In 1939, the experimental composer John Cage published an essay called “Goal: New Music, New Dance,” in which he wrote that dance should be freed from dependence on musical structures. Music and dance should be performed in unison, but allowed to wander independently during the time and space of performance. Each should contain separate tones and rhythms. At the time, Cage was 27 years old and working as a piano accompanist for dance classes at the Cornish College of the Arts in Seattle. There he met a 19-year-old dance and drama student named Merce Cunningham, who became his lifelong partner and artistic collaborator.

Cunningham grew up studying vaudeville and folk dance at a small-town dance studio in Washington State, but people often mistook him for a professionally trained ballet dancer. When Martha Graham, the greatest choreographer of her day, saw him perform at Mills College in 1939, she immediately offered to put him in a piece. He had a light carriage, strong legs, expansive arms, and quick-moving feet. There was also something animal about him. The Los Angeles choreographer Bella Lewitzky noticed when she met him that he had “the longest neck I’d ever seen.”

The concert to which Cunningham “dates his beginning” was a joint solo concert with Cage in 1944, consisting of six dances. Cunningham created one of them, Root of an Unfocus, by dividing the dance into three sections, and then dividing each section into numbers of beats corresponding to a certain square root. Cage’s accompanying score followed the principles laid out in his 1939 essay, relating only loosely to the dance. The dance and score began and ended at the same time, but shared no intuitive rhythmic correlation. Another dance on the program, Tossed as It Is Untroubled, consisted of rapid trembling movements and “a great deal of going up and down on your heels,” Cunningham later recalled.

Like Martha Graham’s, Cunningham’s dances emphasized expansive physicality and unusual shapes. His work also had the clean lines and technical precision of ballet. But his style was all his own: leaving behind the emotive storytelling that characterized both Graham’s dances and traditional ballet, Cunningham treated gesture itself as a vehicle for meaning. His dances had no climaxes or resolutions. Much of the movement originated from chance. He rolled dice to decide which way his dancers’ heads, arms, legs, and feet would combine in a particular position, or whether a duet would follow a group phrase or a solo. He instructed his dancers to face sideways, backward, and diagonally, because any direction in the theater could be considered “front.” There was no central unfolding action on which to train the eye, merely a field of activity. Dancers deftly maneuvered around one another like ants in a hive or crowds on a city street. It was, as Cage put it, “nature in her manner of operation.”

A cult audience of musicians, visual artists, and dancers supported Cunningham’s early work. Critics were harder won. In 1964, Cunningham and Cage scraped together enough funding to take Cunningham’s small company on a tour of Europe. High praise from European critics made its way back to America, and Cunningham found New York audiences changed upon his return. For the rest of his life he remained in the spotlight, touring and choreographing relentlessly, debuting fresh work nearly every year.

The intensity of Cunningham’s style facilitates a slow-burn conversion experience: it takes time to appreciate, but once it hooks you, his work intoxicates. His dancers move against the background of Cage’s musical din with athletic concentration, dashing in swift, tiny steps, sailing in massive leaps across the stage, and executing one serene balance after another. They do not always look graceful, but the commitment to exactitude is riveting. Arms and legs cut geometric patterns in the air, torsos wildly arch and bend. When the dissonant movement aligns for a moment amid Cage’s roars of static, it is like something tender happening at a construction site.

Cunningham took painstaking notes on paper before beginning a rehearsal, but left his notes behind when rehearsals began. In his studio, he simply used words to map movements onto his dancers, and those words were notoriously devoid of qualitative detail. He issued simple instructions: “Leg back!”—“Arm up!”—“Be bigger!” His dancers strayed as little as possible from literal executions of his commands, but each inevitably brought his or her own interpretation to the mechanics. There were countless small decisions and adjustments—some conscious, some not—of timing, spacing, and shape, and these elaborations circled invisibly through many hours of rehearsal and months or years of performance. If another company wanted to reconstruct the material, a Cunningham dancer would have to visit them personally to impart the dance. There was no score.

