복합문법의 차원을 통한 한국어 교재 분석 연구

김의수(Kim, Uisu),허은혜(Eunhye Heo) 언어과학회 2017 언어과학연구 Vol.0 No.81

This study aims to introduce the dimensions of complex grammar used in Korean language textbooks. A complex grammar consists of at least two items; one of these needs to be a grammatical form while the other can be either a lexical or a grammatical form. The number of forms in complex grammar determines its degree of dimension. Analyzing Korean language textbooks from levels 1 to 6 show that level 5 and 6 textbooks have three kinds of dimensions, while levels 1 to 4 show four, five, or six kinds of dimensions. The highest degree of dimension found is nine in level 4. This means that Korean grammar is mainly taught in textbook levels 1 to 4. The most frequent degree of dimension found is three, two, four in order of frequency. If we continue to analyze more Korean language textbooks in this way, we will be better able to understand the complexity of grammar in overall Korean education.

교육과정에 따른 문장의 다양성과 복잡성 추이

김의수(Kim Uisu),이로사(Lee Rosa) 한국언어문학회 2009 한국언어문학 Vol.69 No.-

This paper aims to investigate aspects of diversity and complexity of sentences in Middle School Korean Textbooks published from the period of the 1st Curriculum to that of the 7th Curriculum. There are relatively small-sized texts(henceforth, ‘guiding text’) in the beginning of every chapter, which introduce the main purpose of those chapters. They are normally written by writers or editors of the textbooks. Thus, we can induce the overall degree of diversity and complexity of sentences in those chapters through analysing the guiding texts. Analytic grammar, which I invented previously to make syntactically tagged corpus in Korean, is used to parse the guiding texts. I illustrate the detail procedure of making data which reveal the development of degree of diversity and complexity of sentences in every chapter of the textbooks. From these data, I extract crucial information about synchronic and diachronic trends of diversity and complexity of sentences in the text books.

의미역 위계의 대조언어학적 연구

김의수(Kim Uisu) 한국언어문학회 2007 한국언어문학 Vol.62 No.-

  This paper has two aims; the first is to extract their own Thematic hierarchies from Japanese and Chinese with the methodology already applied to Korean. Generally speaking, Lexical Cases are closely related to Theta roles, unlike Structural Cases. So if you substitute Lexical Case markers with Structural Case markers, a certain word order restriction due to Thematic hierarchy, suddenly appears. Two different Thematic hierarchies resulting from this methodology are as follows;<BR>  〈Japanese Thematic hierarchy〉  Agent〉Experiencer/Location/Source/Recipient〉Theme〉Goal<BR>  〈Chinese Thematic hierarchy〉 Agent〉Experiencer/Goal/Recipient/Location/Source〉Theme<BR>  In terms of word order, Japanese is similar to Korean. Also interestingly, their Thematic hierarchies are very similar. This is also true between Chinese and English.<BR>  Until now, only one Thematic hierarchy is believed to exist among all languages. However, the conclusion of this paper challenges this traditional idea; each language has their own Thematic hierarchies.<BR>  The second aim of this paper is that Thematic hierarchy controls the so-called Subject-honorifics in Japanese. Previous studies on Korean in this subject matter have found this to be also true. This approach is also interesting in that most linguists believed that Spec-Head agreement must control Subject-honorific until recently. However, my research shows that the highest argument in terms of Thematic hierarchy should be the antecedent of the honorific A"-anaphora in Japanese as well as in Korean.

TOPIK 읽기 영역 지문의 난이도와 균질성에 관한 통사론적 접근

김의수(Kim Uisu),정은주(Jung Eunjoo) 한국언어문학회 2009 한국언어문학 Vol.71 No.-

This paper aims to investigate whether the reading texts in TOPIK have been correctly selected according to the different levels and whether the level of reading texts have sustained coherence within the same test levels. For this purpose, the 13th to 15th TOPIKs' reading texts were taken and analysed into syntactically-tagged corpus from which syntactic complexity of matrix clauses and embedded clauses were extracted. The syntactically-tagged corpus were classified into 9 different texts, according to the different levels and the times the tests were taken. After comparing those texts from various points of view, it was concluded that intermediate level texts were more difficult than those of advanced level. In addition, only the advanced level texts were coherent from 13th to 15th TOPIK. Following these results, it is suggested that the intermediate reading texts be made easier and the advanced texts made more difficult by modifying embedded clauses in the reading texts. Accumulation of text corpus can help towards solving these types of problems in selecting texts for TOPIK. (Hankuk University of Foreign Studies)

