Task-based language teaching (TBLT) is a communicative approach to language instruction, using the successful completion of communicative “tasks” as its primary organizing principle. In short, instruction is organized in such a way that students will improve their language ability by focusing on getting something done while using the language, rather than on explicitly practicing language forms, as in more traditional methods of instruction.
Challenges of CLT
TBLT has grown out of Communicative Language Teaching (CLT), which, in our opinion, has suffered from several disadvantages completely unrelated to its core philosophy—namely, that learners learn a language best by using it to try to negotiate meaning.
It has been argued for decades now that the authentic negotiation of meaning should be a crucial component of language classes. However, this ideal has not always been met for several reasons. Perhaps the most important of these is the simple fact that looking at language as a collection of describable forms that can be easily presented, practiced and tested is easy to do, while describing what happens during authentic communication, not to mention then structuring it into a syllabus and assessing it, is not. As a result, there has been resistance to or considerable adaptation of truly communicative syllabuses.
In our experience of teaching EFL in Japan, where rote-learning and repetition drills have long ruled language studies, CLT has often been misunderstood or oversimplified. The CLT movement, which began taking root here in the late 1980s, seems to have given cause for some teachers and institutions to de-emphasize structured language studies altogether, and focus simply on “making conversation,” preferably with native speakers. Conversely, other instructors have continued having their students repeat set textbook dialogs and newly learned grammatically correct examples to each other, but now calling this “oral communication.” It seems to us that classroom realities (student motivation, class size, instruction time) as well as professional realities (a lack of understanding of second language acquisition methods and theory, a lack of preparation time), have often combined with the apparent simplicity of CLT to ensure the general failure of the approach in some Asian contexts.
The basic tenet of CLT, however, remains sound. Languages are best learned when there is an opportunity for students to focus on meaning rather than form, at least some of the time. The problem only exists in how to present these opportunities to students. “Making conversation” is oftentimes too vague. Not only do the students themselves become confused as to what exactly they should talk about and why (e.g., Talk about your summer vacation. Go!), from the teacher's perspective there is also little basis for evaluation or indeed systematic instruction. On the other hand, repeating set dialogs or substitution drills (e.g., Which _____ do you like best? 1) car 2) food 3) sport) allows for very limited meaning-making. Consider what the student doing such a drill would be concentrating on most: Whether they like Volkswagens or baseball, or whether they are forming their sentences correctly?
Advantages of TBLT
A task-based approach solves many of the criticisms traditionally associated with Communicative Language Teaching. To begin with, consider these characteristics of tasks:2
- Meaning is primary.
- Learners are not restricted in their use of language forms.
- Tasks should bear a relationship to real-world activities.
- The priority is on achieving the goal of the task.
- Tasks are assessed based on their outcome.
TBLT provides a structured framework for both instruction and assessment. Using tasks as the basic building blocks of syllabus design allows teachers to both sequence lessons and assess their outcomes, while at the same time creating reasonably authentic parameters within which students can communicate with each other for a purpose. Most importantly, it allows them to focus on what it is that they are saying to each other, rather than on how they are saying it. A task may be short and self-contained (e.g., ordering a pizza by telephone) or longer and more complex (e.g., organizing and publishing a student newspaper), but the tasks always involve a clear and practical outcome (e.g., The pizza arrives with the correct toppings, or the newspaper is printed and is recognizably a newspaper).
In a task-based approach, specific language forms should never be the primary focus, because it is important that students be allowed to make meaning in whichever way they see fit, at least at first. Teachers may assist or even correct students when asked, of course, but may not restrict the students' choice of which forms to use by explicitly teaching, say, the present continuous before the task is attempted. A post-task phase, on the other hand, is generally recognized by TBLT practitioners as useful. During this segment of the lesson, after the students have attempted the task, the teacher may choose to go over the language used, correcting specific errors and/or highlighting particularly well-suited forms that students may have attempted to use.
When considering TBLT, it is crucial to focus on the fundamental notion of authenticity, as tasks attempt to simulate, in a way that is as authentic as possible, what happens when students attempt real-world activities.
This has several advantages:
- Authentic tasks are intrinsically motivating. That is, students attempt them because they see that the task is, in itself, interesting and applicable to their lives.
- Targeted real-world tasks have much clearer outcomes that can be more easily assessed, unlike more general, or “open,” tasks such as having a conversation. For example, when a person attempts to order a pizza on the telephone in a second language, that person knows if he or she has “passed” or “failed” within a very short time—when the pizza does or does not arrive, with the correct toppings or not.
