Task Based Learning

The Task-based Approach in Language Teaching
AQUILINO
SANCHEZ*
University of Murcia
The Task-Based Approach (TBA) has gained popularity in the field of language teaching since
the last decade of the 20fhCentury and significant scholars have joined the discussion and
increased the amount of analytical studies on the issue. Nevertheless experimental research is
poor, and the tendency of some of the scholars is nowadays shifting towards a more tempered
and moderate stand on their claims. Reasons for that are various: the difficulty in the
implementation of the method in the classroom, the difficulty in elaborating materials following
the TBA and the scarcity of task-based manuals count as important and perhaps decisive
arguments. But there are also theoretical implications in the TBA which do not seem to be fully
convincing or may lack sound foundations. In this paper 1 will attempt to describe the TBA
criticaIly, pointing out what 1consider positive in this approach, and underlining the inadequacy
of some assumptions and conclusions. The design of a new TBA model is not the goal of this
study. But the conclusions suggest that tasks may contribute to the production of a more refined
and complete foreign language syllabus, helping to motivate the students and focus the attention
of teachers and learners on meaning and communicative language use.
KEYWORDS
Language teaching, methods, methodology, communicative methodology, tasks, task based
approach, process approach, language learning.
Addressfvrcorrespondence: Departamento de Filologa Inglesa, Facultad de Letras, Campus de Lu ,I,lerced, 30071

Murcia. Spain. Tel. 968-3631 75. E-mail: asanchez@um.es
O Servicio de Publicaciones. Universidad de Murcia. All rights reserved.
IJES, vol. 4 (l), 2004, pp. 39-71
Changes and shifts in language teaching have been present throughout the history of this
discipline. At the basis of this apparently unending uncertainty about the efticiency of methods
at specific historical moments there is also a permanent search and striving to find better ways
of teaching and learning languages, which implies acknowledging dissatisfaction with ongoing
methods and procedures. In the second half of the 20th century those changes in methodology
were more frequent and pressing for teachers and learners. The need for communication arnong
people of different cultures and languages. triggered by travelling and globalisation, puts
pressure on people to learn languages more quickly and efficiently. Leaming a new system of
communication is also substantially different from what it used to be in previous centuries: we
have more need to communicate orally (not only in writing and reading) and we cannot wait for
years before we engage in real communication. This urgency to learn languages is felt
everywhere within society al1 over the world. The search for new and more efficient methods
is a consequence of our social organization and the requirements for fluid communication.
Methodological changes follow each other within short periods of time. Even though the
majority of educational innovations end in failure (Adams, R. and Chen D., 1981) positive
effects can be expected from most of them. But it is true that new methods do not appear al1 of
a sudden or disconnected from the world into which they are born. They overlap for some time
with current methodological practices. This 'incubation' period is a real test for new ideas: some
of thern pass the test, others do not. Many discussions, arguments and counterarguments are
exhibited in the process. But sometimes what was considered a decisive gain against existing
practices at a given moment, proved to be wrong a few years later, and a new theory or method
replaced it in its turn. Once more? Where will the end lie. if there is to be one? The methods
which prevail are usually those that are best suited to the challenges, demands and needs of the
time.
In other writings (Snchez 1992; 1997) 1 have outlined two main trends in language
teaching methodology: the 'grammatical' and the 'conversational' approach. Both approaches
have been permanently in tension with each other and are representative of a dichotomy that
seems to reappear again and again in different ways and formats: written vs. oral language;
learning grammar vs. learning how to speak; and formal vs. informal language use. In the last
part of the 20thCentury the dichotomy focus on form vs. focus on content, teaching and learning
language for accuracy vs. teaching and learning language for meaning developed as the new
paradigm. Emphasis on one or the other end of the scale tends to be cyclical, so that if form.
structure and accuracy prevailed in the sixties and seventies, meaning and communicative
potential gained momentum in the eighties and aftenvards.
The Task Based Approach (TBA) must be placed within this context, at the end of the
20Ih Century. It is not an isolated or 'unique' methodological event. TBA can only be fully
understood if you contrast it with preceding methods and analyse it within mainstream
communicative methodology. Some background information will therefore be needed, and that
IJES: vol. 4 ( l ) , 2004, pp. 39-71
41
is the goal in the first section of this paper. A detailed discussion will follow on what a task is
and on the various definitions proposed. This will lay the ground for a 'balanced criticism' of
the TBA and will allow the reader to draw some positive and realistic conclusions on the issue.
BACKGROUND AND RATIONALE BEHIND THE TBA IN LANGUAGE

LEARNING
The emergence of the TBA is connected to what became known as the 'Bangalore Project'
(Prabhu 1987) initiated in 1979 and completed in 1984. The word 'task' is often used here to
refer to the special kind of activities carried on in the classroom. Such activities are
characterised. among other features, by the emphasis put on meaning and the irnportance
assigned to the process of doing things (how) vs. the prevailing role given to content (nzhut) in
the teaching practice of that decade. The purpose of the project is to investigate new ways of
teaching which sprang from
a strongly felt pedagogic intuition, arising from experience generally but made concrete in the
course of professional debate in India. This was that the development of competence in second
language requires no systematisation of language inputs or maximization of planned practice, but
rather the creation of conditions in which learners engage in an effort to cope with
communication.
Piabhu (1987: 1)
The pro-iect aimed at improving the SOS ('situational oral approach') and the emphasis lay on
competence and communication. Prabhu stated explicitly that competence is to be understood
as 'grammatical competence' ('the ability to conform automatically to grammatical norms') and
communication as 'a matter of understanding or conveying meaning'. Communicative
competence was to develop 'in the course of meaning-focused activity'. It should also be borne
in mind that grammatical competence was to be built through 'intemal self-regulating processes'
and for that it would help to convey meaning in 'favourable conditions'. The most important
responsibility of the teachers was to create the conditions for the leamers to engage in
meaningful situations. Any prior regulation of what had to be learnt according to a predefined
formal o grammatical syllabus was to be excluded.
Emphasis on meaning and authenticity of communication appeal to many teachers and
learners of languages. After all, we use language for transmitting messages, which is content.
and association of meaning and language is perceived as close to reality. The problem is that the
transmission of meaning cannot be separated from the formal 'vehicle' through which it is
conveyed. The role of each one of those elements in communication and their mutual
relationships are at the root of a problem that has never been fully solved.
Most methods are heavily rooted in linguistic theories, theories of leaming or theoretical
assumptions on the nature of linguistic communication. It would be unfair to study the
O Servicio de Publicaciones. Universidad de Murcia. Al1 rights reserved.
IJES, vol. 4 (l), 2004, pp. 39-71
communicative approach without linking it to such names as Vygotsky (1962; 1978), Austin
(1962) or Halliday (1973; 1978), or Krashen (1983), among others.
Vygotsky thinks of language as a social event, a shared social activity through which
individuals develop their personality within a community. In his view, language is not the result
of 'isolated' learning; it requires a social basis. And linguistic ability is built inside our mind to
communicate with the outside world. Children enhance their own personality as 'different'
human beings precisely through contrast and interaction with their environment, and particularly
with other human beings around them. In fact, Vygotsky considers that thought, as separate from
language, takes root when interaction gives way to or turns into monologue. Through
monologues children communicate with themselves, and they do that with words or sentences
that might be difficult or impossible for others to understand. While this 'inner speech'
consolidates and strengthens linguistic thought, communicative interaction allows for the
consolidation of the social dimension of the 'speakers' as human beings. This social perspective
should never be absent in the learning process of languages. The conclusion, then, should be that
when we learn languages other than the native one, the social dimension can only be reached
through interaction and interpersonal relations with others.
Vygotskyan assumptions lead to conclude that interaction belongs to the very nature of
language, because language is socially based. From this perspective, content is important, but
interaction is still more important: you cannot reach true linguistic achievements if opportunities
for interaction are not present. It is obvious that the kind of interaction needed must be
'meaningful' and relevant. What else can be expected from interaction with others? One might
raise the problem of how the learner will manage to integrate and assimilate knowledge coming
from outside. And the answer to that is that nature provides the learner with the necessary
capacity and resources, as needed. You may further raise the issue of how different it is learning
an L1 andan L2. Within the Vygotskyan perspective, however, both processes of learning share
the same object (language) and some basic conditions of learning (shaping reality through
language, establishing connections to the outside world), and it is only to be expected that some
pedagogical 'therapy' has to be applied. Language learnt (both L1 and L2), on the other hand,
serves the same purpose for the learner: it makes communication possible.
Thejunctional nature of language is highlighted by Firthian and Hallidayan linguistics.
And it is through the work of Wilkins (1 976) that thisfunctional dimension is incorporated into
methodological issues on language teaching. Wilkins does it by contrasting synthetic and
analytic syllabuses. He describes synthetic language teaching as a strategy
in which the different parts of the language are taught separately and step-by-step so that
acquisition is a process of gradual accunlulation of the parts until the whole structure of the
language has been built up.
Wilkins (19 76t2)
This strategy requires that the language be 'broken down' into its parts 4 . e . structures
IJES, vol. 4 ( 1 ) . 2004, pp. 39-71
The Tusk-busedApprouch in Lunguuge Teuching
43
and lexical items-, that these parts be ordered following specific criteria and that they be
presented to the learner. Samples brought into the classroom will be necessarily limited and at
least partially disconnected from the whole, which makes the task of the learner still more
difficult when attempting to put al1 the elements together in order to build meaningful 'chunks'
of language.
Analytic approaches do not emphasize such a tight control of elements learnt or learning
itself. In Wilkins' words,
In analytic approaches there is no attempt at this careful linguistic control of the learning
environment. Components of language are not seen as building blocks which have to be
progressively accumulated. [...] In general, structural considerations are secondary when
decisions are being taken about the way in which the language to which the learner will be
exposed is to be selected and organized.
Wilkins (1976i2)
And he adds later:

The prior analysis of the total language system into a set of discrete pieces of language that is a
necessary precondition for the adoption of a synthetic approach is largely superfluous in an
analytic approach. [...] They are organized in terms of the purposes for which people are learning
language and the kinds of language performance that are necessary to meet these purposes. [. . . ]
The units [...] are not primarily labelled in grammatical terms.
Wilkins (1 976:1 3)
Wilkins advocates notional and functional syllabuses, which are analytic in nature -he
says-and, in doing so, he moves significantly away from current teaching practices, based on
audiolingual methodology and attached to the learning of linguistic forms. It may be questioned
whether his 'notional-functional' syllabus is fully analytical, as he claims, since it requires a
previous analysis oflanguage, decomposing it into notions, functions and other smaller linguistic
units, which learners must later put together into larger situations or notions. Notional syllabuses
are still 'elaborated syllabuses' and derive from a previous analysis of the language. The result
of this analysis is then ordered according to particular rules or principles, and this is the final
product that must be taught and learnt. Selection of the content is controlled by principles other
than grammatical or structural ones, but there is a content organized in 'pieces' of language of
different kinds, stratified at different levels, some of them, it is true, of a semantic nature
(meaning). From that point of view Wilkins' notional-functional syllabus is more synthetic than
analytical, as Long and Crookes (1985; 1992) claim. But it is only fair to admit that Wilkins
paved the way towards an approach focused on functions and notions, and hence, on meaning.
The situation is less satisfactory if we approach the teaching scene from a
'communicative' perspective. In this view, it is said that in the process of language acquisition
the role ofthe learner is central and decisive. 'Acquisition' -as opposed to 'learning', following
IJES, vol. 4 (l), 2004, pp. 39-71
44
Aquilino Snchez
Krashen (Krashen. 1985)- is the true goal in language learning, but it takes place only if
knowledge (language) is integrated by the individual into his own set of values and idiosyncrasy
and if a linguistic systern is built. 'Leaming' alone (Krashen, 1985) is not enough. The principie
taken for granted in synthetic approaches, 'what is taught = what is (or ought to be) leamt'
(Prabhu 1984:273 cannot be accepted any more, since it comes frorn outside (externa] syllabus)
or is irnposed on the learner (by the teacher). Under those circurnstances, 'acquisition' will be
hindered, or at least not favoured. This is a job that only the learner can do; he rnust be 'invited'
to collaborate in this purpose, assurning this goal by hirnself. As Long and Crookes (1985:34)
put it, 'language learning is more a psycholinguistic than a linguistic process'.
LANGUAGE LEARNING AS A PROCESS

