LeafMusic & Computers Magazine,
1997 Conversation with Wendy Carlos

(a special online full text transcript)
Interview by Jim Aikin
From the Nov-Dec 1997 Music & Computers issue, originally posted at: <www.music-and-computers.com>
This reposting of the original full transcript is being made available exactly ten years after the original. It's interesting
to compare and note the many changes which have occurred during one decade of this rather quickly evolving medium.

Music & Computers: Nov/Dec '97: Features

[Ed. Note: Because space was limited in the Nov/Dec '97 issue of Music & Computers, large portions of our conversation with Wendy Carlos had to be omitted. What follows is a full transcript of the interview, edited only to remove false starts and clarify points that might be obscure. For more information on Wendy Carlos and her music, be sure to visit Wendy's Web site.]

The close links between music and technology are nothing new. Nineteenth-century advances in metalworking, for instance, gave violinists the brilliant tone of wire-wrapped strings. More recently, the rapid changes brought about by electronics have opened up far broader creative vistas. And just as the violin needed a Paganini to show the world what it could do, just as the piano needed a Liszt to sweep away the cobwebs of harpsichord technique, electronic music technology needs visionaries. People like Wendy Carlos.

Carlos has been using computers in her studio since the early '80s. Her upcoming CD, Tales of Heaven & Hell, was recorded on a pair of Macintoshes using Mark of the Unicorn's Digital Performer sequencing/audio recording software. But her love affair with technology goes back much further. In 1968, synthesizers were found in only a few universities, where electronic music was an arcane specialty taught only to graduate students. Carlos changed all that, almost single-handedly: Her landmark album Switched-On Bach, recorded entirely on an analog modular synth built for her by Robert Moog, became the first classical disc in history to sell platinum (a million units). SOB, as she affectionately refers to it, was the recording that first introduced the public to the sound of the synthesizer.

Imitators quickly jumped on the bandwagon. But most of them were trying to capitalize on a gimmick; few had any grasp of the extraordinary level of musical vision and sheer painstaking craft that had gone into SOB. Today, Carlos brings the same meticulous care to projects that use far more advanced technology. One of her passions is alternative tuning systems, a field that has only become practical to explore with the advent of digital instruments. Another is additive synthesis. Her more recent albums, including Digital Moonscapes (1984) and Beauty in the Beast (1986) have explored the phantasmagorical world of electronic orchestration.

JA: I'm excited to hear that you've got a new album coming out. Tell me about it.

WC: A couple of years ago, some people I know from Seconds magazine come over. I had known them because when the last major project I had done on Telarc, the Bach 2000, had come out, one of them had done an interview, and I like the fellow. He and his associate wanted me to be involved -- if I would -- with doing something for people in their twenties, because they felt that there's not very much intelligent music being done for Generation Xers, and could I do something that wasn't classical, wasn't Baroque, wasn't rock, wasn't jazz, wasn't any of those things, but sort of had little elements of the kinds of things that the younger audience was listening to, and yet still be myself? And I'm afraid I disappointed them, because this was going to be somewhat more of a techno album.

But the idea of making something that was based loosely on Clockwork Orange [the Stanley Kubrick film for which Carlos did the score] had come around, and it just didn't work with something paced fast with an "up" beat, because Clockwork Orange is very slow music, and themes, and ideas, and it's kind of a dark, somber affair, and a lot of the [new] album is that way. It's very melodramatic, dark, over-the-top, scary stuff. The line I've been using with a lot of people who've asked me over the last two years about it is that my nephews, who are in their mid-twenties, got nightmares on listening to the pieces of it that I let them hear. It was kind of a chuckle, because I didn't expect that. I did it more or less like Stephen King, who is basically a fairly accessible, easy-to-take guy who writes these ridiculously convoluted, scary things with all sorts of unsavory things going on.

Well, I'm obviously not a Satanist or a devil-worshiper; I don't believe in that kind of crud, but, at the same time, I know how to pull the strings, I think, to make a little bit of mellerdrammer [melodrama], so it was a lot of fun. And in doing it, I decided that it was really such a dark thing, there were so many things dealing with the afterlife -- there's even a piece I added called "Afterlife," and a very beautiful thing at the end called "Seraphim," which is more heavenly, and [has] a few rather sweet themes -- that the whole thing fits very well with the title that I've given it, Tales of Heaven and Hell. It's a great big, fat thing. It's even longer than Beauty in the Beast, which for new music is unusual; most of the time, long albums are compilations or older albums. It's seldom that creative, new music is done in a very long form, because it's too much work, too much for the audience to take in all at one time. But this is an hour-long project.

You're one of the people who I kept thinking about, because you had given me the most lucid, intelligent review I'd ever gotten in my life on Beauty in the Beast when that came out, and this one has an awful lot things similar to it. Some of the strange tunings are in there, and some of the same kinds of timbral klangfarbenmelodie [according to the Harvard Dictionary of Music, "a succession of tones (even if with only a single pitch) treated as a structure analogous to a melody, which is a succession of pitches"] are in there.

It really, in the end, didn't go down to the "trying to get the twenty-year-olds at their own level," but there are things that they maybe will enjoy listening to while staying steadfastly [with] my own kind of cerebral, eclectic, intellectual kind of music. So it's something I'm really tickled pink about. The whole project, it's got all of the kinds of stuff that I've tried to put in projects in the past which, when I hear many years later and haven't heard them in the interim, I listen to and say, "Wow, I did much better than I remember doing." That's always a nice feeling, and it's something that I've experienced enough times now to make me feel that my obsessions are justified in the end, even if it's not true.

JA: You said you were searching for a label. Do you have any news on that?

WC: Frankly, the project took so long, mostly because of the way I insisted on doing it. Because of all of the complexities of the way it was put together, it couldn't really have been done prior to having DAWs [digital audio workstations] with built-in sequencing and all that stuff, the very thing that's epitomized by [Mark of the Unicorn] Digital Performer, which is what I use.

Unfortunately, on my old [Macintosh] FX, which is what I started the project on, with the big files that these things were turning into, with all the complexity that was going on, the newest versions [of Performer] were totally unstable. I had to go back to an earlier version, which was moderately stable, but wouldn't allow more than four minutes of music with all of these sounds and everything before the files would self-detonate. So I had to do the thing in little bitty chunks, and then try and link them together later on, and every time I tried to link them together, the new file would become unstable, and it would crash everything around it. Anyway, it became the worst software nightmare I have ever experienced. Plus, it was a fairly complicated hardware thing. It was truly a MIDI hell project, the worst I've ever encountered. And I've been at this for a long time, but this is the worst I can think of technology biting back, and almost driving me off from being able to finish. But in the end, I got some help from some friends who gave me a lot of equipment and stuff; we finally got it up and working.

It finally got finished, and by that time some of the contacts they had made with a few companies -- and I don't want to mention names right now -- had sort of lost interest for the time being. They thought the project would never happen, and a few other companies had become interested, and so I'm kind of in the junction of not knowing who to go with right now. We've just begun talking with a few people.

I'm not sure how fast the business negotiations are going to be on this one. It's not something I did knowing which company would take it, as in the past had been the case until the one that Telarc released. They saw either your or Dominic's mention [M&C editorial director Dominic Milano] in the magazine, and contacted me about the Bach 2000. I did not go after them. In this case, I haven't mentioned it to anyone. No one knows about this project, really. And so no one has contacted me except through word of mouth, and from that I'm getting quite a few interested people. But I'm also a little bit bored with the usual thing that happens with artists and companies in the publishing business of records, and so my businesspartner and lawyer is going to make sure that, unless we have a really decent deal, we're not going to sign. Even if it meant we had to wait a little longer than we would normally be tempted to do, we'll do that. So I'm in no hurry, except for the fact that I'm kind of bursting to get reactions from people, but I'm in no hurry to get a deal cut unless it's a moderately good deal. I'm not looking for anything gluttonously greedy, I'm just looking for something reasonable.

JA: I think that's what all artists want, and unfortunately, it can be hard to achieve.

WC: It's almost impossible to achieve.

JA: This is off the subject, but have you read a book called Who Killed Classical Music?

WC: But that's of course what this album is not. I didn't even bother bringing it to Telarc, although we have sort of an open-door agreement to bring some projects to them if we wish to, but I don't believe Telarc or any other classical division would know what to do with this new project, because it ain't classical music at all. It's barely even crossover, it's just something that... It's its own thing. It's probably more closely related to cinema music, except it isn't a soundtrack.

