, judgments about other individuals, things, and
circumstances would have been important for human survival. The cate-
gorization process is fundamental to generalization and schema forma-
tion. Cordes (2017: 276) explains that the basis of categorization is a
prototypical member of a category which embodies most of the shared
elements of other members of the category. Members of a category will
either be closer to or further away from the prototype.
10.5 Analogy
A similar process but one which is distinct from categorization is analogy.
A categorization is made when two things have clear physical similari-
ties, and thus, they can be put in the same class. On the other hand, an
analogy is made when two things may not be obviously similar but may
have functional similarities. One or more elements of the known thing
(the source) are aligned with and mapped onto the unknown thing (the
target). Cordes (2017: 280) gives the example of comparing the solar
system to a nuclear atom. At its centre, an atom has a nucleus, and elec-
trons revolve around the nucleus. Similarly, at its centre, the solar system
has the sun, and the planets revolve around the sun.
154 I. Pemberton
10.6 Schematization
The cognitive processes of categorization and analogy lead to the
process of schematization. This involves the abstraction (or extraction) of
commonalities from two or more exemplars to create a linguistic proto-
type. These commonalities then form a higher level of abstraction. In
usage-based linguistics both the concrete forms and the more abstract
constructions are retained. Eskildsen (2009: 336) notes that learners
progress through increasing levels of schematization: they begin by
employing formulas (unanalysed multi-word units with specific mean-
ings) in familiar situations. They then progress to the use of low-scope
patterns. A low-scope pattern is a schema which is partially fixed and
partially productive; for example, ‘I don’t VERB.’ Usage-based theorists
then hypothesize that learners progress to fully abstract constructions,
e.g., noun phrase-auxiliary-negator-verb.
10.7 Summary
Usage-based linguistics draws upon Dynamic Systems Theory, consid-
ering language to be non-linear, self-organizing, and emergent. With
echoes of Steven Krashen’s Natural Approach, it puts a strong emphasis
on the importance of input, but it goes beyond the Natural Approach
as it does not consider comprehensible input to be a sufficient condition
for acquisition. In fact, it does not even consider comprehended input to
be a sufficient condition. In order to acquire, a learner also has to attend
to the form of the input.
In addition to the dimensions of Dynamic Systems Theory, usage-
based linguistics draws upon constructionist grammars. In particular,
it adopts the three key concepts of constructions, continuum, and
constructicon. Advocates assert language knowledge consists of construc-
tions. A construction is a symbolic representation of syntactic form
and semantic meaning. They may be found at all levels of specificity:
morpheme, phrase, clause, sentence, or even at a suprasentential level.
Secondly, usage-based linguists make no distinction between grammar
and vocabulary. They consider grammar to be nothing more than a
linguistic description of language phenomena rather than something

with cognitive reality for language users. Instead, usage-based linguists
consider language patterns to lie on a cline with highly abstract patterns
at one end and concrete patterns at the other. Thirdly, constructions
of all levels of complexity are represented in a structured inventory
called a constructicon. The constructicon has a high level of redundancy,
containing both abstract linguistic patterns and the concrete examples
that instantiate them.
The process of establishing constructions is called entrenchment. This
takes place as a result of learner external factors and learner internal
mechanisms. The most significant learner external factor is frequency of
occurrence. While frequently recurring tokens result in the emergence of
highly specific lexical patterns, repetition of construction types results in
the emergence of more schematic and productive constructions. In addi-
tion to frequency, memory for language is affected by other factors such
as salience and transparency.
Usage-based theorists believe that learners use general cognitive skills
to learn language. The most basic of these is association. There are four
different kinds of association. The first of these is symbolic associa-
tion which is a holistic association of form and meaning. The second
is syntagmatic association, which is an association of words in a partic-
ular order in a pattern. The words can be adjacent, but they can also
be spatially separate, leading to the creation of more schematic struc-
tures. The third kind of association is paradigmatic association. This is
the association of words and phrases to other items such as synonyms and
antonyms. These may compete for use both in reception and production
and can lead to the creation of variable schema. The final type of asso-
ciation is pragmatic association. These are associations between language
and social and cultural situations, and they underlie the ability to make
judgments about style and register and vary language use based on the
situation.
Another skill which is fundamental to language learning is catego-
rization. This is the process of assigning words to categories based on
perceptual similarities. A related skill is analogy, which is making asso-
ciations between language items which are not perceptually similar but
may be functionally similar. Both categorization and analogy lead to the
156 I. Pemberton
creation of schematic representations and ultimately to productive and

creative language use.
10.8 Implications
Drawing upon complexity theories and constructionist grammars, usage-
based linguistics presents a compelling account of first, second, and
additional language acquisition which has the power to explain a range
of language phenomena. To begin with, it explains why teaching and
practising grammar doesn’t lead to fluency. While the abstract slot and
filler patterns of grammar may be used to create original language, they
only represent a fraction of the patterns that exist within a language and
cannot imbue a learner with the ability to produce conventional ways of
saying things, or for that matter to understand them when they are used
by other speakers particularly where the language is idiomatic.
In addition, usage-based linguistics explains why over-focus on spoken
output activities in pairs or groups (as advocated by Communicative
Language Teaching and Task-Based Instruction) doesn’t lead to fluency.
A fluent language user has a tacit knowledge of language patterns of
all levels of specificity, from the highly abstract to the highly concrete,
including their relative frequencies and contexts of use, and can deploy
this knowledge to process and produce conventional ways of saying
things which identify them as legitimate language users in a speech
community. This knowledge is acquired through repeated exposure to
language in its contexts of situation and culture. This kind of knowledge
cannot be learnt through classroom activities which focus on production.
Furthermore, the concept of emergence explains why learners plateau
for significant periods of time and then experience breakthroughs as they
build a critical mass of language knowledge and then move to the next
level of proficiency.
This begs the question. What then, should learners do to develop
the proficiency to do well in language examinations, or to interact as
legitimate members of a target language community? The foregoing
suggests that to attain fluency, language learners need to experience and
understand language patterns repeatedly in their situational and cultural
contexts and that this kind of practice may allow them to build a crit-
ical mass of language knowledge which will precipitate phase shifts in
their language systems. But what does the research say about usage-based
linguistics? This is the focus of the next chapter.
References
Baicchi, A. (2015). Construction learning as a complex adaptive system. Springer
International Publishing.
Cordes, A. K. (2017). The roles of analogy, categorization, and generalization
in entrenchment. In H. J. Schmid (Ed.), Entrenchment and the psychology of
language learning. De Gruyter Mouton.
De Bot, K., Lowie, W., Thorne, S. L., & Verspoor, M. (2013). Dynamic
Systems Theory as a comprehensive theory of second language development.
In M. D. P. García-Mayo, M. J. Gutierrez-Mangado, & M. Martínez-
Adrián (Eds.), Contemporary approaches to second language acquisition. John
Diessel, H. (2017) Usage-Based Linguistics. In: Aronoff, M., ed. Oxford
Research Encyclopedia of Linguistics. New York: Oxford University Press:
language teaching. In X. Gao (Ed.), Second handbook of English language
Ellis, N. C. (2007). Dynamic systems and SLA: The wood and the trees.
Bilingualism, 10 (1), 23–25.
Ellis, N. C. (2017). Cognition, corpora, and computing: Triangulating research
in usage-based language learning. Language Learning, 67 (S1), 40–65.
Ellis, N. C. (2019). Essentials of a theory of language cognition. The Modern
Language Journal, 103(S1), 39–60.
Eskildsen, S. W. (2009). Constructing another language—Usage-based linguis-
tics in second language acquisition. Applied Linguistics, 30 (3), 335–357.
Goldberg, A. E. (2009). The nature of generalization in language. Cognitive
Linguistics, 20 (1), 93–127.
Ibbotson, P. (2013). The scope of usage-based theory. Frontiers in Psychology,
4 (255), 1–15.
158 I. Pemberton
Javadi, Y., & Kazemirad, F. (2020). Usage-based approaches to second language

acquisition: Cognitive and social aspects. Journal of Language Teaching and
Research, 11(3), 473–479.
Kassenberg, T., Galati, F., de Vries-Zhuravleva, D. & Menke-Bazhutkina,
I. (2020) 12. Film Language Integrated Learning: A Usage-Inspired L2
Teaching Approach. In: Lowie, W., Michel, M., Keijzer, M. & Steinkrauss,
R., (Eds.). Usage-Based Dynamics in Second Language Development. Multi-
lingual Matters: 271–294.
Kusyk, M. (2017). The development of complexity, accuracy and fluency in
L2 written production through informal participation in online activities.
CALICO Journal, 34 (1), 75–96.
Langacker, R. (1999) Grammar and Conceptualization. Berlin, New York: De
Gruyter Mouton.
Langacker, R. (2017) 2. Entrenchment in Cognitive Grammar. In: Schmid, H.
J., ed. Entrenchment and the Psychology of Language Learning. Berlin, Boston:
De Gruyter Mouton: 39–56.
Larsen-Freeman, D. (2011). A complexity theory approach to second language
development/acquisition. In D. Atkinson (Ed.), Alternative approaches to
second language acquisition. Taylor & Francis.
Larsen-Freeman, D. (2012) 5. Complexity Theory. In: Gass, S. M. & Mackey,
A., (Eds.) The Routledge Handbook of Second Language Acquisition. Taylor &
Francis Group:
Laufer, B., & Nation, I. S. P. (2012). Vocabulary. In S. M. Gass & A. Mackey
Francis.
Schmid, H. J. (2017). A framework for understanding linguistic entrenchment
and its psychological foundations. In H.-J. Schmid (Ed.), Entrenchment and
the psychology of language learning (pp. 9–36). De Gruyter Mouton.
Schmid, H. J. (2020). The dynamics of the linguistic system: Usage, convention-
alization, and entrenchment. Oxford University Press.
Schmitt, N. (2008). Instructed second language vocabulary learning. Language
Teaching Research, 12(3), 329–363.
Torrent, T. T. (2012). Usage-based models in linguistics: An interview with
Joan Bybee. Revista Linguistica, 8 (1)
Tyler, A. (2010). Usage-based approaches to language and their applications to
second language learning. Annual review of applied linguistics, 30 (March),
270–291.
11
Empirical Research
The previous chapter set out how usage-based linguistics explains second
language acquisition. It explained that usage-based linguistics is founded
upon Dynamic Systems Theory, a general complexity theory. It also
draws upon the tenets of constructionist grammars. It considers the
fundamental elements of a language to be symbolic pairings of form and
meaning called constructions. Usage-based theories do not dichotomize
between grammar and lexis and constructions can be placed upon a cline.
The cline has two dimensions: from atomicity to complexity, and from
specificity to schematicity. Furthermore, the constructions exist within
a constructicon, which is theorized to be a hierarchically structured
taxonomy of language items. Within the constructicon, there is consid-
erable redundancy brought about by the inclusion of concrete language
items and the more abstract schemas that are created from them.
On a theoretical level, usage-based linguistics exhibits considerable
explanatory power, but what does the empirical research suggest about its
efficacy? In comparison to other areas of linguistics, it is a relatively small
field but one which is growing in importance. A review of the existing
literature reveals multiple research approaches as well as a variety of areas
of inquiry. Approaches to research include case studies, cross-sectional

Switzerland AG 2024
160 I. Pemberton
studies, and corpus analyses. Among the most prominent areas of inquiry
are learner variation and variability, as well as pedagogical approaches,
including the use of autonomy, and the employment of multimedia.
11.1 Case studies

A common way to investigate usage-based theories is to use individual
case studies. While it may be a commonsense in the research community
to favour larger studies of multiple participants, usage-based linguists
consider that these are not representative of the development of indi-
vidual learners. Although the argument can be made that studies with
more participants are likely to provide more generalizable results, Bulte
and Housen (2020: 74) note that developmental patterns across learners
may be ascertained by combining the results of multiple case studies.
Case studies are frequently undertaken to research developmental
trajectory. According to Roehr-Brackin (2014: 773), in contrast to
generative grammar, usage-based linguistics takes a maximalistic view
of cognitive representation. That is to say, cognition contains multiple
representations of the same language, both concrete examples and more
abstract patterns, and includes a high degree of redundancy. A ques-
tion posed by usage-based linguists is how the constructicon, a user’s
knowledge of language, is created. This is generally hypothesized as a
three-stage process, although each stage of the process is described by
different researchers using different terminology.
A learner begins with the acquisition of concrete examples of language.
They are able to apply these in a limited way to the same or very similar
situations. In the literature, these concrete examples are variously referred
to as:
• Formulas (Li et al., 2014: 613)

• Exemplar based formulas, chunks, or entirely specific linguistic
constructions (Roehr-Brackin, 2014: 774)
• Concrete constructional instantiations or fixed recurring multi-word
expressions (Eskildsen, 2012: 337)
The key points here are that the constructions are language patterns that
describe specific phenomena and are completely fixed. The constructions
in the second stage of development are variously referred to as:
• Low scope, semi-productive patterns (Li et al., 2014: 613)

• Item-based schemas (Roehr-Brackin, 2014: 774)
• Partially schematic, partially concrete patterns, utterance schemas or
item-based patterns (Eskildsen, 2012: 337)
The key difference here is that these constructions are partially fixed. For
example, they may contain the same verb, but the object of the verb may
be an open slot which can be filled by various objects. The constructions
in the final stage of the process are variously referred to as:
• increasingly creative and abstract constructions (Li et al., 2014: 613)

• abstract linguistic constructions (Roehr-Brackin, 2014: 774)
• increasingly generalizable schematic constructions (Eskildsen, 2012:
337)
The key points here are that these constructions are abstract, and that
they can be generalized to unrelated phenomena. While this might be
confusing to the novice reader, usage-based linguistics envisages a move-
ment from concrete examples of language to expressions with a fixed
element and an open slot, to entirely abstract representations. A number
of case studies have investigated this developmental trajectory.
11.1.1 Mellow (2006)
A 2006 study by Mellow (p. 654) investigated the linguistic development

of a 12-year-old Spanish-speaking child called Ana (pseudonym). Ana
was accompanying her mother who was studying at university. The study
took place over seven months (201 days) and focused on the develop-
ment of relative clauses. The researcher collected data samples every two
weeks. In total, this consisted of fifteen written summaries of wordless
picture books. Initially, Ana used a small number of concrete exam-
ples of language, but these gradually increased over time. Mellow (2006:
162 I. Pemberton
666) concludes that Ana ‘began with the ability to use individual items,
perhaps as limited scope patterns, and later developed abilities that may
be characterized as grammaticalized linguistic constructions.’
11.1.2 Eskildsen (2012)
A 2012 study by Eskildsen investigated the language development of

two Mexican learners living in the United States, Carlos and Valerio
(pseudonyms). These two learners both attended ESL classes at Port-
land State University, Carlos for three and a half years, and Valerio for 2
years. The research analysed data from the University’s Multimedia Adult
English Learner Corpus (MAELC). The specific focus was the develop-
ment of the participants’ negation (no and not) constructions. In total
Eskildsen analysed 63 hours of video footage from the corpus. In the
footage, the participants were either wearing a microphone or sitting next
to someone wearing a microphone.
At the beginning, Carlos demonstrated some target-like formulaic use
(e.g., I don’t know) together with non-target-like use (I no remember).
Over time, his target-like use increased, and his non-target-like use
declined (e.g., ‘no worry’ became ‘don’t worry’). In Valerio’s case,
Eskildsen analysed his use of pre-negated constructions with (this no
is always) and without a subject (no is melon). The data showed that
Valerio used the pre-negated constructions without a subject throughout,
whereas those with a subject went in and out of use. Eskildsen concludes
that this demonstrates that there is no linear route from ‘no is’ to ‘’is
no,’ in other words, it demonstrates the phenomenon of non-linearity in
second language use.
Eskildsen (2012: 335) concludes that the two learners’ negation
expressions progress over time from item-based expressions to increas-
ingly schematic representations, in line with the predictions of usage-
based theory.
11.1.3 Li et al. (2014)
A 2014 case study by Li et al. also used Portland University’s Multimedia

Adult English Learner Corpus (MAELC) to investigate development in
the verb-argument constructions (VACs) of a learner. This research drew
upon Talmy’s Motion Event typology. Talmy analysed motion events into
four components: the Figure (the thing that moves), the Motion (the
act of moving), the Path (the trajectory followed), and the Ground (the
destination). A Path is encoded using either a preposition or a satellite
and a preposition, e.g., go out to the garden.
Li et al. analysed 120 hours of video footage of one individual (Carlos,
the Spanish-speaking learner of English from Eskildsen’s study) span-
ning a time period of three and a half years. The results showed that he
initially used a limited number of motion verbs, primarily ‘go’ and ‘come’
and that the Path was limited to prepositions (e.g., to, in) or missing
(e.g., go the bank). Over time, Carlos began to use a wider range of
motion verbs together with a wider range of prepositions and satellite
plus preposition combinations. Li et al. conclude that Carlos’s acqui-
sition of verb-argument constructions was item-based, and these items
formed the basis of more complex constructions which emerged over
time. However, they caution that in the period under investigation, it was
not possible to assert that he began to produce abstract constructions.
11.1.4 Roehr-Brackin (2014)
A 2014 study by Roehr-Brackin looks at developmental trajectory from

a slightly different angle. In this study, she looks at the impact of
explicit knowledge on the anticipated usage-based developmental trajec-
tory. The participant was a speaker of British English who resumed
learning German in his 40s, six months before the study began. He was
taught weekly or fortnightly on a one-to-one basis by the researcher over
a period of three years and four months. The researcher made and anal-
ysed eighty-six audio recordings of the participant speaking German to
track his use of two verbs—farhen (go, walk) and gehen (go by car).
The results were as follows: with regard to farhen, the expected devel-
opmental trajectory was observed. The learner progressed from a smaller
number of item-based constructions to increasingly schematic use of the
verb. On the other hand, he used gehen in abstract constructions from
an early stage. Roehr-Brackin speculates that this was due to explicit
learning including L1 transfer and translation. She suggests that such
164 I. Pemberton
use of explicit learning might lead to cognitive overload and to over-

generalization, or what is known in complexity theory as ‘overshooting,’
where the system moves beyond an attractor state leading to a period of
oscillation before the achievement of a stable state.
11.2 Cross-Sectional Studies

While many usage-based researchers use case studies, some favour cross-
sectional studies. Unlike a case study, a cross-sectional study can high-
light general developmental trends in a population. In terms of tracking
development, studies can focus on general linguistic categories (e.g., lexis
and syntax) or they can focus on more specific linguistic phenomena,
e.g., the use of fully fixed and partially schematic chunks.
11.2.1 Verspoor et al. (2012)
A 2012 cross-sectional study by Verspoor et al. made a detailed inves-

tigation of learner development across multiple levels of proficiency.
The study analysed 437 texts written by 12- to 15-year-old Dutch high
school students. To create the texts, the participants were asked to type
responses to writing prompts about general topics into a computer. The
responses were limited to 1000 characters (approximately 200 words).
The researchers allocated the texts to five levels of proficiency from
beginner (CEFR A1.1) to intermediate (CEFR B1.2). The texts were
then coded for 64 different variables. There were five main kinds of
variable: sentences, clauses, verb phrases, lexical items, chunks, and
errors.
The results of the study showed that:
• Mainly lexical changes occurred between levels 1 and 2.

