2025

Concertino for Vibraphone, Marimba, and Orchestra

Berceuse[BeSecure] for octet

I was barely out of my teens when I met Tom Johnson in NYC and we started exchanging ideas about music and mathematics.  We continued doing that for more than four decades.  He left us on the last day of 2024 and I miss him greatly.  

One of his enthusiasms was "logical music."   Now this might seem a category error - after all, shouldn't it be be arguments and proofs, and not works of art, that are logical or not?  

Actually it turns out that formal logics in themselves just don't care about truth and correctness. Rather they're just oblivious rule-based systems that produce new thing from old things.  Well, sometimes they always produce new true statements from old true statements, and then that's very useful. So those kind of  logics are what people think of when they hear the word. but really the realm of logic is actually much larger.

So I understand Tom's program thusly: instead of having rules bound and bind our art, and also instead of attempting to reject rules altogether (if that's even possible), how about making works of art that, by design, faithfully project faithfully their own sui generis and hopefully intrinsically interesting rules?  With music, that means that both the sounds and the rules, so the logic, become potential objects of contemplation.  

Consider the duality of mind and brain, two distinct things with nevertheless thoroughly entangled existences.  So likewise the sound and structure of music.  By itself that's not new of course, music has had structure for millennia.  But treating musical structure as a first-class aesthetic object for contemplation by the thinking/listening subject, that was new.   Now structure could be more than a mere package or organization for the sonic experience, more than the mere bottle for the wine.  

My Elegy here is based on the fact that (11/8)^37 = 2^17 very nearly (the discrepancy is about .07%).  Both the rhythm and the harmony of this piece project this coincidence.  All the details emerge logically from the effort to project this little arithmetic coincidence.  

While the generating interval 11/8 is approximately 5.5 half-steps (actually more like 5.513), so roughly midway between a perfect 4th and a tritone (11/8 is the freshman sum of 4/3 and 7/5), if you pay close attention you will hear that the *logic* of this piece is actually not the logic of quarter-tone music, even though moment-to-moment it might have a sonic quality similar to quarter-tone music.  The faint, high-pitched wood-block type sound keeps track of this.  As Tom would say, "Just count the notes!"

Cover art from Tom's great book, Imaginary Music: 
https://www.editions75.com/pdf/books/Tom_Johnson_-_Imaginary_Music.pdf

Elegy for Tom Johnson (in 37TET)

méson kaí ákron

Lately I'm making pieces for a synthetic percussion orchestra with a virtually infinite number of instruments.

This is a proof of concept piece.

Perkodhuskurunbarggruauyagokgorlayorgromgremmitghundhurthrumathunaradidillifaititillibumullunukkunun

This is the soundtrack to an abstract video work called Grow.  

The cover image is one frame thereof.

The music and visuals are equal partners though, so "soundtrack" is misleading.  But so is "music video."  The music can stand on its own.

Contact me for a link to the video.

Two Forests

The most basic rhythmic transformation consists in subdividing a single unit of time into two equal halves.  Repeating this operation produces an infinitude of patterns of which this piece might be considered a partial catalog.  The tempo is chosen so that the beginning sounds like some sort of virtuoso drumming (but these are my own computer generated sounds, and no sampling was involved), yet gradually, as the patterns grow more complex, execution would require superhuman ability and the result verges on the biological.  The catalog has its own emergent shape, so there is rhythmic interest too on larger time scales.

Halve and Whole

A companion piece to Halve and Whole, but where that piece takes a deep dive into duple rhythm, here (often nested) triplets balance the story.  If the basic concept seems very similar, well, arguably, the overall effect not so much.

Heart and Mind

Branches

Leaves

Sound component of an abstract film of mine (the cover art is the first frame).  Please inquire if interested.

My Heart Is Beating Fast And My Thoughts Are Racing

There is a slender stream in my output, but extending back five decades now, of music based on scales repeating, not at the octave, but at the fifth (or fourth).  In particular, while the effect here is always "locally pentatonic," as the piece progresses it grows increasing pantonal.  By the end all the pitch classes get equal treatment.  

By a "hypercanon," I mean a canon whose theme is itself a canon, and perhaps, and so on.  

Here the  ultimate theme and patterns of canonic repetitions have been chosen very carefully so that music has an overall constant density.  Get the math wrong and the music will just grow muddy or thin out prematurely.

A pulse unifies the rhythmic surface, but it beats so fast, on the order of 50 clicks per second, that the audible rhythm will not imply it even the pulse arguably still informs the rhythm.  So just as the tonality here doesn't count as either diatonic or traditionally atonal, the rhythmic profile is neither entirely "free" nor based on a perceptible count.

Another quite different interpretation of the underlying structure is forthcoming.  Watch this space.

Sometimes the New Place to Go Sits In Between the Places You Know, for piano

A fact of nature: one and the same thing may admit both the simplest of specifications, yet also exhibit rich behavior verging on the quasi-random.  