It's been completed for a couple of months, although I'm doing the final mastering only right now. If the company will agree, I've even done the cover artwork for it myself, which is a take-off on the Clockwork Orange original artwork. It's posted on my Web site, so you can take a look at it there if you wish.

JA: Let's talk more about the production process of the album, You said you started it on an FX, and you had problems. Did you later switch to a Power Mac?

WC: That's what I'm using now to finish the project, and will probably continue. It's an 8500. Unfortunately, for some reason, either the machine itself is a lemon, or there's some instability with the system I've got in there, because the latest Digital Performer upgrades have come with a FreeMIDI version that will not work in the machine. I don't know what it is, and they claim that they've got 8500s there that it has worked with, so I don't know what that means. I've got [System] 7.6, and I'm going to get the 7.6.1 updater, and I'm thinking of putting it in, but there have been Finder glitches and instabilities in the machine, losing icons and losing aliases, which is troublesome, even though the machine seems stable while it's within Digital Performer 1.7 or whatever it is I'm using right now, 1.8. Anyway, it at least it's stable within that realm, and I still use the FX interactively. In fact, I always have liked two computers in a studio at one time because they can do different tasks, and it's handier than even the best multitask single machine.

JA: What two tasks would you be using two computers for at the same time, typically?

WC: Well, sometimes I might be using [Coda] Finale, because it's still my notation program of choice, even though Performer does a pretty reasonable job of printing out music, and I like to sketch out and get printed-out copies of what I'm doing, so that I can then figure out where the piece is going from there, and with alternative tunings and that, you get into such difficulties in trying to keep track.

I don't write simple music. I know I don't, and I cannot just do head arrangements, which is what most so-called electronic composers do. Many of them don't read or write music anyway, so they don't have a need to or don't have any choice in the matter of using notation. But notation is a very fine tool [even] if you don't write "eye music," as the Germans used to call it. What is it, augenmusik? Anyway, if you don't do that to yourself, if you don't do what all those dull 20th Century composers did when they adopted Schoenbergian methods and used to write stuff that would bewilder the eye, but which sounded like s-h-i. . . .

Anyway, if you use notation as the proper tool that it is, it's very handy. Even in multi-tonal alternative tuning musics, where the notes don't necessarily correspond in sound with what you see on the paper, it's good to have something so you can put together the structure of the piece. Otherwise you're always wallowing, and it's just much harder and slower to try and give up that tool. So I like to have that going. Also, it's sometimes very stable to run audio on one machine and MIDI on the other, or some parts of the audio on one and some parts on the other. The machines behave better, and you get more channels multiphonically, and it's easier to have alternative places where you lock them down, rather than trying to do it all within one machine, where, yeah, the task can sort of be done, but it's clumsy, With two machines running at the same time -- and they all can lock to SMPTE, and they can all lock to house sync, and all of that -- that works for me very well. So there are a lot of little ways I work that way that I'm glad to have it. It's just another approach to doing things.

JA: For this album, you were using a combination of [MIDI] sequenced stuff and audio tracks? Is that right?

WC: Exactly. It's what I think everybody's doing. I can't brag about it, because it's the first time I've gone extensively into it, but I've been fooling around with this since 1991. I've now spent a lot of years so that I can really make mostly any sound that I wish to make. There are still things that I find more expedient to use Kurzweil sampling for, although I've never liked samplers in the past because of their lack of expression, which is something I spoke of in Bob Doerschuk's article [for Keyboard]. Do you remember the piece that was done about two or three summers ago? I mentioned that some of the newer, sample-based synthesizers are at least now becoming quite expressive, so you're not just getting that "dant-dant-dant-dant-dant-dint-dant-dant-durnt-durnt-durnt," you know. It's just terribly dull: It's always the same sound. Now I have a lot of tools within the V.A.S.T. business [Kurzweil's synthesis architecture] that let me actually start plastically manipulating sounds and making them respond to what I'm doing with the controllers, and with the notes and velocities and all of that. So I find that textural sounds which are going to be moved around and pitched differently, and layered rhythmically, those are best run off of things like Kurzweils.

JA: Do you use a K2000, or a K2500?

WC: I've got two 2000s. I prefer having 32 MIDI channels, because I do a lot of multitimbral stuff, and one 2500 would definitely be not enough. So I opted for the two 2000s since there's very little difference between the two machines that I can see. I think I can find three things that are different, and one of them is not the sound quality. Anyway, be that as it may, other people might have different experiences, but for myself, I prefer having two 2000s. They're both loaded machines, they have a lot of RAM in them, and they have a lot of sounds, which are mostly musique concrete things that I've built here myself since years and years ago with Rachel Elkind. I used to use concrete a great deal. So I'm getting back to using concrete, which is what sampling seems to allow once again, and then a lot of real synthesis. So it's like those two worlds have been joined together. The world of the audio on hard disk is more concrete, because then you're playing back textural things that are so long that you don't really want to put them in a sampler. What's the point of having something that runs for two minutes be on a note in the sampler? That seems like a foolish place to put it.

JA: Plus, you have to sit there while it loads every time you do a session.

WC: That's right. So I'd rather use the memory for great big, long, fat samples of single-event notes and the like, and then the things that are probably only going to have to appear once, or maybe at most twice in a piece of music, put those in hard disk audio. You can tweak that to a fare-thee-well, and it seems like the right tool for the job. Also, of course, you can bounce them back and forth. You can take sections of a long audio file, poke it into the sampler, play with it there, and then sample out what you've now played with it back into a hard disk file, and then that can run in sync with the new things which you're adding.

The piece de resistance is that this new project, for instance, has a great deal of people singing, including some Gothic-type touches with chanting, almost like a priest intoning some kind of a requiem mass, and that is just stupid to handle in any kind of sequencer. While I have a delicious number of vocoder parts I put into the piece, vocoder does not sound like a real human being singing or chanting. So for those things, hard disk is again the right way to go.

For all the reasons that I think most of your readership will be aware of, I'm having a great deal of fun reaping the benefits of this hybrid, mature technology. Well, I guess "mature" is... It's not stable mature, but it's at least conceptually mature. This is a very, very decent way to make music, having all these kinds of tools linked together so you can do all of the personal, expressive stuff that we can do now with an elaborate MIDI setup, and with responsive synths get that to play back in a way that is every bit as alive and natural as a live ensemble would be, using sounds that go everywhere from the wildest electronic, "electronic-y"-sounding things, through things which are actually acoustically replicate-sounding, or concrete things that are somewhere in the cracks, or hybrids of three of the above put together, and then layer that in with the live performance parts, which are better handled with a multitrack audio metaphor in the DAW. And it's all synchronizable; you can change your mind, you can vary the tempos. Digital Performer allows all kinds of time and tempo-warping anyway, and pitch-warping. It's good stuff. They're really powerful tools. If it only weren't so damn slow, but the scalpel is very sharp, and it allows you to cut very precise layers and put together a piece of music that can be very plastic, and very alive, and have all the stuff that I'm interested in. So I'm really quite pleased by all this. I like the trend, I like the direction things are going in. The only thing now I'm chafing to see is a really big additive [synthesis] engine put into one of these machines so that the final, last missing link can be included.

up(Top of the Page)

JA: Speaking of additive synthesis, are you still using your old Synergies for that, or have they been retired?

WC: No, they're still... They're not on this current project, but I've learned in the past that Stravinsky was right when he said when somebody asked him about composing so many ballets, if he didn't find that restrictive? His comment was, and I'm paraphrasing, "No, I don't mind at all. I like exact specifications." In other words, an arbitrary set of boundaries, within any form of art, can be an asset, not a disadvantage. So forcing yourself to do something as a stunt, like a painter deciding, "This will only be done using red, green, and black pigments, or white and black and red/green pigments, and that's it," or a composer deciding to do a monochrome work for only string orchestra. No woodwinds, no percussion. Those kinds of limitations are very, very good, and it isn't even like. . . . Who is it who wrote, what was it? I can't even think of the name of the book. It was the book that had no "e" in it.

JA: It was a French avant-garde thing in the '30s. I forget who it was.

WC: That's more a stunt, obviously. That's too drastic a step to make really be workable. But that type of thing has worked for me in the past. So for this project, I decided I would make available only the sounds of the two Kurzweils and the digital audio workstation. Well of course, in the end, I couldn't keep to that promise. There are places where the 150s just simply have much better sounds for things like pianos, and certain kinds of harps, and certain additive voices that are built in it. So I started using that, and then I remembered some of the voices I had built for my [Yamaha] SY77, which were semi-additive and semi-FM'd. I started putting some of that in, but then I stopped. I decided I wouldn't use any of the Synergies or the Slave 32s, and I wouldn't use the [Yamaha] TX802, That was where I drew the line on this particular project. And it's just arbitrary, but the Synergies are still working fine, and sometimes the 150s are the better choice for additive anyway.

