This dissertation reexamines certain constructions containing the Korean morphemes -i/ka and -ul/lul, traditionally analyzed as nominative and accusative case markers, within the theoretical framework of Chae (2020, 2023). Chae argues that they are de...
This dissertation reexamines certain constructions containing the Korean morphemes -i/ka and -ul/lul, traditionally analyzed as nominative and accusative case markers, within the theoretical framework of Chae (2020, 2023). Chae argues that they are delimiters rather than case markers and that their distribution exhibits a complementary pattern conditioned by the stativity of the context: -i/ka appears in stative contexts, while -ul/lul is restricted to dynamic contexts. This study provides theoretical and empirical validation for Chae’s hypothesis, employing a comprehensive methodology that integrates grammaticality judgment experiments with extensive corpus analysis. The primary contributions are twofold. First, this study validates and further refines Chae’s delimiter approach through a methodical investigation of auxiliary predicate constructions (APCs). Crucially, this study expands upon Chae’s (2023) original scope by analyzing the previously underexplored [VP-auxiliary predicate] unit. This detailed examination demonstrates the applicability and consistent explanatory power of Chae’s delimiter approach in this intermediate structural domain, establishing an analytical precedent for subsequent analyses. Second, building upon this validated and refined theoretical foundation, this study demonstrates that analyzing -i/ka and -ul/lul as delimiters provides a more efficient, unified, and coherent explanation for a diverse set of constructions that have historically presented significant typological and analytical challenges. These constructions, all involving -i/ka or -ul/lul, encompass case alternation (CA) phenomena, case stacking (CS) phenomena, multiple nominative constructions (MNCs), multiple accusative constructions (MACs), and multiple verbal noun constructions (MVNCs). Under this analysis, MNCs and MACs, for instance, are reinterpreted not as violations of case-theoretic principles, but as regular manifestations of multiple context-sensitive delimiters. Consequently, this approach obviates the need for construction-specific stipulations, attributing these patterns to general syntactic principles grounded in the nature of the delimiters -i/ka and -ul/lul. The findings challenge conventional case-based analyses and demonstrate the empirical advantages of a delimiter-based reanalysis, which offers a more parsimonious and consistent account of the observed linguistic phenomena under consideration. By assuming that -i/ka and -ul/lul are delimiters, this study contributes to clarifying issues concerning case and grammatical function. Ultimately, it illustrates how a critical reevaluation of grammatical primitives can resolve long-standing, seemingly anomalous typological puzzles.