Sensing Angels

by Charles Gran
in three movements
for solo clarinet and electronics

Performed by Jesse Krebs
Recorded Sunday, November 8, 2009

Part 1 (excerpt)

Part 2 (excerpt)

Part 3 (excerpt)

Sensing Angeles is a piece for clarinet and computer. The clarinet is heard acoustically, unamplified. It's sound is also brought into a computer through a microphone which provides some basic, and not so basic, effects such as delay, reverberation, and real-time transposition which creates harmony or a second voice. There are no prerecorded materials used.

When I was nine, my father taught me to use a soldering iron. With it, my brother and I made a computer from a little kit Dad bought from a mail-order catalog. This was at the height of the Cold War industry in California and Los Angeles was thick with engineers and scientists. I guess I was raised to be an engineer too, but ended up being a composer. A piece like this is probably inevitable.

While in grad school I read an ethnography—a criticism—of IRCAM, Pierre Boulez's music think-tank in Paris.1 While the tone of the book was negative, describing the culture as obsessive compared to the more intuitive climate of pop music engineering, I found the world described fascinating. This lead to an interest in Csound, the music synthesis programming language I have used to make the digital effects in this piece. This is “the hard way” of doing electronic music, as interesting and alternately wonderful and vexing as it is infinitely configurable.

Electronic music has interested me for a while now. But the highly abstract nature of much of the music that is the manifestation of all this research means that most of us can't make the connection between imagined possibilities and actual music. More to the point, the music is so exotic, the sounds cold and alien, and little of it is immediately attractive—this hinders a wider interest.

Through pop music, a small repertoire of sounds and textures have entered the canon of common techniques. Many people understand, or at least recognize, the sound of delay, of reverberation, of chorusing. Sensing Angels is meant to exploit this.

It is a piece which is built on these simple acoustic devices as it is built on the common uses of the clarinet. There are no “advanced techniques” here. My interest is in treating the computer like any other instrument that is known and composing in a language that is also known. It is still meant to be progressive.

The work on this project was partially funded through a Summer Research Grant to Dr. Krebs and myself from Truman State University. We would like to thank the University for facilitating our collaboration.

Sensing Angeles was conceived as a solo piece in which the acoustic sound of the clarinet would be manipulated in real-time by a computer. Today, this isn't so remarkable and can be achieved by a variety of means. For the project I chose to use Csound primarily as a challenge, but also for a certain kind of historical connection.

If the clarinet is an instrument with history so is Csound. A direct descendant of software created at Bell Labs in the 1950s,2 many would consider it antique in the same way the clarinet might be seen. Of course, it turns out they are both quite modern. Like most things, modernity is about use. Today, the use of a command-line interface and computer code for the creation of music is as much a point-of-view as a practicality.

There are many that mourn unrealized dreams of the computer and Internet ages, ages which began with an optimistic zeal for a world in which people would instruct computers to help them live lives of transcendence (what was transcended exactly varied greatly depending on who you asked). The ease and stimulation which computers provide most of us has turned out to be mostly about commerce and is actually quite generic. That those of a certain vintage remember our basic programming classes as merely charming rather than useful is just one more testament that our dreams of being liberated into a vibrant e-would of idea exchange and revitalized collectivism have devolved into a tawdry virtual peepshow and the quasi-tribalism of "friends.".

As for music, the electronic revolution has followed the same trajectory, first leading to the creation of compelling music but now the tools mostly are used for inexpensive art that is often redundant. More and more music software is about merely making things easy that once took time and consideration rather than facilitating the creation of exciting new musics. It is disappointing when one considers that music has always been closely tied to innovation.

At this moment, the commercial marketplace for computer music tools is about emulation of beloved hardware tools, the computer iterations becoming increasingly glossy and prescriptive. In this context a spare software interface and the typing of lines of text for the creation of a musical instrument make Csound a bit of a revelation.

The design of the instruments I created required painstaking thought and consideration of many of the laws of acoustics that I have been able to ignore in the past. (This type of slow going isn't natural to me, I love technology, but don't read manuals.) What is interesting about musical training is that the theory of music doesn't involve theory at all really. That many schools have come to use the term "analysis" is somewhat more accurate though what we really study is the ways composers of past centuries have realized their compositions using Guido's notation system. Levels beneath this are mostly ignored.

The signal path is quite simple: microphone, analog to digital conversion, computer, digital to analog conversion, amplifier, loudspeakers. The clarinet itself isn't amplified as the piece is meant to exploit the room acoustics in the same way most concert music does.

The performer controls Csound using a midi interface and a foot pedal. The clarinet requires both hands (most of the time) and the choreography of these controls are built into the piece. There are certain notes that can be played with one hand and those are moments when knobs can be turned and buttons pressed. However, most of the controls are modified very little during the movements.

I hadn't thought about improvisation for a long time, but when Dr. Krebs expressed his interest I thought the time was right to try my hand at integrating that kind of freedom into a composition.

For several years now collaboration has been a defining aspect of the way I work and I have collaborated with writers and choreographers on a variety of dance and theater projects. But I don't collaborate on music. I want control of that—it is what I do after all. So this piece moved me outside my comfort zone.

The improvised section in Part 1 is the solo starting at 3:10, I composed the solo up to that point, as well as the accompaniment. In Part 3 there are two section of improvisation between 1:48 and 2:14 and also between 4:39 and 5:01. All the improvisations can be as long as the performer wishes them to be.

Movement One
A chaconne-like progression is built through a long delay line. As the performer builds these loops, each is panned in place in the stereophonic field using the MIDI controller. Csound has the ability to transpose sound in real-time (no small feat) and this is how the low tones are generated. The clarinet sound is analyzed and then re-synthesized using opcodes developed by Richard Dobson and Victor Lazzarini.3 There is a bit of delay while the computer does the math for this. I would guess it is less than a tenth of a second using a macbook, but it is enough to be a bit of a demand on the performer's timing. The performer also controls the reverberated fade-out.

Movement Two
In this movement the transpositions are at a sixth and fourth below played pitch, what is called first inversion, with the performer being able to select the major or minor version of the sixth (the third of the triad) with the foot pedal. All are tuned to simple ratios 3/2, 6/5, 5/4. The score is planned so that pitches that can be achieved with one hand are the points where the volume of the harmonization can be controlled.

Movement Three
This uses a combination of short delays and the transposition opcodes. Again, the glissandi are possible because certain pitches on the clarinet can be played with only one hand, so the other hand can turn a dial.

— Charles Gran
November 14, 2009
Kirksville, Mo.


  1. Georgina Born, Rationalizing Culture: IRCAM, Boulez, And The Institutionalization Of The Musical Avant-Garde.
  2. See Barry Vercoe's forward to The Csound Book: Perspectives In Software Synthesis, Sound Design, Signal Processing, And Programming. Richard Boulanger (Editor).
  3. "The pitch detection algorithm implemented is based upon Schouten's hypothesis of the neural processes of the brain used to determine the pitch of a sound after the frequency analysis of the basilar membrane." Alan O Cinneide, "Introducing PVSPITCH: A Pitch Tracking Opcode for Csound." Csound Journal, Winter 2006.


©2014 Charles Gran, all rights reserved.