T of C 
Qin as
Qin in
/ Song
Analysis History Ideo-
Personal email me search me
Qin poetry     Gu Yuan     17 Ci Songs     10 Ritual Songs 網站目錄
Songs of the Whitestone Daoist
- A collection of songs and song lyrics by Jiang Kui 2
白石道人歌曲 1
Baishi Daoren Gequ, 姜夔歌曲集  
  A page explaining different forms of music notation 3    
Jiang Kui (ca. 1155 - 1221), introduced separately, was a major poet, music "composer/creator" and critic. However, although his songs apparently received high praise in his day, it is not certain how much the music actually circulated even at that time; and amongst those who did play from his notation, one can also speculate as to what extent these were treated as "creations" and to what extent they were treated as "compositions".4 In fact, there seems to be no record of them having been played outside his own presence, at least not before attempts were made beginning in the 18th century to revive them, a task that therefore had to be based largely if not completely on the written documents, not on living examples.5 Now a number of studies of his work have been published and, for at least some of his songs, performances have been available for some time of the lyrics together with interpretations of the melodies.6

It seems particularly worthwhile to study whether the rhythms suggested in the tablature for Jiang Kui's qin melody Gu Yuan can be used to gain a better understanding of the rhythmic style in Jiang Kui's other melodies. Whether or not they were guided by such details or perhaps simply inpired by Gu Yuan, some players today have adapted for qin (or sung with qin accompaniment) some of Jiang Kui's ci songs.7

My own work in this area8 began with an interest in seeing whether music in the qin tradition could shed light on other early Chinese music traditions, particularly the Tang dynasty music preserved in Japan.9 Most of the information on this page was put here to help consider the possibility of applying this interest to doing my own interpretations of Jiang Kui's songs. However, much of the detail came from material very difficult to understand, and any conclusions given here are very tentative, though hopefully worthy of pursuing further as well as being challenged.10

To begin a practical exploration of rhythmic possibilities within Jiang Kui's scores I did my own transcriptions of all 17 of his ci songs (and made midi files of the same). As with my qin transcriptions, all are in staff notation but treat the notes as though they are solfeggio (as singers might have done). The notes follow Pian (also consulting Picken) except that for purposes of comparison they are all transposed to scales without accidentals and the rhythms are my own (though see #7). Under the notes I have paired the original lyrics, their romanization and, for most of them, a translation adapted from Picken; sometimes there are added comparative comments. Since doing my initial versions of all 17 ci songs I have periodically made revisions, trying to make them more memorable musically without violating the written texts as I understand them. I have also made midi versions of each transcription, and listening to these, as well as hearing them sung or played on other instruments, has provided further inspiration for these revisions.11

For all of these tentative ci song reconstructions reconstructions I have written out staff notation; for five of them I have devised qin tablature. Recordings of these five adaptations for qin can be found by following these links:12

Having done this, and found the melodies interesting but difficult to adapt for qin idiom without adding notes, I have since then worked basically with the midi files, mainly in order to use these to help imagine how they might have sounded, but also to be able to make them available to people who might also have an interest in exploring this music. It would be especially good if flautists (especially players of xiao or di, but not limited to that) would play - and adapt - them (more under treatment and intended results).

In all cases the rhythms follow my own feelings about the songs, but they are not intended to be prescriptive any more than I think Jiang Kui's own notation was intended to be prescriptive: these were creations, not compositions. My understanding of traditional Chinese music, other than perhaps ritual music, is that the results should not be staid and in unison, but a free interplay between instruments (or as here, voice and instrument). When played on a flute or other instrument the melody should also be memorable enough that it can be played without having to look at any notation. 14

A number of musical differences become readily apparent when reconstructing songs from Jiang Kui's notation after doing the same with what seems to be the more complete information about the music in early qin tablature (including Jiang Kui's own qin song Gu Yuan, which I have also recorded). The following outline touches on some of these differences:

  1. Rhythm:15 The biggest difficulty in finding out how these songs might have sounded in Jiang Kui's time is deciding on appropriate rhythms, as the form of notation he used may include some rhythmic indications but it is incomplete information that is not well understood. Of course, the texts of Jiang Kui's songs may themselves also give hints at the rhythm, but here the qin tablature paired with qin song texts generally gives even stronger rhythmic hints through its fingerings and ornamentation.
  2. Mode:16 At present it is too early for me to make my own conclusions about mode, but so far the differences seem to include the following: There are differences of opinion regarding the interpretation of Jiang Kui's note names and I am not an expert in their interpretation. However, I wonder whether the complexity of modal names for the surviving early qin repertoire compared to the regularity of its modal structures might suggest that Jiang Kui's songs, similarly complex in modal terminology, might have been similarly restricted in their actual modality.
  3. Intonation:17 Jiang Kui's ci melodies may have a different intonation from what is played on qin (where this may depend in part on tuning). Such differences are suggested by other reconstructions I have heard, but once again I have not been able to study this issue closely.
  4. Instrumentation (and range):18 With the Jiang Kui songs it is known that he played most of them on flute, but at least once he specified another wind instrument and one can speculate about others that might have been used. As for range, untrained voices typically have a range somewhat more than an octave (compare the octave plus two or three notes available in Jiang Kui's notation system); skilled voices, not to mention flutes, have a much greater range. Is it likely that Jiang Kui would have limited himself to only those notes he could notate?
  5. Syllabic setting:19 As can be seen either from qin tablature paired to lyrics or from transcriptions of the ci songs, both have a largely syllabic setting: one character for each note; if ornamental notes are added (whether directly written in the tablature or not) they add more notes for one syllable, never more syllables for one note.
  6. Repeated notes:20 In the ci songs two notes of the same pitch almost never follow each other, while in the qin songs there are extended passages where several syllables (sometimes as many as seven) are sung to the same repeated note (or octave thereof).

