零起點日籍學習者之漢語聲調習得

Mandarin Tonal Acquisition of Novice Japanese Learners

林華一(Hua-Yi Lin)

指導教授 ： 曾金金博士

國立臺灣師範大學/國際與僑教學院/華語文教學研究所/碩士(2006年)

摘要

本文主要探究零起點日籍學習者在漢語聲調習得上的偏誤與習得情形。過去的文獻對於日籍學習者的聲調偏誤分析或是難點已有不少討論，但可惜的是大多數研究多以橫向分析，了解學習者的聲調偏誤類型，較缺乏縱向歷時的觀察。其次，很多研究僅單向地就學習者在發音方面產生的偏誤討論，而忽略了聽辨感知產生的偏誤情形。因此本文透過十週連續的實驗，進行錄音與聽辨測驗，了解受試學生每週在聽辨與發音的偏誤及其習得情形。此外，也進而探討聽辨感知與發音在習得上的相關性，以及音節組成成分與漢語聲調習得的關係。
　　本文研究結果發現：
(一) 零起點日籍學習者在習得聲調的過程中，一聲與四聲在十週的發展過程中，不論是聽辨感知或是發音，正確率均有顯著的提升，而且正確率高達97%以上，推論是較早習得的聲調。而二、三聲在十週的發展過程中，正確率雖有提升，但二聲聽辨仍有7%、發音11%的偏誤率，三聲聽辨為17%，發音為16%的偏誤率，推論為較晚習得之聲調。發展後期偏誤類型最明顯者為二聲與三聲的相互混淆。
(二) 我們發現實驗初期調型的偏誤是主要問題，但是隨學習時間增加，調域會成為學習者偏誤現象中較為主要的問題。
(三) 另外我們透過每週固定追蹤之音節，試著了解聲調習得與音節組成成分的關連，結果發現對於受試者熟悉之音節，四聲在聽辨方面之正確率與學習時間呈現正相關的走勢，也就是正確率隨著學習週數逐漸提高。但對於受試者不熟悉之音節，四聲之正確率與學習時間呈現正相關的走勢，在第三週才出現。推論對於不熟悉的音節可能分散學習者判別聲調的資源，導致正相關走勢的發展延遲。而發音方面，走勢則較顯曲折，兩組音節至第八週以後，才呈現較為正相關的走勢，可能需要更長的時間才能看到較為完整的習得過程。

關鍵字

零起點日籍學生 ； 漢語聲調 ； 中介語音 ； 聽辨感知與發音

並列摘要

This thesis investigates Mandarin Tone acquisition with Japanese novice learners. Many reports have discussed the tonal errors by Japanese students; however, most of the studies use horizontal analysis to see the error type but lack longitudinal analysis. Most reports concentrate on tonal pronunciation errors, and ignore the importance of the perception. This thesis tests eight novice learners through 10 weeks of tonal acquisition. It also discusses the relationship of tones to perception and production and relationship between syllables and the acquisition of tones.
The findings in the thesis are:
(1) Through the 10 weeks, we can see that the correct rate of T1 and T4, either in perception or in production is much higher than T2 and T3, even above 97%. As for T2 and T3, in spite of the fact that their correct rate is getting higher, T2 still has the error rate of 7% in perception, 11% in production; and 17% in perception, 16% in production for T3 in the 10th week. Therefore we suppose T1 and T4 can be learned earlier than T2 and T3. And we find out the error type is that the participants confuse T2 and T3 with each other.
(2) The errors of the tone contours are mainly a problem during initial study with novice learners; however as learning time progresses, the errors of register become a bigger problem than contours.
(3) We use the 4 tones of the 2 syllables “sha” and “ma” every week to be our secondary testing materials. And in perception, the correlation between the duration of learning time and the syllable “ma” is more obvious than the syllable “sha”. We think the reason is because the learners need to work harder to deal with those unfamiliar syllables in their learning process. In production, the correlation between the duration of learning time and the syllables is not obvious until the 8th week. It will take much longer to see the whole acquisition of production.

並列關鍵字

Japanese novice students ； Mandarin tones ； interlanguage phonology ； perception and production

參考文獻

宋時黃（2003）。以偏誤分析為主的華語語音教學網站設計-以撮口呼之發音教學為例-。國立師範大學華語文教學研究所碩士論文。

陳氏金鸞（2005）。越南學生漢語聲調偏誤分析。國立師範大學華語文教學研究所碩士論文。

張顯達（2000）。三至四歲兒童對國語輔音的聽辨與發音。Language and Linguistics，1.2:19-38。

Hall?, P., Vhang, Y., Best, C., (2004). Identification and discrimination of Mandarin Chinese tones by Mandarin Chinese vs. French listeners. Journal of Phonetics 32, 395-421.

Jayeon Lim, (2004). Ultimate Attainment in L2 Speech Perception and Production. Kate Forum. Vol.28. No. 2, 10-11.