The Voices in Your Head: Some Thoughts on Choral Music
A recent Chorus America report estimates that there are over 250,000 choruses in the U.S.―12,000 (professional and community), 38,000 (school), and 200,000 (church). Those numbers suggest two things to me: 1) there must be something fun/interesting/satisfying about that kind of music making if it attracts so many people, and 2) that’s quite a large group of potential performers for new music.
There’s always room for new good literature in any genre, but composing for chorus can be particularly rewarding. You become part of a truly established tradition, choral concerts have a built-in audience of friends of relatives, and choruses exist everywhere.
Speaking personally, my attraction to choral music is just like my attraction to anything else: nothing sounds quite like it, and it can do things that no other ensemble can do. The following observations and guidelines are a reflection on my own perceptions, and I share them with you in the hope that they will stimulate your own thinking about writing for chorus. Shouldn’t your composing diet include choral music?
A Working Definition
The word “chorus” is often loosely applied to various vocal ensembles, so it might be good to start out with a definition. The classic choral sound consists of multiple singers on a single voice part. While there are excellent one-on-a-part ensembles, multiple voices on a part really defines the choral sound.
Mixed chorus (traditional soprano, alto, tenor, and bass voices) is what most people think of when they imagine the choral sound. Treble chorus (soprano and alto voices) and male chorus (tenor and bass voices) are also standard ensemble types. Sometimes treble chorus is used interchangeably with children’s chorus, although not all treble choir music is intended for children. Most of my comments will apply equally to all types of choral music.
A chamber choir (whether mixed, treble, or male) usually numbers between 12 and 20 individual singers, distributed evenly between the voice parts. It may range up to about 30 singers. Larger choruses (up into the hundreds) are actually quite common. In some cases, 8 singers can make a good choral sound, but music with independent lines tends to emphasize the individual voices in an ensemble that small.
Overall Sound and Texture
Voicing that patterns itself after the harmonic series (wide intervals on the bottom, narrow ones on the top) can keep things clear, especially in more complex sonorities. However, this must be weighed against tessitura issues, the most important of which is the tendency of the basses to drop off severely in volume as they get to the lower end of the staff. Chords forte or louder may be unbalanced if the basses are at F (sometimes even G) or lower. At the other end of the scale, high tenor notes may be too prominent in louder dynamics. They have a ring unlike anything else (which can be used to an advantage). In general, at loud dynamics, keep the tessitura conservative for all voice parts if you want pitch clarity throughout. Also, be aware that long passages of unrelieved extreme tessitura (high or low) are fatiguing to sing.
A fundamental aural concept of a chorus is that it is a more diffuse band of sound in comparison to almost any instrumental ensemble. Even simple simultaneous cross-rhythms are often perceived as mere texture. It takes fewer individual notes in a closely voiced chord to produce a dense cluster, but even a single voice part singing a unison line can be a satisfying and complex sound. Two voice parts singing in octaves or twelfths also has a richness all its own.
One of the surest ways to dull choral writing is to have all voice parts singing all the time, especially in homophonic passages. Using fewer voices doesn’t necessarily mean fewer pitches; divisi passages are common throughout the literature. Divisi a 2 is the most natural and is easiest to balance. Many choirs, especially large community choirs, will find divisi among the sopranos and altos a little more manageable than among tenors and basses. Although with more advanced choirs, divisi in all parts is idiomatic. Another way to obtain textural variety is to consider a short passage for solo voice. Incidental solos within the chorus are found frequently.
Doubling voices with instruments, especially piano, is not necessarily the best way to ensure pitch stability. Voices in a chorus hear and tune to each other more comfortably than they do to an “external” sound that’s 15 feet away. Tuning with the person next to you happens without too much conscious effort. Remember, it’s a built-in part of our biological equipment to be particularly sensitive to vocal sounds. For any chorus and instrument combination, more independence in the instruments and less straight doubling will result in a better overall sound. A simple reinforcement of important pitches in a line or decoration around the sung material are possibilities to explore.
Note that consecutive wide melodic intervals are difficult to manage vocally, even for professional ensembles. Whether chromatic or diatonic, a preponderance of stepwise motion is more natural for singers. For non-professional choruses especially, consider using repeated patterns or intervallic motion that implies a perceptible harmonic structure. Individual voice parts that are relatively simple melodically can be layered to produce a very complex (and ear-catching) vertical texture.
