Interlanguage

Date: 2015-10-07; view: 472.

Error analysis

This weak version of the hypothesis switches the emphasis from prediction of difficulty to observation of difficulty, which is then explained with reference to contrast-ive analysis. In other words, the emphasis switched to a primary focus on the analysis of learner errors, error analysis (see e.g. Richards 1974). Much error analysis, how-ever, especially in the US, did not regard the L1 as influential at all, or regarded its influence as minimal, on learner errors (e.g. Dulay and Burt 1972, 1973, 1974a, 1974b, 1974c). Instead, the learning process and the learner's active part in it were placed in focus, and the new concept of interlanguage became central in theorizing about second language learning.

The notion of an interlanguage was first broached by Nemser (1961/1971) and Briere (1964/1968), but it is best known through the work of Larry Selinker (1972, 1992, 1996). It arises from the observation that learners often produce structures that exist neither in their first language nor in the language they are learning and which (it seems) no native speaker of any language ever produces (Selinker 1996: 97).

Interlanguage competence is of three types: fossilized competence, functional com-petence and transitional competence (Selinker 1996: 97). The notion of fossilized com-petence derives from Corder (see Selinker 1996: 98). The idea is that many L2 learners appear to reach a plateau in their learning, ceasing to improve any further. On their way there, they pass through a number of stages, which are therefore transitional. Some learners achieve competence in restricted domains only, enabling them to use the new language mainly for specific purposes, and it is this kind of competence that Selinker refers to as ‘functional com-petence' – the notion appeared originally in Jain (1969, 1974).

In Selinker's work (1996) the idea of L1 influence remains in his claim that there is firm evidence that L2 learners' preferred learning strategy is the search for inter-lingual identifications, a notion derived from Weinreich (1953/1968) (Selinker 1996: 97).