Music With A Purpose (1988?)

Tell a tale. Choose a dream story or recreate a yarn. Tell it with sounds and words.

The sounds: based on numerology or randomization or equations or patterns.

Organize multiple structures like Polish collapsing Communism — a dictatorial administration transforming [via a trade-union movement] into a popular representative democracy. Recognize the bureaucratic and corrupt inefficiencies of the first, their obstructions to change; the suppression and (later) inexperience of the second; and the factionalism and movement towards chaos of the third.

Net.control--+                               (synths)
 (composer)  +--Consuming/producing public--5 voices
an allegory  +--Gov't bureaus & army--------6 voices
             +--forces for change-----------5 voices
             +--Soviet influences-----------3 voices
             +--world opinion---------------2 voices

Automatic Music (1991)

My creative process: Devise a process that, auto­mati­cally, produces interesting outputs with no further manual intervention. None at all. This might require at lot of artificial intelligence, or an audience with low standards.

Process Music (5 May 1993)

Anyone can write music for humans — you can look at a scribbled / graphic / textual / nonsense score, and decide how to interpret it. Notation for computers is another kettle of spuds. We must be specific — if preparing a graphic score, be aware that the system will play exactly what we draw, and no more. If writing programs to generate sounds, we must carefully choose the amount determinancy and randomness to be allowed in the piece. It is very difficult to be precisely random.

My Music — Some Inclinations: (1995-2001)

1. I'm mostly a guitarist - folk, blues, pop, some jazz, experimental.
2. As a lyricist, it's some lyric poetry, lotsa punk rants, word & sound games.
3. I like making loud noises with electrified stuff.
4. I like making analog synths go BEEP-BRAP-BURP!
5. I have sounds in my head like:
   • loud punk w/ industrial noise features
   • classic string & wind chamber ensembles
   • space music, consonant whispers
   • Zappa-esque orchestrations
   • mad, fast, impossible dances in unlikely metres - and exploiting waltz, tango, polka, twist, lambada, etc
   • brownian 12-tone (or microtonal)
   • stretched & malformed & twisted samples
   • Kurt-Weill-ish somgs (oom-pah-fizz-decay)
   • the most subtle of microtonal soundscapes
   • sampled voices subjected to the severest manipulations
   • PROCESS MUSIC! or, grow some musics...
   • themes, song-cycles, reducto ad absurdum of whatever meme takes hold
6. Sometimes I study, sometimes I practice, sometimes I record, sometimes I don't.

Echo Arpeggiation (1993?)

Generate a succession of equal-tempered pitches; apply rules determining the allowed vertical intervals. Thus, given note X, make note Y if X:Y is allowed, add Y to chord, else use Y as the root of the next chord; then arpeggiate downwards. Note range of intervals between top and root notes in chord, ie (with notes on all lines of treble staff) the distance between N1 (F5) and N2 (E4) is 13 semitones. Play the chord, then play the arpeggiation N1,N2,N3,N4,N5 (F5,D5,B5,G4,E4). Then for notes N2 (D5) thru N4 (G4), adjust the intervals microtonally to provide 4 steps.

Or, if N1-N5 = 34 semitones, then for the first echo, simply make a descending arpeggiation of the notes in the chord. Because the interval span is 34 in the second echo, the pitch span between the lowest tone in the range of the piece and the next highest pitch in the chord is divided into 34 equal intervals. The resulting structure preserves the relative size of adjacent intervals of the first chord, but now includes tones OUTSIDE the 12-tone, equal-temperment scale.

Successive arpeggiations using the same method of interval division are used for the 3rd, 4th, 5th echoes. The structure is thus echoed the same number of times as the total number of notes in the chord.

12-Tone Forever (1993?)

"Total control" ala 12-tone technique involves specifying not only the pitches, but also the thythms, durations, timbres, intensities et al of all the elements of the composition. It's trivial to write programs that select non-recurring elements from tables of such qualities. Some of the elements may include:

• 12 duration classes: 16th note, 8th note, dotted 8th, 1/4th note, 1/4th + 16th, dotted 1/4th, double-dotted 1/4th, 1/2 note, 1/2 + 1/16th, 1/2 + 1/8th, 1/2 + dotted 1/8th, dotted 1/2
  
• Rhythm patterns: Morse, ASCII, Baudot, BCD, EBCDIC, etc
  
• Some rhythm classes: 1:2, 1:3, 2:3, 1:4, 2:4, 3:4, 1:5, 2:5, 3:5, 4:5, 2:1, 3:1, 3:2, 4:1, 4:2, 4:3, 5:1, 5:2, 5:3, 5:4

Also amenable to automatic control/selection are such elements as: volume levels; timbre qualities (voice sets); spatial positioning; harmonic/chord patterns; tempo; et al. All these can be encapsulated in lists, and the composition software can select the desired qualities for the music according the the programmer's dictates.

