System 31 (2003) 13–28

www.elsevier.com/locate/system

CALL—past, present and future

Stephen Bax
Department of Language Studies, Canterbury Christ Church University College, Canterbury,
Kent CT1 1QU, UK

Received 26 February 2002; received in revised form 28 June 2002; accepted 26 August 2002

Abstract
This article offers a critical examination and reassessment of the history of CALL, and
argues for three new categories—Restricted, Open and Integrated CALL. It offers definitions
and description of the three approaches and argues that they allow a more detailed analysis of
institutions and classrooms than earlier analyses. It is suggested that we are currently using
the second approach, Open CALL, but that our aim should be to attain a state of ‘normal-
isation’ in which the technology is invisible and truly integrated. This state is defined and
discussed. In the final section the article proposes some ways in which this normalisation can
be achieved—using ethnographic assessments and action research, for example—thus setting
an agenda for CALL practice in the future.
# 2003 Elsevier Science Ltd. All rights reserved.
Keywords: CALL; ICT; Technology; Language teaching; Language learning; Integration

This article seeks answers to three questions: where has Computer Assisted Lan-
guage Learning (CALL) been, where is it now and where is it going? It starts from
the premise that if we are to maximise the benefits of CALL in future, we need to
establish an agenda and a set of aims towards which to work. This in turn requires
an analysis of where CALL has been in the past and where it is now—the future
must learn from the past and present. Accordingly, it is structured chronologically,
beginning with a reassessment of the history of CALL, followed by a brief assess-
ment of CALL now. It concludes with a detailed discussion of where CALL could
go in future and how we as a profession could help it achieve its maximum potential.

E-mail address: [email protected] (S. Bax).

0346-251X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
PII: S0346-251X(02)00071-4
14 S. Bax / System 31 (2003) 13–28

Part 1: where has CALL been?

In his attempt to answer the first of these questions, Delcloque (2000) refers to five
main literature sources and divides them into essentially two kinds:

1. The properly researched, objective historical accounts which attempt to

summarize the progression and might include precise dates and a comprehen-
sive list of sources.
2. The interpretative type which tends to draw more subjective conclusions
about advances and trends in the field, thus analyzing its progression in a less
objective manner. (Delcloque, 2000)

It is clear from Delcloque’s tone that he prefers accounts of the first type for their
supposed ‘objectivity’—he considers Levy (2000) and Ahmad et al. (1985) to represent
type 1 (as is his own), while Warschauer and Healey’s (Warshauer and Healey, 1998;
Warschauer, 2000) is considered to represent type 2. This attitude represents a depress-
ingly limited view of the nature and role of history. If we are to use our past experience
not only to know what has happened in CALL but also to understand it, then any history
surely needs to be more than a mere recounting of supposed facts concerning software
and hardware developments. It needs to offer analysis of what occurred and also why,
particularly with regard to the actual use of computers in language teaching and learning.
I contend, then, that CALL needs something far closer to Delcloque’s type 2 history.
It is remarkable, in fact, that there exists no in-depth analysis of the history of CALL
along the lines of, say Howatt’s history of ELT (1984), and it is arguable that without it
we cannot properly formulate an agenda for the future use of CALL. Of the main
accounts mentioned by Delcloque, the first (Ahmad et al., 1985) focuses almost exclu-
sively on the factual aspects of technologies—many acronyms but little analysis; the
CALICO study (Sanders, 1995) relates only to North America and is again weighted
towards ‘facts’ rather than analysis, while Levy’s (2000) account admits to being only a
review, rather than an in-depth discussion. Chappele (2001) is similar. Other reviews of
the history of CALL—too numerous to consider in detail here—do exist but are for the
most part of the ‘review’ type, with insufficient attempt at critical analysis.
This—remarkably—leaves Warschauer and Healey (Warshauer and Healey,
1998; Warschauer, 2000) as the only substantive, systematic attempt to analyse and
understand the history of CALL in anything more than ‘factual’ terms. This is
regrettable—CALL surely needs to look in far more depth at its past and present in
order to develop a fully appropriate role for computers in language teaching and
learning. For this reason this article will revisit the history of CALL, and will take
Warschauer’s categories as the starting point, since they are the best available, as
well as being familiar to many in the profession.

The three phases of CALL

Warschauer and Healey famously identifies three phases of CALL which he terms
‘‘Behaviouristic’’, ‘‘Communicative’’ and ‘‘Integrative’’ (Warschauer and Healey,
 S. Bax / System 31 (2003) 13–28 15

1998). He has discussed these phases of CALL in various places over the years, but
has recently produced a summary reproduced in Table 1 (Warschauer, 2000), which
we can use as a reference point. Despite the popularity of his analysis, this frame-
work seems to me to have a number of significant weaknesses, which we can discuss
briefly in turn.

