Please use this identifier to cite or link to this item:
A Comparative Study of L2 Acquisition of Three Mandarin Pronouns by Foreign Students in Taiwan in Taiwan
|Abstract:||在中介語研究中(Selinker, 1972)，第二語言學習者有共通的語言錯誤類型和必經的發展階段，但因母語之差異各階段所花的時間略有不同，且第二語言學習者之表現亦可能與與Eckman(1977)的標誌假設有關。因此，本研究探討在台之外籍學生習得華語特殊人稱代詞的過程，是否亦支持第二語言習得理論中的語言共通性(linguistic universals)、語言轉移(language transfer)及標誌性(markedness)等假設。研究對象為120位就讀國立台灣師範大學國語中心，母語為英語、印尼語及韓語的學生，每組40位，每組並依該中心分班測驗成績分中初及中高兩級，並納入20為母語人士為控制組，進行一個兩年期的華語特殊人稱代詞習得研究。今年以反身代詞「自己」與旁稱代詞「人家」為主，分析三組外籍學生對此兩代詞了解與使用之差異。 研究發現，外籍學生習得華語反身代詞和旁稱代詞時，三組表現相似，英美學生對華語反身代詞「自己」和旁稱代詞「人家」的完全/不完全長距離指涉及有/無阻擋效應的表現並不亞於韓語組及印尼組，母語影響並不顯著。此外，研究發現外籍學生較具標誌性的位置(如：主語的反身代詞「自己」及賓語的旁稱代詞「人家」)，都較無標誌之位置(如：賓語的反身代詞「自己」及主語的旁稱代詞「人家」)來得難些，支持標誌假設。再者，研究發現，外籍學生之表現未受母語影響，且一致呈現主/賓語不對稱性，說明語言共通性。外籍學生對兩個代詞之語意與語法表現相近，但整體表現語意指涉仍較其語法特性來得好。最後，研究發現，外籍學生在影片文法判斷測驗之表現比圖片文法判斷測驗來得佳，呈現題型效應。|
According to Selinker (1972), it seems quite evident that L2 learners pass through similar sequences or stages, though the time they spent at each stage may vary. In addition, it has been found that L2 learners’ performance is constrained by Eckman’s (1977) Markedness Differential Hypothesis. Hence, the purpose of the present study is to conduct a two-year project to investigate foreign students’ L2 acquisition of special pronouns in Mandarin to argue for linguistic universals, language transfer, and markedness. The subjects of the present study are 120 foreign students of National Taiwan Normal University whose first languages are English, Korean, and Indonesian, respectively, together with a group of 20 native controls of Chinese. The L2 learners were further divided into two levels (i.e., mid-low and mid-high) according to the placement test given by the Mandarin Training Center. Two pronouns were investigated (i.e., the reflexive pronoun ziji and the personal pronoun renjia) in the first year to see if the L2 learners show any differences in their comprehension and production of the two pronouns. The preliminary findings are as follows:First, it was found that the subjects comprehended the two pronouns in a similar way. The English-speaking subjects did not perform worse than the Indonesian group and the Korean group on the long-distance binding and the blocking of ziji and renjia, showing that L1 influence was not significant. Second, our subjects performed better on the unmarked positions (such as the object position of ziji and the subject position of renjia) than the marked positions (such as the subject position of ziji and the object position of renjia), supporting Eckman’s markedness. Third, linguistic universal was found obvious since all the subjects demonstrated no effect of L1 influence and a clear subject/object asymmetry in their judgment. Fourth, there was no statically significant difference between the subjects’ responses to semantic and syntactic properties； however, the mean scores of semantic properties were slightly higher. Finally, a task effect was found since our subjects performed better on the film-cued grammaticality judgment task, which provided the subjects with more information, than the picture-cued grammaticality judgment task.
|Appears in Collections:||教師著作|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.