Months after Cunningham’s 90th birthday, the Cunningham Dance Foundation made an important announcement to the company and the public. The Foundation had devised a “Living Legacy Plan,” with Cunningham’s blessing, that would determine the future of his work. Upon his death or inability to lead the company, his dancers would perform his material for two more years. Then, the company would dissolve. In its place, the Foundation planned to assemble an online archive of “Dance Capsules,” which would “document his legacy for future generations and allow ongoing study and enjoyment of his work.” These capsules would in theory contain everything needed to re-create his dances “with their original integrity intact.” Instead of continuing to invest in the company, the Foundation would raise more than $8 million to preserve programs, performance footage, interviews, lighting plots, and set and costume designs—any piece of knowledge about the dances that could be converted into a digital file and posted online.

No one had done anything like this before. Most choreographers are as loath to consider the dissolution of their companies—the living repository of their work—as they are to plan ahead for something as literal as a digital archive of their pieces. A dance company operates almost as an extension of the choreographer’s body itself, and one dance critic compared Cunningham’s Legacy Plan to an act of bodily harm: “[pulling] the plug while the heart was still beating.”

“It deeply saddens us to think about a future without Merce,” one dancer from his company said, reading a collectively written statement to the press. As knowledgeable as they were—Cunningham built his dances for and on his dancers’ bodies; they were his dances, to a certain degree—they knew that while they were more than mere vehicles, there was a limit to their authority.

On the day after Cunningham’s death, the 6 PM dance class took place on time in the Cunningham Studio, an airy, sunlit space on the eleventh floor of the Westbeth building on Bethune Street. Cunningham’s desk sat in the back of the studio, covered in houseplants. A four-person trust had been appointed to handle the rights to his choreography and act as custodians to the Dance Capsules. They were also tasked with figuring out how else to perpetuate Cunningham’s legacy in dancer’s bodies through the teaching of his technique. But the legacy plan gave no indication of what was to become of the practice of teaching Cunningham’s style to students, or what was to become of the studio and the school it housed.

Every Cunningham technique class begins the same way, with a teacher’s light but authoritative finger snaps. Students respond with sixteen “bounces” in a simple forward bend. The teacher may demonstrate the starting position for a newcomer, but after a while he can simply say, “The bounces!” and everyone tilts forward and begins. The beauty of this or any dance technique is the way secondary explanation becomes superfluous once the movement is ingrained. To overhear a dance class is to catch the phrases that frame action and deteriorate in its presence: “And now you go like this . . .” In the silence, new dancers learn what this is. “And then like this . . .” The positions learned in class were building blocks of Cunningham’s technique, the movements from which all his dances were constructed. The tilt forward, backward, left, and right, the swivel (called a twist), and the series of deep curves, combined in endless permutations with positions of the torso, legs, arms, and head—that was it. There was a time when taking class was a way to get noticed and asked to join the company as an understudy. After Cunningham’s death, while a dwindling roster of classes continued at the studio, students just came to learn.

I began working for the Cunningham Foundation one year after Cunningham died, in 2010, and took class as often as I could. Every day at six I waited with other students at the threshold of the studio, where the company spent their days reconstructing works for their final tour. Seven historic works had been selected for reconstruction. After a day’s work, Robert Swinston, Cunningham’s assistant, named director of choreography after his death, parted the crowd of waiting students by wheeling a bulky TV monitor from the studio floor back to the closet.

The reconstruction process looked, in a way, like a Cunningham dance. Many things went on at once: clusters of younger dancers stood in different parts of the room, practicing movement in front of older ones, or gathered around a video of a past performance to scrutinize their parts. Former dancers consulted videos of themselves performing the original work decades ago, squinting at themselves through the static. Stepping back, they reached through some invisible barrier, eyes closed, to physically reformulate a gesture. Sometimes they pored through Cunningham’s notes, revealed to them for the first time; they contained the kind of descriptive adjectives Cunningham never used in rehearsal. Camera crews filmed the reconstructed works. This reconstruction footage, in turn, would be folded into the eventual Dance Capsule for the piece being relearned. On the sidelines of a filming session, a cameraperson once silently admonished me to back up after I accidentally stepped into the frame.