Inconsistencies

Warschauer’s discussions of the phases of CALL show significant differences in

different publications—for example, Structural CALL was previously called Beha-
viouristic CALL (e.g. Warschauer and Healey, 1998). In the table reproduced here
(2000) it is dated as 1970s–1980s, whereas previously it was described as ‘‘conceived
in the 1950s and implemented in the 1960s and 1970s’’ (Warschauer and Healey,
1998). Likewise, communicative CALL—here dated to the 1980s and 1990s—was
previously said to have ‘‘emerged in the late 1970s and early 1980s’’ (War-
schauer and Healey, 1998). This slippage occurs with ‘Integrative CALL’ also,
which in the table is dated to the 21st century, while in 1998 it was said to be
already in existence.
Though these inconsistencies are not particularly important in themselves, they
are peculiar, and avoidable. A more through historical analysis should surely
attempt greater consistency in terms of chronology.

Table 1
Warschauer’s three stages of CALL

Stage 1970s–1980s: 1980s–1990s: 21st Century:

Structural Communicative Integrative
CALL CALL CALL

Technology Mainframe PCs Multimedia and

Internet

English-teaching Grammar- Communicate Content-Based,

paradigm translation and [sic]language ESP/EAP
audio-lingual teaching

View of language Structural Cognitive Socio-cognitive

(a formal (a mentally (developed in
structural constructed social
system) system) interaction)

Principal use of Drill and practice Communicative Authentic

computers exercises discourse

Principal objective Accuracy And fluency And agency

(Warschauer, 2000).
16 S. Bax / System 31 (2003) 13–28

What are these phases?

Throughout Warschauer’s discourse these categories are considered ‘phases’, with

rough dates attached. However, he also offers disclaimers as to the historical validity
of these phases. For example:

The three stages mentioned above do not fall into neatly contained timelines.
As each new stage has emerged, previous stages continue. Current uses of
computers in the language classroom correspond to all three of the paradigms
mentioned above. (Warschauer and Healey, 1998)

This seems ambiguous—are they historical phases or are they not? If they are
historical phases, how is it that all three coexist together today? And if they are not
in fact closely related to historical periods, then it is surely unwise to speak of them
as phases at all and to attach dates to them. In some places, furthermore, War-
schauer speaks of these three categories as ‘paradigms’ or ‘perspectives’ (e.g. War-
schauer)—which only adds to the conceptual confusion. This, then, is a second
aspect of the analysis which needs clarification.

Unclear criteria

Of the three categories, the first, Behaviouristic CALL, is perhaps the most plau-
sible and would attract most agreement. But the other two categories are far less
satisfactory. In the first place, language teaching in general still operates today very
much within a communicative framework in many teaching contexts, so it is confusing
for teachers to hear that ‘communicative CALL’ is no longer with us—the use of the
term itself is not helpful. Secondly, it is not at all certain that the term ‘communicative’
is being used as language teaching methodologists would use it. Warschauer refers to
Underwood (1984) for his definition, where it is suggested that communicative CALL

* focuses more on using forms rather than on the forms themselves;

* teaches grammar implicitly rather than explicitly;
* allows and encourages students to generate original utterances rather than just
manipulate prefabricated language;
* does not judge and evaluate everything the students do nor reward them with
congratulatory messages, lights, or bells;
* avoids telling students they are wrong and is flexible to a variety of student
responses;
* uses the target language exclusively and creates an environment in which using
the target language feels natural, both on and off the screen; and
* will never try to do anything that a book can do just as well. (Underwood, 1984)

This list is curious for a number of reasons—for one thing, there is no mention of
communication at all, which is obviously central to communicative language teach-
ing (CLT). Although the list includes aspects which are certainly part of CLT, and
 S. Bax / System 31 (2003) 13–28 17

has a general communicative ‘flavour’, without the central features of human

communication and interaction it would be difficult to term this ‘communicative
CALL’ in any useful sense.
More significantly, this is no more than Underwood’s wishlist—it is obviously mis-
taken to define a historical phase on the basis of wishes alone, so if we want to establish
the validity of the view that the 1980s represented ‘communicative CALL’, we need to
explore actual practice at the time. Warschauer reports that during the ‘commu-
nicative CALL phase’ there were three main uses or ‘models’ of computer use:

1. First, there were a variety of programs to provide skill practice, but in a non-
drill format. . .. . . In these programs, like the drill and practice programs men-
tioned above, the computer remains the ‘‘knower-of-the-right-answer’’. . .. . .; thus
this represents an extension of the computer as tutor model. But—in contrast to the
drill and practice programs—the process of finding the right answer involves a fair
amount of student choice, control, and interaction.

