In a throat, a note is forming. A puff of air, a pulse of the lungs, rushes up the windpipe and through the vocal cords, parting them like a pair of lips. As the cords begin to vibrate, they’re stretched taut by muscles to either side, raising the pitch. The diaphragm pumps more air, rocketing the note up the vocal tract, making its walls hum like the barrel of a woodwind. The sound ricochets back and forth as it rises, gaining resonance with each rebound, till it bursts into the hollow chamber of the mouth, the ringing cavities of the sinuses, and careens off the palate into the open air.
The human voice is the world’s most astonishing instrument, it’s often said. It’s capable of everything from a trill to a bark to an ear-splitting scream, from growling harmonics to liquid acrobatics, lofted on the breath like a lark on an updraft. Instrument is the wrong word, really. The voice is more like a chamber ensemble: winds and strings and blaring horns, strung together end to end. It’s a pump organ, a viola, an oboe, and the bell of a trumpet, each instrument passing the sound along to the next, adding volume and overtones at every step. Throw in the percussion of the lips and tongue, and the echoing amphitheatre of the skull, and you have a full orchestra playing inside you.
When I joined a chamber choir a few years ago, my first thought was just how much could go wrong. The people weren’t the problem. Most were experienced singers, and there were nice voices all around. It was more the nature of the endeavor. A single voice is complicated enough, but adding others—we were about fifteen singers—multiplies the complexities. Who missed that cutoff? Why do we keep going flat? Even if the notes are right, the rhythms can go slack, the chords refuse to ring. Hitting pitches only gets you so far. The singers have to keep the beat, modulate volume, match vowels, and articulate consonants in perfect synch or sequence. Is that an ah or an ae, a ch or a k? And even then their unison can be soured by a single tenor.
Choral harmony is a kind of miracle when it works. It’s not just that every voice is different—that pitch and blend can vary disastrously, that memory is fitful and sight-reading uncertain, that there are too many sopranos and not enough basses—but that we hear things differently. The sound that travels from your mouth to my ear passes through yet another set of sonic permutations. It enters the auricle—the crumpled cone of the ear—and echoes through the auditory canal, strikes the eardrum, chimes the bones of the middle ear, and goes spinning down the sousaphone of the cochlea, tripping nerves inside like keys on a piano. The sound we hear is one set of instruments translated by another, like a game of telephone at the U.N. By the time the brain has decoded and combined it all—Happy Birthday!—it’s a wonder we even recognize the tune.
In a room, a chord is forming. Two voices, faint but pure, are joined by a third, softly keening. They rise and swell, twist into dissonance and fade. Deeper voices awaken with a groan, then settle into a slow, thrumming chant. Others join in, then peel away like rain from a spinning tire. They shout and wail, gather into pulsating chords and resolve, at last, into words: “There will be always something you can lean your weight into. There will be always something you can rely on.” The singers, four men and four women, are seated in a half circle around a conductor. As the final chord crests and ebbs, they dart their eyes at one another and grin, the sound still surprising after so many repetitions.
“Can someone describe what you just sang?” the conductor asks.
“They’re all just major triads. The A pattern is G, F, A-flat—all major chords. Then the B pattern is G, F, A-flat, G-flat. It’s always alternating.”
“And do we know the voicing?”
“I remember I was on the third.”
“Yeah, and the bass was on the tonic.”
“That’s what my body was telling me, but it was also telling me that I’m on the ninth.”
“And what were you telling your body?”
It’s early June, and the rehearsal room, in a rustic lodge on the shores of Lake Dunmore, in central Vermont, has been left unheated, somewhat optimistically. The indoor temperature is in the high fifties, and the singers are bundled up as if for a hike. The conductor, Brad Wells, has on a worn baseball cap and a sweatshirt, and the singers are wearing hoodies, woollens, and puffy vests, some with scarves and beanies. They have the ragamuffin look of Christmas carollers, despite their fierce-sounding name: Roomful of Teeth. They’ve been invited to the lake for a weeklong residency at the New Music on the Point festival. (When the adults leave, their cabins will be taken over by hordes of summer campers.) Mostly, though, they’re here to blow the other singers’ minds.
Roomful of Teeth is a kind of lab experiment for the human voice. Its eight singers cover a five-octave range, from grunting lows to dog-whistle highs. Three have perfect pitch, all have classical training, and Wells has brought in a succession of experts to teach them a bewildering range of other techniques: alpine yodelling, Bulgarian belting, Persian Tahrir, and Inuit and Tuvan throat singing, among others. Because the group writes or commissions almost all of its pieces, it can create vocal effects that most singers would never attempt. Roomful of Teeth’s first record won the Grammy Award for Best Chamber Music/Small Ensemble Performance in 2013. That same year, one of the group’s two mezzo-sopranos, Caroline Shaw, won the Pulitzer Prize for music for her piece “Partita for 8 Voices.” She was, at thirty, the youngest person ever to win the award.