Downstairs, the Cunningham archivist David Vaughan tended to every scrap of ephemera relating to the company. For Vaughan, “preservation” meant not only recording the dances but retaining a historical picture of the company. He began collecting material for his own interest in 1959. The archive he developed chronologically would now need to be sorted by dance, and then separated on the basis of whether the dance had enough material to digitize for potential reconstruction. To assemble the Capsules, the archivists tried to imagine all the information someone might want to ask a person.

Over the years, there was plenty for Vaughan to save. The archive contained 948 film reels, 548 videos, 330 hours of rehearsal and performance footage, all of Cunningham’s papers, and footage from nineteen interviews with Cunningham recorded just before his death. And as he sorted his archives, Vaughan continued to develop them. My grandmother, who was one of Cunningham’s earliest dancers, has several minutes of silent, black-and-white rehearsal footage from the ’50s. Black rings of sweat bloom on the gray of her shirt as her husband films her from the sidelines of the decrepit rehearsal room. The Cunningham Foundation leapt to get copies—anything from those early years is precious. Glimpses of the company’s beginning give color and counterpoint to the hundreds of hours of tape of Cunningham filmed in the last two years of his life. In these later videos, he speaks throatily from under turtle eyelids and fizzles of white hair, his arthritic hands presiding gracefully over the slope of his cane.

Toward the end of the company’s final tour, as the offices were closing down, there were still two enormous storage units left. We spent a day wading through them, dragging hefty bins back to the office to comb through their contents more carefully. Most of the material was considered too trivial for the archive: old receipts for classes in the ’80s, yellow envelopes with pungent glue no one had ever licked. This was the stuff of bureaucracy—the stuff we threw away. For several days it sat by my desk in the downstairs office, awaiting the dump.

Access to the Capsules is available through the Merce Cunningham Trust, which owns the rights to Merce Cunningham’s name, his technique (although it was never explicitly codified), and all the configurations of steps that constituted his dances. Behind a paywall on the trust’s website, each Capsule features, from a simple drop-down menu, all the QuickTime videos, hi-res photographs, and other archival materials the foundation has on file relating to a specific dance. A good video recording is instructive, but licensing a work for reproduction elsewhere, either in a professional or educational setting, entails more than making the Capsules available for a fee and granting permission to label a performance a “Merce Cunningham work.” Former dancers must act as delegates, accomplishing what video preservation and photographs cannot. They translate, demonstrate, and supervise the archive’s dissemination by physically accompanying the licensing of archival materials to new institutions as reconstructors. A “reconstructor’s fee” is routinely factored into the cost of licensing, as it was when Cunningham was alive. With the help of the Capsules, the process of reconstruction is more organized; a reconstructor has the archive at his or her fingertips. But it is in large part the reconstructors, and not the Capsules, that are keeping the dances “alive.”

In order to be eligible for copyright, law requires that a piece of choreography be “fixed in any tangible medium of expression, now known or later developed, from which [steps] can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.” In other words, a dance is only legally considered eligible for copyright if it is presented as something other than itself. It must be “fixed” in record, following the logic that without a physical copy a choreographic work is no different from an idea no one has written down yet. The Cunningham Trust needs to play along in order to protect Cunningham’s intellectual and artistic property, but it’s understood in the dance community that the information practitioners hold in their bodies can never be stored in an archive. What makes the online archive essential—its odds of surviving the degradable human memory, not to mention the human body—is the reason it will someday facilitate only the most speculative reconstructions. The reconstructors are members of a dying breed.

