Earticle

Based on two cognitive categories, activation and identifiability, Lambrecht (1994) distinguishes four different types of discourse referents, i.e. active, accessible, inactive, unidentifiable, which represent different mental representation of those referents in the minds of the speaker and addressee, and he suggests the topic acceptability scale, which states that there is a general correlation between the activation and identifiability states of topic referents and the pragmatic acceptability of sentences. According to the scale, the most easily acceptable sentences are those whose topics are active since the mental cost to process active topics is low, and the degree of acceptability decreases in the following order: accessible, inactive and unidentifiable topics. In fact, unidentifiable referents cannot be topics since it is impossible because the addressee has to interpret the proposition of a sentence which is about something he cannot even identify. The main purpose of this paper is to show how topics in different activation/identifiability states are expressed in Korean, and it is shown that topic referents belonging to different activation/ identifiability states adopt different strategies to be expressed as topics. Active topics, which are assumed to be the most easily accepted, are generally expressed as zero pronouns and bare NPs, which requires the minimum amount of mental cost to process them. Accessible referents, which are frequently used as topics in Korean, show that they need the marker -nun to be expressed as topic. Finally, Korean uses the marker -nun or the maliya construction to encode inactive topics.

The aim of this study is to illustrate the relationships among topic-interest, topic-usefulness, and motivational orientation. A total of 138 students in six different English language courses in a university in Korea took part in the study. Two sets of identical questionnaires were used before and after the semester. They asked the learners to rate the degrees of motivational orientation and the levels of interest and usefulness on the topics in the textbook. The results show that there was an interesting pattern that the level of correlation between topic-interest and integrative motivation was higher than that between topic-interest and instrumental motivation. In case of the correlations between topic-interest and topic-usefulness, moderate levels of positive correlations are shown. However, it was difficult to find a strong relationship in the levels of topic-interest between before and after the use of the textbooks. The average correlation is calculated as .27, indicating a very low level of relationship. The overall results suggest that teachers need to identify their learners" reasons for studying English to choose a textbook that meets the learners" motivational orientation. Teachers also need to evaluate how textbook writers maximize their creativeness in presenting topics in textbooks.

The purpose of this paper is to find out whether the sentence topics of the English originals are translated into sentence topics in their Korean translations and explain what causes changes, if any, in the translations. Topics are identified in relation to sentence types: topic-comment, identification, event-reporting, presentational, and topicalized sentences. A total of 480 clauses from The Grapes of Wrath, and Great Gatsby and their corresponding clauses from the Korean translations are analysed. Ninety percent of the topics of the topic-comment sentences and eighty-six percent of those of thetic (event-reporting and presentational) sentences in the originals are found to be translated into sentence topics in the translations. Two main factors causing changes in the translations are the presence of inanimate subject sentences in the originals and thematization in translations. Thematization is seen as a result of pragmatic structuring of the author's communicative intention in accordance with the syntactic structure of the target language. Translations of inanimate subject sentences can also be understood in relation to the acceptability of the information states of the inanimate subjects, but more intensive study is required to confirm their relationship.

본 연구는 일본어 담화 내에 존재하는 주제를 대상으로 하여 담화 안에서 주제가 적절 하게 제시되는 방법을 고찰하는 것이다. 주제란 담화 내에서 ‘무엇을 이야기 할 것인가’ 에 대한 화제를 제시한 것이다. 전형적인 주제 제시 방법은 담화의 첫 부분에서 제시되고 그 이후로 생략된 상태로 대화가 이어진다. 그러나 이러한 주제 제시 방법은 절대적이라 고 할 수 없다. 본 고에서는 일본어 모어화자의 담화자료를 바탕으로 하나의 담화 안에서 한가지의 주제가 지속적으로 명시되는 경우에 주목하였다. 이때 주제가 가지는 기능을 세분화하고 각 기능과 주제의 명시 현상과의 관련성을 중심으로 주제 제시 방법을 고찰 하였다. 담화문의 주제를 분석할 때 ①주제명시횟수 ②발화자 ③언어형식 ④주제기능 ⑤주제 명시 필요성의 항목을 토대로 분석하였다. 본 고의 조사에 따르면 하나의 담화 안에서 지속적으로 주제가 명시되는 경우가 적지 않은 것으로 나타났다. 명시된 주제는 ‘담화 내 에 화제로서 제시한다’는 ‘주제 개시’ 기능 이외에도 ‘주제의 환기’ ‘주제를 강조’ ‘주제를 유지’하는 기능들이 작용하는 것으로 분석되었다. 이 중에서도 ‘주제 개시’ 및 ‘주제 환 기’ 기능을 가진 주제는 반드시 명시되어야 하는 것으로 판단되었다. 이러한 기능들은 주 제의 전형적인 ‘제시’ 기능과 유사한 것으로 보이며 이러한 주제는 명시되어야 회화가 성립된다고 분석된다. ‘주제 강조’ 및 ‘주제 유지’ 기능이 작용하는 주제는 화자의 주관적 인 판단이 개입된 것으로 보이며 주제를 명시하지 않아도 회화 성립에 문제가 없는 것으 로 판단되었다.