This is certainly less restricted than previous software, but how is it ‘commu-
nicative’ exactly? If we take as a benchmark two notions criterial to CLT, namely
that learners learn language in order to communicate, and that they probably learn
to communicate best through the process of communication itself (Littlewood,
1981), then the examples given by Warschauer may have had useful roles in lan-
guage learning, but are not noticeably communicative except in very limited com-
puter–student communication. Furthermore, the actual interaction involved was
extremely limited at the time. The second aspect of computer use in ‘Communicative
CALL’ is described as follows:

2. In addition to computer as tutor, another CALL model used for communicative

activities involves the computer as stimulus. . .. . . In this case, the purpose of the
CALL activity is not so much to have students discover the right answer, but
rather to stimulate students’ discussion, writing, or critical thinking.

Here there certainly may be realistic and valuable communication, but this is not
an inherent feature of CALL itself, nor is it particular to this phase. It could be
achieved by other means, without a computer at all, so it hardly counts as evidence
for this period being a communicative phase of CALL. The third area is this:

3. The third model of computers in communicative CALL involves the computer

as tool. . .. . . or, as sometimes called, the computer as workhorse. . .. . . In this role,
the programs do not necessarily provide any language material at all, but rather
empower the learner to use or understand language. Examples of computer as tool
include word processors, spelling and grammar checkers, desk-top publishing
programs, and concordancers.

This, though useful, is clearly not particular to CLT. The computer acts as a tool
for manipulating language or analysing it, not for communicating in it.
18 S. Bax / System 31 (2003) 13–28

This suggests that the CALL software itself during this ‘phase’ in fact had very
little to do with realistic communication at all. This suspicion is reinforced by exam-
ining literature from the period (e.g. Davies, 1982; Sanders and Kenner, 1983;
Meinhoff, 1990; Kenning and Kenning, 1990) as well as from anecdotal evidence from
those of us involved in CALL at the time. So if the two key aspects of CLT, namely the
notion that learners learn in order to communicate and that they may learn to do this
best through the process of communication itself, are apparently absent from so-called
Communicative CALL, this must raise the suspicion that the term ‘communicative’
may not be appropriate to the software or the uses of CALL during the 1980s at all.
The case for this phase being actually ‘communicative’, other than in the wishes
and ambitions of its proponents, is therefore far from proven. Particularly unclear is
whether this phase is to be defined according to the wishes of its proponents, or to
the software itself, or to the use of the software in class—the first area seems irrele-
vant, the second and third are by no means clearly communicative as we usually
understand the term. During the period which Warschauer has termed ‘commu-
nicative’ (mostly the 1980s) I would argue that there was relatively little pedagogi-
cally useful communication going on in CALL, unless the teacher was unusually
inventive in taking students away from the technology to get them communicating.
This is furthermore confirmed by Warschauer’s own analysis:

by the end of the 1980s, many educators felt that CALL was still failing to live
up to its potential. . .. . . Critics pointed out that the computer was being used in
an ad hoc and disconnected fashion and thus ‘‘finds itself making a greater
contribution to marginal rather than to central elements‘‘ of the language
teaching process (Kenning and Kenning, 1990, p. 90). (Warschauer, 1996)

Warschauer uses this to explain why ‘Communicative CALL’ gave way to his
third phase, but we could equally take it as evidence that communicative CALL in
the 1980s was never actually communicative at all in any significant way. I would
argue that this should prompt us to re-evaluate the phases altogether, both in their
nomenclature and in their dating—if the 1980s show little evidence of commu-
nicative CALL in practice, we need either to reassess Warschauer’s dates, or rename
the phase, or completely revise the analysis.

Integrative CALL

Most doubtful of all, however, is the third category—‘Integrative CALL’. War-

schauer suggests that in the late 1980s and early 1990s

[m]any teachers were moving away from a cognitive view of communicative

teaching to a more social or socio-cognitive view, which placed greater empha-
sis on language use in authentic social contexts. (Warschauer and Healey, 1998)

This is a doubtful assertion. It would be difficult to adduce evidence that teachers

at the time had a ‘cognitive view of communicative teaching’, or that teachers had
 S. Bax / System 31 (2003) 13–28 19

previously failed to think of ‘language use in authentic social contexts’. On the

contrary, the use of language in authentic social contexts had surely been stressed
from the very beginnings of CLT, and taken on board by some if not all teachers,
and it is certainly a central part of CLT today— to imply that it is somehow ‘post
communicative’ is odd.
Warschauer and Healey go on to enumerate a number of approaches which he
claims to be ‘integrative’:

Task-based, project-based, and content-based approaches all sought to inte-

grate learners in authentic environments, and also to integrate the various skills
of language learning and use. (ibid)