Chromatic music needs to be notated carefully. Often, you have to choose between being clear vertically and being clear horizontally. Two adjacent chords may be easy to parse individually, but if there’s a leap of a diminished sixth between the two chords in one voice part, for example, you may want to respell. However, if you are only looking at the individual voice line, the altos may glance at the tenor part and miss the fact that their Db is in unison with the tenor C#.
Rhythm must be considered differently when writing for chorus, especially rhythm in an active texture. First, much of the rhythmic activity is determined by text (more about text below). Second, singers sing vowels (this is an oversimplification, but a useful one). The individual notes of a melisma will have a roundness that makes the rhythm less differentiated than the same line would be on a wind instrument, for example. This does not imply any lack of energy in the sound, but will influence the rhythmic perception in the listener. Setting text syllabically is one solution, but to avoid melisma entirely ignores an important musical feature of the chorus as sound.
Choruses frequently sing in acoustically reverberant spaces―more resonant than those used for chamber music. This also diffuses the rhythmic perception. Even in a drier space, choruses (and their directors) have the resonant acoustic in their aural image and will often try to evoke that kind of space.
Here are two tricks I’ve encountered that will sharpen up the rhythmic profile. First, breaking up short phrases of an active texture with silence will heighten the rhythmic effect. Even if it is only a beat or two, a repeated stop-start effect gives the listener a set of guideposts that frame the rhythm. Second, I often see (and use) textures where one voice part is delivering the text while the other parts are humming or singing “ah”. Consider using “ka” or “ta” (or “km” or “tm” for humming) occasionally, either at the phrase or sub-phrase end, or when the pitch changes. “k” and “t” are consonants that carry effectively when sung by a group. They give a much sharper explosion than “b” or “p”.
Another thing to consider is dynamics. For chorus, fast and loud can easily be just loud energy―which may be your intent. There is excitement in a large group of musicians (vocal or instrumental) playing loud, even if the actual decibel level isn’t high: 4 brass players may drown out 40 singers, but the 40 singers communicate something beyond the volume just because of their numbers. However, fast and soft is a marvelous choral texture when done well. Again, 40 singers communicate something very intense when singing a fast texture softly. A classic example: the beginning of “Variation VI—Finale” from Britten’s A Boy Was Born.
Text is the most important characteristic of any sung sound. However, as composers, our mechanism is sound and making sonic connections, and one’s motivation for writing choral music should be more than just presenting the text. Our job is to illuminate the text, marrying it with music to make something more.
This brings me to a blessing and a curse that is at the heart of choral writing: the best way to ensure the intelligibility of text is to set it homophonically at a moderate tempo. But the sure way to turn listeners off is to have many or all pieces on a choral concert where the text is set homophonically at a moderate tempo. So instead, let me posit a continuum, with absolute clarity of text at one end and no clarity at all at the other. Each piece will fall somewhere along that continuum, and I believe it is our job to constantly rethink this aspect of choral writing carefully, determining where a particular piece (or section of a piece) will fall along that continuum. I think it is a mistake to assume that absolute clarity is desirable all the time.
It can be helpful to ask yourself why or what about a particular text appeals to you. Why do you think this text is important enough to set? Is it the sound of the words? Is it the images? Is it the narrative? Is there a narrative? Is it merely a convenient vehicle for your musical ideas, given that you’re writing for chorus and have to use something? The answers to these questions might help you find a place on the intelligibility scale, and they might also suggest musical character, tempo, gesture, etc. Text can even govern form and flow. Connections in the text at a level beyond the individual words or individual poetic lines can suggest musical connections. Don’t settle for just setting line after line of text with your eyes closed (as it were).
It’s probably best to avoid declamatory, information-leaden texts. They will push you toward the “must be intelligible at all times” end of the continuum, and thus may limit your musical choices. At the other end of the scale, intensely personal texts can be a challenge; such texts are effective when sung by a soloist, but can become mawkish when sung by a choir. I used to have a rule (which I’ve relaxed) that I didn’t set texts for chorus that included the word “I”.
