以改良的Wug測試再探華語三聲變調的語音變異不全現象

Revisiting Phonetically Incomplete Tone Three Sandhi in Mandarin Chinese: Insights froma Revised Wug Test

本文探究華語三聲變調在假字實驗中的語音變異不全現象，是否能歸因於詞彙熟悉度效應所導致的發音執行問題。本研究透過改良後的wug測試創造了新的物品及顏色名稱以誘發40位臺灣華語母語成年人士產出三聲變調字詞。語音聲學分析的結果顯示，連續三聲雙音節詞的組合類別是影響三聲變調語音變異是否完整應用的關鍵。當雙音節詞假字中的第一個音節是臺灣華語真實存在的音節時，三聲變調在語音層面上能完全應用；當雙音節詞假字中的第一個音節不是臺灣華語真實存在的音節時，三聲變調在語音層面上則無法完全應用。實驗的結果支持了華語三聲變調在假字實驗中的語音變異不全現象是由詞彙熟悉度效應所導致的發音執行問題。本文將討論改良後的wug測試在研究變調議題上的益處，也同時討論變調假字衍生力與其心理表徵之間的關係。

This study explored the possibility that the phonetically incomplete application of Tone Three Sandhi (T3S) among pseudowords in Mandarin Chinese could be reduced to the articulatory implementation factor triggered by lexical familiarity effects. A revised wug test with newly-constructed color and objects was used to elicit T3S productions from 40 native Taiwan Mandarin adult speakers. The results from acoustical analysis showed that the nature of the word type of the first T3 syllable in the disyllabic T3 word was the key to the phonetic (in)completeness of T3S. When the first syllable of the pseudo-disyllabic word was an actually-occurring syllable, the application of T3S was phonetically complete. When the first syllable was an accidentally-gapped syllable, the application of T3S was phonetically incomplete. These results support the view that the articulatory implementation, which is triggered by lexical familiarity effects, exerts strong influences on the phonetic completeness of T3S. The benefits of the revised wug test used in this study as well as the relationship between the degrees of pseudoword productivity and mental representations of tones are discussed.