Learner Roles and Characteristics
Richards and Rodgers (2001) and Scarcella and Oxford (1992) identified possible task roles for learners, such as group participant, monitor, risk-taker/innovator, strategy-user, goal-setter, self-evaluator, and more. Others (Honeyfield, 1993; Nunan, 1989; Oxford, 1990) have also discussed learners' task roles. A particularly important learner role in a task situation is that of task-analyzer. The learner must analyze task requirements and find suitable strategies to match them.

The learner can take control of the task-that is, be responsible for his or her performance on the task-by considering the task requirements and employing learning strategies to accomplish the task more efficiently and more effectively (Cohen 1998; O'Malley & Chamot, 1990; Oxford 1990). On the part of the learner, this involves a serious commitment, motivation, confidence, clarity of purpose, and willingness to take risks (Dörnyei 2001; Dörnyei & Schmidt, 2001; Honeyfield, 1993; Oxford, 1996; Skehan, 1998b; Willis, 1996a, 1996b, 1998), but these may be dampened by language anxiety (Arnold, 1998; Oxford, 1998; Young, 1998).

Learning styles are likely to affect choice of strategies for accomplishing tasks (see Oxford, 2001). Learning styles also make a difference in which tasks are perceived as difficult by individual learners. For example, face-to-face communication tasks might be viewed as easier for a person with an extroverted learning style than an introverted learning style. Learners whose learning style is highly analytic, concrete-sequential, and/or closure-oriented might perceive greater ease in accuracy- and form-focused tasks than fluency tasks.

Overall Task Difficulty
Honeyfield (1993) specified the following influences on general task difficulty: procedures to derive output from input; input text; output required, such as language items (vocabulary, structures, etc.), skills, or subskills; topic knowledge; text-handling or conversation strategies; amount and type of help given; roles of teachers and learners; time allowed; and learner characteristics, such as motivation, confidence, and learning styles. For Skehan (1996a), factors related to task difficulty include: code (language) complexity, cognitive complexity (cognitive processing, cognitive familiarity), and communicative stress (time, modality, scale, stakes, and control).

Summary of Analyzing Tasks for Task-Based Teaching and Learning
This section has discussed factors that are often analyzed with regard to L2 tasks. Some of the major factors are complexity (linguistic and cognitive); overall difficulty, which is not the same as complexity; and roles of learners and teachers. How we can sequence tasks and parts of tasks is the topic of the next section.

3. Sequencing Tasks for Task-Based Teaching and Learning
As noted by Richards and Rodgers (2001) and Willis (1996a, 1996b, 1998), a task has a natural series of stages, such as preparation for the task (pre-task), the task itself, and follow-up (post-task). Many L2 learner textbooks now follow this practice. In addition, tasks are often placed into a sequence as part of a unit of work or study. Sequencing is a major issue in a task-based syllabus. Swales (1990), tasks are "…sequenceable goal-directed activities…relatable to the acquisition of pre-genre and genre skills appropriate to a foreseen or emerging . . . situation" (p. 76, in Salaberry, 2001, p. 102). Skehan (1998b) noted that tasks have discernable implementation phases, for which there should be clear criteria for outcomes assessment.

The traditional presentation-practice-production (PPP) teaching/learning cycle was at one time virtually the only acceptable L2 task sequence. In the PPP cycle, grammar presentation came first, followed by controlled and less controlled practice and then by actual production. However, Willis' (1996a, 1996b, 1998) task-based model offers a task cycle that opposes the PPP sequence. In this model, which effectively combines meaning and form, the communicative task comes before the focus on form (language analysis and practice). Another special feature is that students not only do the task but also report on it. Willis' framework consists of the following:

*Pre-task - introduction to the topic and task.
*Task cycle

o Task planning
o Doing the task
o Preparing to report on the task
o Presenting the task report

*Language focus - analysis and practice (focus on form).


Nunan (2004) argued in favor of units based on topics or themes in which Halliday's (1985) three groups of macrofunctions are divided into microfunctions, each linked with certain grammatical structures. Nunan's task-based syllabus contains six stages per unit:

*schema building,
*controlled practice embedded in a context (unlike traditional controlled practice),
*authentic receptive skills work,
*a focus on form (lexical and/or grammatical),
*freer practice ("communicative activities"), and at last
*the (communicative) task itself.
It is interesting that Nunan, unlike Ellis (2003) and Long (1985, 1991, 1997), waited until the very end of the process to include the communicative task. In Nunan's model, the task is a culmination of all other work. In this sense, as noted by Feeney (2006), this is not too far from the PPP format, except that Nunan's controlled practice occurs within more of a communicative context than is usual with the PPP arrangement. Nunan's focus on form occurs before both freer practice and the task, whereas Willis's (1996b) model employs a focus on form after the task.

Long's (1985, 1991, 1997, 2005) task-based language teaching model presents a focus on form, which involves meaning, structure, and the context of communication. The model follows the following sequence of task development, implementation, and assessment/evaluation:

*Needs analysis to identify target tasks
* Classify into target task types.
*Derive pedagogic tasks.
*Sequence to form a task-based syllabus.
*Implement with appropriate methodology and pedagogy.
*Assess with task-based, criterion-referenced, performance tests.
*Evaluate program.
In Long's model, tasks are selected based on careful analysis of real-world communication needs. Such tasks are particularly important-even catalytic-for L2 learning because they can generate useful forms of communication breakdown (Long, 1985). The teacher offers some kind of assistance to help the learner focus on form at the point when it is most needed for communication. This is the moment when meaning meets form. While not explaining the learner's error, the teacher provides indirect assistance so the learner can solve his or her own communication problem and can proceed to negotiate meaning still further. Long (1997) presented the following typical instructional sequence for a "false beginner" class of young adult prospective tourists.