Some of differences in style and intonation seem to originate with the physical nature of the qin itself compared to that of other instruments.21

Although many recents attempts to revive the Jiang Kui songs have diverged greatly from the principles of historically informed performance, this of course does not necessarily make them less appealing.22

What we do know of the songs attributed to Jiang Kui is almost completely based on the materials in Songs of the Whitestone Daoist. According to Prof. Rulan Chao Pian, who transcribed them all in her book Sonq Dynasty Musical Sources and their Interpretation, the official date given for this publication, 1202 (i.e., while Jiang was still alive; other sources have put the dates as 1176-1197), is based on a colophon by Qian Xiwu, but it is not clear in what form they were then originally published.23 She added that the earliest surviving publications were hand-written copies by Tao Zongyi dated 1350 and 1360 (which have not survived), and that the latter then became the basis for later printed versions, of which at least three with the music notation have survived but only from the 18th century.24 Of these latter, I am familiar with two versions, one in six folios,25 the other in four.26 Pian writes that one of the six folio versions seems to be the earliest one.27

In any case, the material in the two collections is almost identical, the main contents being:

The possibility that this was all compiled in 1202, while Jiang Kui was still alive, emphasizes the fact that Jiang Kui may have also done settings for more than the 28 pieces included here. Unfortunately, there is no evidence that any such further songs have survived.

Of most interest here are the 28 pieces with the music included. These are further discussed in a three-part Appendix. The three parts are:

These 28 now seem often to be published separately but under the full title, Songs of the Whitestone Daoist. The fact that they do survive today in written form seems to put them amongst the very few known examples prior to the Qing dynasty where we have lyrics and actual surviving music known to have been created by the same person.

As for the music itself, commentary with the surviving scores suggests that while in many cases Jiang was creating music to fit his own lyrics, in others he was arranging the lyrics to fit melodies that he had heard in various places. In fact, among the 51 ci titles, only the last 13 are directly attributed to him, with the first nine of these said to be "self-done" (zidu) and the last four "self-made" (zizhi); the first 12 of these 13 have surviving music settings. Pian, p.35 says that the terms zidu and zizhi seem to be interchangeable, but does not clarify whether this refers specifically to the music, the words, both or a combination of these. It is also important to note that, although today there have been a number of efforts to reconstruct and perform these songs, and clearly the reprintings just mentioned suggest they have always been highly regarded, it is not clear whether they were actually ever performed outside of Jiang's own circle of friends; and by the time efforts were made in the 18th century to revive them, the actual music was long forgotten (Pian, p. 35).

Although there seems to be no indication that these were intended to be sung without accompaniment by at least one instrument, Jiang only wrote down a single line of music and never gave any indication about how the voice and instrument(s) might or should interact. As for the instruments themselves, and their relation to the vocal line, his preface to the ci song Jue Shao (#13 below) states that he himself always played them on the flute, but his preface to An Xiang and Shu Ying (#10 and #11 below) suggests that he may not have had specific instruments or specific arrangements in mind (translated from Cai Zongqi, How to Read Chinese Poetry, p.287 [Google Books]):

In the winter of the year xinhai, I took a ship through the falling snow to visit Stone Lake (style of Fan Chengda, 1126-1193). After I had stayed for a month, he handed me paper, requesting poetry and new tunes. I composed these two song lyrics, which Stone Lake held, fondling them in his hands, unwilling to put them down. He ordered a musician and singing girl to practice them. The melodies were harmonious and graceful, and he entitled them "Anxiang" (Secret Fragrance) and "Shuying" (Dappled Shadows).

This passage was then followed by the two poems (Cai also included a translation of these in his book). It would be interesting to know whether Fan Chengda was sight reading the songs, or whether Jiang Kui sang the songs himself before or while writing them down. The lyrics of the two songs do not mention the words "an xiang" or "shu ying", so perhaps the titles were inspired by their mention in a poem by Lin Bu. (Fan Chengda himself also wrote a poem about qin called 聞琴 Hearing a Qin).

白石道人歌曲 Baishidaoren Gequ

Details of the 28 songs with lyrics

All 28 of these songs have been translated and transcribed, and most of their respective commentary translated, in three articles by Laurence Picken, as follows:

Pian also transcribes and discusses all 28 songs on pp. 99-129, 147-154 and 173-188. Her transcriptions also do not include the original notation (except with Gu Yuan). In addition, none gives any indications of note values and she does not include any translations.

Modern Chinese publications with transcriptions and analysis of these songs begin with the two works listed here.

The following information has been put together as I go through the songs looking for connections, or possible connections, to the qin.

Pian's book also includes her transcriptions of the music for all of the other songs known to have survived in other collections said to date from the Song dynasty.37

Footnotes (Shorthand references are explained on a separate page)

1. Songs of the Whitestone Daoist (白石道人歌曲 Baishi Daoren Gequ)
23191.129 白石道人歌曲 outlines the contents of this book, in four folios plus an addendum, as follows (.127 is 白石道人詩集; .128 is 白石道人詩說):

The lülü and popular notations tell the basic notes, but do not specify how they were to be performed. For transcriptions and further details see:

Modern Chinese publications with transcriptions and analysis of these songs begin with these two works:

Regarding Pian's transliterations, such as "Sonq" for "Song" (or more specifically for "Sòng" or "Song4"), she is using "Gwoyeu Romatzyh" (Wiki), the "Harvard romanization system" devised largely by her father 趙元任 Zhao Yuanren. It indicates tones through different or extra letters rather than through numbers or diacritical marks.

The present Appendix in particular has detail that extends beyond the aim of reconstructing some of the songs. It was put here in part to help decide which songs to work on first, in part in hopes that it might help interest others in doing this kind of work.

2. 姜夔 Jiang Kui (1155-1221; Wiki; Renditions; silkqin)
James J. Y. Liu began his review of Lin Shuen-fu's book as follows (see JSTOR):

Chiang K'uei 姜夔 (ca. 1155-ca. 1235), poet, composer, musicologist, critic, calligrapher, and connoisseur of art, epitomizes the aesthetic sensibiity of his age....

6335.117 姜夔 begins "(江西)鄱陽人字堯章 from (Jiangxi) Poyang, style name Yaozhang, 號白石 nickname Baishi (White-stone)". See also Nienhauser, ICTCL (Chiang K'uei), p.262ff. QSCB Chapter 6b1-6 says he belonged to the School of Poetic Meter (格律派 Gelü pai, but on this ICTCL 675, 858 are both Ming references), also known as the Delicately Restrained School (婉約派 Wanyue Pai, see ICTCL 263 [Jiang Kui], 327 [same time]).

