All of the questions (and many of the proposed answers) that were raised in these early
investigations of tasks and their role in training and teaching mirror similar discussions in
relation to Task-Based Language Teaching. In this chapter, we will outline the critical issues
in Task-Based Language Teaching and provide examples of what task-based teaching is
supposed to look like.
Approach
Theory of language TBLT is motivated primarily by a theory of learning rather than a theory
of language. However, several assumptions about the nature of language can be said to
underlie current approaches to TBLT. These are:
LANGUAGE IS PRIMARILY A MEANS OF MAKING MEANING
In common with other realizations of communicative language teaching, TBLT emphasizes
the central role of meaning in language use. Skehan notes that in task-based instruction (TBI),
"meaning is primary . . . the assessment of the task is in terms of outcome" and that taskbased
instruction is not "concerned with language display" (Skehan 1998: 98).
MULTIPLE MODELS OF LANGUAGE INFORM TBI
Advocates of task-based instruction draw on structural, functional, and interactional models
of language, as defined in Chapter 1. This seems to be more a matter of convenience than of
ideology. For example, structural criteria are employed by Skehan in discussing the criteria
for deter-mining the linguistic complexity of tasks:
Language is simply seen as less-to-more complex in fairly traditional ways, since
linguistic complexity is interpretable as constrained by structural syllabus
considerations. (Skehan 1998: 99)
Other researchers have proposed functional classifications of task types. For example,
Berwick uses "task goals" as one of two distinctions in classification of task types. He notes
that task goals are principally "educational goals which have clear didactic function" and
"social (phatic) goals which require the use of language simply because of the activity in
which the participants are engaged." (Berwick 1988, cited in Skehan 1998: 101). Foster and
Skehan (1996) propose a three-way functional distinction of tasks — personal, narrative, and
decision-making tasks. These and other such classifications of task type borrow categories of
language function from models proposed by Jakobson, Halliday, Wilkins, and others.
Finally, task classifications proposed by those coming from the SLA research tradition of
interaction studies focus on interactional dimensions of tasks. For example, Pica (1994)
distinguishes between interactional activity and communicative goal.
TBI is therefore not linked to a single model of language but rather draws on all three models
of language theory.