The use of traditional Latin texts, especially mass parts, can be problematic. I have seen Kyrie settings where the text choice seemed offhand, as if the composer felt “oh, that’s what choruses do”. Remember, our job is to illuminate the text. If you are writing for a liturgical setting and aren’t specifically asked to write a mass, consider less well-known texts: a passage from the book of Revelation, Hildegard of Bingen (try the original Latin), Mechtild of Magdeburg, a passage from one of the Old Testament prophets (perhaps in Hebrew). Also consider a troped text, where lines of the original text are interspersed with commentary on that text. Or create your own trope by mashing one text up with another.
For something truly new, consider collaborating with a poet, having someone write a new text specifically for your piece. Poets are less familiar with writing on commission or to order, but many of them will welcome the challenge. Although the poet’s work will probably be done before you start yours, a made-to-order text means there is some flexibility: you can go back and ask for a rewrite (perhaps to get some more words with a “k” sound to emphasize your rhythmic idea), work with the poet if you need to truncate text, or ask for more text as your piece develops. I’ve done this several times, and I found the process both stimulating and educational.
You may be surprised at how different poets regard their words. Just as we composers are all up and down the spectrum concerning our notes (some of us are very tied to the notes we write, others don’t mind so much if they get jostled around a bit by the performers), so are poets with their words. Some feel that their text must come through. Others are more concerned with mood or flow. In any case, it is an interesting topic for discussion between you and your poet-collaborator. And what better way to get into a text than having the author right in the room with you (or at the other end of the email chain).
And finally, if the choral sound appeals to you as pure sound, treat it that way by omitting text altogether. Invent your own “language” or just use phonemes that have no meaning. This approach isn’t for everyone, and some directors are put off by such pieces, but it can be an intense experience to bury yourself in the choral sound divorced from word meaning. Even if you don’t write an entire piece using phonemes, there might be a situation where that’s just the ticket for a section of a piece.
Text: Practical Considerations
Be aware that text is mostly lost at the extreme registers. Vowel sounds more or less merge together once the sopranos get up into the ledger lines. Low basses can keep words well-defined as they rumble down into the ledger lines, but pitch starts to lose a central focus as they get low, in addition to the dynamic problem mentioned above.
For basic non-text passages, such as “ah” or “m”, normal letters can be used. However, if you have extended passages (or a whole piece) that uses phonemes, consider notating it using IPA symbols (International Phonetic Alphabet). Not only will you have more precise control over the resultant sounds, but it will be easier for the performers as well. Most trained singers are familiar with IPA symbols. Also, there are many IPA fonts available (some free). Alfred Blatter’s Instrumentation and Orchestration (see “Resources” below) has a basic table of IPA symbols; Google “IPA font” for the latest information and sources.
Always obtain permission to use any text that isn’t in the public domain before you start working. See Stephen Paulus’s “Before You Set Those Words to Music” for specific guidance on negotiating the legal authorization to set text that is under copyright.
Choral singers almost always read from a score. If your piece is for chorus and piano, everyone reads from the score. If you are writing a piece for chorus and only a few instruments, full scores are fine for the chorus. The one exception is a large orchestra piece with chorus, where you sometimes encounter one-line individual voice parts. Singers hate this. Best is a reduced score, which includes not only all the voice parts, but also a one- or two-staff cue line. The cue line may be laid out as a piano reduction, which is also useful in rehearsal situations.
There’s no hard and fast rule about including a piano reduction as a rehearsal aid for unaccompanied choral pieces. However, its inclusion may imply that you sanction a performance of your piece with piano doubling the voice parts (the “for rehearsal only” next to the reduction staves notwithstanding). Whether to include one or not may also be dictated by the content of your piece. If your music contains tongue clicks, whispered, spoken, and other non-pitched material, many glissandi, or if the writing is such that the resultant reduction is mostly unplayable, a reduction has limited usefulness.
If you do include a reduction, spend some time making it piano-friendly if possible: redistribute the notes so they fit into both hands, simplify where complex voice leading/overlapping creates fussy keyboard writing, and omit a few notes here and there if they’re beyond reach. Letting your notation program create your reduction is a timesaver, but that is only the first step.
