Russell Hartenberger

[Note: Some of the information in this article was previously published in my book, Performance Practice in the Music of Steve Reich (Cambridge University Press, 2016).]

Mallet Phase
Russell Hartenberger and Garry Kvistad performing Mallet Phase at Shalin Liu Performance Center, Rockport, MA, May 4, 2016. Photo: Bonnie Sheckter

Marimba Phase origins

Steve Reich composed Piano Phase in 1967. According to Reich, “…Arthur Murphy a musician friend, and I, both working in our homes, experimented with the performance of this phase shifting process using piano and tape loops. Early in 1967, we finally had the opportunity to play together on two pianos and found, to our delight, that we could perform this process without mechanical aid of any kind.”1 Reich and Murphy played an early version of Piano Phase in January of 1967 at Fairleigh Dickinson University in Madison, New Jersey. They premiered the final version of Piano Phase a year later at Phillips Exeter Academy in Exeter, NH,2 and performed the piece regularly on concerts until 1972.

In January of 1972, following the premieres of Drumming in New York City in December 1971, Steve Reich and Musicians toured several countries in Europe performing Drumming and other Reich works, including Piano Phase with Reich and Murphy playing the piano parts. Following a concert at the Hayward Gallery in London, the ensemble had a recording session of Drumming at the BBC. Jim Preiss and I spent time during a break by jokingly playing the parts of Piano Phase on two marimbas. We soon realized it not only worked well but had a clarity of articulation that the piece did not have when played on two pianos. Reich heard us experimenting with the piece and liked what he heard. He asked us to work on this marimba version and play it for him, so Jim and I spent our free time on the rest of the European tour practicing Piano Phase. Jim wanted to play the steady part and I was happy to play the moving part since that was my role in the phases in Drumming.

In the fall of 1972, Reich revamped the personnel in Steve Reich and Musicians. He was beginning his composition of Music for Mallet Instruments, Voices, and Organ and realized he needed professional percussionists for this new piece as well as for all the percussion parts in Drumming. Reich understood that the best way to create a successful music ensemble was to bring in friends of the current members, so he asked Jim and me if we had suggestions for percussionists who might want to join the group. Jim brought in students of his, David Van Tieghem and Glen Velez, and I recommended Bob Becker, who was my colleague in Nexus and who had just joined me in the World Music graduate program at Wesleyan University. They were followed by Tim Ferchen, Garry Kvistad, Gary Schall, and Thad Wheeler, all of whom became core members of Steve Reich and Musicians.


Knowing how to phase is a key to understanding how to play Reich’s music. Reich said that when he auditioned a musician for Drumming, he did so by phasing with them to see if they could adapt to this new technique. Reich explained, “The musical material in Drumming was so easy that you could pick it up very quickly. So, for the people I didn’t know, I said, ‘let’s try phasing,’ then I could tell right away by trying to phase with them if it’s going to work or it’s not going to work. Even to this day, I find with people – pianists or others – who do Piano Phase or Marimba Phase, if you can do phasing everything else will work out.”3 This was certainly the case with Bob Becker.

When I told Bob about the possibility of playing Reich’s music, he asked what he should know about it before attending his first rehearsal. I told him about the phasing process and suggested we try phasing together. Bob was living in a house with a silo attached just outside Middletown, CT, and I brought my marimba to his place so we could practice. When I arrived, Bob had a music stand set up in front of his marimba waiting for his part. I told him he didn’t need a music stand and he didn’t need a score to the piece. I explained that it would be simpler if I demonstrated the part he was to play. In his book, Writings on Music, Reich wrote:

“To perform [Piano Phase], one learns the musical material and puts the score aside because it is no longer necessary, it would only be a distraction. What you have to do to play the piece is to listen carefully in order to hear if you’ve moved one beat ahead, or if you’ve moved two by mistake, or if you’ve tried to move ahead but have instead drifted back to where you started. Both players listen closely and try to perform the musical process over and over again until they can do it well. Everything is worked out there is no improvisation whatever, but the psychology of performance, what really happens when you play, is total involvement with the sound: total sensuous-intellectual involvement.”4

Since I was already playing the moving part in my practice sessions with Jim Preiss, I showed Bob the steady part and we were on our way. Of course, Bob quickly understood the concept of phasing and was able to pick it up without much trouble. That morning was the beginning of a relationship in the phasing process between Bob and me that has continued for fifty years.

