This article originally appeared in the November 1986 issue of
Should you ever meet her, do not tell Wendy Carlos she changed your life. Even if it's true. Her good honest Rhode Islander reserve won't let her stand still to enjoy your equally good and honest compliment: instead it'll be duck and run, and miss the point, and let's talk about anything else, shall we?
So let's talk about something else! For now, carefully ignore that 18 years ago Wendy Carlos, in a mind-boggling display of focus, virtuousity, and sheer physical endurance, created a record called Switched-On Bach and changed all our lives. It didn't happen; it never existed; and you and this magazine are face to face because synthesizers have been a part of human pop culture since way back in the early MiDizoic, when shambling proto-Californians discovered the microprocessor and bandwidthed together to defend themselves from vicious sawtoothed tigers. (Facts are just facts, as Professor Peter Schickele is fond of pointing out; can't argue with facts.) Instead of debating whether the field's debt to Wendy is vast or merely gigantic, let us focus on some of the things about her that you don't know, that you couldn't possibly know, because no one has ever come right out and said them in print. Little things, true, but telling.
Wendy is an avid astronomer who has traveled all over the globe pursuing solar eclipses; in fact, along with her partner Anne-marie Franklin, she has produced some of the best eclipse photos ever taken. She's a big fan of Carl Barks, the comic-book artist who chronicled the adventures of Donald Duck. Much of her recording gear is self-designed and self-built or so thoroughly modified as to blur original distinctions (the Olde English lettering on her mixing board is a dead giveaway). She loves hard science fiction as practiced by masters like Arthur C. Clarke, but has little taste for fantasy. To recuperate after an auto accident, she took to inventing brand-new map projections of the globe and then wrote her own plotter driver software to print them out. She loves theatre organs. She hates (conceptually) yuppies. She loves garlic. A sheriff in Texas once proposed to her by mail. She lets her three Siamese cats drape themselves decorously over her outboard gear; never mind the fur. She has a phenomenal capacity for attention to detail and, like most visionaries, little patience with the limitations of the world. Her tape editing is sharp enough to get a razor in between a hic and a cup. She paints and draws, a little. And even more than never telling her she's changed your life you must never, never ever ever, get her started on puns and Monty Python routines. Not unless you do that kind of thing too. And have no pressing engagements. For days.
In short, Wendy Carlos is a fascinating and diverse person. That she's a brilliant composer whose newest (so far unreleased) music promises to do for tuning what her first record did for timbre...hey. Gravy.
Wendy is one of the few people I've met whom I have no qualms about calling a genius, with all the pluses and minuses automatically attendant upon the term. Genius is a mixture of qualities combined in such a way that the genius pulls things out of the world that are A) starkly obvious when you know where to look and B) completely hidden until the genius shows you how to look. Usually by example.
In 1968, Wendy did it with Switched-On Bach. It went platinum, the first classical record ever to do so, and whether people bought it for love or novelty matters not one whit. All that matters is the door was opened. Glenn Gould called it the "album of the decade"; I'd go further and call it one of the ten seminal recordings of the century.
It is also the worst thing that could possibly have happened to her career.
Think about it. SOB was a step, not an end in itself. Wendy was exploring possibilities. Her point was to conquer a demanding, finicky, fragile collection of wire and knobs and patch points called a modular Moog synthesizer, demonstrating along the way that synthesis and electronic music were good for real art as well as ingrown academic exercises. This the record did, making much money for CBS... a fact CBS executives seem to have misunderstood, supposing the secret was in Bach and the synth, rather than the synth-player.
They weren't alone. A wave of "Switched On Thus-and-Such" spread locust-like through record stores (even today we aren't free, as witness such derivative efforts as the recent Bachbusters CD). But the real problem for Wendy was not incompetent competition: it was confusion of identity in the one place no recording artist can afford, her own record company. I feel Wendy's catalog since 1968 is virtually a case study in artist mismanagement, with her strengths downplayed and her weakness-repetition, which she simply has no talent for-accentuated. This is not to say that her original work has not seen release. But just try and find her two-record environmental sound piece, Sonic Seasonings; her scores for Tron and Clockwork Orange; her By Request (which is something of an historical hodgepodge, marred by some more-gasp-Bach; but to be treasured nevertheless for the riveting snap of "Geodesic Dance" and the intricate musical joke called "Pompous Circumstances"). And if you find any of these, send me a note, because my copies are long since worn out, and I hanker to hear Wendy's music clean.