• Mainly syntactic changes occurred between levels 2 and 3.
• A mix of syntactic changes occurred between levels 3 and 4.
• Only lexical changes occurred between levels 4 and 5.
This data suggests that at the beginner stages, lexical knowledge acts as
a precursor to the emergence of syntax, but by the time learners have
reached an intermediate level, their syntactic system stabilizes, allowing
them to focus on expanding their lexis further.
11.2.2 Verspoor and Smiskova (2012)
Some longitudinal studies have looked at the acquisition of chunks

(formulaic sequences). Chunks are abundant in L1 output and are neces-
sary for fluency production. A 2012 study by Verspoor and Smiskova,
defines chunks as: ‘a combination of two or more words expressing
an idea (concept) in a particular context in a grammatically correct
way, which is an authentic, native-like way of expressing that idea’
(2012: 22). They divide chunks into referential, textual, and commu-
nicative phrasemes. A referential phraseme is used to express ideas or
concepts and may refer to real-world people, actions, and events. Textual
phrasemes are used to organize the content of a spoken or written text.
Communicative phrasemes are used to express attitude or beliefs or
interact with others e.g., to greet, take leave of, include, or influence
them. They (ibid: 23) provide the following taxonomy:
Referential Phrasemes
1. Compounds sunbathing, dressing rooms, deep
blue, forest fire, after-sun cream,
two-week holiday, ice-cream
2. Lexical collocations heavy rain, closely linked, apologize
profusely; the sun goes down, take
a dive, strong current, pretty hard,
real close, went wrong, hurt badly
3. Particles Aim at, afraid of, involved in, at
school, in English
4. Complements avoid –ing; necessary to; want/going/
have/manage to; go –ing; keep –ing;
would like to; be able to; know +
clause; say that + clause
5. Phrasal verbs blow up, make out, crop up
6. Idioms to spill the beans, to let the cat out of
the bag, to bark up the wrong tree
7. Similes as old as the hills, to swear like a
trooper
(continued)
166 I. Pemberton
(continued)
Referential Phrasemes
8. Irreversible bi-and trinomials bed and breakfast; kith and kin; left,
right and centre
9. Structures even ADJ + er than; as ADJ as, it is
easy to do, a year ago, two metres
high, so happy that
10. Variable idioms think nothing of –ing; pay a price for
–ing; end up –ing
11. Constructions The sooner we are finished, the
sooner we can go
12. Conventionalized sentence stems one thing I know for sure is…; all
they can do is…
13. Conventionalized sentences It‘s hard to explain. I‘m just who I am.
I (really) like her as a friend
Textual phrasemes
14. Textual prepositions with respect to, in addition to, apart from,
irrespective of
15. Textual conjunctions so that, as if, even though, as soon as,
given that
16. Textual adverbs in other words, last but not least, more
accurately, what is more, to conclude, the
reason for, however
17. Textual sentence stems the final point is …; another thing is …; it
will be shown that …; I will discuss …
Communicative Phrasemes
18. Communicative speech act formulae good morning; take care; you‘re
welcome; suggesting (why don’t
we), concluding (that’s all)
19. Communicative attitudinal formulae in fact, to be honest, it is clear
and sentence stems that, I think that…
20. Commonplaces it‘s a small world; we only live
once; the sky is the limit
(Verspoor & Smiskova, 2012: 23)
Verspoor and Smiskova’s study consisted of two parts. In the first

part, the researchers conducted a cross-sectional longitudinal study of
22 Dutch high school learners of English. The learners were in one of
two groups: a low input group or a high input group. The low input
group received approximately two hours of input per week of conven-
tional foreign language instruction giving a total of 220 hours over two
years. The high input group were part of a semi-immersion programme
and were taught half of their school subjects (History, Mathematics,
Science, etc.) in English. They received 15 hours of English input per
week providing a total of 1320 hours over two years. The researchers
collected writing samples at the beginning and end of the study. The
writing samples were hand-coded for types of chunk and tokens. They
then calculated how many examples of each type of chunk were present
in every hundred words. In the second part of the study, the researchers
analysed the writing samples of two individual learners. In order to do
this, they collected a greater number of writing samples.
The results of the cross-sectional study were contrary to the
researchers’ expectations. While the analysis showed that the high input
group used more chunks by the end of the study, they did not use signif-
icantly more. There was only one significant difference in the type of
chunk—they used four times as many conventionalized sentence stems,
e.g., ‘one thing I know for sure is’ and ‘all they can do is’ (Verspoor &
Smiskova, 2012: 41).
With regard to the individual learners, the researchers investigated
development in chunk tokens and types. To begin with tokens, neither
learner showed a steady increase in chunk use. Using a moving correla-
tion and min–max graphs, Verspoor and Smiskova (2012: 37) looked
for shifts in learner proficiency. The data for the lower input learner
exhibited no clear signs of development showing consistent bandwidth,
neither widening nor narrowing throughout the time period. On the
other hand, for the high input learner, the analysis indicated clear signs
of development. The variability band both widened and narrowed: there
were peaks in the use of chunk types followed by plateaus as the learner’s
language system oscillated and stabilized.
In addition to investigating the number of tokens used by the learners,
the researchers also analysed the number of chunk types. This anal-
ysis showed that the high input learner employed a greater range of
types. Furthermore, they looked at the development of fully fixed chunks
168 I. Pemberton
and partially schematic chunks. They categorize fully fixed chunks and
partially schematic chunks as follows:
Fully fixed chunks (FFCs) Partially schematic chunks (PSCs)

Lexical Collocations Complements
Particles Structures
Compounds Constructions
Phrasal Verbs Conventionalized sentence stems
Textual Adverbials Attitudinal formulae and sentence stems
Speech act formulae Conventionalized sentences
Textual conjunctions Textual sentence stems
Idioms
Variable idioms
Bi and trinomials
Commonplaces
(Verspoor & Smiskova, 2012: 25)
For both FFCs and PSCs, the low input learner was consistent in using
three types each throughout the period of investigation. Furthermore,
there appeared to be no correlation between development in the use of
these two types of chunks. On the other hand, the high input learner
began with using two to three FFCs at the beginning of the period but
increased their use to a maximum of ten types. With regard to PSCs,
the high input learner began by using three types but progressed to use
five types later on. Also, there was a clear correlation in the patterns of
development with peaks and plateaus.
All in all, these research results indicate that the acquisition of chunks
is a slow process. While the researchers were able to demonstrate a
developmental pattern for the high input learner, the cross-sectional
study suggests that, in a shorter time frame, exposure to the target
language alone may not be adequate to build enough language knowl-
edge to support phase shifts in learners’ proficiency, and that it might be
necessary to combine other strategies with an input-based approach.
11.2.3 Gustaffson and Verspoor (2017)
A 2017 study by Gustafsson and Verspoor provided further analysis

of the data from Verspoor & Smiskova’s, 2012 study on a number of
measures. As noted previously, the raw token frequency of both groups

increased over time but there was no significant difference between the
two groups at the end of the period. Additionally, there was an increase
in the variety of types of chunks used by both groups. At the begin-
ning of the study, the high input learners employed four different types,
whereas the low input learners used an average of just under three and
a half different types. Both groups increased their use of different types
and by the end of the study, the low input group actually used more
complements. However, the high input learners used four times as many
conventionalized sentence stems. In overall terms, the high input group
increased their usage to seven different types, and the low input learners
increase their usage to an average of just over four and a half different
types.
The analysis highlighted a significant development in mean chunk
length for both groups. Initially, there was little difference between the
two groups (high = 1.65 words; low = 1.02 words). During the study,
the length of chunks produced by both groups increased. At the conclu-
sion of the research, the high input group were producing chunks with
an average of over three words in length, whereas the low input group
were producing chunks with an average length of 2.6 words.
The analysis also revealed a significant increase in chunk coverage for
both groups. At the beginning of the study, the texts written by the high
input group contained a greater proportion of chunks. Approximately
one third of their texts consisted of chunks whereas texts written by the
low input group contained only 12.4% chunks. At the end of the study, a
greater proportion of texts by the high input group consisted of chunks.
Almost half were chunks (46%), whereas for the low input group around
one third of their texts were composed of chunks (33%). The figure for
the high input group is not far off the proportion of chunks used by L1
writers (55%) (Erman & Warren, 2000: 50).
Overall, this deeper analysis of the data by Gustafsson and Verspoor
builds on the first analysis, revealing development in chunk use for both
groups but more development across all measures for the higher input
learners.
Cross-sectional studies have also been used to compare the attainment
of students on usage-based programmes with those on conventional
170 I. Pemberton
programmes such as structure-based and communicative programmes.

Rousse-Malpat and Verspoor (2018: 71) make the point that such studies
should be longitudinal. They (2018: 56) comment that studies which
show the efficacy of structure-based programmes tend to be short-term
interventions which test discrete items of grammar. It is not surprising
that learners do better on discrete tests of explicitly taught items and such
studies do not take into account the development of general language
proficiency in the longer term. Therefore, it is key that usage-based
research is longitudinal. Attainment studies have investigated second
language acquisition across a range of learners from children to adults.
11.2.4 Dahl (2015)
A 2015 study by Dahl investigated the impact of a usage-based approach

on six-year-old Norwegian school children. The study took place over
eight months. The children were split into two groups: a native language-
based group and a bilingually-based group. The native language-based
group were given conventional lessons once a week for thirty minutes.
All classroom management and instruction took place in Norwegian. In
addition, they received an additional 45 minutes per week of discussion
of English words in their mother tongue during morning meetings. The
bilingually-based group also received one thirty-minute lesson per week
but classroom management and instructions were given in English. Addi-
tionally, discussion of English words (as far as possible) in English took
place during morning meetings totalling an additional 70 minutes of
input per week.
Attainment was assessed using three measures: receptive vocabulary,
sentence comprehension, and sentence repetition. The bilingually-based
group out-performed the native-language-based group on all three
measures but most significantly on sentence comprehension. This high-
lights the importance of input but it should also be noted that the
bilingually-based group not only received more input in lessons and
meetings but they also received it for a significantly longer time each
week.
11.2.5 Piggott et al. (2020)
A study by Piggott et al. (2020) investigated the effect of reducing and

delaying form-focused instruction. The study was conducted on two
intakes of Dutch secondary school students with an average age of 12
years old who were learning English as a foreign language. The study was
longitudinal, covering two school years (a period of one year and nine
months). The first cohort, known as the explicit group, received 86 hours
of form focused instruction out of 233 hours of class time (37%). This
included the use of metalanguage and explicit instruction in grammar
rules. The second cohort, known as the implicit group, received no form
focused instruction until the very end of the programme. During the last
seven weeks of the programme, they were given one hour of grammar
instruction per week.
A grammar test was administered twice to the explicit group (at the
end of the year, and then again at the beginning of the next academic
year. The same test was administered three times to the implicit group,
once before the grammar course, after the grammar course, and then
again at the beginning of the next academic year. The implicit group
did significantly worse on the test before the intensive grammar course,
but then achieved similar results to the explicit group on the post-course
test and the delayed post-course test. These results indicate that students
can achieve the same level of accuracy whether they are taught grammar
extensively or intensively. They appear to suggest that students taught
implicitly still require explicit instruction to achieve the same level of
accuracy, but this is on a grammar test consisting of discrete items.
11.2.6 Rousse-Malpat and Verspoor (2012)
A 2012 investigation by Rousse-Malpat and Verspoor studied a group of

seventy-eight 14-year-old Dutch speakers learning French over a period
of two years. The students studied French for three hours per week. They
were split into two groups. The first group used a communicative text-
book which provided reading and listening input as well as grammar
explanation and practice. The second group were taught using a method
172 I. Pemberton
known as the Accelerated Integrative Method (AIM). This is a commu-

nicative method which utilizes stories, plays, music and gesture. All input
is in the target language.
After two years the students were tested using the Student Oral Profi-
ciency Assessment (SOPA). This consists of a 20-minute interview which
provides oral free response data. Rousse-Malpat and Verspoor (2012:
273) report that the second group, the AIM students, outperformed the
first group on general proficiency and scored equally well on a test of
grammatical accuracy. This indicates that a focus on meaning rather than
a focus on form has a positive effect on general proficiency.
In a follow-up to the 2012 study, Rousse-Malpat et al. (2021) inves-
tigated a group of forty-one 14-year-old Dutch students learning French
over a period of three years. The participants were split into two groups.
In a similar manner to the 2012 study, the first group were taught using a
conventional structure-based approach and the second group were taught
using the AIM approach. The students were again evaluated using the
Student Oral Proficiency Assessment which provided twenty minutes
of free oral production that was transcribed and then analysed by the
researchers. The results showed that the AIM group scored significantly
higher than the structure-based group on general proficiency. They had
a higher speech rate, used a greater range of structures, and produced
more target-like language. They performed as well as the structure-based
group on grammatical measures, actually out-performing them on use of
the past tense. On the basis of the results, the researchers suggest that
teaching using a structure-based approach may be less effective than a
usage-based approach. They advocate a stronger focus on the lexicon,
and in particular the teaching of multi-word expressions with sufficient
repetition for students to learn them.
11.3 Corpus Analyses

Usage-based linguistics posits that a language is a complex system
involving the accretion of hundreds of thousands of elements of linguistic
items. A number of researchers have used corpus linguistic techniques to
address the problem of how to analyse such a large amount of data.
11.3.1 Romer and Berger (2019)
A 2019 study by Romer and Berger set out to measure the development
of verb-argument constructions (VACS) across a wide range of learners.
In order to do this, they used the EFCAMDAT corpus (English First
Cambridge Open Language Database). This allowed them to analyse
samples from more than 12,000 learners. The researchers looked at
texts written by German learners (28,000 texts, 2.8 million words) and
Spanish-speaking Mexican learners (40,000 texts, 3.2 million words) at
five proficiency levels from beginner to advanced. They used a concor-
dancer to investigate the development of the 19 frequent VACS. All have
a verb-preposition-noun pattern.
VAC EFCAMDAT example (and subcorpus)

V about n I always hear about a sore throat caused by nervous cough.
(German B1)
V across n The supermarket is across the train station. (German A1)
V after n I go out to run after the square to buy clothes. (Mexican A1)
V against n The Corleone family was fighting against the other criminal
families in New York. (Mexican A2)
V among n Today, his paintings rank among the most expensive in the world.
(German A2)
V around n I want to travel around the world. (Mexican B1)
V as n I have been working as personal trainer since 1999. (Mexican B2)
V between n Isn’t it very complicated to differentiate between deliberate and
accidental discrimination? (German B2)
V for n I’m looking for a job as a Marketing Assistant. (German A2)
V in n I am in four extra classes. (Mexican C1)
V into n If there is a flood, do not go into a basement. (Mexican B1)
V like n And a complete trim will make you feel like the king. (German
C1)
V of n I also dream of having enough time for painting. (German B1)
V off n The girl grabbed his laptop off him and ran off the street.
(Mexican B2)
V over n I hope you get over your addiction. (Mexican A2)
V through n Therefore we need to push through changes to improve.
(German A1)
V toward n This money can also count toward paying back the loan. (German
B2)
V under n The keys are under the plant in front the door. (Mexican A1)
V with n He experimented with bold colours on his paintings. (Mexican C1)
Romer and Berger (2019: 1095)

174 I. Pemberton
Romer and Berger used three main analytical steps:
1) type-token comparisons
2) construction growth analysis
3) correlation analysis of verb-VAC associations
In order to do the construction growth analysis, for each level of profi-

ciency, the researchers identified the ten most frequent verbs found in
each verb-argument construction. From this it was possible to estimate
at which level a verb emerged as an occupant of a VAC and whether
it continued to play an important role. In order to do the correlation
analysis, the researchers looked at which verbs learners with different
L1s used in specific VACs. They also compared the learner data with
the 100-million-word British National Corpus (BNC) to assess learners’
movement towards native speaker norms.
The researchers found that moving from lower to higher proficiency
levels, learners used an increasing number of VAC types, starting with a
small set of high frequency general verbs and then progressing to use
a wider range of more specific verbs. At more advanced levels, they
moved towards native speaker norms. The comparison of the German
and Spanish L1 speakers showed that they used verbs in VACs in a similar
way, initially using a smaller set of the same verbs before broadening their
repertoire as they became more efficient.
11.3.2 Jo and Oh (2021)
A similar study in 2021 by Jo and Oh followed Romer and Berger (2019)

in using corpus techniques to investigate the development of Korean
learners’ Verb-Argument-Structures (VACs). They analysed data from the
Yonsei English Learner Corpus (YELC). This corpus features over three
thousand argumentative essays written by candidates taking a placement
test for English study at university. To assess whether the Korean students
approached L1 usage, they compared the data from the Korean students
with essays written by British and American students in the Louvain

Corpus of Native English Essays (LOCNESS).
In addition to looking at the students’ writing, the researchers also
collected data from three spoken fluency tasks conducted in L1 and
L2 English, as well as Korean. These tasks collected data from fifty,
one hundred and fifty, and seventy five participants respectively. The
participants were given a verb frame (e.g., ‘She/He _____ about the …’;
‘It _____ with the …’) and asked to think of as many verbs that could
fill the slot as possible within one minute (Jo & Oh, 2021: 10).
The research findings were similar to those of Romer and Berger
(2019). The study found that lower proficiency learners used a smaller
number of more general verbs and produced a higher number of uncon-
ventional expressions. The VACs became more schematic at higher
proficiency levels with the use of less general, more semantically specific
verbs. In addition, at higher levels, the Korean students’ VACs increas-
ingly approximated L1 writers’ usage. Furthermore, the researchers found
evidence of L1 transfer as a result of typological and collocational features
of Korean.
11.4 Variation and Variability Studies

Two-related areas of interest in usage-based linguistic research are varia-
tion and variability. Variation refers to inter-learner differences, whereas
variability refers to intra-learner differences in language use over time
(Verspoor et al., 2012: 240). Traditional linguistics, which has been
concerned with general trends applicable to all learners, has tended
to regard variability as noise to be ignored or discarded (Van Dijk
et al., 2011: 60). The usage-based perspective is that variability is a
normal feature of complex systems such as languages and should be
studied because it provides important data that helps to understand
when and how a system changes, how subsystems interact, and individual
development patterns.
176 I. Pemberton
11.4.1 Bardovi-Harlig (2002)
A 2002 investigation by Bardovi-Harlig compared the language develop-

ment of different learners. The main purpose of the study was to examine
the usage-based progression route from formulas to low-scope patterns to
abstract constructions. There were sixteen participants of a lower profi-
ciency who were either Arabic, Japanese, Korean, or Spanish speakers.
They received 23 hours per week of language classes and were observed
for approximately one year. The researcher collected 1576 written texts
and recorded 175 hours of conversation.
The research focused on how the learners expressed the future. The
results showed that over time only a third of users analysed the formu-
laic expression ‘I am going to write about’ into ‘going to.’ Even though
eight of the students attended the same class, five were frequent users
of ‘going to’ but three rarely used the expression. Bardovi-Harlig (2002:
189) concluded from the study that the use of formulaic expression varies
from user to user.
11.4.2 Larsen-Freeman (2006)
A 2006 study by Larsen-Freeman looked at the development of five

female high intermediate Chinese learners over ten months. Her study
had a time-series design—she assessed the students frequently enough
during the period of the study to estimate their development. Larsen-
Freeman taught the class for two and a half hours once a week
and provided a private tutorial for each learner once a month. She
assessed their fluency, accuracy, and lexical and grammatical complexity.
While the group data resulted in a smooth ascending curve, individual
data showed considerable variability: some individuals progressed and
regressed. Other individuals showed no progress.
11.4.3 Spoelman and Verspoor (2010)
In 2010, Spoelman and Verspoor published a case study of a 19-year-

old Dutch learner of Finnish. This was a longitudinal study conducted
over three years. The researchers collected 54 texts written by the subject.
They analysed the texts for accuracy and complexity. The Finnish system
of case was selected for analysis of accuracy as it has 15 different cases.
To measure complexity, they looked at word, noun phrase, and sentence
level complexity. They also examined whether there was a correlation
between accuracy and complexity.
In terms of accuracy, the learner demonstrated greatest variability in
the early stages. From the mid-point of the study there was a decrease
in variability in the learner’s accuracy as it appeared to stabilize. In
terms of complexity, it was observed that there was a correlation between
word, noun, and sentence complexity which all increased over time. The
learner began by using a greater number of shorter items which became
progressively longer. With regard to the relationship between accuracy
and complexity, examination of the data did not show a correlation.
Overall, the analysis highlighted that development in all of the measures
was characterized by progress and regress, as well as peaks and plateaus.
11.4.4 Lowie et al. (2018)
A 2018 study by Lowie et al compared the variability of two learners. The

learners were 15-year-old Taiwanese identical twins, Gloria and Grace.
The study was conducted over a period of eight months and during that
time, the girls were asked to produce a spoken text three times a week
providing a total of 100 texts each. In their analysis of the texts, the
researchers used the length of T-units to measure syntactic complexity.
A T-unit is ‘one main clause plus any subordinate clause or non-clausal
structure attached to or embedded in it’ (Hunt, 1970, in Lowie et al.,
2018: 115). To measure vocabulary diversity, they used VocD which
employs type-token ratios to measure the lexical complexity of a text.
The results of the study showed that Gloria was slightly more profi-
cient at the beginning of the study and that she stabilized without
progressing in proficiency. On the other hand, Grace began to catch
up with Gloria, showing signs of development including an increase in
variability and an increase of level. These results indicate that even for
identical twin sisters living in the same home and being in the same
178 I. Pemberton
class in the same school, the developmental trajectories of the girls are
different.
11.4.5 Bulte and Housen (2020)
In a 2020 study, Bulte and Housen drew upon data from Verspoor et al.’s
(2012) corpus of student writing for their investigation of learner devel-
opment across multiple levels of proficiency. Bulte and Housen’s analysis
was neither a cross-sectional nor a case study but sought to make a
detailed comparison of the development of ten Dutch learners of English
over a period of two years. The main purpose of the study was to inves-
tigate the relationship between different components of complexity. In
particular, the existence of a relationship between lexical and syntactic
complexity: whether one of these precedes the other, whether they
develop in parallel, or whether their relationship is a random one.
The researchers examined eleven texts per learner according to four
different metrics. Two of these were syntactic. One was the level of
clausal subordination measured by the Sub-Clausal Ratio (SCR). The
other was the Mean Length of T-Units (MLTU). The other two metrics
were lexical: one was the level of phrasal elaboration, measured by the
Mean Length of Noun Phrases (MNLP), and the other was the Guiraud
Index which is used to measure lexical diversity.
As expected, the study found a strong positive relationship between
the two measures of syntactic development (SCR and MLTU). Similarly,
there was a positive relationship between lexical diversity, as measured
by the Guiraud Index, and the MNLPs. In other words, both pairs
of metrics run in parallel to each other. In addition, the researchers
note that all four metrics show an increasing trend over time. They
also found a competitive relationship between clausal subordination and
phrasal elaboration. That is to say, learners either use more subordinate
clauses or more phrasal elaboration, but not both. Unlike some earlier
studies, they didn’t find any evidence of one component acting as a
precursor to another. They concur with other researchers that individual
learners demonstrate substantial variability and thus, general trends from
cross-sectional data cannot be applied to individual learners.
11.5 Multimedia Studies

