English

“Conversations” with Kevin

by Barry Schrader

Nobody writes letters any more. The time and temperament it takes to do this seem to have passed along with the dial telephone, S&H green stamps, and eight-track tapes. Of course, like letters, all of these have been “upgraded” to the digital age, and instead of physical letters, we now send email to or, even more cursorily, “text” each other. Some of the romance has been lost in the process.

Still, there are some interesting exchanges that go on through email messaging. I don’t know if, in the future, collections of email correspondence will have the same cache as the letters between, say, Pliny and Trajen, Hamilton and Burr, or Emily Dickinson’s communications with her sister-in-law, but some of them may be of note.

In late 2006, Kevin Austin and I had, at least to us, an interesting exchange of email messages that stemmed from a dispatch posted to the CEC listserver on the subject “Originality in Computer/Electronic Music?”. The reason that I wrote privately to Kevin regarding this was that I didn’t have the time or energy to deal in a public arena with what I suspected would create a long-term brouhaha on the list over what are very contested and contentious ideas. At the end of our discussion, Kevin suggested that what we had written about would make a good article for eContact!, and Jef Chippewa has been trying to get me to write something about this for over a year. I’ve finally capitulated, and, I hope, the following will be of some interest.

The original post of “Originality in Computer/Electronic Music?” was from Mark Piszcek, and dated September 26, 2006. It was actually posted to the CEC listserver by Kevin, as Mark Piszcek wasn’t a member, but he wanted to relate ideas to and ask questions of the electroacoustic community. As is sometimes the case, Kevin wrote a long and detailed response to this message, and as both are available in archives, I’ll not quote them here. But I will quote one part of Kevin’s response that made me take notice:

Er … this is naive. Sorry. In the model I use, there is no such thing as “timber”, as such. Timbre requires the quantization of time. The brain then segments the incoming sound into units (windows) which will be examined and handled either as ‘streaming’ (counterpoint) or ‘integration’ (harmony — sic!).

Now this upset me not a little, as my concept of timbre is very important to me in my work. So I responded to him on September 27:

Kevin,

It seems to me that there is a distinction among the perceptual gestalts that we give referential names to in common parlance (pitch, for instance), the “scientific” or laboratory measurement of this (frequency), and the function of a particular element of this dimension in a given organization.

Although one can comprehend the complicated process of gustatory and olfactory chemoreception that occurs in the identification of tastes and smells, it seems of little value to me to think of what’s happening in these processes in terms of what one smells and/or tastes. Of course this is of some value to chemists who try to fake flavorings, but, I think, it is not important to the chef or the diner. While perhaps being able to be analyzed in terms of what’s happening at a chemical/molecular level, the perception of “salty” qua “salty” is quite sufficient as both nomenclature and remembered perceptual reference.

Similarly, “timbre” serves (for me, at least) to describe that dimension of perceived sound which can be successfully described by the areas of spectrum (which does change with time) and what I call the “event envelope” which is the result of all of the envelopes of all of the partials in a spectrum (envelope being a description of amplitude in time). I do not believe that thinking of timbre in this way is “naive”. If it is, I’ve been happy naively concentrating on both time-variant timbres and (linear) timbral transformations in my music since 1976. The fact that a good number of people have commented on this aspect of my work makes me think that I’m not entirely in a state of self-deception in this regard.

I have the highest respect for science and the scientific method. But to think that it has now presented us with the final “truth” about anything, is, I believe naive. Science is a way for humans to try to overcome their limitations and subjectivity in order to discover and test ideas. There are many things learned through science (or other types of quasi-objective observation) that we take for granted, and many others that we constantly revise. As a baby (I was told by my parents) I was exposed to radiation in the hospital because the scientific belief was that this would create a stronger human being. Instead, thanks to science, I am now quite more likely to contract certain forms of cancers than others who did not have the benefit of a ﬁrst-rate hospital birth in 1945. But no matter what you’re thoughts on science and the scientific method, it makes, I think, a terrible paradigm for music or any other art.