The piece they’re rehearsing this morning was written for them by Merrill Garbus, who performs with the bassist Nate Brenner under the name Tune-Yards. The chants and howls and panting rhythms alternate with moments of sudden beauty—luminous plains swept by shimmering chords —but it’s not always clear if that’s what Garbus intended. She taught the piece to the singers a few years ago, mostly by ear, and never gave them a score. “It’s always a bit of an adventure reconstructing it from collective memory,” Wells says. Although Wells is the group’s founder and artistic director as well as its conductor, his role in rehearsals tends to be more watchful than controlling. He sits facing the singers, hands in his lap, eyes following the score on an iPad, only occasionally offering a cue or an interpretive note. “I want this to be true, but I don’t know if it is,” he tells them at one point. “I’ve heard that when the sun rises in the rain forest, the dawn chorus starts with the insects, then the birds, and then the next more-evolved species.” It would be terrific, he seems to be saying, if this piece could convey the same sense of dawning consciousness, of buzzing chaos resolving into order.
The singers ponder this for a moment.
“If I wake up at noon, does that mean I’m the most evolved?”
“I was up at six-thirty this morning, so I have webbed toes.”
Darwin believed that music and speech developed in part out of people mimicking natural sounds. Roomful of Teeth, you might say, is another step in that evolution. The group’s exotic techniques not only use different parts of the vocal anatomy; they’re sound prints of the landscapes where they were developed. In Sardinia, in a tradition known as cantu a tenòre, four men stand in a tight square and sing harmonies pitched to the sounds of their livestock: the lowest voice is a cow, the middle voices are a sheep and the sound of the wind, and the soloist on top is like a shepherd singing to his flock.
By combining such sounds with the echoing cliffs of a Swiss yodel, or the lashing rain of Korean P’ansori, and refracting them through the mind of a composer, the group produces music that’s both primal and sophisticated, ancient and startlingly modern. “There were these hard lines that had been drawn around classical voice pedagogy,” Wells says. “The mentality was: everything else is an inferior use of the voice. But if people have been using these techniques in different parts of the world for so long, how could they be wrong? I love a throaty or a belchy voice. I love voices when they crack. I love hearing grit in the voice. It’s really just pushing the bounds of what’s beautiful.”
Choral directors make unlikely revolutionaries. Their repertoire is rooted in church music, in which every interval can have meaning and harmony was once a matter for papal intervention. Christian monks sang in unison for nearly a thousand years before they allowed themselves a second vocal line, and then only in lockstep with the melody. Three-part harmony had to wait another three centuries, when English and French clergymen added a third or a sixth to the chord. And what we think of as classical harmony, with major and minor keys and chords that follow the bass line, didn’t emerge until the Renaissance. As late as the seventeen-hundreds, the tritone—a dissonant interval of two notes, three whole steps apart—was reviled as diabolus in musica: the devil in music.
Wells, in his way, is like a Protestant Reformer, nailing his radical theses to the conservatory door. He belonged to the church before he rebelled against it. Growing up in Northern California in the nineteen-seventies—he is fifty-seven—Wells played violin, trumpet, and jazz guitar, and sang everything from barbershop to the Beach Boys. But his most formative musical experiences were in his family’s Christian Science church. Mary Baker Eddy, the founder of Christian Science, took a dim view of choirs—“just more fodder for people to play their power games,” Wells says. She liked her congregations to do the singing, not as a performance but as an expression of divine harmony. But the services did allow for occasional solos. By the time Wells was in high school, he was working as a paid cantor at churches throughout the area, practicing or going to rehearsals four hours a day and thinking of a career as a singer.
“I didn’t miss a day,” he told me one afternoon. “I thought, Let’s see what this instrument is capable of.” We were sitting in a windowless studio in the library at Williams College, in Williamstown, Massachusetts, where Wells directs the vocal program and is an artist-in-residence. Tall and spare, with a shaved head and a trim beard, he still has something of the seminarian about him—a studious self-containment, a cerebral serenity. He loved the discipline of Christian Science, he told me. “It’s a kind of full-contact sport for religion. If you’re in it, you’re in it every day, studying, praying, talking—it saturates your life.” But his faith eventually foundered on some of its core tenets. At the Christian Science school where he studied voice—Principia College, in Elsah, Illinois—one of his closest friends was gay and tormented by the Church’s sexual intolerance. And Wells had known a number of Church members who died after refusing medical treatment. (Mary Baker Eddy insisted on faith healing, though she took morphine for her own illnesses.) “I just thought, If there is some universal intelligence manifested on this plane, why isn’t medicine an expression of that?” Wells told me. “And if we’re all ascending toward a form of love, why isn’t sexuality part of that?”
