This study examines the relationships between referring expressions and their referents in English and Korean language. Using the Accessibility as well as the Givenness frameworks (Ariel, 1991, 1994; Gundel et al., 1993), this study investigates the distribution of the referring expressions in each language and how they are related to the framework. Unlike most languages, in Korean, third-person pronouns are rarely used and definite constructions are not used, with the absence of definite and indefinite articles. With these differences in mind, this study aims to examine whether the framework of Gundel et al. (1993) bears its universality to account for the actual use of the distant languages such as English and Korean. Following Mulkern (1996) and Lee (2010), this study restricts its data in political news media discourse from the contrastive perspective. The data consists of ten English political news articles and twenty Korean political news articles published in 2017. The result confirms that both languages follow the general convention of news discourse to introduce a new entity with full names and prefer family names in subsequent mentions. However, further investigation is needed in terms of the use of pronouns in Korean because the distribution from this study differs from Lee's (2010).More contrastive research on different languages in different discourse contexts would shed more light on the universality of the Accessibility and Givenness frameworks.

The purpose of this study is to evaluate the empirical adequacy of the event-based language typology with special reference to English and Korean. The event-based language typology classifies languages into I- and D-languages based on the importance of the initial and terminal bounds of an event with respects to the way the relevant languages use in the recognition of eventiveness. In I-languages, in which the initial bound of an event is important, activity and accomplishment are recognized as events, while in D-languages, where the terminal bound of an event is important, achievement and accomplishment are so recognized. English and Korean, the target languages of this study, show the characteristics of I- and D-languages respectively. In the case of English, we can confirm that it falls into D-languages based on the fact (1) that event cancellation is impossible and (2) that causative reading is possible in terms of the delimitation of predicates. In the case of Korean, we can tell that it belongs to I-languages as (3) event cancellation is possible and (4) only NPs that can work as an initiator can be the subject in the position of [Spec, FP-init].

본고의 목적은 일본인 학습자에게 효율적인 한국어 연결어미 ‘-아/어서’와 ‘-고’에 대한 학습방안을 모색하는 데 있다. 먼저 한국어 연결어미 ‘-아/어서’와 ‘-고’에 대응하는 일본어 접속조사 ‘-て’의 의미 용법을 한일 대조의 입장에서 살펴보았다. 일본어의 ‘-て’의 용법을 분석한 결과, 순서, 나열, 수단․방법, 원인․이유, 역접 등 거의 모든 연결 관계에 걸쳐서 나타났다. 이에 오류의 원인을 과잉일반화와 모국어의 간섭에 주안점을 두고 학습자의 오류양상을 형태상의 시제오류와 의미상의 대치오류로 분류하였다. 시제오류에서는 ‘-ので’의 간섭으로 보고 반복적 활용연습에 초점을 두고 학습방안을 제시하였다. 대치오류에서는 ‘-て’의 과잉일반화로 ‘-아/어서’와 ‘-고’에 대치되는것으로 보고 초급에서 중점으로 다루는 순차성과 수단 방법의 용법으로 범위를 제한하여 교수 학습방안을 제시하였다.