What you’ve said also reminds me of Stockhausen’s ideas from the early 50s where all musical dimensions can be reduced to and quantized as frequency and amplitude. He ignores any differentiation among measurement of data, perception of elements within a dimension, and the function of elements within a given organization. In fact, in his thinking at this time, the latter two are really irrelevant. If you read some of what he wrote around this time, he’s actually a little concerned on how he can apply the concept of serial control of frequency and amplitude that he used in Studie I and Studie II to instrumental composition. Even though he says that his “method” should serve as a model for all electronic composition, we find him 14 years later telling people to “listen to their molecules” and breaking twigs as a musical gesture. This is a long journey predicated upon, perhaps, observing the failure of his previous belief system to sustain continued interesting output. Or, perhaps, it’s the result of a message from his home planet.

If you extrapolate your thinking on timbre to other physiological processes, you might try to be aware of precisely how all areas of perception are handled by the brain and also how all of your other organs are technically performing all of their functions every second of every day (many of which processes are not completely understood as yet). To try to contemplate this on an ongoing basis would, I think, drive one close to madness. It even brings up the old question of who “I” am as an organism: an individual, or a colony that seems to partially (largely?) function without any “thought” or understanding on my part?

So, I believe, to say that, in general, a piano sounds like a piano because of its spectrum and event envelope (explaining what those are and that each piano may be a little different) will suffice for understanding in most areas of the world of music. And, I believe, that people successfully perceive the timbral information that allows them to identify a piano (once learned) as such with or without being cognizant of what the brain is doing in making this identification. And I don’t believe that hearing a single “note” played on the piano is heard as some combination of melody and harmony. It’s perceived as a gestalt, and that gestalt can be broken down into its constituents after the fact. But here we come to the word “believe”.

[Quoting Kevin] “Belief systems, or as those here know from the Windoze / Mac debates, religious systems, are not open to logical discussion, as they are emotional in nature and content. There is no more way to convince me that my beliefs are ‘wrong”’ than I can convince some of my friends to try tofu."

And I must agree with you here. In my experience, most people believe what they do for one or both of two reasons (causes): indoctrination and fear. We are far less rational than we would like to believe. I have found this just as true inside the academy as outside. Once someone has invested themselves in an belief, idea, or point of view, they are very unlikely to be moved from it. In extreme cases, people would rather sacrifice themselves instead of their belief systems because real death seems preferable to ego death.

So where does this leave us? Not far removed, if at all, from where we began. But it does lead into a myriad of needed amplifications and examination of other topics, all of which I’ll eschew for now since I’m really tired and feel (note that word) in danger of making continuously less sense the longer I go on.

Best,
Barry

Now, having known Kevin for many years, I knew better than to simply congratulate myself on my cleverness and not expect a good and witty response. Sure enough later the very same day, Kevin responded:

Barry, thanks. This is clearly worth going to the list.

It is (at root, as I see it), the discussion about ‘core knowledge’ in electroacoustic studies.

Part of my response will be that in university the education of composers is about training “the chef”, not the cook, or the general eater.

Lucasfilm spends $20,000,000 plus developing a sound technique studio, they do not want ‘cooks’ working in there, they want the highest level of ‘sonic chef’.

mp3 was developed by people who understood that “timbre” is psychoacoustic. To understand psychoacoustics, it is necessary to understand the difference between the acoustic (metric) and the perceptual (psychometric).

To develop the mp3 coding and format, it is necessary to have a fundamental grasp of the quantization of time. Can mp3s be created in the analog domain, that is, in the time domain, or is it necessary to move to the frequency domain (with imaginary numbers) and back out?

Last year in a third year class a student brought in a project that compared a number of data compression examples of a short file. He played them for the class, once. The general response was laughter. Almost everyone picked out the original immediately, and then gave more or less detailed critiques of each of the data compression methods, without their being identified.

Sadly, two or three of the students identified most of the data reduction techniques before they were announced. These students have the “golden ears”, and this was in an ordinary classroom situation.

Using a model does not give it a “reality”, and this is the point of this paragraph. The writer here “really believes” that timbre exists outside of perception. His position is “naive”. Sunsets are nice, but they are the result of perceptual and physical forces. One can be the ‘naive’ painter and paints what one sees, or one can be the “???” painter and ‘know’ what one is going to paint.

Unlike Stockhausen, I learned from Hugh Le Caine, that the dimensions are amplitude and time, frequency being a derivative. The example I use is that of a camera taking a picture of a ball being thrown. What direction is it going, and what is its velocity.

I address ‘experts’ and teachers on this list. At the movies the average moviegoer may or may not be interested in how the magic is created (Wizard of Oz…), but part of my point here was that a composer and performer (from the SCI list), with little knowledge or exposure to ea, was putting out the kinds of questions that our classes deal with in the first three weeks of the first year.

