Issues in applied linguistics

Michael McCarthy
published by the press syndicate of the university of cambridge
The Pitt Building, Trumpington Street, Cambridge, United Kingdom
cambridge university press
The Edinburgh Building, Cambridge CB2 2RU, UK
40 West 20th Street, New York, NY 100114211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
Ruiz de Alarcon 13, 28014 Madrid, Spain
http://www.cambridge.org
Cambridge University Press 2001
This book is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 2001
Printed in the United Kingdom at the University Press, Cambridge
Typeface 9.5/13.5 pt. Swift Regular [vn]
A catalogue record for this book is available from the British Library
Library of Congress Cataloguing in Publication data applied for
ISBN 0 521 58487 6 hardback
ISBN 0 521 58546 5 paperback
Contents
Foreword page iv
Acknowledgements vii
1 Applying linguistics: disciplines, theories, models, descriptions 1
2 Language and languages 22
3 Modelling languages: the raw material of applied linguistics 44
4 Language acquisition: methods and metaphors 68
5 Language as discourse: speech and writing in applied linguistics 92
6 Applied linguistics as professional discourse 118
References 145
Index 173
1
Applying linguistics: disciplines, theories,
models, descriptions
1.1 Applied linguistics as problem-solving
In their day-to-day business, professionals whose work involves language
in some way or another often face problems that seem to have no immedi-
ate or obvious solution within the habitual practices which demarcate
their professional expertise. One avenue open to those who nd them-
selves in this position is to have recourse to the discipline of linguistics. It
is the belief that linguistics can offer insights and ways forward in the
resolution of problems related to language in a wide variety of contexts
that underlies the very existence of the discipline usually called applied
linguistics. Applied linguists try to offer solutions to real-world problems
in which language is a central issue (Brumt 1991:46), however tentative
or implied those solutions may be. What, then, might fall within the
domain of typical applied linguistic problems? A list of such problems will
certainly be wide-ranging and potentially endless, but might include the
following:
1 A speech therapist sets out to investigate why a four-year-old child has
failed to develop normal linguistics skills for a child of that age.
2 A teacher of English as a foreign language wonders why groups of
learners sharing the same rst language regularly make a particular
grammatical mistake that learners from other language backgrounds
do not.
3 An expert witness in a criminal case tries to solve the problem of who
exactly instigated a crime, working only with statements made to the
police.
4 An advertising copy writer searches for what would be the most effec-
tive use of language to target a particular social group in order to sell a
product.
5 A mother-tongue teacher needs to know what potential employers
1
consider important in terms of a school-leavers ability to write reports
or other business documents.
6 A historian wishes to understand the meanings of place-names in a
particular geographical area and how they have changed over time.
7 A person constructing a language test for non-native speakers for entry
into further education needs to know what the key linguistic or
psycholinguistic indicators are of reading ability in a second or foreign
language.
8 A literary scholar suspects that an anonymous work was in fact written
by a very famous writer and looks for methods of investigating the
hypothesis.
9 A dictionary writer ponders over possible alternatives to an alphabeti-
cally organised dictionary.
10 A computer programmer wrestles with the goal of trying to get a
computer to process human speech or to get it to translate from one
language into another.
11 A group of civil servants are tasked with standardising language usage
in their country, or deciding major aspects of language planning policy
that will affect millions of people.
12 A body is set up to produce an international, agreed language for use
by air-trafc controllers and pilots, or by marine pilots and ships
captains.
13 A zoologist investigates the question whether monkeys have language
similar to or quite distinct from human language and how it works.
14 A medical sociologist sets out to understand better the changes that
occur in peoples use of language as they move into old age.
The list could continue, and with professional diversication of the kind
common in modern societies, is quite likely to grow even bigger over the
years. What all these professional problems have in common is the possi-
bility of turning to the discipline of linguistics to seek insight and poten-
tial solutions. If they were to do this, the professionals directly involved
would become, even if only temporarily, applied linguists. This is different
from saying that there is a community of applied linguists (usually asso-
ciated with university academic departments) whose job it is to mediate
(and teach) linguistics and to suggest applications. That there is such a
community is not questioned here; the existence of academic journals
such as Applied Linguistics and International Review of Applied Linguistics, and
the provenance of the majority of articles published in them, is ample
2 Applying linguistics: disciplines, theories, models, descriptions
evidence (for further argument on this aspect of the mediation of theory
see Block 1996). But in this book I shall advocate that doing applied
linguistics should not be only the responsibility of the academic commu-
nity.
Over the last few decades, more and more people working in different
professional areas have sought answers to signicant problems by inves-
tigating how language is involved in their branch of human activity. This
has been especially notable in very recent years in areas such as (3), (10)
and (14) in the list of possible problems above (e.g. the growth of forensic
applications of linguistics, see Kniffka et al. 1996; the growth of interest in
language and the elderly, see Coupland et al. 1991). Other areas, such as (1),
