John Adams on his Violin Concerto
(interview conducted and transcribed by Rebecca Jemian and Anne Marie de Zeeuw, Louisville, 1995) Reprinted by courtesy of Perspectives of New Music.
An Interview with John Adams
The following interview was conducted on 24 October 1995 while John Adams was in Louisville to receive the University of Louisville Grawemeyer Award for Music Composition for his Violin Concerto.
Adams composed the Violin Concerto between January 7 and November 1 of 1993 at the request of Jorja Fleezanis, concertmaster of the Minnesota Orchestra. Fleezanis gave the premiere performance with the Minnesota Orchestra in January 1994. The work is dedicated to David Huntley of Boosey & Hawkes; Hendon Music of Boosey & Hawkes publishes the score. A recording with Gidon Kremer and the London Symphony Orchestra, conducted by Kent Nagano, was released on the Nonesuch label in spring 1996. The three movements of the concerto are marked: I. q = 70, II. Chaconne: Body Through Which the Dream Flows, and III. Toccare.
Did the concerto genre and the expectations associated with it—especially that of virtuosic writing for the solo instrument—affect your customary compositional process in any ways that you can single out?
I think it did affect it a lot. I was very slow to come to the concerto form. In fact, even after finishing this work, I'm not entirely sure I'm comfortable with the form itself. There were several issues that tended to make me shy away from the format; one was the problem of handling the dialogue between the instrument and the orchestra, something that really stood quite apart from—almost alien to—my normal way of experiencing musical discourse. It certainly would have been unthinkable for me to write a concerto ten years ago, for instance, around the time of Nixon in China. But in the intervening years my language has become less monolithic and ultimately more melodic. In retrospect Klinghoffer seems to have been the dividing line, the watershed.
I would say very frankly that there are things about the concerto that even to this day cause me trouble when I hear it. I continue to conduct it quite often, and each time I come back to it I find myself going through oscillating periods of doubt or insecurity over certain aspects of the piece. There are moments when it seems to me very satisfying, very true to what I'd want from a concerto written in my own time. But then there are times when the choice of the conventional form irritates me and makes me wish I'd struck out into less well-charted territory. In a certain sense the work is quite archaic in its form; it's a throwback to very traditional means of discourse and syntax, even down to the placement of the cadenza in the traditional location and the fast-slow-fast scenario. One could look at it and say, in a sense, "Well, Adams didn't take many risks in this piece. He bought off the rack when it comes to the formal discourse of the piece." But then there are moments, especially after a genuinely inspired performance, when the almost Platonic perfection of the traditional form produces an intense sense of pleasure. At those times I can appreciate it for its genuine freshness and novelty.
I think my perception for the violin is not conventional, and that the very idea of a composer who is decidedly NOT a violinist writing a piece of this sort is rather provocative. I've told this story very often so I won't go into it deeply again, but for many years I was rather awed by the violin, which seemed to be a very mysterious instrument. From the point of view of a non-violinist, every gesture or fingering that seemed to make sense turned out to be unnecessarily difficult, and everything that seemed ridiculously complex turned out to be no sweat, no trouble. And this is part of the wonderful alchemical mystery of a string instrument. Over the years I've developed a confidence in writing for the violin, and I think in a sense the concerto is an expression of that final sense of, if not mastery, at least confidence that came about from many years of struggling with it.
It certainly sounds as if you make the violinist work very hard.
If there is something that's exhausting about experiencing this work, it may not even be in the difficulty of the work, but in the fact that the violinist virtually never stops playing. In a weak performance, the piece can assume a claustrophobic, oppressive quality. But in a good performance—a really electrifying, committed, and illuminating performance—there is something that's really very cathartic about the work, that adds up to an enormous effect by the end.
In the short lifespan that this work has had so far—which is less than two years—it has already been played by eight different violinists. I've noticed that already a second generation of performers has come to this piece—younger people, some of whom are fresh out of Juilliard—and they seem to have negotiated its nastier passages with very little difficulty. So it's interesting how from generation to generation even the most unforgiving of technical problems seem to find their way into the general discourse of the instrument. I am just astonished at how quickly this has happened with my piece.