Also, there are many ea-m courses taught in the USA that start with Max/MSP, or cSound or some equivalent, and some have The Computer Music Tutorial as the first year text.

You also seem to attribute what I write as to being “what I think” or “what I believe”. The ‘earth-rat’ has the capacity to see, and present both sides of the issue. This (below) is one of my seldom used signatures:

From Dilbert:

Pointy-haired boss: “What’s this I hear about you hating the software integration project?"

Dilbert: “I don’t hate it. I simply mentioned both the pros and cons. People are so conditioned to take sides that a balanced analysis looks to them like hatred."

Pointy-haired boss: “How come you hate it so much??!!

Dilbert: “This is one of those days when it’s hard to be me.”

Thank you Scott Adams

http://www.dilbert.com/comics/dilbert/archive/dilbert-20051031.html

Once minds are made up, the brain is loath to change the perception without really good reason … inertia / hysteresis…

I have proposed a number of models and the models have upset all kinds of people, and then I get blamed. See Dilbert above.

Hang in, but please feel free to post this.
Kevin

Well, I didn’t expect to convince Kevin of anything easily, or, perhaps, at all. I tend to read all of his CEC postings, as I love the way he thinks, even if I don’t agree with him. But I wasn’t yet ready to let this discussion go. And, so, on September 30, I wrote him back:

Kevin,

I appreciate the Dilbert quote, and I quite understand. But if you thought I was addressing my comments on this aspect of human nature to you in particular, you are mistaken. I apologize if, somehow, I seemed to be inferring this. All of us “earth-rats” suffer from being in similar biological and cultural Socratic caves. You are certainly fair and more learned than most. I was, in no way, negatively criticizing you.

[Kevin] “This is clearly worth going to the list."

There are several reasons why I seldom post to the list, one of them being that I usually don’t have time to respond to or even follow threads. It has been three days since you wrote back to me, and this is the first chance that I’ve had to answer you. The reason that I sometimes, as in this case, write to you alone is that I have a great deal of respect for you and your perspective and know that I can not only have an intelligent exchange with you but probably also learn something in the process. I have posted to the list on occasion, when I think it’s worth it for me and/or the people on the list. But putting something like this out and then seemingly ignoring the responses would, I fear, be interpreted by the list as a negative attitude on my part. (I don’t know how you find time to attend to the CEC list as carefully as you do and also find the time to do all of the other things required of you. I know I could never do what you do in this regard.)

I don’t think you and I have any major points of contention. Since I consider music one of the types of works that exists (in perception) in time and deals with what I call “linear kinetic process”, I’m always aware and most considerate of time. Perception is the foundation of my personal compositional theory as well as the basis for much of what I teach in terms of process, procedure, analysis, etc.

But there’s another aspect to the perception of linear kinetic events in memory. Grosvenor Cooper said (and I’m paraphrasing) that to think of a piece of music that you know very well is to hear it all-at-once, vertically. While, in a way, this removes the perception of a work from its existence in time, it is, perhaps, a valuable and useful consideration. Memory also plays a part in listening in time in that perception can change with repeated listening. Babbitt’s Ensembles for Synthesizer is intended as a work of largely non-redundant serialism, but I’ve heard the piece so many times that I’ve got a lot of it memorized, so it becomes quite redundant. But, just as we like to repeat favorite meals over and over throughout life, most people like to listen to favorite works again and again.

[Kevin] “mp3 was developed by people who understood that ‘timbre’ is psychoacoustic. To understand psychoacoustics, it is necessary to understand the difference between the acoustic (metric) and the perceptual (psychometric).”

While I barely know anything about psychometrics, I am led to understand that, even within the field, there is an ongoing controversy about how well a science involving the derivation of standards of testing things which has been widely applied in terms of intelligence and deals with things like factor analysis and data clustering, can be well-applied to aspects that deal with personality or beliefs. To what extent hearing sound (music) is colored by such things as personality or beliefs I’m not sure, but there are individual differences.

