【作者】 韩维新；

【导师】 石锋；

【作者基本信息】 南开大学 ， 语言学及应用语言学， 2013， 博士

【摘要】 本研究以语调格局理论为基础,从最基本的声学参数音高、时长入手,通过起伏度和停延率,对香港粤语四位发音人三种句型五种语气,共540个样品句的语句、进行全面深入地研究。根据石锋[32]语调层级的理论,本文对香港粤语不同的语调层级进行了细致的考察。本文以语调格局为主线,以语调层级对比的方法进行研究。具体来说,以陈述句(自然焦点句)作为基本模式,无标记疑问句及焦点句等为变化模式。从三个层面进行考察分析,并与普通话相应句型的语调进行对比。从研究结果可以看出香港粤语陈述句、无标记疑问句和焦点句的语调格局与普通话语调格局的共性和差别。主要创新点如下：a)陈述句句末音节的音高是表现全句语气信息的焦点。由于音高下倾和降阶的作用,语句整体呈现音高下倾趋势。句末末字通过边界调的降低,调域的扩展表现出陈述句的韵律节奏和语气。数据表明,普通话语句调域、词调域、字调域均比香港粤语跨度大,二者调域的最大值均在句末词调域和末字的字调域。b)舒声调的陈述句四位发音人平均停延率的排序：句末>句首>句中(1.30>1.06>1.00)。句末词调域、句首词调域末字的停延率分别比平均停延率提高了30%和6%,句中词调域末字没有提高。男性,女性发音人停延率的排序为：句末>句首>句中,男性句末,句首停延率分别比平均停延率提高了24%和6%，句中没有提高。女性句末,句首停延率分别比平均停延率提高了36%和6%,句中没有提高。句末词调域末字音高的显著提高和调域的扩展,时长的延长是香港粤语陈述句语调格局的重要特征。c)香港粤语无标记疑问句以句末末字音高的提高,调域的扩展,时长的延长来表达疑问语气。正像赵元任论及的上升的结尾调并不影响整个结构的语调型式,只影响最后一个音节里的浊音部分。其本身并无音位,要寄生在最后一个语位上。疑问语调先把降调或平调单字调的音高和调形进行调节,以疑问语调的音高和调型输出。如果原来是升调的末字,前段重合叠加,后段声调的斜率上升。最后一个音节的字调和语调完全融为一体。从10个音节音高的起伏曲线来看,发现句首收敛准备,逐渐缓慢上升,从第8和第9个音节的音高开始上升,末字不是孤立地突然上升。这也是符合生理及能量释放规律的。不同的声调,疑问语调提高的幅度不同。男性舒声调末字比左边相邻音节的音高提高了43%,女性提高了49.5%。低调值的末字提高的较多,高调值的末字提高的较少。句末末字时长的延长是疑问语气的重要标志,6个舒声调末字的停延率从1.36到1.69均有显著延长。舒声调句末末字停延率比左侧相邻音节的停延率延长了0.54,绝对时长平均延长了132毫秒,比相应陈述句平均绝对时长延长了57毫秒。d)香港粤语和普通话,均以自然焦点句(陈述句),句首、句中和句末不同位置的焦点句作为对比标量。香港粤语舒声调四种焦点句焦点词调域音高百分比的排序：句末焦点>句首焦点>自然焦点>句中焦点(84>75>71>66)。普通话四种焦点句词调域的排序：句首焦点>句中焦点>句末焦点>自然焦点(94>89>77>61)。普通话句首焦点词调域最宽,自然焦点句句末词调域最窄。香港粤语则是句末焦点句句末词调域最宽,句中焦点词调域最窄,二者焦点词调域的表现完全不同。从对比中发现,粤语句末词调域比普通话宽7个百分点。普通话句首焦点句的句首词调域比香港粤语句首词调域宽19个百分点。香港粤语和普通话四种焦点句词调域音高的对比有两点差别：(1)香港粤语句末焦点句的词调域最宽,普通话句首焦点句的词调域最宽。香港粤语自然焦点句、句末焦点句的句末词调域比普通话宽。(2)普通话句首焦点句、句中焦点句所在的焦点词调域发生了最大化扩展,焦点后音高骤降,调域大幅度压缩。香港粤语没有这些特征。这一特点与Chen,S.W.,Wang,B.,＆Xu,Y.和王玲、王蓓等研究的结论一致。即汉语普通话及印欧语系中大部分语言焦点词调域最大化扩展,焦点后音高骤降,调域压缩。这与普通话长期与满语蒙语接触受阿尔泰语系的影响有关。台湾的国语、闽南话,德昂语布雷方言,香港粤语均无这些特点,可能与语言类型有关。香港粤语和普通话的时长均以自然焦点句(陈述句),句首、句中和句末不同位置的焦点为对比参量。香港粤语舒声调四位发音人四种焦点句焦点词调域停延率的排序为：句末焦点>句首焦点>句中焦点>自然焦点(1.14>1.11>1.10>1.06)。普通话10位发音人四种焦点词停延率的排序为：句末焦点>句首焦点>句中焦点>自然焦点(1.10>1.08>1.04>1.03)。香港粤语和普通话四种焦点句焦点词时长的顺序相同,证明同类型焦点句具有相同的韵律特征。香港粤语和普通话强调焦点句焦点词都发生了延长,同类焦点句焦点词延长的百分比顺序相同。更多还原