In the literature, usage-based linguistics is often set in opposition to the
principles and parameters account of first language acquisition provided
by Chomsky’s generative grammar. While generative grammar has been
widely researched, discussed, and argued about since the 1960s, it has
remained in the theoretical domain, with nothing to offer language
teaching pedagogy. On the other hand, usage-based linguistics offers a
cogent and consistent account of language development which applies
not only to first language acquisition but also the development of addi-
tional languages. The insights that it gives provide a basis not only to
critique and challenge existing pedagogical approaches, including the
hegemony of the Communicative Approach, but also it indicates how
to teach language learners in the classroom. It is a complex theory but
one which could potentially bridge the gap between the Ivorian towers
of academia and the classroom and have a potential impact on pedagogy.
As a consequence, a salient theme found in the usage-based linguistic
research is its application to the classroom, and in particular, the use of
multimedia for teaching purposes.
11.5.1 Bylund and Athanasopolous (2015)
A 2015 study by Bylund and Athanasopolous illustrates the impact of

multimedia on language learning. The researchers looked at the use
of verb-argument constructions (VACs) by eighty-two Swedish univer-
sity students with an average age of twenty-three who were proficient
speakers of English. They were compared with a control group of seven-
teen native English speakers. Speakers of languages without grammatical
aspect such as Swedish tend to focus on the Ground (the end point of
the Motion), e.g., ‘two nuns walk to a house,’ whereas speakers of an
aspect language such as English focus more on the progressive aspect of
the Motion, e.g., ‘two nuns are walking’ while de-focusing the Ground
(Bylund & Athanasopolous, 2015: 132). The participants in the study
were asked to match a scene with an intermediate level of orientation
to the Ground (endpoint) to either a representation with a low or high
180 I. Pemberton
level of orientation to the Ground representation. The research showed

that those students who watched more English language programmes on
television placed more emphasis on the progressive aspect of the Motion
and, at the same time, de-emphasized the Ground. The research suggests
the effectiveness of multimedia use in language teaching.
11.5.2 Nguyen (2013)
A 2013 study by Nguyen investigated the efficacy of using multimedia

for teaching purposes. This study took place at a Vietnamese univer-
sity. It involved 163 participants who were 18–19-year-old non-English
major students. Notwithstanding seven years of EFL classes in secondary
school, the participants were judged at a CEFR A1 level in terms of
listening and reading proficiency. The experimental group consisted of
94 students and 79 students acted as a control group. The programme
of study lasted from 12–15 weeks dependent on whether the groups had
four or five English classes per week. The groups were pre- and post-
tested with multiple instruments including a general English proficiency
test and free spoken and written production tasks.
The teaching method for the control group was a communicative task-
based approach. This focused on grammar, interaction and output but
involved little input (Nguyen, 2013: 106). The teaching approach for the
experimental group was as follows: The students watched a movie. The
movie was broken down into 2-minute segments. Three or four segments
were covered per class with each segment occupying about 15 minutes
of class time. The purpose of the first viewing was global comprehension
of the scene. On subsequent viewings, students focused on the language
used by the actors. They were expected to understand each utterance. As
a rule of thumb, the aim was to repeat each utterance eight times. The
purpose of this was to help students internalize the language.
In terms of results, the experimental group made greater gains than
the control group in terms of both receptive and productive skills. They
made greater gains in the general proficiency test and in writing. Despite
not practising speaking during the study, they performed as well as the
control group on the speaking task.
11.5.3 Koster (2015)
A similar study was undertaken in 2015 by Koster. The participants were

19 German learners of Dutch at a CEFR B1 level of proficiency. Koster
took the same approach as Nguyen. The students repeatedly viewed two-
to-three-minute segments of the movie. A PowerPoint presentation with
the written text was used to analyse each segment. Each segment was
discussed for approximately 15 minutes. Three or four segments were
covered per class. While the students showed significant progress between
the pre- and post-tests, the study had two flaws making it difficult to
compare the results with other research. Firstly, there was no control
group. Secondly, the tests only measured reading and writing but not
listening or speaking.
11.5.4 Irshad (2015)
Two further multimedia studies were undertaken by Irshad in 2015 at a

Sri Lankan university. In the first study, there were 88 participants. The
participants were 20-year-olds in the Faculty of Management Studies and
Commerce. They were divided into two groups with 48 students in the
experimental group and 40 students in the control group. The groups
were given two hours of English per day for two days a week for one
semester.
The experimental students watched a movie in short 2–3 minute
segments with each segment repeated a minimum of six times. A Power-
Point slide deck was used to support the delivery and to present the
dialogue in written form. The control group received conventional
lessons which employed a communicative approach. The groups were
pre- and post-tested. Both groups gained in general proficiency and in
writing proficiency. Irshad (2015: 141) comments that the difficulty in
monitoring spoken production and giving effective feedback to large
heterogeneous classes makes the usage-based approach the more prac-
tical of the two. Irshad also speculates that a longer study might produce
different results.
182 I. Pemberton
To further investigate the use of multimedia, Irshad (2015: 100)

conducted a second study. This time, 228 students took part in the
study. They were assigned to one of three classes: 73 students received
conventional communicative classes and 100 students were shown a
movie in segments supported by a PowerPoint presentation. For the third
group of 55 students, the instructions in the PowerPoint slide deck were
converted into a computer programme which was written to CD-ROM.
The students used the CD-ROMS in 34 self-paced e-learning lessons. All
three groups experienced gains in general English proficiency and writing
but the third group who used the e-learning materials made the greatest
gains. Irshad (2015: 250) suggests that individual self-paced classes might
help to solve the problems of large heterogeneous classes.
11.6 Autonomy Studies

A salient theme in the literature is learner autonomy and, in particular,
the Online Informal Learning of English (OILE). OILE refers to off-
timetable unstructured exposure to and interactions with English that
learners have in their free time. These include watching target language
dramas, documentaries, and movies, reading in the target language and
interacting with others via social media and in forums. This is a new
and developing field of research which draws upon usage-based linguis-
tics for its theoretical basis. The initial questions which researchers have
addressed are what activities learners do, and what impact it has on their
language proficiency.
11.6.1 Toffoli and Sockett (2010)
To determine what kinds of activities leaners undertook in their spare

time, Toffoli and Sockett surveyed 222 students at the University of
Strasbourg. These were non-English major students with little require-
ment for English. The researchers found that learners undertook activi-
ties which involved listening, reading, and writing.
To begin with listening activities, the survey showed that nine-tenths

of the respondents listened to English at least once a month, and more
than half listened to English once a week. A quarter of the respondents
watched movies and a similar figure watched TV series. The researchers
point out that these activities are qualitatively different from listening in
a classroom setting, both in terms of duration and richness of representa-
tion. One episode of a TV drama could be a half hour long, and a movie
might be an hour and a half. In addition, the language in a TV drama
has a narrative structure, occurs in a social and cultural context, and has
continuity and elaboration from episode to episode.
In terms of reading, respondents were most likely to read social media
such as Facebook on a weekly basis. Survey respondents were most likely
to write short (two sentences or less) personal texts on social media
(blogs, instant messages, social networks, email) including commenting
on others’ contributions. By contrast, few respondents used the internet
(e.g., VOIP) for spoken interaction in English. The authors of the study
comment that teachers should validate and encourage informal online
behaviour and also suggest giving their students agency to choose their
own class listening materials, as well as peer-teaching others how to
exploit technologies for language learning.
11.6.2 Sockett and Toffoli (2012)
A 2012 follow-up study by Sockett and Toffoli further investigated two

important areas of interest: how much time learners spend on each
activity and whether the activities are sustained over time. In contrast
to the survey approach of their earlier study, this research focused on five
French university students who were studying English. These students
had 48 hours of formal lessons—2 hours per week over 24 weeks. The
participants were asked to keep a log of what websites they used and
how much time they spent on each website for a period of 2 months.
They were also asked to evaluate what they learnt from each area of their
informal learning of English.
At the end of the study, the participants indicated gains in the areas
of listening and vocabulary. With regards to listening, three of the five
184 I. Pemberton
learners said that listening to music improved their listening compre-

hension. They were able to listen strategically: listening for gist, then
listening and reading song lyrics. All five of the learners watched TV
series (comedies/dramas). They commented that this activity trained
the ear, taught them how people actually speak, and improved their
pronunciation. In terms of other areas of improvement, the learners were
principally aware of making vocabulary gains as they progressed through
a cycle of perception, encoding, and use of chunks of idiomatic language.
On the basis of the results, the authors of the study highlight the
importance of social media, on-demand music sites, and TV programmes
for language learning. While watching TV programmes does not involve
two-way interaction, it provides learners with prolonged contact with the
target language, and exposure to the social values and cultural references
of the language users. They also speculate that the repetitive structure of
situation comedies/dramas helps learners to learn the language. Further-
more, OILE has the effect of highlighting the yawning chasm between
what is taught in the classroom and what students learn from their
extracurricular language learning activities.
11.6.3 Sockett (2013)
In an ambitious study, Sockett investigated whether online learners

might be applying the domain-general skills children employ for first
language acquisition to second language acquisition. These skills include:
1. Establishing joint attention

2. Intention reading
3. Forming categories
4. Detecting patterns
5. Imitation
The study involved nine female postgraduate students with an upper-

intermediate to lower advanced level of English. Over a period of three
months, they were asked to keep a blog which focused on their use of
online learning. This resulted in a corpus of 35,000 words.
First, Sockett (2013: 52) looked for evidence of intention reading

skills which include establishing joint attention, directing attention, and
reading intention. In the early stages of L1 learning adults and older
children will direct the attention of babies by gazing and pointing but
as they grow, children begin to direct the attention of their caregivers
and siblings. Sockett gives a number of examples from online learning:
including chatting on Facebook and watching television. In all cases,
learners use questions to direct their interlocutors’ attention. Next, the
researcher provides a number of examples of how learners use reading
strategies, e.g., skimming, looking for key words, and consulting other
texts to figure out the communicative intentions of the author of the
reading material. This implies that the very act of trying to understand
something is an attempt to ascertain the communicative intentions of
another person.
After that, Sockett looked for evidence of general cognitive processes
including forming categories, finding patterns and cultural learning.
The participants provided a number of examples of using strategies to
understand vocabulary; for example, by creating categories for equestrian
vocabulary and putting extra attentional resources into understanding
technical vocabulary.
With regard to finding patterns, the rich contexts that recurring
patterns occur in in Facebook, TV series and Wikipedia help learners
to identify meaning and use. Finally, the study participants provided a
number of examples of cultural learning including listening to music and
watching TV programmes and imitating the language to improve into-
nation and retention. Also, in a Facebook chat, they integrated L1 users’
written English into their own writing.
11.6.4 Kusyk (2017)
In 2017, Kusyk repeated Toffoli and Sockett’s (2010) study but with a
different set of students. A questionnaire designed to investigate online
habits was distributed to 953 students: 538 French students and 415
German students. The participants were non-English majors. The results
of the questionnaire were as follows. Both groups had similar frequency
186 I. Pemberton
rates for receptive activities but differed in their production activities.

In terms of receptive activities, eight out of ten participants from both
groups listened to music regularly and over half of both groups watched
TV. In terms of productive activities, the German students partici-
pate more regularly in speaking, chatting, emails, and commenting and
messaging on Facebook.
11.7 Summary
The usage-based linguistics literature reveals a wide range of macro-
approaches to usage-based research. One common research approach
is to use case studies. A case study can focus on a range of language
phenomena. For example, a study can look at specific items. One case
study examined the development of relative clauses. The child participant
demonstrated a developmental pattern from the use of individual items
to low-scope constructions. Another study looked at the use of negation
by adult learners. This study demonstrated a developmental trajectory
from item-based expressions to increasingly abstract representations. A
further case study examined the use of verb-argument constructions to
express motion. This study showed that the learner began with a limited
set of general verbs and prepositions to express motion events. Over time
the learner began to use an increasingly specific set of verbs and prepo-
sitions. These studies lend weight to the hypothesis that learners use
concrete examples of language to form more abstract representations.
Whereas case studies can be used to track individual developmental
patterns, cross-sectional studies can be used to look at general develop-
mental trends. In one study, texts written by learners were categorized
into five levels of proficiency, then analysed to track the use of different
language phenomena. This analysis indicated that at lower levels of profi-
ciency, lexical development precedes syntactic development, and that
syntax stabilizes at an intermediate level of proficiency allowing for the
further development of lexis. Another area which usage-based linguists
have investigated cross-sectionally is the use of chunks. An initial study
compared two groups of learners: a high and low input group. Although
the learners in the high input group used more tokens by the end of the
study, they did not use significantly more. In a later reanalysis of the data,
the high input group were shown to use significantly more types, longer
chunks, and to use a greater proportion of chunks per text.
Cross-sectional studies can also be useful to compare different teaching
approaches. In order to compare usage-based instruction to form-focused
instruction, researchers have experimented with delaying grammar input,
and not teaching grammar at all. A study assessing the effect of delaying
grammar instruction found that learners who received an intensive
grammar course towards the end of their second year of instruction
did as well on a grammar test as learners on a conventional structure-
based course. A pair of studies investigating the effect of not providing
any explicit instruction in grammar compared the attainment of two
groups of learners: one group followed a conventional structure-based
course, whereas the other group were given a usage-based course but
not taught any grammar. In both cases, the learners on the usage-based
course became more proficient in general English and did as well on
a grammar test. What these studies show is that even without explicit
instruction in grammar, learners can implicitly acquire the underlying
abstract patterns of the language system.
As a language is a complex system consisting of a huge amount of
data, corpus techniques are well-suited to usage-based linguistic anal-
ysis. A pair of studies examined texts produced by learners at different
levels of proficiency. The specific focus was the use of high frequency
verb-argument constructions. In both cases, the researchers were able to
identify which constructions were used at which levels of proficiency,
giving a plausible account of learner development. The studies showed
that learners began by using a small set of general verbs in a narrow range
of constructions. As they became more proficient, they increasingly used
more specific verbs in a wider range of constructions.
A major area of interest within usage-based linguistics is variation
and variability. A number of studies have examined the variability of
specific linguistic phenomena. One study looked at the development
of future expression. The results of the study showed that development
differed from learner to learner. Other research has examined variability
in general measures of language proficiency. One study looked at the
188 I. Pemberton
development of fluency and accuracy, as well as lexical and grammat-

ical complexity in learners of English. While averaging the learners’ data
resulted in a smooth ascending curve, a comparison of individual data
demonstrated progression, no progression, and progression followed by
regression. Another study focused on the language development of twins.
Notwithstanding their identical personal and educational environments,
the researchers found developmental differences. One twin showed no
progress, but the other twin demonstrated a clear developmental pattern.
Finally, a further study compared learner variability to ascertain whether
lexis and syntax entered into parallel, supportive, or competitive relation-
ships. Overall, these studies show that even where cross-sectional analysis
indicates general trends, they are underpinned by considerable individual
variation.
Academics often contrast usage-based theories with generative
grammar. These are both highly theoretical complex explanations
of language acquisition. However, there is one important difference
between the two theories: generative grammar is of no use to language
teachers. On the other hand, usage-based linguistics can translate into
classroom pedagogy. A number of key studies in the literature have
focused on how to do this. Some research suggests that use of multimedia
enhances language development. For example, advanced L2 learners
who watched more English language TV were better able to use verb-
argument constructions in line with L1 user norms.
A number of studies have looked at introducing multimedia into
the classroom. One study involving Vietnamese university students
compared a communicative teaching approach with a usage-based
approach using movies. When the students were post-tested, those in
the movie group surpassed the students who received the communicative
lessons in listening and reading, as well as writing. Notwithstanding that
the experimental group had had no speaking practice they were equal to
the control group on this measure.
Two similar studies were conducted with Sri Lankan university
students. In the first study, an experimental group watched a movie
in short segments. The control group were given conventional commu-
nicative lessons. Both groups made similar gains in general English and
writing proficiency. In the second study, students were either taught
using a conventional communicative approach, a teacher-led movie

approach, or a self-paced e-learning movie approach. In this case, the
group who used the self-paced movie approach surpassed the other two
groups on a post-test.
A final theme that emerges from the literature is autonomy. A number
of studies have been conducted in the area of online informal learning
of English. Some researchers have investigated what learners do online.
An initial study looked into the activities of French university students.
It found that significant numbers of them listened to English songs,
watched movies and TV series, and used English for communication
on social media. A similar investigation compared French and German
university students. The study found that while both groups engaged
in receptive activities to the same extent, the German students were
more likely to engage in productive activities such writing emails and
commenting on Facebook. A related study asked participants to keep a
log to ascertain the amount of time they spent on activities, whether they
sustained them over time, and how beneficial they thought they were.
The participants positively evaluated the impact of their independent
learning and its impact on their language development.
In summary, in examining the literature, this chapter has looked at
a range of macro-approaches to usage-based research, including case
studies, cross-sectional studies and corpus analyses, as well as micro-
techniques such as the use of traditional statistical measures and tech-
niques developed specifically by usage-based researchers. These studies
have provided evidence for the general usage-based developmental
pattern of moving from concrete items to more abstract schemata. They
have also shown that students taught using a usage-based approach
can do as well as or better than students taught using structural and
communicative approaches. The literature gives insight into how a usage-
based approach can be realized pedagogically: two broad pedagogical
approaches include the utilization of autonomous learning and the use
of multimedia.
190 I. Pemberton
References
Bardovi-Harlig, K. (2002). A new starting point? Studies in Second Language
Acquisition, 24 (2), 189–198.
Bulté, B., & Housen, A. (2020). A DUB-inspired case study of multidimen-
sional L2 complexity development: Competing or connecting growers? In
W. Lowie, R. Steinkrauss, M. Keijzer & M. Michel (Eds.), Usage-based
dynamics in second language development. Multilingual Matters.
Bylund, E., & Athanasopoulos, P. (2015). Televised whorf: Cognitive restruc-
turing in advanced foreign language learners as a function of audiovisual
media exposure. The Modern Language Journal, 99 (S1), 123–137.
Dahl, A. (2015). Input and language competence in early-start foreign language
classrooms. In T. Cadierno & S. W. Eskildsen (Eds.), Usage-based perspectives
on second language learning (pp. 125–152). De Gruyter Mouton.
Erman, B., & Warren, B. (2000). The idiom principle and the open choice
principle. Text (The Hague), 20 (1), 29–62.
Eskildsen, S. W. (2012). L2 negation constructions at work. Language Learning,
62(2), 335–372.
Gustafsson, H., & Verspoor, M. (2017). Development of chunks in Dutch L2
learners of English. In J. Evers-Vermeul & E. Tribushinina (Eds.), Usage-
based approaches to language acquisition and language teaching. De Gruyter,
Inc.
Irshad, F. M. (2015). Second language development through the lens of a dynamic
usage-based approach. University of Groningen.
Jo, R., & Oh, S.-Y. (2021). A Usage-based study of L2 constructional develop-
ment: Combining learner corpus and experimental data. English Teaching,
76 (1), 3–31.
Koster, D. E. S. (2015). A dynamic, usage-based approach to teaching L2
Dutch. Dutch Journal of Applied Linguistics, 4 (2), 257–264.
Kusyk, M. (2017). The development of complexity, accuracy and fluency in
L2 written production through informal participation in online activities.
CALICO Journal, 34 (1), 75–96.
Larsen-Freeman, D. (2006). The emergence of complexity, fluency, and accu-
racy in the oral and written production of five chinese learners of english.
Li, P., Eskildsen, S. W., & Cadierno, T. (2014). Tracing an L2 learner’s motion
constructions over time: A usage-based classroom investigation. The Modern
Language Journal, 98(2), 612–628.
Lowie, W., Verspoor, M., & van Dijk, M. (2018). The acquisition of L2
speaking: A dynamic perspective. In R. A. Alonso (Ed.), Speaking in a second
language. John Benjamins Publishing Company.
Mellow, J. D. (2006). The emergence of second language syntax: A case study
of the acquisition of relative clauses. Applied Linguistics, 27 (4), 645–670.
Nguyen, T. P. H. (2013). A dynamic usage-based approach to second language
teaching. DPhil University of Groningen.
Piggott, L., Tribushinina, E., & de Graaff, R. (2020) 11. The icing on the
cake? Effects of explicit form-focused instruction after two years of implicit
EFL learning. In W. Lowie, M. Michel, M. Keijzer & R. Steinkrauss (Eds.),
Usage-based dynamics in second language development. Multilingual Matters,
249–270.
Roehr-Brackin, K. (2014). Explicit knowledge and processes from a usage-
based perspective: The developmental trajectory of an instructed L2 learner:
Explicit processes from a usage-based perspective. Language Learning, 64 (4),
771–808.
Römer, U., & Berger, C. M. (2019). Observing the emergence of construc-
tional knowledge: Verb patterns in german and spanish learners of english
at different proficiency levels. Studies in Second Language Acquisition, 41(5),
1089–1110.
Rousse-Malpat, A., & Verspoor, M. (2012). Measuring effectiveness in Focus
on Form versus Focus on Meaning. Dutch Journal of Applied Linguistics,
1(2), 263–276.
Rousse-Malpat, A., & Verspoor, M. (2018). Foreign language instruction from
a dynamic usage-based perspective. In A. E. Tyler, L. Ortega, M. Uno &
H. I. Park, (Eds.), Usage-inspired L2 instruction: Researched pedagogy. John
Benjamins Publishing Company:
Rousse-Malpat, A., Koote, L., Steinkrauss, R., & Verspoor, M. (2021).
Parlez-vous francais? Effects of structure-based versus dynamic-usage-based
approaches on oral proficiency. Language Teaching Research, 1–25.
Sockett, G. (2013). Understanding the online informal learning of English as
a complex dynamic system: An emic approach. ReCALL : The Journal of
EUROCALL, 25 (1), 48–62.
Sockett, G., & Toffoli, D. (2012). Beyond learner autonomy: A dynamic
systems view of the informal learning of English in virtual online commu-
nities. ReCALL : The Journal of EUROCALL, 24 (2), 138–151.
Spoelman, M., & Verspoor, M. (2010). Dynamic patterns in development of
accuracy and complexity: A longitudinal case study in the acquisition of
Finnish. Applied Linguistics, 31(4), 532–553.
192 I. Pemberton
Toffoli, D., & Sockett, G. (2010). How non-specialist students of English

practice informal learning using web 2.0 tools. Asp, 58, 125–144.
Van Dijk, M., Verspoor, M., & Lowie, W. (2011). Variability and DST. In
M. H. Verspoor, K. de Bot & W. Lowie (Eds.), Dynamic approach to second
language development: Methods and techniques. John Benjamins Publishing
Company
Verspoor, M., Schmid, M. S., & Xu, X. (2012). A dynamic usage-based
perspective on L2 writing. Journal of Second Language Writing, 21(3),
239–263.
Verspoor, M., & Smiskova, H. (2012). Foreign language writing development
from a dynamic usage based perspective. In R. Manchón (Ed.), L2 Writing
Development: Multiple Perspectives. De Gruyter, Inc.
Part III
Designing a Blueprint for Change
12
Principles
A case for change in the field of language teaching was established in

