국어 명사화 어미의 분포에 대한 계량적 연구 = A Quantitative Approach to Korean Nominalizations

다국어 초록 (Multilingual Abstract)

Much has been said on several types of Korean Nominalization Constructions for the past decade, with emphasis on nominalizers` co-occurrence restrictions with matrix verbs and the semantic properties of their constructions. Some of the previous proposals however have been regarded as somewhat problematic as far as the acceptability of the data they deal with is first concerned. In this paper, under the assumption that the distribution of nominalizers must be first thoroughly investigated to find the morpho-syntactosemantic properties of Korean nominalizers, we have attempted to re-examine the data frequently discussed in the linguistic literature. In doing so, we used the corpus called Sejong Balanced Corpus built as part of 21st Sejong Project. The results of our corpus-based research implicate that some previous proposals concerning nominalizations must be reconsidered due to the incorrect observations of the data or the complete lack of the data. Since the purpose of this paper is not to provide a theoretical analysis of the known data, we left the actual analysis open for future. However, we showed our results can be used as some guidelines for providing formal analyses of Nominalization Constructions.