自發性國語語料中的焦點

Abstract

以焦點為主題的研究，在語言學的各個領域都有相關討論。然而，針對國語中的焦點，大部分研究所使用的語料為大陸地區的普通話，即使有少數針對台灣地區國語所做的研究 (Hsiung, 2002) ，仍以朗讀語料 (read speech) 為分析的重點，且多為發音實驗 (production experiment) 。因此，本研究以自發性語料 (spontaneous speech) ，針對當前台灣地區國語中焦點的現象進行探討。 在實驗一的發音實驗 (production experiment) 中，十六個雙音節名詞涵蓋國語中可能的聲調組合，分別位在句首、句中及句尾的位置，承受窄焦點 (narrow focus) 或是沒有承受焦點 (neutral focus) ，由施測者發問，受試者根據所聽到的問題及圖片為引導回答。在實驗二的感知實驗 (perception experiment) 中，使用受試者CBY的發音語料為刺激項，將每一答句可能對應的兩個問題以ABX順序錄製，受試者根據所聽到的肯定句，反推該句所回答的問句為何。 研究結果發現，在本實驗自發性國語的發音語料中，音長的增加及基頻範圍的增加皆是表現焦點的方式，然而基頻範圍的表現不如音長來的明顯，因此，音長的增加為表現焦點的主要方式。同時，受試者 CBY和CKC的資料顯示：由母音/i/，/u/，/a/的第一共振峰 (F1) 和第二共振峰 (F2) 所構成的母音空間 (vowel space) ，在承受窄焦點時，有擴張 (enlargement) 的現象。本研究同時發現，在自發性國語的感知方面，音長是受試者判斷焦點最主要的依據。

Two experiments are conducted to investigate the effect of focus in Taiwanese Mandarin. In production experiment, two types of sentences are used. There are 16 target nouns with all tonal combinations in Mandarin; each target noun consists of two syllables. Two focus conditions are manipulated: narrow focus on target noun, neutral focus on target noun. Three positions of target nouns are controlled: initial, middle and final. Three subjects participated and produced 288 sentences by answering precursor questions. In perception experiment, the production data of subject CBY is used as stimulus without synthesis or manipulation. The stimulus is two precursor questions and one answer as a set recorded in ABX order. There are 288 questions; subjects are tested and instructed to circle the precursor question that the utterance responds to. The results of two experiments show that in production, both duration and F0 range expansion are salient acoustic cues in spontaneous Taiwanese Mandarin, however, the parameter, F0 range expansion does not seem to be as important as duration. Besides, the vowel space formed by the F1 and F2 values of /i/, /u/, /a/ in target nouns tends to enlarge under narrow focus except for one subject. In perception, duration is the most important cue subjects use to perceive focus condition.