3. Page explaining different forms of music notation
From a modern publication that includes the 6 volume version; it is only in Chinese. For English there is a basic outline more specific to Jiang Kui's music in this appendix, which includes two charts, a general one for note/pitch symbols and another one for secondary symbols (duration/deflection) in suzi notation.

4. Treatment of Jiang Kui's Songs: "creations" or "compositions"? (the meaning of "作 zuo")
This does not concern value judgements. In theological terms it might be compared to discussions about whether God is a "creator", giving people free will, or a "composer", whereby everything is predestined or preordained.

Pian (p.38) writes the following of Jiang Kui as a "composer":

Jiang Kwei is one of the few known composers of the Sonq period (fn.: "The other two are Shyong Pernglai...and Fann Cherngdah...."). His fondness for experimenting with theoretical modes and his interest in recovering old forgotten music show that his approach to music was an intellectual one...."

This may have been true - especially with regard to the ritual songs. But equally likely could he not have been a creative person who simply enjoyed making music with friends, especially with lady friends? He picked up inspiration from here and there (including old and perhaps exotic sources he may not have completely understood), and this resulted in his musical works, which he created and probably modified while playing. Also having intellectual interests, and perhaps a desire to be accepted at court and/or by respected scholars, he then justified his creations by trying to fit them into modes that they could approve of. This also affected the notes he selected to write down: the one word for one note pairing method was apparently required by tradition, but if Jiang was a typically skilled flautist, his flute version would likely have included many other notes, and in addition he would have "ornamented" the melody with different notes each time he played it.

This latter scenario is never described in Song dynasty sources, but then there are also no descriptions to suggest that Jiang, like modern composers, sat down a desk and (perhaps based on thorough studies of how to compose in the correct musical modes) wrote out melodies in such a way that others could play them just as he wrote them.

In re-creating Jiang's melodies, then, deciding which approach to use is clearly crucial. However, among modern scholars little attention seems to be made to the the possibility of the freer approach. My own prejudice, perhaps influenced by my understanding of methods used in the modern re-creation of early Western music, says that only the latter approach will ever be successful in bringing to life the essence of what Jiang created, but this will only be true to Jiang (i.e., HIP) after considerably more rigorous attempts to understand the Song dynasty musical idiom of that time.

So where does one begin a discussion about whether Jiang created or composed music? As further discussed here, the usual Chinese word, with Jiang Kui and elsewhere, is 作 zuo, in general terms commonly translated as "make", but with music most often translated into English as "compose". The problem with this is that it evokes what might be an incorrect image:

Making this distinction is complicated by the fact that academically minded people, such as many traditional Chinese literati, wished at times to treat written materials as sacrosanct. Often not musicians themselves, they may treat Jiang Kui's symbols that apparently suggest ornamentation and perhaps note length as though Jiang Kui was trying to use these to define exactly how the piece was to be played. The result may be transcriptions/performances all in quarter notes, half notes and whole notes where symbols might indicate so. Nowadays the tendency to do this is encouraged by an acceptance of the Western idea that this is the way "great music" should be. But this is not even typical of Western music outside the Common Practice Period. If applied, for example, to early Western music the results would generally be what people of that time would have considered unmusical performances.

There is a comment with Ci #16 that Jiang Kui ordered a palace musician to play one of his songs. If the musician was supposed to follow the notation written by Jiang Kui this would be an example of a piece that might rightly be called a "composition" rather than a "creation", even if Jiang Kui himself didn't always play it the same. However, my own examination of his opus strongly suggests to me (though it does not prove) that what Jiang Kui (or his contempories) actually wrote down was to be considered as "creation" rather than "composition".

There is a very limited amount of information available to pursue this issue. At present, for example, there are almost no specifics about the extent to which Jiang Kui's contemporaries actually heard his music; as for now my comments are based largely on an absence of such information.

5. Revival of interest in Jiang Kui songs
This is out of my area of expertise and I do not know how to evaluate to what extent early efforts were influenced either positively or negatively by the then modern practices.

6. Performances of Jiang Kui songs
Qin performances are mentioned in the next footnote; as for efforts using other instruments, I have not done a survey of this but one can find a number of online examples. See, however, further comment below about Historically Informed Performances.

7. Playing Jiang Kui ci songs on the qin
Chang Ting Yuan Man as played by 袁中平 Yuan Jung Ping (translated as "Lament of Departure") is widely available on the internet; he has also performed a version of 鬲西梅令 Ge Xi Mei Ling.

As part of a "Concert of Ci Songs and Qin Music From Southern Song Dynasty" (University of Michigan, 2011; video formerly online at Vimeo), four members of the 德愔琴社 Deyin Qin Society of Hong Kong presented 10 of the 17 ci (i.e., skipping #s 4, 5, 9, 12, 15, 16, 17). The video does not identify the names of the pieces or of the performers.

An outline of the video content is as follows (qin solos were by Sou Si-tai [蘇思棣 Su Sidi]):

06.50   1. Shan Ju Yin from 1722 (qin solo)
12.30   2. Yu Ge from 1722 (qin solo; ends at 25.20)
27.00   3. Gu Yuan (qin and female voice)
31.40   4. Ci #2 Xinghua Tian Ying (xiao, qin, male voice, percussion [paiban and drum]; sung twice);
34.50   5. Ci #14 Zhi Shao (for same; once through)
38.45   6. Ci #8 Dan Huang Liu (for same)
41.30   7. Ci #1 Ge Xi Mei Ling (singer and percussion switch roles)
43.30   8. Ci #3 Zui Yin Shang Xiaopin (same)
45.45   9. Ci #13 Jue Shao (same)
49.20       Break
50.45 10. Ci #11 Shu Ying (switch back to male voice)
54.25 11. Ci #6 Yangzhou Man (same)
57.50 12. Ci #7 Chang Ting Yuan Man (switch to female voice again)
61.10 13. Ci #10 An Xiang (same)
65.00 14. Fan Canglang from 1425 (qin solo)
70.00 15. Xiao Xiang Shui Yun not from 1425 (qin solo)
82.50       Yang Guan Sandie from 1864 (encore by the ensemble)

Some of these have been extracted and put on YouTube without visuals or credit.