The present study targets topics that exist in Japanese discourse and examines how topics are appropriately presented in the discourse. A topic suggests ‘what to talk about’ within a discourse. A typical method of presenting a topic is described in the first part of the discourse, and the conversation continues with omissions thereafter. However, this method of presenting a topic cannot be said to be absolute. In this paper, based on the discourse data of native Japanese speakers, we focused on cases where one topic is continuously specified within a single discourse. To this end, the functions of the topic were subdivided and the method of presenting the topic was considered, focusing on the relationship between each function and the topic's manifestation phenomenon. The findings of the paper shoed that there are many cases in which topics are continuously specified within a single discourse. In addition to the ‘topic initiation’ function of ‘presenting the topic as a topic within the discourse,’ the topic was analyzed for the functions of ‘evoking the topic,’ ‘emphasizing the topic,’ and ‘maintaining the topic.’ Among these, it was judged that the topics with the functions of ‘topic initiation’ and ‘topic evocation’ must be specified. These functions appear to be similar to the typical ‘presentation’ function of the subject, and it was concluded that such a subject must be specified for a painting to be established. The topic on which the ‘topic emphasis’ and ‘topic maintenance’ functions work seems to have involved the speaker’s subjective judgment, and it was judged that there was no problem in establishing a conversation even if the topic has not been specified.

본 연구는 미카미(1960)의 『象は鼻が長い』라는 문법서에 근거해 일본어와 한국어의 ‘무제 화문(無題化文)’의 차이에 초점을 맞추고 있다. 미카미는 일본어에는 ‘주어’가 없다고 하는데, ‘象の鼻が長いこと’라는 표현에서 ‘象の’를 ‘주제’로 하면 ‘象ハ鼻が長い。’가 되고, ‘鼻が’를 ‘주 제’로 하면 ‘象の鼻ハ長い。’가 된다고 한다. 그러나 일본어와 한국어의 ‘주제 구문’인 ‘象は鼻 が長い。’와 ‘코끼리는 코가 길다.’를 ‘무제화문’으로 만들면 각각 ‘象の鼻が長い。’와 ‘코끼리 가 코가 길다.’가 되는 차이점이 보인다. 이에 본고에서는 일본어와 한국어의 ‘무제화문’에는 왜 이러한 차이가 나타나는 것인지에 대해 고찰해 보았다. 우선 ‘～の’와 ‘～의’에 관해서는 일본어의 ‘～の’에는 <소유･소속>의 의미기능 외 이른바 ‘가주어’의 기능도 있는 반면 한국어 의 ‘～의’에는 <어조사>의 기능밖에 없기 때문에 ‘이중주어’가 인정된다. 이러한 점에 의해 한일 ‘주제문’을 ‘무제화’시킨 ‘주문(主文)’에 차이가 나타나는 것이다. 다음으로 가주어의 ‘～の’와 이중주어의 ‘～이/가’ 역시 일본어의 ‘～の’가 ‘가주어’로 기능하고 있는 것에 반해 한 국어에서는 ‘이중주어’가 인정되는 차이가 있다. 이로써 일본어의 ‘～の’는 ‘가주어’로도 기능 하지만 한국어의 ‘～의’는 <소유>나 <소속>의 의미만으로 사용되는 것이 한일의 차이를 일 으키는 원인이라고 할 수 있다. 또 일본어는 ‘주격’, ‘주어’, ‘주제’를 구별하고 있지만 한국어 는 ‘주격’을 인정하지 않는 연구 견해도 있고 ‘주어’와 ‘주제’의 정의가 명확하지 않다. 또한 한국어는 ‘이중주어’나 ‘이중주격’이라는 표현에서 알 수 있듯이 ‘주어’와 ‘주격’을 구별하지 않기 때문에 일본어와의 차이가 생긴다고 볼 수 있다. 향후 과제로는 일본어와 한국어에 있 어서 ‘주격/주어/주제/이중주격/이중주어/가주어/격 성분의 ‘～の’’ 등에 대해 보다 명확한 정 의가 요구됨을 남겨두는 바이다.