Given that these are all used in CLT classrooms today, how can we accept the
suggestion that these are not communicative but ‘integrative’? The value of inte-
grating and integrative skills was recognised at least as long ago as the early 1970s
(see e.g. Oller, 1972) and was a fundamental part of communicative approaches
through the 1980s (Mitchell, 1994). So it is difficult to accept the suggestion that these
features marked any new departure in language teaching, with or without computers.
The other aspect of integration which Warschauer and Healey mentions is the
place of computers in the syllabus:

In integrative approaches, students learn to use a variety of technological tools

as an ongoing process of language learning and use, rather than visiting the
computer lab on a once a week basis for isolated exercises (whether the exer-
cises be behavioristic or communicative). (ibid)

This, I agree, would be integrative. However, Warschauer and Healey’s discussion

implies that this was part of computer use in the 1990—though again, evidence is
difficult to come by. I shall discuss this in some detail below, but suffice it to say here
that the evidence that CALL use changed significantly at some point in the 1990s, so
as to warrant a new label of ‘integrative’, is contentious to say the least, and calls for
more support. I suggest that the ‘once a week’ model still prevails in most institu-
tions throughout the world.
In short, then, the ‘integrative phase’ is also suspect. If it is defined on the basis of
approach to language and language teaching, then it is indistinguishable from
mainstream CLT as it has developed; if it is defined on the basis of the use of com-
puters in the syllabus or in classroom practice, then I suggest that there was no
actual change at all. If it is defined as a new hope or ambition for CALL, then there
may be some validity in the category—but that is hardly a sufficient criterion for
suggesting that a new historical phase of CALL was born at a particular time.

The need for a new analysis

Warschauer and Healey’s analysis has proved a useful way over several years of
conceptualising the development of CALL, and—as I noted earlier—he deserves
20 S. Bax / System 31 (2003) 13–28

credit for his rare attempt to offer such analysis in the first place. However, as we
have seen earlier, the formulation which he has proposed needs clarification and
amendment in a number of areas:

1. it is not clear whether the phases represent clearly defined historical periods or
even whether they are supposed to;
2. the validity of the characterisation of the 1980s as part of ‘Communicative
CALL’ requires more support and tighter reference to mainstream CLT
methodology if the term is to be acceptable, as well as clarification of whether
we are evaluating aims or use of software or software itself, or some combi-
nation. It seems more satisfactory to rename that phase as it relates to that
historical period;
3. the rationale for identifying a third phase, and then calling it ’integrative’,
calls for more support- in terms of attitude to language and language teaching it
has not been clearly enough distinguished from communicative approaches,
while the implied claims for actual integration of CALL into the syllabus and
classroom practice require more support and, as I shall argue below, are doubtful.

It would seem necessary, as a consequence, to formulate an alternative vision of

the history of CALL, one whose terminology is less confusing, and whose categories
seem to fit better with the historical progression of CALL software, approach and
practice. In addition, it should allow us to see clearly where CALL stands at the
moment and where it can usefully go in the future. This is a large undertaking, and
cannot be completed here; however, it is possible to sketch out how such an alter-
native vision might be framed.

Towards an alternative analysis of CALL

In my analysis I shall not refer to ‘phases’—which implies a greater historical

validity than is warranted—but to more general ‘approaches’. I call the first
approach ‘Restricted CALL’. In terms of its historical period and its main features it
differs little from Warschauer and Healey’s ‘Behaviourist CALL’, as can be seen in
Table 2, but the term ‘Restricted’ is more satisfactory since it allows us to refer not
only to a supposed underlying theory of learning but also to the actual software and
activity types in use at the time, to the teachers’ role, to the feedback offered to stu-
dents and to other dimensions—all were relatively ‘restricted’, but not all were
‘behaviourist’. The term is more comprehensive, more flexible and therefore more
satisfactory as a descriptor.
The key dimensions—theory of learning, software, activity types, teacher’s role
and so on—can all be seen in the first row of Table 2. The ways in which Restricted
CALL fits these dimensions can then be seen in row two—for example, it sees the
teacher’s role as being restricted to monitoring, the feedback restricted to closed
responses and so on.
Row three shows the second approach, which I term ‘Open CALL’, since it is rela-
tively open in all dimensions—from the feedback given to students, to the software
Table 2
Restricted, Open and Integrated CALL: an outline
Content Type of task Type of student activity Type of feedback Teacher roles Teacher attitudes Position in curriculum Position in lesson Physical position of
computer

Restricted CALL
Language system Closed drills Text reconstruction Correct/incorrect Monitor Exaggerated Not integrated into Whole CALL Separate computer lab
Quizzes Answering closed fear and/ syllabus—optional extra lesson