8. My work with Jiang Kui songs
In doing these reconstructions I have tried to explore how understanding the early surviving qin repertoire might provide some clues, and my approach here is more that of a musician than a musicologist. This has not involved original work on the texts themselves: my note selection almost completely follows what is on pp. 101-122 of Pian (at the same time making comparisons with Picken) but transposing them and giving them rhythmic values in accordance with the following guidelines:

There is more regarding this way of interpreting mode here and more regarding such rhythmic interpretation here.

One characteristic fairly easy to see in a number of the ci songs is that many have two verses, with the second verse having music quite similar to that of the first verse. The structural similarity is emphasized in several cases in which the second verse begins with two characters that might stand out as connectors. In Picken's translations this can be seen clearly in ci 17, where the words "That place" are on a separate line at the beginning of the second verse. The same structure seems to apply to the beginning of the second verse of ci numbers 5, 7, 10, 11, 13 and 14 as well as 17.

In my own reconstructions I have tried to use this characteristic. However, this is quite speculative since as yet I have not found any critical comment on this and other people reconstructing these melodies generally have not given it special attention.

9. Inspiration from the Cambridge Tang dynasty music research project
The Cambridge project has re-interpreted gagaku (the Japanese court music tradition) in a way that shows its connection to Tang dynasty Chinese music. Since learning of this I have thought it would be essential to compare that music with music of the qin tradition, a tradition which claims much of its origins in the Tang. However, I have found the writings from the Cambridge project very difficult to follow. In addition, the general attitude seems to be that the court music tradition and the literati music tradition (i.e., qin music) have little in common.

Could studying the musical works of Jiang Kui be an intermediate step, in particular as Jiang Kui himself created some music written in qin tablature, some in other forms of notation? Most of the information on this page started as background information that I thought might help me achieve the aim of finding out whether Jiang Kui ci songs can be made to sound natural on the qin. To my ears, Yuan Jung Ping has proven that if the accompanying instrument is free to add appropriate addition notes, then the sung melody can be beautiful indeed.

10. Difficulties with Pian's Sonq Musical Sources and their Interpretation
Pian's book, based on her Ph.D. thesis, is a marvelous and essential groundbreaking work. However, it is in many places needlessly difficult to follow: she needed a good editor. This is not so much because of the romanization system used as because of the way the book is organized, with specific topics and many explanations difficult to find. Some examples:

The 2003 copy is basically a reprinting.

11. Transcriptions with midi files
As mentioned above, the transcriptions are all tentative but can be made available to interested parties, especially musicians who would like to explore playing them. Transcribing all the melodies in the scale 1 2 3 4 5 6 7, i.e., without accidentals, facilitates one sort of analysis. Transcribing them all with 1 as the tonal center (see below) would facilitate another sort. The transcription program allows them easily to be transposed, whether for the purposes of playing them on other instruments or in order to compare the results with Jiang Kui's own terminology.

It must be emphasized that this is an ongoing process and people with an interest in it are welcome to contact me directly: simply putting the transcriptions and midi files online might suggest that they are considered more than experimental. A major advantage of these files is that they can easily be tranposed up or down to any range or key (there are computer programs for transposing midi files and also for making transcriptions from them). Meanwhile, here is an example (download, then open in separate windows to hear the music together with the transcription):

Another example is linked below.

12. My qin recordings of Jiang Kui ci songs
In all of my qin recordings of Jiang Kui ci I begin by doing a glissando on the seven qin strings to show their relative tuning. Perhaps in future I will re-do this, adding notes in a manner such as that discussed below.

14. Intended results of these reconstuctions
As suggested above, interpreting Jiang Kui's songs begins with deciding whether they are "compositions", in which case the performer is expected to change as little as possible of what was written (unless following clear guidelines from the tradition, presumably ones known to the "composer"), or "creations", where the essence of a successful performance requires much input from the performer. "Composer" is, to my understanding, a concept that originated in the West around the 18th century, and my current feeling is that the best approach to recovering the songs is to treat them as "creations", learn as much as possible from the scores of all of them, then try to expand on this based on the understanding gained by this and other research.

The approach taken by Yuan Jung Ping (further below) in adapting ci # 7 Chang Ting Yuan Man for qin has been to add the sort of notes necessary to allow the song and its melody to sound natural within the qin idiom. An understanding of this can perhaps best be seen by downloading the following files then looking at the transcription while listening to this YouTube of Jung Ping's performance; this further comment concerns the changed 4# to 4, clearly heard on the midi file).

Jung Ping is a master of traditional qin playing style. Perhaps a master of the xiao and/or di (or perhaps another bamboo flute) who has been able to learn something about differences between current and earlier styles could do a similarly informed adaptation of these melodies - or at least have fun trying.

15. Rhythm of Jiang Kui ci songs
It is generally thought that some of the extra symbols with Jiang Kui songs concern rhythm, some concern ornamentation. It has also been said that is not always clear which are which, but some transcriptions try to take them into account anyway. Ornamentation in qin tablature is a major factor in selecting note values, so Jiang Kui's symbols must be similarly important. However, with qin music there certainly are notes without ornamentation that are held a long time, and one always has to consider the possibility that sometimes the ornamentation was written down because these were places where it was not expected, not just where it was most important.

As discussed elsewhere on this page, my own understanding of the rhythms is that they are fundamentally quite regular but they are also often so freely interpreted that listeners may suspect there is no rhythm, though they will be able to sense that the music does have form.

16. Modes in Jiang Kui ci songs
The way modes are discussed here is based on treating the scale 1 2 3 4 5 6 7 as the Chinese equivalent of the relative pitches do re mi fa so la ti. This, however, is simply an analytical tool, and not the same as assuming that Jiang Kui's symbols would have been intended in this way. In fact, from what Pian writes it seems quite possible that the scale 1 2 3 4# 5 6 7 (and/or perhaps 1 2 3 4 5 6 7b?) might also or instead have been at one time considered the equivalent of do re mi fa so la ti. I do not know whether there is any disagreement about this in interpreting Jiang Kui's use of the "popular notation", or whether at that time singers ever sang solfeggio, but clearly if one were to select the wrong type of scale for a selected mode/melody it would in turn make the interpretation of it wrong as well.