This study focuses on “untitled sentences” in both Japanese and Korean. This research is based on “象は鼻が長いZō wa hana ga nagai, in “the elephant has a long nose” in Mikami's (1960) influential grammar book. In it, Mikami (1960) says that there is no Japanese word for the ‘subject’. Additionally in the expression ‘象の’ Zō no, elephant's’ in ‘the elephant's trunk is long’ is the subject of ‘象は鼻が長いZō wa hana ga nagai,. When ‘鼻が Hana ga is used then ‘the nose’ becomes the subject, it becomes ‘象の鼻は長い(Zō no hana wa nagai)’. However, if ‘象は鼻が長いZō wa hana ga nagai, is used and ‘코끼리는 코가 길다 Kokkilineun koga gilda, in the Elephant has a long nose’ and have ‘topic syntax’ elements (in Japanese and Korean) and are untitled, they then become ‘象の鼻が長い(Zō no hana ga nagai)’ and ‘코끼리가 코가 길다(Kokkiliga koga gilda)’. Therefore, in this study, we examined why these differences can be seen in ‘untitled sentences’ between Japanese and Korean. As a result, ‘no’ in Japanese has a ‘provisional subject’ function in addition to the ‘ownership/affiliation’ function, whereas ‘ui’ in Korean has only a particle, that has no substantial meaning and aids another word, so the ‘double subject’ is considered to be recognized. From these points of view, it is speculated that there may be differences in the main sentence that make the topic untitled sentences between Japanese and Korean. In the future, clearer definitions of “nominative /subject /topic /double nominative object /double subject /provisional subject /‘no’ of case” etc. are required in both Japanese and Korean.

The terms ‘theme’ and ‘topic’ in information structure have been used without conceptual distinction. So there was conceptual confusion in many discussions. This paper suggests that it is necessary to analyze how the topic-comment structure introduced by Lambrecht(1994) is related to the theme-rheme structure discussed in Halliday(1967) in Korean sentences. Based on the fact that the most important property of the theme is sentence-initial and that of the topic is aboutness, it is pointed out that the sentence-initial constituent of a sentence is the theme and the topic is an active element about which new information is added to the listener. A preposed object with clitic -(n)un in the sentence-initial is analysed as not a topic but a contrastive focus in Korean. And the sentence-initial that represents the possessor and possession marked with -(n)un is not only a topic as a whole but also its possessor could be a topic.

This paper explores two interface-based values (i.e. givenness and contrastiveness) of information structure, and presents a generalized syntactic structure to license all the discourse notions such as focus, sentence topic, discourse topic, and CT. Topic and focus can be easily identifiable in the sentence since they are phonetically and semantically distinguished from each other. A focused phrase bears a focal stress and is interpreted as discourse-new or exhaustive, while a topic phrase is usually destressed and is interpreted as discourse-given. Traditionally, contrastive topic (CT) has been analyzed as a single discourse phenomenon. In the paper, however, CT is a hybrid discourse notion that is composed of two independent semantic features; namely, givenness and contrastiveness. In particular, a topic phrase (sentence or discourse topic) is G-marked and must be merged to the edge of a strong phase in order to check off the uninterpretable givenness feature for its topic interpretation. Meanwhile, a focus phrase is syntactically unmarked and must remain in its default focus domain (i.e. VP or TP) to be properly interpreted as discourse-new. CT undergoes at least two independent scrambling operations to check off both givenness and contrastiveness features in the edge of a strong phase.