S. Bax / System 31 (2003) 13–28

questions or awe
Minimal interaction Technology precedes
with other students syllabus and learner
needs
Open CALL
System and skills Simulations Interacting with the Focus of linguistic Monitor/ Exaggerated fear Toy Whole CALL Separate lab—perhaps
Games computer skills development facilitator and/or awe Not integrated into lesson devoted to languages
CMC Occasional interaction Open, flexible syllabus—optional extra
with other students Technology precedes
syllabus and learner
needs
Integrated CALL
Integrated language CMC Frequent interaction Interpreting, Facilitator Normal part of Tool for learning Smaller part of In every classroom,
skills work WP with other students evaluating, Manager teaching—normalised Normalised every lesson on every desk,
Mixed skills e-mail Some interaction with commenting, integrated into syllabus, in every bag
and system computer through stimulating adapted to learners’ needs
the lesson thought Analysis of needs and
context precedes decisions
about technology
Any, as appropriate
to the immediate
needs

21
22 S. Bax / System 31 (2003) 13–28

types, to the role of the teacher. Arguably, it is not completely open, but at least its
main characteristic in comparison with Restricted CALL is its relative openness in
these dimensions.
The fourth row shows the characteristics of ‘Integrated CALL’ (not integrative, as
in Warschauer and Healey’s formulation). The key point about Integrated CALL—
which sharply distinguishes it from Warschauer and Healey’s—is that it does not yet
exist to any significant degree, but represents instead an aim towards which we
should be working. I shall argue later that at this moment in the historical develop-
ment of CALL we are still operating within the second approach, Open CALL, our
aim being to move towards Integrated CALL in future. I shall indicate in the later
stages of this article how this might be achieved.

Are these historical phases or not?

In general, my three approaches do coincide with general historical periods—

Restricted CALL dominated from the 1960s until about 1980; Open CALL has
lasted from the 1980s until today, with some Restricted CALL manifestations still
observable and still valuable in their place (e.g. in grammar revision and checking).
Integrated CALL exists in a few places and a few dimensions only, but is far from
common, as I shall argue below. It is therefore possible to use this analysis as a guide
to broad historical developments in CALL
However, this classificatory framework also offers a number of other benefits
when contrasted with previous analyses:

1. The terminology prevents conceptual confusion with behaviourist or com-

municative approaches to learning or teaching [Warschauer and Healey’s did
allow this confusion—even Delcloque mistakenly called the first phase
‘Behavioural’ (Delcloque, 2000)].
2. The classification is, I suggest, more accurate as a description of what hap-
pened in the past and is happening now.
3. The framework allows us to define our practice in some detail. For example,
we might find that an institution is Restricted in some aspects, Open in others
and Integrated in others, giving a clear idea as to its practice in all key
domains of CALL practice.

For this reason I shall use these alternative descriptors in what follows.

Part 2: where is CALL now?

Open CALL

In the case of Open CALL, we can see that from around 1980 there was a gradual
awareness that previous approaches had indeed been Restricted, and that new
approaches were needed. In this sense, attitudes to using computers were more open
 S. Bax / System 31 (2003) 13–28 23

(as can be seen from Underwood’s list reproduced earlier) and were certainly
becoming more humanistic (cf. Stevens, 1992) but mostly owing to technological
limitations related to hardware and software it was not possible to use computers
for realistic communication in a CLT vein until the advent of effective CMC, the
web, widely available email and so on. (see. e.g. Sanders and Kenner, 1983, for
contemporary discussion.)
Nowadays, however, it is indeed possible to use computers for genuine
communication, as we can see in discussions such as Motteram (2000). It would
therefore be possible to argue for a more genuinely ‘communicative’ role for CALL
from around 1995 onwards, at least in terms of software. However, this Open aspect of
the technology and software is by no means matched by an Open attitude in other key
areas of implementation such as teachers’ attitudes, administrators’ attitudes and
timetabling. Furthermore, much software being produced today is still of a relatively
Restricted type. For this reason, we could argue that in general terms we are in an
Open phase of CALL, but that each institution and classroom may also exhibit certain
Restricted and even Integrated features. In terms of true integration of CALL within
language teaching and learning (which has been a goal for decades—see e.g. Sanders
and Kenner, 1983), we are still a long way from achieving it, and it is important
therefore to start to reconsider how the profession can move towards that general aim.

