코퍼스를 활용한 신소설 문체 분석 시론

Application of 'Sinsoseol' Corpus to Stylistics : Authorship and Style of Chiaksan and Hongdohwa

The objectives of this paper is to confirm 'Sinsosoel' authorship through quantitative approach to stylistics. In 1900s when the new form of novel, Sinsosoel emerged, matter of authorship was so complicated. There were a lot of cases that the author named in publisher's imprint was not a genuine one but other author or a publisher himself. Recent research examined this matter with bibliographical approach, for example looking through biography of the author and the publisher, and their relationship, or publication law and market condition in 1900s around. This research significantly contributed to confirm the authorship and the canon among different editions, however it tended to pay less attention to the internal property of a work. This paper starts from this point to argue stylistics should be considered in order to confirm authorship and canon. It is very important to select works for this purpose. Here works with separate volumes of different authors under one title were considered, because two totally different stories can distract us from tracing one author's typical styles. Under one story line, we can easily trace two different author's distinctive style from another. Chiaksan 1's author is recorded as Yi Injik whereas volume 2 is Kim Gyoje, and Hongdohwa 1's author is Yi Haejo whereas 2 is Min Junho or Namgung Min. This two works will be examined to confirm whether they are really written by two different authors and if yes, which distinctive styles they have. The tool for this examination is Sinsosel Corpus established by AKS(Academy of Korean Studies) 2009-2011. Frequency listed from computational treat to Corpus will play an important role to confirm this two works authorship and each author's style. This approach will lead stylistics study to more scientific level.