(2) and (8), have used linguistic knowledge and insight over a much longer
period. In the future, even more professions will almost certainly turn to
linguists for potential solutions to practical problems: the increasing
sophistication of computers is just one obvious example where a corre-
spondingly complex understanding of human language may be benecial.
Thus even more professionals will have the opportunity to become applied
linguists.
No one will need to embrace the whole range of the discipline of
linguistics to nd a solution to their particular problem. Linguistics itself
is now an extremely broad discipline, and we shall see in this book just
how large a number of interests it encompasses. Furthermore, within this
broad discipline, the various compartments into which the subject falls
are themselves quite vast (e.g. see Malmkjaers 1991 encyclopedia of the
discipline), and compartmentalisation creates its own problems for the
application of linguistics (see Brumt 1980 for a discussion). What this
book will try to do in its limited scope is to exemplify how language
teachers and others involved directly or indirectly in language teaching
and learning (such as materials writers, syllabus designers, dictionary
writers, etc.) may approach their problems via the many and varied aspects
of linguistic study. Wherever relevant, I will also mention work done by
other, non-pedagogical applied linguists in the spirit of learning and
beneting fromtheir insights and in the fostering of a shared professional
identity, which can only be a good thing. The book cannot and does not
pretend to offer prescriptions for the solving of every problem. You, the
reader, will, it is hoped, see how and where linguistics might rub shoul-
ders with your own professional preoccupations.
1.1 Applied linguistics as problem-solving 3
1.2 Linguistics and applied linguistics: hierarchy or partnership?
Applied linguistics, I shall maintain throughout this book, is essentially a
problem-driven discipline, rather than a theory-driven one, and the com-
munity of applied linguists has characterised itself in the historiography
of the discipline by variety and catholicismof theoretical orientation. This
is in contrast to linguistics, where association with particular schools of
thought or theories tends to exert considerably greater centripetal force.
Indeed, not least of the questions immanent in a book such as this one are:
Can there be a unitary theory of applied linguistics, or indeed do theories of
applied linguistics exist at all? Is it not a dening quality of applied
linguistics that it draws its theory off-the-peg from linguistics; in other
words, that it should be understood as what Widdowson (1980) calls
linguistics applied? One major difculty in asserting the latter is the viabil-
ity of the view that linguistics exists as a set of agreed theories and
instruments that can be readily applied to real-world language-related
problems. Such a view oversimplies the natural and desirable state of
continuous ux of the discipline of linguistics (e.g. see Makkai et al. 1977),
or of any discipline for that matter, and obscures the two-way dialogue
that the academic applied linguistic community has had, and continues to
have, with its own community of non-academic practitioners and with its
peers within linguistics.
Applied linguistics can (and should) not only test the applicability and
replicability of linguistic theory and description, but also question and
challenge them where they are found wanting. In other words, if the
relationship between linguistics and its applications is to be a fruitful
partnership and neither a topdown imposition by theorists on practi-
tioners admonitions of which are implicit in Wilkins (1982) nor a
bottomup cynicism levelled by practitioners against theoreticians, then
both sides of the linguistics/applied linguistics relationship ought to be
accountable to and in regular dialogue with each other with regard to
theories as well as practices (see also Edge 1989). Accountability can
discomt both communities, and abdication of accountability is some-
times the easier line to adopt. I shall attempt wherever possible to refrain
from such abdication in this book, and bi-directional accountability will
be considered an important constraining inuence on both the applicabil-
ity of linguistics and the evaluation of applied linguistic solutions. Ac-
countability will centre on a set of responsibilities falling on the shoulders
of linguists and applied linguists in turn. These include:
4 Applying linguistics: disciplines, theories, models, descriptions
1 The responsibility of linguists to build theories of language that are
testable, which connect with perceived realities and which are not
contradicted or immediately refuted whenthey confront those realities.
2 The responsibility of linguists to offer models, descriptions and explana-
tions of language that satisfy not only intellectual rigour but intuition,
rationality and common sense (but see Widdowson 1980 for comments
on both sides of this particular coin).
3 The responsibility of applied linguists not to misrepresent theories,
descriptions and models.
4 The responsibility of applied linguists not to apply theories, descrip-
tions and models to ill-suited purposes for which they were never
intended.
5 The responsibility of applied linguists not simply to apply linguistics
but to work towards what Widdowson (1980) calls relevant models of
language description (see also Sridhar 1993, who sees applied linguists
as generating their own paradigms for studying language).
6 The responsibility of applied linguists to provide an interface between
linguists and practitioners where appropriate, and to be able to talk on
equal terms to both parties (see James 1986).
7 The responsibility on both sides to adopt a critical position vis-a`-vis the
work of their peers, both within and across the two communities.
8 The responsibility of both communities to exchange experience with
front-end practitioners such as language teachers, psychologists or so-
cial workers, who may not have a training in linguistics nor the time or