Wells has a deep, oaky baritone that roughens to a grumble in its lower register. He tried speaking at a higher pitch a few years ago, when a voice therapist told him that his tone was causing vocal fatigue. “But it just didn’t feel like me,” he said. As he talked, his hands moved over a laptop keyboard, splicing and shuffling a stack of vocal tracks on a pair of computer screens. He was piecing together some recordings he’d made of one of the oldest known Shaker hymns, called “Solemn Song No. 1.” Four of the tracks were solo voices—Roomful of Teeth’s Caroline Shaw and the folk singers Rhiannon Giddens, Sam Amidon, and Eamon O’Leary. The other tracks were an amateur choir that his wife, Betsy Burris, had put together with singers from her Congregational church in Williamstown. The piece was part of an installation that Wells was creating for a restored Shaker village in Hancock, Massachusetts. When the recording was done, it would play on an endless loop inside an old silo.
The Shakers, like the Christian Scientists, were believers too ardent for ordinary churches. Dissident Quakers impatient with their faith’s passivity, they shipped to New York from England in 1774 and sailed up the Hudson, reaching Hancock nine years later. Celibate by principle—they peopled their ranks with orphans and adoptees—they channelled their passions into religion, handicrafts, and music. “They’d sing as they brought the cows in from the field, sing at daily worship, sing while peeling onions,” Jennifer Trainer Thompson, the director of Hancock Shaker Village, told me. They went on to compose more than ten thousand songs, and invented a new kind of pen nib to aid in their transcription. The later hymns, often based on Appalachian folk tunes, are fairly conventional. But the early Shakers mistrusted fancy lyrics and fussy compositions, preferring wordless tunes like this one, full of the shuddering ecstasies that gave them their name. Those were the songs that Wells loved best.
After graduating from Principia, Wells earned a master’s degree in music from the University of Texas at Austin and a doctorate from Yale. But along the way he switched from singing to conducting and composing. His voice, rich as it was, couldn’t go low or high enough for operatic roles, and opera was one of the few profitable careers he could imagine as a trained singer. “I like opera, but it’s far from where I want to live all the time,” Wells told me. “And the question came up even then: Why aren’t there other places for a classically trained singer to make a living?”
It’s an old singer’s complaint: orchestras get paid but choruses don’t. Violinists can find full-time work in a string section; trombonists can earn their keep with the brass. But aside from a few prominent church choirs and opera choruses, and a handful of small ensembles like Chanticleer or the King’s Singers, choral singing is volunteer work. Even when an orchestra and a chorus perform together, in a piece like Beethoven’s Ninth, the instrumentalists are often the only ones on the clock.
Opera and solo work are the big exceptions. Their underlying technique, known as bel canto, is extraordinarily hard to master. To produce the right sound, countless muscles have to be coördinated, reflexes marshalled, and unseen body parts relaxed, until a column of air flows smoothly and evenly as the singer moves through her entire range. It’s a mixture of zen and jujitsu. First developed in the seventeenth century by the Italians—their clear, open vowels suited it perfectly—bel canto does a number of felicitous things at once. It balances the body’s resonances, fosters a natural vibrato, warms the voice with overtones, and colors it with emotion. (One researcher at Ohio State argued that vibrato might trigger emotions in listeners because it simulates the sound of a voice trembling with fear.) Above all, bel canto is loud. It carries long distances. It’s why the Russian soprano Anna Netrebko can fill a concert hall with her voice, why she can be heard above a thundering orchestra and a wailing chorus. It’s a great feat of human athleticism, beyond the capacity of most bodies. An opera star, no less than a sumo wrestler or an N.B.A. center, is a physical freak—a human bell, ringing in every hollow and bone. You’re either born that way or you’re not.
Wells was not. And the result strikes him as both a practical and an aesthetic problem. It means that most classical singers can’t sing some of the finest and most lucrative vocal repertoire, and that the music they can sing is mostly written for amateurs. Choral singing tends to mean one thing: a clear, straight tone designed to blend seamlessly with others. It can be an exacting skill, capable of great beauty and refinement. But it’s only a fraction of what the voice can do. It’s like a ballet choreographed only for the upper body, a painting with nothing but greens and blues.
