The acquisition of definiteness expressions in non-native Japanese by English- and Korean-speaking learners
Keisuke Kume · White Rose eTheses Online (University of Leeds, The University of Sheffield, University of York) · 2021
This thesis investigates learnability problems and the role of L1 in L2 syntax-semantics mappings, through a comparison of English- and Korean-speaking Japanese learners’ acquisition of two Japanese definiteness properties: overt definiteness marking by the demonstrative sono and covert definiteness distinction through word order change between numeral quantifier constructions. The main objective is to examine distinct predictions based on recent accounts of learnability and L1 influence: the Feature Reassembly Hypothesis (FRH) (Lardiere, 2008, 2009) and the cline of difficulty in feature acquisition (Cho & Slabakova, 2014; Slabakova, 2009). The FRH predicts that both definiteness properties will be acquired faster by Korean speakers due to positive transfer of the L1 corresponding properties. On the other hand, according to the cline of difficulty, the overt property will be acquired before the covert property, irrespective of the L1; and the L1 advantage of Korean is expected with the overt property but not with the covert property. Specifically, the corresponding covert L1 definiteness property (Korean numeral constructions) will not facilitate the acquisition of the same property in the L2, but the L1 functional morphology distinguishing definiteness overtly (English articles) will, hence English speakers will acquire the covert property faster than Korean speakers. These predictions were tested by means of an acceptability judgement task and a self-paced reading task. Although the self-paced reading data offer no evidence of the relevant knowledge even with native Japanese controls, the judgement data indicate that whereas both L2 groups can acquire the overt property equally well, the Korean group tend to exhibit more consistent target-like performance with the covert property than the English group. Based on these results, it is proposed that the necessity of feature reassembly (reconciliation of L1-L2 differences within overt/covert category) plays a bigger role than overt vs. covert feature realisation, contra the cline of difficulty.