T of C
|Personal||email me search me|
Rivers in Spring and Autumn:
An examination of four related Ming dynasty guqin melodies 1
This paper examines four melodically related qin (guqin) melodies:
The focus of the examination is a comparison of melodic and modal characteristics (compare "key") in hopes that this will help determine the source and relative ages of the melodies. This effort has been only partially successful.
On my website, www.silkqin.com, as well as in several publications, one of the earliest being my Rhythm in Shen Qi Mi Pu (1993), I have outlined in great detail my methodology in reconstructing almost 200 guqin melodies published through the year 1539; I have also used these methods to reconstruct later melodies (mainly in handbooks through 1614), and this has helped put the earlier melodies into perspective. The aim in this has been to apply similar principles to those used in Historically Informed Performance (HIP) of Western music from the same period. This means following as carefully as possible the historical records of the music: instruments, notation, known performance practice. Only through strict interpretation of these materials can one find the principles behind the music. Then, once performers have thoroughly absorbed these principles, they should be able to play with freedom and creativity within what might be defined as an early guqin playing style. This is in some ways parallel to the archetypal learning process in an oral tradition: one copies one's teacher exactly until the skills and principles have been absorbed; only then is the learner considered able to be creative within the tradition, at the same time expanding that tradition.
As discussed in greater detail in my other publications, the three most important factors in historically informed reconstruction of pre-modern qin melodies are:
Regarding pitch indication, the practice used now when writing the tablature is to use a decimal system which gives the theoretically correct notes according to cycle of fifths sanfen sunyi) tuning. Because of the nature of guqin construction, as well as the richness of sound with silk strings, for stopped sounds these positions and the resulting notes are rarely so precise. (I have never been convinced by attempts to prove a conscious use of complex combinations of cycle of fifths tuning and just temperament.) In this way the system used in early handbooks such as Shen Qi Mi Pu (1425) had the advantage of absorbing this problem: all but one of the 12 notes of a chromatic scale (see A and A# in the table below) can be simply indicated; as with note values, subtle intonation changes (not necessary following common tuning systems) came mostly from the oral tradition.
The table below shows how, in theory, the old method of indicating finger positions can be used to cover all semitones. My work with Shen Qi Mi Pu suggested to me that a system such as this, with a number of variants, might once have been in use.
Table 1: Standard positions on a qin string tuned to C (from Qin Tunings, Some Theoretical Concepts, where items marked * have comments in the notes)
The most problematic positions in the old systems are the four between 4 and 5 (the fourth and fifth hui) and the use here of "ban" (half) for 4.4 is my own suggestion. (Positions higher than 4.0 are extremely rare, except when playing harmonics.) Common variants within the old system are usually easily understandable, such as 8b, 8+ and 9- for 8-9. It was my conclusion that, in general, with Shen Qi Mi Pu there is rarely any doubt about the note intended (intonation, as mentioned, can be much debated). And strict interpretation of these positions revealed previously unknown modal characteristics. Later, if other early handbook using this system had less clearly indicated notes, I would apply the understanding I had gained from Shen Qi Mi Pu and other tablature that did not have ambiguous notes.
