A Syntactic Account of Category-Changing Phrasal Morphology: Nominalizations in English and Korean*

The focus of this paper is on the proper analysis of lexical and phrasal nominalizations in English and Korean that are built around an identical deverbal element. In English, as is well-known, there are both lexical and phrasal nominalizations that are built around the -ing suffixed form of the verb. In Korean too, there are both lexical and phrasal nominalizations based on the same nominalizing affix, such as -um (or -ki). The theoretical interest in this class of nominalizations is that while both lexical and phrasal nominalizations contain the same deverbal element, they contrast in their syntactic properties. A deverbal lexical nominal in either language behaves as a noun and the syntactic structure projected from such a nominal is fully nominal in its properties. However, a deverbal phrasal nominalization based on the same element shows a mixture of verbal and nominal properties. For instance, the Nominal Gerund Phrase (NGP, also called POSS-ing gerund) in English shows verbal properties up to VP, "switching" to nominal properties above the VP. A similar distribution of verbal and nominal properties is found in one type of phrasal nominalization in Korean (Type III nominalization). Despite the similarities, the two languages also differ in interesting ways. That is, while the deverbal element in NGPs in English behaves morphologically as a verb, the corresponding element in Korean phrasal nominalizations shows nominal morphological behavior. In the first part of this paper, I will propose an analysis that accounts for the similarities and differences between the two languages regarding the morphosyntax of deverbal nominalizations. The leading idea of the analysis is that in both languages, different types of nominalizations arise from an affix being able to combine either with a word or a phrase. Independent differences between the morphology of two languages then explains the observed similarities and differences. In the second part of the paper, the proposal sketched in the first part will be worked out in a framework that is largely minimalist in spirit in which distinct principles for syntactic and morphological structure licensing are employed. Finally, I discuss the difficulties that these nominalizations pose for proponents of lexicalism.