Diagrams and Text Contributed by David Crandall
Zahdi Date and Poppies photographs provided courtesy of Casey Preston
[EDITOR’S NOTE: As part of the continuing conversation surrounding Carrie Preston’s Zahdi Dates and Poppies which premiered on March 30, 2016 at the Tsai Performing Arts Center in Boston, Mr. Crandall contributes additional insight into the composition process for the piece.]
Time has slipped by and TN’s successful performance of Zahdi Dates and Poppies at Boston University is now months behind us. Still, I’d like to continue presenting some of the elements that influenced my compositional decisions before all memory of them fades away. I hope that others drawn to the idea of creating new noh will find these notes useful in their own explorations.
Last time, I wrote about how a typical shidai is organized poetically, and how Carrie modified the traditional form to suit her own purposes. Now I’d like to focus on the rhythmic characteristics of a typical shidai. This opens a vast field of discussion that I enter with some trepidation because it can provoke a glassy stare in readers. But to understand why I ultimately decided not to use the traditional structure for Zahdi’s opening, it’s useful to know what I rejected.
Two Types of Rhythmic Organization in Noh Chant
Noh chant is broadly categorized into two types: au (matched) and awazu (unmatched), which refer to how the drums interact with the text. Whether the chant is matched or not is one crucial element that helps to define different shōdan sections, the building blocks of noh. A shidai, for example, is defined in part by the fact that it is always matched.
In matched chanting, the drum patterns are closely calibrated to the text, requiring the drummers to peg their playing (both drum hits and drum calls) to individual syllables. There are three modes of matched chant: hiranori, chūnori, and ōnori. Shidai are always in hiranori mode, which I will discuss in more detail below.
In unmatched chanting, the singers and the drummers follow two parallel paths that are only loosely connected. The beginning and end of an unmatched section are fixed and clearly defined, and there are some guideposts along the way where certain drum patterns are expected to be played in the vicinity of certain lines of text, but there’s no attempt to bind the two strands into tight rhythmic units. The positioning of the drum hits and calls is approximate, even to the point where drummers can vary the number of patterns they use to accompany a given section of text, playing faster or slower as needed to come out even with the singers in the end.
Here’s a simplified chart that shows where the shidai is situated within noh’s rhythmic universe. The chart is far from comprehensive—many other shōdan exist that don’t appear here, and the entire area of instrumental music is not included.
The word hiranori can be translated as “ordinary ride.” It is the most common and the most complex rhythmic mode in noh. I’ll attempt a simple explanation of its underlying principles, with the caveat that I’m cutting things down to the barest of bones and not discussing any complicating factors such as line length and ornamentation.
The basic unit of noh music is an eight-beat measure. For the purposes of English noh, it can be notated like this:
The basic poetic line of noh text has 12 syllables that are divided into hemistiches of seven and five. On this most fundamental level, then, the task before a composer is to distribute 12 syllables across eight beats so that words and music come out even. In actual practice, there are many ways to accommodate both musical measures of different lengths and poetic lines of different syllable counts, and there’s no hard and fast requirement to come out even. But for now, let’s focus on this basic task.
To demonstrate, I’ll use a familiar English line that closely matches the rhythmic feel of a typical 12-syllable Japanese line:
Old MacDonald had a farm / E-I-E-I-O
How do we distribute these syllables over an eight-beat measure?
The first step is to divide each beat in half, which redefines our task as distributing 12 syllables across 16 half-beats. Once this is done, the obvious solution is to give eight of the syllables a length of half a beat while extending the remaining four syllables to a full beat so that the total number of half-beats adds up to 16. Such an arrangement looks like this:
In practice, only three syllables are actually extended, shown here with tied notes on the syllables “Old,” “Do” and “a.” The fourth extension is inserted as a rest on beat 8 (shown as an open triangle) so that the singers have time to breathe. The choice of where to insert the extensions (called mochi in Japanese) depends primarily on which noh school is being discussed. My examples reflect the traditional practice of the Hōshō school, which is similar to that of the Kita school. The Kanze school inserts its extensions in slightly different places. Perhaps you’ve noticed that the line begins on the half beat before beat 1, which might seem a bit odd. This is the default starting point (honma), a convention I ask you to simply accept for now until we get a chance to discuss what happens when lines have syllable counts other than 12.
