International Journal of TESOL Studies (2024)

Vol. 6 (4) 1-13 [Link]

Key Constructs in Language Teaching

Task-based and Task-supported Language Teaching

Rod Ellis
Curtin University, Australia

Received: 3 September, 2024/Accepted: 10 September, 2024/Published: 15 September, 2024

Abstract
My purpose in this article is to give some clarity to what task-based language teaching (TBLT)
is and how it differs from task-supported language teaching (TSLT). Central to both approaches
is the notion of ‘task’, so I will begin with a definition of this and point to some problems in
distinguishing a ‘task’ from an ‘exercise’. I will go on to discuss how tasks figure in TBLT and TSLT.
I will then address arguments that have been used to claim that TSLT is better suited to some
teaching contexts, such as those found in Asia, where teachers and students are used to traditional
examinations. I conclude by suggesting that while TSLT should not be seen as an alternative to
TBLT, it can complement it in several ways.

Keywords
Task-based language teaching (TBLT), task-supported language teaching (TSLT)

1 What is a task?

The task-based literature is replete with definitions of a task. Here is a selection:

… by ‘task’ is meant the hundred and one things people do in everyday life, at play, and in
between. Tasks are the things people will tell you they do if ask them and they are not applied
linguists (Long, 1985, p. 89)
…. tasks are always activities where the target language is used by the learner for a
communicative purpose (Goal) in order to achieve an outcome (Willis, 1996, p. 23).
A task is a holistic activity which engages language use in order to achieve some non-linguistic
outcome while meeting a linguistic challenge, with the overall aim of promoting language
learning, through process or product or both (Samuda and Bygate, 2008, p. 69).
….. an activity in which: meaning is primary; there is some sort of relationship to the real
world; task completion has some priority; and the assessment of task performance is in terms of
task outcome (Skehan, 1996).
Readers may feel confused when confronted with such mixed definitions. Long’s definition refers to
‘target tasks’, which he sees serving as a basis for the design of pedagogic tasks. However, it is not

Email: [Link]@[Link]
2 International Journal of TESOL Studies 6 (4)

difficult to think of activities – such as spot the difference – which would not be tasks according to
Long’s definition as they do not correspond to any real-life target task but have been widely viewed as
tasks in task-based materials and research. Willis’ definition includes some key attributes of a task (i.e.
‘communicative purpose’ and ‘outcome’ but seems too general as it would seem to include situational
grammar practice exercises like Workplan 1 in Figure 1 below. Samuda and Bygate’s definition identifies
three key aspects of a task, namely that it is ‘holistic’, it has a ‘non-linguistic outcome’, and it involves
a ‘linguistic challenge’ but many tasks – for example, narrating a story based on pictures - have a
linguistic, not a non-linguistic, outcome. Also, its abstract nature will make it hard for a teacher to decide
whether any teaching activity is or is not a task. Skehan’s definition is widely cited but has also been
criticized. Widdowson (2003) argued, with some justification, that Skehan’s use of the term ‘meaning’
is indeterminate as it does not distinguish semantic and pragmatic meaning and what Skehan means by
‘goal’ and ‘real-world relationship’ are not specified. Exercises like Workplan 1 in Figure has a ‘goal’ (to
practise the use of some and any and arguably has some kind of real-world relationship.
Table 1 shows my own definition (Ellis, 2003; Ellis & Shintani, 2014). I framed my definition as a set
of criteria that could be applied to specific workplans to determine whether they can be considered tasks.
To address a common misconception about TBLT, namely that TBLT is all about teaching speaking
skills, I pointed out that my definition of a task is applicable to receptive as well as productive tasks. I
also drew on Breen’s (1989) distinction between task-as workplan and task-as-process to argue that a
task can only be defined in terms of a workplan (i.e. as the language teaching materials given to students)
and not in terms of the process of performing a task because the process is always, to some degree at
least, unpredictable. It is for this reason that I do not think ‘linguistic challenge’ should be included in
a definition of task – as Samuda and Bygate proposed. The extent to which a task challenges learners
linguistically is not dependent on the task workplan - but on the linguistic proficiency of the learners
performing it and is therefore an aspect of the process the task results in. ‘Task’ is a construct that is
applicable to the design of a syllabus, a lesson plan or a test. But it has no life of its own; it only comes to
life when it is performed. TBLT lessons are built around tasks, but they also involve a methodology for
implementing them. Distinguishing the design of a task from its implementations is fundamental to an
understanding of TBLT.

