Kılınış ve Söylem Hipotezlerindeki önermelerin test edilmesi: Orta seviye İngilizce öğrenen Türk öğrencilerin yazılı öykülerinden bulgular

The current study investigates the predictions of the Aspect Hypothesis (AH) and Discourse Hypothesis (DH) in the interlanguage of intermediate level of Turkish EFL learners. Specifically, it aims to examine the interaction among lexical aspect, verbal morphology, and discourse grounding by comparing learner narratives with instructional materials. A corpus-based, quantitative research design was employed with personal narratives written by eighty-eight learners and instructional data drawn from two textbooks and a graded reader. All finite predicates in both corpora were annotated for their lexical class, morphological form, and grounding. This process incorporated both manual analysis and the use of a large language model (ChatGPT-4o), representing a novel methodological contribution to the field. The annotated data was analyzed for significant associations and differences using cross-tabulations, Pearson's Chi-square, and one-way ANOVA. The findings revealed substantial empirical support for the predictions of both AH and DH, most notably the alignment of telic predicates with past morphology in foreground clauses. Furthermore, a high degree of structural similarity was observed between learner output and instructional input, reinforcing the influence of the input on learner production. These findings contribute to a deeper understanding of interlanguage development and yield pedagogical implications for curriculum design, particularly with respect to the selection and organization of instructional materials.