Premieres of Marimba Phase and the end of phasing

In May of 1973, Reich presented four concerts of two works in progress, Six Pianos and Music for Mallet Instruments, Voices, and Organ at The John Weber Gallery in lower Manhattan. Included on the concerts were Piano Phase performed on two marimbas and Clapping Music. It is interesting to note that only a year and a half after the premieres of Drumming, Piano/Marimba Phase was the only work on the program that utilized the phasing technique. Phasing, a technique that is still associated with Reich’s compositions, was only used by Reich for a few years from the early tape pieces in 1965/66 and Piano Phase in 1967 to its most illustrious realization in Drumming in 1971. Six Pianos and Clapping Music both began as phase pieces but emerged from early rehearsals as two of the works that marked the next stage of Reich’s compositional approach.

Steve Reich and Musicians Program
Program courtesy of Gary Schall.

Rehearsals for Six Pianos were held in the Baldwin piano store across from Carnegie Hall late at night after the store had closed for business. Reich long had a dream to write a piece for all the pianos in a piano store. The first drafts of the piece we rehearsed were not for all the pianos in the store, but they were for more than six. We found that the distances between the pianos made it difficult to hear and hence phasing was a problem. Each week we came to the Baldwin store for rehearsals to find that Reich had reduced the number of pianos and the number of players. After a couple of futile rehearsals, Reich eliminated phasing from his drafts of the piece.

When Reich brought the finished score to Clapping Music to a rehearsal at his loft, it had the familiar dots between each measure indicating we were to phase from one interlocking pattern to the next. Reich and I tried several times to accomplish the phasing, but each time we were unsuccessful. Eventually it became comical to us that two people who had performed dozens of concerts of Drumming for the past two years, could not phase while playing the simple Clapping Music pattern. We practically rolled on the floor laughing at ourselves. Finally, Reich said to me, “Why don’t you try just jumping from one bar to the next instead of phasing.” I did that and we found it was easy to play the piece. Steve Reich never used phasing again in his compositions. The process that is still associated with his compositional style was only used for about a six-year period from 1965-71.

Piano Phase, however, is still a phasing tour de force and the inclusion of the marimba version on the concerts at The John Weber Gallery in 1973 was a great opportunity for us to play the piece in front of an audience. Since I had prepared the piece with Jim Preiss and Bob Becker each playing the steady part while I phased, Reich scheduled two performances with both combinations. The performances in this format were very successful and we began playing the marimba version regularly on concerts, usurping the programming of the piano version. Reich calls the marimba version, Piano Phase played on two marimbas, but percussionists still refer to the piece as Marimba Phase. [Note: In the 1980s, Edmund Niemann and Nurit Tilles joined the Reich ensemble and formed the piano duo, Double Edge. They performed Piano Phase occasionally on Steve Reich and Musicians concerts and recorded the piece on a Nonesuch recording titled Reich: Early Works in May of 1986.] 

Russell Hartenberger and Jim Preiss - Marimba Phase
Russell Hartenberger and Jim Preiss performing Marimba Phase at The John Weber Gallery.
Photo: Babette Mangolte
Russell Hartenberger and Bob Becker - Marimba Phase
Russell Hartenberger and Bob Becker performing Marimba Phase at The John Weber Gallery
Photo: Babette Mangolte

Performing Marimba Phase

In the early days of the ensemble, we often began concerts with Clapping Music; Steve Reich played the steady part and I played the changing part. For the second piece on the program, Bob Becker and I played Marimba Phase. I found that the endurance necessary to play Clapping Music affected my control of the mallets on the marimba when I started playing Marimba Phase, so I learned to take a minute to relax my arms before beginning the performance. Playing the music of Steve Reich requires concentration and endurance; this is particularly the case with Marimba Phase. I find that I must focus intently and try to keep my sense of “one” on the beginning of my pattern. It is very easy to let my mind get caught up in the composite rhythm created by the two marimba patterns. I can’t tell if the notes I hear are being played by me or the other player and my hands begin to lose their sense of muscle memory. If I sense that this is about to happen, I concentrate on the “one” of my pattern but I am careful not to accent it. 

