conclusions are clear. For them, "strategies can be taught", and "academic language learning
is more effective with learning strategies". Bialystok who also believes in the instruction of
learning strategies states that the teaching of learning strategies is only possible if it gives
the students the opportunity to "analyse and control their linguistic system" (1990,143),
something that O'Malley and Chamofs studies accomplished. However, Little (1995) still
thinks that "the extent to which learning strategies are teachable and learning styles are
directly modifiable remains an open question" (177). With the intention to encourage a
realistic but still positive view on this issue, Chamot and Rubin say that learning strategies
are not a "magic formula" but they state that
still the evidence...leads us to feel confident that such instruction, properly carried
out, can positively assist language learners to become more actively engaged in
their own learning processes, thus taking on greater responsibility for learning.
(Chamot and Rubin, 1994, 774)
A very sensible comment is found in Horwitz (1987), who does not refer to the learning
strategies but to the use of them: "productive use of learning strategies...is an acquired skill"
(my italics) (120), a comment which stresses the importance of instruction.
Fourth, learning strategies involve cognitive work (they are thoughts), which
sometimes can lead to overt behaviour. Hence, some of them have observable outcomes,
which does not mean that the strategies themselves are visible. From the point of view of the
researcher, this fact is very important for she has to develop methodological devices for the
indirect study of learning strategies.
To sum up, research on the area states that 1) second language learning strategies can
be defined as strategies that are carried out in order to Ieam knowledge and skills, 2) that
their use, at the beginning is conscious but, eventually, they become procedural knowledge,
which means that the user is not aware of using them, 3) that learning strategies can be
learned (and, under the appropriate conditions, taught) and 4) that they involve cognitive
processes, hence they cannot be equated with behaviour, which also means that they are not
directly observable.
53