The following table shows my conclusions about tunings and mode in Shen Qi Mi Pu
Table 2: Shen Qi Mi Pu tunings and their modal characteristics
(From Modality in early Ming qin tablature; 中文)
|mode name||tuning||main note||Secondary note(s)||Main string||Tuning method|
|gong||5612356||1 (gong)||5 (zhi)||3rd (jue)||standard|
|shang||1245612||1 (gong)||2, 5 (shang, zhi)||1st (gong)||standard|
|jue||5612356||1 (gong)||3, 6 (jue, yu)||3rd (jue)||standard|
|zhi||1245612||5 (zhi)||2 (shang)||4th (zhi)||standard|
|yu||5612356||6 (yu)||3 (jue)||2nd (shang)||standard|
|shangjue||1245612||1 (gong)||3, 6 (jue, yu)||1st (gong)||standard|
|manshang||1145612||1 (gong)||5 (zhi)||1 (gong)||slacken 2nd|
|manjue||1235612||1 (gong)||5 (zhi) [3 (jue)]||1st (gong)||slacken 3rd|
|ruibin||2356123||6 (yu)||3, 2, 1 (jue, shang, gong)||4th (zhi)||tighten 5th|
|qiliang||2456123||2 (shang)||6 (yu)||1st (gong)||tighten 2nd/5th|
|guxian||6123561'||1 (gong); 5 (zhi)||6 (yu); 3 (jue)||2nd (shang); 5th (yu)||tighten 2nd/5th/7th|
|mangong||3561235||6 (yu); 1 (gong)||3 (jue)||3rd (jue); 4th (zhi)||slacken 1st/3rd/6th|
|huangzhong||1356123||3, 6 (jue, yu)||1, 5 (gong, zhi)||2nd, 4th (shang, zhi)||slacken 1, tighten 5|
Regarding this chart, note that some of the tunings have been transposed. For example, it can easily be calculated that tightening the fifth string from standard tuning gives 5 6 1 2 4 5 6 , but for ruibin this is played as 2 3 5 6 1 2 3 . (Qiliang, guxian, mangong, and huangzhong are similarly transposed.) I often tested the relative pitch name selection by listening to someone sing the resulting melody in solfeggio
Modes can generally be defined in terms of tonal centers, notes to which the melody most often returns, especially at the end of phrases and sections. In line with this, for most melodies there seems to be a clear main note and secondary note. Also for most modes, the secondary note is almost always one fifth up from the main note, usually 1 and 5 (do and sol) or 6 and 3 (la and mi); second most common is for primary and secondary notes to be 6 and 1 (la and do). Also, although on the surface all these are simply varieties of pentatonic modes, in actuality the old modes also have some special characteristics. Most noteworthy of these is the tendency of the third above the main note to alternate between a flatted third and a natural third. This is most common in shang mode, but can be found in any mode. If the tonal center is 1 the most commonly changed note is 3; if the tonal center is 5 the most commonly changed note is 7; if the main note is 6, the most commonly changed note is 1 (to 1 sharp).
With regard to rhythm, I have written extensively (beginning in 1993 with Rhythm in Shen Qi Mi Pu) that I believe much can be determined about the intended note values. Although the tablature does not directly indicate note values it did not, as some have claimed, simply indicate notes, with the player free to assign any desired note value. Instead it described the way someone played a melody. In doing reconstruction I have always considered that person my teacher, trying to learn as much as possible what the tablature may be describing. The final test of this is to imagine someone transcribing my reconstruction: could it be written down in a manner similar to the description of my teacher's method. Only then do I feel I can make changes (though in fact what I usually do is go on and learn another melody).
There is considerable disagreement as to whether guqin music has, or should have, rhythm, a pulse, or no rhythm. My conclusion from this is that it can have all three. However, it is also my observation that in general most Chinese instrumental music has been rhythmic, specifically double rhythm (2/2 or 4/4). On this basic, in doing my initial reconstructions I have always tried to write it out in 2/2 or 4/4 and look for structures within that rhythm. Almost always they come up, and my transcriptions mostly read accordingly. However, once I have absorbed these rhythms and structures, I begin to play more freely. A listener may then think the music is very free. It should, however, sound like a musical language, not random notes.
The four melodies
Regarding the four titles under consideration, the themes of all but Chun Jiang Qu are connected to living a carefree life apart from society. As the following will show, they seem to be variations on each other. Although there are great differences between some versions, the 12 that appeared in at least 11 handbooks from 1511 to 1589 have clear melodic connections. What is remarkable here is not just that during this time there were at least two completely different sets of lyrics, that they told three quite different stories, and that they were specifically connected to four different men, but also that after a 1609 reprint of Qiu Jiang Wan Diao (perhaps also reprinted in 1802) all the related melodies seem suddenly and completely to have disappeared from the repertoire.