Were you influenced by other violin concertos? I hear aspects of the Berg tonally, and the perpetual motion quality of the last movement reminds me of the Barber.
I actually do not know the Berg concerto very well, although I have heard it several times in concert. I have to admit I've never been attracted to that piece. It has just not been my cup of tea. Believe it or not, I like the Schoenberg concerto much more. So if there is Berg in the work, it must be there from the most unconscious of responses on my part. But that's not unusual with me. In fact, this kind of unconscious absorption of other music is one of the engines that runs my creativity. The Barber concerto … not at all one of my favorites. But I understand exactly what you're saying—it too has a very effective "perpetual motion" last movement and I understand that reference that you make. But my "mobile" has much more in common with the sort of writing that you can find in Shaker Loops: a similar ongoing sense of pulsation. What's new here for me is the use of scales …lots of them in all sorts of different modes. Except for the opening of Nixon I can't offhand think of any other instance of such scalar writing in my music.
What role, if any, does improvisation play in your compositional process?
I think every composer improvises at some level. You may be a composer who composes everything in your head before you write anything down, as Hindemith claimed to be or Bach presumably was. Or you may be a composer who has to write down ideas and push them around, see them scribbled out, like Beethoven or Stravinsky. Or you may be somebody who is deeply committed to the physical act of improvisation in order to generate an idea and for whom the actual act of writing down is a vitrification or crystallization of that physical improvisation. I lie somewhere between the latter two. Before I started working with the computer and the MIDI system, I tended to work at the piano—very much as Stravinsky or Ravel did, and I did a lot of physical improvisation; I had to hear things to know whether they were right. Nowadays I tend to do some work at the piano and some with the MIDI system. Whether you would call this "improvisation" or not is a matter of defining the term more narrowly.
The Violin Concerto, despite my reservations about its affects and its superficial conventionality, in some ways is the most rigorously worked-out piece that I have composed in terms of its internal design, its genetic structure and the way in which the larger structures reiterate. What's the expression we learned in biology, "Phylogeny replicates ontogeny"?—basically the large organism is a picture of the smallest cellular structure. For example, in the first movement, those rising waves of triads become basic genetic material for the entire movement. They make their effect felt everywhere, even in the cadenza. I perform all kinds of operations on these rising waves of triads, transforming them into a multitude of shapes and forms: they change mode, change direction, undergo all kinds of augmentation and diminution, and at one point even become a kind of walking bass line. Changes of proportion, mode, direction, etc. all add up to making the concerto, of all my works, even more than Shaker Loops, an analyst's tidepool.
I've read in Michael Steinberg's program notes that you originally envisioned a two-movement design, which later expanded to three movements. Why? What was there about the musical material that seemed to demand this treatment?
I had this wonderful idea of a second movement: it was going to be a chaconne but like the D minor Chaconne of Bach or maybe the last movement of the Brahms Fourth—one that got wilder and wilder and more ornate. It was a grand idea but somehow I never found the right material to justify the form. Instead I produced the present chaconne, which is a more enclosed piece, a kind of dreamy, filmy, almost diaphanous slow movement. One person who recently heard the piece commented that the British pop composer and producer Brian Eno had created a piece on the Pachelbel Canon, and as soon as he told me this, I realized that about twenty years ago I did indeed hear that piece. So it is possible that the image of a deconstructed chaconne was first suggested to me by the Eno piece.
Which MIDI sequencer and software do you use?