JA: What is the 150?

WC: It was when Kurzweil first came out with the MIDIboard, about 1986 I guess it was. The first box [tone generator] that they had was a fairly large rack affair -- might have been ten inches tall -- called the 150 FS. Then they made smaller versions which were seven inches tall with the green displays, and it was an actual additive engine, and it had one of the best-sounding pianos. The piano was built [i.e. programmed] so that it changed timbre on different dynamic levels, gradually, from one to the next, and none of the sample-based pianos do that very well. You compare the two, and it would only take only a one-minute demonstration to you in the studio. You'd say, "Oh yeah, well there's no argument that the 150 is just better even though it's an old, old machine and doesn't go above 10K [an upper frequency limit of 10kHz]."

So there are things like that that make me be aware that we should have that ability within our next generation of big synths. You ought to be able to take a concrete sound, do an analysis of it, and come up with a quick resynthesis additive patch. It won't be perfect, and you'll go in and tweak it, you will aim it in the way you want it. Then you'll start massaging it and turning it into what you're going to turn it into past that, and maybe merge it with qualities from other sounds, and do what would really be a sound morph, which is something I did back in the Digital Moonscapes, when some sounds turn like from a xylophone, into a clarinet, into a cello, into a trumpet. They actually morphed. The overtones actually realigned and changed dynamic level and envelope, which is very different from crossfading between two different sounds, which is what people claim to be morphing now in all the ads in the magazines.

JA: Well, actually, we now have real morphing going on in this new Yamaha analog-modeling synth, the AN1x. It's not crossfading, but it's morphing an old-style analog voice.

WC: Well, I don't want to go back to that. People keep asking about that. It's something easy to do. In fact, the Kurzweils are unusually adept at making analog-type patches, and while it's slower and more clumsy than working on a Moog, once you get it, you can store it, and use it as a library of sounds. We did a bunch of that when we did the Bach at the Beacon concert in April, which was the live performance that I got involved with eight synthesizer players. We wanted to capture a lot of the sounds of the original performances on the original albums, including the Bach 2000. So we made a lot of analog-type patches, and they work just fine. On stage, they were great. So that family of sounds is still available, and that's nice to know. It's just a little clumsier to work with on an interface that isn't really designed for that task.

JA: Speaking of additive, have you had a chance to look at the Kawai K5000?

WC: No. I think another friend of mine mentioned it also. I have been approached by most of the people who have done additive. I've certainly got a lot of opinions, and actually quite a bit of knowledge and dusty bits of business and heuristics that I'd be happy to get involved with companies who are doing something that's truly the next step, but most of the things I have seen were sadly much worse than the Synergies and the [Crumar] GDS had been, and that seemed to me to be stupid. Why take two or three steps backwards? So I just wasn't interested in getting involved with that. I'm old enough now that I feel the fleeting passage of the years, and if it's going to be that kind of a game, then I'd rather just say, "All right, we don't have the ability to do that now, and I'll just be happy working within limitations that are not my own, but are the ones I have to deal with in current electro-acoustic music."

JA: You brought up being able to do the pitch-warp and time-stretch in Digital Performer, and I'm wondering, because you're obviously extremely sensitive to timbral things, whether the little glitches and artifacts that can be created by that process bother you, or if you have particular tricks for getting around them. How do you deal with that?

WC: You try them, and if they don't work, well, even if the idea was good, you don't use it. It's that simple. Sometimes also, I've found that I can take things that have glitches, bring them into [Digidesign] Sound Designer, and find little spots, and do little crossfades, and hide them. And then you end up with a new sound which really was flawed, but like having scratches on a photograph that you scan and put into [Adobe] Photoshop, you've removed the flaws artificially, tediously, by hand, and it takes forever, but that's the way you do it if it will work.

Even with the old vocoders, there were places where I had things on the new project that I just in the past would've allowed to go on the record, but they bothered me: little places where the transition between non-voiced and voiced [timbral regions] aren't really the way that a vocal chord setup in a human being would be. And in this case, on this project, I didn't have to put up with that. I just came into Sound Designer and grabbed regions, and did crossfades, and went inside and massaged little areas, and put little EQ changes, and little gain changes, and did smoothing, and with a pencil tool tried to write out any little glitches and artifacts that bothered me. It's very tedious. It's very time-consuming, but you can get it quite smooth, and then that's what you hear on the final record, because it's then coming from the hard disk audio, and so it's been cleaned up.

In fact, I tend to clean my things rather meticulously now. Everyone's talking about 20-bit audio, but I don't think many people have really squeezed very much out of 16 yet, and it seems to me that 20-bit would be nice for the masters, for people especially who record live orchestra and live piano and ensembles that are unpredictable in their dynamic range, and maybe actually exceed 96dB of original dynamic range. For those cases, I think it'd be nice to have extra bits to play with, but as long as the final result is going to be 16 bits, I don't care what they say. Yes, you can put the noise in the higher bits more with what they call noise-shaping. But it's a real tradeoff. I don't think the public is made aware of what's really going on there, by and large, and this is a case where knowing where the bodies are buried, I try to do the job rather deliberately.

JA: So you're getting the most you can out of 16 bits.

WC: Yes. And I'm squeezing it in a lot of ways that I don't really want to point to every dead body and where I put some perfume around the corpse so you wouldn't smell it, or I just made a hologram of another corpse so that you have the illusion it's there, but it was really fairly fetid and I decided to toss it. So, anyway, with that kind of a metaphor description, it's possible, if you take the time to learn what all these things can do, to use the tools like a surgeon to do plastic surgery on your music and make it fairly optimized. So I kind of like that thing now. I feel like I almost massage every molecule at the end.

JA: Would that involve noise reduction, EQ, compression, all the above?

WC: All of the above! I don't use automatic compression very much at all. I'll usually go and tweak regions and peaks and stuff by hand. Because sometimes a limiter will just take off all of the peaks. I'll go and find the ones I don't really think would mind coming down a dB or two, then I can enlarge that entire spot so I can give a real "oomph" to an attack, and then with several stepped regions selected, all using smoothing, do a very small decibel change on each to form a ramp, and then it becomes a very gentle, little easing in to feather touch near the top of the all ones of 16-bit [i.e., full-code wave peaks] so that I can give the maximum sound output at that moment for some sounds, without sacrificing anything else in the neighborhood, and having no artifacts, nothing that's pumping, nothing that's doing the things that invariably happen when you use any automatic machine. So I'm really in favor of throwing away most automatic processes, and taking on the tools in my hands raw, and using them very carefully, one at a time. I like that way of working. Some of the stuff I picked up from Larry Fast, because that's the way he's done a lot of stuff, very carefully, tweaking things one step at a time, using things like Sound Designer, and it's a very powerful way to work if you are patient enough with it.

JA: When you look back on the whole process of creating this album on a computer, or on multiple computers, what was the most the most difficult part of it for you?

WC: Putting up with the equipment and with the slowness. Putting up with the fact that all these little tools that I talk about are slower than you can think of them, so you're always fighting [user] interfaces, you're always fighting that you can never get enough speed, and fighting the fact that the more complex we build our edifice, the more unstable it does get to be. The one feature that nobody seems to advertise anything on the basis of is stability. I'd almost prefer to have more primitive tools that are rock-solid stable than all the bells and whistles.

JA: My motto is, "If this stuff was any more powerful, it wouldn't work at all."

WC: That's right. I like that. So, that's what I'm always fighting. Conceptually, then, you get very... the old word is deracinated, which I think means "uprooted." You're always in right hemisphere mode working and composing and thinking it through, and losing track of time, and being really involved with it, and then something goes down. Or the next day you go start working, and things just aren't where they had been the previous night, and you wonder, "Where did all the changes come in?" Something's crashed, something's burned out, something you've got to go and troubleshoot again for.

JA: Right. So it's intermittently pulling you out of the creative process. That brings up a larger question, which is whether the creative process itself is different because you're using computers than it would be if you were using a pen and paper and then hiring an orchestra to play it. Do computers really create a different creative space, or do they simply allow you to do the same things in a different way?

