從句法音韻介面觀點研究 i-世代台語二字格、三字格及四字格之口語變調 (2/2)

Abstract

這個二年期研究計畫的目標是建立一個i-世代台語語料庫，提出句 法音韻介面理論分析。i-世代手握智慧手機 (譬如蘋果iPhone) 而 成長，泛指1995年至2005年期出生者。台語i-世代的音韻行為之所 以值得探討，其因有三：一ヽ台語i-世代長時間透過智慧手機與人 溝通及娛樂, 但缺乏面對面社交接觸；二ヽ台語i-世代在社交場合 的語言經常是台語與華語混雜；三ヽ台語i-世代雖然聽得懂台語 ，但經常以華語回答，即使在家亦是如此。本研究計畫即深入檢視 標準韻律理論與配稱理論如何詮釋i-世代台語的口語變調。計畫第 一年，本人收集i-世代台語的口語變調語料。語料庫總共收集了 29890個詞句樣本，包括1500 個雙音節、6090個三音節、10086 個 四音節，以及 12214 個較長多音節樣本。樣本收集是根據不同的句 法結構、變調位置及相鄰聲調。語料庫顯示，i-世代台語語者傾向 將較長的語句切割成較短的片段，而這些片段本質為韻律結構，形 成變調範域。計畫第二年，本人就第一年所建立的語料庫，分析其 句法與音韻的介面。尤其，本人從輸出值來檢驗各類句法結構、韻 律結構與不通聲調組合之間的關係。句法與音韻之間的競爭主要關 鍵在於一組對整治約及韻律標記制約的互動。本計劃並發現，以配 稱理論的架構分析則會導致錯誤的臆測。

The goal of this two-year project is to establish a database of iGeneration Taiwanese and to engage in a theoretical analysis from the perspective of the syntaxphonology interface. The iGeneration, who grew up with a smartphone (such as iPhone) in hand, is loosely referred to people born between 1995 and 2005. The phonological behavior of the Taiwanese iGeneration (T-iGen) is worth exploring for three reasons. First, T-iGens spend a great deal of time using their smartphones for communication and pleasure, but lack of face-to-face social contact. Second, the language of T-iGens is often a mixture of Taiwanese and Mandarin in social environments. Third, T-iGens, though understanding Taiwanese, frequently respond in Mandarin, even at home. This project engages in the investigation of how the standard prosodic theory and the match theory account for the speech tone sandhi of iGeneration Taiwanese (iGT). In the first year, I collected data of speech tone sandhi from iGT. The database eventually contains 29890 tokens of phrases and sentences, including 1500 disyllabic ones, 6090 trisyllabic ones, 10086 tetrasyllabic ones, and 12214 longer ones. They are based on a variety of syntactic bracketing and projections, different positions of sandhi targets, and diverse combinations of neighboring tones. The established database shows that the iGT speakers tend to parse longer expressions into shorter fragments, which are by nature prosodic domains on which tone sandhi operates. In the second year, I analyzed the data collected in the first year in terms of the syntax-phonology interface. In particular, I examine the relationship between the various syntactic structures, prosodic structures and tone combinations from the outputs. The syntax-prosody competition is keyed to a set of alignment and prosodic markedness constraints. An alternative approach under the Match Theory, on the other hand, renders incorrect predictions.