resources to do applied linguistics themselves, but who may be genu-
inely eager to communicate with both groups.
1.3 Theory in applied linguistics
Posing the question whether applied linguists should have theories and
whether the discipline as a whole should seek a unifying and homogenous
set of theoretical constructs is, in my view, a misleading and unproductive
line to pursue, and one which will be discussed further in Chapter 6. It is
difcult enough to establish a set of central tenets that unites the gen-
erally pro-theoretical community of linguists (but see Hudson 1988 for an
interesting list of such tenets; see also Crystal 1981:2, who takes a fairly
optimistic view of the existence of a common core within linguistics), let
alone bring under one umbrella the diversity of approach that marks out
1.3 Theory in applied linguistics 5
the domains of operation of applied linguistics. Within linguistics, widely
differing theories lay claim to deal with what is important in language: as
we shall see, a sentence grammarian may differ fundamentally from a
discourse analyst over the question of just what is the central object of
study. On the other hand, the sentence grammarian and discourse analyst
may unite in distancing themselves from the more speculative claims of
those trying to map the invisible and largely inaccessible territory of
language and the human mind. However, most linguists would unite in
accepting that they have theories and are theoretical in their work (but
see Gethin, 1990 for an opposing view).
Perhaps then, the right question to ask is: should applied linguists be
theoretical? One response is that they can hardly not be, that we all bring to
any problem-solving situation a perspective, a set of beliefs or attitudes
that may inform, but are separate from, the decisions we take to resolve
the problem(s) of the moment. This seems an eminently sensible view of
things, but it has its dangers. It could encourage an ad hoc and unreective
process that never learns from experience or to induce from varied cir-
cumstances a philosophy that says my set of beliefs and established
approaches will serve me well in the face of any problem and need not
subject themselves to objective scrutiny nor to constant revision; they are
accountable to no one but myself. There is also the risk that action,
however manifestly successful, that does not or cannot justify itself ex-
plicitly in some set of theoretical postulates is to be frowned upon: this is
the critic that says thats all very well in practice, but what about in
theory?.
This book will take the line that being theoretical is a desirable thing,
but that theoretical stance is more useful as a motto than theoretical
allegiance, akin to what Widdowson (1984:30) refers to as having a theor-
etical orientation. Widdowsons (1984:2127) viewthat applied linguistics
must formulate concepts and theories in the light of the phenomena it is
trying to account for will be valuable as long as it retains its plurality.
Applied linguists must certainly account for, and be accountable to, the
contexts in which they work and the problems with whichthey engage. An
important component of this is not to shy away from stating the beliefs,
claims and attitudes that inform their position on any given applied
linguistic activity, whether it be solving a language-teaching problem or
proposing a socio-political language-planning solution that might have
wide humanitarian implications. This is ones theoretical stance. The obli-
6 Applying linguistics: disciplines, theories, models, descriptions
gation to espouse any particular establishment school of thought or ca-
nonical set of beliefs, claims and postulates consistently over time and
across different situations, may be referred to as theoretical allegiance,
which Widdowson (1980:21) rightly suspects is essentially conformist.
Thus the question What school of thought do you belong to? or What is
your theoretical position? will likely be misdirected if put to an applied
linguist. What is your theoretical stance with regard to this problem or set
of problems? is a question we have every right to ask of our applied
linguist peers. Furthermore, there is a very good reason why stance and
accountability go together: we owe it to our membership of a disciplinary
community to be able to contextualise our particular position in relation
to those of others. In short, the theoretical life-blood of applied linguistics
is not allegiance to theories but is more a commitment to a discourse. This
discourse is the communication of varied positions among peers using a
shared language that enables us to nd common ground with the posi-
tions taken by others already reported and established, and to recognise
when new ground is being broken (see Crystal 1981:10ff). As Lantolf (1996)
puts it: letting all the owers bloom. Thus the rhetoricising of stance, that
is to say rendering it into an organised, communicable and persuasive set
of claims, arguments, illustrations and conclusions is the way in whichthe
community accounts for itself member to member and to the outside
world. Being theoretical and being accountable are two sides of the same
coin. Encountering problems and adopting a convincing stance towards
them is what denes applied linguistics as a discipline.
1.4 Approaching problems in an applied linguistic way
It is now appropriate to open up the relationship between the more
theoretical aspects of language study and how they might be applied in
the language teaching context. I shall begin by considering what avenues
withinlinguistics suggest themselves for approaching two of the problems
relevant to language teaching in the list of 14 above. Let us consider
problem no. 2 in the list: that of the teacher trying to understand why
learners from the same language background are having difculty with a
particular grammatical structure in English. The teachers potential re-
course to linguistics is likely to involve different areas depending on what
questions are asked (see Figure 1).
1.4 Approaching problems in an applied linguistic way 7
What is known about the learners rst