Prabhu and others initiated the Bangalore Project in 1979. At that time Prabhu affirmed:
Communicative teaching in most Western thinking has been trainingfoi- communication, which
1 claim involves one in some way or other in the preselection; it is a kind of matching of notion
and form. Whereas the Bangalore Project is teaching thi-o~ghcommunication; therefore the very
notion of communication is different.
Piahhz~(1980:164)
Prabhu's claim is revolutionary regarding synthetic or notional-functional approaches:

you do not provide the learners with previously organized language rnaterials to learn; you do
not pretend to achieve specific cornrnunicative goals through activities previously designed and
sequenced, but rather expect the learners to leam through the activities they engage in while
using the language to carry out the task proposed. The process of cornmunication itself is the
rneunsfor learning to cornrnunicate. Acquisition ofthe formal system of language will take sorne
time, but will be reached 'subconsciously' through the activation of an interna1 systern of rules
and principies by the learner. The condition to be rnet is that cornrnunicative practice rnust be
carried out in a meaningful way (Krashen 1982).
There is therefore no syllabus in terms of vocabulary or stmcture, no pre-selection of language
items for any given lesson or activity and no stage in the lesson when language items are practised
or sentence production as such is demanded. The basis of each lesson is a problem solving or a
task.
Piabhti (1984:273-6)
Formal approaches define in advance what the learners rnust learn. The syllabus is
regulated frorn outside. In a process approach assurnptions are very different: learners regulate
the process of learning by themselves, autonornously. And this self-regulating activity results
in language acquisition, as it happens in a natural environrnent (learning of the rnother tongue).
Process approaches do not separate the object c?f'leurning frorn the process oflearning. To do
/JE.S, vol. 4 (1). 2004, pp. 39-71
45
The Task-based Approach in Lanpage Teaching
that when learning a second language would involve depriving the learners of applying their
previous experiences in language learning. That is rather the case of methods based on formal
systems, in which new models (object-regulated input) are offered, while interactive activities
are absent or adjust to formal pattems and become structurally conditioned. In order not to
divorce the object and the process of leaming, Prabhu (1987) expresses the need for 'enabling'
procedures, that is operational ways and practices to reinforce the potential of learners not only
to fulfil specific communicative needs in carrying out a task, but also communicative needs in
the future when implementing different tasks. Working with tasks should allow learners to cope
with unpredictable communicative situations. In fact, fulfilling a task should necessarily bring
with it the development of the learner's cognitive abilities: this will automatically derive from
the solution of the logical problems implied by the sequence of events inherent to tasks. When
leaming a foreign language, the means to perform the task is precisely the target language. The
object and the process of learning converge in one single event, which is 'holistic' in nature: the
process of (interactive) communication, the use of the suitable communicative elements in a
genuine communicative situation and the strengthening of the cognitive abilities of the
intervening individuals al1 come together in a unique communicative episode. In terms of
syllabus design, tasks are fully inserted within a process syllabus, but they cannot get rid of
discrete linguistic elements. How to solve the tension involved in bringing together and
integrating both components is the main challenge of a task-based approach.
When performing a task in the real world, language is automatically limited: structures
and words to be used will be restricted to the semantic field covered by the task. And so they are
as well the logical steps underlying the fulfilment of the task. Gouin's (1892) 'series method'
and the 'logic of nature' can be called upon here to illustrate the situation. The logic of 'cause
and effect' (any cause produces a specific effect and any effect is the result of a specific cause)
pushes the task forward in a way that the learner can automatically and unconsciously detect. On
the basis of this understanding of the ongoing process, the learner will be able to understand the
language being used and carry out the task, occasionally with the help of his peers or the teacher
(the 'outside world'), finding the right words for the right things or ideas.
Breen (1987a) concludes that the TBA is a result of
i) New views on language.
ii) New views on teaching methodology.
iii) New views on the contribution of the learners to the learning process.
iv) New views on how to plan teaching and learning.
P o i ~ t siii) and iv) deserve some comments. The role of the learner has been
systematically left aside for centuries. And that has not only been the case in language teaching,
but in al1 educational fields. Traditional education centred on the transmission of content, well
defined and laid down by teachers or by the authorities. Not much else was added or considered
O Servicio de Publicaciones. Universidad de Murcia. A l l rights reserved.
IJES, vol. 4 ( l ) , 2004, pp. 39-71
46
Aquilino Snchez
regarding other elements also present in the teaching and learning situation. Research in
language acquisition, among other reasons, has recently demonstrated what nowadays seems
obvious: the most important element in the teaching-learning situation is the learner. The
analysis of learning itself reveals relevant facts. Allwright (1984) concludes that learners do not
necessarily learn what teachers teach, while sometimes they learn what teachers have not taught.
And that is so in spite of admitting that class attendance has an effect on learning (Long 1983).
What do learners do in the process of learning? And how do they do it? Answering those
questions requires an in depth analysis of the participation of the learner in the process of
learning. Research is still incomplete in this area, but it seems that individual and inherent
capabilities of the learners prevail over externa1 factors (say teacher, materials, syllabus) (Ellis
1985). Learners, consciously or not, systematically follow their own patterns of learning and
manage to reprocess the input 'in their own way'. Efficiency in teaching demands a careful reevaluation of the learner's role in the classroom.
Regarding point iv., new theories, methods or ideas on teaching abound. This is only
natural if we take into account the exclusive prominence of teachers in the past. But if learners
enter the scene, the process of learning must also be the subject of a more careful attention and
analysis. Experience reveals that a careful definition of the syllabus does not result in the
learning of such a syllabus. ln other words, the syllabus taught is not necessarily equal to the
syllabus learnt. The elaboration of syllabuses is no longer the work of amateurs. On the contrary,
specialists in syllabus design are responsible for defining and refining syllabuses in the school
system, which is no doubt a guarantee of their quality and adequacy. But apparently this is not
enough to reach a satisfactory leve1 in efficiency. Something must be there that hinders the
achievement of the intended results in the teaching-learning situation. Perhaps the 'learning
dimension' should also be included in the definition of a syllabus, which would imply that
formal teaching should no longer be the prevailing criterion conditioning syllabus design.
Syllabus complexity is well illustrated in the literature of language teaching (see Dubin and
Olshtein 1986, among others); what is taught (content) should perhaps be integrated with the
way the content is taught (procedure). The way content and method, content and procedures are
approached needs reconsideration.
If the way we teach has an effect on learning, process syllabuses have a role to play in
language teaching. Contrary to the 'propositional syllabuses' (based on the definition of
structures, rules and vocabulary to learn), 'process syllabuses' face the teaching situation from
the opposite side: they focus not on what has to be taught, but on how things are done or how
goals are achieved. Goals to be achieved are still there, but the means and skills to reach them
are given priority in the analysis of the situation. It is assumed that if we perform the task
adequately, the goals will be achieved more efficiently.
O Servicio de Publicaciones. Universidad de Murcia. All rights reserved
IJES, vol. 4 ( l ) , 2004, pp. 39-71
The Tusk-bused Approuch in Lunguuge Teuching
47
WHAT IS A TASK?
Theory and practice around TBA are far from being uniform and clear. A review of the literature
on the topic reveals that governing principles are loose and not everybody shares the sarne
defining criteria. The TBA has also been applied in different ways in the classroom. Breen
(1987:157) advocates a difference between task-based syllabuses and process syllabuses,
although he acknowledges roots common to both of them, which are named 'processplans'. That
is, task-based syllabuses are 'process based'. Does the concept of task imply more emphasis on
the process of doing things than on the goal it aims at? Processes and goals both belong to the
nature of tasks. Why not focus on goals more than on processes, or on goals as much as on
processes? Are goals less important than the way we achieve them? Traditional methodology and
school practice have prioritised goals in general and a similar point of view is to be detected in
many other areas of human action. This appears not to be the case in the TBA.
Long and Crookes (1992:27) affirm that 'three new, task-based syllabus types appeared
in the 1980s: (a) the procedural syllabus, (b) the process syllabus, and (c) the task-syllabus',
adding later on that 'al1 three reject linguistic elements as the unit of analysis and opt instead for
some conception of jusk'. Following this statement, tasks are to be considered essential to the
three of them and constitute a common denominator, not just a distinctive element of the taskbased syllabus vs. the other two syllabus types. This view is not easy to match with other views,
in which, for example, task-based syllabuses are seen as different from process-syllabuses, while
both are rooted in 'process plans'(See Breen 1987a; 1987b). Do differences derive from the
underlying concept of tusk?
Tasks, in fact, have been defined in different ways. Prabhu proposes the following
definition:
An activity which required leamers to arrive atan outcome from given information through some
process of thought,and which allowed teachers to control and regulate that process, was regarded
as a 'task'.
Prabhat /IY87:21)
The nature of rask is depicted in quite general traits. Two important features are however
mentioned, tightly connected to what was going on in the project: task completion (an outcome
at the end of the activity) and a process 'of thought' while doing the activity. The activity itself,
curiously enough. 'allowed teachers to control and regulate the process' (Where is the autonomy
of the learner in building his own path of learning?).
Long (1985) defines tasks looking at what people usually do in real life:
A piece of work undertaken for oneself or for others, freely or for some reward. Thus, examples
of tasks include painting a fence, dressing a child, filling out a form. buying a pair of shoes,
making an airline reservation, borrowing a library book, taking a driving test. typing a letter.
weighing a patient, sorting letters, taking a hotel reservation, writing a check, finding a street
IJES, vol. 4 (l), 2004, pp. 39-7 1
destination and helping someone across a road. In other words, by 'task' is nieant the hundred and
one things people do in everyday life, at work, at play, and in between. Tasks are the things people
will tell you they do if you ask them and they are not applied linguists.
Long (IY85:HY)
The definition matches the semantic expectations of normal speakers when using the
word 'task' in daily life ('A piece of work assigned to or demanded of a person', in Web.ster'S
dictionary. ' A piece of work to be done or undertaken', in The New Oxji~rdDictionury of'
English). But such a view of the nature of tasks in real life still needs an adaptation to the
classroom situation. 'Painting a fence, buying a pair of shoes' or thousands of other similar daily
tasks are not likely to be 'naturally' performed in the classroom; some of them -xxtremely
important for communication-cannot even be dramatized in the classroom environment. Long
and Crookes (1 992) keep to that definition to support their proposal for task-based syllabuses
and they apparently also accept the one given by Crookes (1986:l) ('a piece of work or an
activity, usually with a specified objective, undertaken as part of an educational course, or at
work'). These definitions are, however, significantly different: Crookes' definition derives from
a classroom perspective and allows for a pedagogical function and manipulation ('specified
objective', 'part of an educational course'), while Long's definition is rooted in real world tasks.
While Long's definition runs parallel to his claim for a 'needs identification of learners' tasks',
the one by Crookes seems to be more dependent on course requirements or possibilities. At the
end of their analysis, both propose a set of 'pedagogic tasks' as the basis for a task-based
syllabus. Their views and statements lead us to conclude that Long's 'real world tasks' must be
filtered and selected depending on what the classroom situation admits, adding to it an ingredient
that must be carefully administered: formal communicative elements necessary for task
conipletion (basically linguistic forms).
Candlin formulates his own definition from a 'pedagogic and operational' point of view:
One of a set of differentiated. sequenceable, problem-posing activities involving learners and
teachers in some joint selection froni a range of varied cognitive and cominunicative procedures
applied to existiiig and new knowledge in the collective exploration and pursuance of foreseen or
emergent goals within a social niilieu.
Candlin (1987: 10)
He specifies that a task involves a set of activities ('one of a set.. . sequenceable.. . '), that
they imply a problem which must be solved, that interaction of various kinds must be activated
and that a goal will be pursued and can be achieved deploying cognitive and communicative
procedures, either taking advantage of already existing knowledge or creating new knowledge
if necessary to achieve the completion of the task. The task is to be performed within a 'social
milieu'. Candlin's definition also clearly refers to tasks to be performed in the classroom,
preferably of a communicative nature. Procedures and goals are mentioned as two of the
ingredients of a task, although the nature of 'goals' has to be understood as a 'conipletion' ofthe
IJES, vol. 4 (l), 2004. pp. 39-71
The Tusk-bused Approuch in L~~nguuge