Despite these marketplace frustrations, Wendy has never stopped pioneering. Each of her records, including the Bach ones, have pushed at the boundaries of available technology and technique.
Case in point, her last album, Digital Moonscapes. It's been out for a year or so, but don't be surprised if you've never heard it: this marvel of original electronic music was dropped into the American marketplace with little or no fanfare, promotion, or advertising, and on a classical label (CBS Masterworks) to boot. What makes Digital Moonscapes special above all else is that it marks the culmination of two years of intense research into re-creating orchestral sounds through digital synthesis. Using the GDS digital synthesizer system she got toward the end of work on Tron, and a couple of its Synergy offspring, Wendy crafted what she refers to as the "LSl Philharmonic Orchestra": a collection of several hundred simulations good enough to stand with pride against the yardstick of the real thing. (Designers and collectors of myriad DX7 patches, take note: listen to this album and see how it should be done.)
The only place to go when you've gotten this far is farther. Wendy's newest project, Beauty in the Beast, takes digital synthesis beyond timbral recreation into extrapolative invention. More importantly, it casts off the leg-irons of the equal-tempered scale.
Some of you know what that last part implies. Some of you don't. Those that don't shouldn't be ashamed, because the standard 12-note octave is so deeply a part of modern Western music that even talented, well-trained musicians take it for granted. But it's not Holy Writ. It's a so-so bargain between the physics of sound and the physical limitations of acoustic instruments, a 300-year-old compromise designed so that every note, in every key, is equally not quite in tune. In other words, you can get by with it. It'll do. But it's a long way from perfect, and there have always been a few hardy (foolhardy?) souls trying to find their way to a better, more musically consonant system. Unfortunately, the same laws and limits that forced the compromise in the first place have not been repealed, and alternate tunings have so far proven both esoteric and impractical.
Enter the computer. And the computer-based instrument. Although there are still massive hardware problems to solve, largely in the areas of controllers (taking one example, how do you physically play a scale with 31 notes to the octave?), it is finally possible to up anchor, sail off on the sea of tuning, and navigate to safe harbor in new countries of sound.
Exploring. That's what Wendy's new album is all about. But best to let the traveller tell the tale.
EM: What led you in this radical direction?
WC: For a long time-since the early '70s-l felt I had to get away from the compartmentalization of sounds; we had rich acoustic sounds on one hand and impoverished synthesizer sounds on the other. The analog synths of the time, and even the newer digital machines, just didn't have sophisticated enough control over sound, and the timbres they produced quickly became very boring-to my ears, at least. That's the reason for the orchestrational style I developed from the first record on, in which I jump from timbre to timbre to timbre so quickly that by slight of hand it gives the impression of considerably greater timbral resources than really existed within the instruments. There are possibly half a dozen basic sounds you can get out of an analog machine, short of some very silly things. They're like little islands that sound good floating in an ocean of possibility. There's a percussive envelope/filter thing you can do on a bright wave, and it doesn't matter much what kind of bright wave; there's a slow attack filter that's more like an imitation of a brass instrument; there are nice, simple, dull waves like triangles and sines that you can put together with either soft or hard attacks; tuned noise, delayed vibrato, and a few others... just these categories and a few others, and that's it. It's a little like drawing in pen and ink. You can make solid shapes, you can do crosshatching, you can do little dots, you can have outlines, but it's all pen and ink, and everything else you get is from juxtaposing these in more or less intricate, clever ways that give the eye the impression more is happening than really is.
EM: Limited or no, you got a lot out of that repertoire of sound.
WC: My earlier pieces worked well in spite of the repertoire of available sounds, rather than because of them. So by the late '70s, I was trying to find new and different generation techniques and exploring digital synthesis when I met Stoney Stockell and Tom Piggott and the folks behind the GDS and Synergy. They wanted me to get involved in their products, and Disney's Tron provided the opportunity. The Synergy looked like the only commercial instrument that would do the kinds of things that I wanted to do, so after Tron, I began to learn how by developing acoustic replicas on it. After doing 300 voices of the orchestra, I'd pretty much tamed the instrument, and decided to record Digital Moonscapes as a waystation from the old analog world into what the future promised. It was a way of saying "Hey look, we can get very, very close to the orchestra now. That's a big step from where we used to be! And with the ambience techniques now possible through digital time processing and reverberation, you see we've joined the worlds of acoustic and electronic... that's what's possible now, and on the next record we'll see where it can lead."