There is a peculiar syntactic difference between focus phrases and relative words with respect to extractability from the TP-ellipsis site. In order to account for it, this paper suggests a novel analysis of TP-ellipsis which is based on three basic assumptions; First, the split CP periphery in the sense of Rizzi (1997, 2001, 2004) is derivationally constructed by means of Merge in a bottom-up fashion. Second, the TP-ellipsis site must undergo topicalization before ellipsis. Third, in the extended spirit of Johnson (2001), the topicalized phrase is adjoined to TopP, and so it becomes an adjunct island in the sense of Ross (1967). Under this theoretical framework, this paper argues that the extractability from the TP-ellipsis site depends on the landing site for the extracted element. The lower head projects before the higher head does in a derivational left periphery, and so extraction of a focus phrase from the TP-ellipsis site may occur before the TP in question moves to TopP. In contrast, extraction from the TP-ellipsis site to the edge of ForceP is impossible since it is done after the TP in question becomes an adjunct island. This paper also shows that its analysis can be extended to explaining embedded sluicing, split questions, left-dislocation, and the survival of topic phrase in sluicing.

The purpose of this study is to investigate how subjects in a Korean text are translated into an English text. The analyzed texts are a South Korean presidential speech on biohealth industry in Korean and its English translation. From a linguistic typological perspective, Korean is thought to be both subject-prominent and topic-prominent, and English is considered subject-prominent. This means that subjects are frequently omitted in Korean as situation-bound expressions are preferred, while most English sentences must have subjects because of their subject-centered sentence constructions. It is suspected that these differences lead to numerous translation shifts between the two languages. In order to make the analysis consistent, the criteria on which to determine the subject in both languages are discussed. Once the criteria are defined, English sentences whose subjects are equivalent or different to those of their original sentences are extracted to understand the details of the translation shifts.

This paper focuses on two questions: what kind of linguistic features the so-called ‘essential information’ possesses, if embedded in the context and whether professional interpreters have group characteristics in distinguishing between essential and secondary information. To answer these questions, an empirical study was conducted in which seven professional interpreters and 77 students were asked to select essential information from two different types of text. Especially carefully examined were focus information (recognizable by Korean focus-markers according to Lee (2000)), predicate and proper names. All in all, the results of the two groups do not differ much from each other, but in details, professional interpreters showed some group-specific features. Interpreters were more inclined to omit deducible information as well as chose discourse markers more often than students. The rate of topic-information was higher for interpreters than it was for students. As for the sentence functional distribution of the essential information, the predicate, object and subject were frequently selected in this order. Based on these results, professional interpreters’ comprehension process – compared to those of non-interpreters – is remodelled.

일본어에서는 처음 만나는 사람과 이야기할 때 일반적으로 정중체가 선택된다. 그러나 일본어 학습자의 담화를 관찰해보면 처음 만나는 두 사람 간의 대화에서 보통체가 사용되는 경우가 있다. 본 논문에서는 각각 한국어･영어･중국어를 모국어로 하는 ​​학습자 90명의 담화자료를 대상으로 정중체를 기조로 하는 첫대면 양자 간 대화에 나타나는 보통체의 사용 조건을 분석했다. 먼저 보통체의 사용 실태를 분석한 결과, 중국인 학습자의 보통체에 종조사가 부가된 문장이 많은 점, 종조사 ‘요’의 사용도 많으며 이는 특히 동사술어문에 많은 점 등이 밝혀졌다. 이러한 현상에 대해 (a)중국어 어기조사의 용법의 활용, (b)담화구성능력의 습득이라는 두 가지 관점에 대하여 고찰하였다. 즉, 중국어를 모어로 하는 학습자가 사용하는 간투조사 및 종조사의 ‘나’, ‘네’, ‘요’ 등은 중국어의 어기조사의 용법을 활용한 것이며, 친근감을 표시하거나 어조를 약화시키는 역할을 하는 것으로 커뮤니케이션상의 전략적인 용법으로 판단된다. 또한 중국인 학습자의 담화에서 보통체는 화제의 지속을 나타내고 정중체는 화제의 종결을 나타내는 담화표지로 기능한다는 것을 알 수 있다.

In first encounter situations, Japanese speakers usually choose the polite form as the basis for their conversation. In the case of Japanese learners, however, it is often observed the use of plain form in these situations. This research analyzes the use of plain form in speech of 90 learners whose mother tongue are Korean, English and Chinese produced in first encounter situations. All the speeches were collected from the KY corpus. In this article, we will show some of the results concerning the Chinese learners. Two main points were considered in the analysis of plain form in the learners’ speech: (a) interference of Chinese modal particles; (b) acquisition of speech organization competence. As for (a), we must mention that the so called “modal particles” in Chinese have similar functions to the sentence-final particles in Japanese, both playing an important role in face-to-face interaction. In Chinese language, there is no clear distinction between sentence-final particles and interjection particles. This may partially explain the use of ‘yo’ with functions other than this particle’s original constraints observed in Japanese speakers’ speech. Also, the subjects used ‘na’, ‘ne’ and ‘yo’ as a strategy to display affinity and to soften the tone of the speech. (b) It was observed a progression from listing of simple sentences toward production of complex sentences. While the polite form was used as a marker of speech organization, plain form was used as a marker of topic organization. The Chinese subjects tended to use the plain form while talking about the same conversation topic and the polite form to bring the conversation of this topic to an end.

