古文字特殊諧聲研究

A Study of Phonologically Irregular Xiesheng Series in Old Chinese

金俊秀(Jun-soo Kim)

指導教授 ： 季旭昇

國立臺灣師範大學/文學院/國文學系/博士(2011年)

摘要

上個世紀，古文字學和漢語音韻學都取得了輝煌的成就。古文字學的突飛猛進，主要仰賴甲骨、楚簡等地下材料的大量出土；漢語音韻學的飛速發展，新動力來自西方，其歷史比較語言學的理論及音系描寫方法，展開了一個新的研究局面。但同時必須要指出，無論是新的出土材料還是新的語言學方法論，之所以能夠迅速發揮其作用，是因為以乾嘉學派為代表的傳統小學方面，已累積相當厚實的實力之故。古文字學是在傳統《說文》學的土壤上盛開的奇葩，漢語音韻學是在傳統古音學的土壤上扎根的異草。本論文的宗旨，主要在於這「奇葩」和「異草」的稼接。

本論文的內容分為四章：第一章「緒論」首先提出本文的研究背景與動機，接著討論諧聲字的時空問題及漢語音韻史的分期。本文認為先秦傳世典籍及出土文獻所反映的，大抵皆為雅言音系。先秦雖已有方言分歧，但諧聲字所反映的仍是內部統一的古雅言音系。在古代，受過嚴格訓練的少數人才識字、造字，諧聲字當然是依雅言所造。即便如此，由於周秦時期包括長達九百多年的時間，我們不能將所有先秦諧聲字當作是一個平面材料進行研究，而要分辨其早晚。這一點僅據《說文》無法做到，必須適當運用古文字材料，才可以在一定程度上得到解決。至於漢語音韻史的分期，上古音部分本文採取三分法，即分為「諧聲時代」、「《詩經》時代」、「兩漢」；中古音部分採取二分法，即分為「《切韻》時代」與「韻圖時代」。

第二章「相關研究回顧」述評高本漢以來的上古聲母研究的發展情況。第一階段以高本漢、王力為代表，古音學界對高氏的中古擬音系統有高度評價，但一般對其上古音體系貶多褒少，王力體系的上古聲母則幾乎是中古的翻版。較合理的上古聲母系統應當是從第二階段李方桂體系開始，第三階段鄭張尚芳體系則是對李方桂體系的批判地繼承。由於本論文採用鄭張體系，因此花費比較多的篇幅討論其上古聲母系統，間或補充古文字方面的證據。

第三章「特殊諧聲字例考釋」共六節。第一節「「悤 」、「兇」諧聲考」憑藉楚簡所見「 」與「凶」互用之現象，討論上古有諧聲關係的中古清母和曉母的古音來源。第二節「井聲系字諧聲考」討論「創」、「梁 / 粱」、「量 / 糧」、「青」、「靜」、「爭」、「耕」等字的字形來源及古音構擬問題。第三節「「迹」字諧聲考」論證「 」改作「迹」並非單純的形訛，而兼具聲化作用，並對〈古音字表〉狄聲系的古音構擬提出修改建議。第四節「「允」、「夋」、「舜」諧聲考」擬測「允」、「夋」、「舜」三字的古音分化原委，並述評李方桂、包擬古、鄭張尚芳的書母來源研究。第五節「「回」、「亘」古音分化考」擬測「回」、「亘」二字的古音分化原委，並述評龔煌城、鄭張尚芳的眞、文、元部韻尾研究。第六節「「賓」字諧聲考」討論「賓」字和「矜」字的字形來源及古音構擬，並討論令聲系字的上古歸部問題。

第四章「總結」針對第三章所進行的研究做一總結。

關鍵字

古文字 ； 上古音 ； 諧聲 ； 鄭張體系

並列摘要

During the previous century, the fields of Chinese Paleography (the study of the origins and evolution of Chinese writing) and Historical Chinese Phonology saw progress by leaps and bounds.
Advancements made in the field of Chinese Paleography were primarily based upon large quantities of newly excavated oracle bones from the Shāng Dynasty as well as bamboo texts from the State of Chǔ of China's Warring States period, but the impetus which rapidly pushed the field of Historical Chinese Phonology forward and into a new era originated in the West with the advent of modern linguistic methods for phonological description and the Comparative Method from the field of historical linguistics. At the same time, however, it needs to be pointed out that regardless of any newly excavated materials or new linguistic theories and methods, one of the key factors in this rapid development was the academic foundation inherited from pre-modern, traditional Chinese philology, the so-called "Minor Studies" (xiǎoxué 小學), including exegesis of ancient texts, phonology and Chinese paleography, most aptly represented by the scholarship of the Qián-Jiā 乾嘉school. While modern Chinese paleography finds its roots in the traditional studies of the Shuōwén Jiězì 說文解字, contemporary Historical Chinese Phonology finds its in traditional Chinese philology. One of the major aims of the research presented here is to build upon the foundation laid by these great scholars.