Here’s what the above example sounds like. I’ve recited it three times to show how it can be made to dovetail in an endless loop.
Notice that the beat is steady—it’s the syllables that are lengthened in strategic places to make the words and music come out even. This way of matching the 12 syllables to the 8-beat measure has a name: tsuzuke-utai. The term tsuzuke denotes two things: 1) a specific, representative drum pattern that is played with a steady beat, and 2) a category of drum patterns (including the specific tsuzuke pattern) that share the steady-beat attribute. Utai means chant.
If this were all hiranori had to offer it wouldn’t be particularly exciting. Fortunately, whoever devised noh’s musical system centuries ago came up with an intriguing second method of making the words and music come out even. Instead of lengthening the syllables, why not shorten the measure? This is where things get interesting.
Let’s start with the basic framework:
The X’s show where half beats are taken out. Notationally, the reason I’m showing them removed in these particular places is related to the chanting practice of the Kanze school, which is by far the largest shite school. For our purposes, all you need to know is that everything gets shifted to the left and the beats get compressed (as indicated by the arrows). This produces a measure that is 13 half beats long, which matches the 12 syllables of the text plus one rest for breathing:
Let’s focus first on the underlying rhythmic framework, which is still conceptualized in terms of eight beats but is actually (aurally) three half-beats shorter. For Western musicians, it’s an oxymoron: an irregular meter. Here’s what it sounds like, counted out:
The meter is irregular, but look at the syllables: now they’re chanted evenly, without break or extension:
As you might expect, this way of matching the 12 syllables to the 8-beat measure also has a name: mitsuji-utai. Like tsuzuke, the term mitsuji denotes two things: 1) a specific, representative drum pattern that is typically played with an irregular beat, and 2) a category of drum patterns (including the specific mitsuji pattern) that share the irregular-beat attribute. Once one begins to sense the galumphing meter that anchors the mitsuji style, drum patterns that might initially seem random come into organized and predictable focus while maintaining delightful elasticity. The contrast between the halting mitsuji patterns and flowing tsuzuke patterns is one of the most enjoyable rhythmic aspects of noh, providing a sense of tension and release that can be used effectively to amplify the meaning and emotion of the text. The following example doesn’t pretend to have such lofty artistic goals; it simply presents five iterations of our trusty English line while switching between the basic mitsuji and tsuzuke styles. I think it should be fairly easy to identify which is which.
This, then, is the basic framework of the hiranori mode, but there are many other factors that affect performance. Here are a few things to keep in mind:
- In practice, noh rarely sounds as clear-cut as the above audio clip suggests. There’s breathing room for rubato even in tsuzuke patterns, and mitsuji patterns are very flexible, providing plenty of space for melodic ornamentation and expressive variation.
- Mitsuji drum patterns can be played in tsuzuke style. That is, the three missing half beats can be reinserted and the timing of the mitsuji patterns can be expanded to accommodate them. This in fact happens quite often—whenever one of the two paired drums (ōtsuzumi hip drum and kotsuzumi shoulder drum) plays a tsuzuke pattern, the other drum must follow along in tsuzuke style, even if it is playing a mitsuji pattern. Obviously, the opposite is not true: tsuzuke patterns cannot be played in mitsuji style because that would mean removing three half beats, which would disrupt the steady beat that tsuzuke patterns require.