Teuching
49
task, which might be of a non-linguistic character (say. solving a mathematical problem). On that
basis it is to be assumed that the language used for carrying out the task has to be considered as
instrumental. Learners will gain in their linguistic skills through the language practice needed
to perform the task, reactivating their own linguistic resources or searching for new ones when
the knowledge available is insufficient. Eniphasis, as in the case of Prabhu, is put on the process
required to reach a specific goal and the meaningful nature of the resources applied to that goal.
Nunan offers a definition focused more on the language classroom:
A piece o f classroom work which involves learners in comprehending, manipulating, producing
or interacting in the target language; while theirattention is principally focused on meaning rather
tlian form.
Nlrnan (1989: 10)
Such a definition synthesizes some of the most prominent features highlighted by other
authors. as Nunan himself remarks, with the exception of one element not mentioned here: tasks
are not necessarily 'goal-driven or goal-oriented'. In that case, his conception of tasks is hardly
to be put alongside real world tasks, where pursuing a goal is an essential feature.
For J. Willis a task is an activity
wliere the target language is used by tlie learner for a communicative purpose (goal) iii order to
achieve an outcome.
J. Willis (/YY6:23)
With this definition Willis achieves the maximum of simplicity. but does not help to
clarify the issue: in this view a task may be any of the communicative activities, of various kinds,
available in textbooks and often practised in the classroom.
Skehan writes that a task is
an activity in which:
- meaning is primary
- there is a problem to solve

- tlie perforniance is outcome evaluated
- there is a real world relationship
Skehnn (2001: 12-13)
Skehan highlights four key features. which are fully within the main streani in the
literature around this issue.
The variety of definitions surrounding the concept 'task' reveals a significant number of
different points of view. It also appears that one of the reasons for the differences is that scholars
do not approach the topic from the point of view of the nature of the task itself in real life, but
rather froni specific methodological preconceptions, which act as filters to the further description
O Servicio de Publicaciones. Universidad de Murcia. All riglits reserved.
IJES, vol. 4 ( l), 2004, pp. 39-71
or definition of the object of study; thus, they end up with different results.
Tasks in real life and pedagogical tasks in the classroom are not to be fully equated.
Tasks performed in the real world are not necessarily transferable to the teaching situation.
among other things, because classrooms do not offer the same situational environment as the
'real world' does. When applied to the classroom we are not interested, for example, in tasks that
do not require written or oral linguistic communication (for example. a mathematical operation
alone), or in tasks unsuitable for the classroom (for example, brushing the floor). Those same
activities might be thought useful for other educational or learning purposes (developing
cognitive or logical abilities), or for feeling comfortable at home, etc. But the tasks language
teachers are interested in are of a particular kind. It seems necessary. therefore, to identi% the
essential features of real and pedagogical tasks.
A second area in which more transparency is needed refers to the dual conceptual pair
'content-method' (what - how). Assigning clear frontiers to both concepts is an old problem. The
content to be taught and the way it is brought into and taught in the classroom cannot be neatly
separated: both interact with each other and are mutually dependent. ln contrast to what some
authors maintain (Kumaravadivelu 1993:72f), methods carry the teaching of content in concord
with the techniques previously defined, and viceversa; one cannot easily imagine the content of
an audiolingual method together with the habitual comn~unicativetechniques and activities of
a communicative or grammar-translation method; or Direct Methodists conducting the class
together with the teaching of structural patterns. Mutual relationships and dependency, however,
do not preclude a minimum of autonomy, which allows for substantial differences. Methods in
general, as they have been formulated by their authors or consolidated by tradition and practice,
tend to emphasize one component or the other: content (whut to teuch) or method (how to teach).
Grammar-Translation or audiolingual methodologies can be defined as content-driven, while the
Direct Method or the Communicative Approach emphasize the role of method (techniques,
procedures). TBA, basically within the communicative mainstream, shares this view: it matters
more how to do things in the classroom, while the whut to teach (prior definition of a syllabus)
plays a less salient role. For some defenders of TBA the definition of a syllabus from outside
must even be excluded: the learning group should be responsible for their own syllabus. ln fact
the linguistic elements to be taught will be defined by the task selected and should be
subordinated to it. Interaction, the negotiation of meaning, the building of a personal leaming
path. the importante of meaning versus linguistic form become the skeleton of TBA, where the
'linear' addition of linguistic materials to be learnt, typical of content-based methodologies,
gives way to materials relevant for communication (based on the communicative needs of the
learners 'loosely' organized -if at all. Acquisition will take place by practising with those
materials in activities that simulate real life situations and contexts. lt is also believed that
leamers are naturally endowed with the necessary capability and abilities to learn by working
with activities focused on meaning; here it seems that there are too many ingredients that are
difficult to constrain and handle coherently within a single teaching-learning approach.
O Servicio de Publicaciones.Universidad de Murcia. All rights reserved.
IJES, vol. 4 ( l ) , 2004, pp. 39-7 1
The Task-based Approach in Lunguuge Teaching
51
REAL WORLD TASKS AND CLASSROOM TASKS