This work is divided into four chapters: Chapter 1, Introduction, starts by describing the background and motivation for this study and then discusses issues with xiéshēng諧聲 series arising from phonological change over time (diachronic phonological variation) and distance (synchronic dialectical variation). The chapter closes by delineating the various phonological stages in the history of Chinese.
The language appearing in both transmitted as well as excavated classical Chinese texts of the pre-Qín period represents a type of standard, educated speech. Although there were already dialects in pre-Qín times, the phonology exhibited in the xiéshēng series is nonetheless a form of internally unified and standard educated speech. During the Old Chinese period, both literacy and the creation of Chinese characters were limited to a small number of rigorously trained scholars, so it is only natural that the phonology of xiéshēng characters reflects this. Even so, because the Zhōu-Qín period encompasses a time frame of more than 900 years, research of pre-Qín xiéshēng characters should not be based upon the view that they represent a single, uniform and unchanging phonological system. Indeed, this time period should be subdivided into an early period and a late period to accommodate phonological change and determining where to make this division cannot be accomplished by reliance upon the ShuōWén Jiězì alone. Only by making proper use of excavated materials from this period can this issue be resolved with any measure of accuracy.
Phonological stages in the history of Chinese as used in this study are as follows: the Old Chinese period is further divided into three sub-stages: the Xiéshēng period, the Shījīng 詩經 period and the Hàn Dynasty period; while the Middle Chinese period is divided into two sub-stages: the Qièyùn 切韻 period and the Rhyme Table period.

Chapter 2, Research Review, gives an account of the stages of development in the research of Old Chinese initials starting with the work of Bernhard Karlgren, who along with Wáng Lì represent Stage one research. Those within the academic field of historical Chinese phonology generally have a high opinion of Karlgren's reconstruction of Middle Chinese, but assessments of his system for Old Chinese are largely negative. Wáng Lì's Old Chinese initials are a virtual copy of his Middle Chinese initials. More reasonable systems of Old Chinese initials start with stage two scholar Lǐ Fāngguì (aka Li Fang-kuei). The system of stage three scholar Zhèng-Zhāng Shàngfāng can be viewed as a continuation of Lǐ Fāngguì's system and since Zhèng-Zhāng's system is used in the present study, it will be described in detail. From time to time, evidence from Chinese paleography will also be provided to supplement the discussion.

Chapter 3, Investigating Irregular Xiéshēng Characters, consists of six sections. Section 1, Investigating the Xiéshēng Relationship between 悤 cōng and 兇 xiōng, discusses the phonological origins of the contact in Old Chinese xiéshēng series of the Middle Chinese initials 清母 qīngmǔ and 曉母 xiǎomǔ based upon the interactions of cōng and 兇 xiōng in bamboo texts from the State of Chǔ. Section 2, Investigating Xiéshēng Relationships of the 井 jǐng Series, discusses the origins of the character forms for characters such as 創 chuàng, 量/糧 liáng, 青 qīng, 靜 jìng, 爭 zhēng, 耕gēng, etc. and some related issues regarding phonological reconstruction. Section 3, Investigating Xiéshēng Relationships of the 迹 jī Series, gives evidence that the evolution of the character form jī into 迹 jī was not simply a matter of graphic corruption, but that this corruption was phonologically motivated. Additionally, some suggestions for improving the Old Chinese reconstruction of the the 狄 dí series in the Gǔyīn Zìbiǎo 古音字表 are offered. Section 4, Investigating Xiéshēng Relationships of 允 yǔn, 夋 qūn and 舜 shùn, reconstructs the phonological divergence of 允 yǔn, 夋 qūn and 舜 shùn and reviews the research of Lǐ Fāngguì, Nicholas Bodman and Zhèng-Zhāng Shàngfāng into the phonological origin of the Middle Chinese initial 書母 shūmǔ. Section 6, Investigating the the Phonological Divergence of 回 huí and 亘 xuān in Old Chinese, reconstructs the divergence of 回 huí and 亘 xuān during the Old Chinese period and reviews the research of Gōng Huángchéng (aka Gong Hwang-cherng) and Zhèng-Zhāng Shàngfāng into the codas (syllable endings) for the Old Chinese rhyme categories 眞 zhēn, 文 wén and 元 yuán. Section 6, Investigating the Xiéshēng Series for 賓 bīn, discusses the origins of the character forms for 賓 bīn and 矜 jīn as well as their phonological reconstructions. Additionally, there is discussion regarding which Old Chinese rhyme category the 令 lìng series should be placed in.

Chapter 4, Conclusion, summarizes the research conducted in Chapter 3.

並列關鍵字

ancient Chinese characters ； Old Chinese phonology ； xiesheng ； Zheng-Zhang Shangfang's Reconstruction of Old Chinese

參考文獻

1995〈重紐問題在日本〉《第四屆國際暨第十三屆全國聲韻學學術研討會議論文

1933〈Certain Phonetic Influences of the Tibetan Prefixes upon the Root Initials〉《中央研究院歷史語言研究所集刊》第4本第2分，1933年。馮蒸譯〈藏文前綴音對於聲母的影響〉《境外漢語音韻學論文選》，上海：上海敎育出版社，2010年10月第1版。

2007《上海博物館藏戰國楚竹書（一～五）文字編》，北京：作家出版社，2007

2003〈上博簡〈緇衣〉篇「 」字解〉《華學》第6輯，北京：紫禁城出版社，2003年6月。

六屆國際音韻學學術討論會論文集》，2000年。