Table 1.
Criteria for Determining Whether a Workplan is a Task
Criteria Description
The primary focus is on The workplan is intended to ensure that learners are primarily
meaning concerned with comprehending or/ and producing messages for a
communicative purpose (i.e. there is primary focus on meaning-
making).
There is some kind of gap The workplan is designed in such a way as to incorporate a gap
which creates a need to convey information, to reason or to express
an opinion.
Learners rely mainly on Learners need to draw on their existing linguistic resources
their own linguistic and (potentially both L1 and L2) and their non-linguistic resources (e.g.
non-linguistic resources gestures; facial expressions) for comprehension and production.
There is a clearly defined The workplan specifies the communicative outcome of the task.
communicative outcome Thus task accomplishment is to be assessed not in terms of
whether learners use language correctly but in terms of whether the
communicative outcome is achieved.
Rod Ellis 3

I illustrated how my definition could serve to distinguish tasks from exercises by comparing the two
workplans shown in Figure 1 (Ellis, 2019). It could be argued that both workplans involve a focus on
meaning as the student playing Abdullah can only provide the correct response by checking whether in
fact the item asked for is available in the store. This suggests that criterion 1 (‘The primary focus is on
meaning”) will not always distinguish a task from an exercise. The other criteria, however, do. There
is no gap in Workplan 1 but there is in Workplan 2. In Workplan 1 students are given a model of how
to form their dialogue but not in Workplan 2 where they must rely on their own linguistic resources.
Finally, in Workplan 1 there is no communicative outcome as the activity simply involves substituting
words in the model provided. Workplan 1 is a situational exercise designed to practise ‘some’ and ‘any’;
its success depends on whether students use these grammatical items correctly. Workplan 2 is a task; its
success depends on whether the students achieve their separate communicative outcomes. Of course,
students can always subvert the intended purpose of an activity in the process of performing it. But as
workplans, the activities in Figure 1 can be clearly distinguished in terms of my criteria.

Figure 1.
Two Workplans
Workplan 1

Look at Mary’s shopping list. Then look at the list of items in Abdullah’s store. Work
with a partner. One person is Mary and the other person is Mr. Abdullah. Make
conversations like this.

Mary: Good morning. Do you have any flour?

Abdullah: Yes, I have some

Mary’s Shopping List

1. oranges 4. powdered milk
2. eggs 5. biscuits
3. flour 6. jam

Abdullah’s Store
1. bread 7. rice
2. salt 8. sugar
3. apples 9. curry powder
4. tins of fish 10. biscuits
5. coca cola 11. powdered milk
6. flour 12. dried beans
4 International Journal of TESOL Studies 6 (4)

Workplan 2

Work with your partner. One of you be A and the other B. Do not look at your partner’s
instructions.

Student A:
You are going shopping at Student B’s store. Here is your shopping list. Put ticks next
to the items on your list you can buy.

1. oranges 4. powdered milk
2. eggs 5. biscuits
3. flour 6. jam

Student B:
You own a store. Here is a list of items for sale in your store. Find the items that
Student A asks for that you do not stock.

1. bread 7. rice
2. salt 8. sugar
3. apples 9. curry powder
4. tins of fish 10. biscuits
5. coca cola 11. powdered milk
6. flour 12. dried beans
You are going shopping at Student B’s store. Here is your shopping list. Put

However, the distinction between a task and an exercise is not always so clear. If we ask a student
to look at a set of pictures and narrate the story that they depict, is this activity a task? This kind of
workplan is very common in both pedagogy and research. It satisfies the first three of the criteria in my
definition but perhaps not the last as there is no communicative reason for narrating the story and no
outcome depending on its telling. This suggests that the difference between an exercise and a task is not
dichotomous but continuous, with some workplans being entirely exercises (for example, a fill-in-the
blank exercises), some being clearly tasks such as Workplan 2 in Figure 1, and others being more or less
task-like. Workplans where the task outcome is clearly non-linguistic (e.g. completing a map; drawing
picture; making a model) are perhaps most fully tasks. Samuda and Bygate are perhaps right in including
‘non-linguistic outcome’ in their definition and perhaps I should amend mine. But this would exclude
many workplans such as descriptive or opinion-gap tasks where the outcome is entirely or mainly
linguistic but are central to TBLT. Perhaps, we should not be purists, and acknowledge that ‘task’ remains
a somewhat fuzzy construct and acknowledge the legitimacy for inclusion in TBLT of any activity that is
primarily task-like.