Tasks in everyday life are to be found everywhere. Tasks surround us from early in the morning
ti11 late at night. Washing our face is a task, as is preparing breakfast, going to work by car.
preparing a lesson, buying the newspaper, etc. Tasks pervade our lives, so much so that there is
hardly an activity that cannot be called a task. When applied linguists and methodologists began
using that word, they obviously relied in one way or another on the basic meaning it had in usual,
plain speech. It is obvious that applied linguists were taking advantage of the semantic field
covered by 'task', but at the same time they +onsciously or not- used the word restricting and
adapting its meaning to concepts common to the field of language teachingllearning. The
'restricted' (pedagogical) or 'unrestricted' (real world) semantic content assigned to 'tasks' often
leads to some confusion and misunderstandings. Regarding the features of real world tasks, the
following set is suggested:
i) They are goal-oriented or goal-guided activities. Performance is evaluated depending
on the achievement or not of the goal.
ii) They consist most of the time of a sequence of steps, well differentiated but tightly
connected among themselves, mutually conditioned by the logical sequence of the
actions preceding and following each one of the steps. Failure to fulfil one of the steps
can invalidate the outcome of the task.
iii) The process and procedures applied in the fulfilment of the task condition the
effective and efficient achievement of the final goal, which is what really matters when
we engage in a task. But procedures per se do not necessarily invulidate the attainment
of the final goal.
iv) Tools needed and procedures applied vary depending on the goals we aim at.
V ) The goal to be reached might be a problem to solve, but not necessarily.
vi) While performing the task, efficiency is closely connected to the leve1 of attention
devoted to it. Human beings, however, work with limited processing systems, so that if
we concentrate on a specific area or topic, another one will probably be totally or
partially abandoned.
vii) Tasks in real life are fully holistic: in their realization the whole person is involved:
mind and body, thought and action must be coordinated and work together. When
coordination and cooperation is deficient, efficiency in task performance declines.
Do those features apply to pedagogic tasks?
Most authors on task-based methodology emphasize the dichotomy meaning vs. form
(Breen (1984; 1987), Candlin (1984; 1987), Prabhu (1984; 1987), Long (1991) Ellis (2003),
Skehan (1986; 1998 etc.). Such a dichotomy is specific to tasks when they are used in the
classroom for teaching languages. In the case of real world tasks (making the bed, mending a
IJES, vol. 4 ( 1 ) . 2004, pp. 39-71
skirt, etc.) such a dichotomy does not necessarily apply, but dichotomies of a different nature can
be found instead. lt has also often been mentioned that learning in a natural environment centres
on meaning more than on form. Nowadays this is a distinctive label in modern second language
acquisition research, although emphasis on content and meaning was already promoted and
practiced by Gouin's method, by the Direct Methodists and by most defenders of 'conversational
approaches' (Snchez 1997). Recent and specific research on the issue (Van Patten 1990, 1996)
further confirms the importance of meaning-guided activities.
Both components are important in language learning. Meaning is 'contained in' and
'conditioned by' the form in which it is inserted. Form alone is useless for communication if
meaning is not attached to it. How to keep both of them active when learning is a real challenge
for the limited capacity of human beings. Some advocates of task-based and process approaches
insist on the primary role of meaning, while ernphasis on form is left aside. Experience of
learning in the classroom goes often against that claim, as teachers and students feel the need for
more formal teaching and learning. There seems to be a gap in the analysis of the problem. On
the one hand, emphasis on meaning is said to be necessary for more efficient learning. On the
other hand, in learning second languages, emphasizing meaning and leaving form aside does not
work properly. Moreover, whether consciously or unconsciously, many learners do not accept
this method and re-establish a focus on form in some way. Parallel to that, the importance of
process in task performance is also emphasized, while formal linguistic goals move to a
secondary place. This introduces a new problem: if reaching the final stage or goal is what really
matters in performing a task, the process should be secondary and subordinate to this final goal,
since it defines the path towards the goal, but its 'raison d'itre' lies in the goal it serves. From
that perspective, a process 'per se' is meaningless unless it is associated and subservient to the
goals it pursues.
A TBA takes real world tasks as the source and model for pedagogical action. The
question rnust then be posed: In which way can real world tasks 'enter' the classroom and be
adapted to it?
To begin with, not al1 real world tasks are eligible for pedagogical purposes, as said
above. Tasks useful for language learning are those that require or favour communication
through language. The social dimension of tasks to which rnany authors refer finds its roots and
rationale here. Language learning tasks are useless for communicative purposes if they do not
engage learners in communication. And as pointed out above, it is obvious that not al1 real world
tasks involve this social dimension ('real world relationship', as Skehan (2001: 13) puts it).
Tasks are also goal-oriented. Goals belong to the nature of the task itself. In fact they are
the ultimate trigger that moves the student to engage in a task. In pedagogical tasks, however,
we deviate from the primary goals of real world tasks and add new ones: language use in
performing the task is a requirement. 'Buying a ticket' for travelling to New York implies that
al1 the stages in the development of the task will be oriented to 'buying the ticket', and not to
other purposes. Anything that interferes with the pursuit of this goal will be an obstacle and bring
IJES, vol. 4 (l), 2004, pp. 39-71
The Tusk-hused Approuch in Lunguuge Teuching
53
a lower degree of efficiency into the action. When this task is brought into the classroom and
turned into a 'pedagogical task', an important change takes place. Students may devote some
time to finding unknown words in a dictionary; they may ask the students for some syntactical
problem; they may repeat the same word or sentence several times before they find the correct
way of eliciting it; they may waste several minutes in understanding a message, etc. Nobody in
the classroom context is really worried about the positive outcome of the task -buying a
ticket- but about something else: building correct discourse, finding the right words and
registers to ask for a price or a place. The 'linguistic dimension' of the task is what really matters
in the classroom. The primary goal of the task has therefore shifted from its original real world
value to another one centred on language (not on tickets). The pedagogical task takes a real
world task as a pretext for achieving different goals. There is a close relationship between real
world tasks and pedagogical tasks based on them, but their primary goals are different.
Actions for carrying out a task are subordinate to the goal defining the task. At the same
time, procedures underlying the actions and the process as a whole must be designed with
efficiency-guided criteria. The conclusion is, then, that if goals change, the actions to be taken
will most likely have to change as well. This is the case of real world tasks versus pedagogical
tasks. Changes in the nature of primary goals, as found in pedagogical tasks, require changes in
the strategies and actions for attaining those goals as found in real world tasks.
The current literature of TBA does not seem to pay enough attention to this fact (see Ellis
2003, however). Features typical of real world tasks are automatically assigned to and required
from classroom tasks, disregarding decisive differences between them. This will result into a
mismatch, with undesirable consequences for language learning.
Those are some of the most salient and contrasting differences:
i) Real world tasks emphasize the use of the right strategies and rely on choosing the right
actions to achieve the desired goal. The nature of those strategies and actions is not necessarily
linguistic. Moreover, Ianguage use may not be necessary at al1 for performing some tasks.
However, second language learning classrooms are obviously centred on tasks that involve the
use of language. And this is not the only difference to be noticed. Not al1 tasks requiring the use
of language aim at language as the most important goal to be reached. Most often the use of
language is a means to an end, but not the end itself. In those cases language use is of a merely
instrumentaI character. Real world tasks consist therefore of operations with a goal in mind, but
these operations are not necessarily of a linguistic nature; performance requires attention and
skills, but not necessarily linguistic skills; they require a focus on what is being done, but not
necessariIy a focus on linguistic meuning. The tasks language teachers refer to are tasks of a
specific kind and nature and they must be studied and analysed under this perspective.
ii) Sequencing of activities does matter in language teaching (Sanchez, 2001). The right
outcome of a task depends largely on how the various steps aiming to the final goal follow each
other. Sequencing also plays an important role in understanding the process. As the cause-effect
sequence proclaimed by Gouin (1 892) is supposed to be self-explanatory, the goal which tasks
IJES, vol. 4 ( l ) , 2004, pp. 39-71
aim at guides and conveys meaning to the whole, and more specifically to the steps that pave the
way to such a goal. This is an advantage: they help students to understand what is going on in
the process without explicitly being told about it. It is the natural capacity of human beings to
interpret reality 'following the logic of nature and natural events' that makes it possible to infer
the meaning. The problem we rnay find in pedagogical tasks is that the natural sequence of the
operations of real world tasks has often to be adapted to the classroom situation, in which case
the original order of events is distorted. The usual sequence of events expected when performing
the task ('intra-task operations') of 'buying a ticket' in real life and when 'buying a ticket' in the
classroom, as a class activity, illustrates the differences we rnay find. Tasks as carried out in the
classroom tend to deviate from real world tasks; in doing so, the 'natural sequence of events'
changes into a 'pedagogically conditioned sequence of events' fully dependent on linguistic
needs and skills. The potential of natural sequencing for inferring meaning will hardly work in
those circumstances.
Experimental research on the patterns learners adjust to in their learning process is far
from being conclusive. The P-P-P model (Presentation-Practice-Production) is usually
considered to be the most frequent in the classroom. If that is so, we should bear in mind that
task completion rnay follow a different scheme. Can we expect learners to adapt 'automatically'
to a novel and constantly changing sequential pattern in the activities they engage in -as real
tasks most often require? If that is not the case, we would be setting the conditions for a conflict
in language learning. Avoiding such a conflict requires that tasks developed in the classroom
take into account the sequence of the steps leading to the final outcome and how well they fit the
learning sequence the students are used to.
iii) A third feature highlighted in the literature on TBA refers to interaction. Tasks, it is
said, favour or require communicative interaction among people. Again, this assumption refers
to classroom tasks, not to any task. Tasks in real life do not always involve interaction with
others, as we al1 know. It is true, however, that only tasks that involve linguistic interaction are
useful for language teaching and learning. But then it would be wrong to transfer features typical
of pedagogical tasks to tasks in general. TBA must count on important restrictions when
introducing tasks in the classroom and accept as a pre-condition that they have to be carefully
selected and adapted to the teaching situation. Interaction cannot be taken to be an essential
characteristic of tasks, but rather as a feature necessary for a task to be useful when brought into
the classroom.
iv) Real world tasks cover any aspect and field of human activity. Their performance rnay
require the activation of various human resources, physical and cognitive, but language skills are
not always needed to carry out a task. Ifwe take that into account, it is difficult to accept that al1
tasks rnay be equally efficient if applied to the language classroom.
v) If the underlying structural skeleton of real world tasks is taken as a model for
elaborating classroom activities, a further conclusion can be reached. Tasks rnay admit any kind
of goals, including purely linguistic ones. There is no reason to exclude linguistic goals from
O Servicio d e Publicaciones. Universidad de Murcia. All rights reserved.
IJES, vol. 4 (l), 2004, pp. 39-71
The Tusk-bused Approuch in Lunpuge Teuching
55
language tasks. In other words, it does not belong to the nature of 'task' that they be centred on
meaning alone. As already commented above, here we face a conceptual problem in relation to
tasks. Advocates of tasks in language teaching have appropriated the word and adapted its
meaning to their own methodological beliefs. Methodological points of view are, however,
different, as are the interpretation and adaptation of task-based approaches to the teaching scene.
The reasons for associating tasks with activities focused on meaning, or with process-guided
syllabuses must lie somewhere else, not in the nature of the task itself.
It must once more be concluded that TBA needs a re-evaluation and perhaps a
reinterpretation of some of its basic assumptions, particularly the nature of the tasks that are
useful for the language classroom. Among other things, the idea that teaching through tasks is
a 'natural' process, assuming that tasks involve 'natural' procedures does not seem to fit in with
reality. Tasks in the classroom may be useful for teaching and learning, but they are, after al],
'pedagogical' devices, among others available. They need and require adaptation for teaching
purposes. And in the process they are subject to pedagogical manipulation.
COMPLEXITY AND SEQUENCING OF TASKS