Wells remembers the first time he heard a record by a Bulgarian women’s choir, in the late nineteen-eighties. Here was music that broke all the rules of Western choral singing. The women were technically belting, like Broadway singers—pushing their chest voices up into the range of their head voices—but they took it much further. The sound was nasal, emphatic, transfixing—closer to Arab ululation than to the smooth sonority of an American choir. “I mean, holy shit!” he says. “The oranges and reds and pinks! At first I was thinking, No, no, no. We’re always trying to remove those colors from the mix—to get this beautiful, uniform blend from top to bottom. But the end result is that the range of colors and timbres gets more and more narrow.”
The great project of twentieth-century music was to liberate instruments from their prescribed sounds, Wells says, citing the Dutch composer Louis Andriessen. “Now it’s time to liberate the voice.”
When Wells founded Roomful of Teeth, in 2008, he began with a set of strict conditions, like a biologist designing a study. The group would sing neither the standard choral repertoire nor songs from other vocal traditions. “I thought, We’re trying to build a new kind of instrument,” Wells told me. “So let’s force ourselves to come up with clothes cut for our particular physique.” His model was the avant-garde restaurant Alinea, in Chicago. He remembers reading a review of it in Gourmet that fall. “It talked about the hushed perfection of the place, and the silence as the diners were presented these inventive plates,” he said. After a while, the reviewer couldn’t help but start giggling, he was so surprised and delighted. “I thought, That’s the effect I want my group to have.”
Wells began by commissioning pieces from two composers: Judd Greenstein and Rinde Eckert. He persuaded MASS MoCA, a contemporary-art museum housed in an old mill complex five miles from Williams, to host a three-week residency, capped by a concert. Then he arranged for experts in yodelling, belting, and Tuvan throat singing to come teach their techniques. All he needed now was some singers.
“It felt like a very one-time thing,” Caroline Shaw recalled, when I asked about her audition for the group. “And kind of a bad idea, actually. Are we going to learn to throat-sing that quickly? Will the music be written in time? Will it make any sense? I thought, Yeah, sure, I like those Bulgarian singers. I like to yodel. But, looking back, I just didn’t know very much. I knew a lot about Beethoven string quartets.”
For the auditions, Wells set up in an apartment on the Upper West Side, on 102nd Street just off Broadway. Each singer had to perform a classical piece and a nonclassical piece. Shaw sang a medieval plainchant, “Salve Regina,” and Otis Redding’s “I’ve Been Loving You Too Long.” The mezzo-soprano Virginia Warnken Kelsey sang Bach and “Mary Jane” by Alanis Morissette. The baritone Avery Griffin sang Schumann and “Fire in the Hole” by Steely Dan. “So that was fun,” Griffin told me. “And then he had us do sight-reading exercises, and that was horrifying.” Griffin, who hails from Weehawken, New Jersey (“The smallest of the Hawkens”), has foxlike features and a skeptical turn of mind. Like Shaw, he was convinced that the group was a bad idea. “But it was kind of a dead month for gigs, and I needed the money,” he said. Wells first had him sing through some standard choral pieces. Then he tossed him a curveball: a violin part from Bartók’s String Quartet No. 1. Because it was written for a stringed instrument, the part was torture to follow. It was like translating Russian into English while singing a tune for the first time. “I cried,” Griffin said. “I left saying my life is a hollow lie.”
Singers tend to be either the worst musicians in the room or the best. Some are content to sing the same five or six operas over and over, never bothering to learn another. But those who specialize in new music, or in thorny twentieth-century pieces, have to be as adept as they are adventurous. Voices have no frets or keys to place the pitches; no brass or wood—impervious to running noses—to make the sound. Wells invited pop and musical-theatre performers to audition, as well as classically trained singers, but only the latter had the right technical chops. “He needed people who could just look at a score and know it, who could get it together in one day,” Kelsey told me. “They had to be able to name tone rows, or pick out a single note from a tone cluster. It was like an advanced-musicianship test from grad school.”
The eight singers who made the cut were all in their twenties. Four were trained instrumentalists: Shaw was a violinist, the soprano Martha Cluver was a violist, the bass Cameron Beauchamp played trombone, and the tenor Eric Dudley studied piano. Among the others, Kelsey, the soprano Estelí Gomez, and the bass-baritone Dashon Burton all studied voice, but with an eye toward contemporary music. As for Griffin, he had trained as both a singer and a composer, but he thinks it was something else that got him in. Even in the Bartók, he says, when the intervals and rhythms nearly brought him to tears, he kept plowing ahead: “I think that’s what Brad really wanted to see: could you keep on going. Because that openness and bravery may be the primary thing about this group.”
One afternoon this past August, in a loft on the second floor of one of the mills at MASS MoCA, a young composer named Harry Stafylakis was having the group sing death metal. Tall and piratical, with a man bun and a pointy beard, Stafylakis studied piano as a boy in Montreal then dumped it for heavy-metal guitar. “I was obsessed with Beethoven,” he said. “That was my gateway to metal.” Having circled back to classical music in college, he still favored doomsday chords and jackhammer rhythms. He was planning to compose a piece for Roomful of Teeth based on a Web site called the Dictionary of Obscure Sorrows.
