韓国学習者の日本語動詞獲得モデル： 学習者総体モデルとの比較 ― 「多言語母語の日本語学習者横断コーパス」の絵描写作文を用いた検証 ―

Abstract

Using the International Corpus of Japanese as a Second Language (I-JAS), this study analyzed the picture description essays written by Korean learners of Japanese (KLJ) at different L2 proficiency levels to explore the process of their acquisition of L2 Japanese verbs. The results were compared to the verb acquisition process of the whole learners of Japanese (WLJ) reported in Ishikawa (2021). The corpus-based analyses showed that (1) an inverted U pattern existed in terms of the number of tokens/ types of the verbs used in the essays for both KLJ and WLJ, (2) when compared to the verb usage of Japanese native-speakers (JNS), some verbs were used only by JNS and others only by KLJ, and these “marker” verbs were almost the same for KLJ and WLJ, (3) KLJ’s verb acquisition process could be subdivided into four stages: 1 (existential verbs/ verbs of moving), 2 (a greater variety of verbs), 3 (composite verbs and aspectual verbs), and 4 (complex composite verbs and verbs of change), which were almost identical to the patterns for WLJ, and (4) KLJ’s L2 proficiency could be explained by the frequency of seven types of verbs (hairu, suru, miru, shimau, shiru, kuru, ireru) with 63% accuracy, though these verbs were different from a set of verbs adopted in the regression model for WLJ. These findings corroborated that KLJ’s verb acquisition pattern did not differ much from the pattern for WLJ, which may suggest the need for the Japanese learner corpus studies focusing on learners’ L2 proficiency rather than on their L1 types.