Real world tasks vary in leve1 of complexity. And so do pedagogic tasks. It is not a simple and
easy matter to consider that a task may consist of just one or many activities. Moreover, a task
may involve some other tasks or sub-tasks as well. 'Buying a ticket' may imply calling the travel
agent, catching the bus and paying the fare, discussing the options available, comparing prices,
describing holiday resorts, or looking for different options on the internet, reading the types of
tickets available, etc. A simple, one-activity task fits any syllabus or classroom with no problem
(linguistic resources needed for its performance are easy to define); a task that consists of several
activities, or tasks involving other tasks or 'subtasks', may be extremely demanding in terms of
words and structures required for their implementation. Learners will have to cope with a
relatively easy communicative situation when a task requires only one activity, but the
requirements will overcome them when facing a more complex chain of communicative
sequences. Teachers for their part will find similar difficulties in 'organizing' or managing
learning in such circumstances.
Some authors claim that a TBA should not submit to previously defined syllabuses. And
they argue that predefined syllabuses are typically content-based (what to teach is previously
defined), while task-based syllabuses should be process-and meaning-guided, where learners
build their own syllabus. according to their needs, and find their own learning path applying their
innate capacity to fulfil the communicative requirements derived from the task in which they
engage. The protagonism of the learners in defining the syllabus should exclude some of the
common questions in syllabus design, such as the one concerning the sequencing of learning
materials from outside. This issue has kept many authors busy (Skehan 1996; Willis, J. 1996,
among others). It is not always clear when such sequencing should be carried out (before or after
IJES, vol. 4 ( l ) , 2004, pp. 39-71
the materials are used in the classroom), or who will be responsible for it (the group of learners,
teacher, teacher plus learners, or syllabus specialists). Coherence is needed when taking those
decisions and the theoretical tenets of the TBA cannot be left aside. Sequencing is an important
problem. After all, even when children learn their first language, acquisition takes place
following a rather universal pattern, which is apparently based, more than on anything else, on
the simplicity vs. complexity axis. From a formal point of view, more simple is that which is
integrated by a lower number of elements; increasing the number of elements means gaining in
complexity. If we approach the question from the point of view of semantics (meaning). the
criteria are similar: a simple 'thought' consists of fewer ideas or semantic units, while a more
complex 'thought' involves more ideas or semantic units. We know that children first learn
sounds (the most simple phonetic units) and from this they turn into more complex phonological
units or sequences of sounds: syllables, words, phrases, simple sentences, and subordinated
sentences.
Breen (1 987b:163) advocates sequencing tasks
... on the basis oftwo sets of criteria or on the basis of relating the two. These criteria are: (i) the
relative familiarityof the task to the learner's current communicative knowledge and abilities,and
(ii) the relative inherent complexity of the task in terms of the demands placed upon a learner.
One might take those words as the criteria to be applied so as to sequence the syllabus
before teaching or learning begin. And this can be inferred from Breen's discourse, when he
refers, for example, to the 'task designer', or to 'learning tasks planned in advance' (Breen
1987b;164). But, apparently, Breen does not mean what he has previously said: plaming from
outside, or planning before the group of learners sets to work would contradict the basic
principies of TBA. Breen seems to be conscious of that restriction when he adds,
These criteria which may guide planning are only half the story. The sequencing of tasks [...]
cannot be worked out in advance. Sequencing here depends upon first, the identification of
leaming problems or dificulties as they arise; second, the prioritising of particular problems and
the order in which they may be dealt with; and third, the identification of appropriate learning
tasks which address the problem areas.
Breen (19876;164)
This is more in accordance with Breen's thesis, which assigns protagonism to the learners
in defining their own path of learning. How could you sequence tasks that have not yet been
selected? Even the possibility of sequencing is really at stake in this view: the selection of tasks
depends on the daily needs of the students, so that it is not possible to have a list of them in
advance. If such a list is not possible, what can you expect to sequence? Tasks already learnt in
the classroom must be necessarily excluded from sequencing. The fact is that in a learner-centred
curriculum sequencing of tasks has no meaningful role to play: sequencing requires some kind
of organization in advance and some materials to grade, and both are necessarily absent in a
IJES, vol. 4 (l), 2004, pp. 39-71
The Task-based Approuch in Language Teaching
57
'non-existent' syllabus.
But if 'there is to be a syllabus', Breen mentions 'the inherent complexity of tasks' as the
second criterion to be taken into consideration. Here 'complexity' has to be understood in terms
of the 'demands placed upon the learner'. Those demands, 1 assume, will derive from,
the identification and selection of adequate linguistic forms,
the difficulty of those forms. measured especially in relation to the mother tongue (how
'far' they are from regular and usual patterns in the Ll),
the number of formal linguistic elements required by the task,
the difficulty in arranging those formal elements,
the amount of ideas or semantic units the learners will have to manipulate or control,
the cognitive demands that the elicitation of those ideas will require from the learners,
the degree of efficiency and accuracy needed for communicating those ideas to others.
Such sequencing does not deviate much from what should be expected in other types of
syllabuses.
If we take the lesson as the basic unit of analysis, several activities are implemented
during the 50 minutes it usually takes. The same scheme will be found if the unit of analysis is
a whole academic year, or a two, three or four year curriculum. The activities implemented may
have been planned in advance, or may have been selected and developed by the students
depending on what they feel they need at each specific moment of their leaming path. In al1
instances the activities will be chained and constitute a sequence. 1s it necessary or pedagogically
convenient to control such a sequence? Most authors (Candlin 1987; Nunan 1989; Skehan 1996;
Ellis 2003:220 ff.) advocate criteria for 'sequencing tasks'. One of the models proposed is based
on the complexity of the code (formal code) and the complexity of the content (conceptualisation
of what one has to communicate). Other models are more specific (comprehending, production,
interuction, Nunan 1989: 118). In any case the need for sequencing tasks does not derive so much
from the nature of TBA, but rather from the nature of the leaming process itself. Human beings
seem to be conditioned to proceed, when they learn, from the most simple units or elements, to
the more complex ones. And that is so regardless of what we learn. Sequencing, therefore, is
subject to similar principles when applied to task organization, to the selection of the different
steps that may constitute the task, or to the linguistic elements used for performing the task. The
most general principle governing learning is guided by the transition from simplicity to
complexity.
This principle has many facets and offers a wide variety of perspectives. Structural
methodologies used to refer to the number of elements needed to build structures or sentences.
or the number of phonetic features of a sound, or the number of morphemes in a word. A
communicative methodology analysed the same concept from the point of view of elements
needed for communicative functions. A further filter was later applied when the linguistic
IJES, vol. 4 (l), 2004, pp. 39-71
58
Aquilino Snchei
exponents of functions were selected or not depending on the number of elements included
(morphological complexity) and the syntactical patterns involved.
Complexity is not the only criterion, though. The learning of language does not take place
on a fully linear basis, by adding simple elements to the more complex ones that we have already
acquired. We also learn language 'by chunks', that is memorizing and consolidating 'pieces of
language' regardless of their degree of complexity. In those cases, the criterion of complexity
does not apply, while others seem to be present, such as a special or urgent communicative need.
the familiarity of the learner with the topic of the task, the previous experience of the learner
with what is being learnt, etc. From that point of view, the claim of the TBA in favour of a
learner-centred syllabus and the primary role of the learner in deciding on what should be taught
and learnt, and therefore on which tasks must be planned, is reasonable. Some gaps have to be
reserved for coping with factors other than complexity in the learning process. The problem or
difficulty lies in defining how much emphasis should be given to each of the various and relevant
criteria applicable.
Tasks, as already discussed above, vary in complexity and focus. Moreover, language
tasks are not to be equated with other tasks which do not require the use of language for their
development. Specific learners may have the cognitive skills to perform aparticular task, but not
necessarily the linguistic skills to do or explain what they are doing in the foreign language they
learn. And most important, tasks based exclusively on linguistic skills, may not be feasible
because learners lack precisely those linguistic skills, for example, because they do not know the
right words for the concepts they have in mind. In order to cope with this problem, some authors
(e.g. Skehan, 1996) advocate three types of tasks, which have to be introduced into the learning
process in this sequence: pre-task, during-task, post-task. Pre-tash begin the sequencing by
introducing the language needed for task performance. Skehan (1996:54) refers to that stage as
'some form of pre-teaching'. That sounds very much like the classical 'presentation stage' within
the already classical sequence P-P-P. And the activities included would be comprehension-type
activities. Performance during the task takes place when the task is selected and learners engage
in fulfilling the goals of the task. To succeed in that goal, manipulation of language is required
and here severa1 factors must be taken into account for the students to proceed successfully.
Again, Skehan (1996) mentions 'cognitive complexity', degree of difficulty (neither too easy,
nor too difficult) and the correct management of the 'communicative pressure'. Once the goal
has been reached, post-task activities will need to take care of refining what has been learnt.
Students may pay too much attention to fluency, because they already know how to perform the
task; in that case emphasis could perhaps be focused on accuracy or restructuring, for example.
That will require further practice, repetition of the task. analytical reconsideration of what has
been done and how it has been done (the cognitive dimension the implementation of parallel
tasks, etc.
The implementation of tasks in the classroom admits variety. Willis, J. (1996) describes
the 'task cycle' in three phases, but in more detail and closer to real practice in the classroom:
I J B , vol. 4 (l), 2004, pp. 39-71
The Task-based Approach in Lunguage Teaching
59
a) Pre-task phase: the description of the different activities suggested by Willis takes up
again the 'presentation stage' most teachers are used to.
b) The task cycle: three stages are mentioned here: task, planning and report. In al1 of
them students are supposed to produce natural language, gain fluency and confidence in
themselves. Willis and Skehan insist on the need for grammatical accuracy, once the task
has been performed. And this is so because during task performance the learners pay
attention to meaning and tend to forget grammatical correctness.
c) Language focus: this emphasizes specific language features. lt seems as if the author
is considering here a kind of 'remedial' final task. Since focus on meaning should have
been the rule throughout the two previous phases, it is now time for 'language focus
activities'. Language activities refer to semantics, lexis, morphology, syntax and
phonetics/phonology.
Planning in advance the sequencing of tasks, or activities within a task, does not fit well
in a process syllabus, centred on learners and learning, with the students having the main
responsibility for building their own learning path. But some basic facts of leaming and the
classroom environment and practice appear to outweigh theoretical and abstract
conceptualisations. This seems to be the case of TBA regarding sequencing: its theoretical
framework claims to fully transfer to the learners the elaboration of the syllabus and therefore
the organization of the tasks and activities. But 'real teachers and real learners' are reluctant to
accept this approach, sometimes with sound arguments on their side. The more TBA is analysed
and the more this issue is investigated. the more one is forced to admit that a planned sequencing
of activities is needed. The challenge is to offer a sequencing scheme without seriously hindering
the linguistic potential of the learner and his capacity for getting himself involved in the learning
process.
A BALANCED CRITICISM
The role of meaning in language has been heavily emphasized in the communicative approach
and in al1 its methodological variants. From this perspective, earlier practices and methods have
been partially undervalued or distorted. lt is true that in the structural approaches emphasis was
not placed on meaning, but one cannot simply say that meaning was totally disregarded in this
method. A parallel can be found in the way al1 methodologies react against previous approaches:
they tend to build a simplistic and distorted picture of the 'newcomers', stressing contrasts that
will work in favour of the new elements proposed. ln the 'new method' emphasis will also be
placed on a limited set of features, while the other intervening elements will be pushed into the
background. TBA is not an exception to this rule: emphasis on meaning and process carries with
it that grammar and discrete goals are relegated to a secondary place and a subsidiary role. In a
similar way, cognitive factors in leaming, traditionally associated with grammar and form, are
IJES, vol. 4 ( l ) , 2004, pp. 39-71
given a minor part to play or not taken into consideration. Early enthusiasm about TBA has been
tempered somewhat and it is becoming increasingly clear that emphasis on meaning alone does
not result in more effective language acquisition. As usual, a balance must be found between the
role assigned to meaning and the necessary focus on linguistic form. The complexity of the
language acquisition process demands a more realistic approach to the teaching-learning
situation.
Dissatisfaction with learning and its outcomes is and has been common among learners.
But such a feeling is not exclusive to one specific methodology alone. Complaints about poor
results in second language learning are particularly outstanding in the educational system al1 over
the world and are well illustrated in the history ofpedagogy. Students begin the second language
curriculum when they are 6 years old and they keep studying a second language for ten academic
years or more. In spite of that, most secondary students leave school unable to communicate
effectively in the language they have supposedly learnt. lnstruction has not been completely
useless. but it does not meet the standard theoretically required in the curriculum. Such a