With regard to the appearance of an autumn theme appearing amongst melodies on a spring theme it is worthwhile noting that both an Autumn River Departure (Qiujiang Songbie) published in 1585 and a Spring River Departure (Chunjiang Songbie) first published in 1620 are variations of the popular melody Three Partings at Yangguan (Yangguan Sandie).
My website has further information on the various individual themes, the lyrics, and the history. Here the focus is on some specific comments on the music. Although I have looked cursorily at the later versions of these titles, my comments here will be limited to the first version of each.
In my specific comments on these four melodies constant reference is made to my transcriptions, making use of the bar numbers on each of the four. Here is should be emphasized that the notes discussed, though notated in staff notation as C, D, E, etc., and called by these names, must be considered as relative pitches: "C" is actually "do", i.e., "gong" or "1". And it must also be noted that under my transcriptions into staff notation I always put the original tablature. Where I make changes they are indicated as follows:
A general examination of the transcriptions will show that musically these four can be grouped in two, with the 1511 and 1539 versions considered together, then the 1530 and 1525 versions considered together.
Spring on the River: the two shorter melodies
The two shorter melodies have more in common with each other than with the longer two.
The 1511 and 1539 versions are both divided into three sections, starting in harmonics, then with most of the melody in stopped sounds, ending with a brief harmonic coda. The opening harmonics of the two (mm. 1-13 of 1511; mm. 1-17 of 1539) are identical for the first 8 mm. 1511 sets this to lyrics by Guo Zhen then, as the harmonics continue into Section 2, the music is accompanied by the beginning of Zhang Ji's lyrics. 1539 ends Section 1 at m. 17, the end of its opening harmonics. Musically this seems to make more sense. This part of the harmonics passage in 1511 (mm. 1-13) basically cuts out the middle of 1539's mm. 9-17.
At first glance this opening suggested to me that the 1539 melody, or a version related to it, was the basis for the 1511 melody. However, as I became more familiar with the melody, and in particular began to sing the lyrics, I began to doubt this conclusion. For one thing, the harmonics fit nicely with the image of cloudless skies and calm water mentioned in mm. 10-11. In general, the lyrics and music match well. Of particular note is the fact that the music of Chun Jiang Qu Section 3 is paired to three poems that each consists of a paired couplet ([5+5] x 2). In each of the three the opening couplet ends on G (so), then the closing one ends on C (do). None of this is explained by suggesting that the 1511 version is based on an earlier prototype.
This does, however, bring into question the inclusion of the 1511 Chun Jiang Qu as a shang mode melody, as D (here re, i.e., shang) is not particularly significant. Here the 1539 Chun Jiang has more clear shang characteristics, particularly after m. 73, where a number of places a phrase ending on D is followed by one ending on C. It might also be noted here that the ending of 1539 Section 2 seems to come one phrase too early: Section 2 ends with a phrase that has G as its final note, while the first phrase in Section 3 ends on C, thus seeming to form a pair with the previous phrase.
Focusing on the stopped sounds, after the opening harmonics the ensuing stopped sound passages (mm. 14-49 of 1511 and mm.18-104 of 1539) start with a very similar phrase, the major difference being that 1511 seems to begin C F F C while 1539 begins C E E C . However, this brings up a major issue with interpreting finger positions in many of the later Ming dynasty handbooks: whereas in my opinion handbooks such as Shen Qi Mi Pu indicate finger positions just as accurately as the decimal system as used in the Qing dynasty, later Ming dynasty handbooks seem to become more and more vague. Thus, certain handbooks only use ban (half) to indicate finger positions occurring between the markers (hui). Such is also the case with many of the melodies in Taigu Yiyin, including Chun Jiang Qu. If only ban is used, how does one distinguish between, e.g., 6.2, 6.4 and 6.7, as occurs here?