Bonfire Sonata (2005)

Somewhere back in the late 1960s or early 1970s I developed a pyromaniac's musical automaton. Fairly cheap, too. The basic hardware looked like this:

• WHINING: I build some basic oscillators, relaxation RC (resistance-capacitance) circuits where the variable resistor took the form of a cadmium photocell. (Thus varying light on a photocell would produce varying tones from an oscillator).
• THUMPING: I picked up some piezoelectric crystals, old crystal (not dynamic) phono­graph pickups, at junk shops and thrift stores — an obsolete phonograph could then be had for under a buck. I attached wires to the pickups. (Such pickups produce a varying voltage when subjected to varying pressure.)
• VOLUME: I built a couple basic audio mixers, and high-output amplifiers, then wired speakers from each amp. The oscillators and pickups were to be wired into the mixers; the mixers were to be wired into the amps.

Next I built (but did not yet light) a loose bonfire in a large outdoor space. I arrayed the speakers around that space. I took a cracked piece of dry drift­wood and embedded the pickups at various points in the wood, stabilized with a bit of glue. I leaned the driftwood on the fire structure. I put stakes around the fire with the cadmium photocells attached, pointed towards the fire. I made sure the photocells were wired to the oscillators, and that the oscillators and pickups were wired to the mixer. That was the basic setup.

After sunset I lit the bonfire and turned on the amps. Flickering firelight activated the photocells, causing the oscillators to produce different wavering tones. As the pickup-embedded driftwood caught fire, its crackings were amplified. All these sounds wafted from the speakers, echoing off distant objects. As fire reached the pickups, each died with a piercing shriek. As the fire died, the oscillations dropped away. And thus was the Bonfire Sonata.

Yeah, multitrack recording and newer digital technologies would permit much more control over and richness of the sonic experience. But simple is good, eh? Especially if one isn't thinking clearly.

Modal Serialism (2006)

Making 12-Tone Music Sound Good.   In olden days, like a decade and more ago, I wrote music. Actually penciled notes onto score paper. Mornings, I'd arise hungover, eat something, drink coffee, then retire to bed with a couple glasses of wine. Turn the radio on to KPFA for the New Music program, sketch out abstract soundscapes according to various formulae. Later I'd get up, drink more coffee, head for the computer and key what I could into the AdLib Composer software for synthesized realization. Some fraction of it sounded pretty good.

I don't do that anymore, none of it. But I did manage to come up with my own style of 12-Tone (Serial) composition, producing works that didn't sound like tortured cats. How is this possible? With strict serialism, one constructs a row of 12 arranged semitones, then writes music using those notes IN ORDER, not repeating one until all the others are used. This technique natually impresses callow sophisticates and alienates everybody else, giving Serialism a bad name.

My approach: Generate a series of randomized tone rows on the computer, with inversions etc, giving a bunch of 12x12 tables. Decide on a KEY for the music — for example, A diatonic (Aeolian mode). Decide on a harmonic+time structure — for example, that A will have the duration of a whole note; C and E will be half notes; B, D, F and G will be quarter notes; and all the accidentals will be 1/16th notes. Now start writing.

Pick any row. Write the notes in the order seen, as chords or strings of the same duration. So if the row starts with a string of accidentals, write them as a 1/16th note chord or fast arpeggio. Then to natural note(s), singly (maybe doubled at octaves) or chorded. For a simple melody, repeat one or two or three rows this way. Want more complexity? Run through the whole sequence of rows, back and forth and up and down. It's so mechanical, a computer could do it.

The same randomizing row-generation program can also produce durations and thus rhythmic patterns. Sometimes I referred to cheat-books of Latin percussion patterns and imposed that order on the harmony and melody. Yes, a 12-tone rhumba!

Now key all this into the AdLib Composer. Assign different classes of notes to different voices: the accidentals might be flutes, the naturals as horns or oboes or chimes. Lay in the rhythm track. To boost the AdLib's limited synthesis, double many notes a few octaves down. All this is pretty easy with the software's piano-roll interface.

Maybe someday I'll reveal how to turn a text into Morse Code and use THAT as the rhythmic basis of a composition.