stubborn reality triggers a permanent reconsideration and questioning of methods in teaching and
is responsible for cyclical shifts in opposite or complementary directions. Present SLA research
and findings may be taken as a reliable point of reference for a more balanced solution to the
problem.
Expectations in second language learning should be significantly moderated and adjusted
to the achievement of reasonable goals. The learning of a second language by teenagers and
adults should renounce the permanent comparison with native language learning; second
language learners cannot expect to become 'native language learners' or 'native language
speakers'. We should rather work within a scale (See figure 1) with a maximurn goal (native-like
results) and a minimum (communicative effectiveness, in written or oral language, or in both
modalities). Perhaps language goals in the school system will have to be defined taking into
account the real situation of language acquisition in a non-natural environrnent, and specifically
in the classroom environrnent. To that must be added the fact that second language learning is
not usually given priority in the curriculum and the educational environment for using a foreign
language is far from ideal.
f
n a r i v e - I i k e
comnrunicative
effectiveness
n a r i v e - l i k e
communicative
effectiveness
TBA. like al1 cornmunicative approaches, tries to recreate natural learning conditions in the
classroom. The reason is that natural conditions are assumed to be ideal for learning a language,
0 Servicio de Publicaciones. Universidad de Murcia. All rights reserved.
IJES, vol. 4 ( 1 ) . 2004, pp. 39-71
The Tusk-based Approuch in Lunguage Teuching
61
as already proclaimed in the history of language teaching. Now this view is put forward together
with other related or supporting assumptions, some of them connected to SLA research:
Everyone is bom with innate abilities to learn a language (which makes instruction not
totally necessary).
Experience tells us that many people leam two languages and they keep using them
successfully throughout their lives (therefore, one can infer that the learning of a second
language should not be so difficult).
Failure comes when formal instruction takes place, where the focus is on form and
intellectual abilities, instead of focusing on meaningful interaction (moving away from
the natural leaming process).
Personal involvement in the learning process is a key factor for more efficient leaming
(motivation is a decisive factor in leaming).
The issue, however, is whether 'natural conditions' can be recreated in the classroom. If
the kind of tasks we can work with in the classroom are different from real world tasks a s
explained above- it is difficult to agree with the TBA, which aims to recreate such a natural
learning environment precisely by means of tasks, when they have to be necessarily 'pedagogical
tasks'. Other assumptions do not necessarily apply to adult learning. It is a well-known thesis
that innate abilities of human beings for learning languages do not last forever with the same
intensity. And the fact applies regardless of whether we have leamt one or more languages in a
natural environment.
Research on SLA (Long and Crookes, 1992; Pienemann and Johnston, 1987) does not
support the conclusion that instructed leaming is a failure either; rather, the contrary seems to
be the case: instruction helps learning. Instruction, we know, is not essential for acquiring
languages in a natural environment, but it definitely helps if we want to gain in accuracy and,
as a result, also in efficiency and in a more refined communicative capacity.
Most methodologists and researchers in SLA admit today that second language leaming
is favoured when at least three conditions are met:
a) That learners be exposed to the language. There is a direct relationship between
exposure to the language and linguistic acquisition. Exposure counts as a necessary input
phase before the learners are able to generate any output and refers both to the oral and
written language. Research in SLA has reinforced this hypothesis, first formulated by
Krashen.
b) That learners use the language and practise with it, especially in a communicative
context. Practica1 work with language may be quite varied, though. Communicative use
of language is one of the factors that usually increase motivation.
c) That leamers are motivated to use the target language orally and in writing or reading.
IJES, vol. 4 (1). 2004, pp. 39-71
Points b) and c) are not new in pedagogy, at least in their basic claims. Point a) has been
intensively revindicated by the communicative method and the TBA is consistent with its
communicative roots in assuming a similar claim. TBA does not seern to be based on new
learning principies. Rather it offers a novel way of being exposed to and practising the language,
and at the same time involving and motivating the student. This novel way is the task. Practising
and using the language by means of a task is supposed to produce more effective learning.
What about formal instruction? Teaching in the classroom offers some advantages. but
is also subject to important restrictions. Two of them are particularly relevant here: instruction
usually predefines the content of what is to be learnt, and the learning conditions do not favour
simulation of real world tasks or the recreation of the natural environment in the classroom. It
is obvious that not al1 methodologies fit equally well into this framework. The grammartranslation method, as well as the structural approaches, goes hand in hand with what is required
from teachers and students in the classroom. But other methods do not, as was the case with the
Direct Method, or with Gouin's method, and now with the TBA. The reason lies in the nature
of real world tasks; they cannot enter the classroom without undergoing substantial changes. The
adaptations that real world tasks must undergo in the classroom eliminate some of the original
and natural communicative features attributed to tasks in TRAS. As a way of illustrating the
issue. people who perform real world tasks 'in real daily life' already have a good command of
the language, or perform tasks suited to their command of the language; they do not need to pay
explicit attention to linguistic use. When students of a language do a task in a learning situation,
the context varies significantly, since they have to do the task and at the same time learn the
linguistic elements needed to reach the desired goal. Unknown words or structures will not come
automatically to the mind of the learner. There must be some effort on hislher part to look for
those words and restructure thern with other linguistic elements already available. This action
takes place unconsciously in a natural environment, when learning the mother tongue. But then
we should consider how different the situation is in the foreign language classroom (age,
'desperate need' to learn to communicate with the community, favourable context, no time
pressure.. .).
Formal instruction in general should not be underestimated as a useful and perhaps
indispensable tool for transmitting human knowledge. The same applies to language learning.
The challenge for the TBA is how to integrate the necessary instruction within the set of
activities derived frorn pedagogical tasks and centred on meaning. Such models must be still
designed, applied and evaluated.
Advocates of innovative approaches tend to be over-enthusiastic about their potential.
It is rewarding to read such paragraphs as this one on the TBA:
In task-based learning, cornrnunication tasks (where language forms are not controlled) involve
learners in an entirely different mental process as they cornpose what they want to say, expressing
what they think or feel.
J. Willis (lYY3:lCI)
IJES, vol. 4 (l), 2004, pp. 39-71
The Tusk-hused Approuch in Languuge Teuching
63
But there are sound arguments against and serious doubts about every one of the
statements produced by this author. Are they to be taken as 'wishful thinking'? Teachers are well
aware of how difficult it is for a student to express 'what he thinks or feels' in a foreign
language, unless there is a lot of previous work on what has to be said. What the nature of such
'previous work' is remains very much the question methods try to solve. The TBA tries to do
it through task work. But tasks point to a final outcome, and what is to be done on the way to this
is the question: something previous is required to succeed in task performance. 1s that formal
instruction? Or just focus-on-form instruction (Ellis 2002, 2003)? Or formal instruction plus
practice? Or formal instruction plus practice plus cognitive consciousness about the language
being learnt?
Explicit instruction refers to al1 the events and actions affecting teaching in the
classroom, such as organizing the syllabus, deciding on the tasks to be selected and on their
sequencing, elaborating the activities required by each task, giving advice on the suitable
linguistic elements for the task +specially if there are several options, as is the case for most
tasks-, etc.
Instruction implies that the cognitive component of the learners is activated. There is
nothing wrong in that. Cognitive skills constitute a decisive difference between humans and
other animals. They should not be put aside but advantage should be taken of them. It is true that
classroom practices tend to emphasize formal aspects of language. Again, this is not a negative
feature of language use: accuracy and fluency do not contradict each other: rather, they
complement each other.
The importance of comprehensible input in language learning was first emphasized by
Krashen (1982; 1985). Since then. this has been a generally accepted principle. TBA is firmly
rooted in that principle. Tasks are considered ideal tools for providing meaningful linguistic
materials; at the same time, carrying out the tasks requires the use of language materials and
even recursive practice with them. Does that mean that comprehensible input alone is enough?
1s there any evidence that forces the exclusion of instruction as additional cognitive input?
Findings on this issue point in the opposite direction: productive skills are rather poor in students
with a high degree of comprehensible input (Sheen 1983: 136), while receptive skills are certainly
favoured by intensive exposure to the language. Experience by many teachers and learners goes
along with these findings: language learnt in the natural environment gains in fluency, but
abounds in formal inaccuracies; instructed learning produces more accurate output, but is poorer
in fluency (this same analysis is ofien found in the history of language teaching; see Sanchez.
1992).
Instruction, therefore, is not negative per se, as one sometimes perceives when reading
about TBA and process approaches. lnstruction is a helpful 'tool', and so it must be kept. The
history of mankind supports this positive view. But it is nota 'perfect tool'; nor is it the only tool
for learning. However one analyses the issue, the conclusion is simiIar and valid for al1
IJES, vol. 4 ( l ) , 2004, pp. 39-71
64
Aquilino Sunchrz
methodologies: success in teaching and learning depends mainly on how well the complexity of
the teaching-learning situation is handled. A key word has to be mentioned again: balance.
The need for a balanced methodology is not exclusive to the TBA. A review of methods
in the history of language teaching reveals how strongly and sometimes fiercely the new methods
rail against the current ones. And opponents base their claims on the failure of old methods to
facilitate learning. We know, nonetheless, that many people have succeeded in their learning of
languages with al1 methods. This fact forces us to believe that perhaps too often innovative
methods gain their share of prestige in the methodological scene by claiming failure and
dissatisfaction with other methods. The cycle, however, will not end, and the innovative methods
will soon become obsolete. To think of new methods basically as attempts to adapt to the
specific needs of the time would be a more reasonable view. It would also be wrong to reject any
methodological innovation because 'the majority of educational innovation results in failure'
(Adams and Chen, 1981, in Sheen, 1994: 127).
The TBA has gained some momentum in the nineties. The group of scholars 'leading the
way' seem to base their methodological claims on a limited set of assumptions and principles,
which are generously endowed with well sounding words and concepts. Their arguments are not
always convincing and research in the new approach is still insignificant. This pattern is not new
in the history of educational innovations: most of them offer a similar profile in this respect. The
question is whether the teaching-learning situation gains in efficiency. The answer to this
question is far from being conclusive or positive.
The TBA must be included within the inductive methods: learners are supposed to
internalise the linguistic system through practice. The Direct Method, the Audiolingual Method
and approaches generically called 'communicative' are also inductive. Much has been written
on the advantages or disadvantages of both ways of learning. But experimental research on
methods reveals that deductive methods or their variants produce better results than inductive
ones (Sheen 1994:129-130). These findings may already introduce a caveat to the TBA. A
second factor that is relevant here is the role of formal instruction in language learning. Process
approaches -and the TBA is to be placed within this mainstream- are well known because
they stress the role of the students and marginalize or assign a secondary role to the action of the
teacher and instruction. In so doing, research on the positive effects of deductive methods is,
deliberately or not, simply stranded. Since formal instruction is connected to deductive
methodologies, it has to be underestimated, as indeed it is in the TBA. Sheen (1094: 133ff) brings
in some data that help to point out some contradictions in Long's (1988) and Long and Crookes'
(1992) arguments regarding the role of instruction in language learning. 1 referred above to
similar 'conceptual gaps' in the views of some advocates of the TBA, who sometimes seem to
argue on the basis of preconceived, but not proven, principles.
IJES, vol. 4 ( l ) , 2004, pp. 39-71
65
The Task-based Approach in Lunguuge Teaching
CONCLUSION
Looking back into the past illustrates what is really new in the TBA. The emphasis on the
communicative learning or teaching of languages is not new, but it offers at least a partially
different way of being exposed to and practising with the language. This leads us to classify
TBA within the 'conversational andlor natural approach' (Snchez 1992; 1997). But new
methods are not to be taken as innately good and efficient by their nature, or simply because they
are new. It must be admitted that the TBA faces most of the problems inherent in natural
methods, particularly when applied to adult learners of a second language. The difference
between real world tasks and pedagogical tasks is at the very centre of the problem. The
classroom environment cannot be equated to the real world environment, or at least not fully
equated to it. In a parallel way, leaming a language in a natural environment -particularly in
the early stages of life-differs considerably from leaming a language as an academic subject.
The history of teaching languages offers a long list of methods to teach and learn
languages. 1am not suggesting in this paper that we should reject new proposals and novel ideas,
but 1 strongly feel that what we urgently need is to do more research on the mechanisms of
learning and accompany new proposals and methods with more experimental evidence before
we bring them into the classroom. Abstract constructs may be well elaborated and their elements
may also be logically intertwined, but something more than that is needed for them to work in
practice. The TBA adds useful elements and contributes to the communicative language teaching
with valuable procedures. But it would be naive to take it as 'the method' language teachers and
learners have been waiting for.
BIBLIOGRAPHY AND REPERENCES