It is important to consider the requirements of "solfeggio" because, from my experience, notating the pitches in a way that avoids most accidentals gives notes the same names as tend to be used by singers who can naturally sing solfeggio. I have not seen studies of this, but from my understanding many traditional Chinese singers can or could do this. As a result, when transcribing into staff notation my own reconstructions from qin tablature, I have not treated "A" as a fixed pitch (specifically the modern Western 440 Hz) but as a flexible "6/la"). Instead, with my transcriptions one should always consider the notes C D E... as though they are the relative pitches do re me....

The relative pitches in my transcriptions are the same as in Pian but they are transcribed based on the scale 1 2 3 4 5 6 7 the 17 ci songs of Jiang Kui have the following tonal centers (with indications where Picken is different):

Ci  1 : 2
Ci  2 : 5 (Picken: 2)
Ci  3 : 5
Ci  4 : 6
Ci  5 : 5
Ci  6 : 4
Ci  7 : 4 (hexatonic: 1 2 4 5 6 7)
Ci  8 : 2
Ci  9 : 5
Ci10 : 4
Ci11 : 4
Ci12 : 4
Ci13 : 6
Ci14 : 1
Ci15 : 5 (Picken hexatonic: 2 3 4 5 6 7)
Ci16 : 2 (Picken: 5)
Ci17 : 5

A similar comparative chart could also be made whereby, instead of transcribing with the scale as 1 2 3 4 5 6 7 1 ("key of C" scale) each melody is transcribed with 1 as the tonal center (TC). Thus:

1 as TC gives the same result as above
2 as TC with "key of C" scale = 1 as TC with "key of Bb" scale (7b 1 2 3b 4 5 6 7b)
4 as TC with "key of C" scale = 1 as TC with "key of G" scale (5 6 7 1 2 3 4# 5)
5 as TC with "key of C" scale = 1 as TC with "key of F" scale (4 5 6 7b 1 2 3 4)
6 as TC with "key of C" scale = 1 as TC with "key of A" scale (6 7 1# 2 3 4 5# 6# 7)
Of course, analysing the scales in this way does not take into account how the mathematics (and resulting intonation differences) of individual note relations may have been affected if there was a required tuning method for each mode. Pian and Picken transcribe in "keys" (without the implications this word has in Western music, for which see Wikipedia) assigned according to their understanding of the then-contemporary terminology, presumably to take into account how this would have affected the mathematical relationship of the notes to each other, and it may indeed have at one time been possible to determine how these mathematics were associated with mode names in actual practice. However, to my knowledge such detail is either not known or highly contested).

From my examination so far, Jiang Kui's music is largely diatonic (7 note scale of whole steps and half steps), occasionally hexatonic (six note scale; e.g., ci #4). Other notes may result from ornamentation, but there is nothing written that clarifies this. Qin music, in constrast, is largely pentatonic, but there are often extra notes in excess even of a diatonic system. These, however, tend to be bunched in such a way as to invite one to consider the possibility that the tonal center is changing in at least some of those passages.

One problem with this analysis is that some specifics of Jiang Kui's notation system depend on knowing the mode, but the descriptions of modes are at times unclear. (For an example see the differences between Pian and Picken on ci16. This, as well as the incomplete information about ornamentation, brings into question whether Jiang Kui's music is really so strictly diatonic (or hexatonic), but I have no other specifics for questioning those transcriptions.

Another way of looking at the issue of non-pentatonic notes is to compare the note collection of the ci songs with that of qin melodies. Gu Yuan is basically hexatonic (six note scale). With the tuning considered as 7b 1 2 3 5 6 1 (as on my transcription) the tonal center is 1 and the basic scale is 1 2 3 5 6 7b; however, there is also the occasional 3b and 4. To avoid accidentals in the basic scale the same relative tuning could be considered as 4 5 6 7 2 3 5, changing tonal center to 5 and making the basic scale 5 6 7 2 3 4; now, however, those occasional extra notes will be 7b and 1. In other words, there is no way to transcribe all of Gu Yuan without using any accidentals.

Perhaps of note here is the fact that in Gu Yuan the passages with non-pentatonic notes tend to be grouped together (end of section 1 and its harmonic repeat; beginning of section 2; end of section 3). In contrast, the Jiang Kui ci (based on a less complete examination) tend to be hexatonic or diatonic throughout. With the five ci I have recorded so far (as listed above), when transcribing them I have always made 1 (do) the primary tonal center. This has generally led to pieces that have either a sharped 4 or a flatted 7, not both, and no other accidentals. Assigning notes for the Jiang Kui transcriptions I have done so far according to the way I have generally done this for qin songs would mean transposing most of them down a fifth to avoid the F# or up a fifth to avoid the 7b.

Since I have been working with pre-existing transcriptions rather than by examining the Jiang Kui's original notation I do not know what method of assigning modern note names is most correct. Perhaps a reinterpretation of the note names could lead to different conclusions about the modality. And because of a lack of information on intonation issues I do not know whether for the actual music such a reinterpretation could significantly affect understanding of the intonation as opposed to the modality.

With qin melodies studying the relation between relative tunings and mode is complicated by the fact the the word "調 diao is used for both tuning and mode. The next footnote below has some brief discussion about how this may have affected intonation.

17. Intonation (in relation to tuning)
The only descriptions I have seen of qin tuning result in pythagorean tuning, which means all notes played in harmonics follow the cycle of fifths tuning system. With stopped sounds the intonation can be different, but there are no traditional descriptions of mixed intonation systems.

Today other repertoires sometimes use different intonations, and certainly this was also sometimes or often the case in the past, but I have not seen studies of this in relation to the original sound of Jiang Kui's melodies. In addition, it is not yet clear what standards there were in the construction of other instruments. For example, was the quality of a flute evaluated in part by how strictly it followed a certain standard in either intonation or pitch? If so, were there different standards at different times and in different places? If Jiang Kui heard a piece he had created on his own flute played by a flute tuned a pitch higher would he have considered it to be playing in a different mode? If he could play the same melody one pitch higher on his own flute, would that change the mode? To my knowledge there have not yet been any definitive answers with regard to this.