If a 3-note chord has pitches sitting x, y and z semitones below C7, that chord occurs here at time x^3 + y^3 + z^3.

That's the whole score of the work (modulo a simple orchestration strategy).

So all the rhythmic complexity, and all the counterpoint and voice leading simply emerge from the implications of that short sentence. 

I thus invite you to listen for the structure, the generating intention.
But I invite you, equally, to listen for the phenomena, the temporal extension.  

To my ears what unfolds has something subjective, even expressive about it, even though I know it's all implicit already in that simple formula.

The Treachery of Simple Arithmetic, for 10 instruments

When living things gain the ability to extract information from their environment and react to it, they can have a survival advantage.   So did our senses evolved;e.g. an animal might "hear that" a predator approaches and evade it.  

Complex beings process complex stimuli for all manner of information, so the cognitive effort after the raw sensation can take time.   This necessary delay, for human beings at least, means experiencing sense-data-in-itself even while reflecting on the question "What is it?"  For music this gives rise to a duality: even as we surrender to an immersive experience we might parse it and deduce some of its structure (or at least experience the mere presence of structure through its artifacts).  

Perhaps "duality" make this too simple: we can also reflect on the feeling of surrender, we can also reflect on the feeling of deductive experience.  We can "hear that" a sound got made in a certain way, we can hear that a sound is similar to different from some other sound, and we can "hear that" a variety of sounds enjoy some order or pattern.   Music is not just how it sounds, it is at least(!) all that the stratified listening mind does.

To Hear and Hear-That, as a kind hypercanon (a canon on a canon on canon on a canon) depends on "transposition up the steps of the overtone series (as opposed to up or down a fixed interval).  

From another but equivalent viewpoint, the many chords here correspond to solutions of the problem of expressing any given beat number, if possible, as the sum of four distinct fourth powers (a.k.a. biquadrates).   From the choice of that problem, and for mathematical reasons, a kind of rhythmic steady state emerges, a flow that neither thins nor thickens on the average, but nevertheless projects predictable/unpredictable variation (depending on the computational power of the agent who would predict).   In contrast to the rhythmic steady state, as the piece progresses the larger beat numbers require solutions with larger biquadrates, so the timbre of the piece slowly evolves from consonance to some kind of resonant dissonance.

For an ending, both a bit arbitrary but also a bit natural, the piece goes on only as long as 32 harmonics suffice.  



To Hear and Hear-That

Another hypercanon piece.  

This time, from the mathematical perspective, beats get filled when their beat number equals the sum of a square, a positive cube and 6th power.  

Because the powers vary, the music keeps the counts on different instruments. 

The numbers to square get heard on the piano (with 1 the top key and then descending chromatically).

The numbers to cube get heard on pizzicato strings tuned in 19-TET, using an ascending scale that divides three octaves into 19 equal intervals which sound as nearly 10:9 rations.

The numbers to raise to the 6th power, never more than 4, map to a small gamut of tuned drums.

The complexities of the sets of sums of just two kinds of powers asserts itself as the irregular staggered entrances of the canon on the instrument associated with other power.  

If this is too much information for you, the bottom line is an intense listen which, white it has a hypersimple underlying concept, also feature a rather complex surface regardless, and because of the density of the beat numbers with a "yes" answer, intense but also unfamiliar syncopation. 

Big thank you to Peter Thoegersen, who literally wrote the book on polymicrotonal (and polytempic) music.  He's a genius and you should check out his music.  While his music and also his working methods are quite different than mine, he has opened me up to one of his great discoveries, namely that the listening mind can instinctively decouple the layers of a complex music according to the tuning systems girding each layer (at a given moment).  The effect is like passing to an experience in a higher dimension, akin to the discovery of perspective in visual art.   I'm not sure Peter would even count this as "poly" since two of the layers are 12TET and my treatment of 19TET avoids actual microintervals.  But I'd never have come to this way of projecting my ideas without his influence.

Higher Powers, for Piano, Percussion and Pizzicato Strings

This is a kind of polytempic work in the sense of Peter Thoegersen (well, if he thinks so).  The whol time span gets divided into 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610 equal parts with each tempo supporting a layer consisting of only 7th chords that have one particular root.  Since I make chords in the slower tempi sound louder, the interaction of dynamics and periodicity tends to balance the importance of all the different roots.  With the next slower layer always carrying the given layers dominant, a certain amount of root progression emerges.  But it's complicated!

Those Fibonacci Fractal Four-Chord False Relation Blues, for piano four hands

Said the poet: "Euclid alone has looked on Beauty bare."

Push back on "alone" if you like, but beauty may dwell in the immaterial, and from thus the possibility of art formed purely out of ideas. Or not purely: a sensual experience may, by its structure, simultaneously carry ideas. Two art objects, one immersive, the other cognitive, can simultaneously inhabit the same experience.