language or any other language they know

which might be interfering with their
learning of the foreign language?
What do grammarians say about this
structure?
Language teachers
questions What psychological barriers might be
preventing the learning of the
structure?
Are some structures difcult to learn
if they are tackled too early on? Is there

an order in which structures are best

presented?
Figure 1: Potential linguistic questions for the solution of a grammatical
problem

What is the internal structure of the vocabulary of

the language(s) I am dealing with?
What do we know about the mental

organisation of vocabulary in human

beings? Perhaps this can be utilised in
Lexicographic dictionary organisation?
(dictionary-making)
questions

What problems might a non-native user of the

dictionary have with the organising principle
chosen?

What place should information about

grammar have in such a dictionary?
Is a bilingual dictionary along non-

alphabetical lines possible?

Figure 2: Potential linguistic questions for the solution of a lexicographic
problem
If we consider another of the problems, that of the dictionary writer
looking for alternatives to the alphabetical dictionary, we might imagine a
different set of questions, as in Figure 2:
The dictionary writer, like the language teacher, confronts the same basic
questions: Can linguistics offer an approach or a solution to the problem
at hand? If so, which branch(es) of linguistic study, and by what method(s)?
8 Applying linguistics: disciplines, theories, models, descriptions
Howreliable is the information offered by linguists? Howtenable are their
theories and models of the language? How willing and ready are linguists
to contribute to this kind of practical undertaking? The title of a paper by
McCawley (1986), What linguists might contribute to dictionary making if
they could get their act together, strikes a slightly pessimistic tone in this
regard. If there is conicting information to be had from the ndings of
linguists, how does one best evaluate which approach is likely to be most
useful? Can the non-linguist take on such a task, or is this a job for highly
trained specialists?
The concern of this book is therefore to raise to the fore a selection of
problem areas in language teaching and learning where knowledge about
language plays or could play a major role, to review what it is that
linguists do, and to consider whether and how their discipline can be
applied, giving as many as possible practical examples of applications. As a
conclusion to the book I shall consider broader ideological issues within
applied linguistics, and how applied linguists have developed and are
developing a sense of a professional community with common interests,
as well as the predictable debates, factions and divisions, uncertainties
and varied positions that characterise any such community, especially one
as loose-knit as that of applied linguists. I shall exemplify across a variety
of languages, even though, inevitably, many examples will centre around
English, because of the historical fact that a large amount of the output of
linguistics and applied linguistics and writing about language teaching
has been based on English, and also because English is the language of this
book. But it is important to offer examples in other languages in order to
underline the universality of the applied linguistic enterprise and the
underlying bond that unites the work of practitioners across the world
working in a variety of language teaching contexts. It is language as a
human phenomenon that we are attempting to understand, in the
hope that we might teach it more effectively in its many manifestations
around the world, and also produce better dictionaries, materials, and
syllabuses, or make improvements in whatever our area of preoccupation
might be.
1.5 Applying linguistics in language teaching: two examples
Before we enter the more detailed chapters on what linguists do, it may be
useful to look more closely at the two examples of linguistics in applica-
tion briey touched on above (Figures 1 and 2) as a template for the overall
1.5 Applying linguistics in language teaching: two examples 9
purposes and goals of this book. I shall therefore take the two examples
and follow them through to two sets of potential applied linguistic con-
clusions.
1.5.1 Example 1. Grammar: Why do they misuse it ?
Many teachers of English as a second or foreign language will be familiar
with errors such as the following in their students written work:
1 A teacher has set an essay entitled Trafc in Cities. An Italian student
writes the title at the top of the page:
Trafc in Cities
And then begins the rst paragraph of the essay:
It is a very big problem nowadays and many cities in the world suffer from
it. . . . etc.
The teacher crosses out the rst it and puts trafc instead.
2 Another student writes an essay about his specialist university subject
construction engineering:
This essay will show the increasing development of the insert of Glulam
(glued laminated timber). It will help to nd the reasons for the present
boom in Glulam structures. For it*, it is interesting to look at the history,
the properties, the manufacturing process and the types of structures
which are possible.
The teacher puts a red mark against the asterisked it and suggests saying
this essay instead of it.
These two learner errors are typical of many which prompt the teacher
to seek some sort of explanation of the problem, both for their own
professional integrity and satisfaction and in order to be able to hand on a
useful rule or principle to the learner. Let us consider what questions the
teacher might pose and the steps that might be followed:
1 What type of problem is this? Is it:
(a) a grammar problem concerning a particularly tricky English gram-
matical choice?
(b) a problem encountered only by speakers of a particular language or
10 Applying linguistics: disciplines, theories, models, descriptions
group of languages, or one encountered by most learners?
(c) a problemfromthat fuzzy area of style, to whichthere is unlikely to
be a clear, satisfactory answer and which one may therefore just as
well forget?
Question 1(a) is not so simple as it may seem. Many linguists understand
the term grammar to be limited to questions of the internal structure of
sentences, and would consider the it problem as it manifests itself in the
student essays to be outside of the purview of the grammarian and
something to do more with pragmatics, the study of how things acquire
meaning in different contexts (see Evans 1980, for instance). This is one of
the consequences of the pronounced theoretical demarcations we often
nd within linguistics. Others might disagree with shunting the problem
out of grammar and into pragmatics, and see this particular problemwith
it as belonging to the recently developed sub-disciplinary area of discourse
grammar. This is a sort of hybrid way of studying grammar by looking at
whole texts and taking contexts into account (see section 5.6; see also
Hughes and McCarthy 1998 for examples and applications to teaching; see
Carter et al. 1995 for further discussion). Therefore, one of the rst and
most important things for the teacher who would be an applied linguist is
to have a good working knowledge of how linguistics is sub-divided and
how the linguistics community makes its decisions as to what to include
in what. Without this knowledge, it will be even more difcult to answer
question 1(b), which concerns whether the problem is likely to be wide-
spread or limited to learners with a particular rst language background.
Question 1(c), whether to consign the problem to the rag-bag category of
style, will also depend to a large extent on whether a satisfactory solution
can be found within studies of sentence grammar, or pragmatics, or
discourse grammar. Then again, the answers to questions 1(a) and 1(b)
need not be mutually exclusive and it may be very benecial to pursue
both. Finally, we may indeed conclude that the problem is a grammatical
one (in terms of the most appropriate label to attach to it), and thus
challenge whether grammarians who place it beyond their purview are
being properly accountable to their audience. In other words, we might
begin to re-theorise the paradigms of grammar from an applied linguists
point of view.
If the teacher decides initially that the it problem is likely to be one of
grammar, then this decision opens up a further set of possible avenues
towards a solution. One set of choices for investigation might be:
1.5 Applying linguistics in language teaching: two examples 11
Rules presented in course books and
reference

