Chapter 9 Introduction to Collected Essays on Dunhuang Pipa Scores《敦煌琵琶譜論集》序言

In: Harmoniousness: Essays in Chinese Musicology
Jao Tsung-i
Search for other papers by Jao Tsung-i in
Current site
Google Scholar
Editor / Translator:
Colin Huehns
Search for other papers by Colin Huehns in
Current site
Google Scholar
Open Access


Deciphering Dunhuang manuscripts is a recurring theme in Jao Tsung-i’s work, and this essay concentrates on dance notations and their relationship to music. To give a context, included here are also three introductions, the third of which is a moving account of the death of former student Kuang Qinghuan 鄺慶歡. Jao’s research methodology is to lay out all the relevant sections of text and then to proceed through detailed critical comparisons of lexical repetition and organisation to extract data. This he puts into a framework of citations from related material, and without falling into the trap of over-interpreting, comes to a finely-honed explanation of what these notations mean.

In the Dunhuang manuscripts are found three musical scores and of these the most important is a damaged juan folio now held in Paris under the catalogue number Pelliot 3808. On one of its faces is written: ‘In the fourth year of the Changxing era (933), in response to an invitation to lecture on scriptural writings on the official royal birthday.’ 長興四年應聖節講經文. Since 1938, when our Japanese friend Hayashi Kenzō (林謙三, 1899–1976) made an investigative comparison of this manuscript with a Japanese pipa (biwa) piece of the Tenpyō era (天平, 729–749) and confirmed that it was indeed a pipa score, fifty years have now passed. In Monthly Publication of Musical Scores (Gekkan gakufu 月刊楽譜), juan 27, issue 1, he published a long essay on the subject,1 and last year, Li Ruiqing 李鋭清 and I finally made a joint translation of it into Chinese, which was published in Music and Art (Yinyue yishu 音樂藝術) in Shanghai (1987, issue 2),2 and at last Chinese people have an opportunity to read the essay in its entirety.

In 1982, when Ye Dong 葉棟 first penned his thesis, it was hailed as a transcription that had finally cracked the code, and after the circulation of his Research into Dunhuang Musical Scores (Dunhuang qupu yanjiu 敦煌曲譜研究), those who discussed it increased day by day, and other transcriptions appeared, including versions by He Changlin 何昌林 and Jin Jianmin 金建民. In June 1987 at the Hong Kong-hosted International Dunhuang Turpan Academic Conference (Guoji Dunhuang Tulufan xueshu huiyi 國際敦煌吐魯番學術會議), I presented a paper ‘Dunhuang yuepu wupu youguan zhu wenti’ 敦煌樂譜舞譜有關諸問題 that stressed the indivisible relationship between musical and dance scores and the impossibility of jettisoning dance and simply discussing music, given that the pieces in the Dunhuang musical scores were mostly composed to accompany the imbibing of alcoholic beverages; their role was the same as that of vernacular drinking songs, and the rhythms of the pipa pieces were in fact matched to dance steps. In the past, when experts researched into music and dance, it was a case of ‘ne’er the twain shall meet’, a taking of separate pathways that never come to the proverbial meeting point, which is a most unwise stance to adopt and a tendency that requires urgent correction.

As far as the rhythmic aspect of these musical scores is concerned, previous interpretations have plagiarised one other and merely adopted a metrical explanation comprising strong and weak beats. When these various scholars have transcribed the scores, they have simply stopped work once the pitches are recorded, even though the musical results are most unpleasing to the ear. Because earlier transcribers had been insufficiently faithful and often added or removed items at whim, there was no choice but to make a fresh transcription or no worthwhile results would have been achieved, so I invited Prof. Chen Yingshi 陳應時 of the Shanghai Music Conservatoire to have another attempt. At that time, Zhao Xiaosheng 趙曉生 had come up with a novel suggestion regarding the punctuation of musical phrases that corrected the omissions and errors of earlier authorities, and which viewed the symbol · as a short break in the musical line and the symbol as a longer one. Prof. Chen, working according to Zhang Yan’s (張炎, 1248–1320) The Origin of Words (Ciyuan 詞源) combined with Mengxi bitan 夢溪筆談 (by Shen Kuo 沈括, 1031–1095) formulated a new theory of ‘pulled beats’ (chepai 掣拍). His freshly transcribed musical score, only recently finished, was sent to me for early perusal; on looking at the new transcription, three important points are worthy of note:

  1. Through close examination and checking of the characters employed in the notation, in many places authoritative correction has been implemented.

  2. The mysterious secret of multiple occurrences of the symbols and · has been unravelled and solved.

  3. Analysis of six characters to a beat and three characters to a beat in sha ( ‘broken’) phrases and comparisons between recurrences of the same musical mode have led to the discovery of melody produced by notational characters within a musical mode, which could be regarded as a breakthrough in terms of research into Dunhuang musical scores in recent years.

