Complex-NP Islands in Korean: An Experimental Approach

Complex-NP Islands in Korean: An Experimental Approach Yong-hun Lee Yeonkyung Park Chungnam National University Hannam University 99 Daehak-ro, Yuseong-gu 70 Hannamro, Daedeok-gu Daejeon 305-764, Korea Daejeon 306-791, Korea yleeuiuc@hanmail.net withbyk@hanmail.net Abstract This paper took an experimental approach and examined island constraints in Korean. Among many island constraints, this study took a Complex NP island constraint, and the experiment was designed with 3 related factors: presence vs. absence of island, matrix clause vs. embedded clause, and scrambling. The analysis results illustrated that the presence/absence of complex NP island did not play a role by itself in Korean but that it made distinctions through the interactions with other factor such as matrix vs. embedded clause. 1 Introduction Since Ross’s identifications of island constraints in English (Ross, 1967), there have been a lot of debates on the existence of island constraints in other languages. For example, Nishigauchi (1990) and Watanabe (1992) claimed that there were island constraints in Japanese, but Ishihara (2002) and Sprouse et al. (2011) mentioned that this language had no island constraint. Likewise, there have been controversies on the existence of island constraints in Korean. Some have argued for the presence of island effects (Lee 1982, Han 1992, Hong 2004), while others have argued against it (Sohn 1980, Kang 1986, Suh 1987, Hwang 2007). This paper investigated the island constraints in Korean. Our questions were (i) if Korean also has the Complex NP island constraints and (ii) if there are, why there have been so many controversies on the existence of island constraints. In order to answer these questions, this paper took an experimental approach and examined the island properties in Korean. The target sentences were constructed with three factors, and native speakers’ intuition was measured with Magnitude Estimation (ME). After the experiment, all the data for the target sentences were extracted and they were statistically analyzed with R. Through the analysis, it was found that that the presence/absence of Complex NP island did not play a role by itself in Korean but that it made distinctions through the interactions with other factor such as matrix vs. embedded clause. These examples provided an account for why there have been so many controversies on the existence of island constraints in Korean. This paper is organized as follows. In Section 2, previous studies were reviewed especially focused on the experimental approaches. Section 3 includes the accounts for experimental design (research materials and research methods), and Section 4 enumerates the analysis results. Section 5 contains discussions, and Section 6 summarizes this paper. 2 Previous Studies 2.1 Island Effects in Korean Since Ross (1967) identified the island constraints in English, there have been lots of studies on the existence of island constraints in other languages. Those studies primarily focused on examining if the island constraints exist in their languages and why the language escaped the island constraints when the language did not demonstrate the island phenomena. Korean is no exception. There have been lots of studies on the island constraints in Korean. Earlier studies …

Content truncated to 500 words