Stafylakis was one of eight composers who’d been invited to MASS MoCA for that summer’s residency. Some—like Eve Beglarian and Julia Wolfe—were accomplished professionals. Others were promising newcomers, plucked from more than four hundred applications to the American Composers Forum. Wells had spent the previous day introducing them to the vocal techniques that his group had learned in its decade together. Death-metal singing, or supraglottal laryngeal constriction, was the latest addition to the list. The previous summer, a singer named Androo O’Hearn, from the band Shaolin Death Squad, had taught the group’s members to tighten their throats so that their aryepiglottic folds—triangular membranes that encircle the larynx and protect it when swallowing—began to vibrate furiously. Death-metal singers sound as if they’re broiling their vocal cords with a blowtorch, but the technique causes no harm when done right.
Composing for Roomful of Teeth, Wells says, is like writing for a pipe organ with multiple stops: an instrument that approximates the sound of many others. It’s an imperfect simile, he admits: “Mimicry is just a parlor trick. It’s the liquid between these styles that I want them to explore.” He’s often asked if learning a technique like Persian Tahrir—a kind of hiccupping wail—is a form of cultural appropriation. “Yes, these techniques have stylistic fingerprints that are specific to certain musical languages,” Wells says. “But they’re also just ways that the vocal folds vibrate. How constricted is the larynx, and how does air pass through it? Which frequencies are amplified or attenuated? All those things are just basic physiology, and they manifest in different ways around the world.” Flipping back and forth between your chest and head voice—what we call yodelling—comes so naturally, for instance, that countless cultures do it. There are Swiss yodellers, cowboy yodellers, and polyphonic Pygmy yodellers in the forests of the Congo. “No one culture can say, ‘We are the yodelling culture,’ ” he says.
To some composers, this is wonderfully liberating. “I want to write something for them tomorrow,” Julia Wolfe told me during a break. “I mean, I can’t wait.” Wolfe, who is sixty, has a head of brilliant red hair and features lit and charged by a constant current of ideas. She won the Pulitzer Prize in 2015 for “Anthracite Fields,” an oratorio about coal miners in Pennsylvania. On the drive to MASS MoCA from New York, she’d been listening to Aretha Franklin, who’d died a few days earlier. “You can hear her coming out of all the cars as they drive by,” she said. “I’ve been thinking, That blues shout she does—I don’t think I’ve heard that in a choral piece before.” She laughed. “It’s sort of a composer’s dream. I think of all the parts that I’ve had to take out of pieces over the years, thinking, Oh, they won’t be able to do that. Well, these guys can do it.”
To other composers, it’s a little overwhelming. Wells not only wants them to incorporate novel techniques; he wants them to tailor their pieces to the group’s voices. Later that week, Stafylakis and the others would sit down with the singers one by one and take notes on their individual ranges, skills, and timbres. Most choirs ignore such differences or try to smooth them over—to sound like a single voice. But Roomful of Teeth is defined by its idiosyncrasies, its flux of timbres and textures: the deep drone of Beauchamp’s bass with Cluver’s pristine soprano floating above it; Kelsey’s gutsy belting against Dudley’s fluid tenor; Burton’s huge baritone, ember warm, beneath Shaw’s flickering lines; and Gomez above them all, surfing the riff like Robert Plant on “Immigrant Song.” They’re less of a choir than a rock band.
“Could you elaborate on the death metal a little bit?” one of the composers asked. “Could we hear an eight-part death-metal piece?” Wells frowned. They didn’t have one yet, he said. Just a few scattered passages. “Can you do a death-metal chord?” Wells turned to the singers, who obliged with a subterranean growl, like a family of monsters snoring in a cave.
“Can you tune the chord and hold it?”
“None of this is pitched,” Beauchamp said.
“Is it possible to pitch it?”
“I don’t think so. Not yet.” But they gave it a try anyway, now sounding like a swarm of prehistoric hornets.
“That’s just bad Kargyraa,” Dudley said. Kargyraa is the most famous of the throat-singing techniques from Tuva, in southern Siberia. Like death metal, it involves activating parts of the larynx that usually lie quiet—in this case, the ventricular folds that flank the vocal cords. By tightening the throat and pushing air through it, Kargyraa forces these folds to vibrate at half the speed of the vocal cords, producing two notes at once: the fundamental pitch and a groaning sub-harmonic below it. (The Tuvans, like the Sardinians, like to imitate their livestock; Wells once saw a film of a camel herder comforting a rejected calf with Kargyraa.) After a quick demonstration, Wells had the group try combining death metal with another Tuvan technique, known as Khöömei. The singers cupped their tongues, circled their lips, and narrowed their throats, creating high, whistling overtones. It was mesmerizing, but they could only sustain it for a few moments.