In Clapping Music, I usually play the moving part, and I have learned to relax very quickly in the moments where there are two eighth rests together moving from one pattern to the next. In Marimba Phase, there are no rests in the pattern at all and the arm muscles cannot relax until the change from one section to the next. This is even more difficult for the person playing the steady part; that player has no chance to relax until the transition to the third part.

Sometimes I become mesmerized by the fact that, in the first section, my right hand plays a three-note sequence, 1, 2, 3, 1, 2, 3, while my left hand plays a two-note sequence, 1, 2, 1, 2, 1, 2. [Note: These descriptions of right hand/left hand alternation will differ if the percussionist plays Marimba Phase with four mallets.] It is easy for my mind to latch on to either one of these sequences. When this happens, I find I get confused about which way each of my hands is supposed to be moving. To counter this confusion, I once again focus on the first note of my pattern and force my mind to think of it as twelve consecutive attacks rather than groups of threes and twos. 

In describing how he conceived of the pattern for Drumming, Steve Reich speculates that this idea of hand alternation might have come from Piano Phase:

As for the origin of the Drumming rhythm, I knew the African bell pattern, so it might have been related to that. I remember being on the phone on speaker, drumming my fingers, and it happened. It might have had something to do with the way the hands interlock in Piano Phase; I don’t remember working it out. I remember working out the notes as it goes along for sure, but the rhythm itself was one of those things that just happened. The symmetry of it was very attractive and also the fact that the right hand was so completely divorced rhythmically from the left hand. Therefore, by using the appropriate notes in the right hand, it would emerge, as it does over and over again, as a counter-rhythm.5 

As Reich states, hand alternation is important in creating cross rhythms in the Drumming pattern. It is also vital in creating metrical and perceptual ambiguity in Piano/Marimba Phase. 

Endurance and concentration continue to be paramount in performing the second part of Marimba Phase, but players must also deal with new issues. When the moving part fades out after going through the sequence of phases and arriving back in unison with the steady part, the person playing the steady part is left all alone. After several minutes of interlocking with the moving part, the steady player can easily feel exposed. This comes at a crucial time because that player must then change the pattern. The right hand, which had been playing the 1, 2, 3, 1, 2, 3 sequence must switch to a 1, 2, 3, 4 movement while the left hand keeps playing a 1, 2, 1, 2 sequence against the right hand. The moving part player gets to relax for a few seconds while this change takes place, but then must come in with a completely new pattern played only on the natural notes. Normally, both parts in Marimba Phase should be played at an equal dynamic, however, in the second section, the moving part should enter softly and crescendo until the dynamic is slightly louder than the steady part. Reich said this dynamic difference is so the harmonic change created by the new pattern is clear. Once the moving part has reached this dynamic level, the challenge for the performer is that disorientation can occur by playing only on the natural notes. Again, focusing on the “one” of the pattern helps concentration.

The third section of Marimba Phase is the easiest and for me the most fun to play. The various tensions of the first two sections are gone and I can relax a bit and focus on playing extended phases.

The phases in Piano Phase, whether the piece is played on pianos or marimbas, should also be gauged from the perspective of the shape of the overall performance. As Reich indicated in the score, phases in the first section should be the shortest (4-16x), in the second section they should be a bit longer (6-18x), and in the third section the phases should be the longest of all (16-32x). I have always considered these indications of phasing lengths to be suggestions. I do not count repetitions of the pattern while phasing, instead I try to maintain the relative lengths among the three sections. 

As if Piano Phase played on two marimbas didn’t provide enough stressful moments for me, a new version of the piece entered my life in 2014.