EM: What about the way you've jettisoned standard scales?
WC: About the time I began the current record, Stoney finally unbuttoned the frequency tables in the Synergy, so I wrote a bunch of custom control software that made it possible to retune the instrument. That's something I've wanted to do for a long time. A digital instrument is a natural for microtonal tuning because you can be precise to any degree you need and also repeatable. It's idiotic that Western music has remained such a slave to a tempering system which evolved 300 years ago as a satisfactory compromise. We don't need the compromise anymore. We can begin to work in areas that up to now have been forbidden because we only had the equal-tempered scale.
Where these steps may lead is anyone's guess. How important these new areas will be I don't know. We'll have to find out. I do feel they have become very important areas to explore and that their implications transcend me as a human being and a composer and an artist. They're bigger than any one person. These ideas have to be disseminated. They mustn't become a quaint personal system, like Harry Partch's, but must get taken as much for granted as people now take synth work and multi-track recording. These areas are so rich with possibilities, they are an ideal way to get pop, classical, jazz and contemporary music out of the cul-de-sacs they're all in, that I feel I could easily become so filled with messianic zeal I'd probably hate myself.
EM: With all possible timbres, and all possible tunings, where do you begin?
WC: When you're given all possibilities, you're in a worse position than you were before. It's a perfect way to drown. In fact, you've really chosen to drown in the middle of a very large ocean. Several oceans. And there's not even a floating log nearby; you've discarded all that. But there just isn't any way art can work outside of a discipline. Stravinsky's wonderful comment when people asked him how he felt about working on a ballet with Balanchine, because the form was rather restrictive, was "I love exact specifications." That's how it is with all art. It works best when there are limits.
Quite probably my earlier records were aided greatly by the fact that they were done while working within very narrow regions of possibility. But in late 1984, I found myself swamped by the anarchy of total possibility, so I began making choices. I chose to limit myself to some small, selected regions of the palette of "everything." In timbre, I decided to see what would happen if I took orchestral instruments that I understood and began combining selected properties of two or three, creating hybrids. That's a fairly small cast of the line; it's not nearly what the hardware allows, but it's a good way to learn the limits in a disciplined manner. I was mainly concerned with learning what rich things could be developed from models of past good instruments: the best Stradivarius overtone structure merged with the best Steinway action. What does that do? Does it sound good? And the answer is yes, it does, it sounds delightful. There are in fact several ways of doing it, and they all sound wonderful. Or take a good Boehm clarinet and merge it with a Guarneri cello, or... you find a lot of fascinating sounds this way, because you are standing on the shoulders of giants of the past of timbre, and yet they are genuinely new sounds, subtly or wildly unlike anything ever heard before.
EM: And tuning? There's even less to stand on, outside what we're all used to.
WC: I cast my sights: what happens if we move out only into the realm of tunings from other cultures? Cultures like those of Bali, Java, India, Africa, or the Middle East... let's explore what they've found rich for many years. And let's also explore things that are fascinating mathematically, like variations in the overtone series and scales built with different numbers of equal-sized steps in an octave. Let's find what those sound like, doing only a few of them, and use those few as a guide. Everything might be possible hardwarewise, but I'm not capable of that, so while I have taken big steps in tuning and timbre they are not the biggest steps imaginable: just the biggest steps I could take while keeping control.
EM: In retrospect, how do you feel about taking these steps?
WC: As I stand here a little bit away from the coastline, the ocean seems far bigger and far more profound than I imagined it could be. It seems to me in hindsight that this was the right thing to do. A wise step. But I didn't know it was wise when I took it; I was just working from instinct. There is no one tuning to the album, no single timbre. Each selection is an essay that explores one or two ideas fairly deeply, rather than a lot of them superficially. I'm like a blind person in a room, poking a long stick in several places to make sure there's an elephant in here, instead of taking a sharp pencil, poking in lots of shallow places, and deciding there's nothing in the room after all.