I use a program called Performer. Most of these musical software programs are developed for pop music, and it's frustrating because their creators could produce far more pliant and resourceful programs if they took more input from serious composers. Many of the features are designed so that musical illiterates can move around in the environment. It's basically a sketchpad, and I use it because I always like to have material in my hands. I'm like a potter. I want to have my hands on the material while I'm working with it. (I'm stretching the metaphor of course, because what I'm actually talking about is being able to hear the music immediately while I'm working on it.) The image of a composer sitting alone at a desk with nothing but a piece of paper and a pencil is something that has never interested me. I'm inclined to think that, by and large, only composers with extremely defined perfect pitch are able to write coherently away from a sound source. There is something about perfect pitch that seems also to be connected with a certain capacity for aural structuring. Those of us who don't have it often find ourselves widely off the mark if we are not more in touch with a sound body. But I'm a very physical person, and I've always liked to have the sound right there. It gets my juices flowing in a way that I don't think would ever happen if I were just writing at a desk and only imagining the sounds in my head.
How do the occurrences of groups or patterns of simultaneous differing speeds in the first movement interact with your idea of an underlying constant pulsation?
Can you give me a case in point?
In mm. 28-29 you get hints of a grouping of three sixteenth notes in the winds against four sixteenths or two eighths in the other parts. You certainly get that interaction in mm. 40-44 in the violin part. [Examples 1 and 2]
Example 1. First movement, mm. 28-29, partial score.
Example 2. First movement, mm. 40-43, partial score.
Well, that was not any conscious decision. What's happened in the course of my musical development is that from the start I've wanted to confound what I felt to be an overly simple definition of pulsation. I think I've always rested apart from the other minimalist composers in this desire to have a strong sense of pulse but at the same time to confound the listener's expectations. I've been very conscious lately of Nancarrow because he, as well as many jazz composers, has worked very hard to create this wonderful sense of underlying pulsation above which many, many layers of events are simultaneously operating. I realize that this occurs in Ligeti's music too, but I am less drawn to Ligeti because the underlying sensation has never struck me as physical or naturally felt. I can't connect my body to it, it doesn't dance for me in the way that even the crankiest, most complicated Nancarrow or Ives piece does. There’s a very obvious case of a Nancarrow reference in the Toccare movement where I have a group of five against a very complex division of sixteenth notes in the woodwinds.
Is that around m. 63? [Example 3]
Example 3. Third movement, mm. 63-66, partial score.
Yes, that’s exactly it. It often sounds very messy unless it is very, very well played. I’m not even sure it is successful in a large orchestral format, because not only is it very hard to get that level of precision from an orchestra, but even if they are playing very precisely, in the general acoustical environment of a full orchestra, that precision gets muddied by the physical distance between the instruments. This may be one reason why Nancarrow finally chose to devote most of his life to working on the player piano. But it represents an "ideal" world of sorts. Being an orchestra composer means living a life of endless surprises, often not very pleasant ones! So much of one's education has to do with acoustics. I know this from conducting Ives. Once you experience the realities of all those bodies on a large stage, with many of them not being able to hear the others …then you understand that the world of intricate polyrhythms is often more hypothetical than real.
What are you trying to achieve by mixing constant pulsations with patterns that simultaneously project conflicting speeds?
I’m trying to find ways to enrich the experience of perceiving the way time is divided. I’ve never been interested in music that denies pulsation. You can tell me that a Carter work has pulsation and it’s just a very abstracted pulsation, but I’m sorry …if I can’t hear pulsation, if I can’t feel it, then for me it doesn’t exist. It may exist theoretically, but for me it's not there. I need to experience that fundamental tick. What I’m trying to do now is enrich that experience.
The piece that I’m working on right now (Lollapalooza) is a short orchestral work, very minimalist in the way the material is presented. The events make their appearance in a very formal, sequential manner. Then the little engines go through their repetition and evolve into different patterns. But instead of actual polyrhythms such as we have here (in the Violin Concerto)—sevens against fives—in Lollapalooza I’m keeping everything within a grid of four. Despite the regularity of the meter, the piece is full of little spikes of energy in different places, producing a wonderful sense of bristling, popping energy in this piece, while at the same time it’s all in four. I learned a lot from conducting Yvar Mikhashoff's marvelous orchestrations of Nancarrow Studies [for player piano]. The original Nancarrow compositions are meterless, and Yvar had to impose some kind of meter on them in order to keep the ensemble together. He ultimately chose the simplest of meters—4/4 or 12/8 or similar meters—for the overall tactus. But beneath this is a teeming underworld of more complex rhythms. It's also an eminently practical solution to playing the piece.
