analysis’. As Brown (1987: 153) stresses, contrastive analysis, deeply rooted in
behaviourism and structuralism, claimed that ‘…a scientific, structural analysis of the
two languages in question would yield a taxonomy of linguistic contrasts between
them which in turn would enable the linguist to predict the difficulties a learner would
encounter’. However, things in language learning came out to be not as simple as the
audio-lingual method assumed. When submitted to empirical investigation,
contrastive analysis failed to provide a convincing account for the occurrence of
second language learners’ errors as researchers (e.g. Dulay and Burt, 1973; 1974a)
found that a large proportion of them could not be explained by L1 interference. Ellis
(1990: 9) referring to a number of longitudinal studies such as those carried out by
Ravem (1968) and Huang and Hatch (1978) stresses that these assign two significant
characteristics to the errors made by learners. The first is that learners’ errors are
developmental rather than behavioural. That is, learners’ errors are independent of
either their L1 or L2 and express an attempt of them to systematize the input they
receive. Corder (1967) saw the making of errors as evidence of learner-internal
processing. Selinker (1972) used the term ‘interlanguage’ to refer to an interim
grammar that is a single system composed of rules that have been developed via
different
cognitive
strategies
(for
example,
transfer,
overgeneralization,
simplification, and the correct understanding of the target language). At any given
time, the interlanguage grammar is some combination of these types of rules.
According to Adjemian (1976), an interlanguage is a linguistic system in its own right
and as such it is a natural language and it is entirely functional. The second finding is
that there appears to be a natural sequence of acquisition for many grammatical
features, as learners from different linguistic backgrounds were found to display
remarkable regularity in the stages of development of certain linguistic features. Thus,
it was argued that L2 acquisition is a process of creative construction not dissimilar to
that found in L1 acquisition, an issue which Ellis (1999: 8) refers to as L1=L2
hypothesis.
It is worth stressing, however, that, although behaviourism cannot provide a full
account of the way languages are learnt, there are, at least, some aspects of language
learning which could be explained according to this model. An example of this kind,
according to Ellis (1990: 28), is the way learners acquire elements of language such as
vocabulary or ready-made phrases. Littlewood (1984: 47-48) who classifies readymade phrases into ‘routine formulas’ (i.e. utterances which the learner produces as a
20