“What’s the problem?”
“A lot of these techniques create lots of phlegm,” Kelsey said. “We have to have time to swallow that down.” She shrugged. “The amount of time that singers talk about their phlegm is very normalized.” For a technique to survive, it has to be safe—it’s an almost Darwinian rule. But there are exceptions. Students of Korean P’ansori traditionally shout themselves hoarse under waterfalls, till their vocal cords bleed and are covered in nodules and calluses. The more ravaged the voice, the more beautiful it is said to sound.
Roomful of Teeth won’t go that far. Its version of P’ansori is like a band saw with a blade guard: a rasping yowl without the guttural breakup. But every style, no matter how harmless, uses different muscle groups and breathing techniques, different tongue, throat, and mouth positions. Switching between them can be exhausting, and even dangerous if the singer lapses into poor technique. “When we studied Kargyraa, I was just growling over and over to try and access those muscles, but I couldn’t find them,” Kelsey said. “I would thrash myself every time.” Dudley nodded: “Over the arc of a Roomful tour, there is a little bit of fear.”
The question of limits—of how much a singer can take—hovers over all the group’s rehearsals. In 2014, during a residency at Williams College, the composer William Brittelle began working on a piece called “Psychedelics.” The early versions were like a vocal triathlon, alternating between belting, throat singing, and yodelling—impressive to hear but a nightmare to sing. Brittelle needed sixty-four drafts to get it right. In the end, the group’s best moments are often the quiet ones, when the vocal gymnastics settle down and the singers fall back into unison, their tuning so perfect that it glimmers with overtones.
The balance between innovation and polish, individuality and blend, can be hard to find. Learning new techniques is unnerving for professional singers. “All of a sudden, you’re in kindergarten again, and your muscles aren’t strong enough to do things,” Martha Cluver says. And Roomful of Teeth sometimes feels like one too many sharp musical minds in the same room. “We’re this nine-headed amoeba,” Wells says. “We can conference about every fucking thing.” Some of the singers wouldn’t mind a little more traditional, top-down direction. Others would like even tighter collaboration. “I’m used to being in chamber groups where everyone is invested in everyone else’s playing,” Shaw says. “But some people found that offensive. It got to the point where they were saying, Caroline, maybe you should just take a break and check your phone sometimes.”
This fall, I met up with Shaw at Juilliard, on Sixty-fifth Street and Broadway. The school had just hired her to coach students and present compositions. Thin-boned and petite, with sharp eyes behind oversized glasses, she had the high-strung yet unruffled quality of a bird on a wire. More than any other composer, Shaw has cracked the code of Roomful of Teeth’s maximalist style—the world-spanning technique that tries to consume every other, like the universe in Krishna’s mouth. Yet she feels like a bit of an interloper at Juilliard, the land of bel canto. Although she has a bachelor’s and master’s in music from Rice and Yale, and has written pieces for Renée Fleming and Dawn Upshaw, her degrees are in violin, not voice or composition. “I woke up that first morning last week and thought, Do I have to wear makeup and jewelry?” she said. “The opera people in college always had makeup on.”
Shaw grew up in Greenville, North Carolina, both at the outskirts of classical music and at the heart of it. Her mother was a trained singer and Suzuki-violin instructor. (Shaw started playing at age two, on a Fruit Roll-Ups box with a paint stick for a neck.) Her church-choir conductor later became the director of the organ department at Indiana University. When Shaw first moved to New York, in 2008, after graduating from Yale, she went from dance company to dance company, offering to play for their rehearsals. Pianists were the standard accompanists, but Shaw dismissed them as parvenus. The first formal ballet classes, she said, were taught by a dance master under Louis XIV, playing a tiny pochette violin tuned up a fourth. Both N.Y.U. and the Broadway Dance Center hired her. “I would just make stuff up,” she said. “I knew I needed harmony, a bass line, and something melodically interesting. It was just puzzle solving. Puzzle solving in real time.”
When Wells chose Shaw for Roomful of Teeth, ten years ago, he knew that she didn’t have the strongest voice or the best technique. But her pure tone and uncanny musicality more than made up for it. The composing was just a bonus. While the group focussed on new techniques and commissioned pieces, Shaw hung back and absorbed the new sounds, until they felt organic to her music. “We had no idea she was creating this spectacular thing,” Cluver told me.