Do you mean to project a grid, a metrical hierarchy of pulsations?
In all my pieces?
In your music in general.
I don’t really know how I would answer that. It does seem that there is a level of the division of the pulse, and I’m not sure exactly where that is. In the creation of temporal realationships there seems to be a point beyond which musical time itself begins to sound dissonant. This is a funny way of describing things, but I feel that there’s a natural way of dividing time and then there’s a dissonant way of dividing time. I don’t know where that dissonance begins. Two against three obviously does not feel dissonant to me; it has a wonderful texture to it, as does three against four. Three against five begins to push the envelope. I guess when you get beyond that, that we get to an area of cognition—we get into Fred Lerdahl’s area—and perhaps it’s a question of whether something is learned or not. It may be that one can learn these things and get to a point where they feel absolutely natural. I’m not sure.
Another area of rhythm that you’re experimenting with is tempo. First, there’s a gradual accelerando to q = 84 [mm. 56-70], and then later there’s a sudden shift to q = 112 [m. 78], a metric modulation in other words. What effects are you trying to achieve with tempo?
I think there are two ways to achieve a shifting of gears in music. One is changing tempos through rational divisions of time, which is what you call metric modulation. The other is by speeding up or slowing down. I like to use both of them because they each have very different expressive results. One of the things I like doing—and I don’t know of any other composer who does it to the level that I do—is to write long, long, long accelerandi. For example, in the first movement of Harmonium, the conductor is asked to create a very subtle accelerando over a structure of almost ten minutes. The third movement of Shaker Loops is similar. Those are very hard to accomplish. The performers have to possess an innate feeling for the rightness of the rate of change. Edo de Waart, the Dutch conductor who introduced so many of my first orchestral works in the '80s, was extraordinary in his ability to control the larger shape of these accelerandi. He was also a very good Bruckner conductor. Maybe that's why! These accelerandi can also be a real headache in recording sessions, because you get to the same spot and if you get three different takes …
Did you change the time signature at m. 280 to 3/4 as a notational expedient so that the pulsations in the violin can align with the downbeats in the orchestra? [Example 4]
Example 4. First movement, mm. 278-82, partial score.
Notationally it makes perfect sense that way. The difficulty at that moment is the violinist moving to playing the pitches two-thirds slower. But one conductor suggested that I do this passage so it could be conducted in two which would make it very easy for the violin, and the woodwinds would be playing [sings as though in 6/8].
Does that change your metric intention?
Well, it makes life hard for the flute player. I’m a very practical notator. Being a conductor, I take the route of least resistance in terms of notation unless something simply makes no sense theoretically. This moment in the concerto can be a tough one, but once fiddle players get a rough idea where things go, they usually intuit it during performance.
The pitch content at the beginning of the first movement can be analyzed in terms of solo and accompaniment. The solo instrument seems to have free access to the twelve tones, and the accompaniment consists of a hexachord at three different transpositions. [Example 5]
Example 5. First movement, mm. 1-2, partial score. The annotations on the score show the ordered intervallic structure of the hexachord, expressed in semitones, and the transpositional levels of the hexachord; C is shown as 0 rather than the bass G because Mr. Adams refers to the vertical sonorities at this point as major triads in 6/4 position.
This was derived through scale systems. I had created a group of transpositions that I put into my computer, which allowed me to take any scale or pitch sequence and to multiply it by this mode. I think there are actually two or three modes that I keep flipping back and forth to in this movement. I took them from the Slonimsky Thesaurus of Scales and Melodic Patterns; I can’t remember what they’re called now. I think I originally derived the rising figure as a diatonic figure which I then submitted to modal transpositions. And then of course they’re all 6/4 major triads, which is very interesting because a major triad is a very pleasant and sonically user-friendly interval. But when they’re perfectly parallel and follow lines that are essentially atonal, it creates a very interesting effect where I think the atonality dominates over the tonality of the vertical arrangement.