Part I of the book. This began by examining historical language teaching
methods, before analysing modern teaching pedagogies from 1970
onwards. The move in the early seventies from a concept of linguistic
competence to a pedagogical goal of communicative competence ushered
in the current paradigm—Communicative Language Teaching. While it
tends to be accepted unquestioningly by its adherents and all trainee
language teachers will likely receive instruction in its methods, the anal-
ysis led to the revelation that CLT lacks a plausible explanation of
language learning, and as a consequence, many (and perhaps most) CLT
approaches overemphasize output while underemphasizing input.
It is one thing to be critical of an extant paradigm, but another to
set out what could take its place. As the major weakness of CLT is its
lack of a theory of learning, it is a necessity of any alternative peda-
gogy to have a plausible theory of language learning. The nature and
scope of this theory was explored in Part II of the book. The theo-
ries of evolutionary biology, first language acquisition, constructionist
grammars, chaos and complexity, and usage-based linguistics were drawn
upon to build a theoretical foundation.

Switzerland AG 2024
196 I. Pemberton
One of the vexing things about linguistic theory is that it is often

difficult to translate into classroom practice. Being remote from the class-
room, linguists have difficulty in reimagining a space that they don’t work
in, and being remote from the ivory towers of academia, teachers are at
the mercy of educational fads, and tend to settle on intuitive notions
of what works best, even if these lack theoretical underpinnings. The
task at hand is to somehow weave together the collective experiences
and understandings of language educators throughout history with the
cutting-edge theorizing of cognitive linguists.
Part III has the dual function of establishing a set of pedagogical prin-
ciples and applying these to the design and development of classroom
materials. It argues for a contextual approach—a multimedia approach
to language teaching and learning which draws upon the intuitions and
theories of educators over the last two hundred years, but at the same
time, is firmly grounded in current linguistic theory. It begins by setting
out the six pedagogical principles of a contextual approach. The first four
relate to optimizing the conditions for language learning, whereas the
final two principles relate to learner behaviours.
1. The Context Principle

2. The Input Principle
3. The Modality Principle
4. The Autonomy Principle
5. The Iteration Principle
6. The Attention Principle
12.1 Environmental Principles

12.1.1 The Context Principle
The Context Principle states that context promotes a deeper under-

standing of language. Context is defined, not as the physical location in
which the learning takes place, or the linguistic context in which a word
occurs, but as the social, cultural, and situational context in which the
12 Principles 197
language occurs. Thus, a context is a framework which contains a group

of related concepts that logically connect to describe and explain ideas.
These concepts can be people, things, actions, events, or circumstances.
Circumstances can be such things as time, place, ways of doing things,
or the manner in which things happen. The contention is that context is
critical for effective language acquisition.
While context is fundamental for both language processing and
production (Finkbeiner et al. 2012: 2), historical teaching methods have
tended to underestimate its importance. According to Yalden (1983: 26),
the structural syllabus taught the restricted meaning of decontextual-
ized words and sentences, but it did not account for the way in which
they were used in communication. According to Lopez-Sanchez (2009:
30), the Audiolingual Method, which was influenced by both struc-
turalism and behaviourism, regarded language learning as the mastery of
‘a finite set of patterns’ leading to linguistic competence. However, CLT
advocates dissatisfied with Audiolingualism pointed out that contextual
factors were necessary to understand the meaning of a word. Wilkins
(1976: 16) maintains that while languages may be general systems,
language has to be related to its social context to be understood. In his
book, The Lexical Approach, Michael Lewis (1993: 97) argues that it is
important for materials writers to produce coursebooks which feature
language in contexts that make the pragmatic meaning is clear. As can be
seen, proponents of different pedagogical persuasions agree that context
is highly important for understanding the meaning of language.
12.1.1.1 Authenticity
A further point of agreement between advocates of different pedagog-

ical approaches is that input should be authentic. Authentic texts are
most frequently defined as target language materials which have not
been created for language teaching. Authenticity is generally regarded
as a principle of Communicative Language Teaching, but with the many
different approaches this encompasses, it is not clear how consistently
teachers apply it. There are a number of important reasons for using
authentic texts. First of all, they enhance understanding. Usage-based
198 I. Pemberton
linguists Verspoor and Nguyen (2015: 308) state that it is better to teach
language in meaningful contexts which approximate the norms of social
and cultural usage. When learners are exposed to sufficient authentic
input, it helps them to establish links between form and meaning (ibid.:
315). Secondly, authentic texts promote retention. Hill (1967: 118)
points out that ‘it is much easier to learn a word or structure if it grows
naturally out of a context than if it is taught without such a context,
or if the context is a forced, artificial, classroom one.’ In summary,
authentic contexts supply the raw data that becomes the building blocks
for construction of learner interlanguage. They are concrete evidence of
how language is actually used. As with any inputs to a system, and partic-
ularly a complex one, the higher the quality of the inputs, the greater the
chance of creating a well-constructed and functioning system.
12.1.1.2 Paralinguistic Information
A context functions in a number of different ways to promote under-

standing, retention and ultimately, production. First of all, context
provides paralinguistic clues which promote a deeper understanding of
language. In the Natural Approach, Krashen and Terrell (1983: 32) point
out that context allows learners to understand and acquire new gram-
matical structures. Children learning their first language are facilitated
by parents and other adults who restrict the topics of conversation to
a child’s immediate environment. Verspoor and Nguyen (2015: 314)
comment that context supplies visual cues such as eye gaze, gestures,
and posture. They (ibid.: 316) explain that children make associa-
tions between what they hear and what they see which helps them to
remember things. In addition to deictic elements, Krashen (1985: 2)
adds that physical context also activates knowledge in memory including
world knowledge and existing language knowledge.
12.1.1.3 Conceptual Schema
A contextual approach facilitates the construction of conceptual

schemata. A learner can use a context to construct a new schema, which
12 Principles 199
can then be used as a framework to process and produce the language

used to articulate it. Increasingly, usage-based linguists take the view
that contextual information is stored alongside linguistic information in
the constructicon. In Language, Usage and Cognition (2010: 14), Bybee
introduces the concept of rich memory representations. A rich memory
representation contains not only language knowledge but also knowledge
of the social and physical contexts in which the language was experi-
enced. According to Schmid (2017: 21), these properties allow speakers
to vary the formality of their language in different situations. Thus, expo-
sure to language in context provides knowledge of important properties
of words and phrases which supports their use.
12.1.1.4 Language Schema
While schema theories were originally created to account for conceptual

knowledge, the term was extended in the late 1970s to refer to discourse
patterns such as stories (Mandler, 1978; Rumelhart, 1975; Thorndyke,
1977). Notwithstanding a number of criticisms (e.g., Alba & Hasher,
1983; Kintsch, 1988; Nassaji, 2002), the term has been further extended
and has come into common parlance in the conversations of linguists and
language educators to refer to knowledge of lexico-grammatical patterns.
In the general language teaching community, the importance of
learning common lexico-grammatical patterns has been understood for
some time. In the early 1980s, structural approaches came under criti-
cism for only teaching syntactic relations (Yalden, 1983: 28). While lexis
now has greater importance, recommendations to teach decontextual-
ized vocabulary persist (e.g., Nation, 2001: 2). However, a problem with
decontextualized word learning is that it does not reveal the relation-
ships between words. Hulstijn (2001: 285) points out that, in addition
to morphology, orthography, and semantics, context supplies a word’s
prosodic and pragmatic features, and demonstrates interword relations.
With regard to interword relations, Lewis (1993: 103) notes that context
refers not only to the situations in which words occur, but more impor-
tantly, to the co-text with which they routinely co-occur. He (1997:
30) asserts that this co-occurrence (collocation) is a powerful means of
200 I. Pemberton
learning vocabulary but that many learners will need a teacher’s help in
identifying co-text.
12.1.1.5 Frequency
Another key feature of context is that it provides information about

frequency. In the usage-based paradigm, frequency is the major external
vector that accounts for the entrenchment of constructions in learner
cognition. On the one hand, repeat experience of language tokens
strengthens the representations of concrete lexical items in memory. On
the other hand, repeat experience of language types, together with the
general learning abilities of categorization, analogy, and abstraction leads
to the formation of language schema with fixed and flexible items, and
at a higher level, fully abstract schema. Ellis (2008: 103) points out
that neither grammar instruction nor dictionaries supply learners with
information about the frequency of occurrence of individual items—this
can only be gained from experience of language in use. According to
Barclay and Schmitt (2019: 807), deeper knowledge of a word such as
aspects like collocation and derivation may be best learnt incidentally via
extensive exposure. They maintain that this allows learners to develop
intuitions about the pragmatic use of language.
12.1.1.6 Connotation
A context also reveals the connotations of a word. Sinclair (1991: 112)

reports that corpus linguistics reveals many words and phrases occur
in specific environments. As an example, ‘happen’ tends to be used to
describe negative experiences like accidents. In addition, Sinclair points
out that corpus linguistic research shows that words and phrases influ-
ence a speaker’s choice of grammar, for example, ‘the phrasal verb ‘set
about,’ in its meaning of something like inaugurate is closely associated
with a following verb in the –ing form, for example, set about leaving.’
The task facing learners is an immense one. Ellis (2019: 48) notes the
scale of the task, pointing out that language learning is a highly complex
12 Principles 201
activity which involves not only learning ‘tens of thousands of construc-

tions (words, morphemes, lexico-grammatical-functional patterns, etc.)
and of the probabilistic relations between them but also their functions,
their speakers, their contexts, and their genres.’ Similarly, Barclay and
Schmitt (2019: 817) comment that taking into account ‘the number
of word families needed to engage with ungraded authentic texts, the
ubiquity of formulaic language needed for natural production and fluent
decoding, and the various aspects of the vocabulary knowledge construct
needed for appropriate use, learners face a difficult challenge.’
In summary, context is highly facilitating for understanding and
learning new language. It is hard evidence of how language is used for
communication. When authentic texts are used, context provides high
quality data and paralinguistic clues that help learners to understand the
meaning of language and learn its form. Context leads to the creation of
semantic schemata which help listeners to understand language and act
as a compensatory device to understand meaning when other cues are
unreliable, and it exposes learners to a wide range of closely related infor-
mation that is necessary both for language processing and language use.
This includes morphology, prosody, collocations, derivations, semantics,
connotations, functions, pragmatics, probabilistic relations, speakers,
situations of use and genres. Given the amount of language and the level
of information learners need to learn, it is safe to say that in the absence
of context, language learning becomes challenging, to say the least.
12.1.2 The Input Principle
The second principle of a contextual approach is that, in contrast to

communicative approaches, language learning activities should have a
strong focus on input. According to Krashen and Terrell (1983: 56), ‘we
acquire from what we hear (or read) and understand, not from what we
say.’ Ellidokuzoglu (1996) citing Schwartz (1993) concurs, noting that
without exposure to instances (exemplars) of the target language, growth
cannot take place. Similarly, Rost (1994) observes that learning can’t even
begin in the absence of the right level of input (cited in Nunan, 1997).
202 I. Pemberton
As learning a language is a non-linear process, input is a necessary

condition for acquiring language patterns. In the Natural Approach,
Krashen and Terrell (1983: 32) assert that for learners to move up to the
next level, the input they hear has to contain a structure which is a part
of that level. This challenged the orthodox structural view of language
learning which regarded learning as a process where grammar points
could be taught in a set order, with learners progressing incrementally
in a linear manner to the next stage. However, a criticism of the Natural
Approach was made by Howatt (1984: 286) who pointed out that it only
dealt with the acquisition of grammar.
The importance of input is recognized in its codification in eminent
linguists’ taxonomies of language learning principles. In Learning Vocabu-
lary in Another Language, Nation (2001: 2) posits that the first strand of a
well-designed language course is comprehensible meaning-focused input.
Similarly, principle six of Ellis’s (2005) Principles of instructed language
learning, is ‘successful instructed language learning requires extensive
L2 input.’ Likewise, in his Principled Communicative Approach, Dornyei
(2009: 40) declares that ‘PCA should offer learners extensive exposure to
large amounts of L2 input that can feed the learners’ implicit learning
mechanisms.’
Usage-based linguists are also firmly in agreement about the crit-
ical importance of input. Verspoor and Nguyen (2015: 314) assert that
‘the first characteristic of a [Dynamic Usage-Based] approach to second
language teaching is the vital focus on input.’ Kassenberg et al., (2020:
273) point out that ‘this is considered essential because, at its core, usage-
based theories hold that the acquisition of language is exemplar-based,’
adding that input should be contextualized, meaningful, frequent, and
repeated. Verspoor and Nguyen (2015: 307) emphasize exposure to
authentic input which provides concrete examples of language use
including their prosody and situational meaning. In addition, they (ibid.:
315) cite Van Patten and Benati (2010) who maintain that while it is
frequently assumed that output helps language learning, no research has
shown that output is a necessity.
12 Principles 203
12.1.3 The Modality Principle
The third principle of a contextual approach is that use of multimedia

resources leads to deeper learning. This valorizes watching and listening.
12.1.3.1 Audio Materials
Speaking activities have come to dominate classroom practices. A well-

known comment by Nunan (1997) is that ‘listening is the Cinderella
skill in second language learning. All too often, it has been overlooked
by its elder sister: speaking.’ Field (2008: 2) maintains that during
the second half of the twentieth century, in line with the commu-
nicative paradigm, teachers involved their learners in a great deal of
speaking practice. He describes the thinking behind this as ‘muddled,’
commenting that communication is not only about speaking but equally
about understanding messages. In the speaking activities of a commu-
nicative classroom, learners provide input to one another. While these
activities may give them a reasonable command of the target language,
Howatt (1984: 264) maintains that exchanges will be based on well-
formed standard sentences whereas what he terms ‘natural oracy’ or
interaction between members of the target speech community is quite
different. According to Howatt (ibid.: 265), natural oracy is difficult to
achieve even with a native-speaking teacher as this provides a ratio of
native to non-native speakers of one-to-many in a classroom.
There is a strong case for more listening activities in the classroom.
Nunan (1997) citing Rost (1994) remarks that listening is a vital skill
because it provides input to the learner and without input learning
cannot take place. After pointing out that first language learning takes
place exclusively through listening, Lewis (1993: 8) suggests that the
early stages of second language learning programmes should similarly
exploit receptive skills, and particularly listening. However, Field (2008:
1) points out that listening, which is regarded as a passive skill that takes
place in a learner’s cognition, is difficult to teach. As a result, many
language teachers naturally gravitate towards the use of reading materials.
204 I. Pemberton
12.1.3.2 Multimedia Materials
In addition to calls more listening in the classroom, more recently, a

growing number of linguists are arguing for the use of multimedia
resources. Hall (2018: 36) remarks that technology use for language
education is not a new phenomenon. For example, language labora-
tories came into use in the 1950s. In the 1970s, Wilkins (1976: 80)
noted the growing use of tape recorders and authentic listening texts
but commented that these materials lacked images to contextualize the
language and provide extra-linguistic features to help learners understand
meaning. He predicted that ‘as television and video-cassettes become a
more established part of the resources available for language learning, we
can expect that semantically oriented courses will introduce learners to
authentic materials from the beginning and that this contact will not be
deferred until the learner has supposedly mastered all the forms that he
or she is likely to hear.’
Writing in the new millennium, Field (2007: 36) comments on the
huge changes that have taken place in the world including the prolif-
eration of English language TV, DVD, video and Internet resources,
and the concomitant spread of the English language for travel and
work purposes. Similarly, Chapelle (2009: 750) notes that ‘technology
dramatically extends and changes the breadth and depth of exposure
that learners can have with the target language.’ Noting the abundance
of resources, Barclay and Schmitt (2019: 811) suggest the use of televi-
sion, movies, computer games, and online content to provide exposure
for vocabulary learning. Ellis (2019: 52) speculates that ‘digital technolo-
gies give increasing opportunities for rich ‘rewilding’ of education.’ Also
using the metaphor of ecology, Larsen-Freeman (2020: 296) reports on
the growing trend towards an ‘ecological person-centred approach.’ In
this kind of approach, digital resources are matched to learners’ personal
interests.
While digital resources are freely available these days, there is a more
compelling reason for using them. This is that they draw upon learners’
primary biological skills—the evolutionary endowment which allows all
cognitively normal individuals to master their first language. A common
observation language educators make is that you can’t teach a second
12 Principles 205
language like a first one. This is certainly true. But it misses the point. A
first language is not taught, rather, it is learnt. And the good news is that
psycholinguistic theory is reporting what the reformists and other more
enlightened educators have known all along—that second and additional
languages are learnt like first ones.
Multimedia materials create ideational and linguistic contexts for
learners which activate their primary biological skills. They provide expo-
sure to concrete examples of language which can be schematized to
create linguistic prototypes. This is a very exciting idea but even today,
despite their availability, most educators and learners don’t use multi-
media resources extensively in the classroom. And even when they are
used, they are not necessarily used in ways that will lead to language
acquisition. Thus, a second set of principles is needed that relate to how
to use multimedia materials.
12.1.4 The Autonomy Principle
The fourth principle of a contextual approach is that autonomy promotes

deeper learning. Unlike Communicative Language Teaching where there
is a strong bias towards interaction in pairs and groups, this principle
implies that learners should work mostly individually and independently,
even when they are in the classroom.
Academics talk a lot about the importance of autonomy. Allwright
and Hanks (2009: 2) assert that while teachers control the classroom,
they can’t learn the language for the learners. Only the learners can do
that for themselves. They make the point that learners have veto power
over whether to learn or not. Larsen-Freeman (2012: 83) agrees stating
that ‘teachers do not control their students’ learning. Teaching does not
cause learning; learners make their own paths.’
There are conflicting views on the meaning of autonomy. Field (2007:
30) citing Holec (1981) defines autonomy as ‘the ability to take charge of
one’s own learning.’ In other words, autonomy is freedom from control
and direction by others. While this might seem like a generally accept-
able definition, some linguists view the construct as the autonomy of
206 I. Pemberton
the learners from the teacher. This definition places learners in self-
determining groups rather than individuals deciding what they will do.
A contextual approach interprets autonomy in a stricter sense—not the
autonomy of the group but personal autonomy. That is to say, learners
work by themselves, not all the time, but enough to develop a language
base that they can subsequently draw upon in their interactions with
teachers and other learners and depending on their context, perhaps
friends, family, or members of the public.
Autonomy can relate to behaviours inside or outside the classroom.
Modern day classrooms tend to be dominated by speaking activities.
However, a contextual approach favours learner autonomy in the class-
room based on the observation that languages are, for the most part,
not learnt while speaking to others. They are learnt while listening and
mapping acoustic data onto features of physical and cognitive environ-
ments. Working autonomously crucially places a learner’s emphasis on
processing not production.
It is sometimes suggested that learners provide input to each other,
and this is an important element of Krashen’s Natural Approach and
Communicative Language Teaching but there are questions of whether
less proficient learners can provide language input to each other of a suffi-
ciently high quality to promote acquisition, and whether learners can
process input for form while they are attending to meaning.
Autonomy can also relate to behaviours outside the classroom.
Allwright and Hanks (2009: 2) make the point that they want students
to learn how to learn and continue to do so after their educational
programmes finish. Field (2008: 4) agrees, stating that a central goal of
language teaching should be to prepare learners for life outside of the
classroom. This recognizes the limitations of classroom teaching and that
for many learners, most of the language learning is going to occur in the
outside world through the use of reading and listening materials, and
where learners are living in the target language community, spoken inter-
actions. Training learners to learn autonomously turns them into lifelong
learners who can continue to study once they have graduated from their
educational programmes.
12 Principles 207
12.2 Behavioural Principles