Varèse framed music as "structured sound," and I would add, alternatively, "sonified structure." Mathematics is the science of structure, so, if you like, the part of music theory robust (to some extent) against the contingencies of sonic materials. I have a long interest both in bringing to my composition ready-made mathematical objects and mathematical phenomena, and in letting the same emerge from sonic considerations themselves. But truly to infuse music with at least the ghost of "Beauty bare" requires evoking mathematical argument.

So consider communicating something by drumming. Fix a small unit of time. Now say you decide to work with only two durations between strikes, namely 2 units and 3 units. Even with the best coding, that choice limits your bandwidth—limits the rate at which you can transmit information.  Now one can quantify this in a sophisticated way with information theory, but we don't need to get into that now. But here's an interesting and not quite trivial fact: if your durations are 1 unit and 5 units, then you have exactly the same bandwidth as with 2 and 3 units! 

A long and short duration might remind you of Morse code. Well, if you "glue" the "dash", the long duration, of the 2–3 system to the following duration, you will get a drumming pattern in a 2–5–6 system. And that translation is reversible. But if you glue the "dot" of the 1–5 system to the following duration, then you also get a pattern in the 2–5–6 system, and also in a reversible way. So you can get from 2–3 to 1–5 via the trick of interpolating between them with another system you might not originally have thought to consider. There's an episode in the piece that makes precisely this transition/translation audible.

But it doesn't stop there! All the following systems also have the same bandwidth as the three mentioned above:
                                                                1–6–11–16–20, 
                                                                1–6–10, 
                                                                1–7–10–12–16, 
                                                                2–6–7–10–11, 
                                                                3–4–5, 
                                                                3–4–8–9–10, 
                                                                3–5–7–8–9, 
                                                                4–5–6–7–8.
Now hopefully the required "bridge" between, say, 1–6–11–16–20 and 3–4–8–9–10 doesn't seem obvious at all! Nevertheless one will experience it if one attends very closely to the sonic details of the piece.

Yes, you can forget all this and you still have an intense experience reminiscent of ritual drumming, albeit without any periodic underpinning. By the way, he piano part works just as the drums do, but also correlates pitches to the durations.

The title: if you wish to transmit a long random "message" in the 2-3-system, a certain ratio between the frequency of 2's and 3's will emerge.  The famous golden ratio would emerge from the 1-2-system and goes by the name "phi."  Hence "high phi" as one old and mostly forgotten proposal for a name of our ratio here.

High Phi, for drums and piano obbligato

The music can speak for itself, sonically, if you prefer.

But if you care to understand what's going on, here it is in brief.  The durations in the cello are 2, 4, and 8 8th notes; the durations in the bass line of the piano part are 3,5,6, and 7 8th notes.  Any rhythm of any duration of the former sort admits a translation into a rhythm of the latter sort, and vice-versa, indeed a translation that preserves the overall duration (to within a fixed few beats anyway).   I expect that this fact hardly seems obvious!  But the violin part is the intermediary, along with the other voices in the piano part.   Attend to the relationship between cello and violin; and between piano and violin, and, with some concentration, the logic indeed reveals itself.

Elegy for Piano Trio (High Phi II)

In both movements of this quartet, each instrument gets assigned a repeating rhythmic pattern—sequences of notes and rests of equal length (say, all eighth notes). 

The first violin plays a pattern of 11 beats (6 notes, 5 rests), the second violin a 7-beat pattern (4 notes, 3 rests), the viola 6 (3+3), and the cello 5 (3+2). These patterns cycle independently.

One full cycle  takes 2310 beats—the least common multiple of 11, 7, 6, and 5.

But—crucially—beat durations vary.

Indeed, the duration of a beat depends on its density—meaning how many instruments are sounding at that moment (0 through 4). Each density appears a fixed number of times within the full cycle, independent of pattern contents assuming the constraints above. 

The twist: beat durations get adjusted so that each density—whether a silence, solo, duo, trio, or a full quartet—occupies exactly one-fifth of the total time of the movement. Rare textures are stretched, common ones compressed.

Thus the interaction of the parts effects an emergent large-scale rhythm.

Out of millions of possible combinations of patterns, one movement employs the set that yields the most unpredictable overall rhythm (in a precisely defined mathematical sense). The other, the most predictable.

As for pitch: each instrument traces a fragment of the cycle of fifths—in near standard tuning. The quartet plays 17TET, so with the octave divided into 17 equal steps, which makes a "perfect" fifth span 10 steps. Each instrument repeatedly travels through this cycle fragment, alternating between octaves. Individually, the lines sound diatonic; together, they form unfamiliar and highly dissonant verticalities—often stranger than even the most dissonant music in 12-tone systems.

Despite this microtonality, no specialized technique is required. If each player tunes their instrument to a slightly different “A,” then standard fingerings suffice—as long as they resist the instinct to tune to one another.

Waves, Winds, Mind - Mvt. II

Waves, Winds, Mind - Mvt. I

Composer, mathematician, visual artist. 
Seeks collaboration with performers and choreographers.

Please don't spam this account!

2025

David Victor Feldman