이 글은 일상 대화에서 쓰인 담화 표지 ‘있잖아’의 기능을 살펴보는 것이 목적이다. 논항을 취하고 명제 내용에 관여하는 실질 용언의 ‘있잖아’와 달리, 담화 표지 ‘있잖아’는 일상 대화 속에서 다음과 같은 세 기능을 담당하고 있다. 첫째, 담화 표지 ‘있잖아’는 [마디]-[담화 표지-마디]에 놓여 화제를 도입하는 데 초점을 둔다. 그런데 앞 화제와 의미적으로 단절되어 새로 화제를 구성하는 마디의 앞에 놓이어 화제 전환에 관여할 수도 있지만, 새로 화제를 구성하더라도 앞 화제와 의미망이 여전히 연결되어 있어서 화제 추이에 관여할 수도 있다. 둘째, 화제가 바뀌는 지점에 ‘있잖아’를 두어서 청자에게 화제가 바뀐다는 신호를 일부러 보내어 청자의 주의를 환기시킨다. 이와 같이 청자에게 주의를 환기시키는 이유는, 청자에게 갑자기 새로운 지식을 활성화하도록 하는 것은 청자에게 부담을 줄 수 있어 청자의 체면을 손상할 가능성이 커지기 때문이다. 이 때 담화 표지 ‘있잖아’는 화제 이동으로 인한 청자의 체면이 손상되는 것을 막거나 줄이기 위한 공손성 장치로 쓰인다. 셋째, 화제의 내용을 잘 알고 있지만 그 화제를 지시할 정확한 언어 기호를 모를 때 어휘부에서 그 기호를 탐색하는 과정에 담화 표지 ‘있잖아’가 쓰일 수 있다.

The purpose of this paper is to deal with functions of issjanha as a discourse marker. The functions of issjanha can be divided into three types based on Maschler(2009). Firstly, issjanha has a textual function. It introduces a topic, which makes all the utterances keeping coherence in the part of conversation. It is general that issjanha consists in constructions X issjanha or issjanha X, where X is a topic. If hearer takes part in X, the hearer is not a topic but an appellation. Secondly, issjanha has an interpersonal function. Hearers usually focus on the one topic. If speakers change a topic without notifying to hearer, hearers lose face. The loss of face can be prevented with issjanha as a discourse marker. Thirdly, issjanha has a cognitive function. Though speakers know the contents of a topic, sometimes they don’t know the exact linguistic sign referring the contents. In that case speakers can use issjanha while searching for the sign, and issjanha works for keeping their turn in conversation.

This paper aims to establish the grammatical relationship between the syntactic positions of S-adverbs and information structure. The crosslinguistic properties of S-adverbs are as follows: i) S-adverbs have a focal stress only in a post-subject position; ii) S-adverbs cannot occupy a post-subject position when subject is focused; iii) S-adverbs in a pre-subject position are usually interpreted as topic. To analyze these properties, a new approach toward S-adverbs is attempted in this paper. This approach actively applies two important Minimalist theories, that is, Agree and phase. Under the Agree mechanism suggested here, the Probe C0 carrying an uninterpretable feature [uF] seeks for its Goal containing an [iF]. By Agree, every S-adverb bearing an [iF] can be independently licensed by C0, a phase head. On the other hand, under the phase-based focusing mechanism made in this paper, only the highest syntactic element (either subject or S-adverb) can receive a focal stress in the spellout domain of the CP phase (i.e. TP). In other words, the spellout domain is a focus domain for S-adverbs. Unless it has a focal stress, neither subject nor S-adverb can survive in the focus domain. For this reason, S-adverbs occurring in a pre-subject position are the outputs of anti-focus movement, and thus they are usually interpreted as topic.