When doing transcriptions on my computer I can easily transpose a melody up or down. However, when listening to the differences on my computer's midi playback system some of the real differences of such transposition are obscured by the fact that the system uses equal temperament (I have not yet figured out how to explore other midi settings and in any case they would still probably not capture many significant color differences).

18. Instrumentation (and range) of Jiang Kui ci songs
This issue is discussed further in connection with particular songs. Here it should be emphasized that it is unlikely any selected instrument would have been limited by the range of any ci melody as written. For example, the stated range of the Jiang Kui ci songs is about 10 notes (octave plus two), but even most skilled human voices, not to mention most music instruments, have a greater range than this. It thus seems unlikely that the mere fact that the notation can only express a limited range means that the actual music was never intended to exceed that range. For example, the range of the qin for Gu Yuan is about three octaves; presumably the voice narrowed this by changing octaves. It thus seems reasonable in adapting Jiang Kui's ci melodies for qin to expand the range of the melody through upper or lower octaves; likewise with other instruments. Similarly, when transcribing qin songs I have sometimes had to guess at where a singer might not follow the octaves indicated by the tablature; likewise Jiang Kui's own transcriptions might also have indicated leaps up where some singers might have actually gone down.

19. Syllabic settings
Gu Yuan does have an instrumental interlude where the melody of section 1 is largely repeated, but in harmonics and without paired lyrics. The absence of several notes at the end mean the lyrics could not be sung with this harmonic passage.

The most notatable characteristic showing the word intensity of qin songs, as mentioned further here, is the fact that those songs are paired one stroke per word - ornaments may or may not have words. Jiang Kui's ci songs are written in notation, not tablature. How syllabic the setting actually is depends on how significant the ornamentation. However, there seem to be some significant differences of opinion about how to interpret Jiang Kui's ornamentation. Is it possible that they are in there because Jiang Kui this was the only method Jiang Kui could imagine using to avoid the syllabic setting requirement?

According to their research into Tang dynasty music by studying the Japanese gagaku tradition, Cambridge Tang dynasty music research project determined that the Japanese gradually slowed it down, adding and emphasizing the ornamentation until the ornamentation actually became the melody. Is it then possible to understand clearly Jiang Kui's settings without a thorough understanding of his ornamentation?

20. Repeated notes
A fundamental of the qin aesthetic is the color of the notes (which in turn is dependent on the silk strings as well as the quality of the instrument itself). Inherent in the present comparison is the fact that in the qin repertoire no note is ever repeated exactly: do followed by do will always be different in color, whether from playing with a different type of pluck, from playing on different strings, in different octaves (concerning which see more above), and so forth. A flute, for example, may be able to do some of this, but there was no way to indicate it in any surviving notation system.

21. Comparing Jiang Kui's qin song with his ci settings
One can speculate about to what extent the aesthetics of qin music is wrapped up in the physical shape of the instrument. One fundamental issue, as mentioned above, is that within the qin repertoire notes are repeated more often than in other repertoires because the way a note can be played on different strings gives it a significant difference in color.

22. "HIP" and Reconstructions of Jiang Kui's Songs:
My understanding of notation for Jiang Kui's songs is that it almost certainly should not be considered prescriptive of how it should always be played. It is either descriptive of a particular performance or, more likely, provides an outline upon which performers can build/expand. It is also clearly incomplete in other ways as well, leaving out such important information as the instruments to be used (Jiang Kui is known to have played flute, but there is nothing to suggest he would have rejected other instruments) or the relationship between the voice and the music (e.g., whether they should be in unison, interact heterophonically, adjust octaves according to the range of voice or nature of the instruments, etc.). Much Western music outside its Common Practice Period was also written with this intention.

Given this, my own standard for credibility in connecting a song to Jiang Kui's intentions is whether one can say for certain that something could not have been done that way. Thus, for example, the first time I heard a reconstruction of Jiang's songs the singer was a clearly Western-trained Chinese soprano who sang in bel canto style. When I questioned why this was done the answer was that we don't know how music was sung during the Song dynasty but we do know it was great music. Since the greatest music today is sung in bel canto style we should sing Jiang Kui songs in this way as well. Such an attitude makes its difficult for me to have a "willing suspension of disbelief" that what I am hearing has more than a superficial connection to Jiang Kui's actual creation.

23. Transmission of the songs of the Whitestone Daoist (白石道人歌曲 Baishi Daoren Gequ)
See Pian, p.34. The colophon by 錢希武 Qian Xiwu can be found at the end of the 6 folio volume. 陶宗儀 Tao Zongyi is mentioned further elsewhere. (Regarding Pian's transliterations, such as "Sonq" for "Song" (or more specifically "Sòng"), she is using the Harvard romanization system).

24. Available Sources for the Songs of the Whitestone Daoist (白石道人歌曲 Baishi Daoren Gequ)
The three 18th century editions are by 江炳炎 Jiang Bingyan (1737, but only printed in 1913 by 朱孝臧 Zhu Xiaozang), 陸鍾輝 Lu Zhonghui (1743) and 張奕樞 Zhang Yishu (1749). Pian (p.33 ff) says there were many other early editions that omitted the tablature/notation. The only reference in her bibliography seems to be to an edition in Volume XVIII of 彊村叢書 Qiangcun Congshu, a compilation by 朱祖謀 Zhu Zumou (1857-1931). And although she also has detailed discussions about other early editions, see also p.99, it is not clear to me yet how this particular edition fits in with other early versions such as the one in 四庫全書 Siku Quanshu (4卷).

So far I have seen two collections, both in the library of Columbia University in New York. The following details are from the card catalogue:

  1. Baishidaoren ge qu [6 juan]
    Title 白石道人 歌曲.[6卷]
    Author Jiang, Kui 姜蘷, approximately 1155-approximately 1235
    Published Taibei Shi : Shi jie shu ju, Min guo 70 [1981]
    Columbia, East Asian, PL2687.C5 A5 1981

  2. Baishidaoren ge qu: [4 juan] / (Song) Jiang Kui
    白石道人歌曲 : [4卷] / (宋) 姜夔撰.
    Published : [Taipei] : Taiwan shang wu yin shu guan 臺灣商務印書館, [1983]
    Ying yin Wen yuan ge Si ku quan shu; di 1488 ce
    景印文淵閣四庫全書 ; 第1488册 (269-303)
    Columbia, East Asian, AC149 .S699 1983
    partially online)

Both are facsimile reprints; there are certainly other such versions (see, e.g., the unidentified one here), but as yet I do not have further details.