2 Task-based Language Teaching

TBLT is an approach that gives central place to tasks in both the design of a language course and a
methodology for implementing tasks that gives primacy to attention to form while communicating in
Rod Ellis 5

order to complete a task. To understand TBLT, therefore, it is necessary to have a clear grasp of what a
task is and knowledge of options for implementing it.
TBLT is theoretically grounded in second language acquisition research that documents how
acquisition is a gradual process of building L2 systems (once called ‘interlanguages’ but more recently
‘dynamic complex systems’) incidentally (rather than intentionally) and in educational theories that
emphasize the importance of experiential learning and learners as active agents. Incidental learning
leads to implicit knowledge – the kind of knowledge that is found in children acquiring their mother
tongue and that can still be acquired by older learners learning a second language (Long, 2015). Implicit
knowledge is available for use without conscious effort and thus is essential for normal communication.
From this perspective, a structural approach to teaching a language where the language is broken down
into discrete units that are taught and learned intentionally and incrementally, will not be successful,
a position borne out by the experience of teachers who have followed this approach only to find
that after years of instruction their learners are unable to utilize what they have learned in everyday
communication. Such an approach is more likely to lead to explicit than implicit knowledge. The
theoretical base of TBLT is undoubtedly a strong one.
TBLT, however, is not monolithic (Ellis, 2009; Ellis, 2021). It is an approach, not a narrowly
prescribed method. Table 2 summarizes the key characteristics of TBLT in the writings of advocates of
this approach. All four versions of TBLT emphasize the centrality of natural language use (i.e. the use
of language where the focus is on communicating, not on learning). Learner-centredness is also seen as
a central aspect in tasks that involve students in collaborative meaning making although I have argued
that teacher-centred classes can also have a role in TBLT (i.e. there will be times where the teacher will
need to perform a task with the whole class). For me the opportunity for student-student interaction
among learners is an important aspect of TBLT but not an essential one. The fact that it is sometimes
seen as a defining feature of TBLT is the product of an abundance of research that has focused narrowly
on speaking tasks (e.g. Mackey, 2013). I have always argued that input-based tasks involving listening
and reading have a major role to play in a TBLT programme. Long (2015) also acknowledged that input-
based tasks are needed with beginners. Long differs from the other proponents of TBLT in insisting that
tasks must be based on a needs-analysis to identify the target tasks that students will need to perform in
their lives. Such an approach is clearly desirable in contexts where students are living in a country where
the target language is an official language, but it is generally not relevant in foreign language contexts
involving state schools where the target language has no obvious communicative function outside the
classrooms (e.g. in most Asian countries). In general, proponents of TBLT favour unfocused tasks
(i.e. tasks not designed to elicit specific linguistic features) but in fact there are many research studies,
including my own (e.g. Ellis, Basturkmen, & Loewen, 2002; Shintani & Ellis, 2010; Zhao & Ellis, 2022)
that have used focused tasks (i.e. tasks that satisfy the criteria for a task but provide a context for the use
of pre-determined linguistic features, usually grammatical). Later in this article I will explain how I think
focused tasks can fit into a task-based course.
Other differences in TBLT occur at the level of task implementation. One of the main areas of
difference among the proponents of TBLT lies in how a focus on form is handled. A TBLT lesson can
consists of three stages – a pre-task stage, a main task stage and a post-task. Both Skehan and I consider
that attention to form can occur in all three. In the pre-task stage this can take the form of guided
planning, but not direct explicit teaching. In the main task-stage it will typically occur through feedback
on learners’ efforts to communicate and in the post-task stage through direct, explicit teaching addressing
linguistic problems seen to arise during the performance of the task. Long considers what he termed ‘focus
on form’ should occur in the main task while Willis thought it belonged only in the post-task stage. There
are arguments to support all these different positions, but I will not consider these here. Finally, there is a
question of whether traditional, structural-based teaching has a place in TBLT. Only I have argued that it
does have a place in what a call a modular curriculum. I will give my reasons later after I have dealt with
task-supported language teaching.
6 International Journal of TESOL Studies 6 (4)

Table 2.
Similarities and Differences in TBLT
Characteristic Long (1985; Willis (1996) Skehan (1998) Ellis (2003)
2015)
Natural language use Yes Yes Yes Yes
Learner-centredness Yes Yes Yes Not necessarily
Task topics Based on target Based on student - Based on student
tasks interests interests
Task types Unfocused Unfocused Unfocused Unfocused and
focused
Task medium Input-based and Output-based Output-based Input-based and
output-based output-based
Focus on form Not in pre-task Only in post-task In all stages of a In all stages of a
stage; primarily stage lesson lesson
in main task stage
Rejection of traditional Yes Yes Yes No
approaches