Regarding the original meaning of the two symbols and · and their evolution in subsequent usage, they have in fact a relationship to the music of lyric song (ci ) as intimate and mutually dependent as water has to milk. I therefore took Jiang Baishi’s (姜白石; also called Jiang Kui 姜夔, 1154–1221) ‘side-by-side notation’ (pangpu 旁譜) and Zhang Yan’s ‘tablature notation using characters of a pipe’s timbral features in answer to fingering requirements’ 管色應指字譜 as well as Shilin guangji 事林廣記 (by Chen Yuanjing 陳元靚, fl. thirteenth century) and the symbols that they use to notate rhythm: (‘ding’), (‘zhu’), (‘zhe’), (‘che’), and so on, and compared them with those in the Dunhuang scores and realised that the two practices had in fact evolved from a common source, and the symbol · was fundamentally an abbreviated graph of (); · had originally been and later when used became , just as in (Jiang) Baishi’s ‘side-by-side notation’ is also a symbol for ; the character that Zhang Yan employs is , and its usage is so alike that it is as if they had emerged from the same mould; this evidence is furnished here for thoughtful minds as sufficient for the sake of providing an argument.

Here, I propose to take Prof. Chen’s transcriptions and my own inevitably clumsy remarks and combine them in a single string-bound volume accompanied by excellent illustrations that should be sufficient for the time being to provide interested parties with food for thought, nonetheless still seeking instruction and correction from the elegant circle of experts in the hope that together we can later make a deeper investigation.

February 1988, written at the Chinese University of Hong Kong, Institute of Chinese Studies (Zhongguo wenhua yanjiusuo 中國文化研究所), the dingmao 丁卯 year of the old calendar, the night before New Year and the Spring Festival.

Although a metrical explanation comprising strong and weak beats came into being in a later epoch, from the late Tang dynasty onwards, instrumental scores such as Dunhuang Pipa Scores (Dunhuang pipa pu 敦煌琵琶譜) all had the two symbols and · as their backbone, and if the origin of this system is tracked to its source, it is in fact the Han dynasty system of sentence breaks. is the symbol for the end of a passage and can be traced back as distantly as the Warring States period. After the Song dynasty, evolved and became ○ and from an oblong turned into a circle, and the symbols ○ and · used respectively to represent a break and a pause thereupon became the means to mark out the fundamental metrical arrangement of sung genres. Still more when it came to melismatic styles, which are even more plentifully furnished with helpful punctuating symbols, for example, (‘zhe’), (‘zhuai’), (‘ding’), and (‘zhu’), but the cart-tracks from which they had sprung had already revealed their traces in the Dunhuang pipa scores, and the practice was passed down into scores of the ‘drum-(blow) music’ (Gu [chui] yue 鼓 [吹]樂) genre of Xi’an such that they appear to be inherited from the same bloodline.

Most of the old rules and practice of musical notation can be used as corroborating evidence to demonstrate mutual veracity. In the winter of 1988, I went to Xi’an and with the assistance of Mr Li Shigen 李石根 had the privilege to view old scores of the ‘drum-(blow) music’ genre in their original state. The characters used for the notation are in fact vernacular versions of gongche 工尺 characters, and indication of the metrical arrangement is given primarily using the symbols ○ and · (x), which also have the additional functions of serving as marks of stress and coda, and this practice and that of Dunhuang Pipa Scores are actually not of two different kinds. Immersed in the continuance of writing essays and in order to clear up ignorance and misunderstanding, for the time being I can only record the process of discovery I have taken. The rationale of any scholarly research is that if it does not proceed by painstaking accumulation drop by drop, how then can comprehension of a wider flow of understanding be acquired? I dare not yet lay claim even to a brittle belief that I am right, but still hope that those of broad knowledge and sound scholarship will seek to organize and improve my work.

1988, the night before New Year, another note.

As an addition in the second chapter, this book includes investigation of the original manuscript of the pipa score, and in this respect I have availed myself of the assistance of my student in Paris Dr Kuang Qinghuan 鄺慶歡, who made a tracing of the whole manuscript and sent it to me; through this, issues regarding the dating of the manuscript have been solved, and it has become clear that the musical notation was written first, so the fourth year of the Changxing era (933) is the latest possible date for its composition. The manuscript as copied by Dr Kuang looks as if the ink is still fresh on it, and I have brought it out many times to show others as a way of rewarding her labours on its behalf. Recently, however, the terrible news of her death has come; because of a long-standing illness suddenly she could no longer rise from her sickbed and in middle age has passed away, an orchid fragrance now damaged; in memory of her, I could not help but gaze vacantly for many days, barely holding back my sobs.

A memorial service was held for her in Paris, and my eulogy poem on her death reads: ‘I remember I once asked of her the meaning of the characters “mysterious pavilion”; and now all that is left behind is a remnant manuscript on which to shed my tears; awaiting the summoning of her soul at the high green tomb of Wang Zhaojun (54 BCE–19 CE); yet fearing I will hear emana- ting from it the dark distant plaint of a pipa.’ 記問字玄亭,剩有遣箋供涕淚,待招魂青塚,怕聞幽怨出琵琶. She was highly regarded for her research into vernacular ‘transformational texts’ (bianwen 變文; also translated as ‘changed literature’) pertaining to Wang Zhaojun and once went with me to pay a visit to Zhaojun’s tomb in Inner Mongolia. Unexpectedly, boundless Heaven has exterminated my fine and beautiful friend, so I will use this book Dunhuang Pipa Scores to commemorate her for evermore.

7 July 1990, after editing, once more, another note.


Hayashi Kenzō, Hiraide Hisao 平出久雄, ‘Pipa ko fu no kenkyū’, 23–58.


‘Pipa gupu zhi yanjiu: Tianping Dunhuang erpu shijie’ I, 1–15.

  • Collapse
  • Expand


All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 19 19 6
PDF Views & Downloads 22 22 5