Part 3: where is CALL going? Integrated CALL and normalisation

In order to understand the extent to which CALL is truly integrated into a class-
room or into an institution or into a particular teacher’s practice, we can draw use-
fully on research into the diffusion of innovations (e.g. Rogers, 1995), looking at
how an innovation comes to be accepted and effective in its new domain. However,
let us start by identifying an end goal for CALL. In my view the clearest way of
defining this goal is through the concept of ‘normalisation’ (see Bax, 2000). This
concept is relevant to any kind of technological innovation and refers to the stage
when the technology becomes invisible, embedded in everyday practice and hence
‘normalised’. To take some commonplace examples, a wristwatch, a pen, shoes,
writing—these are all technologies which have become normalised to the extent that
we hardly even recognise them as technologies.
Normalisation is therefore the stage when a technology is invisible, hardly even
recognised as a technology, taken for granted in everyday life. CALL has not
reached this stage, as evidenced by the use of the very acronym ‘CALL’—we do not
speak of PALL (Pen Assisted Language Learning) or of BALL (Book Assisted
Language Learning) because those two technologies are completely integrated into
education, but CALL has not yet reached that normalised stage. In other words, one
criterion of CALL’s successful integration into language learning will be that it
ceases to exist as a separate concept and field for discussion. CALL practitioners
should be aiming at their own extinction.
CALL will reach this state when computers (probably very different in shape and
size from their current manifestations) are used every day by language students and
24 S. Bax / System 31 (2003) 13–28

teachers as an integral part of every lesson, like a pen or a book. Teachers and stu-
dents will use them without fear or inhibition, and equally without an exaggerated
respect for what they can do. They will not be the centre of any lesson, but they will
play a part in almost all. They will be completely integrated into all other aspects of
classroom life, alongside coursebooks, teachers and notepads. They will go almost
unnoticed.
Most importantly, CALL will be normalised when computers are treated as always
secondary to learning itself, when the needs of learners will be carefully analysed first
of all, and then the computer used to serve those needs. Elsewhere I have tried to
discuss in detail what this means for CALL and the teaching of grammar, vocabulary
and language skills (Bax, 2000). Technology will then be in its proper place.
These features are not widely observed at the moment—hence my objection to the
notion that we are currently in an Integrative phase. In fact it may take several years
for these practices to become commonplace. However, it is possible—and, I suggest,
imperative—to plan for this normalised state and then move towards it—indeed this
offer and structure our entire agenda for the future of CALL. The first step is to
identify the criterial factors which normalisation requires. The second is to audit the
practice of each teaching context in the light of these criteria; the final step is to
adjust our current practice in each aspect so as to encourage normalisation. Fol-
lowing this procedure will give each institution and teacher a clear framework within
which to audit progress, and within which any obstacles to integration and normal-
isation can be identified and dealt with.

Diffusion of innovations

How can normalisation occur? The literature on the way in which innovations
become gradually accepted in social groups is summarised and best exemplified by
Rogers (1995) (see also McCormack Brown, 1999 for a useful summary). The dis-
cussion of that body of research is highly pertinent to CALL. For our purposes,
however, we also need to add to it certain elements such as the definition of the end-
goal for CALL in terms of normalisation. We also need to identify some of the key
stages towards that goal. Drawing on Rogers (1995), then, but adding and altering
some details to suit our particular purposes and contexts, we can summarise the
probable progress of CALL towards normalisation as follows:
Stages of normalisation in CALL:

1. Early Adopters. A few teachers and schools adopt the technology out of
curiosity.
2. Ignorance/scepticism. However, most people are sceptical, or ignorant of its
existence.
3. Try once. People try it out but reject it because of early problems. They can’t
see its value—it doesn’t appear to add anything of ‘relative advantage’
(Rogers, 1995).
4. Try again. Someone tells them it really works. They try again. They see it does
in fact have relative advantage.
 S. Bax / System 31 (2003) 13–28 25

5. Fear/awe. More people start to use it, but still there is (a) fear, alternating
with (b) exaggerated expectations.
6. Normalising. Gradually it is seen as something normal.
7. Normalisation. The technology is so integrated into our lives that it becomes
invisible—‘normalised’.

Two fallacies in our approach to CALL

Taking account of our analysis above of Restricted and Open CALL, I would
suggest that many teachers and indeed institutions, in many areas of their use of
CALL, are currently at stage 5 and 6 of the progression towards normalisation—
most people in language education would recognise that CALL does have a relative
advantage (now that the communicative potential can at least be realised through
web technology), but we are still at the stage where the majority of teachers are
nervous of it. In addition, they are over-optimistic of its powers and potential, since
the other side of the ‘Fear’ coin is ‘Awe’—many people still see CALL as having a
greater potential role in language learning than is in fact likely. The fact that we are
still in the Fear/Awe stage can be illustrated by the first of my illustrative examples:

Case 1. At a seminar at a university, a teacher was presenting some new software

which he had helped to design for teaching vocabulary. At the end the audience of
academics and fellow teachers asked questions, all concerning things the software
could not do. There was a general air of disappointment that it had no on-line dic-
tionary, or grammar check, and so on.
I have chosen this simple anecdote deliberately, since it represents what I see as a
common attitude to CALL which needs to be addressed. At the time I noted down
the fact that none of the audience commented on the features of the software as
presented, nor discussed the contextual aspects of how it might be used or how
particular groups might respond to it, nor how other classroom factors could be
combined with it. The audience seemed to have two assumptions:

(a) that the software should be able to do everything, and

(b) that if it had more features then it would be inherently more effective.