The essence of an approach lies in the principles that define it and that underlie both the design and
implementation of language programmes based on it. The principles that I see as common to all TBLT
approaches are as follows:
• The primacy of ‘task’:
Task – as I have tried to define it – is central. It constitutes the unit that informs the design of a
syllabus. It also constitutes the unit around which individual lessons are built. Lesson planning begins
with a task.
• No a priori explicit language teaching
While both Skehan and I see room for attention to form in the pre-task stage of a lesson – for
example, in the strategic planning carried out by individual learners – none of the four versions
shown in Table 2 allow for any explicit teaching of language. This is the case for focused as well as
unfocused tasks. Even if a task is set up to elicit the use of a specific grammatical structure, students
do not receive any pre-teaching of that structure and are not told that they should try to use it. A
focused task may involve the use of the target structure, but it may not as how the task is performed
depends on what the learners make of the task; they may or may not attempt use of the target
structure. A focused task must still satisfy the criteria for a task.
• Focus on form
By form I mean any aspect of language – pronunciation, vocabulary, grammar, discourse. In
unfocused tasks, focus on form will be directed at whatever problems arise as students perform the
task. In focused tasks, focus on form will primarily address the structure that is the target of the task.
One of the most common misconceptions about TBLT is that there is no room for language work– for
example, Sheen (2003) claimed that TBLT has no room for grammar. There are, however, differences
in where and how proponents of TBLT think grammar should fit in. For Long (1991) it needs to occur
within the performance of a task, for example by teachers (or learners) reacting to linguistic problems
experienced while a student is struggling to communicate. For Skehan, the ideal site for focus on
form is before the task is performed. For Willis, it needs to be reserved to the post-task stage in order
to allow for students to focus on communicating in the main task stage. For me, focus on form can
occur usefully at any stage of a lesson.
Rod Ellis 7

• Explicit post-task work on language problems

A task-based lesson does not have to include a post-task stage. It can conclude with the performance
of the task. But all proponents of TBLT recognize the utility of a post-task stage. This need not
involve explicit work on language problems, but it does afford an opportunity for students to engage
not just in ‘noticing’ but also ‘understanding’ their language problems. Explicit work can take many
forms. For example, Lau, Aubrey and Ellis (2024) asked students to complete error logs addressing
the errors in a writing task that a teacher had corrected. Even direct, explicit instruction has a place in
the post-task stage. There is a need for studies that have investigated the effect that explicit post-task
work in TBLT has on learning, including the type of knowledge it results in (i.e. implicit or explicit).
Li. Zhu and Ellis (2016) is one of the few studies to have investigated this to date.

3 Task-supported Language Teaching

In a thoughtful discussion of the differences between TBLT and TSLT, Samuda and Bygate (2008)
commented “the TBLT/TSLT distinction is not a property of ‘task’ itself, but a function of the way tasks
are used” (p. 216). They go on to offer an account of the different ways in which focus on form can be
used “at the beginning of a sequence, an initiating role, or a more reactive role if used in the middle and
an integrative role if used at the end” (p. 216). In effect Samuda and Bygate blur the distinction between
TBLT and TSLT, highlighting instead the importance of ‘task’ as a construct that can lead to different
ways of implementing it.
I have taken a different tack, viewing TBLT and TSLT as distinct in two related and fundamental ways:
1. In TBLT ‘task’ serves as the unit for designing a syllabus – without reference to any language
to be taught and learned. In TSLT, a syllabus is constructed around a set of linguistic units to be
taught and learned and ‘task’ is simply a pedagogic device for achieving this.
2. In a TBLT lesson there is no a priori explicit teaching of any language; a TSLT lesson starts with
the explicit presentation of the linguistic target(s) of the lesson.
In other words, the differences between TBLT and TSLT reside in both how the content of a language
programme is established and sequenced and in how lesson plans involving tasks are constructed. These
differences are also reflected in different theoretical positions. TBLT is based on a cognitive view of
language acquisition as holistic, gradual and learner-determined; TSLT is based on a skill-learning view
of language development, where the learning of specific pieces of language proceeds from a declarative
to procedural and automatic stages (Dekeyser, 1998). TBLT aims to provide opportunities for incidental
learning through ‘language use’ (i.e. where communicative intent is primary); TSLT aims to provide
opportunities for intentional learning and automatization through ‘practice’ where attention is focused on
the mastery of language. It is these theoretical differences, reflected in programme and lesson design, that
distinguish TBLT and TSLT.
Let me make one-point clear, however, I am not saying that the difference lies in the fact that TSLT
includes explicit language teaching while TBLT does not. I have already pointed out that focus on form
is an essential feature of TBLT and while in many instances this will involve implicit techniques such
as recasting students’ erroneous utterances during the performance of a task, it can also involve more
explicit techniques both during the main task and the post-task stages of a lesson. Samuda (2001), for
example, described a task-based lesson where she found it necessary to stop students’ performance
of a task to provide a mini-grammar lesson addressing what was the target feature of her focused task
(epistemic modal verbs) before asking them to continue. Ellis, Basturkmen and Loewen (2002) described
how teachers drew explicit attention to form both pre-emptively when nudging learners to attend to form
and reactively by responding to attested language problems during the main task stage of the lessons they
observed. What makes TSLT distinct, is the positioning of the explicit language work at the beginning of
8 International Journal of TESOL Studies 6 (4)