EM: Did you find yourself having to fight not to think in old, familiar ways?
WC: Constantly. I also found that your conditioning and learning, the things you've developed-your strengths, in other words-can be crippling when you're trying to take new steps, because you keep falling back into habit patterns. Although it's self-conscious, you've got to deliberately break your habit patterns. For example, the last album was totally notated first, written out just like any orchestral composition, and then played. This is an approach I know very well, because of all the Bach records. It's a safe way to work; not much risk. So on this new album I chose to work in a scary way. I composed directly on tape, relying on sketches and improvisations which were edited many, many times and re-improvised and reedited until they grew into compositions.. . working that way, things happened that I didn't know how to write down. And I had to hold it all in my memory, which is scary for me because I've always had the same crutch Stravinksy had: he used to say he didn't compose except when he was at the drawing board with the manuscript paper right in front of him.
EM: You mentioned you'd been interested in alternate tunings for a long time. When did you first start?
WC: Oh, way back in my teens, probably from reading some magazine articles. When I was 16,1 bought a piano-tuning hammer and wedges and began retuning my parent's spinet piano in all manner of unorthodox tunings, trying to find out what some of these things I'd read about sounded like (you couldn't then find records with these things on them). In college I got involved in musique concréte pieces with retuned pianos and arbitrarily tuned sine waves and stuff like that. Nothing very profound. But I got inspired to put together a series of special reference tapes with a physicist friend.
We had access to several very expensive audio oscillators in a Brown University laboratory, test devices worth several thousand dollars, and we'd go in and tune one to a 440 Hz reference signal broadcast by short-wave radio station WWV, then tune another against it until an oscilloscope pattern told us we'd reached the particular ratio of our choice. Then we'd tape that. We ended up with a library of something under 100 pitches in an octave, all of which were derived from pure thirds and fifths. Most of the pro tape machines of the day tended to run at pretty much the same speed all the time-if they did vary from day to day it was within the limits of our precision-so we spliced some leader ahead of the recorded strips, labeled them, and from then on whenever we wanted to do any precise intervals or ratios we would use the strips as tuning references. The technology was never intended for this purpose, but there was just no other way to make the empirical tests necessary to take intellectual ideas about tuning and turn them into something that might have practical and pragmatic value. Even then I knew that you have to have a practical application. If there's no way to use something in a real musical context, who cares?
EM: After that?
WC: Well, I readjust about every book on the subject, but after college it wasn't until 1984 that I started experimenting with tunings again. I did have other things going. But when I came back to tuning it was as a gleeful child in a candy store. After Moonscapes I listened to a lot of ethnic records, deciding what direction I wanted to take the new album. I'd sit and try to play along with some of the different scales, but the equal-tempered scale didn't fit very well. It was driving me crazy. I tried minor variations. I even explored quarter-tone scales, but these were even less musically useful than equal-temperment and I'm amazed that so many musicians have bothered to explore them.
Anyway, in July 1984, bingo, Stoney presented us with some new chips and said "look in there at byte so-and-so, perform a write-read, and you'll get the pitch table." At the time, I didn't have any idea what it did, but I pulled out all the numbers, dumped them into the Hewlett-Packard 9825, and fiddled around with them until I figured out how to do some things. There was a small section of 12 x 2 byte values that the Synergy actually used for tuning, with all the values for all the other octaves derived from those through 2:1 (octave) ratios. I found the way to convert those from Hertz into cents, wrote a piece of software that slowly grew and grew and grew but finally made it possible, after much hassle and lots of math and tricks, to move the notes individually with a resolution of about 1.5 cents, It made it very easy to set up all kinds of what-if situations. For example, one piece on the new album, "Just Imaginings," was based on asking "what if all the notes of the scale were tuned to represent the closest natural harmonic overtones from a single fundamental?" That means you have to store the overtone series for every key-note with which you want to work I wound up with an array of 12 times 12 (144) different pitches in the octave, each of which represents a pattern of the 12 closest-fit partials to a particular fundamental. You play the piece by starting in one key and then as you move along to another, you hit a reference key on a special little keyboard and it instantly retunes the whole instrument.