But of greater concern to me is the idea that everything is quantizable. While I agree that everything can be purported to be quantizable, I do not accept that any given quantization accurately represents the thing being represented. The field of AI has, indeed, created a range of artificial intelligences, but I think these are not, and may never be, the same as human intelligence. As they become more biological and self-evolving, they may come closer to us, but they might also become (create) a new species. In a similar way, what I call “translational procedures” of composition (as opposed to historically-evolved “relational procedures") such as chance composition, serial composition, and algorithmic composition can certainly create works of music, these works usually lack the important aspects of function and teleology that have characterized most linear-kinetic works prior to the 20th century. For this reason, I believe, most listeners, including myself, find them boring. There is, for me, a great difference between quantity and quality. Change creates information, but not all changes are equal. A work can be quantized after the fact, but, at most, the direct and limited application of this information can only lead to creating an imitation of that work. To some extent, this is what composers using translational procedures have been doing for decades, and, in my opinion, this does not lead to interesting music. But they can spend many hours in a classroom delineating the procedures and related data used to make these works.

Recently on the CEC site someone cited a great benefit in algorithmic composition in that one could discover music that one would never have come up with without using the algorithm. They said nothing about the quality of this music, however. To extrapolate from this, why not just create a master program to generate 1) all possible algorithms for combining music or sound data and/or 2) all possible combinations of all music or sound data (having defined this in terms of the latest scientific understanding of the perception and quantization of said data)? Translational procedures began removing the composer from the compositional process (Cage was quite honest about this) in earnest around 1950 or 1951. The logical extension of this process would be to do away with composers altogether, as I have intimated above. There would then be nothing left for us to do other than sift through the results and make endless analyses of an endless number of “compositions”. Perhaps we could even earnestly apply the scientific method by creating lists of quantifiable properties which we could alchemically translate into aesthetic values, eventually creating an approved list of masterworks.

[Kevin] “Lucasfilm spends $20,000,000 plus developing a sound technique studio, they do not want ‘cooks’ working in there, they want the highest level of ‘sonic chef'.”

It seems evident to me that all of Lucas’s Skywalker enterprises don’t guarantee a good film, even though they might insure a high level of technical quality. There is, in my mind, a great difference between creating tools and creating art. This is something that has always put me at odds with many in the electroacoustic community as I think that a lot of eam works exist more as demonstrations of technology instead of being interesting musical compositions. You and I will probably differ on this, but, I believe, humans don’t hear appreciably differently now than they did 500 years ago. We may have much greater understanding about how we perceive sound (and music) than before, but on a perceptually functional level of awareness, little, if anything, has changed. We simply don’t usually evolve that quickly.

[Kevin] “Part of my response will be that in university the education of composers is about training ‘the chef', not the cook, or the general eater.”

I agree completely. I couldn’t have composed the way I do 100 years ago because both the tools for creating and controlling the material and our understanding of sound and the perception of it didn’t exist. But even though I can now do this, I want to create something that has the possibility of communicating to a listener, and, I hope, not just another “chef”. This, of course, is a personal choice. I consider myself a “chef”, but I’m not interested in cooking only for other “chefs”.

The process, in my mind, is a little Zen, and I often tell this to my students:

Before I knew of Zen, rocks were rocks and trees were trees.
When I first learned of Zen, rocks were no longer rocks and trees were no longer trees.
Now that I know Zen, rocks are rocks and trees are trees.

A problem that I have with the work of many in the field of eam is that they don’t seem to be able to get to that third stage; they get lost in the second, and they continue to make works about the technology they use. This clearly isn’t a problem with a lack of knowledge of or access to technology. Many of these composers are, to put it simply, lost in the technology. (And this is the case not only with eam but instrumental music as well.) In my music I want, insofar as is possible, the technology to be invisible; it’s a means to an end and I couldn’t create what I do without this technology. But I want to communicate to a larger musical audience, if that’s possible, beyond the tiny in-house crowd of experts.

When I was starting out in the business, about 36 years ago, I knew a great deal. As I have aged, learned, thought, and experienced, I continually feel that I “know” less and less. I’m increasingly less likely to insist on my point of view with others; rather I present it, when asked or when appropriate, and let others, including students, decide on how they want to deal with it. But in the music I create, while it’s still evolving, there’s a consistency in perspective and understanding that informs my work. I believe I have a personal style (and this has been commented on by critics and listeners with no prompting from me) that, to some extent, exists because of this perspective. My way of understanding and working with the dimensional materials of sound works for me, and so I do have some vested interest in it and intend to continue using it.

That you “have proposed a number of models” doesn’t upset me in the least. I’m actually interested in these and find that they help me clarify and, if it seems like a good idea, modify my own ideas. So I’ll continue to read, consider, and appreciate the information and ideas that you so generously impart to the CEC crowd.

