台灣學生英語冠詞使用之研究

Abstract

本研究旨在探討台灣學生在英語冠詞使用上的表現。研究對象為150位英語學習者，依據其英文程度分為初級、中級、高級組。試題包含作文及克漏字填充。研究分析包含(1)定冠詞(definite article) the和虛冠詞(null article) 過度使用(overuse)的程度、(2)三組受試者的冠詞使用正確率、(3)受試者在兩種測驗中的冠詞使用是否呈相關、(4)在各種名詞環境中的冠詞使用。研究結果顯示，定冠詞的正確率偏高，虛冠詞的正確率偏低，而台灣學生最常犯的冠詞錯誤是使用定冠詞the代替虛冠詞。因此，定冠詞的高正確率不代表學生確實了解如何使用定冠詞，而是學生易將定冠詞推廣(overgeneralize)到其他的冠詞環境。另外，冠詞使用會隨著語言程度及言語活動改變。在克漏字測驗中，冠詞使用的正確率和英語程度成正比；在作文中，高級學生的冠詞使用沒有比中級學生正確。此外，受試者在兩種測驗方式的表現不存在任何可以預測的關係 (zero correlation)，這是因為克漏字測驗學生的接受性語言知識(receptive knowledge)，而作文測驗能產性語言知識(productive knowledge)，運用不同的知識會導致不同的冠詞使用。本研究也指出，類指(generic reference)、非慣用語(non-commonly occurring sequences)、抽象名詞(abstract nouns)是最常出現冠詞錯誤的環境。台灣學生經常過度使用定冠詞the於複數類指(plural generic nouns)和非慣用語，並且過度使用虛冠詞於單數可數或複數的抽象名詞(singular count/plural abstract nouns)。 除了探討台灣學生英語冠詞的使用，本研究也檢視英語學習教材如何呈現及解釋英語冠詞，發現教材中許多錯誤的規則和過度依賴單句說明(sentence-level samples)是導致冠詞使用錯誤的原因。根據上述分析結果，本研究提供一些教學原則及具體可行的教學活動以供參考。

The present study investigates the article use of 150 Taiwanese students learning English as a second language, categorized as low, intermediate, and advanced levels according to their English proficiency, on a blank-filling test and a composition test. The analyses focus on (1) degrees of the overuse and  (null article) overuse, (2) overall accuracy at the three proficiency levels, (3) article performance in different test methods, and (4) use of articles in different NP contexts. The results show that ofthe three English articles, the obtains the highest accuracy and  obtains the lowest. Besides, the most frequent article error is overuse of the in  contexts, which explains the accuracy ratios in the and  contexts. Therefore, students’ perfect performance in the contexts does not mean they really know when to use the definite article. Rather, its high accuracy results from the overgeneralization of the into a(n) and  contexts. Second, article performance varies with proficiency level as well as language production settings. On the blank-filling test article accuracy improves with increased L2 proficiency, but such a relationship is not completely confirmed on the composition test. Furthermore, no correlation is found within each group between article performance in the two test methods, indicating that accurate article use in one test does not guarantee equal accuracy in the other. The lack of correlation between the tests is due to different degrees of attention to form elicited in different language tasks: the blank-filling test, requiring learners’ receptive knowledge, elicits a higher degree of attention to form, whereas the composition test, requiring productive knowledge, elicits a lower degree. Also, the study points out that generic nouns, non-commonly occurring sequences, and abstract nouns are the most problematic article environments for Taiwanese EFL learners. More specifically, the is overused for plural generic nouns, which actually require , and for non-commonly occurring sequences, which require a(n) or . As for abstract nouns,  tends to be overused when it is a(n) or the that is required. In addition to Taiwanese students’ use of articles, this study also examines how instructional materials present the article system and points out that inadequacies of instructional materials should be largely responsible for students’ article errors. Problems with existing article pedagogy include inappropriate reliance on misleading rules and sentence-level samples. In light of these defects this study suggests some teaching principles and activities for EFL/ESL teachers to effectively teach the functions of articles.