검색 상세

Constituent Negation and Sentential Negation in English and Korean : Locality and Scope of negation

한국어와 영어의 부정문 연구(구문 부정과 문장부정): 부정어의 국소성과 영향권

초록/요약

내가 본 논문에서 주로 다루고자 하는 바는 한국어와 영어의 부정어인 ‘안’과 ‘not’을 중심으로 문장부정(Sentential Negation)과 구문 부정(Constituent Negation)을 비교하여 각각을 구분하는 기준이 무엇인지 살펴보고 궁극적으로 부정어의 영향권을 바탕으로 한 부정어의 종류 판별법이 바람직하다는 것을 증명할 것이다. 한편, 부정구문이 비문법적인 이유를 domain안에서의 vP구조의 존재유무 및 접근성으로 보아 locality condition으로 증명하였다. 가령, ‘안 늦게’는 문법적이지만 ‘안 빨리’는 비문법적인 이유를 vP 구조의 존재유무로 보았다. 또한 한국어의 ‘안 빠르게’는 시제를 취하지 못하지만 ‘안 빠름’ ‘안 빠른’ 등은 시제를 취하는 현상을 관찰하여 영어의 Sujbunctive 와 유사성과 차이점을 비교하였다. 그리고 한국어의 ‘안 빨리’ 같은 부정어와 부사어짝은 비문법적이지만 영어의 대응표현인 ‘not quickly’는 문법적인 근거에 대해서 ‘구문부정’일 경우에는 언어별로 국소정의역(local domain)의 크기가 다르며, ‘문장부정’일 경우에는 핵(head)의 위치가 다르다는 매개변인(parameter)의 차이에서 기인한다는 점을 논증하였다. 끝으로 단문대답(Fragement Answer)을 살펴보아 문장부정으로 접근해야하는 지 구문부정으로 접근해야 하는 지를 살펴보고, Klima (1964)의 문장부정문판별법에 기초하여 개발한 판별법이 한국어에도 비교적 잘 적용된다는 것을 보였다.

more

초록/요약

In this thesis I mainly explore the comparison between negation in English and Korean in forms such as an-nuc-key (‘not late’) or an coh-key, which can correspond and be compared to not late or not good (well), conventionally treated as constituent negation. In constituent negation, the negative ‘an’ adjoins to an adjectival root such as ‘nuc’ first before the complex head, i.e. ‘an-nuc’, adjoins to ‘–key’, an adverbial marker. In sentential negation, Neg projects into NegP as long as NegP can merge with T. Depending on the type of negation, i.e. constituent negation or sentential negation, I decide which analysis is better, the NegP analysis or the adjunction analysis. There are the two criteria by which to decide whether negation in question is constituent negation or sentential negation. The first criterion is based on the type of complement Neg as the head of NegP selects or the type of element or category Neg adjoins to as an adverbial adjoining to a maximal projection or a head. However, the first criterion is problematic, so the second criterion is considered, i.e. the scope of negation. Only when the scope of negation is taken over the matrix predicate is such negation considered sentential; elsewhere it is a case of constituent negation. However, there are similarities as well as differences between negation in English and Korean. They are similar in that negation is sentential only when the widest scope of negation is considered but different as far as the local domain where a vP structure is to be found is considered. The local domain to make the clause including Neg grammatical is narrow in Korean while it is wide in English. Sentential negation is different between Korean and English because there are the two types of negation in Korean, i.e. short-form negation and long-form negation, which are not available in English while the scope of negation is the same, i.e. the widest scope. I argue that in sentential negation the reason for the (un)grammaticality of the [Neg + AdvP] pair is due to the difference between the parameters in the two languages, i.e. head-initial and head-final. In addition, with regard to ha-support in Korean, which corresponds to do-support in English, a NegP structure can well explain and describe the phenomena in both languages. So, the X(P)-adjunction analysis with Neg as an adverbial adjoining to any element or category as well as the NegP analysis with Neg as the head or a specifier of NegP can be applied both to sentential negation and constituent negation in Korean. However, NegP is assumed only if it can merge with T: NegP is parasitic on T. So, only the adjunction analysis can be applied to constituent negation without TP both in English and Korean while the NegP approach can be used in constituent negation with TP as well as sentential negation in Korean. Nevertheless, I invented the universal structure and account to be applied to both constituent negation and sentential negation as well as short-form negation and long-form negation. Also, in order to determine constituent negation and sentential negation or to distinguish between them, I invented a test for linguistic data in Korean mainly with Klima’s (1964) test. One of the most important conclusions I came to is that constituent negation and sentential negation are cross-linguistic phenomena to be applicable to both English and Korean.

more
반출 Meta View 목록