In Japanese, the “～ni tsuite wa (regarding～)” form comprises thetopic particle “wa” to the phrase particle “～ni tsuite (regarding～)”. The for the “ni tsuite” form is used when a specific topic is discussedbetween the speaker and listener, or when a description is made withinthe boundary of a definition. Likewise, “wa” performs the function ofcontrasting the meanings, limiting the topic to a specific extent, as isdone by “eun” or “neun,” auxiliary particles (or determinatives) in theKorean language. In practice, the “～ni tsuite wa” form, composed byadding “wa” to the “～ni tsuite” form, is in many cases interchangeablewith the simple “wa” , without causing any alterations in grammar ormeaning. This paper examines the uses of the two forms, in which thetwo perform the same function of indicating a specific and limited nounor subject, thereby proposing the use of the topic particle “wa” intranslation and interpreting, for a simpler and more accurate use oflanguage.

이전의 연구들에서 ‘トキ(ニ)ハ’가 조건을 나타내는 일이 있다고 하는 지적은 있었 지만 조건의 어떠한 용법․형식과 관련되어있는지에 대해서 명확히 기술된 일은 없다. 이에 본 연구에서는 ‘トキ(ニ)ハ’의 조건용법을 고찰하고자 선행연구에 의거하여 조건 의 용법을 여섯 가지로 나누고, 특히 ‘ハ’의 용법에 주목해서 조건을 나타내는 ‘トキ (ニ)ハ’가 어떤 조건용법과 관련되어지고, 또한 ‘と’, ‘ば’, ‘たら’, ‘なら’의 어떤 조건 형식과 치환가능한지를 살펴보았다. 그 결과를 서술하면 다음과 같다. ‘トキ(ニ)ハ’는 가정조건, 항상조건, 우연조건, 과거의 습관을 나타낼 수가 있는 한 편, 확정조건, 반사실조건과는 관련이 없다. 가정조건을 나타낼 경우 대비의 ‘ハ’가 쓰 인 ‘トキ(ニ)ハ’는 ‘たら’나 ‘ば’와 치환가능하고, 주제의 ‘ハ’가 쓰이면 ‘(の)なら’와 바꾸어 쓸 수 있게 된다. 가정조건 외의 다른 세 가지 조건용법에 쓰이는 ‘トキ(ニ)ハ’ 의 ‘ハ’는 주제용법으로 보이며, 각각의 조건용법에 맞는 조건형식에 대응된다. 단 ‘ト キ(ニ)ハ’가 어떤 조건용법과 관련이 있다고 해서 그 조건용법의 문이 모두 ‘トキ(ニ) ハ’와 치환될 수 있는 것은 아니며, 또한 ‘トキ(ニ)ハ’와 어떤 조건형식이 치환된다는 것이 반드시 각각의 세세한 의미내용까지도 완전히 같다는 것을 의미하지는 않는다. 이상의 고찰을 통하여 ‘トキ(ニ)ハ’는 몇 가지 조건용법을 나타내고 그에 상응하는 조건형식과 치환될 수 있는 만큼, 상당히 복잡한 의미내용․구문구조를 가지고 있는 표현임을 알 수 있다.

以前の研究において｢トキ(ニ)ハ｣が条件を表すことがあるという指摘はあった ものの、条件のどのような用法形式と関わっているのかについては明らかではな かった。そこで、本研究では、｢トキ(ニ)ハ｣の条件用法を考察すべく、先行研究 を基に条件の用法を六つに分け、特に｢ハ｣の用法に注目しながら条件を表す｢トキ (ニ)ハ｣がどの条件用法と関わり、また｢と｣｢ば｣｢たら｣｢なら｣のうちのどの条件形 式と置き換えられるのかを調べた。その結果を述べると、次のようになる。｢トキ(ニ)ハ｣は、仮定条件、恒常条件、偶然条件、過去の習慣を表すことがで きる一方で、確定条件、反事実条件とは関わりを持たない。仮定条件を表す場 合、対比の｢ハ｣が使われた｢トキ(ニ)ハ｣は｢たら｣や｢ば｣と置き換えられ、主題の｢ハ｣が添えられると｢(の)なら｣と置き換えが可能になる。仮定条件以外の、他の 三つの条件用法としての｢トキ(ニ)ハ｣の｢ハ｣は主題用法と見られ、それぞれの条 件用法に合った条件形式に対応している。但し、｢トキ(ニ)ハ｣がある条件用法と 関わりを持っているからといってその条件用法の文がすべて｢トキ(ニ)ハ｣と置き 換えられるわけではなく、また｢トキ(ニ)ハ｣とある条件形式が置き換えられると いうことが必ずしもそれぞれの細かい意味内容までもまったく同じであることを 意味するのではない。 以上の考察を通して、｢トキ(ニ)ハ｣は、いくつかの条件用法を表し、それに 合った条件形式と置き換えられるだけあって、かなり入り組んだ意味内容構文構 造を持っていることが分かった。