25. Six volume edition
The contents of my copy of Baishidaoren Ge Qu in six folios are as follows:

Opening preface
I am not yet sure of the source, the text begins and ends as follows:


The date does not seem to be indicated.

Table of contents (.jpg version, which does not name the 14 Sacred Songs)

卷之一 Folio 1

聖宋鐃歌鼓吹曲十四首 14 Sacred Song (Dynasty) Military Wind and Percussion Songs (lyrics only)
琴曲一首 One qin piece (in standard qin tablature)
側商調 Ceshang mode
調弦法 Tuning method
古怨 Gu Yuan (music and lyrics)

卷之二 Folio 2

越九歌 Yue Jiu Ge: Nine Songs of Yue (also translated as "Ten Ritual Songs"; 10 poems, all with music)
(The music is all written in lülü notation)
古琴譜法 Old and New Notation Method (compares the lülü and gongche systems
折字法 Rules for the character "zhe" (

卷之三 Folio 3

Ling: 19 titles, five of them with two or more examples
(Four of these have music, all using suzi notation:
#5, #14, #15, #16)

卷之四 Folio 4

Man: 19 titles, one with two examples
(Only the first one (
Nishang Zhong Xu) has music, again in suzi notation)

卷之五 Folio 5

自度曲 Zidu Qu: "Self-done songs", 9 titles
(All include music written in suzi notation)

卷之六 Folio 6

自製曲 Zizhi Qu: "Self-made songs", 4 titles
(The first three have music, all written in suzi notation)

Closing line: 嘉泰壬辰至日,刻於東岩之讀書堂, 雲間錢希武。
Dated 1202, Qian Xiwu

Extra comment: 歌曲特文人餘事耳....今將善本勘讐方可人意後十一年庚子夏四月也。

After this there is a 別集 bie ji, an additional collection including 10 more titles (lyrics only), some of them having multiple examples

26. Four volume edition
The contents of my copy of Baishidaoren Ge Qu in four folios are as follows:

Opening prefaces (different from the above; no table of contents)

Folio 1: 皇朝鐃歌鼓吹曲十四首 Imperial Military Wind and Percussion Songs, a somewhat different title, but the same songs as above, followed by Nine Songs of Yue and the qin melody Gu Yuan

Folio 2: The same items as under the above-mentioned 19 ling titles

Folio 3: The same items as under the above-mentioned 19 man titles

Folio 4: The same Self-done Songs and Self-made Songs as above.

There is then the same 別集 bie ji as above, at the end with the same extra comment as above, but without the line from Qian Xiwu giving the date as 1202.

I am working from a photocopy of both the six and four volume versions.

27. Pian, p. 34

28. 皇朝鐃歌鼓吹曲十四首 14 Imperial Military Wind and Percussion Songs
Also called 聖宋鐃歌鼓吹曲十四首 14 Sacred Song (Dynasty) Military Wind and Percussion Songs.Their titles read more like prefaces than titles:

  1. 上帝命太祖受命也五季亂極人心戴宋太祖無心而得天下也
  2. 河之表破澤州也李筠不知天命自憑其勇不能降心以至於叛而死也
  3. 淮海濁定維揚也李重進自謂周大臣不屈於太祖作鐵券以安之猶據鎮叛
  4. 沅之上取湖南也湖南有難乞援於我至則拒焉我師取之
  5. 皇威暢得荊州也我師救湖南道荊州高繼冲懼歸其土
  6. 蜀山邃取蜀也孟昶恃其國險且結河東以拒命兵加國除
  7. 時雨霈取廣南也劉鋹淫虐我師弔其民俘鋹以歸
  8. 望鍾山下江南也李煜乍臣乍叛勢窮乃降而我師朱嘗戮一人也
  9. 大哉仁吳越錢俶獻其國也
  10. 謳歌歸陳洪進以漳泉來獻也
  11. 伐功繼克河東也始太祖之伐河東誓不殺一人又哀劉氏之不祀故緩取之至太宗始得其地
  12. 帝臨墉親征契丹於澶淵也
  13. 維四葉美致治也
  14. 炎精復歌中興也

As yet I have not seen any punctuated versions or translations.

29. Tablature explanation with Gu Yuan
It is called 調絃法 Tiao Xian Fa (Tuning strings method) and is different from the 定絃法 Ding xian fa (Fixing strings method; 21570.xxx) discussed by Hsu Wen-Ying, The Ku-ch'in, p. 327. Here she writes (text edited here) that Jiang Kui "petitioned to the Royal court of the Southern Song dynasty to regulate the musical tones for rituals, and made a list with analyses of theories on different ways to tune the guqin." Hsu goes on to discuss the different tunings.

30. Ling
"Ling" literally means "command", but Picken (p.132) writes, "this term is applied to short tz'ŭ with a comparatively small number of syllables to the stanza".

31. Man: "idling tune?", "slow tune?", "short tune?", "drawn out melody?"
"Idling" is suggested by Kroll, who has "neglectful", "indolent", "unhurried", "dilatory", "arrogant", etc., as well as "slow". As for Chinese dictionaries, 11385 慢 has 13 entries, none dealing with songs, but see 11385.21 慢詞 man ci and 11385.35 慢調 man diao. The latter seems to say it is a 小令 short form of a ling:


Picken (p.133), however, has the following about "man":

"This term was reserved for longer songs than ling, with a larger total number of syllables to the stanza. The usual meaning of the term is 'slowly' or 'gradually', but in a musical context it indicates not pace but a particular kind of structure: a heterometric stanza with a large number of short lines, and with a comparatively small number of lines of seven or more syllables."

Here man is rendered as "idling tune" in part because it could refer to the mood expressed in the melody (with ambivalence about that mood) or to the nature of the melody itself. The comments above suggest that "man" is a term meaning a song of a certain type, but unfortunately this does not say clearly what type of melody it is and so it does not help very much in reconstructing the melody. For example, an irregular number of characters per phrase does not necessarily signify an irregular rhythm for the melody.

