A Computer Learner Corpus-based Analysis of the Acquisition of Irregular Verbs : Optimizing the Learning List and Monitoring the Learning Process

Abstract

In this paper, we explored the possibility of developing EFL learning lists based on the frequency data obtained from both L1 and L2 corpora. We especially focused on the learning list of irregular verbs originally developed by Grabowski and Mindt (1995), which was based upon L1 corpora (Brown and LOB) only arid investigated if we could optimize the list by comparing the findings from L1 corpora with those from the Japanese EFL Learner (JEFLL) Corpus data. We also argued that further steps should be taken to monitor the interlanguage to see how these irregular verbs are actually used in L2 learners' performance data. For this purpose, we investigated how irregular verbs were actually used in the learner corpus to see if there was any difference in the rate of correct use of these verbs regarding the classifications based upon the Vendler framework. The results were not very conclusive, due to the treatment of the past tense only, but still showed an interesting possibility of using a learner corpus for monitoring learner performance after introducing the items to be acquired. The paper concludes that while the primary source of target language performance is L1 corpora, L2 learner corpora have a great potential to provide valuable data for optimizing and monitoring the learning items for L2 learners.