문장의 구조와 해석문법

김의수(Kim Uisu) 한국언어문학회 2008 한국언어문학 Vol.67 No.-

Grammatical theories about sentence structures can be divided into two types: one is for parsing sentence, the other for producing sentence. Generative grammar has tried to deal with them both through Universal Grammar. Many theories about natural language processing are also interested in production as well as parsing. However, if you want to make a good model for producing sentences, you need to know beforehand how sentences are organized. Unfortunately, generative grammar makes assumptions on a priori grounds which have not been proved correct yet. Thus, I think that we have no choice but to find a new way to understand sentence structures. This led me to a model called ‘analytic grammar’ which only concerns itself with parsing sentence. In this paper, I propose a system that can immediately be applied to parsing sentences in Korean. It consists of a few letters and numbers which represent characteristics of sentence constituents such as their grammatical functions, parts of speech, and sorts of modifiers. I analyze real sentences from a poem and a novel. The analysis shows that analytic grammar works well without having to make many assumptions and that the results can be easily understood. This model also shows potential for processing huge corpus.

단어 형성의 한 축으로서의 단일어화

김의수 ( Uisu Kim ),김건영 ( Gunyoung Kim ) 한국외국어대학교 언어연구소 2017 언어와 언어학 Vol.0 No.76

The aim of this paper is to argue that it is not only mulitiverbation but also univerbation that functions as an important axis for word formation. Univerbation in this paper covers a simplex which belongs to simple words but within which does not have any morphological structure. The extended concept of univerbation is a subtype of lexicalization, and shares some partial phenomena with grammaticalization. Word formation can be divided into univerbation and multiverbation. Both have three processes as their subtypes: extension, reduction and switch. The process of word formation can be totally systemized by setting up univerbation and multiverbation as the two axes of word formation.

외국인 유학생 유치·관리 역량 인증제의 성과 분석에 대한 연구: 인증대학과 미인증대학의 차이를 중심으로

김의수 ( Kim Uisu ),이정현 ( Lee Junghyun ) 충남대학교 교육연구소 2017 교육연구논총 Vol.38 No.1

본 연구는 외국인 유학생 유치·관리 역량 인증제의 성과를 분석하기 위해 인증대학들과 미인증대학들을 나누어 그 차이에 따라 외국인 유학생 수와 유학생의 한국어 능력 수준에 미친 영향이 어떠한가를 살펴보고자 한다. 또한 인증제의 시행 결과가 국내 대학들의 교육 역량, 국제화 역량 그리고 학생 충원율 부문에 어떠한 영향을 미쳤는가를 실증적으로 살펴보는데 그 목적이 있다. 이를 위해 EDSS 고등교육통계 및 대학정보공시 자료, 대학 알리미 자료 등을 사용하여 국내 185개교 4년제 대학을 분석하였다. 연구자는 SPSS 20.0을 이용하여 다중회귀 분석을 실시하였으며, 주요 연구 결과는 다음과 같이 요약된다. 대학의 인증 여부는 유학생 유치와 유의한 관계가 있는 것으로 나타났는데, 인증대학의 경우 미인증대학에 비해 외국인 유학생 수에 부정적인 영향을 미치는 것으로 나타났다. 각 학교의 유학생 중 TOPIK 합격생의 비율은 대학의 인증 여부와는 유의미한 관계가 있지 않았으나 최근 들어 인증대학에 한국어 실력이 우수한 유학생 수가 더 많아졌음을 알 수 있었다. 본 연구 결과, 대학의 인증 여부가 유학생 수에 중요한 영향을 주는 요인임을 규명했으며, 최근 인증된 대학들에 한국어 실력이 우수한 유학생 수가 증가하였음을 알 수 있었다. 본 연구 결과는 고등교육의 국제화 관련 연구로서 유의미한 기초자료로 활용될 수 있을 것이다. This study investigated the effects of the International Education Quality Assurance System (IEQAS) on international student recruitment at 4-year universities across the country. It was analyzed whether certification for IEQAS affected recruitment and management of international students among 185 4-year universities in Korea based on various sources. Using SPSS 20.0, the researcher conducted reliability analyses, correlation analyses, and multiple regression analyses, which led to the following results. First, it was revealed that certification had a significant relationship with the recruitment of foreign students in such a way that certified universities were affected negatively in terms of the number of foreign students compared to uncertified ones. The proportion of those who passed the TOPIK among foreign students did not have a significant relationship with certification, though findings showed that the number of foreign students fluent in Korean on campuses of certified universities has increased recently. Research findings suggest that certification is a significant factor affecting the number of foreign students, and those certified recently recruited more foreign students with Korean language proficiency. It is expected that the findings and results of this study will be utilized as meaningful material for research on the globalization of higher education given the sparse literature on the IEQAS..

