중국어 연동문은 그 성분 간의 의미의 복잡성으로 한국어 번역과 교수에 많은 어려움을 준다. 이 논문은 연동문이 한국어 번역에서 나타나는 여러 가지 오역현상을 지적하고 이런 문제 의식에서 출발하여 선행연구에서의 구조주의 문법이론의 연구방 법에서 탈피하여 기능문법에서 타동성의 정도성 이론에 근거하여 연동문을 타동성의 정도성에 따라 기본구조 연동문과 선행동사가 ‘着’를 가진 연동문, 그리고 이동동사 연동문으로 나누었다. 그리고 기본구조 연동문은 선행동사가 빈어(宾语)를 가지는 연동문과 후행동사가 빈어를 가지는 연동문, 선후행동사 모두 빈어를 가지는 연동문으로 하위분류를 하여 살폈고 이동동사 연동문은 선행동사가 이동동사인 경우와 후행 동사가 이동동사인 경우를 나누어 살폈다. 이 분류를 바탕으로 연동문의 한국어 번역양상 분석에서는 각 연동문의 특성과 연동문에서 구성성분의 의미관계에 대하여 처음으로 개념구조와 의미지도를 그려 보이고 연동문의 한국어 번역전략을 제시함으로써 연동문의 한국어 기계번역과 인공번역연구 및 번역교수에 유익한 자료를 제공하였다.

The purpose of this study is to investigate serial verb constructions (hereafter, SVCs), and provide how two verbs combine to represent a single event. Examining the argument structures of verbs, we propose that SVCs fall into two types: one is total argument sharing, and the other is partial argument sharing. These different types of sharing should have a distinct merger considering the argument structures. When two verbs in an SVC share their all arguments (internal and external arguments), namely total sharing, the serialization of two verbs takes place at the level of (categorial) head (head-head merger). In the case of partial sharing, the merger occurs at a phrase level (phrase-phrase merger): that is, after each verb meets their internal argument requirement. Previously asserted arguments about the relations between two verbs (namely, adjunction and complementation) are reexamined, and this study will employ the labeling algorithm to resolve their shortcomings.

Crosslinguistically, basic motion verbs similar to the English go and come have been shown to be highly susceptible to grammaticalization. However, there has been relatively little emphasis in previous research on the grammaticalization of the English motion verb come. This paper attempts to redress some of the balance by investigating the grammaticalization of come, with particular emphasis on its occurrence in Serial Verb Constructions. The major aim is to provide a detailed account of how the semantic change undergone by come in such contexts is effected by a number of diverse mechanisms, including metaphor, metonymy and subjectification. In addition, certain functional properties of the grammaticalized come will be discussed, along with certain morphosyntactic changes that occur by way of reanalysis. The result of this analysis should demonstrate that come in its serialized form has undergone extensive grammaticalization, shifting from a fully lexical verb to a functional item in its semantic, categorial and morphosyntactic aspects.

There have been many linguists who studied Korean two-full-verb constructions. Some of them found an apparently meaningful similarity between these constructions in Korean and other serializing languages with respect to tense sharing and argument sharing and proposed serial verb constructions (SVC) in Korean. However, there are several phenomena which cannot be accounted for within the SVC framework, and one of the important properties of the SVC―the unaccusative second verb which takes the object of V1 as its subject―does not exist in Korean. In this paper we analyze the so-called SVCs into compound verb constructions (CVC) and covert coordination constructions (CCC) which already exist in Korean syntax. We also propose adequate structures for them without introducing a new structure to an already existing system. Based on this structural analysis, we not only explain the problems that the previous analysis could not solve but also provide further evidence for our non-serializing approach.