The last pan of this track ends by going through the entire cycle of 12 pitches-the "circle of fifths," if you will-except all in perfect tuning. That's one of the things you aren't supposed to be able to do, which is exactly why I did it! The nice thing about this Harmonic Scale, as I call it, as opposed to normal just intonation, is that you can put down a cluster, just play anything, like a three-year-old kid, and it will always be in tune. The accompaniment at the end is a continuous cluster. Every note remains down, but from moment to moment all the notes are slightly retuned to match whatever note the melody and bass is playing, so the same cluster always harmonizes the melody perfectly.
EM: What about the other tracks? You've got an African-inspired piece, a gamelan-inspired piece.
WC: In 1983, chasing an eclipse, we spent some time in Bali and fell in love with the place. I had to write music "about" the experience. "Poem For Bali" is in authentic tunings taken from cassette recordings I made and bought when I was there, and also on a few records I've found since. (The tunings aren't based on any textbook descriptions; it turns out there are a lot of discrepancies between what's in print and what's really used.) The piece has ten sections written wholly in several varieties of Pelog (and one Slendro) tunings used by Balinese gamelans. Near the very end is a dance which I turn into a concerto for gamelan with symphony orchestra. That's a stunt that can't be done in real life because of tuning differences, but when you hear it here you'll wish it could be, because its a really great stunt
The funny thing about gamelans is that their scales sound pretty horrible on our Western instruments but are really damned good scales on theirs. And it isn't because they are somehow quaint savages or primitives who only have five notes-you know, that condescending "hey, maybe someday they'll get to seven" attitude Western music takes-but because they found a way that empirically fits the overtone structure of their instruments. And their tunings work well there. Ours don't. In fact, our pure octavecommandment number 1, "thou shalt not use anything but 2:1 for an octave"sounds pretty awful on their instruments. It sounds flat.
EM: If each step of exploration is built on the previous one, I expect the last track you finished must have been pretty unusual. Which one was it?
WC: "Beauty in the Beast." It's son of a rondo-like form, with two main themes that come round and round again, always in motion. This track could only be written for the electronic medium. It's built on two scales I discovered. One uses 78 cents per step, which is what you get if you split a pure minor third into four equal pans; it happens that if you do that you have virtually perfect triads, but no octaves, creating beautiful harmonies and very exotic melodies, because the steps are so strange. Motion from chord to chord is unlike anything you've ever heard, and yet the arrival points are so perfectly in tune that you know it's something very natural to us, for all its wildness, something distant and strange and yet at home and peaceful. The other scale is derived very much the same way, but from a a perfect fourth broken into four equal steps of about 125 cents, and then splitting each of these in half. We have a hard time describing a split-fourth melody because we've never heard that in Western music. But it works! And the track as a whole is kind of a whimsical blending of two different quasi-grotesque ideas in the very best "Ballet Ruse" style.
EM: You actually started dispensing with the octave entirely, then?
WC: After the album was recorded I started exploring tunings in a more analytical fashion, using the Hewlett-Packard 9825 and some programs I wrote to plot the "fit" of different intervals as you change the number or size of equal steps in an octave. Twelve steps in an octave happens to hit the fifth well, as we know. It doesn't do as good a job on the thirds; sixths are a little better. The next good fit occurs as we move on to 15 steps. That one actually misses the fifth, the third, and the minor third by being a little too small, but it is also equally a little too large for the fourths and the sixths-not bad, but kind of equally out all the way around.
Nineteen steps fits the minor third almost exactly, but is less good on the major third and the fifth. After that, it isn't until 31 steps that things start to get interesting: the minor third is not quite perfect, but the rest all group together very high in consonance. That includes the seventh harmonic. There's an oddly near-perfect group near 34 steps, but the next most useful is perhaps at 53 steps-it sits really nicely on a crest-and another one that sits a little less well occurs at 65 steps in the octave. And so on, to an infinite number of steps!
But if we take away the restriction of having only scales which form an octave, if we throw out the octave completely, use the hardware to get our octaves through 16', 8', 4', and so on, and say okay, just within one octave let us make equal step divisions, because equal steps are lovely: they allow you to modulate conveniently and linearly all over the place...whooptedoo! We start finding some really remarkable configurations, including the tunings I used in "Beauty in the Beast." I hadn't plotted and figured these out when I composed that piece, but afterwards I wanted to know why it worked so well, and here it is.