books designed for teaching English as a

foreign language
(pedagogical grammars).
Rules presented in grammar books that
simply describe the English language
(descriptive grammars).
Rules offered by theoreticians who create
models for understanding particular areas of
It as a grammar grammar (in this case for the pronoun system,
problem for example) and who report their conclusions
in books and learned journals.
An action research project by the teacher in
which he/she sets up a variety of tests and
experiments and observations to see if the
problem is recurrent, if it can be pinned down
and made more specic (e.g. perhaps it only
occurs at the beginning of essays) and
whether such action research can provide an
explanation without further need for
theoretical investigation.
Figure 3: Paths of investigation in solving a grammar problem

Does the learners rst language have a

grammatical choice similar to but not entirely
overlapping with English it for contexts such
Questions about the as the two example essays?
learners rst
language Does the learners rst language have
quite a different set of grammar rules
for expressing English it and this?
Figure 4: First-language-related questions
Pursuing the problem in terms of question 1(b) (Is it a problem encoun-
tered only by speakers of a particular language or group of languages, or
one encountered by most learners?) raises yet another set of questions:
This assumes that the problem is one to do with the learners rst lan-
guage. Another question might be: Is the learner transferring something
from the rst language (which may or may not be viewed as a positive
strategy), is the rst language interfering in some way (whichwould usually
be interpreted in a negative way), or is it possible that it is not a case of
12 Applying linguistics: disciplines, theories, models, descriptions

