형용사 논항 의미부류 표준화를 위한 기초 연구 - '크다, 작다, 많다, 적다'를 중심으로 -
A Study on the Semantic Classification of Argument Nominals of Adjectives in Korean: Focused on khu-/cak- 'big/small' and manh-/cek- 'many/few'
1
조회
0
다운로드
10
인용
0
참고문헌
키워드
초록
This paper aims at establishing standard usages of Korean adjectival predicates khu-/cak- ‘big/small’ and manh-/cek- ‘many/few’ based on their syntactic and semantic behavior. The paper focuses on the proper/improper combination of subject and the adjectives, in other words, their semantic/syntactic selection of the subject argument. In order to get an objective and generalized result, we collected data from a large corpus built by National Academy of the Korean Language (2002), which includes a variety of written/spoken texts of contemporary Korean. First, we argue that the four adjectives should be analyzed as one-place predicates, and thus we are interested in the semantic classes of their subject argument in particular. We examined the previous semantic descriptions of three main dictionaries of Korean, and reinterpreted them in terms of semantic classes of subject nominals. On the other hand, we classified the noun classes of subject arguments collected from the large corpus, so to illustrate a comparison between the previous descriptions and the actual uses of the adjectives. The data from the corpus show that the adjectives khu-/cak‘big’/‘small’ semantically select a subject noun in terms of ‘size, magnitude, volume’, whereas manh- ‘many’ and cek- ‘few’ select it in terms of ‘quantity, amount.’ The denotation of the subject nominal extends from the concrete domain to the abstract domain, whose semantic classification is given under [Table 2 and 3]. The tables suggest that the adjectives do not have many distinct senses but their polysemous readings derive from the different semantic types of subject nouns. The paper also notes asymmetric behaviors of khu- and cak- on the one hand and manh- ‘many’ and cek- ‘few’ on the other. That is, khu- and cakdo not exactly share the same classes of subject nominals, and further they may select different nominals in its predicative uses and its adnominal uses. We noted that this last asymmetry between the adnominal and the predicative uses can be accounted in terms of degree of frozenness
Abstract
This paper aims at establishing standard usages of Korean adjectival predicates khu-/cak- ‘big/small’ and manh-/cek- ‘many/few’ based on their syntactic and semantic behavior. The paper focuses on the proper/improper combination of subject and the adjectives, in other words, their semantic/syntactic selection of the subject argument. In order to get an objective and generalized result, we collected data from a large corpus built by National Academy of the Korean Language (2002), which includes a variety of written/spoken texts of contemporary Korean. First, we argue that the four adjectives should be analyzed as one-place predicates, and thus we are interested in the semantic classes of their subject argument in particular. We examined the previous semantic descriptions of three main dictionaries of Korean, and reinterpreted them in terms of semantic classes of subject nominals. On the other hand, we classified the noun classes of subject arguments collected from the large corpus, so to illustrate a comparison between the previous descriptions and the actual uses of the adjectives. The data from the corpus show that the adjectives khu-/cak‘big’/‘small’ semantically select a subject noun in terms of ‘size, magnitude, volume’, whereas manh- ‘many’ and cek- ‘few’ select it in terms of ‘quantity, amount.’ The denotation of the subject nominal extends from the concrete domain to the abstract domain, whose semantic classification is given under [Table 2 and 3]. The tables suggest that the adjectives do not have many distinct senses but their polysemous readings derive from the different semantic types of subject nouns. The paper also notes asymmetric behaviors of khu- and cak- on the one hand and manh- ‘many’ and cek- ‘few’ on the other. That is, khu- and cakdo not exactly share the same classes of subject nominals, and further they may select different nominals in its predicative uses and its adnominal uses. We noted that this last asymmetry between the adnominal and the predicative uses can be accounted in terms of degree of frozenness
참고문헌 (21)
- "주격 중출문과 형용사의 내부 주어", 제24권(제4호), pp.531~556 (1999)
- "한국어 형태소 및 어휘 사용 빈도의 분석1", 4 (2000), 서울: 고려대민족문화연구원
- "21세기 세종계획 전자사전 개발분과:2002년 연구 결과물" (2002)
- "처소논항 교체의 의미론:자동사와 형용사를 중심으로", 제38권(제1호), pp.295~318 (2002)
- "국어의 보충어와 부가어 판별 기준", (제34호), pp.75~111 (2002)
- "국어 형용사 연구:연세대학교 언어정보개발연구원", 3 (1998), 서울:한국문화사
- "The Description of Adjectives for Natural Language Processing: Theoretical and Applied Perspectives:In TALN '99:Workshop on Description des adjectifs pour les traitements informatiques" (1999), Cargèse, Corse, France
- "The adjective vieux: The point of view of Generative Lexicon:Breadth and Depth of Semantic Lexicons", pp.148~166 (1999), Dordrecht[DOI]
- "Projecting the Adjective:The Syntax and Semantics of Gradability and Comparison" (1999), New York: Garland Pubishing, Inc
- "The Science of Words" (1996), New York: Scientific America Library
- "방송 언어 실태 조사 보고서" (2000)
- "비디오 자막 오용 사례 보고서" (2000)
- "신문 언어 실태 조사 보고서" (2000)
- "어문 규범 실태 조사 보고서" (2000)
- "방송 언어 오용 사례 보고서" (2001)
- "어문 규범 준수 실태 조사 보고서:아동 도서" (2001)
- "어문 규범 실태 조사 보고서:홈페이지" (2002)
- "21세기 세종계획 균형말뭉치, 연구 교육용 현대 국어 균형 말뭉치" (2002), (재)한국어세계화재단 21세기 세종계획 결과물 보급 관리 센터 제작 배포
- "표준국어대사전" (2001), 두산동아
- "연세한국어사전" (1998), 두산동아
- "「뉘앙스풀이를 겸한 우리말사전」" (1994), 아카데미하우스
