This paper examines the (a)symmetries in the realization of focused subjects and objects in Korean. Through rating experiments, we demonstrate that native speakers' judgments of acceptability of sentences containing case-marked or case-ellipsed subjects and objects are sensitive to the contrastiveness strength and the discourse accessibility of focused arguments. However, our experiments also show that focused subjects exhibit stronger preference for explicit case marking over case ellipsis and that contrastiveness strength and discourse accessibility have weaker influence on the case marking and ellipsis of focused subjects compared to focused objects. We propose an account of variable case marking that is capable of subsuming both the similarities and differences between focused subjects and objects under the universal theory of markedness. In particular, it is shown that the similarities between focused subjects and objects are predicted by the proposed account based on the contrastiveness strength and the discourse accessibility of focused arguments and that the differences between focused subjects and objects follow naturally from the relative markedness of focus as subjects.

Every language employs its own morphological and syntactic ways of expressing gradable concepts and making comparison between properties of two objects. Korean uses the adverb te 'more' and the post-position pota 'than' to express such relations objects, but displays quite different grammatical properties from a language like English. This paper shows how a constraint-based grammar, HPSG, can provide a robust basis for the grammatical analysis of Korean comparative constructions.

It has been observed that not all negative polarity items (NPIs) are licensed in the same negative contexts, and different NPIs may be licensed by different negative expressions. This shows that Ladusaw's (1979) downward entailment is not precise enough to account for the distributional patterns of NPIs (van der Wouden, 1997; van der Wouden and Zwarts, 1993; Zwarts, 1986, among others). One well-known attempt to deal with this issue is to divide negative expressions into several subtypes. Using boolean semantics, Zwarts (1986; 1998) distinguishes three kinds of downward entailing licensors and accounts for heterogeneous NPI-licensing conditions by means of the semantic strength of negative expressions. This approach has been adopted to define Korean negation (Nam, 1994; Chung, 1993; Chung, 1997; Hwang, 2009). In this paper, however, we argue that the boolean semantic approach for negation is not adequate in characterizing the properties of Korean negative expressions and explaining the contexts of licensing Korean NPIs.

The type of a given sentence indicates the speaker's attitude towards the listener and is usually determined by its final endings and punctuation marks. However, some 6na1 endings are used in several types of sentences, which means that we cannot identify the sentential type by considering only the final endings and punctuation marks. In this paper, we propose methods of finding some other linguistic clues for indentifying the sentential type with a morphological analysis. We also propose to use these methods to implement a system that automatically classifies sentences in Korean according to their sentential types.