Two recordings are attached here. Either one can be used for any of the four ci poems given below: the two directly connected to these melodies and the two by Bai Juyi given further down.

1. From 1676: listen 聽 (qin only; 1.06; prelude 11"; 五線譜 transcription; tablature at right)
2. From 1682: listen 聽 (qin only; 1.20; prelude 11"; 五線譜 transcription)

1. Chang Xiang Si in Japan
   In Japanese publications of Chang Xiang Si, such as this one, the melody has the subtitle Chun Gui⁶ (Spring Day in the Boudoir, commonly used as a metaphor for a woman; no relationship to Chun Gui Yuan). The lyrics are by Feng Yansi⁷ (903? - 960; ICTCL). :
   紅滿枝，綠滿枝，宿雨厭厭睡起遲，閑庭花影移。
   Hóng mǎn zhī, lǜ mǎn zhī, sù yǔ yàn yàn shuì qǐ chí, xián tíng huā yǐng yí.
   憶歸期，數歸期，夢見雖多相見稀，相逢知幾時。
   Yì guī qī, shù guī qī, mèng jiàn suī duō xiāng jiàn xī, xiāng féng zhī jǐ shí.

   Red (petals) fill the branches, green fills the branches,
       last night came rain in layers and I am slow to awaken, in the quiet courtyard the appearance of flowers is transformed.
   I recall your expected day of return, many are the expected days of your return,
       In dreams although I see you often yet we seldom do see each other: when will I realize our meeting?

2. Chang Xiang Si in China
   The versions of Chang Xiang Si published in three Chinese handbooks all seem identical to each other; they also have the same ci pattern as the version preserved in Japan (3,3,7,5 repeated once), but their lyrics (said to have been added by Wang Danlu⁸) and music (by Cheng Xiong) are completely different. In the editions from China the lyrics are:
   客滿庭，酒滿觥，故斷危弦聲不成，侯門空復情。
   Kè mǎn tíng, jiǔ mǎn gōng, gù duàn wēi xián shēng bù chéng, hóu mén kōng fù qíng.
   山層層，水泠泠，一曲龍吟萬慮清，風微玄鶴鳴。
   Shān céng céng, shuǐ líng líng, yī qū lóng yín wàn lǜ qīng, fēng wēi xuán hè míng.

   Guests fill the room, wine fills the goblets,
       and so when a lofty string breaks the sound does not form, only when the mansion empties do feelings return.
   The mountains are in layers, the rivers are cold,
       (Just playing) the melody Dragon's Intonation and all my worries are cleared, the air is profound and black cranes call out.

Even after a careful examination of the original tablatures for both these melodies as well as their modern interpretations it is difficult to say which of these two versions is earlier. The ones published in Japan were probably brought there from China at the beginning of the Qing dynasty by the monk Jiang Xingchou (Shin-Etsu), suggesting they are earlier; however, it is also possible that Shin-Etsu created these melodies after he arrived in Japan.

In addition to my own, several other transcriptions for these melodies are also available. These include those by:

1. Wang Di, two, both from Hewen Zhuyin Qinpu: in Qin Ge, p.61; and Xiange Yayun, p.124
   Musically both seem the same, with small musical changes from the original, but the first is in staff notation with no tablature, the second is in number notation with tablature. In the first the lyrics are changed to the Bai Juyi lyrics quoted above; the second uses the original Feng Yansi lyrics. The latter is recorded on CD 1/track 5: it is played three times, all with ensemble, each time with the last phrase played as a preamble; the first is without singing.
2. Wu Wenguang, from Songfengge Qinpu: Yushan Wushi Qinpu, p.445.
   This reconstruction of the melody has one change (correction?) in the tablature and a few in the transcription; it includes the original lyrics.
3. Li Xiangting, from Toko Kinpu: in Chapter 7 of his Guqin Foundation Course (李祥霆，古琴實用教程 Guqin Shiyong Jiaocheng)
   According to the commentary there, Li 訂指法 re-arranged the melody and added new fingering. Specifically, although it is basically the same melody, he has transposed it down a fifth (starting on the open second string instead of open fifth), then re-done all the fingering mainly to fit this. He also takes the closing phrase (no longer in harmonics, as the original) and puts a version of it at front to start the piece. The lyrics are omitted. (Thanks to Dr. Tse Chun Yan for pointing out the connection to me.)