Instead of addressing the role which the software could play within the wider
classroom context (a small role, but a useful one) their expectation seemed to be that
it should either do everything and replace current technologies such as dictionaries
and even the teacher, or it was not useful. This attitude, which is relatively common
in my experience, illustrates what Healy has identified as ‘‘unreasonable and
unfounded fascination and belief in . . . [computer] technology’s educational power’’
(Healy, cited in Haughton, 1999: 2). I would characterise it as a key fallacy in users’
perception of computers in education (and in other spheres) at this moment in our
progression towards normalisation. This is our excessive ‘awe’ of computer tech-
nology and the belief that it can do more than it can; it is to put technology on a
26 S. Bax / System 31 (2003) 13–28

pedestal—the Omnipotence Fallacy. (cf. the ‘wow’ factor identified by Murray and
Barnes, 1998).
Parallel to this first fallacy is a second, which is characterised in the common
assumption that the key or only factor in successful implementation of the technol-
ogy is the technology itself—what we could call the ‘Sole Agent’ fallacy. It tends to
be accompanied by a neglect of the many other factors required in successful
implementation—and explains why in the illustrative example above the audience
ignored all issues to do with how the software would mesh with students’ needs, or
the teaching and learning environment. In principle, professionals would nowadays
recognise that many other factors are crucial in successful CALL, but in practice
many teachers, students and administrators and others in language teaching and
learning are still at the Fear/Awe stage, part of which includes this Sole Agent fal-
lacy. Here is another illustration from my own experience:

Case 2. A small private language school proudly announced the opening of a new
computer centre with 12 new computers, the latest technology and the latest soft-
ware. Four years later, it is almost unused and is being redesigned and redesignated
to teach computer skills to local businesses.
Why was the facility underused? Interviews with teachers indicate that although a
lot of time and money had gone into the purchase of equipment and furniture, all
the other crucial factors in successful CALL implementation had been ignored—
training for teachers, administrative and pedagogical support, integration into the
timetable and so on. In short, the director of the school had made the same falla-
cious assumption that the mere existence of the hardware and software is the only
relevant factor in achieving successful implementation or normalisation, when
clearly it is not.

Overcoming the two fallacious attitudes to CALL

If—as I suggest—these mistaken ideas are widespread in language teaching and

learning, at all levels and in all contexts, and represent an obstacle on the road to
successful normalisation, it is clearly imperative for the profession to seek to over-
come them. What can we do to counter these two mistaken ideas, which (a) place
CALL on a pedestal and (b) assume that the technology alone will solve the prob-
lem, and thereby ignore all the other key factors in achieving integration?
I would argue that a crucial first role for CALL is to emphasise and analyse the
host of other factors, besides the technology and software, which must be accounted
for in the move towards normalisation. All those involved need to be made aware
that successful implementation of CALL requires close attention to many areas—
technology, software, teachers’ attitudes and so on. It should be the role of CALL
practitioners to study those factors and then to communicate them to the wider
language teaching profession in an ongoing attempt to undermine the fallacious
view that computers alone are sufficient.
To this end the kind of ethnographic study exemplified by Chambers (2000)
demonstrates how a meticulous ethnographic analysis of a context can draw out the
 S. Bax / System 31 (2003) 13–28 27

factors encouraging the integration of CALL and the factors standing in its way.
This is in line with the argument put forward by Motteram (1999), who is surely
correct in arguing that we need more careful qualitative—I would argue for ethno-
graphic—analyses, in order to understand CALL better. Ethnographic studies can
help to identify the many interlocking and overlapping factors which have to be
taken into account in implementing change in a target institution, and allow us to
target our efforts more precisely. This is the effect of Chambers’ study (op. cit),
which can serve as a useful model for other institutions seeking to move towards the
normalisation of CALL.
At the level of the individual teacher a potentially fruitful procedure for achieving
normalisation is to carry out careful action research into integration. This is well
illustrated by recent work by Cole (2001), which demonstrates that a careful attempt
to use CALL in a thoroughly integrated way can not only yield results in terms of
student response, but can illuminate the factors which affect or hinder normal-
isation.