It’s now 5:23 A.M., I’m quite tired, and I think I’ve said everything I need to about all of this (and more, I’m sure), so I’ll stop here. If you have additional comments about any of this, I’d appreciate your response. If you think we’re at a good ending point on this for the present, that’s fine also. As for the board, I rather think that all of this would make quite a stir, and I’m not sure how welcome that would be.

Best,
Barry

Before going on, I think I need to clarify how I understand what I quoted as a “Zen” saying. “Before I knew of Zen, rocks were rocks and trees were trees.” implies to me that we all know how to deal with our perceptual reality from a combination of practical experience and inherited knowledge (the result of genetic transference and limitations). Assuming we’re born with the ability to hear, we learn what “music” is in terms of the limitations of our perception combined with aural experience. (Actually, I think this is somewhat more complicated, and that we really learn the “meaning” of sound in terms of it being an accompaniment to a viewed action. I call this the “action-response mechanism”, and have written about it elsewhere (“Live/electro-acoustic music: a perspective from history and California,” in Live Electronics (Contemporary Music Review vol. 6, pt. 1), pp 91–106.), so I’ll not go into this in any detail here.) The point is, from whatever are our personal, historical, and geographical experience, we “know” what music is to us at a rather early age, having heard hundreds of works by the time we’re a few years old. So, before seriously studying music theory and composition, we already know what music is. But this is at the gestalt level, a knowledge from practical experience, not from careful, conscious, and detailed understanding. “When I first learned of Zen, rocks were no longer rocks and trees were no longer trees.” is a state that I have experienced with students time and time again. They enter the academy already thinking they know a great deal about music. But when they start to learn theory, compositional technique, and the like, they often become unhappy and confused because their musical world has been seriously complicated. They used to comprehend music only on the gestalt level, and now it’s being torn apart, fragmented, and broken down into its constituent parts. Many students don’t like this and they resist the process. What is hoped for is that, having taken things apart and learned to understand, master, and control the various elements of musical dimensional information, they will be able, with the instructors’ help, to put things back together so that they can reach the state of “Now that I know Zen, rocks are rocks and trees are trees.” This doesn’t always happen.

On October 4, 2006, Kevin replied, making succinct comments on some of my points:

Barry, over my head too… but

The term psychometrics has been highjacked and politicized. When speaking to a knowledgeable group I mention this, but have lately found that this meaning is best know among people who were paying attention in the 80s and 90s.

In a basic way, psychometrics is statistical in nature … the question: “Is it warm?” is not a ‘measurement’ of a physical type, but an evaluation done by people.

[Barry] “While I barely know anything about psychometrics, I am led to understand that, even within the field, there is an ongoing controversy about how well a science involving the derivation of standards of testing things which has been widely applied in terms of intelligence and deals with things like factor analysis and data clustering, can be well-applied to aspects that deal with personality or beliefs. To what extent hearing sound (music) is colored by such things as personality or beliefs I’m not sure, but there are individual differences. But of greater concern to me is the idea that everything is quantizable.”

This is simply a model.

[Barry] “While I agree that everything can be purported to be quantizable, I do not accept that any give quantization accurately represents the thing being represented.”

Alfred Korzybski… map and territory.

[Barry] “… The logical extension of this process would be to do away with composers altogether, as I have intimated above. There would then be nothing left for us to do other than sift through the results and make endless analyses of an endless number of ‘compositions.’”

This is detailed in Stanislaw Lem’s Cyberiad stories.

[Barry] “… as I think that a lot of eam works exist more as demonstrations of technology instead of being interesting musical compositions.”

I think Mannheim School here.

[Barry] “The process, in my mind, is a little Zen, and I often tell this to my students:

This is a version of the Cage story in Silence. It is one of the first I tell my basic theory class. Mine also has clouds, and another ending.

“But what is the difference?”

“My feet are no longer on the ground.”

And that is the end of this “conversation”. As usual, Kevin got the last word.

So did we solve important problems and definitively answer crucial questions? Will this exchange become of historic note? Probably not, although there are some interesting ideas discussed. I think Kevin and I probably raised more questions than we answered, but, then, that’s one of the important functions of enlightened discourse.

eContact!

eC!

Social top

“Conversations” with Kevin

Other Articles by and about the Author

Print

Online

Social bottom