Adams, R. aiid Clien, D. (1981). The Process qfEducationa1 Innovation. An international Perspective.
Loiidoii: Kogan PressUNESCO Press.
Allwright, R. L., (1984). 'Wliy don't learners learn wliat teachers teach? The interaction hypothesis'. iii
D. Siiigletoii and F. G. Little (eds). Language Leurningin FormalandInformal Contexts. Dublin:
IRAAL.
Austiii, J. L., (1962). How to Do Things with Words. Loiidoii: Oxford University Press.
Berwick, R. F., (1988). The Efect of Task Variution in Teacher-Led Groups on Repair ofEnglish as A
Foreign Language. Uiipublislied doctoral dissertation. Vancouver, Canada: University of Britisli
Coluinbia.
Boydell. T. H., (1970). A Guide to Job Analysis. London: Bacie.
Breeii M. P., (1984a). "How would we recognise a coinmunicative classroom?". En B. Coffey (ed).
IJES, vol. 4 (l), 2004, pp. 39-71
Teacher Training and the Curriculurn. The Dunford House Seminar. 1982. London: The British
Couiicil.
Breeii, M. P., (1984b). Process syllabuses for the language classroom. 111C. J. Brumtit (ed.). General
English Syllabus Design. (ELT Documents No. 118: 47-60). London: Pergarnon Press aiid The
British Council.
Breen, M. P., (1987a). Learner contributioiis to task design. In C.N. Candlin and D. Murphy (eds.).
Lunca.srer Prucfical Papers in English Languuge Educufion. Vol. 7. Lunguage Learning Tasks,
23-46. Englewood Cliffs, NJ: Prentice Hall.
Breeii, M. P., (1 987b). "Conteinporary paradigms
157-1 74.
iii
syllabus design",
iii
Lunguage Teaching, 20 (3):
Breen. M. P., C. N. Candlin, A. Waters, (1979). "Cornrnunicative rnaterials design, sorne basic
principles", RELC Journal, 1 012.
Breeii, M. P. and C. Candliii, (1980). The essentials of a commuiiicative curriculum
Applied Linpislics, l(2): 89- 1 12.
iii
language teacliing.
Brindley, G., (1 989). Assessing Achievernenf in the Learner-Cenfred Curriculum. Sydney, Australia:
Macquarie University, National Centre for English Language Teaching and Research.
Brown, G., A. Anderson, R. Schilcock, and G. Yule, (1 984). Teaching Talk. Strategies for Producfion
and Assessment, Cainbridge: Cambridge University Press.
Bygate, M., (1994). 'Adjusting the focus, Teacher roles in Task-based learning of grammar', in Bygate,
M, A. Tonkyn and E. Williains (1994). Grarnrnar and [he Lunguage Teacher, New York:
Preiitice Hall International.
Candlin, C. N,, (1984). Syllabus Design as u Critica1 Process. ELTDocurnenfs No. 1 18: 29-46. London:
Pergamoii Press and The British Council.
Caiidlin. C. N., (1987). Towards task-based language learning. In C. N. Candlin and Murphy (eds.).
Luncasfer Pracficul Pupers in English Language Educufion. Vol. 7. Language Learning Tasks,
5-22. Englewood Cliffs, NJ: Prentice Hall.
Candliii, C. N,, J. Brutoii, and J. M. Leather, (1976). Doctors i i i casualty, Specialist course design froin
a database. InfernafionalReview ofApplied Linguistics. 14: 245-272.
Candlin, C. N. and D. Murphy, (eds.). (1987). Lancusfer Practica1 Papers in English Language
Education. Vol. 7. Language Learning Tasks. Englewood Cliffs, NJ: Prentice Hall.
Carr, T. and T. Curreii, (1 994). 'Cogiiitive factors in learning about structured sequences, Applications
to syntax'. Studies in Second Language Acyuisifion. 1612: 205-230.
Chaudron, C., (1988). Second Lunguage Clussroom.. Research on Teaching und Learning. New York:
Cambridge Uiiiversity Press
Crookes, G. and M. Gass Susan, (eds) (1 993a). Tasks in a Pedugogical Confext.Clevedon: Multilingual
Matters.
IJES, vol. 4 (I), 2004, pp. 39-71
67
The Task-bused Approach in Languuge Teuching
Crookes, G. aiid M. Gass Susan. (eds) (1993b). Tasks and Lunguage Learning. Clevedon: Multilingual
Matters.
Crookes, G.. (1 986). Task Classij'icafion. A Cross-Disciplinary Review (Tech. Rep. No. 4). Honolulu:
University of Hawaii at Manoa, Social Science Research Institute, Center for Second Language
Classroom Research.
Crookes, G., (1 989). Plaiiiiing and interlaiiguage variation. Sf~rdiesin Second Languuge Acquisition.
1 l(4): 367-383.
Crookes, G. and Loiig, M. H., (1987a). Task-based secoiid laiiguage teaching. A brief report (Pt. 1).
Modern English Teacher. 24(5): 26-28.
Crookes, G. aiid M. H. Long, (1987b). Task-based second language teachiiig. A brief report (Pt. 2).
Modern English Teacher. 24(6): 20-23.
Doughty, C. and T. Pica, ( 1 986). "lnformation gap" tasks. Do they facilitate second laiiguage acquisition?
TESOL Quarferly. 20(2): 305-326
Dubin, S. and E. Olslitain, (1 986). Course Design. Developing programs and materials for language
learning. Cambridge: C.U.P.
Ellis, R., ( 1 985). UndersfandingSecond Language Acquisition. Oxford: OUP.
Ellis, R.. (2002). 'Does form-focused instruction affect the acquisition of implicit knowledge? A review
of tlie research. Sfudies in Second Language Acquisition 24 (2): 233-236.
Ellis, R., (2003). Task-based Language Learning and Teaching. Oxford: Oxford University Press.
Fernandez, S., (ed.). (2001). Tareas yproyectos en clase. Madrid: Edinumen
Foley, J., (1991). 'A psycholinguistics framework for Task-based Approaches to Language Teaching',
Applied Linguistics, 12, No. 1 : 62-75.
Freire, P., ( 1970). Pedagogy o f f h e Oppressed. Harmoiidsworth: England, Penguiii.
Gass, S. M., (1989). Second Laiiguage vocabulary acquisitioii. Annual Review ofApplied Lingzrisfics.
1988.9: 92- 106
Gouiii, F., ( 1 892). The Arf of Teaching and Studying Lanpages. Londoii: George Philip and Son
(Englisli traiislation by H. Swan and V. Betis).
Halliday, M. A. K., ( 1973). Explorations in the F~rnctionsufLanguage. Loiidoii: Edward Ariiold.
Halliday M. A. K., (1978). Langiruge as Social Semiotics. London: Longinaii.
Hyltenstain. K., (1988). Lexical cliaracteristics of near-native second laiiguage leariiers of Swedisli.
Journal qf Mulfilingual and Mulficultrrral Dcvelopment. Y( 1 and 2): 67-84
Johnson, K., (2003). Designing Language Te~ichingTasks. Houndmills: McMillan.
IJES, vol. 4 ( 1 ), 2004, pp. 39-71
68
Aquilino Snrhez
Karmillof-Sinitli, A., (1986). 'From meta-processes to conscious access. Evidence for children's
metalinguistic and repair data'. Cognition. 23: 95-147.
Kellerman, E., (1985). 'If at tirst you do succeed ...', in S. M. Gass and C. G. Madden (eds.). Input in
Second Lunguuge Acquisition. 345-363. Rowley, Mass: Newbury Mouse.
Kouraogo, P., (1987). EFL curriculum renewal and INSET in difficult circumstances. ELT Journul.
41(3): 171-178.
Krashen, S. D., (1 982). Principies und Pructice in Second Lunguuge Acquisition. Oxford: Pergamon
Press.
Krashen, S. D., (1 985). The Input Hypothesis. London: Longman.
Krashen, S. D. and T. D. TerreII, (1983). The NuturulApprouch. Lung~rugeAcquisition in the Clus.sroom.
San Francisco, CA: Alemany Press.
Kumaravadivelu, B., (1993). The name of tlie task and the task of naining: Methodological aspects of
task-based pedagogy, in Crookes, G. and M. Gass Susan (eds). Tusks in u Pedugogicul Context.
Clevedon: Multilingual Matters.
Larsen-Freeinaii, D. and M. H Long, (1991). An Introdzrction to Second Lunguuge Acquisition Reseurch.
Londoii: Longman.
Long, M. H., (1 983). Does second language instruction inake a difference? A review of research. TESOL
Quurterly. 1 7, 3.
Long, M. H., (1 985). A tole for instruction in second language acquisition. Task-based language teaching.
111 K. Hy ltenstain and M. Pienemann (eds). Modelling und Assessing Second Lunguugc
Acquisition. 77-99. Clevedon, England: Multilingual Matters.
Long, M. H., (1988). Instructed iiiterlaiiguage developinent. Iii L. M. Beebe (ed.). Iss~resin Second
Lunguuge Acquisition. Multiple Perspectives. 1 15-1 41. New York: Harper aiid Row.
Long, M. H., (1 989). Task, group, and task-group interactions. UniversiS ofHuwui Working Pupers in
ESL. '(2). 1-26. (Repriiited in S. Aiiivaii (ed.). Lunguuge Teuching Methodologyfor the Nineties.
3 1-50, 1990. Singapore: SEAMEO Regional Language Ceiiter.)
Long, M. H., (1 990). The least a secoiid language acquisition tlieory needs to explaiii. TESOL Qlrurterly.
24(4). 649-666.
Long, M. H., (1991). Focus on form, A design feature in language teaching inethodology. 111 K. de Bot,
D. Coste, R. Giiisberg, and C. Krainsch, (eds.). Foreign Lunguuge Reseurch in 'ross-Culturul
Perspective. 39-52. Ainsterdain: Joliii Benjainiiis.
Long, M. H. and G. Crookes, (1987). Interventioii points iii second language classrooin processes. Iii B.
K. Das (ed.). Putterns in clussroom interuction in Southeust Asiu. 177-203. Siiigapore: Siiigapore
Uiiiversity PressIRELC.
Loiig, M. H. and G.Crookes, (1992). 'Three Approaclies to Task-Based Syllabus Design'. TESOL
IJES, vol. 4 ( l ) , 2004, pp. 39-71
69
Loiig, M. H. and G. Crookes, (1993). Units of analysis in syllabus design. Iii G. Crookes and S. M. Gass
(eds.). Tuslu in lunguuge leurning. Clevedoii, England: Multilingual Matters.
Loiig, M. H. and P. Robiiison,, (2002). 'Focus on forin. Tlieory, research aiid practice. In Doughty G. and
J. Williams (eds.). Focus onfirm in clussroorn SLA. Cambridge: Cambridge University Press.
Martiii Peris. E., (1999). Libros de texto y tareas. In Zann, J., (ed.) 1999: 25-52.
McLaughliii, B., (1990). 'Restructuring'. Applied Linguistics. I I : 1 13-128.
Nunan, D., (1989). Designing Tusks fir the Comrnunicutive Clussroorn. Cambridge: Cambridge
Uiiiversity Press.
Parker, Cecil L. and Rubin, L. J., (1 966). Process as Content. Curriculurn Design und the Applicution oJ'
Knowbdge. Chicago: Raiid McNally and Coinpany.
Pica, T., (1983). Adult acquisition of English as a second laiiguage uiider different coiiditioiis of
exposure. Lunguuge Leurning. 33(4): 465-497.
Pica, T., L. Holliday, N. Lewis aiid L. Morgeiitlialer, (1989). Coinprelieiisible output as an outcome of
linguistic demands on the learner. Sfudies in Second Lunguuge Acqtrisition. 1 I(1): 63-90.
Pica, T., R. Kanagy aiid J. Faloduii, (1993). 'Clioosiiig aiid usiiig communicative tasks for secoiid
laiiguage instruction', i i i Crookes, G. and M. Gass Susan (eds) (1993).
Pieiieinaiin, M., (1984). Psycliological constraiiits oii tlie teacliability and leariiability of laiiguages.
Studies in Second Lunguuge Acquisifion. 6: 186, 2 14.
Pieiiemaiiii. M., (1 987). Psycliological coiistraiiits on the teachability of languages. Iii C. Pfaff(ed.). First
und second lunguuge ucquistion processes. 143-168. Rowley, Mass: Newbury House.
Pienemanii, M. aiid M. Johnstoii, (1987). Factors iiifluencing the development of language proficiency.
I i i D. Nunaii (ed.). ApplyingSecondLunguugeAcquisition Reseurch. 45-14 1. Adelaide, Australia:
Natioiial Curriculuin Resource Centre.
Pienemaiin, M., Joliiistoii, M. aiid Brindley, G., (1988). Constructing an acquisition-based procedure for
secoiid laiiguage assessinent. Studies in Second Lunguuge Acquisition. 1 O(2): 2 17-243.
Prahbu, N. S., (1980). 'Reactioiis aiid predictioiis' [Special issue]. Bulletin 4(1). Bangalore: Regional
Institute of Eiiglish, South India.
Prabliu, N. S., (1984). Procedural syllabuses. Iii T. E. Read (ed.). Trends in lunguuge syllubus design.
272-280. Siiigapore: Singapore University Press/RELC.
Prabliu, N. S., (I987a). 'Language educatioii, equipping or eiiabling', en Das, B. K. (ed). Lunguuge
Educution in Humun Resource Development. Anthology Series 20. Singapore: RELC.
Prabhu, N. S., (1987b). Second Lunguuge Pedugogy. Oxford: OUP.
IJES, vol. 4 ( l ) , 2004, pp. 39-71
Robiiison, P., (1991). Motivating fheories of'task complexi@. Unpublished manuscript, University of
Hawaii at Manoa: Departinent of ESL, Hoiiolulu.
Sanchez, A,, (1 992). Historia de la enseanza del espaol como lengua extranjera. Madrid: SGEL.
Snchez, A,, ( 1997). Los mEfodos en la enseanza de idiomas. Evolucin histdrica y anlisis didcrico.
Madrid: SGEL.
Siicliez, A., (2001). Sequeiicing of activities and motivation', in V. Codiiia aiid E. Alcn, (eds.).
Languuge Learningin the Foreign Lungtrage Classroom. 107-126. Colleci Estudis Filologics/8,
Castellii: Universidad Jaume 1,
Sato, C. J., (1 990). The Syntax of'Conversation in Interlangltage Developmenf. Tbingen: Gnter Narr.
Sclimidt, R. W., (1990a). Tlie role of consciousness in second language learning. Applied Linguistics.
1 l(2): 1 7-46.
e
Sclimidt, R. W., (1990b, July-August). Input. Inferaction. Attention. and Aww-eness. The C u ~ jOr
Consciousness-Ruising in Second Language Teaching. Paper presented at the 1 Oth 'Encuentro
Nacional de Profesores Universitarios de Lingua Inglesa': Rio de Janeiro, Brazil.
Schmidt, R. W., (1992). 'Psychological inechanisms uiiderlying second language fluency'. Studies in
Second Languuge Acyuisition. 14: 357-385.
Sheen, R., (1 994). 'A critica1analysis of the advocacy of the Task-Based syllabus'.
28(1): 127-151.
iii
TESOL Quarterly,
Skehan, P., (1 994). 'Second language acquisitioii strategies, interlanguage development and Task-based
leariiiiig', iil Bygate, M, A. Tonkyii aiid E. Williams (1994). Grammur and [he Language
Teucher. New York: Prentice Hall Iiiternational.
Skehan. P., (1 996). 'A frainework for the implementation of Task-based instruction', AppliedLingiristics.
1996, 1711: 38-62.
Skehan, P., (1 998). 'Task-based instruction'. Annltul Review oj'Applied Lin,qwistics. 1998, 18: 268-286.
Skeliaii, P., (2001). 'The role ofa focus oii form during Task-Based instruction'
(eds). Trabajos en lingistica aplicada. 1 1-24. Barcelona: AESLA.
Iii
Celaya, M" L. and Al.,
Stenhouse, M., (1 975). An infroduction f o curriculum research and deialopment. London: Heineinaii.
Van Patteii, B., (1990). 'Attendiiig to coiitent and form in tlie input. An experiment
Studies in Second Lan~wageAcyuisition. 12: 287-30 1 .
iii
consciousiiess'.
Van Patten, B., (1994). 'Evaluating the role of consciousness in SLA. Terms, linguistic features and
researcli inethodology'. AILA Review. 1 1: 26-37.
Van Patteii, B. and T. Cadierno, (1993). 'Explicit instruction and input processiiig'. Stzrdies in Second
Lanl?.tage Acyitisition. 15: 225-243.
Vygotsky, L. S., (1 962). Thought and Langirage. Cambridge, Mass.: MIT Press.
IJES, vol. 4 ( l ) , 2004. pp. 39-71
71
The Tusk-bused Approach in Lunguuge Teuching
Vygotsky, L. S.. (1978). Mind in Sociew. The Development of Higher Ps.vchologica1 Processes.
Carnbridge, Mass.: Harvard University Press.
Wells, G., (1 981). Learning through Interaction. Carnbridge: CUP.
Wliite, L., (1987). Against compreliensible input. The input hypothesis and tlie developinent of L2
coinpeteiice. Applied Linguistics. H(I ): 95- 1 10.
Wliite, R. V., (1988). The ELT Curriculum. Design. Innovation and Management. Oxford: Basil
Blackwell.
Wilkiiis, D. A,, (1976). Notionul Syllabtrses. Oxford: OUP.
Willis, Jane ( 1996). A Frumework for Tusk-Based Learning. London: Loiigman.
Willis, D. aiid J. Willis, (2001). 'Applying linguistics to task-based learning, six propositions in searcli
for a inetliodology', in De la Cruz, I.& Al., (2001). La lin~q?isticaaplicada afinales del S. XY
Ensayos y propuestas. 16 I - 168. Alcal: Universidad de Alcal.
Zanii, Javier (ed.) 1999. La enseanza del espaol mediante tareas. Madrid: Edinurnen
Yalden, J., (1987). Principies qf Course Design for Language Teuching. Carnbridge: Carnbridge
University Pres
Yule, G., Powers, M. aiid D. McDonald, (1992). 'The variable effects of soine Task-Based learning
procedures o11L2 cornmuiiicative effectiveness'. Languagc Learning. 42-2: 249-277.
IJES, vol. 4 (1). 2004, pp. 39-71