There is no ideal solution, so what I have done with questionable notes in Chun Jiang Qu is largely to follow the pitches of Chun Jiang. As mentioned above, Chun Jiang has two non-pentatonic pitches, F and B flat. Chun Jiang Qu has these same two, again as indicated above. But what about m. 23 of Chun Jiang Qu, which has a 6 ban that could be interpreted as 6.4 = B natural? Since this note does not occur elsewhere, I interpret it as 6.2 = C.
Melodically the two pieces come together again at mm. 18-23 of 1511 and mm. 39-46 of 1539. Both are a descending passage from the 7th to the 1st strings, largely played at the 7th position, then returning to the G on the fourth string. In addition, the passage in 1511 mm. 44-47 is almost the same as 1539 mm. 74-77.
The harmonics codas of 1511 and 1539 are different, but the similarities elsewhere are sufficient to establish a strong relationship between the two.
Evenings on the River: the two longer melodies
Comparing these two melodies, by some statistics they are quite different from each other:
4. Autumn River Evening Fishing (1530), in 8 untitled sections, has lyrics of unknown origin. 1530 does not name the mode, but the later two versions are in shang. 1530 is the longest melody and has the most non-pentatonic notes (B flat [mm. 55, 87, 175], E flat [mm. 96, 90], F [mm. 44, 54, 65, 147, 153-4, 194] and F sharp [F# / 4#; m. 29]; in addition there is perhaps a G# (G sharp) in m. 167 and there are numerous other questionable notes). It is printed poorly, and it is written imprecisely (finger positions between hui are in four cases indicated precisely [mm. 54, 67 repeated 131, 144], otherwise they are either indicated by ban or are omitted altogether).
In spite of these differences, one can still find parallels throughout.
Both have harmonics at the beginning, middle and end. Both first section harmonics are related to those of the 1511 and 1539 handbooks, but are much more similar to each other. Section 5 of 1530 repeats with different lyrics the harmonics in its Section 1; Section 3 of 1525, also in harmonics, changes the opening phrase and drops the last one, otherwise it, too, is a repeat of its first section. The closing harmonic passage in 1525 resembles the latter part of the closing harmonics of 1530.
The stopped sounds openings (Section 2 of both melodies) begin very similarly both to the stopped sounds openings to 1511 and 1539, but again more similar to each other. After this mm 30-37 of 1525 have much in common with 1530's mm. 31 through the end of its Section 2. Then mm 38-54 of 1525 even more closely resemble the opening of 1530 Section 3 (mm. 43-50). Continuing on, Section 3 of 1525 (mm. 62-85) at first glance seems quite different from the rest of Section 3 plus Section 4 of 1530; however, it can be seen that 1525 mm. 62-65 and especially its elaboration in mm. 66-70 resemble in outline 1530 mm. 81-84 (which in turn is a variant of its mm. 51-54). Then 1525 mm. 79-85 also resemble in outline 1530 mm. 96-104, bringing both melodies to the harmonic passage discussed above.
After the central section harmonics, 1525 Section 5 is a variant of its Section 3, while 1530 Section 6 is a variant of its mm. 81-84, perhaps even more closely resembling the opening of 1525 Section 5. It then has new material before at mm. 144-147 going into another variation on mm. 81-84; then another transitional passage is followed by a near repeat at mm. 155-161 of its mm. 89-95, but with the former E flat changed to E natural. 1530 Section 7 has more repeats and variations on earlier material: mm. 162-165 repeats mm. 96-99. Then mm. 166-171 are answered by a similar passage at mm. 172-176. The end of 1530 Section 7 is a repeat again of mm. 155-161.
For the closings, 1530 Section 8 and 1525 Section 6 can be seen as variants of each other. Each begins on the same four notes. The first phrase of 1530 ends on D, the second phrase is a variant ending on C. The third and fourth phrases elaborate on these, with the third ending on D and the fourth again resolving to C. 1525 expresses the same concept of ending with three phrases, the first ending on E, the second on D, the third on C.