WC: That's the kind of a question, unfortunately, which gets so top of the head, I'm not certain that I'd be saying something that has any validity at all. I mean, the centipede with a hundred legs doesn't really know how it moves all the legs, and when it tries to think of them all, it can no longer walk. I don't really know how composition works. I've told people my verbalizations in the past, how you feel like themes and the like are filtering through your mind. You're aware of being an editor, you're aware of saying, "No, not that. That's not good enough." You're aware of throwing things out that you've spent days, weeks, even months doing, because something better just pops into your head. You're aware of that, you're aware of the time you've spent, you're aware of every little bit in the music at the end, that it all went through you, but the process itself is somehow still almost magical. That's just the act of being creative. I've done a little bit of creative writing, I've done a little bit of creative artwork, I've done a lot of messing around with Photoshop. . . . It's not music composition per se, and even good performances have a lot of that same thing, too, where all of the stuff that you've forgotten becomes your wisdom, and the wisdom then becomes intuition, which is not to say that it's not valid and real, it's just not verbally accessible.

In the case of the tools of computerdom, obviously we can make music that has a different medium, or different media -- multimedia, too, and all. So the canvas and the colors are different, but the steps of making great music are still the same steps. Form is still the biggest bugaboo in any piece of music longer than a minute or two. Anyone who hasn't attacked anything but doing a song -- be it an instrumental, or most of them of course have lyrics, but songs seem to be the main output of music in this country right now -- they don't know what I'm talking about. But the minute you try and get into a longer, abstract piece of music, even if it's program or film music, you're facing formal questions: How do the structures relate to one another? Does this lead to the other logically? Does it lead to the other logically enough that it's going to sound self-conscious when you hear it more than a few times, or does it flow in a way that seems more instinctive? Do themes come in and never repeat that almost seem like they're gratuitous, and therefore like you threw in too many ideas instead of working them through? Or, did you work them through enough, but the things that tie them together don't really quite work, so the bridges are bad? Or, are the bridges evident, but the material sort of dull, and so you've got a polished job on mediocre material? Or, do you have some great things here, but you ought to pull it apart into shorter pieces because it's really not making it as a large structure? Or, is it exactly the opposite problem?

Questions like that have never changed in all the years I've been writing music, and from reading biographies and interviews, and talking with other composers, and other people in other creative arts, no difference. I don't think the tools of computers change it at all, it's just for the disturbing one that, because they're more brittle, you're more likely to find that that stream of consciousness will be interrupted constantly, making you sometimes wonder why you don't just take the whole shmear, open the window, and drop it down to the sidewalk below. There were many times on any project that I've ever worked on in the last many years that I've wanted to do just that.

JA: Did you ever want to do that to your old modular Moog?

WC: Yes. Oh, God, that was a nightmare, too. In fact in some ways, it was a worse nightmare, because you didn't have the wide variety of sounds. It was much more limiting. I feel now I almost can get any sounds that I. . . . I mean, it's such a large family now, I don't feel the sting of my old teachers telling me that if you're interested in composing for timbre, stay with the orchestra, but if you're interested in composing new, odd things, you can stay with electronics, but it will not give you the range of timbre that you can get with an orchestra. That's no longer true. So, for that reason, no, I'm not unhappy using this medium, and I don't feel necessarily that I would rather use a pencil and paper for an orchestra. But for some themes and ideas that come along, it would be nice to have a live orchestra waiting to add its contribution to the electronic elements. If I were a wealthy composer, that's what I would probably try to do. But when you come to alternative tunings and the like, and I use a lot of those, I'm not sure that the hassles of dealing with musicians who can't cope with anything but the equal-tempered scale wouldn't kick that off.

up(Top of the Page)

 JA: A few years ago, in an interview, you were talking about how you had started with an original Macintosh. Was it a Plus?

WC: No, it was the 128. That was our first Mac. My bizpartner got one for us right at the beginning of the first year they were available, 1984, I think in March, we were using one. So, we saw the first interface, when in order to copy a floppy, it used some of the screen RAM, and so the screen filled up with black and white dots while you made a copy of the floppy. I mean, that is really going way, way back. And then that machine got added onto. It became the 512, and the 512KE, and then a Plus, and then it was turned into a Prodigy Prime. And then I updated to a IIx, then to a IIfx, and now to the 8500.

JA: So you've been --

WC: I've been there. Before that, there were the CP/M machines I used with the GDS and the Synergy.

JA: Those were Kaypros?

WC: Like Kaypros. Bigger versions of machines using Z-80s [microprocessors], which were a little more business strength than personal, home hobbyist-style computers.

JA: And they were used mainly for controlling the Synergies with a GDS-type operating system?

WC: Yes. That's right. In fact, you're overlooking the fact that every one of these synths, since the Moog, is a computer also. The Synergies are computers, and then you would also use another computer to voice that computer. The GDS is actually two computers, one of which is controlling the sounds, and the other of which is controlling the interface and the parameter changes. That's actually the way a lot of the current machines work. So V.A.S.T. is really two different chips. It used to be Calvin and Hobbes, and no, it's Calvin and Janus, and they each are computers that are dedicated to steps necessary for doing the sounds, and then you've got an overall 68K, or 68030, or 040, or even I think now they're going to be coming with Power PCs that handle the interface, and the MIDI Ins and Outs, and allocating the hardware resources.

So we're working with multiprocessor environments. A good MIDI interface box like a [Mark of the Unicorn] MIDI Time Piece, that's a computer. And then Jim Cooper's various little [JL Cooper] devices, those are each computers: The FaderMaster, which is damn helpful to have, and all of the merging boxes, and all of that. Each of those has its own little computer, and sometimes they will go down. Sometimes the little simple passive [MIDI] merging box that you're using for putting together a lot of controllers to squeeze the most out of what you're playing, that'll go down, and it won't affect anything else except that the notes that you're trying to input suddenly are either blocking up, or being polluted in some funny way. It's hard to find when things go wrong where the problem is, because there are so many places it can happen. Sometimes it's just a patch cord, or a dirty spot on a fader, or a jack, or something like that.

JA: I've had some very strange malfunctions. I haven't even bothered to track it down yet, because it hasn't been an urgent thing, but one of my synthesizers occasionally drops a note. It just forgets to play it.

WC: Oh, that easily can happen. Yeah, yeah, yeah. You'd have to look and see what's really going on at that moment. If it's in the sequencing, you can repeat the situation, but I find when I've had those dropped events, if I play it back another time, I hear it fine, and it drives me bonkers.

JA: We could talk about malfunctions all day.

WC: Oh yeah! That's not the focus of the article, but at least you could make it plain, especially for people who think that this is a quick, cheap ticket to success and fame and fortune and all that, which it's none of, honestly. In fact, it's almost exactly the opposite direction, especially at this stage, where it's highly competitive. Like with the profession of law, too many people have gotten into the field right now. They should be discouraged, really, because there's nothing to be gained except if you're on a hobbyist level, and you enjoy it for your own pleasure, and you don't mind these problems that we were talking about, that's fine, but the truth is that it's a very onerous, difficult way to make music. At least what I describe as being music. The only way around it is to give in to the technology, and just go with what's easy, which is what new age music is about. It's doing what happens to be lamely easy to do with the current synthesizer technology, or was in the '80s. You know, it's all the cheap shots, and cheap shots does not make great art.

JA: There's another whole level below that. Technology is being oversold to consumers as, "Now you can make professional-sounding music, even if you've never had a lesson!"

WC: Oh, dear. There was a magazine, I think it was called Compute, which was dealing with, I think, Apple IIe's and the like and -- or was it the Apple Plus? Anyway, somebody was quoted as saying that there were going to be so many more Beethovens now, thanks to the new technology. And I looked at that, very puzzled. It's wanton lies! That is not true. Art comes from creative people doing a job well. Some part of it is a gift, in which case you can take no bows for it, other parts it's just pure sweat work, in which case you can take a bow for it. In any event, human beings are not decided on their value in life by whether or not they can put together a good string of notes as a piece of music. Some of us do it well, some of us don't do it well. And you can still enjoy the act of music and do it for your own recreation, without being in any way a great composer. Why should you take on that onus anyway? Would you think you have to be an Olympic champion able to run a mile in under 2-1/2 minutes? Is that really the only way you define yourself as being an acceptable human being? And why shouldn't there be people who can do it a little better than others?

Why is it when you talk this way about the arts, and particularly music, it sounds elitist to some folks? I don't understand that. I've spent all my life in this field, and I think I can do it better than most people; not all people, but most people, and it's because I've taken where I think my strengths lie, and have invested a great deal of my life in doing it. And that should be the reward one gets, in the same way that the Olympic champion I mentioned earlier has put in a great deal of their young lives into doing something exceedingly well that they were gifted in from childhood. And that's all it's about. It's no more than that.