외국어로서의 한국어 학습 교재의 문장 분석

김의수(Uisu Kim),김태성(Taeseong Kim) 한국중원언어학회 2009 언어학연구 Vol.0 No.14

This paper investigates aspects of diversity and complexity of sentences in low to middle level Korean language text books for non-native speakers. We hypothesize that the higher level text books will have more sophisticated sentences than lower level ones. However, we do not know if this is really the case. So, I have analysed sentences from four text books, levels 1 to 4 published by Yonsei University Press, from the syntactic point of view, namely the diversity and complexity of sentences. Looking at the degrees of difficulty of sentences, I point out the following: 1) evidence contrary to our hypothesis; 2) differences have occurred; 3) how issues that arise from the findings can be resolved. Analytic grammar, which I had previously proposed to make syntactically tagged corpus in Korean, is used to parse sentences in the text books. In addition, I detail the procedure of making data which reveal the development of the degree of diversity and complexity of sentences in every chapter of the text books.

한국어교육을 위한 언어범주의 재정립: 어법범주를 중심으로

김의수 ( Kim Uisu ),허은혜 ( Heo Eunhye ) 한국외국어대학교 외국어교육연구소 2018 외국어교육연구 Vol.32 No.2

이 연구의 목적은 한국어교육이 국어교육과 달리 필수성과 용이성이라는 언어 외적 직관에 의해 어법범주라는 새로운 언어범주를 추가적으로 가지고 있음을 구체적인 한국어교재 분석을 통해 실증적으로 논의하는 데 있다. 국어교육에서 어휘범주와 문법범주의 이분체계를 전제한다면, 한국어교육에서는 어휘범주와 어법범주, 문법범주의 삼분체계를 설정해야 한다는 것이다. 어법범주는 의사소통에서의 필수성에 근거한 어법범주 유형Ⅰ과 학습에서의 용이성에 바탕을 둔 어법범주 유형Ⅱ로 나뉜다. 본고에서는 이들 두 가지 유형이 3종의 한국어교재에서 구체적으로 어떻게 나타나 있는가를 고찰하였다. 그 결과, 어법범주 유형Ⅰ은 초급 교재에서 체언의 8가지, 용언의 2가지, 수식언의 4가지로 실현되어 있었다. 어법범주 유형Ⅱ는 1급 교재에서는 2차원부터 5차원까지의 복합형태로 나타나며 총 15개 타입과 51개 토큰으로 실현되고 있음을 확인하였다. This study aims to introduce phraseological category in addition to lexical and grammatical category in KFL. Phraseological category has two subtypes: TypeⅠ is based on necessity in communication while TypeⅡ on easiness of learning. We investigate three kinds of KFL textbooks in order to verify the existence of these two subtypes. It is found that TypeⅠ is realized as nominals, verbs and modifiers while TypeⅡ has complex forms varying from two to five dimensions.

SVO 언어권 한국어 학습자들의 작문에 나타난 문장의 통사적 복잡성과 다양성 연구

김의수(Kim Uisu),채문숙(Chae Moonsook) 한국언어문학회 2010 한국언어문학 Vol.75 No.-

We gathered and parsed well-formed clauses in writings from Chinese, English and Spanish learners of Korean in order to investigate syntactic complexity and diversity revealed in their clauses using Analytic Grammar. It was concluded that the Korean language learners from the three SVO type languages showed similarity of syntactic aspects of their clauses. First of all, syntactic complexity of matrix clauses was found to be constant from beginners to advanced learners. That of embedded clauses, however, differed according to learners' proficiency; that is, higher level learners showed higher syntactic complexity than lower levels. Syntactic complexity of sentences depended not on matrix clauses but on embedded clauses. Secondly, the Korean language learners from the three SVO type languages also showed similarity of syntactic diversity. The matrix clauses played the main role in sentences written by beginners but the embedded clauses caught up with the matrix clauses in those of the advanced learners. It is also interesting that these conclusions support the results from our previous study on intermediate learners from China for academic purpose.