Singers usually hold their music and only rarely use music stands. This means that large size and landscape format scores are awkward. Traditional for chorus is octavo size is (6¾ x 10½), but I’ve never encountered an objection to plain old 8½ x 11 (or A4). If your piece includes handclaps or some other technique where one or both hands need to be free, mention in the front of the score that the chorus will need music stands. One thing to note in this regard is that children’s choirs usually perform from memory, which eliminates the stand issue.
A related problem is the number of pages in a score. If your piece is long and contains a lot of divisi, you’ll end up with a thick score that is fatiguing to hold. I frequently start by notating my piece on eight staves (SSAATTBB), then go back and collapse to fewer staves where there are passages that are non-divisi, or where the divisi is such that a single voice part can be notated clearly on one staff. Obviously, a different staff layout for each system makes for a confusing score, but if there is an extended passage that can be reduced, your singers will appreciate your effort to cut down on the number of pages. Consider using a smaller staff size than you would for an instrumental piece. Singers are not limited by the distance to the music stand; they can hold the score closer to their eyes. This may allow you to get multiple systems on a page and reduce the number of pages.
Resources: Recommended Books and CDs
Orchestration books are notoriously spotty on the voice and especially the chorus. One that does cover the voice is Alfred Blatter’s Instrumentation and Orchestration (Macmillan, 1997 [2nd edition] ISBN-13: 978-0028645704). Another (older and expensive—check your library) is Andrew Stiller’s Handbook of Instrumentation (Univ of California Press, 1985, ISBN-13: 978-0520044234). There is one survey of recent choral literature: Nick Strimple’s Choral Music in the Twentieth Century (Amadeus, ISBN-13: 978-1574671223). It’s comprehensive, available in paper, and is relatively inexpensive.
Some excellent anthology CDs:
- The Dale Warland Singers are no more (alas!), however their recorded legacy is still available from Gothic Recordings. Of particular interest: Bernstein & Britten (also includes works by Paulus, Albright, and others) and Reincarnations (Fine, Avshalomov, Finney, Barber and Ives). Also, check out Choral Currents from Innova Recordings (Barnett, Franklin, Hodkinson, Larsen, and others)
- Gregg Smith has been a friend of composers for a long time. His 20th Century Choral Music in Space is a fine anthology (Druckman, Talma, Hawley, Gould, and others), and his I Hear America Singing (Rorem, Schuman, Talma) is also worth seeking out.
- Of Eternal Light from Musica Sacra is a wonderfully diverse collection, ranging from Monk (Meredith, not Thelonious) to Messiaen. Also includes works by Ricky Ian Gordon, Kim Sherman, and Robert Moran. And you get the Ligeti Lux Aeterna, a true masterwork of choral literature.
- Masters of 20th Century A Cappella, Danish National Radio Chamber Choir, directed by Stefan Parkman on Chandos: masterful choral writing beautifully performed; includes works by Schoenberg, Poulenc, Henze, Lidholm and Nørgård.
- Colors of Love, Chanticleer (Teldec), a fine cross-section of new music: Steven Stucky, John Tavener, Bernard Rands, Zhou Long, Chen Yi, Augusta Read Thomas, and Steven Sametz.
- Baltic Voices (vol 1, vol 2, vol 3, separate CDs), Estonian Philharmonic Chamber Choir, directed by Paul Hillier (harmonia mundi): also wonderful samplers; composers range from Górecki to Pärt to Rautavaara to Saariaho.
The following choruses have a strong commitment to living composers. Their websites (especially their repertoire pages) will suggest further listening:
- The Esoterics (Seattle)
- VocalEssence (Minneapolis)
- Volti (San Francisco)
- New York Virtuoso Singers (New York)
- Piedmont Children’s Choir (Piedmont, CA)
- San Francisco Girl’s Chorus (San Francisco)
Special thanks to composers Elliott Gyger and Stacy Garrop, and especially to conductor Robert Geary for their thoughtful discussions and input to this article.
Mark Winges has been resident composer/advisor for Volti (formerly The San Francisco Chamber Singers) since 1990, and has written works for them ranging from a cappella to chorus and chamber orchestra. His choral works have also been performed by the Piedmont Children’s Choir, the San Francisco Girl’s Chorus, the Pharos Music Project (NY), Carmina Slovenica (Slovenia), the Guangdong Choir (China) and many others.