Perhaps that's especially true when the triads are in 6/4 position—like the Stravinsky Octet which ends on a C major 6/4 chord.
That’s interesting. One of my pop songs in Ceiling/Sky [I Was Looking at the Ceiling and Then I Saw the Sky] ends on a 6/4 chord and it’s quite wonderful. I just did it by chance; I don’t know of any other song that ends on a 6/4 chord with such power, and I didn’t realize until I heard it performed how remarkable that was. I want to do it again!
Did you use the same sort of transformational process that you’ve just described on the ground of the Chaconne?
I did. I did multiply it—when I say multiply, I mean exposed it to my mode-changing multiplier—so you can find it changing to different modes. I also shrunk it and expanded it rhythmically. This movement again is extremely tight compositionally. You’ll find the violin theme, which starts out as a very free counterpoint to the chaconne, coming back all through the rest of the movement in the woodwinds or, at the very end, in the violins. It is in a different mode, and it has been altered rhythmically. In this case, I couldn’t tell you offhand exactly what the proportion is, but you can see if you actually follow these pitches that they appear first in the opening violin solo. I forget what the rhythmic relationship is; it’s probably 3:5 or something like that. I do these neither arbitrarily nor with exceedingly specific intent. It’s somewhere in the middle, a means of creating a sense of what I call genetic integrity in the work and also a kind of almost subliminal recall. Now once I call attention to them, they’re not subliminal, but that’s a technique that I learned, I suppose, from Wagner more than from anybody else. This sense of having been somewhere and having experienced something in the past, coming back to it, is really motivic recall.
Do you find that you’re particularly drawn to ostinato forms such as the chaconne because of the background and associations you have with minimalist composition?
No, I don’t think that there’s any relationship at all there. In fact, this is the only case anywhere in my work where a borrowed archaic form appears. I have no idea why I did it. I think it was that original idea I had about the big Bach Chaconne. Sometimes on a bad night, if I’m conducting this piece, when I get to this movement, I feel as though I’m in an area that’s not genuine John Adams, and it always makes me a little nervous. But then on other nights it strikes me as quite effective. I’m very ambivalent about this movement. I will say that I don’t intend to write any more chaconnes.
You've mentioned the ground of the Pachelbel canon as an "artifact" upon which this movement is based. Did you mean that you adopted the procedure or the actual ground as a model?
I think it’s pretty close to the Pachelbel. Actually, I found this particular version of the chaconne bass in Grove’s dictionary! I found it in an article on ground bass where there were about ten examples of the classic grounds that constitute the traditional ciaconna, and this was one of them. This both sounded and looked like the Ur-type to me, so I used it.
Your theme differs from Pachelbel’s in a couple of respects; one that seems especially interesting is that your ground, unlike Pachelbel’s, is tonally closed. [Example 6.] You bring it back to the D, whereas Pachelbel ends his statement on the dominant each time. That kind of organization has implications for the continuity and the shape of the movement as you compose it. I wonder if you’d care to comment on that.
Example 6. Second movement, mm. 1-6, Synthesizer I.
You mean that you keep experiencing closure with this, over and over again?
Potentially. I don’t think the movement does.
Well, that’s interesting. I hadn’t thought about that. Maybe that’s why Pachelbel's got a hit on his hands. That’s really true. I basically don’t like the variation form for just that reason. I do love the Goldberg Variations, but even in the Goldberg Variations, that stop-start mechanism causes me trouble. And the Diabelli Variations too. I love all the imagination and the wild extremes to which Beethoven takes the original material, but the stringing together of thirty-two variations, the Aneinanderreihen...it’s not my favorite art form.
