Subjects and Predication in Korean and Japanese

The assumption that nominative case is assigned by INFL has proved fruitful in explaining the distribution of nominative case in a number of languages , and in particular in accounting for the differences between the nominative and other cases. Korean and Japanese, however, provide evidence that this method of case-assignment to the subject is not universal. In this paper we argue that nominative case-assignment in Korean and Japanese is independent of INFL, and claim that –ka/-ga marks the syntactic subject of a predication structure independent of the argument structure of the clause. Among the advantages of this unified treatment of –ka/-ga marking is that it leads to an account of the multiple nominative construction found in both languages and to a principled explanation of the impossibility of this construction in the European languages. Further, we argue that the ability to 'license non-theta-marked lexical NPs by predication alone is not unique to Korean and Japanese: Non-argument non-expletive predication subjects can be found in English as well