An Interview with Gottfried Michael Koenig

Gottfried Michael Koenig; C. Roads

Computer Music Journal, Vol. 2, No. 3. (Dec., 1978), pp. 11-15+29.

Stable URL:
http://links.jstor.org/sici?sici=0148-9267%28197812%292%3A3%3C11%3AAIWGMK%3E2.0.CO%3B2-%23

Computer Music Journal is currently published by The MIT Press.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/mitpress.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Wed Dec 5 19:16:13 2007
 Interview with

Gottfried Michael Koenig

by C. Roads
0 1978

Introduction sionally. After my studies in the conservatory 1 con-

sidered myself a composer until 1 nlet electronic
Gottfried Michael Koenig (b. 1926, Magdeburg, music-which really meant to study anew.
Germany) is one of the directors of the Institute of Sonology,
Utrecht State University, in Utrecht, Holland. CBR When did you first become involved in electronic
Koenig's work as a composer includes instrumental as music?
well as electro -acoustic compositions. From 1954 to 1964
he worked at WDR, the Cologne Radio station. Among other GMK It was about in 195 1, 1952 when the first night pro-
projects there, he assisted Karlheinz Stockhausen in the grams were broadcast by the Cologne radio station.
construction of his Kontakte (1960).
Mr. Koenig's compositions available on record include CBR When did you first start to work with electronic music?
his Terminus (1966167) and his Funktion Griin (1967) which
features computer-generated control voltages regulating GMK I went to Cologne in 1953 and started working in the
analog equipment. Both compositions are found on DGG Cologne electronic music studio one year later in 1954.
13701 1 (the Avant Garde Series). In addition, his Funktion
Gelb (1967168) is available on a small (7") disk from Wergo: CBR Is it not true that you also studied computer pro-
WER 324. gramming at one time?
Most recently, his concentrated h u n g filr Klavier
(1969) was performed at the UNESCO Workshop on Com- GMK Yes, that was about ten years later: 1963 and 1964.
puter Music in Aarhus, Denmark. Until then there was no need. There was a large field
The Institute of Sonology is a center for contemporary of electronic music; computers were barely known;
electro-acoustic and computer music. Distributed among its and when computers came into my field of view. it
five studios are several computers, including small-scale took some time before I met people who really made
(LSI- 11) and medium-scale (PDP- 15) cpu's, as well as analog me aware of the possibilities. And that was at the
equipment. The Institute has its own technical department and beginning of the sixties. I met somebody from the
staff. University at Bonn and started studying computer
This interview took place in Utrecht at the Institute programming.
for Sonology during the early afternoon of August 25, 1978.
CBR When did you first get involved writing composing
Background programs?