JA: I think maybe in America we have both a fascination with excellence and a sort of pseudo-egalitarian urge to level it, to claim that we are just as good.

WC: Yes. So we have to handicap as though... There's a little short story that I read just two weeks ago, when I was sick, and I pulled out this old collection of interesting speculative fiction stories, and it was in a world in which the beautiful had to wear funny noses like clowns wear. . . .

JA: I think that's a Kurt Vonnegut story.

WC: Yes, you're right, it is. And the bright people had to have the thing in their ear that made loud noises [so they couldn't concentrate], and people who could run fast had to wear heavy sash weights, and people who were musically skilled would have to have cotton, so they couldn't hear the sounds properly, so they couldn't play well, or put rubber bands around their fingers so they couldn't play well.

I think if you make the analogy with the Olympic athletes, which to me, is something spectacular, that takes my breath away -- I wish, I wish, I wish I could compete in that, but I know I'm a klutz, and cannot do those things any more than I could fly. But I can offer the world my small contributions in this little area. But it's fine if you want to try doing music also, in the same way that I think we should all try to do a little running for our health, and the like. It's something that should be recreational. But your ad that you mentioned that got us going on this topic, about now anyone can make great music even if you've never -- that's a lie. Would you be able to sell an exercise machine and regime and health pills and the like under the notion that you now could beat the Olympic world record?

JA: I don't think anybody'd believe it.

WC: No. But they do in this case. What you should say is that the technology allows even people who'd never trained in music to investigate and make and learn to do music on their own, and by their time, and sweat, and natural talents, they will achieve or not achieve varying degrees of success.

JA: And technology certainly makes it easier for people like Brian Eno, who perhaps had no conventional music training, to achieve some wonderful things that would not have been available to us otherwise.

WC: Yes. Right. The only thing is, don't be surprised that some of his material sounds like it's lacking performance value, because he doesn't know how to put in that aspect unless he wished to hire some performers to do that part for him. I have felt very bad about the fact that a great deal of electro-acoustic music seems to come from a mindset that champions getting rid of the performer, whereas in truth, the technology is slowly coming around to giving us the great control that a great performer needs to put the notes and the sounds and the events together in a fluid, graceful, elegant manner, and that is not what a composer does. The composer says what to do, but not how to get from point, to point, to point.

JA: That creates more of a burden, perhaps, on the electronic artist, because it's not enough to be a great composer; you also have to become a great performer -- in non-real time, but you now have to wear those two hats.

WC: You have to wear the two hats. And you should, if you're going to do an electro-acoustic thing, become a great orchestrator and instrument-builder as well, too. Otherwise you'll have to just write to the companies and buy their pre-packaged sounds and samples and CD-ROMs and all of that -- which is fine, except then you're using canned art, in the same way that people who design brochures and little business publishing notes might use clip art, because they don't know how to draw. Unfortunately, a great deal of Photoshop work nowadays seems to be being done by people who actually can't draw. And you can see it. You can see immediately that they fall into algorithmic things, using far too much of Kai's Power Tools in a way that it's letting the tool control them instead of them controlling the final result, and that is not what the hardware is at its best about, although it's nice that it can do that as well.

I'm all for democratization. One time, years ago, in some way I was misquoted out of context, and I got a lot of shit in the mail because of it, because I was sounding elitist, and, well, I mean, I'm elitist to the extent that I said about the Olympic people, but in the real way, I love the democratization that the technology has produced. I think making all fields open to a great deal of the public is an extremely important service.

JA: And that's something that the computer has made possible for amateur musicians.

WC: Absolutely right.

JA: You were talking about people getting canned clip art music off of CD-ROMs, and that made me curious: What are your sources of musique concrete materials?

WC: Usually just from myself, what I set up microphones to record here.

JA: So you will drop a set of keys, or clank the bottom of a sauce pan, or all of that stuff?

WC: Exactly. Not so much just stunts. I know what I'm looking for, and I'll try and come up with things that can give me the raw material that can be useful in building that thing.

JA: Can you give me an example?

WC: Well, yeah, there was a sound that over at Tom O'Horgan's. He has a replica of Benjamin Franklin's glass harmonica, which is done with a lot of nested bowls. There's a little trough in front that you wet your fingers, and then there's a foot treadle you rock. The bowls are on a shaft which runs from left to right, with the smallest nested bowl to the right. Almost like a bell tree, they get bigger and bigger as they go the left. And so, like a keyboard, you kind of run your fingers on it, and there are color codes, in the same way that harp strings show where the C string and the F string are so you know where to pluck. The bowl rims are colored, I think, red and blue, to show where the C and the G are, or something like that. You can actually sit down and play music, more or less, but it's clumsy. Anyway, I liked the sound of it, and so I tried to do that with my finger on a crystal goblet, and I discovered that I could get a much better sound with a cello bow with only a small amount of rosin on the edge of it, but it was hard to control, and I had to do many [recording] sessions before I got several beautiful-sounding notes.

I used different-sounding goblets, and I put them all together to form one great big keymap, which has this wonderful sort of glass harmonica sound. After finishing that, I discovered that that sound could be convoluted and metamorphosed into a lot of wonderful other concrete sounds, which just came about.... Well, now having done the work for the glass harmonica, what else can we get out of this? And that was the same with the sound that I had wanted to get which was sort of a cembalom, the Hungarian mallet-style [instrument], you know, the Eastern European style of strings that are played with a little hammer? I tried to do that on my autoharp, and I got some lovely sounds out of that, but it turned out if we removed the attack, and changed the envelope, and went in with the V.A.S.T. and started messing with the actual sound itself, there are some wonderful sounds I was able to get out of that that were unusually good for microtonality, because the overtone structure was so terribly organized, and similar to the simple overtone scale. There are things like this that just happen all the time.

In fact, Tom has been nice enough to also let me go over to his place, and since he has probably every instrument that ever existed in any country of the world, there are quite a few little sounds that we were able to get that I was able to massage and put into this project, and will be working with for many years to come, that have turned into some lovely, rich sounds which take full advantage of what concrete should be about, what Pierre Henry and Pierre Shaeffer were about back in the '50s in France. It's something that can be done again now, and so I'm trying to do that as well. It's just I wish that that could be combined with additive, as I keep saying. Somehow, it seems to me there's a missing link that's been absent notably here. All that should be tied together. And I hope before I cease to have a productive career and care about these things, that some degree of a technology step can be made in that direction so that it'll be fun to work with that. But in the meantime, it's still possible to be a hybrid with all of these technologies that maybe don't tie together in a single engine, but you can simulate some of that same thing, just by having access to all of them at the same time.

up(Top of the Page)

JA: Is there any chance that there's going to be a Wendy Carlos sound CD-ROM for the Kurzweil K2000?

WC: Oh, I don't know. I've talked with a fellow from Ilio [Entertainments] named Mark Hiskey. I wanted to use some genuine, old-fashioned, fat string sounds, and he had sent me one of their string volumes, and I found that while the original sounds were well-recorded, the programs, as they call them, were aimed for a rock audience, and so were, to my ears, terribly different-sounding from a string orchestra, from anything I recognized. So I took many months to massage them and build my own sounds, and in the end got them to the point where I could play along with some of my fat English orchestra recordings, where they have these beautiful-sounding strings, and I could play in and insinuate myself into their performance, and not be able to tell by ear, "Was it real or was it Memorex?"

When I got to that level, I stopped, because I now have a good string orchestra, with soloist strings and the like, that I could use on some of the tracks of this album that is quite plastic and quite amenable to doing, you know, a certain limited range of what a string orchestra could do rather nicely. It's slow and difficult to play it properly, but it works very well, and it's a source of sounds that I've been hoping to have, and it's clearly using sampling as sampling, but it's combined with so much in the V.A.S.T. technology to plastically allow the expressive use of these sounds, that you really can go anywhere from a real loure to a real marcato, to a real detache, to playing with a lot of wiriness that a real ensemble would use, and just the usual string machine things that you and I associate with a lot of years of pretty awful-sounding devices that would almost be like putting rosin in your ears. It would just. . . . Real? There'd be no doubt what you were hearing. Anyway, the line there is getting a little foggier, although certainly it'd be nice, as I said earlier, to have access to a full live ensemble that could combine with this kind of thing as well, so that you could have a complete continuum.