- Real-world activities can be looked at and sequenced in much the same way as grammar forms can—from simpler to more complex. For instance, ordering from a menu at a restaurant is easier than ordering by telephone for several reasons—students can use gestures, text and sometimes pictures; there is less information to convey (e.g., no address or credit card number); students may resort to single-word utterances. In the same way, telling a story is more complex than both examples above, because students now need to use connected sentences, time markers, pronouns and so on. It can be reasonably assumed that a student who can tell a story in English can also telephone for a pizza or order at a restaurant (but not vice-versa), in much the same way as we can reasonably assume that a student who can use conditionals can also use the present continuous (but again, not vice-versa).
Therefore, when a series of connected, themed tasks are sequenced in such a way as allows students to simulate a real-world context and perform at an increasing level of complexity, a variety of benefits occur. These include a purpose-driven recycling of vocabulary and language forms, a heightened sense of overall motivation, a marked increase in communicative confidence, scaffolded autonomy-building and a truly student-centered classroom. Much of the language learning thus occurs implicitly, as noticing on the part of the student, rather than as explaining on the part of the teacher.
A “strong” approach to TBLT
It has been argued that TBLT may not be the best way to develop basic language skills in the lowest ability levels, nor for very young learners. Because ours is a “strong” approach to TBLT, we generally agree with this perspective. Although many textbooks on the market today claim to be task-based, and are targeted across many levels including children and beginners, we should stress that these are, almost without exception, and often admitted to by the authors, a “weak” approach to TBLT. In other words, they make compromises with some of the tenets of TBLT in order to target children and beginners, especially by providing language-based activities such as embedded grammar points. We do not find fault with this practice beyond simply articulating the fact that these are not, strictly speaking, task-based courses. In fact, we think that many of these courses are very well designed in their own right and serve their purposes well.
We believe that a strong TBLT approach is built squarely upon a foundation of authenticity. Tasks must be as realistic as possible in order to engage students so that their meaning-making is also as authentic as it can be. A “weak” approach may be effective in an ESL environment, since the forms students learn in class may be authentically used outside the classroom almost immediately. In an EFL environment, however, such opportunities are usually limited, and therefore the more authenticity created inside the classroom, the better. For this to happen, tasks must not simply be authentic in their own right, but they must be authentically linked to each other as well, thus creating a sustained authenticity which allows for the recycling and reinforcement of the language forms used.
One other important aspect of TBLT is the underlying assumption that task complexity is somehow linked to language complexity. This is supported by various “language descriptor” systems developed over the years in different countries. Perhaps the first and best known of these is the ACTFL Guidelines. We recommend the Canadian Language Benchmarks (CLB) and the Common European Framework of Reference for Languages (both available online). The CLB document—The Canadian Language Benchmarks 2000: English as a Second Language for Adults—is especially clear and accessible for teachers new to TBLT.
The idea of task complexity as an organizing feature for language instruction is simple and intuitively clear, though the possibility of objective correlations between task and language complexity are still sometimes disputed by researchers. Briefly, however, we can generally assume that the complexity of any task is linked to the language needed to perform it. As an example, and at a very basic level, most teachers may agree that, if a student can successfully describe a photo of familiar, concrete items, then that student can: Make simple sentences (e.g., It's a dog.); modify nouns with adjectives (e.g., It's a red shirt.); use a couple of basic verb forms (e.g.; He is walking.); etc. This student would not necessarily, however, be able to tell a story because he or she may not be able to use more complex language, such as connecting sentences together coherently (e.g., The man walks to the store. He buys a drink there. Later, he goes home.). Certainly, there is no indication that this student may be able to give a formal presentation using a variety of verb forms and an expanded vocabulary, since that is, again, a more complex type of task. Thus, by organizing a syllabus according to increasing task complexity, even though there is no specific focus on language, we can slowly increase the language complexity attempted by students.
However, there are shortcomings to this approach. The debate continues on how tasks should be sequenced in terms of linguistic, cognitive and other complexities. There is also no strong consensus on whether students being able to perform one particular task are necessarily proficient at performing all tasks of the same type. Despite these and other such caveats, however, we strongly believe that, from the perspective of the classroom teacher, TBLT presents a clear and easy-to-use resolution to some of the problems inherent in Communicative Language Teaching.
Much more can, and indeed should, be considered with regards to TBLT, but we hope that the above will be enough to get teachers into the right frame of mind to approach Widgets.
Our primary consideration when organizing Widgets has been authenticity, which means that where there has been a conflict with the conventional interpretation of the literature, we have tended toward what “felt” more real. For example, we did not follow an explicit form of the pre-task/task/post-task structure, because we felt that by doing this we would be breaking some of the illusion of authenticity provided by the overall simulation. However, we have tried to stay true to the spirit of this framework by making our tasks fit into and support each other incrementally, so that each individual task serves to some extent as the pre-task phase for the one following it, and also as the post-task phase for the one which came before.