12.2.1 The Iteration Principle
How does it benefit learners to work autonomously? This leads to the

fifth principle of a contextual approach, the Iteration Principle. The Iter-
ation Principle states that repetition of a task promotes deeper learning.
To begin with, the writings of linguist John Field are used to illustrate the
importance of iteration. One of Field’s important contributions to class-
room pedagogy has been his analysis and critique of the comprehension
approach to the teaching of listening skills.
12.2.1.1 The Comprehension Approach
In his ELT Journal article, Skills and strategies: towards a new methodology
for listening, Field (1998: 110) contrasts classical and modern approaches
to the teaching of listening. He begins by setting out a typical format for
listening lessons in the 1960s and 1970s which he calls ‘the compre-
hension approach.’ A typical lesson would proceed according to the
following stages:
1. pre-teaching of new vocabulary

2. extensive listening → questions about general context
3. intensive listening → detailed questions
4. examination of vocabulary and/or exponents of grammar
5. use of play and repeat/play and predict/recall words
He contrasts this with a modern listening lesson format as follows:
1. pre-listening (for context and motivation)

2. extensive listening → questions to establish the situation
3. pre-set questions or pre-set task
4. extensive listening
5. review of questions or task
6. inferring new vocabulary/examination of functional language
208 I. Pemberton
Field (1998: 111) makes the point that the lesson format has been
modified, whereas it actually needs rethinking completely.
12.2.1.2 The Comprehension Approach: Problems
While the comprehension approach gives students practice at listening,

it does very little to improve their listening abilities. Field (2008: 38)
explains that the classical comprehension approach manifests several
problems. First of all, the activity is teacher-centric. The teacher plays
the CD, determines the number of replays, and decides how much time
to spend on each part of the recording. At times, the teacher may spend
too long on one section that has been understood well but not enough
time on another section that causes problems for some learners. Another
serious issue is that listening is a real time activity. A learner has to keep
pace with the recording irrespective of whether they have understood
it. Unlike reading, learners cannot check unknown words. Similarly,
learners cannot re-read parts of the text to resolve ambiguities.
12.2.1.3 The Comprehension Approach: Solutions
Field (2008: 45) suggests giving learners control over the pace of the
recording. This would:
• Make learners responsible for their own learning

• Ensure that they have enough time to complete the task without
having to work at the same pace as the group
• Allow them to play and replay parts of the recording that they
personally have difficulty with
• Give learners more time to analyse and extract miscellaneous informa-
tion from a text
• Accustom their ears to ‘the sounds, rhythms and intonation patterns
of the target language’ (Field, 2008: 48).
12 Principles 209
From a usage-based perspective, iteration is a key tenet which is

considered to promote deeper learning. When learners control the mate-
rials, they are able to review those parts of an audio or video recording
that they experience difficulty with. Verspoor and Nguyen (2015: 309)
contend that iteration gives rise to complex patterns. In relation to this,
Larsen-Freeman (2012: 83) comments that iteration is not simply about
revisiting the same material over and over but about each revisit to the
same place having a different initial state. Verspoor and Nguyen (2015:
309) explain this in terms of a movie scene where learners first get
the gist of a scene, then identify some of the expressions in it before
finally gaining a deeper understanding of some of those expressions by
considering their relationship to the context.
12.2.2 The Attention Principle
12.2.2.1 Incidental Learning
For many years incidental learning has been thought to be an impor-

tant part of language acquisition. In contrast to deliberate word learning,
incidental learning is the acquisition of vocabulary while engaged in
other activities such as listening, reading, or speaking. Hulstijn (2001:
266) reports it is customary in the L1 and L2 literature to state that
most vocabulary is learnt incidentally. In fact, Nation (2001: 232) asserts
that ‘incidental learning via guessing from context is the most impor-
tant source of all vocabulary learning.’ However, research casts doubt on
the effectiveness of incidental learning for vocabulary growth. Schmitt
(2008: 348) reports on reading and listening research which shows that
the rate of incidental vocabulary learning is low and even when learners
do learn new words, they have trouble using them productively. Verspoor
and Nguyen (2015: 313) citing Nagy (1997) report that the chance of
remembering a word from a single reading is no more than five to four-
teen per cent. Hulstijn (2001: 285) concludes that incidental exposure
to new words is not a necessary nor sufficient condition for vocabulary
growth.
210 I. Pemberton
Incidental learning is difficult because it requires learners to do two

things at the same time. The concept of multi-tasking has been around
for a long time, but recently, the psychology literature has cast doubt
on its efficacy. While people are able to do different tasks at the same
time, in particular when they are in different modalities (e.g., watching
while listening) this tends to impact negatively on both (or all) of the
tasks they are attempting to do either in terms of time taken or quality
of completion. During incidental learning, language learners are required
to do two things simultaneously. Skehan (1998: 24) cites research by Van
Patten (1990) which shows that it is hard for learners to pay attention to
both semantic and syntactic features during language processing. Schmitt
(2008: 342) states that when learners understand the gist of something,
they don’t pay attention to individual words. One reason for this lack
of attention to form is limited attentional capacity. Schmidt (2001: 13)
explains that in most situations, meaning is most important to a listener.
Given that attentional capacity is limited, listeners will concentrate on
meaning, focusing on lexical items first, and only then, if capacity allows,
paying attention to form. Thus, contrary to popular belief, picking up a
language incidentally is an insufficient condition for the bulk of acqui-
sition as learners have difficulty attending to meaning and form at the
same time. This raises the question of how learners can focus on form.
The answer is that the key to learning is attention. Schmidt (2012: 35)
remarks that input must be noticed to become intake. In other words,
learners learn what they attend to but not what they don’t. The impli-
cation is that a learner must intentionally pay attention to the form of
an utterance. Field (2007: 34) citing Schmidt and Frota (1986) remarks
that language learning depends on focusing learners on the words that an
L1 speaker uses. Similarly, Skehan (1998: 40) citing Ellis (1994) points
out that a learner must attend to linguistic form.
However, Skehan (ibid.) notes that asking learners to attend to form
by teaching them the rules of grammar has not been very successful.
Likewise, it seems that asking learners to attend to form during speaking
activities is similarly ineffective. Ellidokuzoglu (1996: 5) observes that
the CLT practice of having learners focus simultaneously on both
meaning and form while producing language doesn’t work either because
working memory limitations mean that they can only do one or the
12 Principles 211
other. To summarize, on the one hand, the solution seems obvious—

pay attention to form, but on the other hand, this is not as straight
forward as it appears. To understand how to operationalize the construct
of attention for pedagogical purposes, a basic understanding of memory
is helpful.
12.2.2.2 Working Memory
A well-known model of human memory is the multi-component model

which was proposed in the 1970s by Baddeley and Hitch (Baddeley,
2015: 18). This consists of three components. It makes a distinction
between short-term and long-term memory and further divides short-
term memory into a component for auditory input and a component
for visual input. These are referred to as the phonological loop and
the visuospatial sketchpad respectively. A fourth component, known
as an episodic buffer, was added more recently (Baddeley, 2015: 20).
This combines information from the phonological and visual short-
term memories and long-term memory. While there are other models of
memory, they possess similarities, and most cognitive psychologists can
broadly agree on the concepts of short and long-term memory (Wen,
2015: 44).
Some linguists use the term working memory to refer to short-term
memory, but they are essentially describing the same thing (Randall,
2007: 15). Short-term memory refers to the temporary nature of
the memory store whereas working memory refers to its function of
combining different inputs during processing and production.
An important point about working memory is that it has a limited
capacity. It can only retain about two seconds of input before new input
flows into the mind to replace it. This has also been estimated in terms of
five words plus or minus two, or approximately fifteen words if the input
is perceived in chunks (Baddeley, 2015: 20). Wen (2015: 43) remarks
that more recent estimates put the number of elements at four plus or
minus one.
Another point about phonological working memory is that it has two
main features. One is a phonological store which holds language, but
212 I. Pemberton
it decays rapidly. Schmidt (2001: 16) notes that stimuli which a person
doesn’t attend to will only persist for a few seconds. The second feature is
a process of articulatory rehearsal. Articulatory rehearsal means repeating
items (e.g., a phone number) using inner speech. The process of articu-
latory rehearsal keeps items in memory for longer periods of time (Ellis,
2001: 33).
Articulatory rehearsal in phonological working memory is an impor-
tant driver of both first and second language acquisition. To begin with,
children learning their L1 use rehearsal to internalize new language. Wen
(2015: 47) reports that cognitive psychologists in Europe have found
rehearsal facilitates the acquisition of vocabulary and the development of
a learner’s first language. Furthermore, an increasing number of linguists
view working memory as playing an equal or greater role in the acquisi-
tion of an additional language. Schmidt (2001: 10) cites Baddeley et al.
(1988) who contend that rehearsal in short-term memory is a neces-
sary condition for learning new language. According to Randall (2007:
161), rehearsal in working memory is not only a necessary condition for
learning but actually leads to fluency. Hulstijn (2001: 261) reports on the
results of a number of studies which have shown that briefly repeating
the target language in working memory leads to better retention. These
studies indicate that repetition can be either vocal or subvocal. More-
over, Hulstijn (ibid.: 66) cites research by Ellis and Sinclair (1996)
which found that when second language learners listen to and use the
target language, they learn about the frequency of its phonemes and
their probable sequences, and repetition consolidates words in long-term
memory.
In summary, while the rate of incidental learning is insufficient, and
either asking learners to focus on form by teaching them grammar or
during productive speaking exercises isn’t effective, research is increas-
ingly showing that attending to linguistic form and rehearsing it in
working memory provides better retention. Thus, the sixth and final
principle of a contextual approach is that attention to form is a necessary
condition for language learning.
12 Principles 213
12.2.2.3 Shadowing
Clearly, repetition of target language in transient short-term memory

helps to create stable representations in long-term memory. The ques-
tion is how to realize this in classroom practice. One way to do this is to
draw upon an existing teaching technique known as shadowing. Shad-
owing originates from the field of interpreter training. Murphey (2001:
146) defines shadowing simply as ‘the repetition of an utterance by a
listener.’ Hamada (2016: 36) citing Lambert (1988) defines shadowing
as ‘word-for word repetition in the same language, parrot-style.’ This may
give the impression that shadowing is a simple activity with echoes of the
rote repetition of Audiolingualism, but it is a far more complex process.
A pioneer of shadowing, Tamai (1997), defines it as ‘an active and highly
cognitive activity in which learners track the speech that they hear and
vocalize it as clearly as possible’ (in Hamada, 2016: 36). To summarize,
shadowing is a pedagogical practice which involves mindful repetition of
the target language.
Shadowing has a number of pedagogical benefits. Schaefer (2015:
174) points out that ‘it forces the listener to notice more carefully what
the speaker is saying.’ Murphey (2001: 131) cites Kurz (1992) who
explains that shadowing is an exercise which enhances ‘listening and
short-term memory skills.’ Hamada (2016: 36) notes that it is particu-
larly useful to teach prosody to learners whose L1 differs markedly from
the L2. It can improve listening, intonation, and accent (Tanaka, 2002)
and increase pitch range (Yamane, 2004) (both cited in Wiltshier, 2007:
499).
Shadowing as a language learning technique can be practiced in a
number of different ways depending on learner level. Murphey (2001:
129) sets out the scope of shadowing practices in terms of three
continuum: complete to selective, silent to out loud, and non-interactive
to interactive.
1. Complete to selective: a listener can repeat everything that they hear,

or they can repeat only certain phrases.
2. Silent to out loud: at one end of the spectrum, a listener can listen
only, or at the other end, they can vocalize everything that they hear.
214 I. Pemberton
3. Non-interactive to interactive: At one end, a listener can repeat faith-

fully what they hear, or at the other they can repeat and ask questions
of the speaker in response to what they hear.
While drawing upon existing practices, shadowing in a contextual

approach differs in some ways from pedagogical suggestions in the
shadowing literature. In the literature, shadowing can involve simply
repeating language, even if it is unknown, and without stopping to check
meaning. On other occasions, given the conundrum of one L1 speaking
teacher to many non-native speakers, there are suggestions that learners
should shadow each other. As pedagogical practices, both activities are
vulnerable to criticism. While listening and repeating language where
the meaning is not clear may improve aspects of prosody, it is unclear
whether it can contribute to lexical growth. In the second scenario, where
learners shadow each other, there is a question of how much reciprocal
learner inputs can contribute to target language proficiency.
Operationalizing a contextual approach involves the use of devices
such as tablets, laptops, and personal computers to be able to
autonomously view multimedia resources. This allows learners to control
the pace of the materials, and to watch recursively, focus on what they
have difficulty with and use any cultural tools at their disposal—paper,
electronic or online dictionaries, or target language or first language
subtitles—to solve any problems that they encounter.
12.3 Summary
This chapter introduced six principles for realising a contextual approach
inside and outside the classroom. The first four principles strive to
create the optimum conditions for language learning to take place.
The first principle of a contextual approach is that context is vital for
learning. Notwithstanding its contribution to language learning, histor-
ical teaching methods have tended to underestimate its importance.
Theorists and educators of all stripes are agreed that contexts should be
authentic. This is because they provide the building blocks for learner
interlanguage. The facilitating features of context are considered key
12 Principles 215
for understanding, remembering, and producing language. First of all,

context provides the learner with paralinguistic cues which promote a
deeper understanding of language. Secondly, context allows learners to
develop schematic knowledge. Thirdly, context illustrates the relation-
ships between language items. Context demonstrates the connections
between lexical items and also helps learners to learn suprasentential rela-
tions. Fourthly, context provides information about how frequently items
occur. Fifthly, context can reveal the connotations of words. Taking into
account all the possible relations that can exist between concrete items of
language, the task facing learners is an immense one but the presentation
of language in context pulls all of the necessary information together. A
context acts as a glue to bind the semantic, syntagmatic, prosodic, and
pragmatic features of constructions together.
The second principle of a contextual approach is that learning
materials should have a strong input bias. While input is almost
universally regarded as important for language learning and is codi-
fied in linguists’ taxonomies of language learning principles, the current
paradigm valorizes output activities. Conversely, taking a contextual
approach recognizes that language learning is a non-linear complex
process, and that a substantial amount of input is necessary to promote
a phase shift in learner interlanguage.
The third principle is the Modality Principle. This requires that
learners use multimedia resources. While speaking activities have come
to dominate classroom activities, a strong case has been made in recent
years for more listening activities. However, given the wide diffusion of
TV, DVDs and videos, and the rapid evolution of the Internet, the case is
made not for use of listening materials but multimedia resources as these
allow for parallel processing of both auditory and visual input. There is
an even more compelling reason for the use of multimedia resources and
that is that they activate the primary biological skills of listening and
intention reading which all cognitively normal individuals use to learn
their first language.
The fourth principle is that learners should work (mainly)
autonomously even when they are in the classroom. Academics make
a lot of autonomy, but it often comes to be defined as freedom from
the teacher and the autonomy of the group, rather than the autonomy
216 I. Pemberton
of the individual. A contextual approach goes further in requiring that

learners work mostly individually. This recognizes the scale of the lexical
learning problem and the importance of context and input in helping
learners to acquire language. While a contextual approach doesn’t imply
that learning can’t take place during productive activities, it regards them
as ineffective when a teacher uses them as a main pedagogic strategy.
A further perspective on teaching learners how to acquire language by
working autonomously is that they can continue to study outside the
classroom. Even where they do not reside in a target language commu-
nity, they can study in a way that affords them to access to L1 user norms
of proficiency.
The final two principles relate to learner behaviours. The fifth prin-
ciple is the principle of iteration. To illustrate, the teaching of listening
has traditionally drawn upon the comprehension approach. This scenario
will be familiar to many language teachers. It involves the teacher
controlling the auditory input, deciding how many times to play a
recording and where to stop it before nominating learners to answer
questions to test their comprehension. The problem with the compre-
hension approach is that while it provides correct answers, it doesn’t
reveal what problems individual learners have or help them to solve
their problems. If learners have access to an individual copy of a
recording, they can control the pace, stopping and replaying areas where
they personally have difficulty in understanding. This will address their
specific learning problems. While the comprehension approach is asso-
ciated with listening skills, the principle of iteration applies equally to
watching and listening during the use of multimedia resources.
The sixth and final principle is the Attention Principle. There is a
common belief that second languages are learnt incidentally, especially
where the learner is in the target language community. However, there
are many counterexamples of individuals who have failed to achieve
proficiency or who are proficient but make persistent errors during
production. Research also casts doubt on the efficacy of incidental
learning for language acquisition. While the solution to this problem
is quite obvious—pay attention to language form—teaching grammar
and involving learners in speaking activities have both failed to effec-
tively solve the problem. With regard to improving accuracy through
12 Principles 217
speaking activities, listeners have difficulty focusing on both meaning

and form. The issue lies in the structure of memory. A person has to
use their short-term memory both to process input and to focus on
form. It has a highly limited capacity and can’t do both at the same
time. If the key to language learning is to pay attention to linguistic
form, the question arises how this can be done. Working memory has
two components, a phonological store for storing language briefly and
an articulatory rehearsal process. Articulatory rehearsal has been shown
to be an important aspect of both first and second language learning.
A pedagogical practice known as shadowing can be adapted to push
input from the working memory into long-term memory. This involves
listening intently and repeating exactly what a speaker says. Unlike some
pedagogical suggestions—to repeat spoken text when the meaning is
not understood and for learners to shadow each other—in a contextual
approach, learners working autonomously are free to stop and confirm
meaning, and they only shadow higher quality recordings of expert
language users.
References
Alba, J. W., & Hasher, L. (1983). Is memory schematic? Psychological Bulletin,
93, 203–231.
Baddeley, A. (2015). Working memory in second language learning. In Z. Wen,
M. Borges-Mota, & A. McNeill (Eds.), Working memory in second language
acquisition and processing (pp. 17–28). Multilingual Matters.
Barclay, S., & Schmitt, N. (2019). Current perspectives on vocabulary teaching
and learning. In X. Gao (Ed.), Second handbook of English language teaching
(pp. 799–819). Springer International Publishing.
Chapelle, C. A. (2009). The relationship between second language acquisi-
tion theory and computer-assisted language learning. The Modern language
journal (Boulder, Colo.), 93(s1), 741–753.
218 I. Pemberton
Dornyei, Z. (2010). The 2010s communicative language teaching in the 21st