【Abstract】 Based on the Theory of Intonation Patterns, deep and thorough research was conducted on the sound materials (including three sentence types and five moods, total540samples) of Hong Kong Cantonese intonation, as spoken by four native speakers. This was done through calculating the acoustics parameter pitch and duration of the final syllable, using the undulating scale and lengthening rate. The different intonation hierarchies of Hong Kong Cantonese were investigated in the paper according to the view on intonational hierarchy put forward by Shi Feng (2012). This research is based on the Intonation Patterns, using comparisions of intonational hierarchy to conduct. It included:the declarative sentence (natural focused sentence), which was the basic pattern; unmarked interrogative sentence; and focused sentence——both of which were varied patterns. These three types of sentences (declarative, unmarked interrogative and focused) were used for investigation and analyze, after that were compared with corresponding intonations of sentence patterns in Standard Mandarin. The comparative analysis between Standard Mandarin and Hong Kong Cantonese revealed their similarities and differences in universal intonation traits.The following new ideas were brought forth in the paper:a) In the declarative sentence, the final syllable was the most important carrier of the declarative mood. Because of the declination and down step, the whole sentence showed the pitch dropping from the beginning to the end. The final syllable represented the mood and prosodic rhythm by the boundary tone lowering and range expansion. The data suggested that the ranges of sentence, phase group, and syllable in Standard Mandarin were all wider than those of Hong Kong Cantonese. The Maximum values of both ranges were all at final phrasal and final syllable of the last character.b) Average lengthening rates of declarative sentences composed by Shusheng decreased in the following order:final phrasal group> initial phrasal group> middle phrasal group (1.30>1.06>1.00). Compared with the average lengthening rate, the lengthening rates of the initial and final phrasal groups increased by6%and30%, respectively. The lengthening rates of middle phrasal group did not increase. Average lengthening rates of male and female subjects was as follows: final phrasal group> initial phrasal group> middle phrasal group. Male subjects’ lengthening rates of the final and initial phrasal groups were24%and6%above the overall average lengthening rate. By comparison, female subjects’lengthening rates increased by36%and6%above the average. Neither did lengthening rates of middle phrasal group increase. The intonations of Hong Kong Cantonese were characterized by rising pitch and range expansion of the final syllable of final phrasal group, and duration of the final syllable.c) The unmarked interrogative sentence expressed its interrogative mood by the rising pitch, range expansion, and duration of the final syllable. This echoed the viewpoint raised by Yuen Ren Chao, that the rising ending had no effect on the whole sentence intonation pattern, but only on the voiced part of the last syllable. The pitch and the contour of the last syllable would be adjusted by the interrogative intonation. If it were a rising tone, the former part would overlap with the intonation, and the latter part would rise with the higher gradient. The pitch rising did not begin from the final syllable, but from the8th and9th syllables. It corresponded to the regularity of physiology. Moreover, the rising extent of the interrogative intonation varied with the different tones. The pitch of male subjects’final syllable (as composed by Shusheng) increased by43%relative to the adjacent syllable on the left, and for female subjects increased by49.5%. The low tone of the final syllable increased much more than the high tone.The duration of the final syllable was one of the important features of interrogative sentences. As for the6final syllables of Shusheng’s sentences, the lengthening rates ranged from1.39to1.69, making them longer than the average lengthening rate. The lengthening rate of final syllable of Shusheng’s sentences extended by0.54over those adjoined syllables on the left. The absolute duration of the final syllable of the interrogative sentence extended to132ms, which was longer than that of the corresponded declarative sentences of57ms. d) By comparing the natural focused sentence (declarative) with focused sentences with initial, middle and final focuses, of Hong Kong Cantonese and Standard Mandarin, several discoveries were made. In Hong Kong Cantonese, the order of the pitch range of the four kinds of focused sentences was:final focused sentence> initial focused sentence> natural focused sentence> middle focused sentence (84>75>71>66). In Standard Mandarin, the order of the pitch range of the four kinds of focused sentences was:initial focused sentence> middle focused sentence> final focused sentence> natural focused sentence (94>89>77>61). In comparing Standard Mandarin with Hong Kong Cantonese, there were two main differences:(1) For Standard Mandarin, the initial focus was the widest, and that of the natural focus was the narrowest; but for Hong Kong Cantonese, the final focus was the widest, and the middle focus was the narrowest. The sentence ranges of the natural focus, and final focus in Hong Kong Cantonese were both wider than those in Standard Mandarin.(2) In Standard Mandarin, the ranges of the initial focus, and middle focus were the widest, and the pitch after the focus dropped suddenly, so the range after the focus compressed significantly. This feature was the result of contact with Altai language family, including the Manchu language in Mongolia. This phenomenon is called PFD (post-focus depression) and could be embodied in different language types, though it didn’t exist in Taiwanese Mandarin, the Min dialect, or other dialects and languages.The comparison of lengthening rates in natural focus, initial focus, middle focus, and final focus between Standard Mandarin and Hong Kong Cantonese revealed a pattern. In both dialects, the order of the four kinds of focused sentence was the same: final focus>initial focus>middle focus> natural focus, proving that they have the same prosodic pattern.更多还原