Quantitative Variation in Korean Case Ellipsis: Implications for Case Theory

Abstract

It is well-known that hierarchies of person, animacy and definiteness have effects on case marking systems in various languages, where certain classes of subjects and objects are marked, but not others. This paper presents evidence of frequency effects of those hierarchies on case ellipsis in Korean. The two major aims of this paper are the following. First of all, the significance of variable case ellipsis patterns of Korean, as found in the CallFriend Korean corpus (LDC 1996), will be demonstrated when looked at from a functional-typological perspective: variation in case marking between style levels within a single language reflects variation across languages. In a second step, the findings from a comparative study of Korean and other languages are integrated into a coherent theoretical framework – Stochastic Optimality Theory (OT) (Boersma 1998; Boersma and Hayes 2001). It is shown that quantitative patterning found in Korean case ellipsis can be analyzed within the stochastic OT framework in a way analogous to an account of categorical differential case marking effects proposed by Aissen (2003). In this analysis, categorical differential case marking found in various languages is viewed as conventionalization of the same universal pragmatic tendency to mark disharmonic elements, which is also present in the variable case-marking systems of languages like Japanese and Korean.