Korean shows very different sentence structure from Indo-European language, but unfortunately, its characteristic features are not yet clearly described. This paper claims the following structural characteristics: (1) No obligatory sentential element can be posited for the Korean sentence. (2) A Korean sentence consists of necessary elements in consideration of context. (3) A discussion of sentential semantics presupposes the distinction between meaning conveyed by the linguistic expression and information conveyed through the context. (4) Practically all Korean sentences are described not in terms of 'subject-predicate' structure, but in terms of 'topic-comment' structure. (5) The topic can be defined as 'field of judgement', the comment as 'content of judgement', and the sentential ending, as 'intention of utterance'. (6) The shared information between the speaker and the addressee is conveyed mainly through the context. (7) Examples of translation between Korean and English reveal the characteristic features of Korean sentence.

This paper examines the question of whether the order of the left peripheral elements causes difficulties in EFL adult learners’ using WH-structures. Furthermore, it suggests a method to solve the problem by applying A'-chain (which is a chain linking an s-selection to a criterial position) based on the Criterial Hypothesis (Rizzi (1997, 2004a, 2004b)). To justify the suggestion, the survey targeted 81 EFL adult learners. The results indicate that the subjects have difficulties in using grammatical structures which include wh-elements, focus- elements, and topic-elements in the leftmost edge of the sentence. It has been identified that the fundamental reason for making the difficulties is due to the different way to operate the left peripheral elements between Korean and English. To tackle this challenge, this paper suggests that EFL adult learners should recognize the order of operating the left peripheral elements (Top-WH-(Top)-Foc) and apply the order of them to build the wh-structures in terms of the concept that the standard of the operation is wh-movement which is always preceding the focus and topic movement.

Modal expressions such as CAN and V-(u)l-swu iss ‘can’ are ambiguously interpreted. That is, The English modal can is ambiguous between a root modal (i.e., RM) reading denoting an ability/capability reading and an epistemic modal (i.e., EM) reading denoting a possibility/probability reading. In the same vein, The modal external negation disambiguates the modal reading: only the RM reading survives, while the EM reading is suppressed. The embedded focus bearing element in the EM reading cannot bear a presupposed interpretation in spec-C, since the plain topic cannot be compatible with the embedded spec-C. Thus, the focus bearing element between an interrogative C and wh-in-situ in the EM reading always induces the intervention effect. More importantly, when the nun-marked subject occurs in the modal V-(u)l-swu iss ‘can’ context, only the RM reading survives. That is, the plain topic plays an important role that disambiguates the modal reading. This leads us to believe that the real reason why the modal external negation context induces the sentence to only bear the RM reading is because the subject in such a context plays a topic. It logically follows that the subject in the modal external negation context such as V-(u)l-swu eps ‘cannot’ should be a topic, which always induces the intervention effect to be cancelled.

Korean is an agglutinative language where grammatical categories are realized with morphemes. As compared to Indo-European languages which are inflecting languages, Korean grammatical forms are relatively simple and regular. Further, ellipsis is rare phenomena in Korean grammar. Taking these into consideration, issues and methods for a study of Korean grammar must be determined by careful examination of grammatical characteristics observed in the empirical data rather than by grammatical theories proposed for Indo-European languages. The word order of Korean sentences has been described as ‘SOV’, but this generalization is couched in terms of sentence structure of Indo-European languages. There are many utterances that do not have ‘subject’. Further some sentences appear to have more than two ‘subjects’, and there is no fixed ordering between subject and object. This necessitates a careful reexamination of the traditional generalizations: the generalization that a sentence consists of subject and predicate and the generalization that the concept of ‘subject’ is necessary for a sentence structure in Korean. This study analyses sentences as consisting of ‘topic+comment’, not of ‘subject+predicate’. Accordingly, word-order in Korean should be described in terms of ‘topic’ and ‘comment’. This ordering of ‘topic-comment’ is very strict, and so is the ordering of ‘modifier-modified’.