*Intensive listening practice: The task is to identify which of 40 telephone requests for reservations can be met, and which not, by looking at four charts showing the availability, dates and cost of hotel rooms, theater and plane seats, and tables at a restaurant.
*Role-playing: The learners take roles of customers and airline reservation clerks in situations in which the airline seats required are available.
*Role-playing: The learners take roles in situations in which, due to unavailability, learners must choose among progressively more complicated alternatives (seats in different sections of the plane, at different prices, on different flights or dates, via different routes, etc.).
In this model, the exact sequence of any given task or set of tasks would depend on the learners' needs, which shape the goals of instruction.

Ellis (2003b) distinguished between (a) unfocused tasks (e.g., ordinary listening tasks or interactions) and (b) focused tasks, which are used to elicit a particular linguistic feature or to center on language as task content. He cited three principal designs for focused tasks: comprehension tasks, consciousness-raising tasks, and structure-based production tasks. Elsewhere (Ellis, 2003a) presented a sequence of tasks for helping learners become more grammatical, rather than for attaining the elusive goal of mastery. The sequence includes:

*Listening task, in which students listen to a text that they process for meaning).
*"Noticing" task, in which students listen to the same text, which is now gapped, and fill in the missing words.
*Consciousness-raising task, in which students discover how the target grammar structure works by analyzing the "data" provided by the listening text.
*Checking task, in which students complete an activity to check if they have understood how the target structure works.
*Production task, in which students have the chance to try out or experiment with the target structure by producing their own sentences.
Johnson (1996), Skehan (1998b), and Willis (1996b) discussed sequencing of tasks according to methodological task features, such as extent of communication (negotiation of meaning), task difficulty, and amount of planning allowed. Others have discussed how to sequence tasks to reflect the developmental sequence of language acquisition. Skehan (199b) suggested targeting a range of structures rather than a single one and using the criterion of usefulness rather than necessity as a sequencing criterion.

Salaberry (2001) argued that a successful task sequence leads learners to: (a) communicate with limited resources, (b) become aware of apparent limitations in their knowledge about linguistic structures that are necessary to convey the message appropriately and accurately, and finally, (c) look for alternatives to overcome such limitations. Building on the work of McCarthy (1998), Salaberry offered a pedagogical sequence of four stages, which for the learner would be involvement, inquiry, induction, and incorporation. For the teacher the corresponding four-step sequence is introduction of the topic, illustration, implementation, and integration. See Table 3. This sequence is very detailed and includes multiple tasks at each stage.

Table 3 Four stages of teaching/learning showing sequence of tasks





Teacher  Learner Salaberry's example
1. Introduction of topic  1. Involvement (motivation to
participate in the task)
Teacher illustrates particular features; students rate various movie reviews written by movie critics on a scale from the most positive to the most negative.
2. Illustration  2. Inquiry (communicative analysis of language in communicative context; mostly initiated by learners, not the teacher) Teacher reads a movie narrative and asks students to identify events in the plot (in infinitive form); students separate main events from minor events; students reconstruct story in writing in present tense.
3. Implementation  3. Induction (development of hypotheses about structure and functions of the language) Students do a listening comprehension task: place pictures of main movie events in correct order. Then they listen to the tape again to write down as many plot events in past tense as possible while tape is played to reconstruct whole plot, including minor events (modified dictogloss). Students have not yet had a formal explanation of past tense endings, but teacher can informally give past tenses of various verb types from student narratives in #2.
During the [essential] debriefing stage students may be given the actual script that was read to them so that they can compare it to their transcription; this is crucial for allowing students to verify, modify, or reject their hypotheses (from induction). Learner controls the learning process.

4. Integration  4. Incorporation (assimilation of knowledge about new L2 features in a way productive to the overall L2 system) Students produce their own movie scripts (incorporation). For instance, they can be asked to write a dialogue for a series of (scrambled) pictures that recount a possible eye-witness account of an event parallel to the movie plot (#3). They act out the scene (concrete outcome).

Source: Summarized from Salaberry (2001, pp. 108-110).

It is evident that no consensus yet exists about the best way to sequence tasks or to sequence elements within tasks. This is one of the key areas of research needed in the field. The next section offers a set of implications for research.

4. Implications for Future Research
Researchers have made significant strides in this field. However, it will be important to keep focusing on what is meant by "task-based L2 teaching and learning." The term can evoke many different images, depending on which theorists and models are involved and on various and locations in which such teaching occurs. We have seen many variations and possibilities above. The definitional and conceptual question,

*What do we mean by task-based learning and teaching?" can be broken down and elaborated as a series of questions:
*What are optimal or at least relevant types of task-based teaching to fulfill different learning goals of diverse students of different ages, genders, L1 backgrounds, cultural backgrounds, backgrounds, needs, learning styles, interests, and occupations?
*What are the most relevant criteria for sequencing tasks in task-based teaching? Do these criteria differ by any of the factors just listed?
*With a focus on form, does a given sequence of tasks work better, or should tasks be spontaneously determined based on evident learner needs at the time?
*How does the ordinary teacher find (or create) a task-based syllabus that fits the authentic language needs of his or her students?
*Can an off-the shelf task-based syllabus ever work for multiple age groups in diverse settings in different parts of the world?
*How much does cultural background influence the acceptability of different task types, input, and sequencing?
From these questions and from the whole article it is clear that task-based teaching and learning as a field is an exciting field that is experiencing much ferment at this time. Task-based teaching and learning potentially offer great riches if explored by teachers in their dual roles as instructor and action researcher. Professional researchers can provide additional answers to the questions raised here. The answers will enhance the teaching and learning of languages around the world. The ultimate beneficiaries will be the students whose needs will be more fully met if the questions are clearly raised, explored, and answered.