CBR Mr. Koenig, can you tell us, when did you first become GMK Right from the beginning. In the computer program-
involved with music as a composer? Can you tell us ming class, exercises were given mostly in fields not
about your early development as a composer? related to music. When I said I would like to have some
programming exercises in the music field, the teacher
GMK I started in a music conservatory and studied besides was very happy about it, of course. So I made little
music theory, acoustics, analysis and piano, also com- programs about making 12-tone series and some
position as a student of music; I started making or harmonic control and things like that. And after a
composing music as a child together with my piano couple of experiments like this, I started to write a
education and developed that up to a certain level larger composition program which later would be the
when it became necessary to study music profes- PROJECT 1 program.
Curtis Roads: A n Interview with Gottfried Michael Koenig
PROJECT 1 and PROJECT 2 design electronic music completely different from the
design of instrumental music.
CBR When was PROJECT 1 actually written?
On the other hand, if we don't talk about electronic
GMK I suppose in the years '64, '65, and '66. music, then it's a question of whether we could try to
use the VOSIM system to add a sound part to PRO-
CBR Can you tell us about PROJECT 1 itself? JECT 1 which would replace the string quartet or the
piano or the orchestra. I think that that is theoreti-
GMK Yes. The basic idea was to make use of my experience cally possible, not at the present stage, but in the
in serial music and also the aleatoric consequences future when all those instruments are properly de-
drawn from serial music. I tried to describe a model scribed and can be just inserted in the program. But at
in whlch certain basic decisions were described, that moment I think I would lose interest, because if
and where besides that the user-the composer- I would want a piano I would take one, and not try to
would have some influence over the musical variables. imitate it with any VOSIM generator in the world.
In that way you could ask for any number of variants
which would evolve from the same basic principle; So what I am expecting from devices like VOSIM
you could compare them and see to what extent the generators is a new field of sound, not speech sounds,
musical characteristics laid down in the program not instrumental sounds, but something which I've
were really experiencable in music. tried to achieve in the electronic music studio and
towards which I will keep striving. The VOSIM gener-
CBR And how did the PROJECT 1 program lead to PRO- ator is just a convenient means of having a sound
JECT 2? output for PROJECT 1 because the VOSIM system was
developed at this Institute and it's just there.
GMK
Sometimes I call PROJECT 1 a closed program because
the composer can exert very little influence on the CBR Can you tell us about the relationship between PRO-
program itself. In contrast to that, I wanted a program JECT 1 and the VOSIM system through the program
which would, according to the experiences gathered VOSACS? What is the VOSACS program?
with PROJECT 1, allow a composer to fill in his own
variables, not only with respect to the musical material GMK The VOSACS program is a routine which enables you
-what kind of pitches or loudnesses or durations- to test the VOSIM variables, nine of them, at least
but also with respect to compositional rules-the way in a small number of steps. There's a certain variable
the elements are put together to form a musical con- which can have about two thousand states or steps,
text. To that purpose I designed PROJECT 2, which of which the VOSACS routine only uses maybe four
is actually a questionnaire of more than sixty questions. or eight spread over this range. Still, it gives a general
According to those questions, which refer to the view, and enables the composer to make up his mind
musical material and the rules, the program would what kind of VOSIM sounds he wants to use when
combine or compose a piece. playing a PROJECT 1 score.

