L2 Errors in Korean Nominative and Accusative Case Particles：A Learner Corpus-Based Analysis

다국어 초록 (Multilingual Abstract)

This study investigates error patterns in the use of Korean case particles by English-speaking learners across three proficiency levels, using data from a learner corpus. In Korean, case particles are morphologically realized as -i/ka for the nominative case and -ul/lul for the accusative case. In contrast, English lacks overt case marking, except in certain personal pronouns. While Korean case particles can be omitted, such omissions are determined by specific discourse/semantics-related conditions. Thus, English-speaking learners must not only acquire the structural properties of Korean noun phrases but also internalize the conditioning factors required for case particle use. This study adopts the Feature Reassembly Approach (Lardiere, 2009), which emphasizes the complexity of the learning process involved in acquiring relevant features. This study has two primary aims: (i) to identify learning challenges faced by English-speaking learners based on crosslinguistic differences, and (ii) to determine whether learners exhibit distinct error patterns depending on their proficiency level. Using error-annotated data from the Korean Learner Corpus (KLC), the study examines four types of error patterns. The findings show that omission and replacement errors are the most common across all proficiency levels including the advanced level. In particular, omission errors are associated with various conditioning factors, and the lack of further development at the advanced level may suggest that it takes considerable time to completely acquire them.