Suggestions for Contemporary Composers on Setting English to Music

As an American singer that sings primarily English-language repertoire, I've really gotten a lot of satisfaction out of being able to deliver quality music to people in my native language - to an audience that understands that language. I've participated in a lot of world premieres for someone my age, and I'm acquainted with a good number of contemporary composers - many of whom want to write for voice. This post is mostly dedicated to those people, offering insight into what you can do to enhance the ease, intelligibility, and linguistic faithfulness of your English settings.

I Can't Stress This Enough

The most common problem I see in contemporary English settings is the misuse of linguistic stress. Stress can be applied to a given musical sound in these ways (among others)

1. Increasing the duration of the sound
2. Increasing the volume of the sound
3. Making the sound happen at an extreme pitch level
4. Articulation
5. Changing tone quality
6. Placement on strong beats
7. Assigning one of the above characteristics (duration, volume, pitch, articulation, tone, beat placement) in a manner that is obviously different from what surrounds it

There you go. Short list. Essentially there's only one common way to apply stress to a sound in spoken English: duration. Think of that word, "duration". The middle syllable is slightly longer, which gives it its stress. Longer syllables give English its rhythm, making shorter syllables unstressed and less important by comparison. In order to set this word sensibly, you would use items from my "Stress List" above to add stress to long/stressed syllables, and avoid applying items from the list to the short/unstressed syllables. To put it simply, you don't want it to sound like "DUraTION" or "duraTION" or anything else weird. Here are some options for how to set this word:

In the first example, the unstressed syllable "du" is being accented by its placement on beat one, and the stressed syllable "ra" is being un-accented by its placement on a weak beat. This is awful text setting. The second example is better, as the stressed and unstressed syllables are better aligned with correspondingly weak and strong beats. The third example is best, as the written rhythm mirrors that of speech. The fourth example demonstrates an effort to write correct stress allocation, but when speaking this, you can hear that the center syllable sounds unnaturally long.

This is also apparent on a more macro level, as entire words can be stressed or unstressed in order to affect meaning. Consider these options:

Obviously the word "tide" is the most important. In the first example, all words are given equal duration, but the placement of "the" (one of the least important English words) on the downbeat is incorrect. The second example shows the massive effect that duration has on stress, as shortening "the" by just an 8th note unstresses it quite a lot, thereby transferring more stress to the important word. The third example is better yet, and shows a more complex rhythm that one might see in a contemporary composition. This one gives both a strong beat and a rhythmic accent to the important word. Again we have an idiot-proof version at the end, perhaps not the most "artistic" rhythm, but one that will be clear and hard to mess up.

The Harder it Is, The Harder it Is to Understand

What is the goal of competent text setting? Intelligibility. It doesn't do you any good to pick or write a brilliant text if no one can understand it. Consider the following:

Melodically and rhythmically active, however...

...this version is simpler in every way, while still showing proper linguistic stress.

In a more demanding register, EVERY syllable becomes stressed.

The 5/4 meter of the first example does nothing to aid text delivery and only adds complexity for complexity's sake. The second example isn't interesting, but is a very simple rhythmic setting that is faithful to the text's inherent rhythm. The third example accents everything by being in a difficult register, as according to my Stress List, pitch is a way of adding stress. 

NB: Just as with every other instrument, the four main types of singing voices each have ranges, registers, prime registers, sub-prime registers, and extreme registers. It is your responsibility to know them just as you you do wind instruments, etc.

Flying Planes Over the Orchestra

Bernoulli's Principle, the scientific theory explaining the mechanics of flight, says that "as speed increases, pressure decreases". In other words, heavy things move slowly and light things move quickly. This means that if you want something to be loud, it will be easier to do if set in a slow rhythm or tempo. If you want something to be fast, it will be easier to do at a lighter dynamic. Don't get mad at me if you want to set a coloratura passage over a brass ensemble and electronics and it's coming out too quietly... I didn't invent physics.

This would be a difficult passage to understand, because it would be difficult to simply hear. A high voice (soprano or tenor) wouldn't be able to project this to any great degree, and it would be especially awful if written "forte" or heavily orchestrated. It is placed in a sub-prime area of the voice (decreasing the amount of weight than can be added), it is set rhythmically fast (decreasing the time able to be spent on each tone), and is wordy with a different sound on every note (shortening the length of each vowel). All of these issues affect the text, as the lightness, shortness, and sub-prime range conspire to obscure the sound - therefore obscuring the text. Consider this instead:

Here we have a better register (for a high voice), important words and syllables are stressed, and vowels have more time to be heard. This passage would be much more easily sung forte or with thick accompaniment, even the melismatic passage. The melisma is delivered on one vowel - serving to add a bunch of fast notes without obscuring the text whatsoever. Even though the second measure will have to be sung lighter than the first measure, the length of the vowel will help it to be heard and understood even if it can't be sung heavily. Obviously not everything can always be set in the absolute best possible way, so let's see if we can't alter the the previous passage rather than abandoning it completely.

Perhaps we're dealing with a play, short story, or poetic submission from a friend, and we don't want to change the text. Here is an alternate version that preserves the pitch landscape, rhythmic attitude, and text, without sacrificing intelligibility. This is in a better range, and the rhythms are better matched to the stresses of the language without being totally perfect.

WWBD? (He Would Speak It)

"WWBD?" of course stands for "What Would Britten Do?" Regardless of one's personal opinion on Britten (mine is that I'd be fine singing no other music), much has been written praising and proving his aptitude for setting English. Britten had one big advantage that most of us don't enjoy, which is that he worked closely with both the performers and the poets/librettists involved. He would work on a W.H. Auden setting, with Auden speaking the composition rhythmically and giving feedback on whether it was understandable. Though you won't be talking through texts with Maya Angelou and Diana Damrau any time soon, you will probably learn a lot from simply speaking through your texts. Does it sound like English? Do your meters and rhythmic divisions clarify, or do they obscure? Do the stresses, pauses, and pitch contours sound somewhat natural? And if not, is that unnaturalness justified by the text or the dramatic action at the time? Will this text be audible?

That's all I've got. And thanks for setting English, it's my favorite language in which to sing.