Mallet Phase Instruments
Garry Kvistad with Mallet Phase instruments and Olympos® Wind Chimes. Photo: Carol Judson

Mallet Phase

Mallet Phase is a re-orchestration of Steve Reich’s Piano Phase using instruments designed by Garry Kvistad. Garry has designed and built instruments in just intonation since1979 when he created the Olympos® wind chime tuned to an ancient Greek scale. This wind chime led to the founding of his Woodstock Chimes® company that same year. Garry describes the instruments in Mallet Phase this way:

My latest experiment in designing instruments with just intonation is an instrument I built for a new arrangement of Steve Reich’s Piano Phase, called Mallet Phase.  I have played this piece many times in the Marimba Phase version, but I wanted to hear what it would sound like in just intonation and on different materials. 

For the first section of the piece, I built an amadinda-style xylophone out of five pieces of cherry wood that are between 17” and 26” long, 3.25” wide, and 2” thick.  I tuned the bars to the just intonation equivalent of the five pitches in this section of the piece: E, F#, B, C#, and D, beginning a third above middle C. The second partial was tuned an octave and a fifth above the fundamental as in the western xylophone (quint tuning). I built a stand for the bars and mounted PVC resonators tuned to the pitch of each bar. To play the bars, I used large wooden dowels made from a Chinese soft wood and struck the bars on the ends at a forty-five-degree angle (as one would normally play on an amadinda). As with all percussion instruments, the beaters had to be softer than the bars to keep from chipping them.

For the second section of the piece, I tuned aluminum tubes in just intonation to the required pitches: E, F#, A, B, C#, D, and E, beginning a third above middle C. The tubes are between 24” and 35” long, with .25” thick walls, and an outside dimension of 1.75”.  Again, I built a separate stand for the tubes with built-in PVC resonators tuned to the pitch of each bar, and the tubes were played on their ends with the wooden dowel sticks as well.

My original idea for the third section of Piano Phase was to use slap tubes, pieces of PVC tubes tuned in just intonation, and played with foam paddles on the top of each tube. I tuned these slap tubes to the required pitches for the last section: A, B, D, E, starting a sixth above middle C. 

I recorded Mallet Phase with Russell Hartenberger in a studio in Toronto. We sent a copy of the recording to Steve Reich who liked the piece but felt the third section with the slap tubes was a much weaker sound than the amadinda bar and aluminum tube sections. We agreed with Reich’s comment, so I made additional amadinda bars in order to play the last part of the piece on the wooden bars, giving it a stronger feel. To create a different timbre from the first section of the piece, we played the amadinda bars with large, yarn marimba mallets, not on the ends, but on the top of the bars as one would strike a marimba bar. Since resonators amplify the odd-numbered overtones of quint tuning, the wooden bars were amplified even more, and the psychoacoustics of the instrument were enhanced by the brilliance of the overtone.”6 

In Mallet Phase, both performers play on the same set of instruments. In order to accommodate the various patterns, the amadinda bars and aluminum tubes had to be arranged in a non-sequential order. This means that the players have to figure out which bars or tubes to hit without reference points as there are on a standard keyboard configuration. It also means that the bars and tubes are of uneven length so that the players must move in and out from bar to bar and tube to tube as well as playing laterally.  In spite of the various performance issues, the end result is a beautiful combination of sounds and textures. You can listen to a clip from Mallet Phase below:

Mallet Phase
Garry Kvistad and Russell Hartenberger performing Mallet Phase.
Photo: Bonnie Sheckter
  1. Steve Reich, Writings on Music, (Oxford University Press, 2002), p. 24
  2. David Chapman, “Improvisation, Two Variations on a Watermelon, and a New Timeline for Piano Phase,” in Rethinking Reich, Sumanth Gopinath and Pwyll ap Siôn (eds.) (Oxford University Press, 2019), pp. 219-20.
  3. Interview with Steve Reich (December 18, 2003); Steve Reich, “Thoughts on percussion and rhythm,” in Russell Hartenberger (ed.), The Cambridge Companion to Percussion (Cambridge University Press, 2016), p. 179.
  4. Reich, Writings on Music, p. 24.
  5. Reich, “Thoughts on Percussion and Rhythm, pp. 177-78.
  6. Garry Kvistad, “Instrumental Ingredients,” in The Cambridge Companion to Percussion, pp. 63-65.