In Ming dynasty qin tablature one often sees the instructions "入慢 ru man" and/or "漸慢 jian man". Perhaps the above comments on "man" suggest that these terms mean that the music becomes more free rather than (though not precluding) slower.

32. Flute
Ci Song #16 mentions a reed-pipe. One can only speculate as to how appropriate Jiang Kui himself would have found it to hear his songs played on other instruments, but there is no reason to believe he would have objected to it.

33. Singing
The suggestion that perhaps the songs did not need to be sung is largely speculative: due to their brevity anyone who liked the melody would presumably have enjoyed hearing it played on an instrument. One can also cite the evidence that Jiang Kui sometimes listened to them played on instruments before they were sung (see, e.g., under ci #16).

34. Percussion
There is little or no mention of either rhythm or percussion in Jiang Kui's own writing. My own attempts at reconstructing his melodies, as with my qin melody reconstructions, are based on an assumption that the music is rhythmic but that the rhythms are freely interpreted. I have not been able properly to study Jiang Kui's rhythmic indications.

As for percussion instruments, if a drum was used to mark the beat, this presumably would emphasize the rhythm, but the most common percussion instrument at that time seems to have been clappers, and evidence suggests that, at least later, these may generally have been used only to mark the ends of phrases. A poem by Lin Bu refers to the percussion instrument/clappers as 檀板 tanban but the more generic name is 拍板 paiban (Wiki).

35. 霓裳中序 Nishang Zhong Xu
Nishang is best known in the title Nishang Yu Yi (details), but at least one old list also includes as a qin melody title Nishang Yin.

Recently I found on YouTube this interpretation (posted by Tao Liu) for flute and percussion of Section 1. The extremely slow style (4'46" for that section only) suggests that the interpretation was inspired by Japanese court music (gagaku) as played today. Gagaku did include music brought to Japan from Tang dynasty China. However, the extremely slow style of its contemporary performance most likely developed in Japan several centuries after its arrival there.

Other melodies attributed to Jiang Kui have many non-pentatonic notes, but the music of his Nishang Zhong Xu is basically diatonic throughout (scale 1 2 3 4 5 6 7b with 1 as the primary tonal center or 1 2 3 4 5 6 7 with 5 as the primary tonal center); this made it awkward adapting it for a qin using standard tuning. As for my own reconstruction, linked above, the melody seems so different from other melodies I have worked on that I am not convinced I have yet found the right clues to its interpretation.

36. Recording of 長亭怨慢 Chang Ting Yuan Man by 袁中平 Yuan Jung Ping (YouTube)
As mentioned above (together with a copy of Yang Yinliu's transcription), in his YouTube performance ("Lament of Departure") Yuan Jung Ping begins with a 30 second prelude, then plays the melody twice on the qin, the first time reciting the lyrics as he plays, then on the repeat singing them.

Without accidentals (and not considering possible ornamental notes) Jiang Kui's Chang Ting Yuan Man as generally interpreted is a hexatonic (six tone) melody with the note collection 1 2 4 5 6 7 (1) and the tonal center 4. With the tonal center 1 the note collection becomes 1 2 3 4# 5 6 (1; compare "Lydian" and see my own transcription). Picken, Pian and Yang have basically the same result: Picken and Yang transcribe the same Chinese note symbols using the note collection 1 2 4 5 6 7; Pian (pp.106-7) transcribes them as 1 3b 4 5 6 7b, making 3b the main tonal center.

Jung Ping largely follows the interpretation of Yang Yinliu, who writes out his interpretation of what he considers as ornaments. However, based on this understanding of the original note collection being 1 2 3 4# 5 6 7, with 1 as the main tonal center, then what Jung Ping has done is change the mode by flattening all the 4#s to 4. There are 8 occurrences of this relative pitch: with the words 香 xiang, 人 ren, 情 qing, 高 gao, 去 qu, 玉 Yu, 一 yi and 並 bing.

Jiang Kui himself says the mode is 中呂宮 gong on zhonglü; if someone interprets gong to mean "C" then that might be justification for changing the mode to C (or Ionian). Since I have not examined all other transcriptions I do not know whether in making this change Jung Ping is actually following a different existing intepretation, whether he made this change for the sake of qin tuning, or whether it was for another reason, such a personal taste. To my knowledge he has not discussed his reasons for this change.

To play this piece on the qin Yuan Jung Ping has in fact devised an original qin tuning: from standard tuning 1 2 4 5 6 1 2 he has raised the forth string a whole tone and the 5th string a half tone, giving 1 2 4 6 7b 1 2 (with A=440 his first note on the video is actually B, making the open first string A=55 or 110); the harmonics on the 5th and 9th studs (hui) thus become 5 6 1 3 4# 5 6. If he had instead used the tuning 1 2 4 6 7 1 2 then the harmonics on the 5th and 9th hui would have been 5 6 1 3 4 5 6. This suggests that the change was not done for the sake of tuning; perhaps it was for personal taste combined with a belief that for a gong mode piece Jiang would not have actually called for the F#s.

37. Melodies transcribed in Pian's Sonq Dynasty Musical Sources and their Interpretation
In addition to the Jiang Kui Songs of the Whitestone Daoist discussed above, Pian also has transcriptions and analyses of the other known sources of music published in the Song dynasty.

Pian wrote (p.VII) "Eighty-seven different musical examples are found in Sonq sources." The following is a full outline by category of her transcriptions of these in her book:

  1. Music in Popular Notation (俗字譜 suzi pu [sample]; pp.99-137)

  2. Pieces in Qin Tablature (pp.137-154)

  3. Music in the "pitch-pipe notation" (律呂譜 Lülü pu [sample-center: 七月]; pp.154-173)

  4. Music in "gong che notation" (工尺譜 Gongche pu [sample-left side: 七月]; pp.188-231)
    A popular form functioning much like the lülü notation in the pieces just mentioned

Much of the qin music in such sources as Shen Qi Mi Pu is thought to date from the Song dynasty or earlier, but there are no specific details of their Song dynasty publications/manuscripts, and so were not considered here.

Return to Qin poetry and songs, or to the Guqin ToC.