JA: But you don't have any plans to bring anything out through Ilio that contains any of your own sounds, or programs, or anything like that?

WC: We talked about that, and they were thinking of it. It's up to me to take the next step, and it's a lot of time and work, and until I get a few of these recording projects out of my door, I'm not going to face it. But I am thinking of doing it, because I have a lot of sounds here, and I think it would be kind of a marketable item. And they sounded very excited about it. And they would help me in any way necessary with massaging the sounds, or collecting them and collating -- a lot of the bookkeeping work, which is terribly tedious to do well, and they do a remarkably good job, so, they're a fine company.

JA: We'll keep our fingers crossed. Before we wrap this up, we need to talk about tuning.

WC: Okay. You're going to write a whole article on it [for M&C], right?

JA: I'm hoping to. I want to try to introduce people to the idea of alternate tunings, starting with the basics: "If you don't know how to listen for beats, here's how to listen for beats."

WC: Ooh. What a good thing to start off with!

JA: And then talk about the ratios of the harmonics series, and how to do the math of ratios.

WC: Please include some small line, even if it's only a throwaway, just for my purist thing. Every article always makes it sound as though the just ratios are inherent, period. No. The just ratios work for sounds which have simple harmonic numbers for their overtones. Even the piano is not tuned by pure octaves, because it does not have simple harmonics. They're stretched, especially in the higher and the lower notes, and so what you're doing with inventing tunings is finding ways -- as I tried to speak of in that Computer Music Journal article from ten years ago -- finding ways to have simultaneous timbres merge smoothly. That's all. And simultaneous timbres, when they have simple harmonic overtones, will tend to form ratios that are simple harmonic ratios that we call just. But the minute that they're not that kind of a timbre, as indeed most physical instruments are not, at least in some ranges, then you're not going to want to use just. You're going to want to use whatever it is! The best example, still, to me is that a gamelan in Java or Bali is not tuned to anything like just, or equal-tempered. It's tuned to what sounds good on those instruments, which is a fairly odd, funny set of scales called slendro and pelog. They do it by ear, which is what everyone should be doing! But it is hard to tune accurately by ear. You have to kind of learn how to do that. It's a skill. JA: Can you remember the first time that you encountered alternative tunings?

WC: I was playing with my parents' piano back when I was an adolescent -- I don't know exactly what age. First I thought, "Well, I'll go to the music store and get a tuning thing, because I bet I could tune a piano," and then I discovered, "Oh, my God! This is hard!" And I went and got the books from the library, and really then was worried, because you've learned that it's even harder than you think it is. I actually managed a few times to tune it fairly well, but in doing it, I had remembered some of the chapters of Helmholtz's On the Sensations of Tone, which spawned a whole lot of people like me into getting an interest in alternative tunings. And so I started playing around with just, and with meantone, and with some extended oddball scales, and it was slow and tedious to do on a piano, but I played with a lot of them. For three or four days I'd leave it in one tuning, then the next week I'd try some other tuning, and some other tuning.

And finally, back in the '70s, Rachel [Elkind] found a used Novachord, which is something made by the Hammond Organ Company in the late '30s. That had 12 oscillators with octave dividers, and you could turn a simple brass knob and change each of the 12 notes extremely easily. It was an electric device, very easy to tune, very nice rich harmonics, sawtoothy waves, and that was a good source of my experience with fooling with alternative tunings. Boy, did I learn fast what works and what doesn't work there.

JA: Do you still have the Novachord?

WC: No. When I moved here, it went into storage with one of Rachel's people, where they were storing some of her equipment and furniture and stuff, and I believe it was later sold, which is a damn shame. I didn't have room for it, though, here, very honestly. It's a big, big cabinet. It's the size of a kitchen table, or a credenza or something. It's a big heavy affair, and by then I had a set of oscillators that Bob Moog built on the Moog [modular synth] for me which let me do much of the same thing. And finally, I was getting involved with some of the earlier digital machines like the Synergies, and by 1985 already I could do complete computer-controlled tuning, with exact precision, on the Synergies. And that wiped away any of the need to want to use these unstable analog methods of tuning. Really, until digital synths -- and digital only because of the precision of the quartz clock -- microtuning of any kind was a pain in the you-know-what. Now, it's gotten rather easy.

JA: I've tuned my Korg 01/W several times to different scales, and I've noticed that although they give you a nominal plus or minus 50 -- in other words, one-cent resolution for each of the twelve notes of the scale -- in fact, the way the circuitry works, you aren't really getting one-cent resolution. You're getting something like 1-2/3 cent resolution. [Ed. Note: The interval of one cent is 1/100 of an equal-tempered half-step.]

WC: Yeah, but that shouldn't be of any concern at all. The coarseness of the steps on the Kurzweil 1000 series, those were as much as 5, 6, even 7 or 8 cents. That's too coarse a step to be too useful. But the 150 had very tiny resolution, I mean smaller than one cent. The Synergies are about 1-1/2 to 2 cents per step, which is fine, but not super-fine. On the K2000 and K2500, they're less than one-cent steps. They're not exactly a cent, but it doesn't matter, because what your goal is is to get within roughly two or three cents, unless you're holding the chord very long and don't want even a very slow throb, then you maybe want to get down to one-cent accuracy. But we're talking about [retuning] intervals that were way off, like the major third is more than 14 cents off in the equal-tempered scale, and the harmonic seventh is more than 33 cents -- that's a whole third of a semitone off. So, for those big steps, even if you don't get down better within one or two cents, if you can get better than 33, that's making a big improvement.

And for some of the sounds and some of the timbres that you're going to be using them on, you simply will not be aware -- it will not be palpable less than about 3 cents or so; you won't know. So fooling around with the fifth on equal-tempered is a waste of time, because that one's very close to being a pure thing; it's only two cents off, a little better than that. But fooling around with thirds, and with sevenths, and with seconds, and with elevenths, and thirteenths, that's where you'll get the juice. And of course there are a lot of scales that don't even begin to use anything like ours, like the Indonesian scales, where slendro, for example, is more like 240 cents per step. Now that's such an odd scale, or my own little alpha, beta, and gamma scales, that you really just have to get within the ballpark of them, and you'll immediately find that there's something interesting happening.

So don't worry about the accuracy; you're going to go for something where, you know, you've never had red paint in your life, and now suddenly you're given a pretty good color of red. It might not be the ultimate color of red, but don't sweat it, it's a good color of red. That's how I feel about it, anyway.

JA: What scales did you use on the album? Is there any way to describe them?

WC: Oh, sure. There's nothing particularly exotic about them. For music that forms a moderately conventional-sounding Western harmonic progression, I still prefer using some version of meantone, or the Werkmeister scales. And so I've used quite a bit of those on this album, because the music is done with singers, so it's something I wanted to be similar to what traditional musicians play, and the meantone and the Werkmeister scales very nicely get around the difficulties that just intonation has with the chord on the second, and with the modulations that go around a circle, or go through a I-VI-IV . . . I-VI-II-IV . . . I-IV-II-V7-I, and suspended sevenths, and. . . . There are a lot of problems with the just scales in certain Western-style harmonic progressions, and these are gotten around very nicely with meantone and with Werkmeister, so it works. Also, I have never done a piece in 15-note tuning, which Easley Blackwood had done a lot of, and since I like a lot of his stuff, I thought I would try that, and so some of the music is in 15-note tuning.

JA: Fifteen equal-tempered?

WC: Yeah. Didn't you use that one once? No, you used 19.

JA: I've used 19, and I've experimented a little with 17, which I quite like.

WC: Yeah, well, you might enjoy 15 also. Fifteen equal -- it's not as smooth-sounding to me, it's not like a just tuning, but it's a fascinating tuning to work with. It's also very close to my own invention, the alpha tuning, except the alpha tuning is almost just; 15 is not. In fact, 15 is quite rough. But it is a cycle, so that you can go around through the octave and come back on the other side, which alpha cannot do. [Ed. Note: Carlos's alpha tuning is an equal temperament, discovered through a computer search, that has between 15 and 16 steps per octave.]

JA: You've got five steps in every major third, rather than four.

WC: That's right. There are also some random tunings that I've invented just to try and find, by ear, intervals the way the Indonesian musicians might find them, where that pitch for that timbre is kind of in a little niche that's just about right off of equal and off of any expectation to be the most provocative next step I could have. So where I'm really trying to flavor for spice, and I don't care about the theory, I'm caring only about what the aural effect is, and so there I've used pitch-bends on individual notes within the sequence where they're all monophonic notes per track, and I can bend a note in any direction I want, and so that's what I've done. That's a slow way of working, but it's extremely open-ended and flexible, and you can come out with some very arresting, beautiful-sounding things that way.