The struggle to faithfully restage a dance stems partly from the fact that no one—dancers, choreographers, historians, critics, or audience members—can express movements they’ve seen or done in a universal system of notation. Without a stable version of a dance, a “faithful” reconstruction strives to re-create more than it does to interpret. Even for Cunningham dancers relearning the work themselves during the Legacy Tour, there was a palpable absence of the master’s galvanizing ability to make resounding creative choices. “That gravitational center has dispersed,” company member Silas Riener told the Brooklyn Rail, describing what it felt like to reconstruct an old dance now that Cunningham was gone. “There’s no gavel.”

A lack of written notation limits the opportunity for creative deviations. An agreed-upon system of notation—one that secured a legible, easily interpretable version of the work—might make a seminal dance more like Shakespeare: delightful in its many interpretations that steer us far from the original production. In this context the proposition of notation seems logical: if a dancer is executing a movement she or he knows how to do, and has planned to do, and can do again—if she is executing a choreographic work—why can’t she just write it down like a play or a piece of music?

Dance notation does exist, and is passionately pursued by an esoteric niche of dance scholars and former dancers; but the practice is widely dismissed by the dance community as an illegitimate preservation method. To most choreographers, writing down a dance seems like an academic misappropriation. Cunningham, who broke the body down into discrete units, might seem like a prime candidate for a written notation system, but it didn’t interest him. “More than the museum, I like the actuality,” he once said. Dance notators counter that physical compositions are no different than any other sort of artistic language—or any language, period—with its written form that stands in a rich relation to a live expression. The conflicting positions are, essentially, a debate about writing itself. Dancers say they will never consider a word capable of corresponding to the motion it describes.

While proponents of dance notation tout its practical use for preservation, there’s also an unspoken feeling among them that widespread use of notation could provide a stronger foundation for the scholarly study of dance—a historically unacademic art form. Modern and postmodern dance, compared with their counterparts in visual art and music, are largely absent from mention in the theoretical canon. For their part, members of the modern dance community have never been particularly attracted to scholarly study of the form. This may be a result of the fact that creating and maintaining a professionally trained dancer’s body is a Sisyphean, life-consuming task, one that takes precedence over other aspects of the discipline. But to notators, this lack of emphasis on scholarly study—and by extension, notation—is just stubbornness. They think dancers should commit to notation the way music students commit to sight-reading early in their careers. Until they do, outsiders will continue to confuse dance with mere displays of sexiness and technical acuity. It is as though notators believe that dance will remain in the dim, unenlightened shadow of other art forms until it learns to write.

Several Cunningham dances have been notated, but these, like most of the holdings in the Dance Notation Bureau’s library, are the result of projects initiated and funded by the notators themselves. The Dance Notation Bureau in New York City is the oldest notation bureau in America, almost as old as modern dance itself. The Bureau’s notators use a form called Labanotation, developed by Hungarian choreographer Rudolf von Laban in the 1920s as a universal written system for coding and representing the way a body moves in space. The system is detailed and laborious, but it supposedly can be used to code any form of locomotion. The language consists of blocks corresponding to different body parts, shaded according to level (high, medium, or low). The length of time a movement is to be performed is indicated by the size of the block. Notators claim this method is rich enough to enable sight-reading; based on a written score, they say they can stand up and execute the steps to a dance.

Contemporary notators tend to personalize their transcription practice. One dance notator insisted to me that her notated scores “were the dances” to her, since they included the choreographer’s verbatim instructions alongside the universal language. This, she said, made her scores “more like photographs than texts.” Another notator has begun adding color to her scores, calling it the only way to adequately express the nuance of motion. The usefulness of these personalized scores seems to lie in their deviation from a strictly codified symbolic form—in the hybridization of the symbol with the interpretive flourish of the notator herself, all based on the contingencies of a particular piece. But this, in a sense, defeats the purpose of notation, since a personalized notation ceases to be universal. Laban’s form of “movement analysis” purportedly appealed to Nazi Party leaders as a way to organize mass demonstrations—an idea of a uniform movement language that is fundamentally opposed to the subjective ways that choreographers remember their material. (By contrast, the American choreographer Alwin Nikolais took pains to hide his personal notes on choreography as he crossed between countries during World War II; he had scrawled them in symbolic shorthand he created himself, and was nervous he would be suspected of smuggling information in enemy code.)

