Learning Japanese as a Second Language: A Processability Perspect ...

A major contribution of the EA approach to SLA is its promotion of the status of the learner’s IL from an undesirable version of the target language to a product of the active process of L2 learning. Despite this contribution, EA also has a few problems and limitations. First, it fails to grasp the whole picture of language acquisition because it focuses only on errors. An SLA theory should be able to explain what the learner can do as well as cannot do. Second, the category of errors proposed by EA researchers is problematic because it is difficult to identify the cause of errors; in fact, the source of error usually involves multiple factors, rather than a single factor (Schachter & Celce-Murcia, 1977). Third, a fatal problem associated with EA is its inability to explain ‘avoidance’ behaviour by L2 learners. This is best exemplified by Schachter’s (1974) study of the acquisition of the relative clause in English L2 by two groups of different L1 backgrounds: Group 1 (Chinese L1 and Japanese L1 learners) and Group 2 (Spanish L1 and Persian L1 learners). According to the study, Group 1 exhibited fewer errors than Group 2 in terms of error frequency. Schachter pointed out an ‘error’ of EA: Based on the result, EA would predict that Group 1 (Japanese/Chinese speakers) would have fewer problems with the production of English relative clauses because they made fewer errors. But in fact the opposite was true: Japanese/ Chinese speakers made fewer errors simply because they avoided the construction of English relative clauses because they knew that this structure was problematic for them. In conclusion, EA can explain what the learner does wrong but it cannot explain neither ‘avoidance’ nor what the learner can do correctly. SLA theory should be able to explain what the learner does incorrectly (errors) as well as what he/she can and cannot do (including what they may avoid), in order to look at the whole picture of L2 acquisition.

2.1.3. Creative Constructionism:
Morpheme Studies and Krashen’s Monitor Hypothesis

Morpheme Studies

The limitations exhibited in EA naturally led SLA researchers to consider the L2 learner’s whole performance rather than focusing only on errors.