If you’ve sung for long in a church choir, you’ve probably been told to make sure you pronounce all final consonants. You might also have been told to make sure your t’s don’t sound like d’s. You might even have been told to drop your r’s before consonants.
If you’ve been told these things, you’ve been told wrong.
You’ve been told to sing according to outdated rules originating two centuries ago to standardize the pronunciation of “stage English” by Americans from different parts of the country, by immigrants whose English was often poor, and by non-English-speakers singing English-language opera. When the rules were written, Victorian Britain set the standard of culture for the English-speaking world, so American theater people, panting as always to appear cultured, consciously imitated English ways, even to the point of assuming English accents.
Things changed with the advent of talking motion pictures, which were mostly made by Americans for Americans. By then, however, the old rules had become entrenched in high-brow East Coast theater schools like Juilliard. As late as 1946, Juilliard’s authority on singing diction, Madeleine Marshall, wrote a book spelling out rules like “never sing r before a consonant (no exceptions).” Marshall judged the singing of r’s before consonants “unpleasant” and instructed students to sing “chahm” for charm and “haht” for heart, declaring it “obvious” that these patently British pronunciations are more “euphonious and natural.”
This is nonsense, of course. The proof is right here in Jack Jones’s 1964 hit “Dear Heart.” Jones sings the r’s in both “dear” and “heart,” and the result is both natural and euphonious.
Unfortunately, Marshall’s outdated-when-written book is “still the standard” for some amateur choir directors and singers who may know a lot about music but don’t know enough about the English language, especially in its American form.
I’m no authority on music, but I have a degree in English, I make a living writing and editing English to be both printed and spoken, I collect and read books on the language, and I have several decades of experience speaking, reading, singing, and chanting in church. Here are a few things I would think to pass on to people who sing or chant in church. They mostly concern solo chanting, but some also apply to singing as part of a choir:
- Chant clearly and naturally. That means avoiding laziness, hastiness, and affectation. Pronounce words clearly and correctly, but don’t assume a foreign accent or some other odd pronunciation you wouldn’t use in normal speech.
- Use American English, not British English. One major difference between British and American English is the pronunciation of intervocalic t’s — t’s between vowels. Technically speaking, Americans voice their intervocalic t’s; Brits do not. In other words, Brits stop the flow of air completely for intervocalic t’s, so that “it is” becomes simply “tis.” Americans do not stop the flow of air completely and instead pronounce intervocalic t’s just like intervocalic d’s, so that “it is” sounds just like “id is.” It’s no use arguing that t’s should sound different from d’s; they just don’t in American English when they come between vowels. So American deacons should say, “Led us attend,” and not, “Let tus attend,” as an English deacon might.
- Don’t over-do the pronunciation of final consonants. Sometimes choir directors stress the pronunciation of final consonants for the sake of control: They want to hear everyone finishing at the same time. Singers sometimes respond by making their final consonants more noticeable than is natural in English. The most enduring and determinative characteristic of English is heavy frontal stress, de-emphasizing the ends of words, which are often muted, slurred, or simply dropped. For Marcus, we say Mark. For valiant we say VAL-yent, not vay-lee-ant. For single syllable words, we stress the initiation of the word, not its conclusion. For example, when we say or sing the word Lord, we stress the first half of the word and let the volume fall off at the end. This falling off of stress at the end of words was an aspect of human speech that computers could not imitate when they were first taught to speak, which is why they sounded so unnatural. When chanters consciously stress the ends of English words, they also sound unnatural. Instead of simply singing lord or wisdom, they will sing “Lor-duh” or “wisdom-muh.”
- Don’t turn d’s into t’s. D’s in both American English and British English are voiced consonants, meaning the flow of air is not stopped abruptly. This makes final d’s sound much softer in English than in other languages. A sales girl from Israel once tried to sell me soap made with salt from the “detsy,” by which she meant the Dead Sea. Likewise, Afrikaners in South Africa will tend to say “lort” for lord, as will some American chanters who consciously try to make their final d’s distinctly audible. Instead of chanting, “Lord have mercy, Lord have mercy, Lord have mercy,” they will chant, “Lort have mercy, Lort have mercy, Lort have mercy.”
