Exploring the development of 'verb gram' sequences and grammar patterns in L2 writing: A case study of L1 Korean speakers

Abstract

Usage-based approaches claim that language acquisition is sensitive to frequency. Thus, learners can utilize and produce more schematic language structures as they gain more exposure to the language. This understanding has made a significant contribution to learner corpus research. To date, a plethora of studies has explored various linguistic structures, such as constructions, in terms of usage. However, some known research gaps should be addressed. First, most of the studies utilized a smaller learner corpus. Second, studies used a top-down approach where the analysis was based on pre-selected constructions. However, a bottom-up approach could shed a valuable understanding of the kind of language structures that the learners can produce at different developmental trajectories. Lastly, such a study has not been conducted in the L1 Korean context. Therefore, this research aimed to explore the use of verb grammar patterns across L1 Korean speakers through the Cambridge Learner Corpus (CLC). The CLC is the largest corpus to date and contains written exam scripts from six different proficiency levels. The current study implemented a bottom-up approach to explore the verb 4-gram sequences in L1 Korean speakers. This study is quasi-longitudinal, where the proficiency levels were used as a proxy to measure language development. To do so, an L1 Korean speaker subcorpus was created from the CLC, which consisted of 1,973 learners and 914,343 words. The 4- gram verb sequences were extracted from A1 to C2 proficiency levels, and the examples from the key sequences were categorized based on pattern grammar. The findings were compared with other L1 speakers from the CLC and L1 English speakers from the written academic portion of the British National Corpus (BNC). The findings reveal that there is a clear transition towards a greater variety of complex grammar and a wider range of meanings that become more prevalent in higher proficiency levels. Furthermore, idiomatic expressions, phrasal verbs, and collocations were present across all levels, although they become more prevalent with higher proficiency levels. The comparison iii with other L1 speakers and L1 English speakers shows a similarity in the use of most frequent verb 4-gram sequences, although they tend to diverge as the structures become less frequent. Overall, this study proposes that there is a clear development pattern in L2 usage among L1 Korean speakers and that the teachers and curriculum decisions should be made based on such findings. The study also calls for further research into different parts of speech, such as nouns and adjectives.