Task Based Learning

Uploaded by

Copyright:

Available Formats

Task Based Learning

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Task Based Learning

Uploaded by

Copyright:

Available Formats

The Task-based Approach in Language Teaching

Addressfvrcorrespondence: Departamento de Filologa Inglesa, Facultad de Letras, Campus de Lu ,I,lerced, 30071

O Servicio de Publicaciones. Universidad de Murcia. All rights reserved.

IJES, vol. 4 (l), 2004, pp. 39-71

IJES: vol. 4 ( l ) , 2004, pp. 39-71

The Task-based Approach in Language Teaching

BACKGROUND AND RATIONALE BEHIND THE TBA IN LANGUAGE

IJES, vol. 4 (l), 2004, pp. 39-71

IJES, vol. 4 ( 1 ) . 2004, pp. 39-71

The Tusk-busedApprouch in Lunguuge Teuching

And he adds later:

IJES, vol. 4 (l), 2004, pp. 39-71

LANGUAGE LEARNING AS A PROCESS

Prabhu's claim is revolutionary regarding synthetic or notional-functional approaches:

/JE.S, vol. 4 (1). 2004, pp. 39-71

The Task-based Approach in Lanpage Teaching

IJES, vol. 4 ( l ) , 2004, pp. 39-71

O Servicio de Publicaciones. Universidad de Murcia. All rights reserved

IJES, vol. 4 ( l ) , 2004, pp. 39-71

The Tusk-bused Approuch in Lunguuge Teuching

O Servicio de Publicaciones. Universidad de Murcia. All rights reserved.

IJES, vol. 4 (l), 2004, pp. 39-7 1

IJES, vol. 4 (l), 2004. pp. 39-71

The Tusk-bused Approuch in L~~nguuge

- there is a problem to solve

IJES, vol. 4 ( l), 2004, pp. 39-71

IJES, vol. 4 ( l ) , 2004, pp. 39-7 1

The Task-based Approach in Lunguuge Teaching

REAL WORLD TASKS AND CLASSROOM TASKS

IJES, vol. 4 ( 1 ) . 2004, pp. 39-71

IJES, vol. 4 (l), 2004, pp. 39-71

The Tusk-hused Approuch in Lunguuge Teuching

IJES, vol. 4 ( l ) , 2004, pp. 39-71

IJES, vol. 4 (l), 2004, pp. 39-71

The Tusk-bused Approuch in Lunpuge Teuching

COMPLEXITY AND SEQUENCING OF TASKS

IJES, vol. 4 ( l ) , 2004, pp. 39-71

IJES, vol. 4 (l), 2004, pp. 39-71

The Task-based Approuch in Language Teaching

IJES, vol. 4 (l), 2004, pp. 39-71

I J B , vol. 4 (l), 2004, pp. 39-71

The Task-based Approach in Lunguage Teaching

IJES, vol. 4 ( l ) , 2004, pp. 39-71

IJES, vol. 4 ( 1 ) . 2004, pp. 39-71

The Tusk-based Approuch in Lunguage Teuching

IJES, vol. 4 (1). 2004, pp. 39-71

IJES, vol. 4 (l), 2004, pp. 39-71

The Tusk-hused Approuch in Languuge Teuching

IJES, vol. 4 ( l ) , 2004, pp. 39-71

O Servicio de Publicaciones. Universidad de Murcia. All rights reserved.

IJES, vol. 4 ( l ) , 2004, pp. 39-71

The Task-based Approach in Lunguuge Teaching

BIBLIOGRAPHY AND REPERENCES

IJES, vol. 4 (l), 2004, pp. 39-71

Lunguage Teaching, 20 (3):

IJES, vol. 4 (I), 2004, pp. 39-71

The Task-bused Approach in Languuge Teuching

IJES, vol. 4 ( 1 ), 2004, pp. 39-71

IJES, vol. 4 ( l ) , 2004, pp. 39-71