a lesson, prior to the performance of the task. When teachers do this, they are signalling to students that
the real purpose of the lesson is to learn the target structure. This is likely to fundamentally affect how
they then perform the task (i.e. by making the task an opportunity for ‘practice’ rather than for ‘language
use’). It should be noted, however, that students may elect to pay explicit attention to form in the pre-task
planning stage but they determine what language to focus on, not the teacher.
TSLT is often equated with PPP (present-practice-produce) where the practice stage involves discrete-
point practice exercises and the production stage a task. However, TSLT and PPP are not isomorphic.
TSLT need not involve the intermediate practice stage; that is a lesson can progress directly from the
explicit presentation of the target structure to the performance of the focused task designed to elicit its
use. When this happens, TSLT moves closer to TBLT but is still fundamentally different because the aim
of the lesson remains the mastery of the target structure. An interesting issue is whether the inclusion
of a priori explicit teaching has any effect on how students perform a task. Do they pay attention to the
target structure during the task, do they use it correctly, and does allocating attention to it impact on other
aspects (e.g. fluency and complexity) of how a task is performed? There has been relatively little research
comparing learners’ performance of a task with and without a prior explicit instruction but see Ellis, Li
and Zhu (2018), which found that it did impact on how the same task was performed, which is exactly
what a cognitive theory of L2 acquisition would predict.
TBLT and TSLT derive from competing theories of language acquisition and differ in both how
tasks are used in the design of course and in how they are implemented in lessons. Is one more effective
than the other? This would seem a central question to address but it is difficult to answer both because
there have been relatively few evaluation studies of the two approaches and because of the difficulty of
designing evaluation studies that take account of all the variables involved in the operationalization and
implementation of both approaches. Ellis et al (2020) reviewed several comparative evaluation studies,
finding that overall TBLT was more effective than traditional approaches. However, they also warned
against hasty conclusions and pointed to the need to examine the process features of how the approaches
were implemented (i.e. what happens in the classroom) and not just measures of learning outcomes.

4 The Importance of Instructional Context

Since the 1900 there has been considerable resistance to the idea that there is a single method (or
approach) that is appropriate for all instructional contexts. Bax (2003), for example, proposed what he
called the Context Approach as an alternative to the adoption of communicative language teaching. He
described the Context Approach like this:
The first priority is the learning context. and the first step is to identify key aspects of the
teaching context before deciding what and how to teach in any given class.
He goes on to identify what aspects of the context teachers should consider – the needs, wants, styles
and strategies of individual learners, the coursebook, local conditions, the classroom culture, the school
culture, and the national culture. He does not, however, specify how teachers should use their analysis of
context to make decisions about what and how to teach. In a similar vein, Kumaravidivelu (2001) argued
for what he termed a “post-method pedagogy’, where teachers are not constrained by the prescriptions (and
proscriptions) of any method, but instead have the freedom and flexibility to make their own pedagogic
decisions.
These perspectives underlie commentaries about the perceived problems of introducing TBLT,
leading to the suggestions that it is ill-suited to some contexts. Ellis et al (2020) summarised the problems
that teachers and students face along with structural problems relating to teaching context. Teachers have
been found to have difficulty in understanding what a task is (Hu, 2013; Erlam, 2016) and, reflecting their
own learning experiences, often manifest an overriding concern for grammar (McDonough, 2015). As a
Rod Ellis 9