“Partita for 8 Voices” was written one movement at a time in the course of three residencies at MASS MoCA. The movements are named for Baroque musical forms—“Allemande,” “Sarabande,” “Courante,” and “Passacaglia”—and the piece begins with spoken instructions, as if Shaw were back in the dance studio: “To the side, to the side, to the side and around . . . two-three-four.” When the voices suddenly break into high, belting harmonies, it’s as if the dancers are swept into the air, to spin weightlessly around one another. “The detail of the pattern is movement,” a voice says, and the music lands back on its feet.
Roomful of Teeth concerts sometimes sound like musical playdates—a bunch of composers tossing their new sonic toys around a sandbox. But Shaw’s pieces never feel arbitrary. “Partita for 8 Voices” is a pattern book of vocal styles: its movements are stitched together from plainchant, percussive breathing, Early American hymnody, and half a dozen other techniques. Yet the piece is so carefully constructed, like the quilts that Shaw grew up with in North Carolina, that even its showiest bits fit together.
To Wells, the piece is both a work of art and a capsule history. “The movements are like geological strata,” he told me. “In ‘Passacaglia,’ which was written first, the way the voices stretch past bel canto to belting sounds and the yodel break—those two techniques were the ground floor. Then it’s another year and we’re learning P’ansori, and that’s ‘Sarabande.’ ” It’s as if Shaw, like Wells, is rebuilding choral singing one layer at a time.
How will the singers of the future sound? Will they be specialists or generalists, sprinters or decathletes? Wells is hardly alone in rebelling against bel canto. At MASS MoCA, one composer all but declared a ban on vibrato, and the rest seemed to share her skepticism. “Part of my argument with bel canto is that it runs roughshod over intelligibility,” the composer Eve Beglarian told me. Beglarian is composing a piece for Roomful of Teeth and the Dessoff Choirs set to text from Walt Whitman’s “A Song of the Rolling Earth.” Ordinary speech has a frequency range of between a hundred and three hundred hertz, she said. Above a thousand hertz, the brain has a hard time distinguishing vowels. Yet bel canto focusses the voice at a frequency more than three times that high—between three and four thousand hertz—so that it can vault over the sound of the orchestra to the audience. “It’s a rare breed of singer who’s able, through feats of diction, to make English intelligible in opera,” Beglarian said. “But if you’re going to set a text, presumably you care about the fucking text!”
Bel canto is an anachronism, some say—a throwback to an unamplified age. Why sing so loudly when you can use a microphone? Like Broadway shows and pop concerts, Roomful of Teeth performances are amplified. It’s the only way for styles like death metal and Khöömei to be heard from the stage, and it dramatically expands the group’s expressive range. “You can just do more with a microphone,” Kelsey told me. “We can juxtapose very loud belting with delicate breath sounds and then equalize them with the soundboard. Sopranos, when they were belting, used to top out at a C above middle C. Now they can go up to F-sharps and G’s with the microphone, and sometimes even higher.”
Yet for most trained singers, bel canto is still a singular focus. “Vocally, it’s the Olympics,” a young mezzo-soprano named Ariana Stultz told me at the New Music on the Point festival. “If you’re sick or take a vacation, or time away to explore other music, instead of keeping in constant, super-operatic vocal condition, the falloff is so quick.” As Jascha Heifetz famously said of the violin, “If I don’t practice one day, I know it; two days, the critics know it; three days, the public knows it.” Caroline Shaw, for her part, has no interest in forcing her Juilliard students to give up bel canto—to turn their great brass instruments into woodwinds. “I want a horn to sound like a horn,” she told me. “I can push it to be breathy or push it to a squawk. But it took a few hundred years to get that instrument to that point, and at some point I want that horn to do what it can do. And the same for the voice.”
After Shaw’s Pulitzer in 2013, and Roomful of Teeth’s Grammy that same year, the group’s obscure experimentation phase abruptly ended. Concerts at Lincoln Center and Disney Hall followed for the group, and, for Shaw, collaborations with Kanye West and the National as well as appearances on “Saturday Night Live” and “Mozart in the Jungle.” “Is Caroline Shaw Really the Future of Music?” a profile in the Guardian wondered. Yet Shaw still lives in a three-hundred-and-twenty-five-square-foot one-bedroom, and Roomful of Teeth relies on grants and donations to break even. Wells has managed to make good on one of his goals: to pay his singers a living wage. Not every member of the group can sing at every concert—Wells has a pool of regular substitutes. But the core members average about sixty thousand dollars a year, for roughly a hundred days of rehearsing and performing. Still, that’s only a fraction of what any successful pop or country singer would get. And the audience for opera still dwarfs the one for contemporary choral music.
