Recurrent phrases in learner English: A corpus driven approach

Insights from corpus linguistics have contributed considerably to the understanding of recurrent nature of language, that is, language use, to a great or lesser extent is marked by routine or recurrence. Studies designed on native language have inspired the researchers from the field of second language acquisition, and investigations of recurrent word combinations, the way words co-occur with other words, have recently gained more and more attention in the field of second language acquisition. Advent of Contrastive Interlanguage Analysis (CIA) has highlighted unprecedented features that characterize interlanguage. Within the framework of CIA, adopting a corpus-driven recurrent word combination method, the present study has focused on the 2-to 6- word combinations in both spoken and written interlanguage to investigate Turkish learners' tendencies in designing their discourse in English. The main objectives are to explore the use of recurrent phrases in Turkish EFL learners' interlanguage and to compare and contrast them with native speakers' use of recurrent phrases across written and spoken language in terms of both quantitative and qualitative variation. The primary source of material for this study is two major corpora: International Corpus of Learner English (ICLE) and Louvain International Database of Spoken English Interlanguage (LINDSEI). For non-native data, Turkish components of these corpora have been used. The overall findings have revealed some common aspects of Turkish learners' interlanguage in terms of recurrent word combinations. Comparisons with native speaker language have painted a complex picture of similarities and differences regarding frequently used word combinations both in spoken and written language. The results are discussed referring to the possible effects of first language, instructional factors and register interference.