century: The ‘principled communicative approach.’ Perspectives, 36 (2), 33–
43.
International ELT Conference: From diversity to synergy. Ankara, Turkey.
Ellis, N. C. (2001). Memory for language. In P. Robinson (Ed.), Cognition and
second language instruction (pp. 33–68). Cambridge University Press.
Ellis, N. C. (2008). Usage-based and form-focused SLA: The implicit and
explicit learning of constructions. In N. C. Ellis & P. Robinson (Eds.),
Handbook of cognitive linguistics and second language acquisition. Routledge.
Ellis, N. C. (2019). Essentials of a theory of language cognition. The Modern
Language Journal, 103(S1), 39–60.
Ellis, R. (2005). Principles of instructed language learning. System, 33(2), 209–
224.
Field, J. (1998). Skills and strategies: Towards a new methodology for listening.
ELT Journal, 52(2), 110–118.
Field, J. (2007). Looking outwards, not inwards. ELT Journal, 61(1), 30–38.
Field, J. (2008). Listening in the language classroom. Cambridge University Press.
Finkbeiner, R., Meibauer, J., & Schumacher, P. B. (2012). What is a context?
Linguistic approaches and challenges. John Benjamins Publishing Company.
ed.). Routledge/Taylor & Francis.
Hamada, Y. (2016). Shadowing: Who benefits and how? Uncovering a
booming EFL teaching technique for listening comprehension. Language
Teaching Research, 20 (1), 35–52.
Oxford University Press.
University Press.
Hulstijn, J. (2001). Intentional and incidental second-language vocabulary
learning: A reappraisal of elaboration, rehearsal and automaticity. In P. J.
Robinson (Ed.), Cognition and second language instruction (pp. 258–286).
Kassenberg, T., Galati, F., de Vries-Zhuravleva, D., & Menke-Bazhutkina,
I. (2020). 12. Film language integrated learning: A usage-inspired L2
teaching approach. In W. Lowie, M. Michel, M. Keijzer, & R. Steinkrauss
(Eds.), Usage-based dynamics in second language development (pp. 271–294).
Multilingual Matters.
12 Principles 219
Kintsch, W. (1988). The use of knowledge in discourse processing: A

construction-integration model. Psychological Review, 95, 163–182.
Larsen-Freeman, D. (2012) Complexity theory. In S. M. Gass & A. Mackey
Francis.
Larsen-Freeman, D. (2020). Epilogue. In W. Lowie, M. Michel, M. Keijzer, &
R. Steinkrauss (Eds.), Usage-based dynamics in second language development.
Blue Ridge Summit: Multilingual Matters.
Lewis, M. (1997). Implementing the Lexical approach. Language Teaching
Publications.
Mandler, J. M. (1978). A code in the node: The use of a story schema in
retrieval. Discourse Processes, 1, 14–35.
Murphey, T. (2001). Exploring conversational shadowing. Language Teaching
Research, 5 (2), 128–155.
Nassaji, H. (2002). Schema theory and knowledge-based processes in second
language reading comprehension: A need for alternative perspectives.
Language Learning, 52(2), 439–481.
University Press.
Nunan, D. (1997). Listening In language learning. The Language Teacher,
21(9).
Rumelhart, D. E. (1975). Notes on a schema for stories. In D. Bobrow & A.
Collins (Eds.), Representations and understanding: Studies in cognitive science
(pp. 211–236). Academic Press.
Schaefer, M. (2015). Conversational shadowing. New Directions in Teaching
and Learning English Discussion, 3, 173–181.
Schmid, H. J. (2017). A framework for understanding linguistic entrenchment
and its psychological foundations. In H. J. Schmid (Ed.), Entrenchment and
the psychology of language learning (pp. 9–36). De Gruyter Mouton.
220 I. Pemberton
Schmidt, R. W. (2001). Attention. In P. J. Robinson (Ed.), Cognition and second

language instruction. Cambridge University Press.
Schmidt, R. (2012). Attention, awareness, and individual differences in
language learning. In C. Wai Meng, C. Kwee Nyet, B. Sunil, & W. Izumi
(Eds.), Perspectives on individual characteristics and foreign language education
(pp. 27–50). De Gruyter Mouton.
Schmitt, N. (2008). Instructed second language vocabulary learning. Language
Teaching Research, 12(3), 329–363.
Sinclair, J. (1991). Corpus, concordance, collocation. Oxford University Press.
Press.
Thorndyke, P. W. (1977). Cognitive structures in comprehension and memory
of narrative discourse. Cognitive Psychology, 9, 77–110.
Wen, Z. (2015). 3. Working memory in second language acquisition and
processing: The phonological/executive model. In W. Zhisheng, M. Mailce
Borges, & M. Arthur (Eds.), Working memory in second language acquisition
and processing (pp. 41–62). Multilingual Matters.
Wiltshier, J. (2007). Fluency through shadowing—What, why, and how? In K.
Bradford-Watts (Ed.), JALT 2006 conference proceedings. JALT.
tation. Pergamon.
13
Materials Design
If someone were to assert that a language can be learnt better outside of

the country where it is spoken, I’m sure most people would disagree. But
let’s think about it in terms of the principles of a contextual approach.
Each of the six principles has an associated hypothesis. The main hypoth-
esis of the approach is that learning through context results in deeper
learning. In this case, a context refers to a rich ideational context together
with its associated language. Intuitively, I think most people would agree
with this statement.
The second hypothesis is that a substantial amount of input is neces-
sary for language learning. Once again, I’m sure most people would
agree. Pretty much anyone you ask, and it doesn’t matter if it is a language
teacher, a teacher trainer, a teacher manager, a classmate, a parent, a
sibling, or a person you pass in the street, will give you the same advice.
If you want to learn a language, watch movies in the target language,
listen to the radio, read the newspaper, and make friends with people
who speak the target language. Better still, go live in a country where
the language is spoken. When you are there, you are surrounded by
sights, sounds, and symbols which activate, extend, and reinforce your
knowledge of the language.

Switzerland AG 2024
222 I. Pemberton
However, there are many expatriates who live in a country for decades
who still fail to learn the target language to a degree of fluency. So, being
in a country where a language is spoken is not a sufficient condition for
learning the language. There are other important factors such as moti-
vation and effort which determine engagement with the target language,
and these are things that a learner can draw upon whether in the country
or not.
The third hypothesis is that use of multimedia resources promotes
deeper learning. This statement may divide opinion. However, it was
argued earlier that watching and listening are the primary biological skills
that children use to acquire their mother tongue, and that reading and
writing, the skills that are generally applied in second language learning
classrooms, are much less effective. And being in the country of the
target language provides a rich visual and auditory context within which
learners can employ their innate capacities to watch and listen, and to
map sound onto image.
But do we actually need to be there? We live in the twenty-first
century, an era when most everyone in higher resource contexts has access
to the Internet and can vicariously travel through cyberspace to pretty
much anywhere they wish. Many of us have or can get access to any
number of films, documentaries, dramas, and animations written, acted,
and narrated by expert users of the target language. Things have changed
a lot since 1625 when Francis Bacon wrote, ‘if the mountain won’t come
to Mohammad, then Mohammad must go to the mountain.’ It seems
that, on this occasion, should we wish to do so, we can bring the moun-
tain, whether it be the Matterhorn, Mount Fuji, or Mont Blanc, into our
homes, or for that matter, anywhere we happen to be.
But surely, it is still better to be in the country? The fourth hypothesis
is that autonomous learning results in deeper learning. This is a position
many will disagree with this. Any teacher schooled in Communicative
Language Teaching will have been exhorted by teacher trainers, teacher
managers, and even fellow teachers to use pairwork and groupwork activ-
ities on the basis that language is learnt through interaction with others.
They might have come to believe that this is true. Admittedly, while
learning autonomously, you can’t have conversations with people, you
can’t test hypotheses, and you can’t negotiate meaning with people.
But you can do better than that. The fifth hypothesis is that itera-
tion promotes language acquisition. This is more difficult to disagree
with. Put simply, watching a movie, documentary, drama, or animation,
learners can stop the programme when they don’t hear something, they
can take as long as they like, and they can replay it as many times as they
like. If they can’t catch something, they can use target language subtitles,
or if there are no target language subtitles, they may be able to use subti-
tles in their own language. Once a learner has identified the language,
they can check the meaning in a dictionary, or dictionaries, or by using
a free or paid version of an online AI translator. Now, learners can’t
interrupt a target language speaker every time they don’t hear, or don’t
understand something. If they did, the target language speaker would
quickly get frustrated with them.
The sixth and final hypothesis is that attention to form promotes
language acquisition. I think most people would agree with this state-
ment too. It was pointed out earlier that language learners have difficulty
simultaneously listening for meaning and attending to form due to
the limitations of working memory. As already mentioned, a learner
can’t keep stopping an L1 interlocutor while they focus on the form of
what they are saying. But with a video recording, learners can stop the
recording, repeat, and imitate speakers as much as they like.
To summarize, if someone asserted that a person could learn a
language better outside of a target language community than within it—
most people would disagree. But we live in an era where we have easy
access to multimedia materials which can provide rich cultural, situ-
ational, conceptual, and ideational contexts for a target language. So,
with motivation and effort, there is really no reason why you can’t learn
a language to a higher level of fluency even outside a target language
community. Of course, a learner can have the best of both worlds. That
is, live in the country where the language is spoken, but at the same time
make systematic and principled use of multimedia materials to attain
higher levels of proficiency. This is certainly what expatriates focused on
mastering a target language should do.
This chapter describes how to realize the principles of a contextual
approach, not only for the individual language learner, but also in the
design of classroom materials for learners (of an intermediate level and
224 I. Pemberton
above) who are focused on passing language examinations or wish to be

regarded as legitimate members of target language communities.
13.1 The Role of the Teacher

A question you might ask is if students are expected to learn
autonomously, does that make the approach purely constructivist? If so,
what role does the teacher play in a re-imagined twenty-first-century
classroom? The answer to the first question is, surprisingly, no. As an
educator, you can still work within the framework of Sociocultural
Theory. Vygotsky’s Zone of Proximal Development is a space where
a learner works with a more knowledgeable other who scaffolds the
learning so that they can achieve more than they otherwise would do by
themselves. As a teacher, you create the frame of learner focus, and scaf-
fold the learning. O’Keeffe (2021: 265) explains that the teacher chooses
the material for the learners, the target language they focus on, and the
tasks that they undertake.
13.2 Multimedia Materials

When working with learners, the first step will be to select mate-
rials for them to use. In a contextual approach, learners work mostly
autonomously, so it is important to select the right materials for them.
There are a number of criteria for selecting materials, and these are set
out below.
1. Relevance: To begin with, consider learner aims. For example, are

they preparing for a general language proficiency exam, or are they
preparing to study specific academic subjects in the target language?
This will determine what to focus on. Learners may benefit from
studying about a general topic such as climate change, or a specific
academic subject such as business or economics.
2. Interest: If students are attending a general language course and are

not preparing for a language proficiency exam or for future studies,
then the interests of the learners should be considered. These will
be influenced by student demographics such as age, gender, family
background, education, and social and cultural background. Addi-
tionally, is a video visually engaging? Are the images interesting?
Are there deictic elements? Is there the potential to map sounds
to images? Do scenes in the video disambiguate the meaning of
vocabulary?
3. Duration: The length of a video is a key criterion. Learners will work
intensively with the materials, and, if the materials are at an appro-
priate level of challenge, quite short videos of between three and five
minutes can suffice for an hour of class time.
4. Coherence: Are the materials easy to understand? Is the organization
clear? Are the main ideas easy to identify?
5. Clarity: Are the images sharp? Is the sound clear?
6. Learner control: Can learners navigate the video easily? Can they
pause it? Can they select a point to play it from? Are there fast
forward and rewind controls?
7. Access: Are the materials on an easily accessible platform? Are they
freely available or behind a paywall? Are there advertisements that
the students will have to watch?
8. Orthographic support: Are there target language subtitles? Are there
learner L1 subtitles? Can they be turned on and off? Is there a
written transcript?
9. Informational content: Is the macro-organization clear?
10. Lexico-grammatical content: Does the video contain an adequate
amount of useful lexico-grammatical content for the learners.
The first principle of a contextual approach is that context promotes

deeper language learning. One important affordance of learning language
through context is that it helps learners to develop schematic knowledge.
In other words, context helps learners to develop organized knowledge
of the conceptual and ideational content of written and spoken texts.
The informational content of a video helps learners internalize new
language forms. One way to think about how this works is to consider
226 I. Pemberton
improvements in Artificial Intelligence as applied to translation software.

In the not-too-distant past, employing literal word-for-word translation,
AI produced highly unusual translations that were markedly different
from L1 writing and speech. More recent iterations of AI translation
software are now able to produce much more accurate translations. This
improvement was due in part to the establishment of semantic fields that
group together related concepts. This helps translation software to make
the right lexical choices.
Similarly, using a topic as an ideational frame for teaching vocabulary
allows learners to make connections not only between words in a phrase
which expresses a specific concept but between phrases which are used
to articulate related concepts in the same sentence or other sentences. In
other words, the semantic content acts as a glue not only to bind together
the words in a phrase but also to connect related phrases in more distant
locations in the text.
An important part of materials selection is to focus on the content of
the video materials. The following semantic categories may be used to
identify suitable materials which provide high informational content.
Semantic Category Features

Argumentation an opinion for or against something and several
supporting arguments
Discussion (a) compares two or more phenomena
Discussion (b) benefits and drawbacks
Challenge counterargues an argument for or against
something
Factorial the reasons for something
Consequential the results of a particular phenomena
Solution solutions to a problem
Process the steps in a process
Narrative articulates a story
Working with semantic categories makes it possible to group mate-

rials for different lessons thematically. That is to say, after presenting
the ideas of proponents of something, learners work with the ideas of
opponents. Equally, learners can work with semantically related episodes
in a story. This kind of work helps learners to develop the ability to
talk about specific topics. For example, they can learn about the causes
of climate change, they can learn arguments for and against alternative
energies such as solar, wind, and wave power, or they can learn about the
steps in a process such as the greenhouse effect.
13.3 Ideational Content

After identifying the macro-organization, the next step is to determine
the most important ideas to make them transparent for learners. There
are several different cues which can be used to identify the important
informational content.
13.3.1 Title(s)
To begin with, use any titles to identify main ideas. These can be
converted to content questions to pinpoint the most important informa-
tional content. In some cases, this will be straightforward, as illustrated
by the following examples.
Semantic
Video Title Category Content Question
How playing sports benefits Factorial How does playing
your body and brain sports benefits your
body and brain?
Competitive sports: harmful or Discussion Are competitive sports
healthy? harmful? Why? Are
competitive sports
healthy? How?
Why languages die Factorial Why do languages die?
Five things we can do to Solution What are five things
control climate change we can do to control
climate change?
Make great coffee with a Process How do you make a
moka pot great coffee with a
moka pot?
228 I. Pemberton
13.3.2 Introductions
If titles do not point to the main ideas, it may be possible to pick

these up from the introduction to a video. For example, the main focus
may be apparent from a statement of purpose in the introduction. The
introduction might also provide further cues such as an outline of the
organization of the content.
13.3.3 Text Types
On other occasions, the title and the introduction might still not
provide enough information to identify the most important informa-
tional content, but it may be identifiable from the text type. While
the topic may be clear, the specific content may not be. An inspection
of the video transcript in terms of the semantic categories introduced
earlier, might reveal the text type, for example, a discussion (benefits and
drawbacks) structure.
13.3.4 Topic Sentences
Topic sentences are generally associated with written texts and a video is
a spoken text. However, many videos are scripted. They are not purely
spontaneous spoken interactions but are organized into spoken para-
graphs in much the same way as a piece of writing. Furthermore, these
paragraphs often have a conventional structure and include elements
such as topic sentences and supporting sentences. This raises a related
question of how to identify the breaks between different paragraphs.
There are a number of possible cues: a pause, a discourse marker, a
change of speaker, a change of location, or a return to a main presenter.
13.3.5 Logical Division of Ideas
Another technique to identify important informational content is to

thematically analyse the transcript. For example, a video with a title such
as ‘The future of language’ has a clear topical focus, but it is still broad,
and requires narrowing down. An analysis of the transcript might indi-
cate the logical division of ideas, for example, the script writer might
organize the text using different languages such as Chinese, French,
English, Spanish, and Arabic.
13.4 Constructions
A context facilitates the acquisition of conceptual and ideational knowl-
edge. On a language level, a context does more than disambiguate the
meanings of words. It also facilitates the acquisition of the language
necessary to articulate the concepts and ideas that the content expresses.
Furthermore, a context helps learners to connect concepts within the
same sentence, across sentences, paragraphs, and even whole texts. It
allows learners to develop knowledge not only of the semantic meaning
of words and their lexico-grammatical patterns, but also to begin to build
up an intuitive knowledge of the frequency with which specific words
and expressions occur when they are used to talk about particular topics,
subjects, and fields. This raises a question of how to choose vocabulary
from context for learners. A range of criteria can be applied to help do
this.
1. Number: How many constructions should be chosen from a text?

Students have a lot to learn. They will have to learn a large number
of constructions to lead to a phase shift and allow them to jump to a
higher proficiency curve. However, too many constructions will over-
burden them and probably not be realistic within the time allowed.
As a rule of thumb, an upper limit might be twenty constructions per
one-hour class.
2. Topic-specific language: Avoid the time-honoured practice of
presenting words simply because they are unknown. Constructions
don’t have to contain unknown words. Of more importance is that
any words chosen, whether known or unknown, are important to the
explanation of the topic. For example, if the topic is about the benefits
of exercise, the language chosen helps learners to talk about exercise
230 I. Pemberton
or to talk about benefits. While some of what learners learn will be

tied to a particular topic (in this case, exercise), a part of what they
learn (in this case benefits) will be transferable to other topics.
3. Content-specific language: Having identified topic specific language,
it may be necessary to narrow down the number of constructions
further. A way to do this is to preferentially select constructions that
form a part of the most important ideas in a text.
4. Completeness: Does the construction represent a complete idea? In
other words, is it a fragment or does it represent a complete noun-,
verb-, adverbial- or adjectival phrase? While we can present fragments
to learners, e.g., sentence starters, unitary bounded constructions that
represent specific ideas may be easier to remember.
5. Proximity: To further narrow down the choice of constructions, look
at how close particular words, such as subject-verb or verb-object
combinations, are to each other. Collocates can often be spatially
separated. Given the associative nature of word learning, it is more
likely that learners will remember words that are next to each other.
6. Transparency: Preferentially choose constructions which display cue
reliability. That is to say, the words in the construction are being
used with conventional meanings, and the meaning of the whole
construction is clear.
7. Surrender value: What is the value of the constructions to the learners.
Are they new, partially known, or known? Will they be able to use
them to articulate topics and ideas? Are they likely to use them in
written and spoken examinations, or in interactions with L1 users?
8. Exemplar value: Does the construction represent a common conven-
tional word order, e.g., does it display conventional noun phrase word
order, or verb phrase word order? Will a learner be able to compare it
with like items to form schematic representations?
9. Frequency: Is either the syntactic pattern or the lexical pattern
repeated elsewhere in the text? Are there similar patterns with system-
atic variations that a learner can use to create schematic representa-
tions to flexibly vary their spoken and written outputs?
13.5 Pre-Watching Activities

As discussed earlier, approaches such as Task-Based Instruction have a
clearly defined set of pre-task activities. These usually take place in a
shorter lesson segment prior to the main activity.
1. Activating background schemata

2. Eliciting relevant vocabulary from students
3. Listening to model conversations which provide further language for
the main task
While they might be considered ‘good practice’ elsewhere, it is

doubtful whether these activities are appropriate or even necessary in a
contextual approach. First of all, schemata already exist within the mind
of a learner, and they will be activated by sound and image as soon as
learners start to watch a video, so the value of pre-activation is question-
able. Secondly, the main purpose of a contextual approach is to introduce
unfamiliar lexis in context, so it doesn’t make much sense to elicit decon-
textualized vocabulary from learners. Thirdly, the limitations of teacher
controlled listening exercises have already been discussed at length, and
learners struggle to process both meaning and form, so it is doubtful how
useful listening to a model conversation will be. A task-based approach
simply sets students up to practice language immediately upon hearing
it, irrespective of whether it is known or only partially known. If in
fact, language is a complex, non-linear and emergent system, then ques-
tions might be asked about how useful the decontextualized activities
and emphasis on output of a task-based lesson are.
So, what might preview activities in a contextual approach look like?
The lesson will be markedly different from any conventional language
class that learners have previously taken, and some may struggle with
the level of autonomy they are granted. So, it is important to explain to
learners what they are going to do, and how they are going to do it. In
addition, it is necessary to think of a straightforward way of explaining
to students in simple terms why they are doing it.
232 I. Pemberton
13.6 While Watching Activities

This section covers while watching exercises in a contextual approach.
13.6.1 Activity Types
There are four core activity types:
1. Transcript completion: The simplest form of output activity is to ask

students to complete a gapped transcript. This kind of activity is more
appropriate for a shorter video and provides more time to analyse the
whole transcript.
2. Note-completion: To abridge a longer task, use a note comple-
tion activity. If the notes focus on the main points and the main
supporting points, this helps learners to build up an outline of the
most important information in the text.
3. Summary completion: A similar exercise which also focuses students’
attention on the main ideas of a text and the language used to
articulate them.
4. Content Questions: Sometimes, a text is not neatly hierarchically
structured to provide a clear set of notes or summary, but it may
be highly engaging for learners. Identifying the main points might
also require a bit more thought. Use content questions to do this.
These differ from conventional comprehension questions which are
designed to do no more than test comprehension. The purpose of
content questions is to draw attention to the most important points
of the text and in answering the questions learners create a coherent
knowledge framework.
13.6.2 Captions
To see or not to see? That is the question. The use of captions is a contro-
versial issue in language teaching and learning. There are arguments both
for not using them, and for using them. Let’s think about this in detail
and also review relevant research.
The sixth principle of a contextual approach is the Attention Principle.
This hypothesizes that attention to the form of an utterance promotes
deeper learning. Most experts agree that memory is divided into long-
term and working memory. Wen (2015: 41) explains that working
memory allows us to maintain a small amount of information in our
minds while we manipulate it to do such things as understand language,
do mathematical calculations, or solve problems. Working memory has
two components: a phonological loop which processes sounds, and
a visuospatial sketchpad which deals with images. Another feature of
working memory is that it has a limited capacity, which is generally
considered to be seven bits of information plus or minus two. However,
more recent research indicates that its capacity could be four plus or
minus one. In addition to having a limited capacity, information held
in working memory decays within seconds unless refreshed by rehearsal
(ibid:43).
A number of studies have shown that parallel auditory and visual
inputs support learning. A series of studies by Mayer (2014: 6) consis-
tently found that learning is enhanced when people use both their
auditory and visual channels simultaneously to process information. He
offers two explanations: One is quantitative—it doubles the amount
of information a learner processes. A second explanation, which Mayer
favours, is qualitative—a learner must go through the cognitive process
of integrating the auditory and visual information, which leads to deeper
learning. Clark and Mayer (2016: 71) add that ‘presenting words alone
may encourage learners - especially those with less experience or expertise
- to engage in shallow learning, such as not connecting the words with
other knowledge.’
Some investigations have looked at the parallel use of visuals, audio
narration, and subtitles. While it might be expected that increasing the
number of modalities from two to three would lead to better under-
standing, Clark and Mayer (2016: 131) summarize research that shows
that people learn better from graphics and audio narration, than from
graphics and audio narration plus onscreen text. This is because the
presentation of both images and the pictorial form of words overloads
234 I. Pemberton
the visual input channel as people try to focus on both of them at the
same time. This seems to suggest that when designing classes, it might
be better to limit the use of subtitles.
However, the participants in the studies were L1 users. There is
a significant amount of research that looks at the effectiveness of
using closed captions for second language learners. A meta-analysis by
Montero-Perez et al. (2013: 720) found that the use of subtitles had a
large effect on both the comprehension of L2 learners and their retention
of vocabulary. A meta-analysis by Kanellopoulou et al (2019) also found
that bimodal subtitles (L2 audio and L2 subtitles) improved listening
skills and long-term vocabulary retention. In a study of caption use,
Vanderplank (2019: 421) reports that participants made use of captions
to increase enjoyment, improve understanding, develop their listening
skills, and build vocabulary knowledge. While captions might have a
negative impact on the understanding of L1 viewers, studies show that
they have a positive effect on the understanding and vocabulary retention
of second language learners.
13.6.3 Recording Vocabulary
The recording of vocabulary is a key activity in the Lexical Approach.