And since this is virgin territory, like Christopher Columbus I hereby christen these three peaks in the plot Carlos Alpha, Beta, and Gamma. Alpha is the temperment that "Beauty in the Beast" is written in, the equal splitting of the minor third. Beta is the interlude that starts it, the divided fourth. Notice that these things, in cents, are simple numbers. Alpha has steps of 78 cents but that's equivalent to something like 15-and-a-third steps in an octave, which makes no sense. How do you put a third of a step in an octave? Build in scales with hiccups? It has to be treated as a special case. But throw out this one unorthodox quality by handling it with the hardware, leaving only the harmonic point of view, and it's a great tuning. With an approach like this, we can get very close to just intonation without any of the problems that prompted people to say "oh,just intonation simply doesn't work." Well, in a practical world, here it is.
EM: What about Gamma? Worked with that one yet?
WC: No, I'm waiting for Stoney to get me a way to play more than 12 notes meaningfully at one time. Gamma has something like 34.5 steps per octave, and arranged on a normal keyboard, even someone with huge hands simply couldn't physically span more than a third. A fourth would be out of the question.
EM: Of course, these things can always be explored with multitrack recording.
WC: For final performances, sure! But for composing and gaining familiarity, it isn't an easy way to work. I'm a composer who very much believes in the Debussy dictum: "do whatever pleases the ear, and the rules be damned." There have been a lot of proposed tuning variations in the last 200 years, different kinds of keyboards and controllers, but nothing has actually changed. The number of people using any of these alternative systems has always been appallingly small. The feedback I'm getting now suggests that a sizable number of people are getting interested in alternative tunings, but will a real majority adopt it? Probably not. There's one very good reason: laziness. In these areas, most musicians you encounter-and most musical theoreticians, for that matter-are just very lazy. It's a human foible we all share. I can tell you that after doing the perfect tunings piece, "Just Imaginings," in which there was a passage of only two measures that took over 12 hours to compose six chords, I'm not so sure 1 won't flee to their side very soon!
In some ways it's preposterous that the difficulty of using these things is so great. But that's the price of admission. You're not going to find anyone looking for a fast dollar coming into this at all. Those who do will run in and run right back out again. On the other hand, with Alpha-and maybe even Beta-and a practical keyboard, you could become an empirical musician. You could just comp on it and start finding things that sound good to your ear, then put them in your MIDI sequencer or what have you, it would actually allow you to write music in these tunings without quite knowing what you're doing, which is how I went about "Beauty in the Beast."
EM: Sequencers would be ideal for just slinging experiments around for editing later.
WC: Even so, I find I like the idea of not having anything stored rigidly, because as I move from tuning to tuning, the way the melody wants to move is different for me. It's less interesting to take some existing tune and move it around from one temperment to another. That might be useful as an exercise, as an etude, as a quick way of getting some results out. But I think that if you're going to really explore the depth of these things, you've got to allow the implications of the particular tuning to steer you. You can't be dogmatic. You can't go in with preconceptions, or you'll just be overlooking the true beauty and power of the particular scale.
EM: What about the next generation of instrument technology? Could it knock down enough barriers to attract a lot of people to new tunings?
WC: It might. If the manufacturers get feedback from people who want to cut with their cutting edges, instead of sloughing. Me, I'm very impatient. I'm discovering how different timbres demand different tunings, such as the Balinese examples. You can put together any kind of sound and hear what sorts of tuning it cries out for (literally). In this arena, tuning and timbre are really kind of the same thing: overlapping and combining over\tones in a pleasing way. It's a very exciting place to be, but also very frustrating, because the support hardware, new keyboards and the rest, are not at all in place and it will require the expenditure of much time and money to get there.
EM: You are on the way, though.
WC: If, as it often seems, everybody else wants to waste these new tools doing diatonic new age equal-tempered tunes and triads, fine. Let them. But before I die I want to find out what lies beyond all these new horizons. And I'm doing it for the best motive in the world: I'm curious.
Freff lives in Brooklyn with three friends, three cats, seven computers, and a recording studio. Aside from drowning in article deadlines, he writes documentation for synths and software, is the American reporter for a BBC show about computers, and is working on various book and record projects.