result, they tend to fall back on explicit language instruction. Students who are used to teacher-centred
approaches may fail to see the point in performing tasks. They experience difficulties in performing
tasks when their linguistic proficiency is limited and are likely to resort to their L1. Structural issues
concerning the teaching contexts include the large size of classes, the lack of suitable teaching materials
and the need to prepare students for traditional, discrete-point tests. These attested problems have led
some commentators to propose the abandonment of TBLT in favour a structure-oriented approach such
as TSLT (Littlewood, 2014).
Debates about language pedagogy are often characterized by a tension between universalist positions
based on theories of language acquisition and general educational principles and localist positions that
emphasize the importance of context and individual differences in teachers and learners. The former
lead to the advocacy of methods or approaches – such as TBLT – while the latter reject straitjackets and
emphasize the need for flexibility. It seems to me, however, that universalist and localist positions need
not be seen as antithetical but can be complementary. It is surely sensible to be guided by an approach
that is supported by what we know about learning and sound education. But in implementing this
approach there must be and indeed always will be room for teachers to make both offline and online
decisions about how to implement any approach. Thus we do not need to reject the value of a well-
theorized approach such as TBLT, which can guide the design of a syllabus and the choice of language
teaching materials. But when implementing TBLT it will be necessary to take account of local factors.
This need not mean the substitution of TBLT by some other approach, such as TSLT, but rather utilizing
methodological procedures that can help to address the practical problems teachers experience. This is
the position both I and Long have adopted. In Ellis (2018) I spell out ways in which teacher, learner and
structural problems can be dealt with. Long (2015) provided a list of general methodological procedures,
arguing that teachers will need to draw on these flexibly in the process of implementing TBLT. Thus, I do
not accept the strong positions that Bax and Kumaridevelou have adopted. Context is important but so is
a theory-grounded approach that can shape the design of courses and guide (but not constrain) teachers
as they implement TBLT. The question remains, however, as to whether there should be struct adherence
to TBLT – what Long (2015) calls ‘pure TBLT’ – or whether there is also a place in a curriculum for a
language-oriented approach such as TSLT. I address this question in the next section.

5 Combining TBLT and TSLT – the Case for Modularity

I do not see that replacing TBLT with TSLT is a viable option; these two approaches are based on
theoretical positions about the nature of language learning that are incompatible. However, I do think that
a case can be made for combining TBLT and TSLT in what I have termed a ‘modular curriculum’ (Ellis,
2019).
Figure 2 shows how TBLT and TSLT can be combined in the design of a curriculum. The vertical axis
represents the students’ language proficiency; the horizontal axis the approach I suggest is appropriate at
different stages of a complete course from beginner to advanced level. Thus, for beginner-level learners,
the curriculum consists of only TBLT to foster confidence in using the L1 in communication and to equip
learners with the implicit knowledge needed for this. The task-based component of the curriculum is
also needed at later intermediate and advanced stages but, to address entrenched interlanguage forms, I
argue it needs to be supplemented with the structured component (e.g., TSLT) where explicit attention is
given to those specific language items that have not been mastered incidentally. In a modular curriculum,
the task-based and the structured component sit side by side with no attempt to integrate them. In
this respect, it differs from the kind of curriculum found in many popular textbooks where there is an
attempt to integrate an explicit focus on language with communicative activities as in PPP or TSLT. The
rationale for rejecting the attempt to integrate the two approaches is that they address different ways
of learning leading to different types of linguistic knowledge (implicit in the case of the task-based
component; explicit in the case of the structured component). The task-based component involves a task-
10 International Journal of TESOL Studies 6 (4)

based syllabus. The structured component consists of a checklist of linguistic problems that learners are
likely to manifest at intermediate and advanced levels of proficiency and a bank of teaching materials,
including focused tasks, for address each problem. It serves a resource that teachers can refer to, but
the selection of linguistic targets is guided by teachers noting the actual problems learners experience
when performing tasks. For example, if students continue to make errors in the third person singular of
the present simple tense (e.g. *Mary has left Japan and now live in London) when performing tasks, the
teacher could devise a separate lesson focusing explicitly on this problem and including a focused task to
provide opportunity for communicative practice. The theoretical basis for resorting to explicit teaching
is that, without it, certain types of errors will remain entrenched in learners’ L2 systems and that this
problem may persist but can be addressed by explicit teaching. The explicit teaching that takes place is
reactive, not proactive. It builds on learners’ own attempts to use specific forms. TSLT (or PPP), then,
does not replace TBLT but supplements it, serving a remedial function.
The modular curriculum model is advantageous in another way – it provides a basis for taking
context into account. For example, in a context where the teacher is constrained by the need to prepare
students for a traditional test, greater emphasis can be given to the structured component. Perhaps too
students can be provided with supplementary materials addressing attested language problems which
they work though individually outside of class time, reserving time in class for TBLT. Or if context
requires adherence to a more traditional teaching approach, teachers could elect to compartmentalize
by including activities that are task-based and activities involving explicit instruction within a single
lesson or in different lessons. In accordance with the principle of modularity, however, teacher will need
to keep the different kinds of activities separate, ideally allocating more time to TBLT. Another way of
addressing contextual constraints, might be to vary the extent to which the tasks chosen are focused or
unfocused. Unfocused tasks are preferable for low proficiency learners, but where the context involves
a language-based syllabus, more focused tasks could be employed in the task-based component. Being
responsive to context should not involve untheorized eclecticism but a principled way of balancing
universalist principles that shape TBLT and local requirements that may require a more structured
approach such as TSLT.