Lewis (1997) goes into great detail about organizing principles, and
about formats. Organizing principles include topic, situation, colloca-
tion, person, and grammar. Formats include collocation boxes, pattern
displays, and discourse structures. While it may be taken for granted that
learners should record vocabulary, consideration should be given to the
purposes of recording vocabulary.
Group Lessons
In a group lesson, the teacher selects the vocabulary for learners to
focus on. The vocabulary chosen should have a surrender value, either
in a language examination or in the target community. A vocabulary
record can form the basis of spaced repetition. It can be reviewed in the
next lesson, and it can also be converted into language tests for students.
Setting numerical targets and testing vocabulary can be used to judge
individual progress.
Individual Study
While it is more important to increase exposure to the target language
than spend time reviewing what has already been heard and understood,
there is a value in recording vocabulary. When studying online, an elec-
tronic medium, such as an excel spreadsheet, is convenient to record
vocabulary. Have one column for the target vocabulary, and a hidden
column for meaning, which allows for review.
An important reason for recording vocabulary is that while listening,
checking meaning, and writing it down, learners attend to form. A
listener holds the vocabulary in short-term memory while moving
between the video, dictionary or translation software, and spreadsheet.
While it is useful to review the vocabulary, it is not critical because
learners have a lot to learn, and as they watch more videos, they meet
the same words and phrases in different contexts and gain an intuitive
understanding of the frequency of each construction.
Format
It is useful to record vocabulary in the order that you hear it. While
watching an instructional video, documentary, drama, or animation, a
learner builds up a cognitive representation of the content (a content
schema) and, at the same time, builds a parallel linguistic schema. The
content schema helps to remember and recall the language.
An important point is to record new words or new patterns together
with their co-text to associate a word not only with its conceptual
context, but also its lexico-grammatical context. The objective is not only
to remember constructions but to relate them to other constructions that
form a part of the surrounding conceptual and ideational context.
13.7 Post-Watching Activities

A cursory reading of the principles of a contextual approach, and in
particular the assertion that language is learnt through primarily receptive
activities, might give the impression that there is little or no produc-
tion. While the approach prescribes a strong focus on receptive activities,
productive activities help learners to remember specific constructions
236 I. Pemberton
which are useful in cases such as preparing for an examination. Where

learners have little opportunity to produce language, the inclusion of
productive activities can also provide motivation for learning.
13.7.1 Flashcards
Flashcards are a time-honoured activity in language teaching and

learning and digitisation has made them even more convenient to create
and use. If new vocabulary is recorded in an excel spreadsheet, it can be
exported directly into a vocabulary learning app and converted into elec-
tronic flashcards. The flashcards can then be used on any of a learners’
devices.
With student groups, flashcards can be utilized after analysing a video
and directly before a production activity to increase the chances that
students integrate new vocabulary with their existing interlanguage. A
teacher can also display digital flashcards via an electronic whiteboard to
quickly check learning at the end of class, and to review learning at the
beginning of the next class.
13.7.2 Pronunciation
For an individual learner, rehearsing language in short-term memory

while recording it allows them to focus on form, which is a necessary
condition for acquisition. At the same time, considering the audiovi-
sual medium, it also helps learners to internalize the pronunciation of
constructions.
The teacher can also make use of flashcards to improve pronunciation.
Unlike traditional pronunciation training which is likely to be decontex-
tualized and may have a main focus on linguistic phenomena such as
minimal pairs, in a contextual approach, all aspects of pronunciation can
be dealt with. That is to say, not only phonemic differences but also
the pronunciation of words, phrases, clauses, and sentences. In teaching
pronunciation above a word level, a teacher can deal not only with word
stress but also phrase, clause, and sentence stress. This also allows for the
teaching of rhythm and intonation, and for the teaching of linguistic

phenomena such as assimilation, elision, linking, and intrusion.
13.7.3 Recounts
One of the key output techniques in a contextual approach is recounts.

A convention throughout language teaching history and codified in the
Present-Practice-Produce paradigm has been to present new language,
usually a grammatical structure or form before asking learners to prac-
tice the language by working with decontextualized examples before
producing the language by transferring it to new and possibly unrelated
contexts.
A contextual approach follows this commonsense approach of prac-
tising newly presented language, but it keeps the language contextualized
in the Practice phase. That is to say, it asks learners to use the informa-
tional content of the text as a guide to producing the language. The main
purpose of the process of recounting is to create, strengthen, and main-
tain neurological representations within the brain of the content and the
language used to articulate it. The process of recounting information is
as follows:
Stage One: To begin with, learners work individually to write a
summary of the content. The purpose of this is to check their recall of
the informational content video and to give them the opportunity to
remember and use language patterns which are new to them from the
video.
Stage Two: As learners finish their written summaries, pair them with
classmates who have also finished in order to practise a spoken recount.
They should not read their summaries for this but will most likely need
a prompt card. One student recounts while their partner listens, then
they change roles. The listener can help out the speaker if they get stuck.
For instance, consider a note completion activity on the pros and cons
of social media. The written recount will look something like this:
238 I. Pemberton
Write a summary
Discuss the pros and cons of social media. You should explain:
• the benefits of social media
• why young people are at greater risk
• the dangers of social media
And finish by making a conclusion
The prompts for the spoken recount will follow a similar pattern:
Give a 2-minute talk about the pros and cons of social media
You should discuss:
• the benefits of social media
• why young people are at greater risk
• the dangers of social media
And finish by making a conclusion
There is an intentional strategy in the materials of increasing the

frequency with which students process and produce the same language.
Learners will watch and listen for language iteratively, practise it with
flash cards, use the language in writing, and also use it speaking.
13.7.4 Transcript
Once students have written a summary, they tend to want the teacher to
check whether their work is correct or not, but this misunderstands the
purpose of the activity. Provide a transcript at this point and tell students
to check what they have written against the transcript. This will allow
them to compare their own language with the actual language. In other
words, they will be able to notice the gap between what they can produce
and the target language, which will allow them to begin the process of
bridging the gap. In some ways this might seem similar to suggestions
made in Merrill Swain’s (1995: 26) Output Hypothesis:
First, it is hypothesized that output promotes ‘noticing’. That is to say, in

producing the target language (vocally or subvocally) learners may notice
a gap between what they want to say and what they can say, leading them
to recognize what they do not know, or know only partially. This may
trigger cognitive processes which might generate linguistic knowledge that
is new for the learners, or which consolidate their existing knowledge.
The important difference is that in a contextual approach, learners

are working autonomously. They are not speaking in pairs, and not
attempting to process input for meaning. They are not impeded by the
limitations of working memory. Another major difference afforded by
autonomous learning is that learners do not only notice that there is a gap
between what they wrote and what they wanted to write but, through
inspection of the transcript, they are provided with the means of bridging
the gap.
13.7.5 Knowledge Integration
This final activity type is consistent with the Produce phase of a conven-
tional language class. However, students are not expected to attempt far
transfer. Rather, it maintains and extends the existing context while rein-
forcing the language. This activity requires students to integrate new
knowledge from two or more lessons with their background knowledge.
For instance, if students have seen two videos about marriage rates such
as Rise in singletons in South Korea and Average age of marriage for women
in South Korea hits 30, they can be asked to do the following activity.
Write an essay on the following topic:

Nowadays, many developed countries are experiencing a decline in
marriage rates. What are the reasons for this? Do you think it is a positive
or negative development?
240 I. Pemberton
It can be seen that this particular essay does not require far transfer of
knowledge. Both videos refer to the Korean context. While maintaining
the original context of marriage rates, it requires learners to combine
knowledge and language of the Korean context with their own experience
and understanding.
13.7.6 Assessment
The primary purpose of any test in a contextual approach should be

assessment FOR learning, that is, spaced repetition of vocabulary rather
than a simple test of proficiency.
A useful test format is Paul Nation’s (2001) vocabulary tests. In
Learning Vocabulary in Another Language, Nation includes a receptive
vocabulary test which has been designed to test knowledge of the 10,000
most common words in the English language. He also includes a produc-
tive vocabulary test which similarly tests productive knowledge of the
first 10,000 words of the English language. The basic format of this test
is to provide sentences which include common collocates but with only
the first, second, or third letters of the second part of the collocate. This
format underlines the knowledge that L1 users possess of which words
are combined and with which frequency. While an L1 user would be
expected to correctly identify all or almost all of the collocates, even
advanced level L2 learners can struggle. Whereas they can sometimes
make logical word choices, these are different from the choices L1 users
make due to the differences in knowledge of conventional ways of saying
things and sensitivity to frequency.
A test can be designed following the same format to test the vocabulary
in a particular lesson, or to test vocabulary from a series of lessons. When
testing some time (say a week or a month) after the class has taken place,
it might be necessary to include additional letters or even whole words
to disambiguate between alternative correct answers, with more letters
being added according to the difficulty of the particular construction,
and whether the students have been given warning of a re-test.
13.8 Limitations
Unlike some of the major methods of the twentieth century, a contex-
tual approach is not intended for all learners in all situations. Thus, it is
important to recognize some of the limitations of the approach.
13.8.1 Level
First of all, a contextual approach is currently only intended to be used

with learners of an intermediate level and above. These will be students
who have made steady progress as they have gradually mastered the basics
of a language, such as its grammar, but have plateaued as a result of lack
of vocabulary, not simply a lack of word knowledge but also a lack of
knowledge of what lexico-grammatical relations words enter into, and
what topics, subjects and fields those words are used to articulate. They
will be able to communicate their meanings slowly and with frequent
errors but will lack automaticity. They will experience difficulty under-
standing what is being said to them due to lack of knowledge not only
of vocabulary, but of the prosody of a language.
A contextual approach does not eschew grammar or the teaching of
structural patterns in the early stages of the language learning journey.
The teaching of structural patterns of a high level of generality allows
beginners to function in a limited way in a language. As it is theoretically
underpinned by Dynamic Systems Theory and usage-based linguistics, a
contextual approach rejects the idea that grammar has a cognitive reality
for language users. Grammar is a human artefact. It is a partial descrip-
tion of language phenomena created by descriptive linguists. However,
a contextual approach does not reject the usefulness of grammar to
learners. It regards structural patterns as a set of tools with which learners
can mine a language for more information. When they have acquired
a sufficiently large inventory of language items, they can use these to
directly understand and produce language without the need to mediate
the input with such tools.
242 I. Pemberton
13.8.2 Resources
A contextual approach is only possible in higher resource contexts. It

might invite some initial confusion but taking a contextual approach
does not imply matching teaching methods to a local context. It refers
to presenting students with a rich and immersive multimedia context to
allow them to develop both ideational and lexico-grammatical knowl-
edge at the same time. In order to do this, learners will need a device
to work on such as a PC, laptop, or tablet. Mobile phones are possible,
but they tend to be too small to manipulate the video buttons. Thus, a
contextual approach will not be appropriate for low-resource teaching
contexts where there is a lack of access to digital devices or to the
Internet.
13.8.3 Mental Flexibility
As a contextual approach involves a high level of autonomy with little

teacher intervention, some students, depending on their previous educa-
tional experiences, may perceive it as having a lack of face value. For
this reason, the rationale for using the approach may need to be care-
fully explained at the beginning. On the other hand, video resources
are known to be highly engaging, and they allow learners to progress
at their own pace, and to focus on their own problem areas. Therefore,
it is anticipated that even where there might be some initial resistance,
students will make tangible progress leading to acceptance of a contextual
approach as a valid pedagogy.
13.8.4 Motivation
While not a problem that is limited to a contextual approach, a lack of

student motivation will short circuit learning no matter how engaging
the videos are, or how carefully crafted the materials are. As learners
are offered a high level of autonomy, less mature students may need
to be supervised carefully to ensure that they meet lesson objectives.
However, learners will be working mostly autonomously, so this offers

opportunities to speak to and engage them individually.
In summary, a contextual approach will not be suited to all learners in
all contexts. It is most suited to intermediate learners and above in higher
resource contexts who have the necessary technology, mental flexibility,
and motivation.
References
Clark, R. C., & Mayer, R. E. (2016). E-learning and the science of instruction:
proven guidelines for consumers and designers of multimedia learning (4th ed.).
John Wiley & Sons, Inc.
Kanellopoulou, C., Kermanidis, K. L., & Giannakoulopoulos, A. (2019).
The Dual-coding and multimedia learning theories: Film subtitles as a
vocabulary teaching tool. Education Sciences, 9 (3), 210.
Lewis, M. (1997) Implementing the Lexical Approach. Language Teaching
Publications.
Mayer, R. E. (2014). Introduction to multimedia learning. In R. E. Mayer
(Ed.), The Cambridge Handbook of Multimedia Learning (2nd ed., pp. 1–
24). Cambridge University Press.
Montero Perez, M., Van Den Noortgate, W., & Desmet, P. (2013). Captioned
video for L2 listening and vocabulary learning: A meta-analysis. System
(linköping), 41(3), 720–739.
University Press.
O’Keeffe, A. (2021). Data-driven learning—a call for a broader research gaze.
Language Teaching, 54 (2), 259–272.
Swain, M. (1995). Three functions of output in second language learning. In
G. Cook & B. Seidlhofer (Eds.), Principle & practice in applied linguistics:
Studies in honour of H. G. Widdowson. Oxford University Press.
Vanderplank, R. (2019). ‘Gist watching can only take you so far’: Atti-
tudes, strategies and changes in behaviour in watching films with captions.
Language Learning Journal, 47 (4), 407–423.
244 I. Pemberton
Wen, Z. (2015) 3. Working memory in second language acquisition and

processing: The phonological/ executive model. In Z. Wen, M. Borges-
Mota & A. McNeill (Eds.), Working memory in second language acquisition
and processing (pp. 41–62). Multilingual Matters.
14
Conclusion
14.1 Making the Case for Change

A reaction to the rituals and chanting of Audiolingualism, Communica-
tive Language Teaching, took the world stage by storm in the 1970s.
Today, it has become so well-established that it is difficult to voice criti-
cism of it, so much so that anyone disavowing CLT risks being labelled
a heretic. By claiming that we live in a post-method era, prominent
language educators have effectively formed a praetorian guard around
CLT, shutting down any form of a challenge to its hegemony.
The main aims of this book have been to build a case for a change and
to imagine what that change might be. At the beginning, the following
questions were posed:
1. Is Communicative Language Teaching the answer to our pedagogical

prayers, or is it a vain faith that promises more than it can deliver?
2. If CLT is not the acme of language teaching pedagogy, then what is?
Answers to these questions were sought by an examination of five

areas of inquiry: the history of language teaching methods, first language

Switzerland AG 2024
246 I. Pemberton
acquisition theories, chaos-complexity theory, construction grammars,

and usage-based linguistics.
The rationale for an examination of the history of language teaching
methods was that a knowledge of history informs the present and avoids
repeating the mistakes of the past. An analysis of the major language
teaching methods from Grammar-Translation onwards considered the
strengths and weaknesses of each one to identify important elements of
methodology.
The analysis of the Classical and Reform Periods provided insights into
the intuitions of innovative language educators who looked towards first
language learning as a source of inspiration. The Scientific Period drew
attention to the tension between knowing the structures of a language
and being able to use them, and Audiolingualism represented an attempt
to apply theory to practice. The Communicative Period from the 1970s
onwards marked several major milestones in the history of methods.
Each method highlighted something of value for language teachers.
Wilkins’ Notional-Functional Syllabus established the importance of
teaching the functions of a language. Krashen’s Natural Approach drew
attention to the importance of input. The Communicative Approach
highlighted learner interaction. Content-Based Instruction encouraged
educators to consider the role of content in language learning. Task-
Based Instruction stimulated debate about intentional and incidental
learning. The Lexical Approach drew attention to the nature and scope
of what learners have to learn.
For such a public phenomenon, an analysis of Communicative
Language Teaching was not straightforward. As if a bodyguard of
eminent opinion were not enough, CLT has always stayed one step
ahead of its critics. It has shape-shifted, beginning as the Communica-
tive Approach before metamorphosing into Communicative Language
Teaching and then splitting into Classical CLT and Current CLT. Along
the way, it has absorbed many disparate methods to sustain its growth.
This has given rise to a formless entity with fuzzy edges making it
difficult to define or evaluate.
Difficult, perhaps, but not impossible. By drawing together the
common features of communicative methods, it was possible to stand
14 Conclusion 247
CLT out in the cold light of day. A close examination led to the revela-
tion that CLT lacks a coherent theory of learning. Despite this serious
flaw, somehow it has continued to thrive unabated. But lacking the guid-
ance of a theory, it overemphasizes productive speaking activities in the
belief that these lead to fluency, misspending class time, and preventing
learners from achieving their full potential.
14.2 Imagining a Contextual Approach