The cipai (ci pattern) for 長相思 Chang Xiangsi

A study of this one cipai and the qin songs that use it shows some of the difficulties of trying to understand how ci might have been set to qin melodies. Starting with background on Chang Xiang Si, there is a famous poem by Li Bai of this name, but it has no connection to the ci pattern.

Regarding the two qin songs, both sets of lyrics have been arranged above as two lines (four in English) in order to show their parallel nature. Although each line is patterned 3,3,7,5 repeated, each has different music as well as the different lyrics. A question that must be asked in reconstruction is whether to try to give each line a parallel musical rhythm in accord with the parallel ci pattern. My own inclination, especially given the style of the first two short phrases of each line, is first to see whether there is a single rhythm that will fit both halves of one poem. I have done so for each of my two transcriptions. The reconstructions by Wang Di and Wu Wenguang may have originated this way, but then in execution some variation was introduced. Li Xiangting's, perhaps being instrumental, does not seem to try to keep that pattern. None of the other reconstructions I have seen has a discussion giving reasons why they did or did not try to do this.

In addition I wonder if there should be one musical rhythm that will fit both halves of both poems; mine does so fairly closely, but not exactly.

As mentioned here, almost all qin songs pair words to music by a nearly syllabic method using one character for each right hand stroke and certain left hand plucks. With this as a constant, as any particular melody title continues in the repertoire, several types appear. With long melodies that have lyrics the melodies usually change somewhat from version to version, as do the lyrics, but often it is not clear that these were ever intended to be sung.

In theory, ci patterns were based on the melodic setting of a specific poem/song; then for that one melody (even after it was lost) there were eventually many different textual settings. Most short qin songs of one title are different in this aspect: the lyrics seldom change but the melody may change somewhat, or even be completely different in some versions.

Chang Xiang Si, however is different from both of the above types in that, while the ci pattern remains, both the words and melody are completely different between the two versions. This phenomenon, of keeping a ci pattern but providing both new lyrics and new music, may have a longer history but it seems first to appear in print at the beginning of the Qing dynasty, in particular with the ci melodies connected to the Songfengge Qinpu.

However, all of the above types differ from the supposed ci archetype, wherein the ci pattern ("melody") remains the same but the words change. Perhaps there was once an oral tradition of creating (perhaps spontaneously) new lyrics for an existing qin song melody, or even of matching existing ci lyrics to a melody that already has different lyrics in the same pattern, but I have not yet found specific evidence for this.

 
Footnotes (Shorthand references are explained on a separate page)

1. References for Evelasting Longing (長相思 Chang Xiang Si)
Also translated as Everlasting Love. References include the following:

• Mention in other qin melodies, such as Xiang Fei Yuan.
• 42022.324/3 長相思 first discusses Chang Xiang Si as a section (within Folio 69) of Yuefu Shiji (pp. 990-995); Folio 69 has numerous poems of this name, including one by Bai Juyi (九月西風興，月冷露華凝。....). However, none of them has the present cipai structure, the most common one it has being 3,3,7,3,3,5,5,5,5. The one by Bai Juyi is (5+5) x 11.

• 42022.324/4 then discusses Chang Xiang Si as 詞牌名 the name of a cipai, saying that its pattern came from a poem by Bai Juyi named 汴水流 Bianshui Liu (17525.1 Bianshui river in Henan). Here Bai Juyi's poem is as follows:
  汴水流，泗水流，流到瓜州古渡頭，吳山點點愁。
  Biàn shuǐ liú, Sì shuǐ liú, liú dào Guā zhōu gǔ dù tóu, Wú shān diǎn diǎn chóu.
  思悠悠，恨悠悠，恨到歸時方始休，月明人倚樓。
  Sī yōu yōu, hèn yōu yōu, hèn dào guī shí fāng shǐ xiū, yuè míng rén yǐ lóu.