JA: Do you have any documentation on your alpha, beta, and gamma scales?

WC: You can find some stuff on the Web. The big article is not there yet; Matthew Davidson said he was thinking of putting that up fairly soon now. If you have access to a library and can get the Computer Music Journal from exactly ten years ago, it would have been the spring or summer issue that had a lot of articles on tuning -- I'm in that one. And there were also a couple of articles in Pitch magazine, which Johnny Reinhard publishes with the American Festival of Microtonal Music [318 E. 70th St., Ste. 5FW, New York, NY 10021; 212-517-3550, fax 212-517-5495], that speak specifically of alpha, beta, and gamma. I can't find those right now myself. I've been looking around, and Johnny said maybe he could find a copy of them. But they're covered in some depth also in... Kurzweil has that manual. I think that's online. I think on my Web site [www.apocalypse.org/~wendy] I have the Kurzweil [tuning] manual. You can get that and download it and print it. For instruments that have only twelve generic pitches -- not the individual, tunable notes like Yamaha does -- you can use that article to fool around with some of the stuff that I've done, and it goes through the history of some of these things too.

up(Top of the Page)

JA: Since a lot of the readers of Music & Computers may never have even considered alternate tunings, how would you explain the attraction or the benefit of alternate tunings to a naive listener? You talked about the sounds smoothly blending into one another because their overtones mesh. Is that really the essence of it, or are there other aspects of it that somebody should be aware of and listen for?

WC: It's a little bit like you've only eaten at a fast food restaurant all of your life, and you're suddenly being told that the whole world of haute cuisine exists, and that many countries -- Northern Italy, France, certainly, and Asian countries, like Thai food -- that there are all these values of food that you've never sampled, but they're different, and they're much more subtle than what you've had to do to your tongue and taste buds and expectations in having to live on a diet of fast food only, where the subtleties are all totally removed by the temperature of cooking.

So you've got to kind of get back in touch with the real world of hearing, and not just say, "Ooh, that's different! That's not my fast food place!" You've got to to kind of let yourself let go of those things, and start listening to them first. There will be a tendency for you to have the same effect that you have when you first try ethnic food, and so like your first piece of sushi, it's strange. But at the same time, these things are based on very real acoustic values, and they're better than anything you've gotten used to in the past, so given time, if you listen closely enough, you will become aware that something subtle is occurring that you never have experienced before in music. Some of it can be extremely hard to hear, so that in some records, you're really never going to be aware in a double-blind test whether you're hearing something that different from equal temperament or not. Others will be smoother and nicer-sounding in some ways, and you'll say, "Oh, yeah, I can hear it right away."

But because of the tunings, the musician or composer will have stopped doing things like certain types of modulations and the like, so it'll almost have a new style of music that you'll hear, which came about because of the tuning, in the same way that Western music tends to be pretty busy about moving its notes from note, to note, to note all the time, because none of the notes really sound that great, and so we've evolved a certain kind of music and an expectation that's based on trying to hide the inadequacies of our scale. That actually, historically, occurred just exactly that way. So you'll be able to investigate kinds of music which don't need to do that sort of stuff. That's fascinating, too.

Finally, there are some types of music expression that you just couldn't have imagined at all in the past that are every bit as strange as hearing your first electronic music if you've only heard acoustic instruments in your life, and suddenly you hear your first real synthesizer stuff. It'll be just like that; it's something that you've never heard before. So it's not necessarily going to be something you want to stay with, but you really owe it to yourself to sample it first, and see if this special thing, which many people have grown extremely fond of, might not be an acquired taste that you want to acquire.

JA: Is there any software that you use currently for setting up tunings, or are you still doing it all from the front panel?

WC: The [Opcode] Galaxy editors have a pretty good one for the [Yamaha] TX802/DX7II machines. That's probably the best computer tuning software I've seen. It's not possible for the [Yamaha] SY77, though. You have to use the tuning in the 802 mode, then port it in to the module that supports the 77, and download it into the 77, which is less than interactive. Yet it works fine. I mean, it ports it right in there. But the SY77's tuning table editor is not bad. The Kurzweil's quite easy to use. The SY77 is harder to use, but they tune every note differently, so it's doing a bigger task than just changing the 12 base notes. As you know, if you've gone into most of the ones that have 12 base notes, you've got like a little keyboard, and each keyboard note has like a zero if it's tuned to equal temperament, and a plus or minus in value by how many cents it's deviating from the equal temperament. Well, there's only twelve numbers you can put in there, and that's it. Whereas Yamaha lets you put in a different number for every single note of the scale, which is the efficient way to do really wonderful scales, but it's a lot of work, so I don't know how to get around that one.

I've used spreadsheets. I find spreadsheets very handy because you can put in ratios and have them propagate all the way down there, so that you can find the numbers in Yamaha tuning units and in cents, and it's real easy to do that way. I'm not going to teach you how here, it's not the point of your article, but it's something that anyone curious enough and who knows what these things all mean, they can get into a spreadsheet and spit out values and frequency in Hertz, based on A-440, or C-261 or 264, whichever you're using, and they can spit out Yamaha tuning units, or even the tuning units that other machines use, like the Synergies are all different from that. I think they use hexadecimal, in fact. Anyway, you can make all those things happen on a spreadsheet, and it's a very efficient way to generate a lot of tables, but in the end you still have to sit down with the machine and type it all in, and that's extremely tedious. Galaxy's editor will help you as long as you're using instruments that are based on the Yamaha or similar modules.

I believe you can also, though, take those modules and strip the Sys-Ex parts that don't apply to a particular synthesizer, and download it into a different synthesizer. I do recall doing that with a lot of tables I built for my TX802 that I wanted to then play in my SY77. I just ported them through Digital Performer, and I just yanked the hex code portions, the header and the suffix, or appended on the new ones. I don't remember. But it was just an editing job in Sys-Ex. And then you took the table and put it right in.

JA: I can't imagine that would work if it wasn't two Yamaha units. If you were trying to send it to something else, like an Ensoniq, I can't imagine if it would work.

WC: You'd have to try. I think some other machines use the same assumptions Yamaha did.

JA: Well, supposedly there's a MIDI tuning standard now, but I don't know whether anybody's actually using it yet.

WC: Well, some people suggested that the [MIDI] polyphonic aftertouch business should be used for microtuning. Each note then has an additional byte associated with how it's going to be pitch-bent, and that would work just fine. For the moment, you can simulate that by breaking your sequences into single monophonic lines, like I've been doing, and put regular Pitch-Bend on each one.

JA: Right, and then you have to use your spreadsheet to figure out how deep the Pitch-Bend should be, and then use algorithmic software to add the right amount of Pitch-Bend to all the D#s.

WC: That's right. But once you've done the homework, once you have the values built, they sit there for you. I mean, all of my homework now does me very well, because I can go in and bap out almost anything I need, so it's worth doing the homework.

JA: Do you have any upcoming projects, or anything that has been of passionate interest to you lately that you wanted to mention to people?

WC: Well, I did have the American Festival of Microtonal Music perform the "Afterlife" piece from the new album that is not yet out, and they did a live version with many parts being done live onstage, and the rest coming from a DAT tape, which was specially made with the parts that were not to be done live, and that worked very, very well. In fact, it's a nice way to do an ensemble piece which is complicated, so you can't have everybody onstage, but want to have some of them onstage. And some of the Bach at the Beacon concert, which we did in Bach's authentic tunings, I was aware that we can probably put on a concert and may do that soon for some of the new stuff on the album, so I expect that we might get involved doing some live performance stuff of the alternative tuned things, doing a lot of the stuff just live.

There are a lot of options suddenly becoming open, and I'm finding some other musicians who seem to be eager to work with me on doing this stuff. It's kind of exciting from that point of view. So even though it only would be a very wealthy reader of the magazine who could afford to hire an orchestra and train them to do all kinds of strange tunings, and play along with their MIDI sequences -- I doubt if very many people at all would be able to do that -- nonetheless, there are some other options coming along now that most of us should be able to do the same kind of thing on.

I'm going to do a Tales of Heaven and Hell Part II. I mean, I don't know what I'll call it, but I really want to tap the vein of this extremely dramatic-based, picture-painting-style music -- tone poems. I have a lot of ideas that I've already started working on for the next project, but we're going to try to get the first one out as soon as we can, and then I will get the next one out, probably much more rapidly now that the new machine seems to be working. I can probably get the next album done without quite the same pain that the first one took.