A proposed agenda for CALL

In conclusion, then, we have identified a possible future agenda for CALL. Our
aim can be the normalisation which I have described, in which CALL finally
becomes invisible, serving the needs of learners and integrated into every teachers’
everyday practice. This will almost certainly require changes in technology, in the
size, shape and position of the classroom computer. It will require change in atti-
tudes, in approach and practice amongst teachers and learners; it will require fuller
integration into administrative procedures and syllabuses.
Secondly, we have identified the means of achieving this goal. I argue that we need
more in-depth ethnographic studies of individual environments to elucidate the
relationship between the factors just mentioned. We also need action research in
individual environments to identify barriers to normalisation and ways of over-
coming them.
This will not be an easy process. However, if we take our aim to be normalisation,
and then work for ways of achieving it efficiently, computers can finally achieve their
proper place and true potential in the classroom.

References

Ahmad, K., Corbett, G., Rodgers, M., Sussex, R., 1985. Computers, Language Learning and Language
Teaching. Cambridge University Press, Cambridge.
Bax, S., 2000. Putting technology in its place. In: Field, C. (Ed.), Issues in Modern Foreign Languages
Teaching. Routledge, pp. 208–219.
Chambers, A., 2000. Current Practice in CALL: Teachers’ Attitudes and other Factors that Limit the
Potential of CALL. MA dissertation, Canterbury Christ Church University College, UK.
Chapelle, C., 2001. Computer Applications in Second Language Acquisition. Cambridge University Press,
Cambridge.
Cole, D., 2001. Implementation of CALL in Classroom Practice. MA dissertation, Canterbury Christ
Church University College, UK.
28 S. Bax / System 31 (2003) 13–28

Davies, G., 1982. Computers, Language and Language Learning. CILT, London.
Delcloque, P., 2000. History of CALL. Available from http://www.history-of-call.org/ (Accessed 27/1/02).
Haughton, E., 1999. Look what they’ve done to my brain, ma. The Independent Education supplement 3/
6/99, p. 2.
Howatt, A.P.R., 1984. A History of English Language Teaching. Oxford University Press, Oxford.
Kenning, M.-M., Kenning, M.J., 1990. Computers and Language Learning: Current Theory and Practice.
Ellis Horwood, New York.
Levy, M., 2000. CALL Context and Conceptualization. Oxford University Press, Oxford.
Littlewood, W., 1981. Communicative Language Teaching. Cambridge University Press, Cambridge.
McCormack Brown, K., 1999. Diffusion of Innovations. Available from http://hsc.usf.edu/ kmbrown/
Diffusion_of_Innovations_Overview.htm (Accessed 1/2/02).
Meinhoff, U., 1990. Television news, the computer and foreign language learning. In: Anivan, S. (Ed.),
Language Teaching Methodology for the Nineties. SEAMEO, Singapore, pp. 250–268.
Mitchell, R., 1994. The communicative approach to language teaching. In: Swarbrick, A. (Ed.), Teaching
Modern Languages. Routledge, London.
Motteram, G., 1999. Changing the research paradigm: qualitative research methodology and the CALL
classroom. In: Debski, R., Levy, M. (Eds.), WorldCALL: Global Perspectives on Computer-Assisted
Language Learning. Swets & Zeitlinger, Amsterdam.
Motteram, G., 2000. Communicating with computers. In: Brett, P., Motteram, G. (Eds.), A Special
Interest in Computers. IATEFL, Whitstable, 2000, pp. 73–91.
Murray, L., Barnes, A., 1998. Beyond the ‘‘wow’’ factor—evaluating multimedia language learning soft-
ware from a pedagogical viewpoint. System 26, 249–259.
Oller, J., 1972. Assessing competence in ESL: reading. TESOL Quarterly 6 (4 Dec), 313–323.
Rogers, E., 1995. Diffusion of Innovations, fourth ed. Free Press, New York.
Sanders, D., Kenner, R., 1983. Whither CAI? The need for communicative courseware. System 11 (1), 33–40.
Sanders, R. (Ed.), 1995. Thirty Years of Computer Assisted Language Instruction: Festschrift for John R.
Russell. CALICO Monograph Series, Vol. 3.
Stevens, V., 1992. Humanism and CALL: a coming of age. In: Pennington, M., Stevens, V. (Eds.), Com-
puters in Applied Linguistics. Multilingual Matters, Clevedon, pp. 11–38.
Underwood, J., 1984. Linguistics, Computers, and the Language Teacher: A Communicative Approach.
Newbury House, Rowley, MA.
Warschauer, M., 1996. Computer-assisted language learning: an introduction. In: Fotos, S. (Ed.), Multi-
media Language Teaching. Logos, Tokyo, pp. 3–20.
Warschauer, M., 2000. ‘‘CALL for the 21st Century’’ IATEFL and ESADE Conference, 2 July 2000,
Barcelona, Spain. Available from http://www.gse.uci.edu/markw/cyberspace.html (Accessed 25/1/02).
Warschauer, M., Healey, D., 1998. Computers and language learning: an overview. Language Teaching
31, 57–71. Available from http://www.lll.hawaii.edu/web/faculty/markw/overview.html.