Causative constructions in Turkish International Corpus of Learner English (TICLE)

This study investigated the usage of English periphrastic causative constructions by Turkish learners of English as a Foreign Language (EFL). The grammatical area of causative constructions was chosen due to the significant differences between how Turkish and English encode causation. Furthermore, English periphrastic causative verbs possess a rich array of syntactic and semantic nuances that can be particularly challenging for non-native speakers. Given these reasons, it was believed that Turkish learners of English would struggle with this area of grammar. For the purposes of this analysis, the four most frequently used periphrastic causative verbs of "make," "have," "get," and "cause" were chosen and using corpus analysis methods, data from Turkish International Corpus of Learner English (TICLE) and Louvain Corpus of Native English Essays (LOCNESS) were compared. This comparison was a twofold process. First, in order to see any potential over or underuse, a quantitative analysis was conducted to compare the overall frequencies of the four verbs across both corpora. Second, the learner corpus TICLE was analysed to identify and categorize any syntactic and semantic misuses of the four periphrastic causative constructions. This was done to assess the extent of the learners' grasp of these constructions and to determine if there were any systematic error patterns. To further contextualize the findings, the results from the Turkish learner data were then compared to those of two other learner groups, namely the German and Japanese subcorpora of International Corpus of Learner English (ICLE). The purpose of this interlanguage comparison was to ascertain whether the characteristics identified in the Turkish learner data were a common property of interlanguage development in general, or if they were specific to the Turkish learners' linguistic background. Our analysis has found that there was a high number of both syntactic and semantic errors in Turkish learners' usage of English causative constructions, indicating a serious struggle with these structures. When contrasted with the German and Japanese learner groups, it was apparent that Turkish learners were more prone to errors. Moreover, it was found that Turkish learners used the cause causative significantly more than both natives and other learner groups who, in stark contrast, significantly underused the same verb. This finding, also coupled with the fact that Turkish learners exhibited a high rate of overuse of the "be" verb complement with cause strongly suggests that they are transferring the equivalent of the Turkish "neden olmak" (to cause to be) structure from their L1 to their L2. The findings and their significant pedagogical implications for English language instruction are discussed in detail. Keywords: Corpus linguistics, Turkish EFL learners, causative constructions, periphrastic causatives