Nicolas Collins's book, Handmade Electronic Music: The Art Of Hardware Hacking, is a brilliant, hands-on guide to electronic music making. I've known Nic since the mid-1970s; he's been a friend, a colleague, a collaborator and a mentor, yet he never ceases to amaze me with his latest bit of music or techno-logical innovation.
Were you an electronics hobbyist as a child? Did you build crystal radio sets and fuzz boxes and hang out at Radio Shack?
Not really. For an academic art historian, my father was quite the handyman — good carpenter, plumber, car mechanic, basic household electrical — and these were the skills he tried to pass on to his kids. He was never happier than when something broke down or got clogged up. I got a toolbox for my sixth birthday, and on Sundays we repaired appliances, built shelves, and went to the dump instead of church. I think my having learned to solder and patch together a component stereo were, for my father, the most impressive achievements of my very expensive college education. In high school I briefly tried playing electric guitar (I was terrible, even for a 13-year old), which got me interested in electronic sound. I bought an old used Tandberg reel-to-reel tape recorder and discovered a hidden switch, accessed by a screwdriver through the back panel, that when thrown set up some kind of undocumented feedback loop from output to input — adjusting the volume controls while recording manipulated weird, totally un-Scandinavian shrieking oscillation, further modulated by head echo (3 speeds!) and tape saturation limiting — very cool. I was hooked. I have a theory about the rise of American homemade electronic music: it involves the Signetics NE/SE566, a chip I became aware of in 1972 in my last year of high school.
When did you make the jump from simple analog circuits to the world of digital electronics?
Almost from the start. For a long time one of the biggest problems for non-engineers making audio circuits was power supplies — they were difficult to build, and always seemed to contribute excessive hum and other noise. Early integrated circuits [ICs] consumed a fair amount of power, only functioned within a narrow range of supply voltages, and required two "symmetrical" supplies. "CMOS" [Complementary Metal Oxide Semiconductor] was the first family of digital ICs designed for low power operation over a wide range of voltages — they run for days on a 9-volt battery. CMOS chips became available on the hobbyist market in the mid-1970s. A number of us sound guys figured out how to make noises with these chips, which was surprisingly easy, if never intended by the designers. In fact, when it came time to develop a course in "electronics for dummies" at my art school almost 30 years later, I went back to these mid-'70s circuit designs — they run the first time, every time — and they now form the core of my book on hardware hacking. Because they were digital chips, they suggested "digital ideas". My own circuitry started to go in the direction of "proto-computer instruments". I built circuits that combined sound generation with logical operations of various sorts. I was inspired by the "coordination" scores of Christian Wolff from the early 1960s — I saw his pieces like "For 1, 2, or 3 People" [1964] as presaging computer music: all those notations that instructed the players not to play notes by following a metronomic time base, but by linking their actions to those of other players as if by logic circuits. So I built instruments that could only be played by multiple players — sounds were generated when two players did the same thing at the same time — for example, pressing the same key on two separate keypads ["ANDS", 1978]. I took a course in computer programming in college in 1974 or so. I figured this would be the future of electronic music, so I should learn it, but sitting in front of a mainframe seemed miles away from the concert stage. A few years later Paul DeMarinis said I should get a KIM-1, one of the first generation of affordable, single-board microcomputers. I demurred, saying I didn't like computers. Paul replied, "Don't think of it as a computer, think of it as a big, expensive digital chip." That did it. I bought a VIM — a slightly turbocharged KIM: 1 MHz 8-bit CPU, 1 k RAM (expandable to 4 k!), 6-digit numeric LED display, a small calculator-style keypad for writing programs, audio jacks for connecting a cassette recorder for saving and retrieving data files using finicky fax-like tones. It really was not so different from designing circuits — programming machine code makes you think about waveforms, about voltages going on and off one at a time, hopefully in the right sequence — very different from Max, SuperCollider or Pro Tools. I couldn't figure out how to make a D/A converter, so my "computer music" consisted of direct digital output from the computer — just 1s and 0s, square waves at different frequencies. To vary the timbre I'd make trains of different patterns of ons and offs. My first computer music was as much about the hardware of the machine as it was about the software I wrote. It was pretty raw. I listen now to pieces I did then and think, "Wow, that is so much more extreme than anything I've done since" [like "Little Spiders", 1980]. Most of the composers of my generation who started soldering their own circuit instruments in the 1970s went over to programming by the end of the decade; they rode the wave from KIM to AIM to Commodore 64 to Apple II to Atari to Macintosh and never looked back at hardware. I always kept one foot in hardware and one in software. One thing I learned from David Tudor was that music seems to get more interesting when you add more connections: duos generally are more engaged than solos; a guitar sounds better though a distortion pedal and amp than it does unplugged; and lots of cheap circuits patched together with some feedback loops sound way more interesting than any MIDI keyboard. Unfortunately, computers tend to be permutationally impoverished in the connection department. Yeah, I know someone's going to say, "what about plug-ins?" But it's not the same without real patch cords. So after a few initial forays into pure, computer-generated sound I settled into an impractical, but richer strategy of combining programs I wrote with hardware I built or hacked. Around 1980 I got interested in early hip-hop DJs like Grandmaster Flash, and I wondered what it would sound like to have a dozen turntables cross fading on beat, instead of just two. So I built a crude automated mixer that could detect the beat in 16 channels of audio material and cut between any two channels when they came into sync on a downbeat. The performance consisted of constantly "feeding" the mixer with different audio material — tape loops, drum machines, electronic toys, radios, etc. — and letting the computer work out a perfectly seamless, if illogical, mix of these otherwise unrelated sounds — a sort of schizophrenic form of dub ["Is She/He Really Going Out With Him/Her/Them", 1982]. Later I used the system to tighten up the punch-ins on a more overtly rhythmic record of mine [Devil's Music, 1986] and a few tracks on an LP I produced for your band, Western Eyes [Western Eyes, 1984].
