One of my most vivid early encounters with Roscoe’s music took place in my car. I tuned the radio to WKCR as I was driving out of New York, and what came through the speakers was so hypnotic and beguiling that I had to pull over to finish listening to it. What is this!? A nine-note passage was repeated, again and again, with wide leaps and shifting rhythm that was articulated with slight variations; it was static yet constantly changing. And that sound! The alto saxophonist clearly had great technical command, but it sounded as if he was deliberately approaching the instrument in a way that forced him to confront it—each high-E that begins the phrase burst through in a different manner (more on this later). I had to wait until the DJ announcement afterwards to find out what I was listening to, but the recording turned out to be the solo version of Roscoe’s “Nonaah” taped live in Willisau in 1976, released on a double album of the same name on Nessa Records. 

The circumstances of the recording are legendary—Anthony Braxton could not make his solo concert in Willisau and the organizers had asked Roscoe, who had just performed there with the Art Ensemble of Chicago, to do the concert instead. The crowd was unruly, and increasingly so when confronted with the unwavering repetition. Even through the recording, the tension is palpable. In the liner notes, Roscoe notes that “the music couldn’t move till they respected me, until they realized that I wasn’t going anywhere, and if someone was going it would have had to be them . . . and it helped to create the environment the piece was to take place in . . . building tensions . . . and when I finally did release it my alto had just given in to me (it said, ‘OK, you can play me now’).” This arc takes place over twenty minutes, and the “release”’ is exhilarating. 

I immediately set out to find a copy of Nonaah (not a trivial task twenty-five years ago, the double record was already long out of print). When I finally did, I studied everything about it closely. The cover photo of Roscoe with his foot propped up on a cafe table towards the viewer is both casual and confrontational. There were score fragments in the inner cover. Much to my delight, it also included a version of “Nonaah” for four alto saxophones. At that moment I happened to be working on a series of pieces for this particular instrumentation, and here was this masterpiece that I was not aware of. Expanded from the nine-note figure in the solo version, the quartet “Nonaah” configured the altos in complex rhythmic relationships—stacked, hocketing, and dovetailing. Sometimes the four voices generated compound patterns that were too dense to parse, but are perceived as an aggregate (e.g. the opening and the ending quick section). At other times they form a single line that is fragmented between different voices. While the DNA from the solo version is easily recognizable, the quartet version is very much its own beast. Over the next forty-plus years, Roscoe would continue to mine the implications germinated by that solo piece, reworking it into a cello quartet; a trio for flute, oboe, and piano; bass saxophone quartet; and an orchestral work among others. 

When Mills College presented a portrait concert of Roscoe in 2012 including chamber and orchestra works, I had the opportunity to pay tribute to this piece that was so influential to me, performing “Nonaah” with my alto quartet. Characteristically, Roscoe made a new revision for the performance. It was both thrilling and nerve wracking to learn the piece, not just because it was difficult, but because it is so iconic to us and we didn’t want to screw it up. We practiced for some time before feeling confident enough to invite Roscoe to a rehearsal. After a run-through, I asked Roscoe if he had any critiques for us. I was somewhat relieved that his only comment was “just don’t get sick before the concert.”¹

It’s perhaps a cliché to say that you can recognize a musician’s sound within a moment, but Roscoe exemplifies this perfectly, particularly on saxophones. It is hard to pinpoint what makes his sound so distinct, but somewhere between his articulation, pitch inflection, phrasing, and tone is a voice that is utterly unique and powerful (especially when he is playing quietly). From his earliest recording, fittingly titled Sound, it seemed to me like he was creating obstacles within the instrument for himself—different “situations” that he had to negotiate depending on the register and dynamics. This stands in stark contrast to classical training, where difficult areas of the instrument are meant to sound easy and fluid. In improvisations you will often find Roscoe constantly shifting his embouchure and the angle and position of the mouthpiece, where the physical interface of the instrument is perpetually unstable. 

When I asked Anthony Braxton about Roscoe’s playing, he told me that Roscoe always used the hardest reeds available, “his reed was like a tree bark!” The recordings certainly sounded like it. The strength, or stiffness, of the saxophone reed determines how much air is required to set it into vibration. There are proponents on both ends of the spectrum, but generally softer reeds allow more flexibility and require less force to activate, with the disadvantage of more instability and being prone to choking up when pushed.² Hard reeds can produce a timbral richness at the cost of more resistance. When I got to work with Roscoe many years later, I was surprised to find that he was playing soft number-two reeds. (“Tell Anthony I gave up on those reeds a long time ago!”) But the fundamental quality of his sound has not changed. His conception of the output determines the sound, regardless of changes in the mechanics of the instrument.

Roscoe’s approach to the saxophone is also obsessively focused on minute details of sound production. This is most clearly illustrated in “S II Examples,” a solo recording from 1978. (Released on Nessa N-14/15). The instrument he recorded the piece on, a Vito curved soprano saxophone, was picked out among many for its unique characteristics.³ Over the course of seventeen-and-a-half minutes, Roscoe examines a small set of resonances from every possible angle, with microtonal inflections produced by closely non-adjacent tone holes, shifts in air pressure and embouchure. Overtone structure, multiphonics, and timbre are articulated methodically, all done at a delicate dynamic level. 

The emphasis on specific playing techniques also extends to his works for close collaborators, such as “L-R-G” (also from 1978), written for his trio with Wadada Leo Smith and George Lewis. The score is tailored to the specific language of the performers, each employing a battery of instruments, sometimes switched several times within a page. Despite these rapid shifts, the sounds proceed at an unhurried pace, often enveloped in silence. In his remarks before a 1998 performance at Emory University, George Lewis noted that “L-R-G” created a meditative space—“instead of making things happen, allow them to occur.” 

When Roscoe wrote Angel City for Willie Winant and myself in 2012, we rehearsed for months as he worked on it. We always recorded the rehearsals, and Roscoe would study them to shape our improvisations in subsequent sessions. The score, then, is a mixture of notation and open elements that are partially composed through rehearsals, built on particular combinations of our languages.⁴ 

A year after I started teaching at Mills College in 2006, Roscoe joined the faculty as Darius Milhaud Chair of Composition. I got to know him and his music more intimately over the next twelve years. For the first five years, we were also neighbors living on campus, and I had the great pleasure of practicing with him, sometimes five times a week over the summer. I was nervous when he first invited me to practice. His regimen was well known (infamous is probably more accurate), and I did not know what I would be confronted with. I urgently worked on sight reading a lot of chromatic music with wide leaps ahead of the session. To my surprise, when I got to his house he pulled out two well-worn copies of Universal Method for Saxophone, probably the most old school set of tonal exercises and etudes for the instrument. We read through it for two hours straight, after which he suggested taking a short break before hitting it again. Over the years we would play through those etudes with every possible combination of saxophones, flute, and clarinet. 

In all of the years I practiced with Roscoe, we played etudes and baroque music for the vast majority of the time. He had a seemingly endless supply of Telemann, J. S. Bach, and C. P. E. Bach scores. What struck me was that, despite his obvious mastery of the instrument, he was always working on something. He approached each piece, even the simplest tonal etudes, with intensity and critical evaluation. Pieces with rough edges were slowed down until we really got it right. Tuners were frequently consulted. Registers that did not speak the way he wanted were scrutinized.⁵ He never pulled out any of the spectacular things he could do during practice—it was work, brutally honest in addressing shortcomings. After a long day of practice, when I couldn’t really feel my lips anymore, he would often say, beaming: “There’s so much to learn in music, James. There’s so much to learn in music.” He seemed happiest in those moments.