- Some final consonants do warrant careful attention. The obvious example is the d in and. If you are speaking or singing very slowly, you should always pronounce the final d, but if you are speaking or singing at normal speed you should pronounce the d before vowels but often not before consonants because the consonants themselves will terminate the sound of and in the place of the d. When I switched from OCA to ROCOR, I had to go from saying “sins and transgressions” to saying “sins and offences,” which meant adding the d to and. The first few times, the phrase caught me by surprise and came out “sins an offences,” which sounds lazy. Even worse, when you drop the d in and before a vowel, you speed up the phrase and lose the distinction necessary for comprehension: “Sins and offences” could be heard as “sins in offences” or even “sins inoffensive.” On the other hand, if you consciously pronounce the d in and before a consonant, at normal speed you will inadvertently insert an extra syllable to mark the end of and: “Sins and transgressions” will be heard as “sins anduh transgressions.”
- Use traditional English pronunciations of ancient names. Some Bibles mark names to show which syllables to stress, but the dictionary is a better authority. You might not find all the names you need, but you will find many, and after looking up a few, you’ll begin to understand how English handles foreign names. For instance, you will find in the dictionary that the traditional pronunciation of Pontius Pilate is PON-chus PIE-let. Saying Pon-tee-us PIE-let mixes Latin and English pronunciations, because the Latin pronunciation of Pilate is pea-LOT-us. There is no reason to do this. PON-chus PIE-let is what we commonly say, so PON-chus PIE-let should be what we commonly sing. Another example: For “Myra in Lycia,” sing MY-ruh in LY-sha (English), not MEER-ruh in lick-KEE-ah (Greek).
- Respect the natural rhythm of English. Heavy frontal emphasis also makes English a “stress-timed” language composed of long and short syllables of varying length, in contrast to syllable-timed languages like French and Spanish. French-speakers say, “Excusez moi” (short-short-short-short); English-speakers say, “Ex-CUSE me” (short-long-short). Allotting every syllable an equal value sounds unnatural in English. We don’t say, “Con-stan-ti-no-ple”; we say “CON-STAN-tiNOple,” expressing five syllables in just three meters or beats. Many short English words are also quite complex phonetically, including four or five distinct sounds or phonemes in a single syllable. For example, the word shouldst makes five distinct sounds, including a glottal stop in the place of the d; its single syllable therefore takes more time to say than a less complex word like in. Chanters should therefore follow the language, reading phrase by phrase, taking the time to say long or complex words, and not rushing every syllable out without respect to the language’s natural rhythm.
- Don’t think glottal stops are always to be avoided. A glottal stop closes the glottis (the gap between vocal cords) to stop the flow of air momentarily. It’s how we say “uh-oh” and “unh-uh.” We all do this more times that we realize, often to reinforce a voiceless consonant like t, also often to replace a t that is difficult to say or obtrusive when said. (The cockney dialect of the GEICO Gecko uses glottal stops to replace many hard consonants, including the t and second c in motorcycle, which becomes “MO-uh-cy-ul” because Brits also drop their r’s.) Americans commonly use glottal stops in words ending with an unstressed t-n or d-n such as mutton, button, batten, bitten, sadden, gladden. For bitten, we don’t say “BIT-ten”; we say “BIT’n” — we touch the tongue to the ridge behind our upper front teeth, stop the flow of air with our vocal cords, then release air through our nose without moving our tongue to nasalize the final syllable. The same is true of the words written and begotten, which are usually pronounced RIT’n and be-GOT’n. There is nothing wrong with this; it’s just how these words are said. The only time we would not nasalize the final syllable in these words is when we are speaking or singing very slowly. If we are chanting at normal speed, the pronunciation of the second t will sound intrusive and unnatural. (Brits sometimes use a glottal stop to replace both t’s in words like bitter, which becomes BI’uh, whereas Americans invariably pronounce bitter the same way they pronounce bidder.)
- When singing or chanting very slowly and avoiding glottal stops in words like written and mutton, don’t shorten the vowel in the final syllable: Sing WRI-ten instead of WRI-tin and MUH-tun instead of MUH-tin.
- Finally, chant clearly and naturally, avoiding laziness, hastiness, and affectation. Pronounce words clearly and correctly, but don’t assume a foreign accent or some other odd pronunciation you wouldn’t use in normal speech.
I’ve said that already, but it’s worth repeating. It’s the first and last rule of chanting. If Nos. 2 through 9 sound too technical, just remember Nos. 1 and 10, and you’ll do well.