Cunningham’s own notes on choreography often look like cave paintings. Stick figures march unevenly across the page, followed by diagrams and phrases, some barely legible, scribbled beside leaning columns of numbers. They communicate the inspired tremor of the hand more than they convey information. Unlike dance notation, they were made for composition, not preservation; no mark is definitive, or arbitrates a question about a particular movement better than a dancer’s memory can. Instead the notes transmit the frenzy of creation, and perhaps the fear of forgetting: a jotted arrow indicating a breathless need to move on to the next gesture before it slips away. Like a journal, Cunningham’s notes are often cryptic and sometimes quite revealing. At other times, they seem as mundane as a grocery list, succeeding in reminding only its writer what needs to get done.

The reason dance notation might be unsuited to dance is not that dancers don’t understand how useful it could be, but that they understand all too well its power to define. Choreographers like Cunningham may want preservation to be impossible. They want to remember dances the way we remember people who have died—framed by loss and yet perfect, untainted, “as they were.”

My grandmother danced in Cunningham’s company between 1950 and 1957, after meeting him as a 19-year-old studying abroad in Paris. Someone had told her a famous dancer was staying in a nearby hotel, so she stationed herself on the steps outside until he appeared. She informed him of nearby studio space he could rent for cheap, and offered herself as his first student. He graciously took her on.

She could barely point her toes, but she studied devotedly with the company, and performed Cunningham’s work during his legendary early years. She rode across the country packed into Cunningham’s VW bus; Cage, an avid mycologist, would spot, pick, and cook mushrooms by the side of the road between tour stops. She left the company to have children just before the European tour that launched the company into fame, but she and Cunningham kept in touch throughout his life. She may have been a novice, and she did not stay as long as some, but she was in his first company: when there are pictures of his early years in books and articles, she is in them. There’s one of her looking severe and earnest in Minutiae, 1954, in front of a blood-red Rauschenberg.

My grandmother was a typical early Cunningham convert, one who understood more about his choreographic concepts than how to make her body perform the routine. She remembers once asking Cunningham to tell her how to leap. “The only way to do it is to do it,” he said. She is proud of the one position in Septet that, in subsequent years, it turns out only she could do. Balancing on the toes of one foot while keeping her other leg straight out to the side, parallel to the floor, she had to slowly lower her body so she ended in a squat with one leg at a right angle to her body, hovering inches above the ground. There’s a picture in her living room of her performing this feat. Framed above the couch, it hangs beside a picture of Cunningham, glowing in a white unitard.

When my grandmother performed her final show with the company in 1957, she reported to her diary that another dancer protested, in Cunningham’s presence, that the company could never again perform Septet or Banjo. As one of the bodies upon whom these works had been realized, my grandmother had played a central role in their creation. Those were her dances, Marianne’s dances, the other dancer had said, implying they could not continue to exist without her. “I don’t really think so,” my grandmother wrote, “but that’s sweet.”

Cunningham got his start dancing for Martha Graham, but their styles could not have been more different, and when she died in 1991, Graham prepared a very different future for her company than Cunningham would for his nearly a decade later. In what the dance community now refers to as “the Graham debacle,” she willed the rights to her dances to a close friend, Ron Protas, and bequeathed her creative directorship to him with the belief that the company would continue on without her. Protas was a photographer, charismatic and reportedly tyrannical, whom Graham had befriended in her later, alcoholic years. He quickly found himself at odds with the rest of the company. When the Graham Foundation asked Protas to relinquish creative directorship of the company, he retaliated by denying them the right to perform her work. The board ousted Protas as artistic director; in response, he locked the Graham sets and costumes in a warehouse to which no one else had the key. The lengthy court proceedings to wrest the rights away from Protas left the company bankrupt and in disarray. World-class dancers at the peak of their ability had no legal right to perform work they could not help but know. For a time, the Graham studio was forced to close.