Studies comparing and contrasting the learners rst

language grammar with English (examples of contrastive
analysis; see sections 2.42.7).
Grammars of the learners rst language, either those
written to describe that language or those written to
teach it (descriptive or pedagogical grammars).
Resources
Studies of typical transfer and cross-linguistic
interference from the learners language to English, as
reported in learned journals.
Studies of grammatical strategies adopted by learners
(e.g. grammatical simplication) at various levels, as
reported in learned journals.
Figure 5: Cross-linguistic resources for the solution of a grammatical
problem
transfer or interference at all, but perhaps a strategic choice the learner
has made to solve a particular problem (a positive strategy)?
In turn, these questions open up possible paths for exploration:
We can already see that the pathways into doing applied linguistics lead
us into complex elds and a multitude of potential resources, and that the
success of the applied enterprise depends on:
1 Identifying and dening problems.
2 Contextualising those problems within linguistic study and developing
a theoretical stance.
3 Harnessing appropriate resources for the exploration of possible sol-
utions.
4 Evaluating the proposed solutions.
We shall also see later in this book that real-world problems are best not
regarded as divorced from the world outside of the classroom, from the
wider socio-cultural and political contexts in which language learning
takes place. As with all problem-solving activities, the solutions may not
come easily or immediately.
Let us nowpursue further the problems with it in the student essays and
consider what happens if we conclude that we are dealing with a gram-
matical problem concerning a rather subtle or difcult choice within
1.5 Applying linguistics in language teaching: two examples 13
English grammar. Our rst and most direct resource might be the
coursebooks and other books in use in the classroom. It is likely, though,
that we shall nd it dealt with under the pronouns of English, where it is
contrasted with he and she in relation to human or non-human entities.
This is also likely to be so in grammar reference books designed for
learners, but the better and more detailed ones may also point to the use
of it in contrast to possible choices such as this and that, as does this extract
from Alexander (1988). Alexander gives us the following rule:
Subject pronouns replacing demonstratives
Demonstratives are replaced by it or they in short responses when
the thing or things referred to have been identied:
Is this/that yours? Yes it is (Not * Yes, this/that is)
Note: An asterisk (*) before a stretch of quoted language indicates
an incorrect or inappropriate form.
This illustration may offer a partial solution to the problem, in that it
seems to suggest the possibility that it cannot be used to refer to things not
already identied, and this principle could perhaps be extrapolated to the
student essays. At this point we are evaluating a linguistic statement,
rather than simply taking it on board wholesale, which is perhaps the
most crucial phase of all in doing any kind of applied linguistics.
However, the evaluation may well be that the concept of things not
already identied is not a very useful (or teachable) one. In both the
examples of errors in student essays, the thing being talked about cer-
tainly seems to have been identied (trafc in the rst case and the
present essay in the second). We might therefore search further aeld
than pedagogical grammars such as Alexanders to nd a more satisfac-
tory solution. One likely area would be the considerable journal literature
onstudent essay-writing whichhas grown up around the college composi-
tion tradition in the United States. Articles within the college composi-
tion eld do indeed treat such apparently puzzling areas as pronoun and
demonstrative usage (e.g. Moskovit 1983; Geisler et al. 1985). When we nd
such studies (either by manually searching indexes or doing key-word
computer searches on electronic media such as CD-ROM bibliographies or
on-line bibliographical services), we see how they, in their turn, draw on
wider areas such as the study of writing as communication, text- and
discourse analysis, and the study of reading. In the case of pronouns versus
14 Applying linguistics: disciplines, theories, models, descriptions
demonstrative and/or full noun phrase, we nd writers such as Hofmann
(1989) and Fox (1987a and b) having recourse to notions such as text
boundaries, segments, topics and focus in the development of the text, rather
than sentences or the identication of things in the real world (see also
McCarthy 1994a). These terms are not the familiar ones of sentence gram-
mar, then, but belong to the world of discourse grammar and text analysis.
What is crucial, it seems, is not so much whether something can be
identied in the text, but what its status is as a topic in the text from the
viewpoint of the interactants (i.e. writer and reader or speaker and lis-
tener): Is it the current topic? Is it a secondary or marginalised topic? Is
there potential ambiguity or confusion as to what the current topic is?
These are quite different questions from: Is it third person? Is it human or
non-human?
In the rst student essay (on trafc) it seems that crossing the gap from
the title to the main text disallows the use of the topic-continuing
pronoun it, and linguists have indeed argued that the it pronoun may not
be able to refer back to something separated by a textual boundary such as
a paragraph division(e.g. Fox 1987a and b). In the second essay, the use of it
in the phrase for it seems to create confusion as to what we are actually
focussing on at that precise moment: is it glulam or the essay itself ? In
other words is this use of it a typical grammar problemof reference or one of
the structuring of information within the textual world shared by writer and
reader(s)? Linguistic descriptions that offer no insight into what seems to
be a crucial distinctionmay be less than useful for the practitioner seeking
an answer to this particular set of problems.
One or two papers on college composition may not, in themselves, be
enough to offer a convincing and generalisable solution to the pronoun
problem, and the teacher doing applied linguistics may feel the need to
explore further in text- and discourse analysis, or may decide to gather
more data from learners. In addition, even if the teacher feels that a
satisfactory explanation is available, there will still be the problem of how
to fashion it into a point for teaching and learning, i.e. the problem of
methodology, which will largely remain outside of the remit of this book.
However, implicit in what this book describes will always be the belief that
teaching methodologies and descriptions of languages should interact to
produce good teaching (i.e. that accountability should not end between
linguists and academic applied linguists, but should apply between all
groupings within the language teaching profession). Good descriptions
1.5 Applying linguistics in language teaching: two examples 15
and practical guidelines should inuence methodology and methodologi-
cal developments should inuence the quest for better description and
more accessible guidelines for learners.
1.5.2 Example 2. Lexicography: the case of the bilingual thesaurus
Let us turn to another problem mentioned at the outset of this introduc-
tory chapter: that of the lexicographer trying to develop an alternative to
the traditional, alphabetical bilingual dictionary. Alphabetical diction-
aries are useful if the user already knows the word in the target language
or has a word in his/her own language to look up. But what if one only has
a vague idea of what one wants to say, i.e. that one has a meaning oating
round in the mind, but no words whereby to access it, either in the rst or
the target language? Among the resources available in such a situation
will be thesauruses and word-nders of various kinds, and dictionaries of
synonyms and antonyms. These types of reference works depart from
purely alphabetical organisation and bring words together on the same
page according to notions of meaning rather than their orthographical
(written-alphabetical) form. The classic model for such organisation is
Rogets Thesaurus (Roget 1852). Roget brought words together according to
their role in describing a philosophically organised world, a model almost
Aristotelian in character (Kjellmer 1990), where the taxonomies of the
natural and human world are reected in an orderly vocabulary. And yet
we react with mild amusement when we note that Roget included the
word stomach under the category container (along with boxes and baskets);
somehow, Rogets classication often seems remote from commonsense,
everyday meanings and how words relate to one another.
The lexicographer in search of alternatives for organising the vast
meaning-stock of any language has available a range of semantic and
cognitive models of meaning. If the thesaurus is, in addition, to be bilin-
gual, then a model which permits the mapping of one languages mean-
ing-stock onto another with all the problems of lack of one-to-one t
which that entails will be a desirable basis from which to work. In other
words, a merely descriptive list of words for each of the two languages in
question will not be enough; it is the model that underlies the description
that is crucial.
The lexicographic problems difference from the grammatical one (that
of students misusing it) is only one of degree. Even though a satisfactory
answer may have been forthcoming frompedagogical or descriptive gram-
16 Applying linguistics: disciplines, theories, models, descriptions
[who has the rst or lowest academic degree]
(human) [young knight serving under
(male)
the standard of another knight]
bachelor noun
[who has never married]
(animal) (male) [young fur seal when without a
mate during the breeding time]
Figure 6: Katz and Fodors description of bachelor