But what I do like about my chaconne treatment, despite the fact that it keeps closing in on itself, is that once it begins mutating, it produces some deeply disturbing events. It is like some piece of kinetic sculpture or a clock that normally functions in a regular, predictable, reassuring, comforting manner, over and over again to the point where one is almost lulled to sleep. Then it suddenly begins to go awry, starts going into very strange modal areas, starts to experience arhythmia, begins to behave in a dreamlike, irrational manner. It becomes Salvador Dali's clock. What happens above the chaconne line, with all these figures whose rhythm is either augmented or diminished, causes a constant sense of overlapping and for me an interesting dissonance to this kind of clockwork ostinato.
The third movement has a noticeable link to the first movement. The scalar material used at the beginning of the third movement is related to the passage after the cadenza in the first movement [m. 320]. [Examples 7 and 8.]
Example 7. Third movement, mm. 1-2, solo violin.
Example 8. First movement, mm. 321-28, solo violin.
I hadn’t thought of that. That’s certainly not intentional, but I see it.
Both passages use a white-note scale on A with Bb and F#.
Well, the last movement again is very modal, very consciously modal. I had two or three modes that I introduced. The opening mode is sort of a six-note mode. Against that, the violins in the orchestra are playing A minor without the b-flat.
Are there any conscious connections between the first and last movements?
No, not that I can think of at the moment.
It seems to me that the style of the third movement differs from the first two in terms of form. Did that happen because the third movement was not part of the original plan?
It is different but not necessarily because it was not part of the original plan. I think it’s just a different movement. I deny any genetic familial links there.
It has a rondo quality.
Yes, I’ve been writing a lot of rondos lately. In my new piece, Road Movies, which I heard last night for the first time, I realized two of the three movements were rondos. Part of that comes just from the way I’m composing now. I’m actually returning to a lot of the techniques I developed in the '80s. Not only structurally, but harmonically and rhythmically as well. So I guess one could say I’m having a second childhood and going back to my minimalist roots. I think this particular movement was the first inkling that I was really breaking out of the mode that I’ve been in for a quite a few years, since Klinghoffer. Also, you know, this is a whimsical virtuoso piece. As the movement progresses, the virtuosity becomes more and more outrageous. It’s still nothing compared to the real showpieces of the nineteenth century. I recently heard some fantasy by Wieniawski that Midori played. Those pieces are simply outrageous, but only a violinist can compose them because only a violinist knows how far one can go. But this movement does go quite far out there, and violinists, once they’ve got their smelling salts and picked themselves up off the floor and started working on it, find that they really like to do this movement because it challenges them to go to the furthest reach of their technique.
You've spoken about the modes that you use in the Violin Concerto. In China Gates and Phrygian Gates, you became famous for "gating." Are there examples of this technique here? Is this one of the returns you've spoken about?
Possibly. I don’t think they’re quite as abrupt or as dramatic as they are in those earlier pieces, like Harmonium for example. But I think the Concerto is a very transitional and very interesting work in a way. All these compositional techniques that I’ve mentioned to you have contributed to produce a sense of compositional solidity. I feel very confident about the tightness of the work because I know how integral and economical the material is. And in a sense, I think it is a far more mature work than my earlier ones. But I also see it as a work of a crisis of language; I see it pointing, particularly in the last movement, in a direction that I’m now struggling to find—one which probably will refine itself and head back toward a simpler use of tonality and harmony, although the rhythmic and structural levels will probably be more complex.
You've compared slow harmonic rhythm to airplanes needing a long runway. You say that minimalism requires slow harmonic rhythm so the ear can rest and function on a different level. Could you describe what this level is like? Do you think that the idea of ears shifting to a different level is like those popular 3-D images where you look at the surface and then, using a different viewing technique, a picture comes out behind it?
Yes, I think that’s a very useful analogy. Of course, I said that eight years ago at least, and my music has begun to belie that. A lot of music I’ve written since 1990 has gone very far afield from those ideals. But I think I was essentially trying to describe the way I was experiencing music at the time. I was writing pieces like Harmonium, Light Over Water, or Shaker Loops, and I’ve recently become enchanted again with that.
Return to top of page >