The Graham case was decided, finally, on a technical detail. As Martha Graham was a salaried employee of the company she had created, the court ruled that Graham’s works were technically “works for hire,” and not even hers to give away. The rights to almost all her dances returned to ownership by Graham’s trust. Today the company exists under new artistic directorship, and supplements Graham’s canonical works with pieces by new choreographers.

While the Graham case may have provided a cautionary tale for the Cunningham Foundation, Cunningham himself was more concerned with the fact that the work of prominent choreographers often ends not in litigation, but oblivion. Erick Hawkins, who died in 1994, famous in his lifetime but largely unknown now, left his works to his wife, a composer and frequent collaborator. When she died, there was no infrastructure in place to preserve Hawkins’s works. Today, his dances are virtually unseen.

Like the Graham debacle, Hawkins’s near-total disappearance is an unpleasant reminder of how much dance depends on institution, in practice and in preservation. Money greatly affects a dance’s “life.” A dance cannot just be known, it must also be seen—performed for audiences at theaters, given ongoing resources and broad access. You cannot order someone not to know what they know, but neither does owning rights to a dance mean anything if no one can be paid—or allowed—to perform it.

“You have to love dancing to stick to it,” Cunningham once said. “It gives you nothing back, no manuscripts to store away, no paintings to show on walls and maybe hang in museums, no poems to be printed and sold, nothing but that single fleeting moment when you feel alive.” Cunningham’s own archive argues the contrary, and the institution that tends to his legacy offers more than a semblance of permanence; the Capsules do give something back, as outlines for potential reconstructions. But Cunningham may have meant that these things are just placeholders, mementos. Cage once described Cunningham’s work as “less like an object and more like the weather.” One is no less present than the other; both are tangible. But an object is good at sticking around. The weather, on the other hand, passes on.

The Cunningham Studio closed in the spring of 2012, several months after the company’s final tour ended. The trust had hoped to keep it open, but no one saw a way to sustain the costly space without a touring company. The new tenants, as it turned out, would be the Martha Graham Dance Company.

Despite its overall success, the Graham Company had continued to falter financially after the legal battle with Protas. It had also struggled to solidify a model for continuing without its founder. Taking over the Cunningham Studio space, said the company’s chairwoman, would allow the Graham Company to “[rise] like a phoenix,” restoring order after years of scattered operations.

Meanwhile, Cunningham technique classes are now held at studios throughout the city. The technique no longer lives in its historical home, but perhaps this is not a bad thing. The classes offered every weekday at City Center are well attended. Cunningham’s presence hovers over the lessons given by his former dancers. From memory, they quote his advice on dancing, imitating his gravelly voice. In their stories, he’s a wise and personable presence, like a relative students have never met.

Shortly after the Cunningham community decamped to midtown and the Graham foundation moved in, Hurricane Sandy struck. The building on the news that had its side sheared off like a dollhouse stood several blocks from the studio. After the storm, when Cunningham technique classes resumed at City Center, a former Cunningham dancer reported the news to the dressing room, tugging a leotard over her tummy: the Graham costumes and sets had been submerged in six feet of water, ruining almost all the original costumes and Isamu Noguchi set pieces housed there. The Cunningham costumes, meanwhile, were safely catalogued in Minneapolis at the Walker Art Center. The rest of the archive was in the New York Public Library and some of it, of course, already online.

After class, I stopped to talk to a former Cunningham production staff member while he buckled his bike helmet under his chin. He had been wary of coming to class in a new studio, he said, but then he’d gone to see the basement of the old Westbeth studio to see how it had weathered the storm. He stood dumbfounded at the sight of the crumbled cinder block. The carefully built pallets, the well-carved space that housed several production offices, the massage room, and millions of dollars’ worth of Graham memorabilia had been lifted up, swirled around in thousands of gallons of river water, and deposited in a single uniform layer on the ground.