Figure 2.
A Modular Curriculum Model (Ellis, 2019)

At the level of task-implementation TSLT involves an initial stage of explicit instruction whereas TBLT
does not, which is the essential difference in the two approaches. In the focus on form that occurs during
Rod Ellis 11

task-performance, both approaches can draw on similar options. In both TSLT and TBLT, teachers may
choose to avoid any focus on form and give precedence to communicative fluency – as Willis (1986)
recommends. Both can use the same range of corrective feedback strategies. So there is room here for
teachers to be context sensitive in deciding which options to employ. Opportunities exist for teachers to
be responsive to their specific context - for example, in deciding whether to correct errors using prompts
or recasts or by means of implicit or explicit strategies (Lyster, Saito, & Sato, 2013). In one respect, the
essential difference between TBLT and TSLT involves a difference in how focus on form during task
performance is implemented; in TSLT, it will be directed at the linguistic target of the lesson, whereas in
TBLT it will be directed at whatever linguistic problems happen to arise.

6 Conclusion

In this article I have tried to give clarity to two key concepts – TBLT and TSLT – and, in so doing,
outline my own views about language pedagogy. I conclude here with a summary of my main points.
1. In discussion of any language teaching approach a clear distinction needs to be made between
two aspects of a curriculum – its design and its implementation. This distinction is crucial for
understanding the difference between TBLT and TSLT.
2. Fundamental to both TBLT and TSLT is the concept of ‘task’. It is important, therefore, to be
clear about what a ‘task’ is. I have proposed that a task is a workplan (it is a design unit, not an
implementation unit) and offered my own definition in terms of four criteria. I have acknowledged,
however, that ‘task’ remains a fuzzy construct, and that it may be necessary to view the difference
between a task and an exercise as non-dichotomous. There is room for more thought and more
research here.
3. I emphasized that TBLT is not monolithic and that there are different versions. Nevertheless, all
versions adhere to a set of common principles, in particular: there is no a priori explicit teaching
of language, that performance of a task involves learners using their existing linguistic resources
and there is focus on form.
4. In both TBLT and TSLT tasks are fundamental but there are differences in how a curriculum is
designed and how tasks are implemented. TSLT involves a structural syllabus, TBLT a task-based
syllabus. TSLT involves only focused tasks; TBLT mainly unfocused tasks but in my version (Ellis,
2003) it can also involve focused tasks. In TSLT focus on form is directed at the target feature
during task performance, in TBLT at whatever linguistic problems occur.
5. The need to adapt teaching to take account of context, does not override the need to ensure that
it is based on approach that is theoretically sound (i.e. supported by what we know about L2
acquisition and sound educational principles).
6. TSLT cannot substitute for TBLT because they constitute fundamentally different approaches.
7. In a modular curriculum, TSLT can supplement a curriculum where TBLT is primary. In such a
curriculum, no attempt is made to integrate the two approaches. They exist as separate strands in
the curriculum.
8. A modular approach is sensitive to context. The emphasis given to each approach will depend on
the students’ proficiency but also on other considerations such as the external constraints teachers
face, the expectancies of their students and the need to prepare students for examinations. Within
each approach, too, there are methodological options that allow teachers flexibility.
In this article I have presented my own definitions of constructs that are complex and my own ideas
about how TBLT and TSLT can function together in a curriculum. I offer them to encourage debate about
what I see as central issues in language pedagogy today.
12 International Journal of TESOL Studies 6 (4)

