Cherubic Hymn in Japanese - first mode

frephraim

Παλαιό Μέλος
With a lot of help and based on my incomplete collection of papadic formulae, I have composed a cherubic hymn in Japanese for a choir director in Japan who believes that Byzantine music is more appropriate for the Japanese language than Russian-style music is. I've attached my composition in Byzantine notation and in staff notation, as well as a recording of me chanting it and a recording of a native speaker reading it.

It was a little tricky in several spots because in Japanese very few syllables are accented. Also, some syllables have vowels tacked on to the end of them that don't quite count as a separate syllable. Using a meaningless "n" in the middle of a syllable held for many notes does not sound right to their ears, so I omitted it. And since Japanese has even more syllables than Greek, when a formula is supposed to repeat syllables, I continued instead with new syllables so that the cherubic hymn would not last too long.
 

Dimitri

Δημήτρης Κουμπαρούλης, Administrator
Staff member
Wow father... who would imagine to hear Byzantine chanting in Japanese. But with God's help everything is possible. May God give you health and strength to continue this useful work. And as always thanks for sharing it with all of us.
 

basil

Παλαιό Μέλος
As usual, great work! Congratulations on your achievement of composing formulaically-correct Byzantine chant in a mora-timed language.

It was a little tricky in several spots because in Japanese very few syllables are accented.

I noticed that in line 3 of page 2, the formula you used for "hodokosu" didn't sound quite right to me. Consulting page 85 of your collection of papadic formulas in Plagal First Mode, I saw that this unusual 100X formula was only used twice in the past: once in a Slavonic adaptation by Nicola Trandafilov in 1847 and once in a Slavonic adaptation by Petar Saraf in 1912. It is clear that these two composers devised this formula due to the lack of similar 100X formulas in Cherubic Hymns in Greek. I respect their effort and ingenuity, but I believe that in this case an optimal formula can be derived by blending together primitives from Greek formulas alone; i.e., without resorting to the creation of a new formula. (Note that I am not necessarily against the creation of new formulas in non-Greek languages; however, I do prefer the use of existing formulas whenever possible.) Here is my derivation.

First, observe that the third 10 formula on page 81 of the Plagal First Mode formulas is fairly flexible. Consider the fragment of that formula that begins with the first high Ni and ends with the end of the formula. At the bottom of that same page, for example, this fragment takes a meaningless consonant on its first high Ni. On page 71, however, it appears in the second 0001 formula on the page with syllable repetition; from the fourth-to-last Ke to the last Ke, it takes four syllables. And on page 77 of the First Mode formulas, the same fragment also appears in the [almost identical] first 001 formula on the page; this time, the fragment takes only three syllables from the fourth-to-last Ke to the last Ke. The flexibility this fragment affords even leads me to question whether the syllable repetition is necessary at all when attaching extra syllables to it. For example, the 1001 formula on page 85 of the Plagal First Mode formulas is very similar in melodic contour and syllabic distribution, and it does not employ syllable repetition at all. But since you have chosen to ignore the Greek rules of syllable repetition when composing in Japanese, we don't have to answer this question right now.

Based on these examples, we can construct a 100X formula for "hodokosu" as follows: start with the second half of the second 0001 formula on page 71 of the Plagal First Mode formulas. Following the example of the second 1001 formula on page 85, discard the use of syllable repetition. Following the example of the 001 formula on page 77 of the First Mode formulas, delete the first 0 syllable so that the fragment takes four syllables instead of five.

We have now derived a valid 100X Di to Di formula for use in papadic First Mode and Plagal First Mode compositions. But before we can employ this derivation in our composition, we must adjust the melody for "inochio" so that it ends on Di. The second 0001 formula we started with on page 71 of the Plagal First Mode formulas gives us a way to do that; however, it only provides two syllables, and we need four. The second 0001 formula on page 71 of the First Mode formulas gives us a way to get to Di with three syllables, but we're still missing a syllable and there aren't any obvious ways to work with that formula further. The first line of the first 000001 formula on page 73 of the First Mode formulas gives us a way to do it with five syllables, but its second unstressed syllable (and its associated melody) can be safely snipped off because it merely repeats the melodic gesture of the preceding syllable. The end of this line gets us to the same melodic point achieved by the first syllable in the second 0001 formula on page 71 of the Plagal First Mode formulas that we are altering. From there, we can take the melody from the second syllable of that formula intact, getting us to a resting point on Di, and then we're in a position to use the 100X formula we derived.

Thus we have derived a valid melody for this unusual syllabic pattern only by blending together primitives from existing Greek formulas and without needing to resort to the creation of a new formula. I have attached the end result to this post.
 

Attachments

  • example.png
    example.png
    11.5 KB · Views: 11
Last edited:

frephraim

Παλαιό Μέλος
Dear Basil,

I was delighted to see your methodology in using my compilation of formulas to come up with a better solution for this tricky problem. I agree that your solution is preferable to the melody I have posted. (I wonder if someone like Ioannis Arvanitis who composes with the old notation could have reached the same solution you reached.)

There is only one slight linguistic problem that might be an issue: When I was composing music for this cherubic hymn, that choir director in Japan (and also a visiting Japanese monk as well) had me redo a few melodic lines because those lines held certain syllables for too many beats. Even though those melodic lines were following the formulaic rules, apparently the problem had something to do with the unique nature of the Japanese language, which you pointed out is a mora-timed language.

One of the things they told me was that when words end with the syllable "o," that syllable doesn't exactly count as a separate syllable. That is why I used a two-syllable formula for the word "ka-ri-o." I suspect that they might not be pleased with your melody that holds the last syllable of the word "i-no-chi-o" for eight beats (although they did let me get away with doing this for the word "che-ru-bim-o"). So to find out, I have sent them a recording of your solution. When they reply, I will let you know what they think.

in Christ,
+Papa Ephraim
 
Top