mars, they in their turn presuppose some model or underlying theoretical
view of how grammar functions, whether it be that sentence-level syntac-
tic structures lie at the core, or whether a more context-sensitive, discour-
sal model is presupposed. Subsequent chapters of this book will explore
these competing claims. In the case of thesaurus design, the lexicographer
is not unlike the grammarian designing a grammar: the key question is
What is the model of language and meaning which will drive the or-
ganisational structure of the thesaurus?. In other words, what theoretical
stance(s) may be adopted to solve the problem? Though this would seemto
place the lexicographer on a higher plain in the applied linguistics rma-
ment than the teacher looking for a solution to a problem of pronoun
misuse, this book does not take that line. The teacher applying a gram-
matical description is doing applied linguistics just as much as the lexi-
cographer applying a model of word-meaning; they are simply working in
different ways.
The various models of meaning offered by linguists all have some
attraction for the lexicographer. For example, Katz and Fodors (1963)
inuential notion of decomposing words into their semantic properties,
epitomised in their description of the meaning(s) of bachelor in English (see
Figure 6), would seem to offer a possible basis for mapping words in
different languages onto one another.
But there is a great deal of semantic overlap and grading in meaning
withinfamilies of related words, and Katz and Fodors technique turns out
to be severely limited for the lexicographer working with thousands of
headwords in a dictionary or thesaurus. The approach to meaning based
on such a notion of componential analysis has been superseded in lin-
guistics by other models of meaning, as we shall see, amongst which
1.5 Applying linguistics in language teaching: two examples 17
the lexicographer might gain insight from frame-theoretical approaches.
In frame theory, the sharp distinction between what we know about
language and what we know about the world is broken down (Lehrer
1993), enabling the lexicographer to include socio-cultural information
within the meaning of a word (see also Schmid 1993). Such a broader-
based model of meaning may well provide a more practical basis for the
construction of a bilingual thesaurus and the mapping of two linguistic
cultures onto one another in a commonsense and intuitively more satisfy-
ing way.
At this point I permit myself to exemplify the applied linguistic outcome
from one of my own published works. McCarthy (1995), in a bilingual
thematic (thesaurus-type) dictionary for Italian learners of English, at-
tempts to map English words connected with poverty onto Italian words
and expressions in the same frame (see Figure 7). In addition to semantic
equivalences, the learner is given circumstantial information that is cru-
cial to distinguishing use, as well as advice on appropriate collocations.
The particular frame embraces adjectives, nouns, verbs and xed expres-
sions. The dictionary entry was constructed from a base English list of
poverty words, andtranslated into Italian by a teamof experts withnative
speaker command of both languages. The experts includedall the informa-
tion which would, theoretically at least, enable the Italian user to distin-
guish accurately among the possible English candidates for an Italian
meaning connected with poverty which the user might wish to word in
English. The extra information beyond the pure semantics includes de-
grees of formality, the contexts in which each word normally occurs (e.g.
bankrupt versus destitute), and the word set includes words such as beggar
and beg, whichare roles and actions that have a real-world association with
poverty. In addition to the thematic grouping, any of the words can be
accessed in Italian or English in the alphabetical index, thus enabling the
resource to be used either as an alphabetical bilingual dictionary or via the
overall theme, as a tool when the learner has a meaning in mind but no
clear words as a starting point. The thematic dictionary is as imperfect and
awed as any other enterprise, and I present it here simply as an example
of a product that began with a problem. The solution involved an applied
linguistic process of starting with the learner (How can he/she get to an
English word starting only from a vague notion of a desired meaning?),
proceeding to the application of a relevant theoretical model (frame
theory), and producing the goods (the dictionary). Its users will be the only
proper evaluators of its success or failure as a piece of applied linguistics.
18 Applying linguistics: disciplines, theories, models, descriptions
Figure 7: Entry for poor in an English-Italian thematic dictionary
(McCarthy 1995)
1.6 Conclusion
One nal important area must be addressed before I embark on the rest of
this book, for which we need to return to the question of who, precisely,
applied linguists are. In section 1.1, I spoke of applied linguists in univer-
sity academic departments, but distanced myself from equating only
those people with the title applied linguist or with the notion of doing
applied linguistics. This is important, for the temptation to ring-fence
applied linguistics within the academic community leads inexorably to a
gulf of suspicion between academics (whether linguists or applied lin-
guists) and practising language teachers out there at the chalkface. Kirby
