An Analysis of the Effect of the NEG Raising on the Strength of NPIs in English and Korean

This study examines the licensing mechanisms of Negative Polarity Items (NPIs) in English and Korean, with a particular focus on the role of NEG raising and its effect on NPI strength. Previous literature has not fully accounted for cross-linguistic variation in NPI behavior, especially concerning the interaction of syntax and lexical properties. To address this gap, the study adopts the strong/weak NPI distinction proposed by Zwart (1990) and Nam (1994), and applies Collins and Postal’s (2014) syntactic framework of binary (reversal) and unary NPIs to Korean data, drawing on native speaker judgments and textual examples. The analysis shows that in English, strong NPIs resist alternation with non-NPI indefinites, discourse anaphora, and VP ellipsis but allow negative inversion and intensives. In Korean, strong NPIs permit VP ellipsis and intensives but restrict alternation with non-NPI indefinites, while weak NPIs alternate with non-NPI indefinites under implicative readings. Crucially, whereas NPI licensing in English relies on clausal neg-raising predicates (CNRPs) and structural configurations, Korean NPI distribution is largely determined by lexical properties and remains unaffected by NEG raising. By integrating Collins and Postal’s framework into the analysis, this study provides a finer cross-linguistic characterization of NPIs and highlights the syntactic dependencies that underlie their behavior. These findings refine the theoretical understanding of NPI strength, bridging gaps left by purely semantic accounts, and suggest broader implications for investigating the interaction of Korean NPIs with lexical negators and verb classes.