A final note on 1530 Qiu Jiang Wan Diao: a glance at my transcription shows numerous changes. Most of the reasons for the changes are very obvious within the context of all the writings I have done on notes played in Ming dynasty tablature, and it is beyond the scope of this paper to consider each one individually. A larger question is the reason for these changes. In some cases it seems as though the transcriber was making a fanciful effort at pairing fingerings to pre-existing music. For example, at the end whoever was responsible for the tablature seems to have run out of lyrics and so, with no words to pair, the writer uses (see mm. 200-201) a standard version of the multi-note cluster 搯撮三聲 tao cuo san shang. However, in mm. 69 to 72, presumably in order to match to lyrics what quite possibly is a similar note sequence, the writer uses some strange figures indeed.
Summary and Conclusion
My aim when beginning this paper was to try to try to see whether they contained evidence that would help date them. In this I have not been very successful. However, comparing them has also been very useful in my attempts to reconstruct them. Earlier I had worked on them separately, the 1511 Chun Jiang Qu and 1530 Chun Jiang Wan Diao in particular, without realizing their relationship or being happy with my selected note values. Later, as I worked on 1539 Chun Jiang I began to see the relationships, and realized that comparing passages would help me decide on note values. 1525 Chun Jiang Wan Tiao provided more help with interpreting other musical phrases. The result of this is four melodies that I consider very lovely. Three I can play from memory, meaning their structure is very logical to me. 1530 is more problematic because of its repeated patterns and variations, not to mention its similarities and differences from the 1525 tablature in particular. To play Qiu Jiang Wan Diao from memory might require also memorizing the lyrics or working with a singer.
All four melodies are in the shang mode and have some of the characteristics of other shang mode melodies I have studied:
Another characteristic found in almost all early surviving shang mode melodies is the occasional flattening of the third, so the melody has both E and E flat. Here the 1530 melody has an E flat (it has two). For non-pentatonic notes, much more common for at least three of these melodies are B flat and F. This is not a common characteristic of early surviving shang mode melodies.
The change within the shang mode from having E mixed with E flat to having E mixed with F is something I first discovered in 2002 while reconstructing the three "new" Songxianguan Qinpu melodies (titles which had not appeared in earlier handbooks) for a conference held that year in Suzhou in honor of Yan Cheng, considered the founder of the Yushan School. At that time I made a presentation suggesting that this might be a characteristic of the early Yushan school – later Yushan tablature mostly eliminates the F as well as the E flat, becoming more purely pentatonic.
This examination of four melodies published between 1511 and 1525 perhaps suggests these modal differences date from earlier, perhaps much earlier. My focus has been handbooks such as Shen Qi Mi Pu, Fengxuan Xuanpin and Xilutang Qintong, with emphasis on melodies that may have been copied from earlier tablature. What I have not done so thoroughly is reconstruct melodies from early handbooks said to reflect the Xumen Zhengchuan, a tradition said to date from the famous player Xu Tianmin at the end of the Song dynasty. This is partly because the Xumen-related handbooks seem to have many more written errors in their tablature. Also, my unsubstantiated assumption has been that, whereas the handbooks I have dealt with often were simply copying old tablature, these other melodies were passed down through oral tradition until being written down in the 16th century. Only a concerted effort to reconstruct such melodies, similar to what I have done with the non-Xumen handbooks, would allow better answers to questions posed here about varieties and changes in the modes within the guqin repertoire of that time.
Appendix: Four transcriptions
Footnotes (Shorthand references are explained on a separate page)
Rivers in Spring and Autumn
This paper is slightly modified from the one originally presented in Hong Kong on 9 December 2010 at the 2nd Interdisciplinary, Intercultural International Conference on Guqin, Aesthetics and Humanism (2010第二屆「古琴的音樂美學與人文精神」 跨領域、跨文化國際學術研討會). The main difference is the addition here of links explaining terms.
Metal strings and the Cultural Revolution
One might argue about the reasons for their introduction: Is it something that eventually would have happened anyway? Was it a natural result of the poor quality of silk strings? Was it part of a political effort to make the qin into an instrument for performance rather than self-cultivation? Or a combination of these and other reasons?
Return to the top or to the Guqin ToC.