SAC counsellor can help most learners is to assign contents to study (according to their
own needs) in order for them to be focused. I believe that this is one of the areas in
which the counsellor needs to deploy her authoritative knowledge.
Second, to a greater or lesser degree, the learners showed the use of inner guides
that let them decide what to pay attention to in order to notice the grammatical form. It
seems to me that following an inner guide in self-learning prevents the problems learners
necessarily confront when teachers’ initial presentation “cannot resemble the learner’s
internal one” (Johnson 1996,111, based on Prabhu, 1985). As one of the learners put it:
Ga: the teacher may not know the way I leam. He has so many
students. He tends to generalise. Γd rather study by myself
This inner guide allows the learners to make decisions along a continuum of noticing.
The extremes of this cline are noticing for the learner, where formulated information is
overt and explicit and noticing by the learner where most things are implicit and have to
be worked out by the learner (Batstone; 1994,72). After all, these are the types of
decisions a self-directed learner is able to make.
Finally, in all the cases of the Oaxaca/97 project analysed in this section, we have
entered into the realms of metalanguage, in the three different kinds that Widdowson
(1997) has identified: metalinguistic description, exemplification and analysis.
According to Widdowson,
The purpose of language pedagogy (is) to establish a metalanguage of
description, exemplification and analysis to offset the disadvantages of
authentic language data (1894)
I believe that making the learners consciously work on metalinguistic awareness
is a good way to “offset the disadvantages” of the common top-down approach to
authentic materials they usually work with5
Metalanguage in a classroom context, to follow Widdowson's argument, is
teacher controlled. By contrast, in the present study, metalanguage was controlled by
three different agents: first, I, as a counsellor, took control of the metalanguage by
“imposing” them a grammatical form to work with; second, the authors of the books
controlled the metalinguistic input by making decisions of what to include, and what to
leave out, in their metalanguage presentations. However, in the case of self-direction,
there was a third agent, the learner, who also controlled the input she wanted to pay
attention to.
In terms of the use of metalanguage, my argument is in line with Widdowson:
220