In the 1980s, after a number of years working in "pure" electronic music and acoustics and feedback-based music, you started tinkering with electric guitars. How did that come about, and what exactly is a "backwards electric guitar"?
Moving back to NYC in 1980 after college I was slowly drawn into a music scene that was more about playing than soldering or programming. I realized that guitar bands had a serious theatrical edge over knob- twiddlers, so I decided quite pointedly to make instruments and pieces that could compete for attention on the noisy New York stage. My first strategy was the "backwards electric guitar": I connected amplifier outputs to the pickups of some pawnshop guitars via some cheap matching transformers so that sounds plugged into the amp would resonate the guitar strings and turn the instrument into a mechanical filter — like a spring reverb, only one you could "play" by fretting chords and muffling strings — as if an EBow had an input jack. I asked some musicians to stand on stage and work their left hands while I sent a whole bunch of different sounds into the strings. It looked just enough like any other band of the time that we managed to open for Rhys Chatham at CBGBs in 1982, but it didn't really sound like one ["A Letter From My Uncle", 1984].
I performed a solo piece of yours at The Kitchen during this period. I vaguely recall a few toy drumming panda bears....
Yeah, you of all people should know how difficult it is to find and keep a good drummer. The six bears worked pretty well together, the computer kept turning each one on and off, so the fact that they were playing waltz time didn't become too apparent. They were mic'ed and sent to resonate the strings — I used that conning tower-shaped ribbon mic [Altec 639] you some-times see in front of a ranting Harry Truman ["Killed In A Bar When He Was Only Three", 1982].
You were the first person I knew who owned an Electro-Harmonix 16 Second Delay, which I promptly borrowed. This was in 1982 or '83. How did Devil's Music come about?
As music technology went digital, got cheaper and proliferated in the 1980s I realized that for a non- engineer such as myself, it was more efficient to modify commercial devices to suit one's personal needs, rather than continue to build from scratch. Hacking gained an advantage over honest design. Let me digress for a moment — from the day I started studying with Alvin Lucier in college I was paralyzed by John Cage — a nice enough guy, but he ruined my life. Under the sway of his maxim that "any sound can be a musical sound," I found it impossible to choose any one sound over another — they all seemed equally valid. My first way out of this conundrum was feedback — "the infinite amplification of silence" seemed a natural way to make sound without making any decision — plug in a mic, turn up, and let the architecture compose the music. I did a lot of feedback pieces in the 1970s ["Pea Soup", 1974-76, revived and reworked 2002] and periodically return to the stuff when I'm stuck ["Second State", 1981; "Charlotte Aux Poires", 1997; "Mortal Coil", 2002]. When Electro-Harmonix made the 16 Second Delay around 1980 — the first affordable, malleable DDL — I saw that live sampling could be another route to "making music without making a sound." I bought one of the first ones out the door of Electro- Harmonix, and made a piece called "Vaya Con Dios" [1984] using the loop as a way to accumulate, intercut and modulate samples from a Reagan speech and some old pop ballads — a ham-fisted political piece about America's clandestine war in Nicaragua. What bugged me about delays and echo, however (and still irritates me today), is their periodicity/repetitiveness — I've never been interested in a steady beat, and DDLs out of the box are way too steady. So I built a very simple 99-cent "stuttering circuit" — whenever there was a peak in the signal connected to the input of the delay it would reverse the playback direction of the loop — it was as if you had an "auto-scratch" function on a turntable that would reverse its direction on every beat from the other turntable, or were scrubbing tape back and forth across a tape head when editing. The hack was easy — much easier than building a sampler from scratch, or writing my own sampling code for any of the personal computers of the time. I added the same circuit to a couple of "Super Replays" (Electro-Harmonix's next product, a 1-second trigger- able sampler), and made a piece called "Devil's Music" [1985]. I'd grab samples of radio stations on the fly, and the circuit would retrigger the loops to create a wonderful sea of shifting rhythmic patterns that made the sample seem much more complicated (and longer) than they were. Within a few years, hip-hop's hiccoughing retriggered samples would become all too familiar, but in 1985 it was still fresh. In 2002 John Corbett, a friend of mine in Chicago, asked me to revive the piece for an all-night, multi-DJ performance at The Empty Bottle. The circuits were rusting away in an attic on Cape Cod, so I threw together a quick and dirty program in Max/MSP that emulated pretty well the original boxes — complete with stuttering circuit and slightly reduced bandwidth. Subsequently I've emailed an improved program to DJs and lap-toppers in the UK, Germany and Italy, and we've done nice long nights with multiple performers, segueing and mixing together. It's the closest I'll ever come to writing a song that can be covered creatively.