(1991) speaks of a growing chasm which separates theoreticians from
practitioners and an end of the honeymoon (a reference to a paper on the
subject by Lennon 1988). One of the central problems Kirby identies is the
feeling that applied linguistic research does not address the practical
needs of teachers, and much of what he says cannot be denied. But the
solution that applied linguists (in the academic sense) and theoreticians
must become more sensitive to the needs of language teachers is only half
a solution: the position this book takes is that non-academic teachers
should become applied linguists, not just look to them for guidance. Only
when the community of applied linguists itself becomes a broader church
1.6 Conclusion 19
will the problems of the current uneasy relationship be able to be properly
addressed and moved towards solutions satisfactory to all parties. That is
why this book is aimed at language teachers and other language practi-
tioners, not just applied linguists in the academic sense of the term. If it
can only speak to this last group, then it has failed.
What I have tried to do in this introduction in considering two quite
different language-teaching problems and how they may be solved by
having recourse to aspects of linguistics, is to emphasise the multi-faceted
nature of applied linguistics, even in just one of its professional branches,
that of language teaching and learning, and to begin to explore the
various levels on which problems may be tackled. In the rst case (the
grammatical problem) I stressed the potential of linguistic description, that
is the sets of observable facts about languages that linguists can offer. In
the second case (the lexicographic problem), I stressed the modelling of
language, that is theoretical constructs that help us to understand how
languages (might) work. Behind models lie theories the mental explora-
tions, speculation and argumentationthat go to build a set of ideas, beliefs
or principles about language. Linguists are in some sense inevitably in-
volved in all three of these activities, though some eschew description of
actual language use, for example early exponents of transformational-
generative grammar (see section 3.3.2), while others would argue that
only looking at real language in use is the proper starting point on the
long journey to a theory of language (e.g. Sinclair 1991; see also Chapter 5).
Most prefer to move in both directions: the good applied linguist not only
starts from day-to-day practical problems and looks for solutions in de-
scriptions, models and theories of language, but also develops his or her
own models and theoretical stances. Behind these there usually develops a
guiding set of beliefs about language, however rooted in practical con-
cerns and however scornful non-academic applied linguists may occa-
sionally be of those for whom language seems to be an abstract, rather
than a concrete, object. The examples we have looked at and the typical
procedures followed to get to the roots of the problems have been peda-
gogical ones, but essentially the same questioning must take place in the
mind of any applied linguist who tries to locate his or her particular set of
problems within the vast array of linguistic theories and descriptions.
We thus travel in this book across a landscape strewn with different
theories, models and descriptions and attempt to build up the complex
picture that is present-day applied linguistics with reference to language
teaching and learning. The book will consider the description of sounds,
20 Applying linguistics: disciplines, theories, models, descriptions
words, and grammars, the modelling of how we communicate and create
texts, how the mind processes language, and theories of what language is
and how those theories shape our day-to-day perceptions and actions as
language practitioners. It will also be concerned with how applied lin-
guists engage in discourse with one another and construct their common
language and professional identity. No one level of activity will be con-
sidered privileged, and the interrelationships between levels of applied
linguistic activity will inform the argument throughout.
The lack of a monolithic denition of applied linguistics, the lack of
unitary theory and of clear disciplinary boundaries will be regarded as a
positive characteristic of the discipline, its very openness to outside in-
uences being its strongest and most enduring quality, and one that has
served it well over the decades that the term applied linguistics has had
currency.
1
All this will take place against the background of a belief that
applied linguists and linguists alike owe accountability to one another,
principally through the fruits of their work, and that the cornerstone of
such accountability is uent and non-obfuscating communication be-
tween the partners in the task of making social sense of phenomena
connected with individual languages and language as a whole.
Notes
1 Exactly when the term applied linguistics came to be established is not clear.
The term linguistics goes back to the middle of the nineteenth century, al-
though the beginnings of scientic linguistics properly go back further (see
Lepschy 1982). The use of applied in the sense of practical applications of
sciences can be dated back to at least the middle of the seventeenth century.
Howatt (1984) looks back to Henry Sweet (18451912) as applying living philol-
ogy, though Howatt dates the rst public use of the term applied linguistics to
1948.
Notes 21