There will always be bel-canto specialists, just as there will be jazz and gospel singers. But in the Juilliards of the future, Wells hopes, most singers will learn a variety of styles. “If you’re going to work with composers, who are constantly exploring different colors, you should be comfortable yodelling and belting and some sort of throat singing,” he told me. “After that, the options become sort of infinite.” Lately, instead of learning new techniques, Wells has been going in the opposite direction. He’s been trying to strip the voice back to the pitches and textures of ordinary speech—to tap the hidden music of everyday life, as the Shakers once did.
Two years ago, he was listening to NPR one morning when a story by the reporter Jasmine Garsd came on. It was about a heroin addict in North Carolina named Bone who’d overdosed numerous times and was once brought back to life after his heart stopped. Listening to Bone, Wells realized that his story was etched into the very sound of his voice: a slow, dehydrated drawl, thick with Southern heat and vocal fry, the indolence of small towns and the drag-footed pace of a heart slowed by heroin. “I thought, This is it. That’s the whole insanity of the opioid crisis right there, in one voice.”
Wells has since gone to North Carolina with Garsd and interviewed a number of addicts, dealers, and social workers for a piece he’s composing called “Visible Speech.” It’s a series of portraits pieced together from tapes of the conversations, interwoven with vocal lines sung by Roomful of Teeth. Wells hopes to have the singers capture some of Bone’s drawling, descending pitches. “We don’t think of French and English as languages where pitch has meaning,” he told me. “But it’s there.” He knew an Irish composer who lived in Paris and who noticed a pattern at his local boulangerie. Customers who said “Bonjour!” in an ascending line, with two notes a sixth apart, got served first. English has similar patterns, studies have found. We use minor thirds when telling sad stories and major thirds when telling happy ones. We match pitches with those we admire and expect the same of those who admire us. We harmonize when we agree—starting our sentences a perfect fifth or an octave from where the last sentence left off—and grow dissonant when we disagree. Our arguments are full of tritones.
Whether we know it or not, Wells said, we’re always singing.
On my drive back to New York in August, after my last day at MASS MoCA, I stopped at Hancock Shaker Village to see the silo installation. The exhibit had opened in early June and seemed to be a hit with visitors. But there’d been a few complaints from the farmers and grounds crew, Wells told me when we met at the village. The sound of the Shaker hymn looping endlessly was driving them all a little crazy.
The silo was about eight feet across and thirty feet high, with boards of unvarnished yellow pine as tightly fitted as cooperage. Built in the late eighteen-hundreds, the building was nearly blown over by a storm in the winter of 2017. When repairs were estimated at a hundred thousand dollars, the village had a choice: knock it down or make it worth the price. Wells was contacted soon after that. Twelve small speakers were now mounted at cardinal points around the wall—four at ear level, another four halfway up, and a third set close to the ceiling—along with two speakers in the grain chute. As we listened, Sam Amidon’s reedy tenor began to hum, and then another voice joined him, and another.
I asked Wells if he’d ever found a religion to replace the Christian Science of his youth, and he said no. He’d tried attending his wife’s church, as well as others over the years, but the services seemed half-hearted compared with the full-body worship he’d known. “I love our young pastor, but I just can’t stomach the Jesus brand,” he said. When he was sick as a boy, he said, his mother used to sing healing songs instead of giving him medicine. The one he remembered best was about not really having a body—just pure spirit. “It was a warm, embracing hymn,” he said, but it has a complicated effect on him now. “When I was feeling nauseous, she would sing this hymn to me. So now when I hear it I feel a sort of malaise.”
Musical legacies are hard to predict. Will choirs still be singing Roomful of Teeth arrangements in fifty years? Or will the group’s focus on individual voices make the pieces sui generis—doomed to have no interpreters? Wells tries not to think about it. He needs to find new music to perform now, he says. The better it sounds, the more likely it is to survive. Art and the human voice can assume almost any shape, if we want it badly enough. When Mozart composed “The Magic Flute,” he wrote the Queen of the Night aria for his sister-in-law Josepha Hofer, who had a prodigiously high and mobile soprano. Now lots of coloraturas do it. Benjamin Britten wrote a number of cabaret songs for his friend Hedli Anderson, whose voice could go both very low and very high. Now other sopranos sing those songs, too. “What Roomful is trying to do is expand the idea of the complete singer in the twenty-first century,” Eve Beglarian told me. “If there is rep that people want to do, they will find a way to do the rep.”
The voices were growing louder, circling the silo one by one with the choir close behind. Wells lifted his head to the cloud of voices rising and swirling toward the ceiling, then stretched out his arms as they joined in a great, ragged chord. When they fell silent, I could hear the tapping of rain on the roof outside. Then a last voice sang out—Shaw’s quiet mezzo, wafting up like a fleck of ash above a flame. Wells smiled and shook his head. “It’s more like church than I expected,” he said.♦