ACQUISITION OF FUNCTIONAL MORPHOLOGY IN L2 RUSSIAN BY ADULT L1 TURKISH LEARNERS: EVIDENCE FROM SPLIT D-LINKED WH-QUESTIONS

Abstract

This study explores the acquisition of functional morphology and adjective agreement in L2 Russian by adult L1 Turkish learners in the generative perspective. Our research focuses on the acquisition of adjective agreement in split discourse linked wh-questions. The question posed is to what extent L2 Russian learners make correct decisions in connecting the wh-word, specified for phi-features, with its headword through employing implicit knowledge of the uninterpretable features realized as an inflection. The above domains, along with NP splitting, are absent from the participants’ L1. The enquiry specifically examines L2 acquisition of short- and long-distance NP splits. This domain of L2 Russian is predicted to be unacquirable by the Interpretability Hypothesis (IH, Tsimpli & Mastropavlou, 2007) and the Shallow Structure Hypothesis (SSH, Clahsen & Felser, 2006), whereas the Bottleneck Hypothesis (BH, Slabakova, 2008) and the Full Transfer/Full Access Hypothesis (FTFAH, Schwartz & Sprouse, 1996) regard it to be acquirable. The data come from the results of a Semantic Entailments task administered via Google Forms, where 64 adult L1 Turkish/L2 Russian learners of A2 through C2 proficiency levels selected a response to split discourse-linked wh-questions supplied with a preceding context. 56 L1 Russian speakers constitute the control group. Six conditions are utilized through manipulating noun genders and the inflection on the wh-word. Our findings demonstrate a decreased accuracy in the L2 Russian group: 84% for short-distance splits and 62% for long-distance splits; the L1 group performs over the top and displays no variability regarding split types. Throughout proficiency levels, we observe a stark difference in L2 Russian learners’ treatment of short-distance and long-distance conditions: accuracy is relatively high for short-distance splits (72% in A2 and 94% - in B2/C2). Conversely, accuracy for long-distance splits is 38% in A2 learners but gradually rises to about 84% in B2/C2 levels. This outcome raises issues as to why L2 learners’ accuracy is decreased with accusative case morphology, which is acquired prior to dative case morphology. We suspect the reason to be thr additional processing load associated with long-distance splits. Hence, this domain of L2 Russian, though challenging at lower levels of proficiency, may be successfully acquired at higher levels, which overall supports the FTFAH and the BH, and casts doubt on the IH and the SSH. The results of the study can be applied in L2 syllabus preparation: special attention should be paid to designing activities aimed at developing processability skills in L2.