Having established a rationale for change, the next question to address
was what form that change would take. The intuitions of educators
throughout history have been to look to first language acquisition
for answers, so this seemed like an appropriate place to start. The
examination of first language acquisition considered three areas: evolu-
tionary biology, classical theories of L1 acquisition, and Social-Pragmatic
Theory.
In evolutionary biology, a distinction is drawn between primary and
secondary biological skills, and it was argued that the primary skills of
watching and listening are far more effective for language learning than
the secondary skills of reading and writing.
The classical theories of Constructivism and Sociocultural Theory
have made important contributions to our understanding of L1 acqui-
sition. Constructivism posits that children actively construct knowledge
from interaction with their environments. On the other hand, Sociocul-
tural Theory postulates that children gain knowledge with the assistance
of more knowledgeable others and cultural artefacts.
These theories have had a strong influence on the use of active
learning in the second language learning classroom. In a traditional
teacher-led classroom setting, learners passively listen while the teacher
explains things. However, Constructivism posits that learners are not
just recipients of knowledge but actively construct it from input. In a
communicative classroom, learner activity is accomplished by putting
students into pairs and groups. However, an important point about
active learning is that it can be either behavioural or cognitive and only
248 I. Pemberton
the second type will result in learning. This can take place irrespective of
whether a learner is working with others or working alone.
A further rationale for learner interactions in the communicative class-
room is drawn from Sociocultural Theory. On the face of it, the theory
suggests that children learn their first language with the assistance of a
more knowledgeable other who listens and corrects them. That is to say,
learning takes place as a result of social interaction. However, this is not
borne out by research findings. These show that input from adults and
older siblings plays only a peripheral role. A final point is that a more
knowledgeable other need not even be a sentient being. It can also be a
computer, a textbook, a dictionary, or even a pencil and paper. Thus, it
appears that Sociocultural Theory does not provide a strong rationale for
pair and group activities.
The classical theories of child cognitive development provide some
guidance for second language learning. However, their main emphasis
is on the learning of concepts, and they lack specifics on the cognitive
processes of language acquisition. An understanding of these processes
can be gained from a more up-to-date theory of language learning—
Social-Pragmatic Theory. This theory defines language acquisition in
terms of three processes. First of all, it argues that children use inten-
tion reading to understand what adults mean. In other words, they don’t
have to go through the time-consuming process of forming hypotheses
and negotiating the meaning of each utterance. Secondly, the theory
explains that children imitate what they hear in order to internalize it.
Thirdly, children encode knowledge in memory using general pattern
finding skills including categorizing, analogy, and schema building.
Social-Pragmatic Theory is a usage-based theory of first language
acquisition. Usage-based theories claim that all languages, first, second,
and additional are acquired in the same way. For its theoretical foun-
dations, usage-based linguistics draws upon Dynamic Systems Theory
(DST) and constructionist grammars.
Dynamic Systems Theory is a general complexity theory which posits
that a complex system is non-linear, self-organizing, and emergent. A
language is clearly a complex system consisting of tens or even hundreds
of thousands of lexico-grammatical patterns which range from general,
in the case of grammar, to specific, in the case of concrete examples
14 Conclusion 249
of vocabulary. A language is also a non-linear system. This means that

there is no direct relationship between its causes and effects. A seem-
ingly insignificant input to the system can bring about a major change.
However, a critical mass of language knowledge needs to be built up
before a change in proficiency can occur. This is known as a phase shift.
A language is also emergent, which means that the many different inter-
actions that provide lexico-grammatical inputs to the system give rise to
patterns of use.
In addition to Dynamic Systems Theory, usage-based linguistics draws
upon constructionist grammars for its theoretical underpinnings. These
grammars share three key concepts: first of all, the fundamental unit
of language is the construction. A construction is a phonological or
orthographic pattern that symbolizes a semantic meaning. Secondly,
constructions can represent all language items. Usage-based theorists do
not draw a distinction between meaningless grammar rules and mean-
ingful lexis. All language is meaningful. No matter how atomistic or
complex, constructions can be placed on a continuum with more abstract
items (traditionally grammar) at one end, and with more specific items
(conventionally lexis) at the other. Thirdly, all constructions are stored in
a constructicon. Most constructionist grammars consider the construc-
ticon to contain a substantial amount of redundancy as specific lexical
items exist alongside their more abstract representations.
Usage-based linguistics theorizes language development as a process
of entrenchment. After initial exposure to a concrete item of language,
subsequent activations strengthen its representation in memory leading
to automatization during processing and production. In a similar way,
when a learner is exposed to different types of constructions with similar
features, they employ the domain-general learning abilities of associa-
tion, categorization, and analogy to abstract those features and create
schematic structures which can then be used creatively in language
production.
A survey of the usage-based linguistics literature suggests its effective-
ness for second language learning. A number of cross-sectional studies
have investigated general trends. These show that usage-based learners
make more progress than learners taught by other pedagogies. When
compared to structure-based programmes, students who followed a
250 I. Pemberton
usage-based programme did better in tests of general proficiency and did

as well on tests of grammar. In comparison with students who followed
communicative programmes, usage-based learners did better in general
proficiency tests and as well in a test of speaking. These results suggest
that neither explicit instruction in grammar nor communicative pair and
groupwork activities are a necessary condition for language development.
14.3 Designing a Contextual Approach

Attempts to bridge the gap between theory and practice have always
presented a challenge. Academics who inhabit the Ivorian towers of
universities find it difficult to visualize classrooms in which they don’t
teach, and practitioners with little access to research and even less time
to look at it are guided by institutional expectations or dependent on
the latest educational fads to inform their teaching. A main aim of
this manuscript has been to bridge the gap between theory and prac-
tice. Insights from multiple overlapping theories have informed the
development of six principles for a contextual approach.
The primary principle of a contextual approach is that a context is
critical for language learning. This may be something which is taken for
granted but the point is that contexts are underused in teaching language
whereas they should be foregrounded. Drawing upon schema theory, a
context provides an ideational framework for language, and use of that
framework in the classroom facilitates the understanding and learning of
the language to articulate it.
The second principle calls for a strong focus on input. This is not
something that many will disagree with but the lack of theory under-
pinning CLT has resulted in the valorization of output. Starting with
Krashen’s insight that comprehensible input is a necessary condition for
language learning, Lewis’s Lexical Approach highlighted the scale of what
language learners have to learn. Dynamic Systems Theory further high-
lights the complexity of the language system and the sheer volume of
language items that have to be learnt. A substantial amount of input is
indisputably necessary for the achievement of fluency.
14 Conclusion 251
The third principle of a contextual approach prescribes the use of

multimedia materials. While there is plenty of research that reveals the
general benefits of multimedia learning, multimedia materials create a
rich visual and auditory environment in which learners can employ their
primary biological skills of observing and listening.
The fourth principle of a contextual approach, and one which might
be contentious given the current paradigm, is that learning should be
mostly autonomous. The underuse of listening in the language learning
classroom, the argument that listening practice should be learner-centric,
and research into learner control in multimedia learning all point to the
effectiveness of autonomous learning.
The final two principles relate to the use of multimedia materials. The
fifth principle states that iteration is key, that is, working over the same
language material repeatedly in order to internalize it. While this may be
self-evident, it requires learner control.
The final principle relates to the cognitive and psycholinguistic prin-
ciple of attention. The importance of focus on form is recognized, even
within the current CLT paradigm. Task-based Instruction advocates fret
over whether form-focused instruction should go before the task or after
it, but this is due to the collaborative nature of tasks which prevent
learners from focusing on form. With primarily autonomous learning,
learners can focus on form at any and all points during classroom
activities.
14.4 Implementing a Contextual Approach

The final question is: Can historical insights and theoretical principles
be transformed into teaching materials? While the answer is yes, doing
so results in reimagining the classroom. The principles of a contextual
approach can be applied to the selection of language learning materials,
and the design of activities, classroom procedures, and assessment.
The Context, Input, and Modality principles guide the design of
materials. The use of video creates a context for the language to be learnt.
It is important to pay attention to its relevance for learners, the coherency
252 I. Pemberton
of its message, whether main ideas are easily identifiable, and whether it
contains clear examples of constructions for the learners to focus on.
The Context Principle also guides the design of activities. There are
two objectives for while-watching activities: focus learners on content
and focus them on language. To focus learners on content requires
specific types of activity. One of these activities is content questions.
These are distinct from comprehension questions in that they focus
learners on main ideas. Other types of activities which will focus
learners on the informational content of a video are note and summary
completion activities.
The Context, Iteration, and Attention Principles guide the design of
post-watching activities. These can include (digital) flashcards to increase
frequency and promote the stability of constructions in memory. To
encourage learners to make sense of the content, written recount, where
learners write down the main ideas, and spoken recount, where learners
articulate those ideas in speech, can be used. In addition to helping
learners build content and language schemata, these activities increase
the frequency with which learners use constructions, and they increase
the stability of those constructions in learners’ minds.
In keeping with ELT tradition, it may be useful to sketch out what
a contextual approach classroom looks like. Classroom procedures are
guided by the Autonomy, Iteration, and Attention Principles. To study
a multimedia context which provides visual and auditory input, and
to work autonomously, learners will need a device—either a laptop
or tablet—and an internet connection. A mobile phone is marginally
acceptable but difficult to manipulate. Alternatively, they could work in
a room equipped with personal computers. Headphones or earphones
are also a requirement.
As learners study primarily autonomously, they control the pace with
which they work through the materials, what they focus on, and the
number of times they review language that they have difficulty with. The
task-based conundrum of when to focus on form becomes a non-issue.
This is integrated with the activities. It is in part determined by what the
teacher selects for deletion from the materials, but also learners have the
agency to decide whether or not to focus on form at other points while
viewing a video. While learners work independently to watch and listen
14 Conclusion 253
to the video, the teacher monitors what they are doing, intervening only
when necessary to keep learners on track.
The one thing some teachers will struggle with, and particularly those
trained in communicative teaching, is the idea of having a primary focus
on input. Does this mean there is no output? A contextual approach
does not preclude output, rather, in line with the tendencies of cognitive
linguistic theories to identify input as the main driver of acquisition, it
backgrounds output to foreground input. There are a number of ways
that learners can output, and learners are not precluded from working
together. To begin with, the use of flashcards as an immediate self-test
requires them to recall the language. Similarly, writing down the main
ideas requires learners to make sense of the input and to use the language.
At this point, having perceived, attended to, and produced language,
learners can work together to recount the main ideas in pairs. Teachers
can also employ free practice activities such as topical discussions and
debates, particularly where learners are studying a foreign language, but
this might be less important when they are studying for exams or are able
to use what they learn in a target language community.
Assessment in a contextual approach is assessment for learning. That
is, testing represents spaced repetition of learnt vocabulary. Teachers
can test learners’ ability to correctly apply learnt vocabulary to gapped
constructions.
14.5 Conclusion
Some fifty years ago, the appearance of Communicative Language
Teaching caught the mood of language teaching professionals the world
over and rescued the field from the totalitarianism of Audiolingualism. In
what is now known as the Communicative Period, CLT has become the
new status quo, not a new totalitarian regime but more a protectorate
ruled over by prominent educators in positions of power and influ-
ence. But Communicative Language Teaching is an approach that lacks
a coherent language learning theory. Without the guidance of a theory,
it overrelies on spoken production, a naïve strategy which is inconsistent
254 I. Pemberton
with recent language acquisition theory. As a result, it misspends class

time and prevents learners from reaching their full potential.
In contrast, usage-based second language instruction is a context-
driven multimedia language learning pedagogy which translates current
theory into practice. It is not suitable for all teaching environments, nor
for all levels of learners. It will no doubt meet resistance from both the
rank and file and the powers that be. But for individuals and teachers
who wish their learners to reach their full potential and be regarded as
legitimate language users of target language communities, it provides the
tools and the directions to continue the journey towards fluency.
Index
A Kusyk, M. 185
Abstract constructions 161 Sockett, G. 184
Abstraction 117 Toffoli, D. 182, 183
Acquisition-Learning Hypothesis 36
Active Learning 56, 111
Adaptation 133 B
Affective Filter Hypothesis 38 Behavioural Principles 207–214
Analogy 153 attention principle 209
Assessment 240 iteration principle 207
Association 117, 150–152
paradigmatic association 152
pragmatic association 152 C
symbolic association 151 Case studies 160–164
syntagmatic association 151 Eskildsen, S.W. 162
Attention Principle 209 Li et al. 162
Audiolingual Method 21 Mellow, J.D. 161
Audio materials 203 Roehr-Brackin, K. 163
Authenticity 197 Categorization 117, 153
Autonomy Principle 205 Chunks 92, 93
Autonomy studies 182–186 Classical Period 8–9
© The Editor(s) (if applicable) and The Author(s), under exclusive 255
licence to Springer Nature Switzerland AG 2024
https://doi.org/10.1007/978-3-031-53414-0
256 Index
Collocations 87 comprehensible output hypothesis

Communicative Activities 59–62 71
discussion 61 Content Questions 232
information gap 59 Context Principle 196
information gathering 60 Contextual Approach 196, 214, 224,
information transfer 60 241, 250
jigsaw 60 Continuum 137, 147
role play and simulation 61 Corpus analyses 173–175
Communicative Competence 24–26 Berger, C.M. 173
Communicative functions 35 Jo, R. 174
Communicative Language Teaching Oh, S.-Y. 174
47–65, 156, 195 Romer, U. 173
classical and current versions 49 Cross-sectional studies 164–172
communicative activities 59 Dahl, A. 170
communicative principles 49 Gustafsson, H. 168
strong and weak versions 48 Piggott et al. 171
Communicative Period 31–44 Rousse-Malpat, A. 172
natural approach 35 Smiskova, H. 171
notional-functional syllabus 32 Verspoor, M. 164, 165, 168
Communicative Principles 49–58 Cultural imitation 116
active learning 56 Cultural learning 117, 185
aim 49
authentic materials 56
inputs 54 D
interaction 57 Direct Method 12
meaningfulness 57 Dynamic Systems Theory 126–136,
theory of learning 50 248
treatment of errors 58 adaptation 133
Complexity 127 complexity 127
Comprehensible Output Hypothesis dynamism 130
71 emergence 135
Comprehension Approach 207 non-linearity 134
Conceptual Schema 198 openness 131
Constructicons 137, 138, 147, self-organization 132
229–230 Dynamism 130
Construction Grammars 136–138
Constructivism 109–111
Content-Based Instruction 69–72 E
Ego-centric speech 110
Index 257
Emergence 135 H
Empirical Research 159–190 Historical Teaching Approaches
autonomy studies 182 7–15
case studies 160 classical period 8
corpus analyses 172 reform period 9
cross-sectional studies 164
multimedia studies 179
variation and variability studies I
175 Incidental learning 209
Entrenchment 148–150 Individuals 235
Environmental Principles 196–206 Inner speech 113
autonomy principle 205 Input Hypothesis 37
context principle 196 Input Principle 201
input principle 201 Intention reading 116, 184, 248
modality principle 203 Interaction Hypothesis 51
Evolution 106–108 Introductions 228
Iteration Principle 207
F
K
Fillmore, C. 136, 141 Knowledge Integration 239
First Language Acquisition 105–122 Krashen, S. 32, 35, 71, 92, 154,
constructivism 109 198, 201
evolution 106
primary biological skills 107
social-pragmatic theory 115 L
sociocultural theory 112 Langacker, R.W. 117, 136, 137,
Fixed (Institutionalized) expressions 141, 145, 148
88 Language Schemata 199
fully fixed expressions 88 Larsen-Freeman, D. 21, 22, 31,
semi-fixed expressions 89 127–135, 149, 150, 176, 204,
Flash cards 236 205, 209
Formulas 118, 154, 160 Level 241
Frequency 148, 200 Lewis, M. 85, 108, 234
Fully abstract constructions 118, 154 Lexical Approach 85–102
Fully fixed expressions 88 learning theory 91
lexical categories 86
methodology 92
G notebooks 97
Grammar-Translation Method 8 theory of language 90
258 Index
Lexical Categories N
collocations 87 Natural Approach 35–43, 92, 154
fixed (Institutionalized) acquisition-learning hypothesis 36
expressions 88 affective filter hypothesis 38
words 86 input hypothesis 37
Lexicon 128 monitor hypothesis 37
Limitations 241–243 natural order hypothesis 37
Logical division of ideas 228 teaching procedures 39
Long, M. 51 Natural Order Hypothesis 37
Low-scope patterns 118, 154, 161 Negotiation of meaning 52, 79, 114
Non-linearity 134
Note-completion 232
Notional-Functional Syllabus 32–35
M
communicative functions 35
Mastery System 11
modal meaning 34
Materials Design 221–243
semantico-grammar 34
constructions 229
Nunan’s Task Cycle 78
limitations 241
multimedia materials 224
post-watching activities 235
O
pre-watching activities 231
role of the teacher 224 Openness 131
while watching activities 232 Other Communicative Approaches
Mental flexibility 242 69–82
Modality Principle 203 content-based instruction 69
Modal meaning 34 task-based instruction 72
Monitor Hypothesis 37
More Knowledgeable Other 112,
114, 224 P
Motivation 242 Pairwork 114
Multimedia materials 204, 224–227 Paradigmatic association 152
Multimedia studies 179–182 Paralinguistic information 198
Athanasopolous, P. 179 Pattern finding skills 117, 185, 248
Bylund, E. 179 Piaget, J. 109
Irshad, F.M. 181 Post-watching activities 235–240
Koster, D.E.S. 181 Pragmatic association 152
Nguyen, T.P.H. 180 Pre-watching activities 231
Primary Biological Skills 107
Principles of a Contextual Approach
195–217
Index 259
behavioural principles 207 Social speech 113

environmental principles 196 Sociocultural Theory 112–115
Private speech 113 Spontaneous concepts 112
Pronunciation 236 Structural Syllabus 17–19
Student groups 234
Subsystems 128
R Subtitles 232
Rational Method 10 Summary completion 232
Recording vocabulary 234 Swain, M. 71, 239
Recounts 237 Symbolic association 151
Reform Period 9–13 Syntagmatic association 151
direct method 12
mastery system 11
rational method 10 T
series method 11 Task-Based Instruction 72–81, 156
Resources 242 defining task 72
Role of the teacher 224 focus on form 79
inputs and outputs 76
learning theory 76
S negotiation of meaning 79
Salience 149 task types 74
Scaffolding 112 terminology 75
Schematization 118, 154 Teaching procedures 39
Scientific concepts 112 Text types 228
Scientific Period 17–29 Theory of mind 116
audiolingual method 21 The Structural Syllabus 17–19
communicative competence 24 Title(s) 227
situational syllabus 19 Tomasello, M. 23, 24, 114–118
structural syllabus 17 Topic sentences 228
universal grammar 23 Transcript 238
Self-organization 132 Transcript completion 232
Semantico-grammar 34
Semi-fixed expressions 89
Series Method 11 U
Shadowing 213 Universal Grammar 23–24
Situational Syllabus 19–21 Usage-Based Linguistics 145–157
Socialized speech 110 analogy 153
Social-Pragmatic Theory 115–118, association 150
247 categorization 153
260 Index
constructicon 147 Verspoor, M. 176

continuum 147 Vygotsky, L. 112, 224
entrenchment 148
schematization 154
W
While watching activities 232–235
V Wilkins, D. 32
Variation and variability studies Willis’s Task Cycle 76
175–178 Words 86
Bardovi-Harlig, K. 176 Working memory 53
Bulte, B. 178
Housen, A. 178
Larsen-Freeman, D. 176 Z
Lowie et al. 177 Zone of Proximal Development 112,
Spoelman, M. 176 224

978-3-031-53414-0

Uploaded by

Copyright:

Available Formats

978-3-031-53414-0

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

978-3-031-53414-0

Uploaded by

Copyright:

Available Formats

Usage-Based

ISBN 978-3-031-53413-3 ISBN 978-3-031-53414-0 (eBook)

Cover image: Peerapat Tandavanitj

Paper in this product is recyclable.

Part I Making the Case for Change

3.5 Communicative Competence 24

Part II Establishing a Theoretical Basis

Part III Designing a Blueprint for Change

It is often said that with the advent of Communicative Language

these foundational theories, the book explores the usage-based account

It may be tempting to think that as ELT moves from method to post-

© The Author(s), under exclusive license to Springer Nature 7

examines some of the antecedents of modern-day teaching methods

1. 1750–1880: The Classical Period

2.1 The Classical Period

2.1.1 The Grammar-Translation Method

difficult vocabulary, problematic grammar, and providing cultural back-

2.2 The Reform Period

modern languages should be taught as written and spoken communica-

2.2.1 The Rational Method

While Claude Marcel is less well-known than Prendergast and Gouin, he

2.2.2 The Mastery System

2.2.3 The Series Method

A French language teacher, Francois Gouin, also made some percep-

conceptions. Language is a means of thinking, of representing the world

2.2.4 The Direct Method

While individual educators were testing out their theories in Europe,

example, Brown (2001: 377) advises trainee teachers to prevent learners

and combined these to communicate messages. Francis Gouin’s Series

All of these features were precursors to later “innovations” in language

3.1 The Structural Syllabus

© The Author(s), under exclusive license to Springer Nature 17

useful patterns in a language could be identified and taught, and then

learning. It freed us from the obsession with words by showing us that

3.2 The Situational Syllabus

A situational syllabus is an inventory of situations in which learners

• the physical context of the language event

There are a number of arguments in favour of using a situational

• language cannot be fully understood without reference to its social

is not adequate for a general language learner. Mackey (1977) cited in

3.3 The Audiolingual Method

regarded its activities as ‘excruciatingly boring.’ Similarly, Richards and

3.4 Universal Grammar

Notwithstanding, the abstruse nature of his theory, Howatt (1984:

3.5 Communicative Competence

1. whether (and to what degree) something is formally possible

In other words, communicative competence is the outcome of the

1. Grammatical competence which includes morphological, syntactic,

It is often said that the significance of Hymes’ proposals for commu-

that ‘the introduction of communicative competence as a guide for

or relative clauses. These can be arranged into levels and taught in

meaning. The Situational Syllabus contextualized language but didn’t

• linking language to its situations of use

Hall, G. (2018) Exploring English language teaching: language in action (2nd

© The Author(s), under exclusive license to Springer Nature 31

• While the teaching of functions is a key feature of communicative

4.1 The Notional-Functional Syllabus

This category includes common semantic concepts with a high utility

4.1.2 Modal Meaning

This category represents interpersonal meaning. That is to say, a speaker’s