References

Bax, S. (2003). The end of CLT: A context approach to language teaching. ELT Journal 57, 278–287.
Breen, M. (1989). The evaluation cycle for language learning tasks. In R. K. Johnson (Eds.): The Second
Language Curriculum. Cambridge University Press.
DeKeyser, R. (1998). Beyond focus on form: Cognitive perspectives on learning and practicing second
language grammar. In C. Doughty and J. Williams (Eds.). Focus on form in classroom second
language acquisition (pp. 42-63). Cambridge University Press.
Ellis, R., Basturkmen, H. & Loewen, S. (2002). Doing focus-on-form. System, 30, 419–432.
Ellis, R. (2003). Task-based language learning and teaching. Oxford University Press.
Ellis, R. (2009a). Task-based language teaching: sorting out the misunderstandings. International Journal
of Applied Linguistics, 16, 221–246.
Ellis, R. (2018a). Reflections on task-based language teaching. Multilingual Matters.
Ellis, R. (2019). Towards a modular language curriculum for using tasks. Language Teaching Research,
23(4), 454-475.
Ellis, R. (2021). Options in a task-based language-teaching curriculum: An educational perspective. Task,
1(1), 11-46.
Ellis, R., Basturkmen, H. & Loewen, S. (2002). Doing focus-on-form. System, 30, 419–432.
Ellis, R., Li, S., & Zhu, Y. (2018). The effects of pre-task explicit instruction on the performance of a
focused task. System, 80, 38–47.
Ellis, R., Loewen, S. & R. Erlam. (2006). Implicit and explicit corrective feedback and the acquisition of
L2 grammar. Studies in Second Language Acquisition 28, 339–368.
Ellis, R. & Shintani, N. (2014). Exploring language pedagogy through second language acquisition
research. Routledge.
Ellis, R., Skehan, P., Li, S., Shintani, N., & Lambert, C. (2020). Task-based language teaching: Theory
and practice. Cambridge University Press.
Erlam, R. (2016). I’m still not sure what a task is: Teachers designing language tasks. Language Teaching
Research 20, 275–299.
Hu, R. (2013). Task-based language teaching: responses from Chinese teachers of English. The
Electronic Journal for English as a Second Language 16 (4)
Kumaravadivelu, B. (2001). Toward a postmethod pedagogy. TESOL Quarterly, 35(4), 537-560.
Lau, W. M. E., Aubrey, S., & Ellis, R. (2024). Error logs in the second language classroom: Exploring
the relationship between learner engagement with written corrective feedback and improvements in
writing accuracy. System, 123, 103329.
Li, S., Zhu, Y., & Ellis, R. (2016). The effects of the timing of corrective feedback on the acquisition of a
new linguistic structure. The Modern Language Journal, 100(1), 276-295.
Littlewood, W. (2014). Communication-oriented teaching: where are we now? Where do we go from
here? Language Teaching 47, 249–362.
Long, M. (1985). A role for instruction in second language acquisition: Task-based language teaching
in K. Hyltenstam & M. Pienemann (Eds.): Modelling and assessing second language acquisition.
Multilingual Matters.
Long, M. (1991b). Focus on form: A design feature in language teaching methodology. In K. de Bot,
R. Ginsberg, & C. Kramsch (Eds.). Foreign language research in cross-cultural perspective (pp.
39–52). John Benjamins.
Long, M. (2015). Second language acquisition and task-based language teaching. Wiley-Blackwell.
Rod Ellis 13

Lyster, R., Saito, K., & Sato, M. (2013). Oral corrective feedback in second language classrooms.
Language teaching, 46(1), 1-40
Mackey, A. (2013). Conversational interaction in second language acquisition. Oxford University Pres
McDonough, K. (2015). Perceived benefits and challenges with the use of collaborative tasks in EFL
contexts. In M. Bygate (Ed.). Domains and direction in the development of TBLT (pp. 225–245).
John Benjamins.
Samuda, V. (2001). Guiding relationships between form and meaning during task performance: The role
of the teacher. In M. Bygate, P. Skehan, & M. Swain (Eds.). Researching pedagogic tasks: Second
language learning, teaching and testing (pp. 119–140). Longman.
Samuda, V. & Bygate, M. (2008). Tasks in second language learning. Palgrave Macmillan.
Sheen, R. (2003). Focus on form–a myth in the making? ELT journal 57, 225–233.
Shintani, N., & Ellis, R. (2010). The incidental acquisition of English plural–s by Japanese children in
comprehension-based and production-based lessons: A process-product study. Studies in second
language acquisition, 32(4), 607-63.
Skehan, P. (1996). A framework for the implementation of task-based learning. Applied Linguistics, 17,
38–62.
Skehan, P. (1998). A cognitive approach to language learning. Oxford University Press.
Widdowson, H. (2003). Defining issues in English language teaching. Oxford University Press.
Willis, J. (1996). A framework for task-based language teaching. Longman.
Zhao, Y., & Ellis, R. (2022). The relative effects of implicit and explicit corrective feedback on the
acquisition of 3rd person-s by Chinese university students: A classroom-based study. Language
Teaching Research, 26(3), 361-38.

Rod Ellis is a Distinguished Research Professor at Curtin University (Perth, Australia) and also, an
Emeritus Distinguished Professor at the University of Auckland (New Zealand). He has also worked in
universities in UK, Japan, and the USA and is a fellow of the Royal Society of New Zealand. He has
published widely in the related fields of second language acquisition research and task-based language
teaching. He has won the British Association of Applied Linguists prize (1985), the Mildenberger prize
(1987), The Duke of Edinburgh prize (1994), and the International Association of Task-based Teaching
prize (2021). His current H-index on Google Scholar is 109.