And then I don't know where I'm going to go from there, but I think there'll always be an interest in alternative tunings in my music in the future for the rest of my career, now that it's become so easy to do. I still will do some music in equal temperament, and some in variations of it, like meantone, but there will also be music...You know, it's like, "What way do you want to eat tonight? Do you want Turkish food, do you want Japanese, do you want Thai, do you want French, do you want Northern Italian, do you want Ethiopian...?" There are so many ways to go, and we have the stuff to do it now, that it would take a rather cowardly, un-curious person not to want to fool with it.

JA: One of the things that interests me about alternate tunings is that when you're playing even three simultaneous notes, the harmonic implications of your new chord progressions --

WC: Are very big. Very big.

JA: I've been studying jazz piano lately, and using some quite thick voicings. Just playing in the equal-tempered 12-note-per-octave scale, you can get some interesting stuff, but then you think, "God, this was all done by Dave Brubeck 30 years ago." Yet when you tune to an alternate scale, all of a sudden you have something that's completely fresh harmonically, and yet it doesn't require these seven-note chord voicings.

WC: That's right. My harmonic scale, which is one of the ones that I spoke about in some of the things I've mentioned to you, is one that would be great for improvised jazz, because, like "You Say, I Say" and "That's Just It," the pieces from Beauty in the Beast that you recall from years ago, they used tall chords that were very much like a Brubeck would have used, but within the realm of a harmonic scale, where you have the overtones going all the way up to 13, 15, 17... You can't do what the equal-tempered scale does, but instead are forced to do things that are very, very different, and very, very provocative. And it implies a new type of jazz harmony which would allow beautiful melodies to be rolled on the top of [it], as the best of jazz has always done. Yeah, it's so ticklish. It leaves me a little giddy with delight at the possibilities.

JA: Is there any chance we might be able to make some of your Kurzweil files available to M&C readers?

WC: Well, I am making online versions of a lot of my files, music, graphics, and text, on my website, and those resources continue to grow, for those who are interested. But to get into any of these things, they're all consuming of time, and time is the one commodity that we have no control over in our lives, and I lost a great quantity of that just in trying to get into doing sounds, and music, and this merger with the digital audio workstation and Digital Performer. It was an extremely high price of admission, which I hope those who come after me will no longer have to pay, but I had to pay that thing, and it's cost me many albums' worth. I mean, it's the main reason that nothing's happened, plus the topic that we didn't get into, which is a few things that I had done after that last Telarc album with Larry Fast, trying to do sound things for motion pictures... It's a whole project that I probably was very ill-advised to try, that lost me a couple of years as well. Things like that make me very cautious on giving my word for anything right now, unless I can get a staff of people to help me with it, to facilitate the fact that these things, none of them, come in less than several days, sometimes several weeks of work. Even just to do one crummy file sometimes can be a whole day that you're polishing it.

So I will try my best, Jim. I'm certainly not trying to hide anything from anyone. But, really, this is just me, this is just my little world of sound. We have a set of tools here now that lets anyone who's curious, who's got some degree of intelligence, who's got some degree of taste and love of timbre, you can explore a lot of these same things just as I have. And you shouldn't be just waiting on me like I'm the only person doing this; it's not true anymore. There's no excuse...It would be like having the only source of clip art be from one particular artist who has to sit there in front of a computer for hours and hours drawing things. That's not true.

So don't bite that lie any more than you're willing to bite the lie that the advertisers have been trying to put off onto people with their products. The truth is actually already good enough that you don't need to lie. It's not even that without a lie, we'd have no business; that ain't the case. Just telling the truth, you'd have plenty of business, because the truth is exciting already. I'm very giddy about this stuff. I certainly lost long ago the feeling that was drifting on me in the late '70s and the early '80s, that this whole field was just sort of worn out, and there wasn't very much in the way of room to go anywhere. That's no longer true.

JA: No, it's a whole new ball game.

WC: So you don't need me to point you in the direction... Well, I've pointed in some directions here. I mean, go and explore it. It's there. It's like the Pathfinder's up there on Mars. Go and do something now! Jeepers! You don't need me to point the way. But it's nice talking about it with you. You're one of the few writers who I know knows a lot about this and has actually gotten your hand into the field, and I'd be damn interested in what your [tuning] article's going to be like. The approach that you've given me already sounds just like what some young, inexperienced people need: Someone to just make the first step as painless as possible. I'm sure you're going to do that very well, Jim.

JA: Well, I hope some people will pick up on it. It's never going to be a... Well, at least not in our lifetime, it's never going to be a mass movement.

WC: No. Never. Never will be. Although Johnny Reinhard tells me... It's amazing. In all of the time over the last five or ten years that he and I and a few others have been talking about it, we kept saying, "Gee, it's too bad there aren't a lot of other people. Gee, it's too bad there aren't a lot of microtonalists. Gee, it's too bad there aren't a lot of young composers trying this." He says, "You know what's happened?" He says they're suddenly getting a lot of good young composers doing a lot of interesting work, and of course, being that they're just starting, the formalities of their pieces are a little shaky, and some other parts of the structure... You know, it's sort of like teething time, but they're working with the field. They're getting into it, and suddenly it's hopeful that in half a generation's time, you know, like 15 years from now, it's going to be a very different ball game. And we won't be having conversations like this. In fact, I hope like with my attempt to try and get people to investigate the Dvorak keyboard for typing, there are a lot of intellectually bright but lazy people like me who've picked up on that, because once you've now discovered you can't hire people to do your typing for you, and you've got to do it yourself, it's like, "Well, I know I can type, yeah, but why not do it the easy way and get rid of the carpal stress syndrome?" You know, why not make it easy?

JA: So you use the Dvorak keyboard layout?

WC: All the time. The article I typed for Computer Music Journal was my learning. That was where I began to type using it. I still can type QWERTY, but it's a little slower than it used to be. But on Dvorak, I go so fast now, I make the kind of glitches that you get in keyboard playing of fast music, where your fingers tumble so that they go in the wrong order sometimes, so you get a strange kind of typos that you start making from going too fast, and that just means that the technology isn't really in your way any longer. You're now having to confront that physical things don't move as fast as thought.


Frequently Asked Questions

[Ed. Note: Below are explanations of some of the concepts discussed in the Wendy Carlos interview.]

What are alternative tunings?

The conventional 12-note scale found on both the piano and synthesizers is called equal-tempered, because the octave is divided into 12 steps that are all of equal size. Many other types of tuning are used in other cultures, and were used in European music in earlier centuries. In many historical tunings, the 12 notes of the scale are not spaced equally. As a result, some chords sound relatively pure, while others sound very harsh. Modern tunings sometimes provide more than 12 notes within the octave, and these may or may not be spaced evenly.

What is additive synthesis?

There are many ways of creating sound electronically. Conventional synthesizers, for instance, use so-called PCM or wavetable synthesis, a method in which short digital recordings (samples) of actual sounds are played back by the synthesizer. In additive synthesis, a complex sound -- whether realistic or other-worldly -- is built up by adding together various harmonics. Each harmonic is a simple sine wave, and each has its own frequency (musical pitch) and amplitude envelope (loudness contour). Wendy Carlos began using additive synthesis in the early '80s with a digital keyboard called the Synergy, which was programmed with the aid of a computer. Additive synthesis has never become a commercial success because it's more expensive to implement than PCM synthesis, and is harder for the musician to program. With the right hardware/software tools, however, and with patience and a great sensitivity to sound, the results can be far more expressive and flexible than PCM synthesis.

What is musique concrete?

With the advent of magnetic tape in the late 1940s, it became possible for composers to use actual acoustic sounds as an element in their scores. This technique was pioneered in Paris by Pierre Schaeffer and Pierre Henry, hence the term musique concrete (literally, "concrete music"). Early musique concrete pieces were created by cutting and splicing dozens or hundreds of short pieces of tape. Today, the same results can be achieved far more easily in a computer, with the added advantages of greater sonic fidelity and the possibility of altering the recorded sound using DSP (digital signal processing).

© 1997 Miller Freeman, Inc.
© 2007 Serendip LLC. No images, text, graphics or design
may be reproduced without permission. All Rights Reserved.

up(Top of the Page)
Back Back to the Wendy Carlos Home Page

Wendy Carlos, Jim Aikin '97 Inteview
© 1997-2008 Serendip LLC. No images, text, graphics or design
may be reproduced without permission. All Rights Reserved.