標準國語的想像？台灣配音表演的語音與社會意涵分析

Abstract

過去有研究指出台灣配音一直以來不受親睞，且有多於一半的觀眾認為配音表演甚不自然(Ishii et al., 1990)，但皆並未探討配音表演具體的特性。本研究結合作者自身之語言學與配音訓練，試圖回答以下三個研究課題：聽眾的具體感受為何？配音的語音特性為何？以及與形塑整個情況的社會成因又為何？聽眾感受的部分，作者撥放標準國語版與日文版的十句對白並設計問卷，請台日各一百名參與者評量對白。結果顯示台灣年輕聽眾普遍認為對白極為標準但甚不自然。日本年輕聽眾則認為既標準又自然。語言特性的部分，本文比較配音對白與日常言談之捲舌音，變調，連併，基頻軌跡及逐字時值變異，結果發現配音對白在變調之外都與日常言談有顯著差異。社會成因的部分，本文闡述配音對白之風格來自歷史脈絡下的實際需求與意識型態，結果則不可避免產生政治聯想。例如配音對白之逐字時值變異高於日常言談，而Low & Grabe (2002) 曾指出高逐字時值變異值是重音計時 (stress-timed) 語言的特性。這足以解釋第一部分問卷中開放式問題收集到不少答案表明配音對白感覺帶有中國大陸腔，因如Tseng (2004) 所指出，中國大陸的普通話比台灣的國語更加偏向重音計時。綜觀之，研究發現聽眾普遍感受台灣配音不自然，而配音的語音展現大於日常語言的抑揚頓挫，背後的社會成因則帶有政治意味，三者其實環環相扣，因配音表演風格受早期語言政策規範之國語發音所型塑，而今日人們對過去政策加以檢視反省的同時，也對依然處在國語發音框架之中的配音對白有了更加複雜的觀感。本研究運用量化研究方法，從問卷具體語音層次分析配音表演與日常言談之疏離，並可從而窺見所謂標準國語的矛盾處境與不同群體對標準國語之解讀，並或許能從語音學的角度提出一種更加自然的配音風格。

Past studies have shown that dubbing performance in Taiwan has not been well-received by audience, and is rejected by more than 50% of the audience as unnatural (Ishii et al., 1990). However a solid linguistic analysis on dubbing performance has been lacking. In this study I combine my training in linguistics and dubbing performance and try to answer the following three questions: How exactly is dubbing received? What are dubbing’s phonetic characteristics? Why is dubbing like how it is now? For the first question, a survey is given to Taiwanese and Japanese audience and they are asked to rank selected dubbing materials in their respective languages. The results show that Taiwanese audience finds dubbing performance very standard but unnatural, while Japanese audience finds their dubbing to be both standard and natural. For the second question, the study compares the retroflexion, downdrifting, contraction, pitch contour and PVI (paired variation index) of dubbing performance and daily speech and finds the two to be significantly different apart from some common downdrifting pattern. For the third question, the study explores the practical and ideological reasons in the history that shaped the dubbing style, which is interpreted politically today. For example, dubbing performance has a higher PVI than daily speech, which, according to Low & Grabe (2002), is a feature of stress-timed language, which in turns explains why some participants comment that the materials sounds like Mandarin from Mainland China in the survey, because as Tseng (2004) points out, Peking Mandarin is more stress-timed than Taiwan Mandarin. All in all, the unnaturalness of dubbing indeed has specific phonetic features, which are influenced by political social cause. Early language policy in Taiwan prescribed the system of Guoyu (‘national language’), and today, when the policy is being reviewed critically, the system also starts to receive mixed comments. The study uses quantitative method to analyse dubbing performance’ reception and phonetic discrepancies from daily speech, showing the current situation of Guoyu, namely different groups of people holding very different opinions on the system, and eventually attempts to propose an alternative dubbing style that would be more natural.