不同語言類型的潛在影響：以日籍學習者對於漢語複合趨向事件表述之認知為例

Typological Differences: a Study on Japanese L2 Learners’ Cognitive Perception of Chinese Motion Event Expressions

本研究以漢語常用兩大類句型作為實驗材料 (cf. Chen and Guo 2009)， 共二十二名漢語母語者及十五名日籍學習者透過線上問卷進行圖片自然 度判斷調查，探究日籍學習者在判定漢語趨向事件表述的不同句式時，是 否受其母語類型特徵的影響。結果顯示：(1) 母語者在「方式句」(MP) 與 「路徑句」(P)的自然度判定出現顯著差異 (M = 4.43 vs. M = 3.99， p < .01)；學習者則無顯著差異 (M = 3.38 vs. M = 3.36， p = .808)，此可能與 其母語「動詞語言」中不傾向提出方式之特徵相關。(2) 在含有趨向指示 詞（來／去）的語句上，日籍學習者受其母語結構干擾，在自然度判斷上 趨於保守。(3) 在病句題組中，日籍學習者明顯接受度較高 (M = 2.23/ 2.76)，而漢語母語則有一致性低接受度 (M = 1.35/1.55)，組間差異具顯著 性 (p < .001)。另，母語者對於病句「*方式 + 路徑 + 來／去 + 處所」 接受度高，則可能受到方言的影響（cf. 陸儉明 2002）。

This study explores the possibility of Japanese learners of Chinese as a second language being influenced by the typology of their first language when judging the relative naturalness of two major types of Chinese motion event expressions (cf. Chen and Guo 2009). Twenty-two Taiwanese speakers and fifteen Japanese learners of Chinese took part in an online survey. The results of the study are as follows: (1) Significant differences were obtained between MP and P patterns within Taiwanese speakers (M = 4.43 vs. M = 3.99， p < .01) in contrast to Japanese learners’ group (M = 3.38 vs. M = 3.36， p = .808); (2) In terms of sentences that contain deictic expressions (lái/qù)， Japanese learners of Chinese were influenced by a differing legitimate structure in their first language; (3) In ungrammatical patterns， the average naturalness rated by the Taiwanese subjects were all lower than 2 points (M = 1.35/1.55)， while relatively-high points were found in Japanese subjects (M = 2.23/2.76) instead; significant differences were found (p < .001) between the two groups. The overall results indicate that Japanese learners of Chinese were to some extent influenced by directional structures in their first language; however， the curious anomaly in the high acceptance rate by the Taiwanese group of the pattern “*VManner + VPath + Deictic + Ground” may be contingent upon indirect influence from the Taiwanese dialect (cf. Lu 2002).