How did you come to start teaching hardware hacking and write the book? We moved to Amsterdam and I had an invitation to be the visiting artistic director of STEIM, a foundation that does specialized music technology research for artists. For four years I coordinated research, organized the visiting artists program and programmed concerts. From there we moved to Berlin, where I had a wonderful composer's residency under the auspices of the DAAD. Money got tight and I decided maybe a job wouldn't be a bad idea, and I was lucky enough to get my first regular teaching position in the Department of Sound at The School of the Art Institute of Chicago. It's a great job — a sound department in an art school (one of the only ones in the USA) attracts a very eclectic bunch of students, much more experimental than you'll find in most music schools. When students realized that I knew something about the ancient black art of circuitry, they started asking advice, mostly about interfacing the real world to their computers. I decided we needed a course in "electronics for dummies". I start with making contact mics, using telephone taps to sniff electromagnetic fields, wiring up tape heads to "scratch" data cards. Then we open up transistor radios and emulate the infamous STEIM "Cracklebox" by licking our finger and touching the circuit board until we find feedback paths that cause it to swoop and squeal (in a most tactile, playable fashion). We do some basic "circuit bending" — hacking toys to make non-toy sounds. We build from scratch a handful of very simple, guaranteed-to-work circuits based on digital logic chips: oscillators, preamps, distortion circuits, tremolos, gates, panners, etc. We finish up with simple passive mixers and hacking sensors onto cheap game pads to make computer instruments. The emphasis is on new ways of hearing things (through alternatives to normal microphones, for example) and making highly playable gizmos — doing easily in simple circuitry what's difficult to do with computers. It was an interesting process, after almost 30 years of making my own circuits, to settle on designs that could be taught with an absolute minimum of theory, no special test equipment, could run on batteries, and that — although individually very simple — could be combined in multiple permutations to yield sophisticated, unexpected results. The class handouts eventually got spiral- bound, and this workbook escaped into the wild — like some invasive walking catfish. It got into the hands of a few producers in Europe who asked me to do workshops, and then things kind of snowballed. I've done a dozen workshops from Brussels to Beijing in the last two years, and an editor at Routledge asked me to develop a book for publication. I added an audio CD, a dozen sidebar profiles of interesting people working with homemade circuitry and a lot of nice pictures.
What are some of the most popular hacks?
That's hard to say — everyone has a favorite. Turning the radio into a synth with wet fingers is a real liberating experience for someone who has never messed with hardware — a roomful of novices with open radios has a distinctly speaking-in-[silicon]-tongues quality. Contact mics are always fun — some of the most unassuming objects (such as Slinkies) make the most extraordinary sounds — as is scratching credit cards with a tape head. Six oscillators on a single 20-cent chip make a glorious din. The workshops and the book try to provide something for everyone, and move along fast enough that people get exhausted before they get bored.
One of the things I like about your approach is that there are loads of people building their own fuzz boxes, treble boosters and tremolos, but your stuff is, even at its most conventional, unconventional.
Empowering as it is, "circuit bending", as inspired by the wonderful Reed Ghazala, has its peculiar orthodoxies. Looking at my instruments as I was setting up a demonstration at the Bent Festival at The Tank Gallery in NYC in 2004, an audience member inquired, "are they bent or hacked?" When I looked baffled he elaborated: "'Bent' means you have no idea what you are doing when you open up the circuit; 'hacked' means you have some idea." I don't care about those kinds of distinctions. I just try to find the best tool for the job, bought or made, acoustic or electronic, analog or digital, software or hardware, hacked or bent. In general, because I'm not an engineer, or a great programmer or wealthy, I look for the simplest, cheapest solution; but because the solution has to serve "art" rather than commerce, I also want a solution that is rich. I've never had to build "a fuzz box", so if I need to distort something I might or might not end up with a fuzz box. Throughout my workshops and the book, for example, I emphasize the importance of doing things "wrong", backwards, of trying anything at least once. I chose designs and components that were cheap, easy to get running, capable of operating over with a wide range of component substitutions and unlikely to burn out no matter how you hook them up. This led to some very non-standard devices and designs — there are no "op- amps" or transistors anywhere in the book, for example. I guess unconventional has been conventional in my own work for so long that I don't even think about it anymore.