- A single line can begin in mitsuji style and switch to tsuzuke style halfway through. Sometimes this happens because the line has more than 12 syllables. It also might be dictated by the melodic setting, when ornaments require an expansion of the beat. Or, it might reflect a drum convention, such as the practice of having the kotsuzumi play a tsuzuke-like uchikiri to accompany the repeated line of a shidai (a feature I’ll discuss in more detail later).
- A line that begins in tsuzuke style always stays in tsuzuke style. It never switches to mitsuji style halfway through.
A Closer Look at Mitsuji Patterns
Let’s look at the mitsuji patterns of the Kadono school of ōtsuzumi players and the Ko school of kotsuzumi players. In the following example, the ōtsuzumi is notated above the top line, with the kotsuzumi below. Just for fun, I’ve written the drum calls in katakana script, as they appear in Japanese. If you’ve read this far, I’m guessing you either know how to read them already, or are willing to invest the minor effort required to learn the very small number of symbols that will enable you to read traditional drum scores. To keep it graphically simple, I’ve left out the dotted half-beat lines.
Mitsuji Drum Patterns (Kadono and Ko schools)
The open triangle in the ōtsuzumi part is a loud, full hit called chon. The open circle in the kotsuzumi part is a round, full sound called po. The black circle is a softer, sharper and higher sound called chi. Since the type of hit isn’t relevant to our discussion of timing, that’s all I’ll say about them here.
I’ll record how this sounds if sung exactly as written, with a straight beat in tsuzuke style. To my knowledge this would not happen in noh because if the whole line is set with a straight beat it makes more sense to use a tsuzuke pattern in at least one of the drums. Still, I’m hoping that if we start this way it will be easier to see how the beat is altered when a line is set in mitsuji style. As you listen to the following examples, please keep in mind that I’m not a professional drummer and don’t pretend to be; I can only hope that my playing is good enough to get the basic concepts across.
Now let’s look at the same line with words added, and see where the drum calls and beats fall in relation to the text when the line is sung with the basic tsuzuke extensions (mochi).
Note that the ōtsuzumi hit on beat 3 falls on the fourth syllable “nald” and the kotsuzumi hit on beat 5 falls on the seventh syllable “farm.” Here’s what it sounds like:
Now, to obtain the mitsuji style, we’ll remove the three half beats as shown by the X’s. This means that the syllables are now chanted evenly, without pause or extension.
Notice that the drum hits continue to be notated as before, on beats 3, 5, 7 and 8. But because half beats are removed at the X’s, the syllables as chanted (but not as notated) shift to the left. This means that the ōtsuzumi hit on beat 3 now falls on “had,” (the fifth syllable of the line) instead of “nald” (fourth syllable). Similarly, the kotsuzumi hit on beat 5 now falls on “E” (eighth syllable) instead of “farm” (seventh syllable). To indicate this shift, it’s customary to insert diagonal lines that connect the drum hits with the respective syllables where they actually fall. I should note here that X’s are not used in traditional Japanese notation—the spaces where the X’s fall are simply left blank.
Here’s what it sounds like:
Some readers might find this to be an unnecessarily complicated and eccentric notational system. I know I thought so when I first learned about it. Why not write it like the following, which more clearly reflects how the line is actually performed?
The answer, at least in part, is that there are many different schools of noh chanting and drumming, so that “how a line is actually performed” varies depending on which schools are involved. There is no definitive version of any noh in the repertory; rather, when we talk about a piece like Hagoromo, for example, we’re talking about a cluster of equally valid versions. In most cases the variation is slight, but sometimes the text of whole sections of plays is different, which of course would also result in completely different musical settings. It would be a hugely daunting and impractical task to make and use scores that are individually tailored to every possible combination of performance forces. In some versions, a given line might be cast in mitsuji style; in others, the same line might be tsuzuke. What is needed, then, is a single format that can always accommodate tsuzuke, which takes up the most space on the page. Then, if a line is set in the more compact mitsuji style, it can be notated with diagonal lines without having to modify the basic format. Once one gets used to it, it’s a perfectly serviceable, ingenious and elegant notation system. Keep in mind, too, that experienced professionals simply note the names of drum patterns alongside the noh text and don’t bother to write out full scores at all. The kind of notation we’re now discussing (called yatsuwari) is used primarily by young professionals just starting to learn their art and by amateur students.