  The Bian waters flow, the Si waters flow;
      they flow by the ancient Guazhou ferry,
      and the Wu Mountains bow in sorrow.
  Endless thoughts, endless grief;
      grief that will have respite only when you return;
      the moon shines down as I lean over the balcony.

  Another Bai Juyi poem in this same structure is:

  深畫眉，淺畫眉，蟬鬢鬅鬙雲滿衣，陽臺行雨回。
  Shēn huà méi, qiǎn huà méi, chán bìn péng sēng yún mǎn yī, yáng tái xíng yǔ huí.
  巫山高，巫山低，暮雨蕭蕭郎不歸，空房獨守時。
  Wū shān gāo, Wū shān dī, mù yǔ xiāo xiāo láng bù guī, kōng fáng dú shǒu shí.

  Eyebrowns drawn too thickly or eyebrows drawn too thin;
      cicada sidelocks or disheveled sidelocks, puffed hair overflowing my clothing,
      along the sunny terrace rain returns.
  The Wu Mountains appear high, the Wu Mountains appear low;
      twilight rain whistles, the young man does not return,
      in an empty room alone I watch the time.

  N.B. In poetry the "Wu Mountains", literally "Shaman Mountains", can represent sexual encounters. Here "rain" also alludes to tears (thanks to Juni Yeung).

The structure in these latter two cipai examples (3,3,7,5 repeated once) is the form of the lyrics in the two Chang Xiang Si songs for qin discussed above; it is also the one generally used for later poems of this title.
(Return)

2. Ci pattern for 長相思 Chang Xiang Si
I have not yet seen a 平仄 pingze structure given for the present Chang Xiang Si. However, even though they have the same word count per phrase, there is no way the two settings above could both be conforming to the same pattern. More likely, neither was.
(Return)

3. Image: Setting for qin of the song Chang Xiang Si
This setting, discussed further below, is from Hewen Zhuyin Qinpu, published in Japan before 1676, probably having been brought there from China. The symbols by the Chinese characters tell Japanese how to pronounce the characters in Chinese. In these first two double lines the setting has one character for each qin note except for the slides on the fourth cluster of each line of tablature.
(Return)

4. Tracing Chang Xiang Si
Zha Guide 35/--/510 listings five occurences of Chang Xiang Si, as follows:

1. three identical ones in handbooks published in China:
         Shuhuai Cao, 1682 (XII/374; part of 1677-87)
         Songsheng Cao, 1697 (XII/392; also part of 1677-87)
         Ziyuantang Qinpu (1802; XVII/549)
2. two nearly identical ones from late 17th century Japan
         Hewen Zhuyin Qinpu (XII/220; see the original)
         Toko Kinpu (XII/257
         also copied in XII/263

If there are any others in this pattern but with another title I have not yet found them.
(Return)

5. Ci structure for Chang Xiang Si
Three 平仄 pingze and 韵 rhyme patterns seem most common:

1. 中中平（韵），中中平（韵）。
   中仄平平中仄平（韵），中平中仄平（韵）。

   中中平（韵），中中平（韵）。
   中仄平平中仄平（韵），中平中仄平（韵）。

2. 平平仄（韵），中平仄（韵）。
   中仄中平中仄仄（韵），仄平中中仄（韵）。

   平平仄（韵），中中仄（韵）。
   中中平中平仄仄（韵），仄平中中仄（韵）。

3. 中仄仄平平（韵）。仄仄仄平平（韵）。
   平仄中中仄，平仄仄中平（韵）。

   平平仄仄仄平平（韵）。平平中仄中平（韵。）
   中仄平平中仄，仄仄中仄平（韵）。

The actual examples here follow one of the first two patterns.
(Return)

6. 春閨 Chun Gui
14146.568 Spring Day in the Boudoir, commonly used as a metaphor for a woman. No relationship to Chun Gui Yuan.
(Return)

7. 馮延巳 Feng Yansi (903? - 960)
See ICTCL.
(Return)

8. 王晫 Wang Zhuo
21295.1340: Qing dynasty Hangzhou poet.
(Return)

Return to Qin Poetry and Song or to the Guqin ToC.