Regarding the “openness” or “closedness” of tasks,3 we have looked at the issue from a more complex perspective. Although most tasks in Widgets may appear to be open, in fact they “draw to a close,” so to speak. In Stage 2, for example, the smaller sub-tasks building up to having students create a new product idea allow a necessary freedom of discussion at first, but at the same time begin to converge students onto a range of outcomes that are more restricted. For instance, the product idea must be original, safe, useful, etc. While these may appear at first to be subjective criteria, students almost invariably come to the realization that there is indeed general agreement as to which products are safe and which are not, effectively creating “correct” and “incorrect” outcomes.
In other words, we have let the content re-inform our understanding of the theory rather than let theory dictate the content. We feel that our ideas have been supported by the positive results we and our colleagues have seen in the classroom.
We, like most teachers today, take it as a given that communicative
ability in a second language must be considered as a whole. That is,
communicative ability includes not only vocabulary, pronunciation and
grammar skills, but also the capacity to use these in real-world
contexts. It is this last point which is often missed by traditional
assessment tools, such as university entrance exams, often because it is
considered too time-consuming and subjective to try to assess
communicative ability. We hope to show you that task-based assessment is
neither time-consuming nor subjective, and in fact includes many other
advantages as well.
Task-based assessment is easy, straightforward and, above all,
meaningful for students and teachers alike. Simply put, one begins by
looking at the appropriate completion of any given task first,
and at the accuracy of the language used to complete it second.
If the student can achieve a particular goal, or “outcome,”
using English, then the student passes. Conversely, if they cannot
achieve the outcome in a generally acceptable manner, then they fail.
Here is a hypothetical task, including some specific parameters to be
||Tell a story
On a simple, familiar topic (e.g., family trip)
To a single sympathetic listener (e.g., a friend)
For a certain length of time (e.g., two minutes)
Note that the parameters are important in order to keep tasks at a
specific level. For example, if you were to change the “single
sympathetic listener” to “an audience of English teachers
during an entrance interview” it would make the same basic task
far more challenging.
Now, let us say you are grading the task outcome on a 10-point scale.
If the student has appropriately completed the task (i.e., They
have managed to tell a story according to these parameters, regardless
of how “good” it actually was.), they pass, and now have a
score of between 6 and 10.
If they could not accomplish the task (e.g., They could not be
understood at all, or if what they produced would not reasonably be
called “a story.”) then they fail. If they spoke reasonably
well but did not stay completely within the parameters (e.g., If they
spoke for only one minute or they spoke on an entirely different kind of
topic.) then they did not complete the task, and they fail. Of course,
as the teacher, you may always choose to make allowances in such cases,
but strictly speaking, in a task-based assessment model, this student
would indeed fail.
The next step is to assess how well the task was achieved.
Now we can look at things such as pronunciation, vocabulary and grammar.
Remember, if an appropriate outcome was achieved, then we already
know that the student's pronunciation, vocabulary and grammar are
at an “acceptable” level for the task, communicatively
speaking. They would not have managed to complete the task otherwise.
Thus, task-based assessment works something like this:
Was the task appropriately completed? Would the
outcome be reasonably recognizable by an “average” native
speaker of English as an example of its “type”? (In this
case, was it a story? Was there a beginning, a middle and an
end? Were the events in the story linked to each other coherently? Was
it clear enough to understand, despite any possible language problems
such as poor pronunciation or grammar mistakes?)
Yes (pass) No (fail)
If “yes,” how good was it? Read the
descriptors below and assign a grade from 6 to 10.
If “no,” were there at least some
redeeming qualities? Assign a grade from 1 to 5.
criteria for a pass:
||No grammar mistakes worth mentioning. Vocabulary use was very
appropriate. Pronunciation was exceptionally clear. Speech was
remarkably smooth and fluent. Gestures, facial expressions and manner
were always appropriate and natural.
small grammar, vocabulary or pronunciation mistakes. However, speech
was still very smooth and easy to understand. Gestures, facial
expressions and manner were appropriate and natural.
noticeable grammar, vocabulary or pronunciation difficulties. However,
overall speech was easy to follow and understand. Gestures, facial
expressions and manner were generally appropriate.
serious difficulties with grammar, vocabulary or pronunciation. Speech
was not always smooth and clear, but quite understandable. Did not
revert to first language. Gestures, facial expressions or manner may
have been somewhat distracting.
difficulties with basic grammar, vocabulary or pronunciation. Speech
was not always clear. Required some support from the listener.