But How Do You Play It?
The uneven mitsuji meter presents a challenge for performers. Counting (or even feeling) the beats is problematic because they don’t proceed at a regular pace. And yet the drum patterns are conceived within an eight-beat framework, with calls and hits tied to specific, numbered beats. How do performers stay on track?
The main way they do it is by using komi: silent points of concentration that take the physical form of an exhaled breath cut suddenly short and held, creating sharp pressure on the lower abdomen at a precise moment in time. (It’s a lot easier to do a komi than it is to describe one!) Komi are used by all noh performers, and by practitioners of other traditional Japanese arts like kabuki and bunraku as well; they help the ensemble stay together, since there is no conductor or other outside eye/ear to provide a guiding hand during performance.
The role of komi is especially crucial in mitsuji style passages because the meter is irregular. Professional drummers know exactly where the beat lies, of course, but they do not count time. Instead, they listen to the chanted syllables and insert komi at predetermined points. The komi act as markers, or springboards, for the drummers’ calls and hits. This generates give and take between the drummers and the singers that enables flexible timing. No one is trying to keep the beat in the sense of counting out a meter; the singers, for example, enjoy leeway with syllable ornamentation and can otherwise alter the tempo, and the drummers can accommodate it without getting lost—all they have to do is wait for the syllable that gives them their cue, and insert a komi. If they get the komi right, the subsequent drum call and hit will naturally correspond with the correct syllables. Komi are often notated as small x’s, like those shown below.
The first komi occurs in the ōtsuzumi part on beat 2, which always coincides with the third syllable of the chanted line (assuming, as we do here, that the line has 12 syllables). The performance of a drum call and hit, then, becomes a three-stage process, with each stage separated by a half-beat. The drummer “takes a komi” (a direct translation from the Japanese) on the third syllable, performs the drum call a half-beat later, and hits the drum a half-beat after that. In this way the hit will always coincide with the fifth chanted syllable.
The komi shown on beats 4 and 6 in the kotsuzumi part are performed in the same way, although in actual practice the drummer will often use the hit of the ōtsuzumi instead of the chanted text as a cue for the komi on beat 4. After hearing the ōtsuzumi hit, the kotsuzumi drummer takes a komi on the next half-beat and executes the call-and-hit from there. That puts the hit on the eighth syllable, providing the cue for the final komi a half-beat later, which in turn sets the drummer up to complete the pattern with hits on the 11th syllable and the final rest.
Komi are usually performed silently, but for demonstration purposes I’ll make them audible in the following recording so that you can hear how they work.
We’re almost done! Let’s go one step further now and introduce how the Kadono and Ko schools play their tsuzuke patterns. Remember, the term tsuzuke refers both to particular patterns played respectively by the ōtsuzumi and kotsuzumi drums, and a category of patterns that are played with a straight beat. I’m talking here about the particular patterns named tsuzuke, which look like this:
The black triangle in the ōtsuzumi part is a soft hit called a tsu.
No komi are notated because komi aren’t needed to perform tsuzuke patterns. The meter is regular, so the drummers need only keep a steady beat to match up with the correct syllables of chanted text. It’s the singers who must adjust by extending certain syllables, as we’ve already discussed.
Here’s what it sounds like:
Keep in mind that both drums have many other patterns belonging to the tsuzuke category that could be substituted for these particular patterns without changing the way the section is sung. I’ve limited the example to these patterns to keep things simple.
Finally, here’s a section that switches between mitsuji and tsuzuke styles, this time with the drum patterns included:
I realize that this has been a long and involved digression, and I thank you for your patience. I think we’re ready, now, to look at the musical structure of a shidai.