Reverted to first language on occasion. Gestures, facial expressions
or manner were often distracting; nevertheless, generally
Example scoring criteria for a fail:
difficulties with basic grammar, vocabulary or pronunciation. Required
considerable support and patience from the listener. Often reverted to
first language; nevertheless, short sections of the speech could
sometimes be understandable.
difficulties with basic grammar, vocabulary or pronunciation. Required
considerable support and patience from the listener. Often reverted to
first language. Understandable only to a very sympathetic listener
familiar with the student's first language, such as a teacher.
not display an ability to use basic grammar structures. Spoke in two-
or three-word utterances using basic, but appropriate vocabulary. Used
other means to support speech, including relying very heavily on first
language. Difficult to understand even for a very sympathetic
listener; nevertheless, displayed some noteworthy quality, such as an
understanding of storytelling conventions.
not display an ability to use basic grammar structures. Spoke in two-
or three-word utterances using only basic vocabulary. Used other means
to support speech, including relying very heavily on first language.
Extremely difficult to understand, even for a very sympathetic
not be understood beyond basic set expressions such as “How are
you?” Made only single-word utterances, if any at all.
The underlying principle at work in task-based assessment is that
tasks can be organized in a hierarchy which parallels
“steps” in language proficiency because the language
necessary to perform any particular task ultimately indicates an ability
to perform that task's “type.”
This means two things:
- We can look at tasks in terms of relative difficulty. For
instance, “ordering a hamburger at the restaurant” is
easier than “ordering a pizza by phone,” which in turn is
easier than “giving one's impromptu opinion in a TV
interview.” This is because the language required for each is
increasingly more difficult. At a restaurant, one needs only to speak
in single words aided by gestures to be reasonably understood. On the
phone, it is necessary to be able to ask and respond to simple complete
questions with no visual cues. Finally, to participate in a TV
interview might require speaking for an extended period on an
unprepared topic, requiring a facility with grammar and a large,
- We can think of tasks as representative of certain
“types” of communicative acts. It is reasonable to
expect that a learner who can order a meal at a restaurant can also
function reasonably well at the dry cleaner's or rent a car in person.
Renting a car by phone, however, would be more like ordering a pizza,
since the learner could not rely on gestures and other means of
communicating. Finally, someone who could give a reasonable TV
interview could also be expected to, say, participate as a student in a
high school algebra course.
Although how to rank tasks according to complexity and how to
organize them into task types are still unresolved issues for
researchers and theorists, we have chosen to follow the lead of language
descriptor systems such as the Canadian Language Benchmarks and the Common
European Framework of Reference for Languages in our Student Book.
Nevertheless, our tasks do not necessarily correlate with either of
these, as our primary concerns were task complexity and authenticity. Widgets
has therefore been informed by what has worked in our classes first, and
by our knowledge and experience with the literature second. In the end,
however, we have noted extremely high correlations with all of the major
TBLT tenets proposed by researchers and theorists.
Marcos Benevides and Chris Valvona
||For an excellent short summary of TBLT (here referred to as TBI, or Task-Based Instruction), we recommend Jane Willis' introductory essay in Betty Lou Leaver and Jane R. Willis, Task-Based Instruction in Foreign Language Education (Washington, D.C.: Georgetown University Press, 2004).
||A more comprehensive, but very readable survey of TBLT, with an emphasis on classroom applications, is David Nunan, Task-Based Teaching (Cambridge: Cambridge University Press, 2004).
||Also comprehensive, but somewhat denser and exhaustively researched, is Rod Ellis, Task-based Language Learning and Teaching (Oxford: Oxford University Press, 2003).
||Visit www.language.ca for more information on the Canadian Language Benchmarks. The Canadian Language Benchmarks 2000: English as a Second Language for Adults by Grazyna Pawlikowska-Smith can be downloaded from www.language.ca/pdfs/clb_adults.pdf. (You will need Adobe Reader to view this document.)
||The Common European Framework of Reference for Languages can be downloaded from www.coe.int/t/dg4/linguistic/Source/Framework_EN.pdf. (You will need Adobe Reader to view this document.)
1. This article is intended to provide a brief explanation of TBLT as it applies to the Widgets course. For a deeper understanding of the theory and research behind the approach, we recommend that teachers read further. The number of articles and books on the topic has been growing exponentially in recent years, but a good starting point would be the three books listed in the “Further reading on TBLT” section.
2. From Nunan (2004, pg. 3). Note that Nunan is here drawing on Skehan, Bygate, Swain, Ellis and others, and that these characteristics of a task are in fact now very widely agreed upon.
3. In short, open tasks are those which do not have a predetermined outcome, such as a free chat, and closed tasks, such as describing a picture, are those which have one (or a few) “correct” outcomes. Research suggests that closed tasks are generally more useful for language practice (Ellis, 2003).