CBR So you not only specify elements, you specify oper- CBR You said that PROJECT 1 was originally designed to
ations in PROJECT 2-predicates, as it were, for make instrumental music. How about PROJECT 2?
composition.
GMK Yes, the same is valid for PROJECT 2. PROJECT 2
GMK
Yes. being kind of an extension to PROJECT 1 also was
designed to make instrumental music. I have made
PROJECT 1 , PROJECT 2, and VOSIM scores with PROJECT 2 and the new version which is
in preparation will also be given a VOSIM extension,
CBR Recently you've combined your PROJECT 1 program so that the results of PROJECT 2 composition can be
with the VOSIM oscillators. Are the VOSIM oscillators heard immediately.
a kind of ideal synthesis output for PROJECT l ?
Extensions to PROJECT 1 and PROJECT 2
GMK
I don't know yet. Maybe not, because the VOSIM
system was designed for a very special task, namely CBR Are you planning a composing program to compose
for making speech sounds. They also offer the possi- electronic music? And if so, what kinds of principles
bility of making a large variety of sounds not being would it incorporate? How would it be different
part of any natural language. But still, it's not original- from PROJECT 1 and PROJECT 2 for instance?
ly designed to replace an electronic music studio for
instance. Now neither is PROJECT 1. PROJECT 1 GMK In instrumental music, my starting point is always the
actually is a program which designs instrumental music relationship to a well-defined musical material. You
scores. But under certain conditions you could use this have the orchestral instruments, and all the division of
kind of structure also for electronic music; but there labor that takes place in an orchestra, which results in
are restrictions in the program which are more on the harmony, in certain rhythmic patterns and even in the
side of instrumental music. I mean, I would personally notation.
Page 12 Coniputer Music Journal, Box E, Menlo Park, CA 94025 Volume I I Number 3
 Electronic music is con~pletelyfree from all those re- Primarily I'm very annoyed with composers using the
strictions. So when I might in the future try to write most modern tools of music-making like electronic
a routine to make electronic music with a computer, I music, voltage-control, even computers, and making
would not start with those traditional language-like twelve-tone series for instance, or trying to imitate
features I just mentioned but from predescribed sound existing instruments. That has, of course, its scienti-
characteristics or ranges in which sounds could move. fic value, but not necessarily a creative value in new
Then I would try to develop the grammar of the music music-making. But if a system is designed to produce
according to the sounds and not the other way around. twelve-tone series and instrumental sounds, then it is
very difficult to avoid that. So just to be able to avoid
Researcher and Composer that, to open up new fields of sounds you would not
be able to produce or would not think of describing
CBR In an interview with NUMUS-WEST (1973) you said in classical terms, I have chosen this non-standard
that you were at that time a researcher, not a compos- approach.
er. Has this changed?
The Influence of Computers
GMK Not really. I have been busy designing programs, cor-
recting them, and testing them for many years besides CBR How have your compositional ideas been influenced
other work I have had to do in these last years. I hope by working with computers?
to become a composer again, maybe next year and the
years after. GMK Not so much. If I compare my own compositions of
the pre-computer era to what I have tried t o describe
CBR Is this prompted by the new possibilities of PROJECT in my programming and what still fascinates me, then
1, PROJECT 2, and the sound synthesis systems here there is not so much difference. I mean I have gathered
at the Institute? experience as a composer for many years, say twenty
years, something like that. A program will not be able
GMK Actually I've waited for many years to put those pro- to describe twenty years of experience but only a
grams to use. There were many interferences, not only week of it or maybe a month. That means that I try to
because of my own work but also because of the com- influence my programming by my experience instead
puter development. Computers break down occassion- of the other way round. I try to lay down what I
ally, and things like that. know about music and at this moment I think I know
more about music than my computer program.
SSP and "Non-Standard" Synthesis
CBR Can you tell us about your sound synthesis program CBR Are your composing ideas influenced by the sounds
SSP? that you're able to create with the systems at the
Institute?
GMK Yes, this program uses what we call the "non-standard
approach" to sound synthesis. That means not refer- GMK Yes, especially in the field of electronic music. Then
ring to a given acoustic model but rather describing the it's difficult to say. Have I chosen electronic music
waveform in terms of amplitude values and time because of sound ideas or have I composed in a certain
values. My first intention was to go away from the style because of the sounds possible in the studio? I
classical instrumental definitions of sound in terms think there is a mutual influence. I think that the com-
of loudness, pitch, and duration and so on, because positional technique should be in a very narrow and
then you would refer t o musical elements which are strong relation t o the sound sources. That's my interest
not necessarily the elements of the language of today. in programmed music-to put possible sound sources
To explore a new field of sound possibilities I thought into a program which combines them instead of in-
it would be best to close the classical descriptions of strumenting a piano score.
sound and open up an experimental field in which you
would really have to start again. It would be the task CBR One kind of artificial intelligence task is that of a pro-
of a later time or other people to map the new pos- gram itself knowing what kinds of sounds that it's
sibilities t o the old experiences. actually dealing with, and altering the program logic
according to these sounds. Do you see this as a pos-
CBR What are the primary values of a non-standard ap- sibility?
proach (or synthesis-by-instruction) vis-a-vis a standard
(or synthesis-by-rule) approach? GMK Not only as a possibility, I think it is even a necessity.
My own programming does not take that aspect into
GMK That's difficult to say. Even though I wrote the pro- account as yet because the programs were meant to
gram, I don't have enough experience (with SSP) at be instrumental music programs; but as soon as sound
this very minute to make a judgement about that. production is involved, you need some kind of feed-
back or relationship between the musical language
structure and the structure of the sounds produced.
Curtis Roads: An Interview with Gottfried Michael Koenig
Composers and Computer Music place, we have research in fields of musical structures
and musical sounds-that means both analysis and
CBR What directions being pursued by other composers are synthesis of musical structures and musical sounds and
of interest to you? of course the relationship between them. We organize
a series of concerts every year both in Utrecht and
GMK When we talk about computer music we have to dis- Amsterdam including not only electronic music but
tinguish between the act of composing and the act of also films in which modern music plays an important
sound production. In many cases, the modem means of role. We gather documents- as many as possible, not
sound production of the computer are used to make only in the form of books and records, scores and
let's say, outmoded music to come into being, into periodicals, but also any kind of document which
sound. It's not always the most advanced structural de- comes to our knowledge is filed and even stored on the
sign which is used in computer music. I think com- computer for easier reference. We have our own elec-
posers I admire most are not busy at this moment in tronic workshop to maintain the apparatus and also for
the computer music field whilst what I have heard in new development of analog or digital hardware.
computer music attracts me more or less in terms of
technical procedures and not so much as aesthetic CBR There seems to be a great deal of collaboration among
results. the people within the institute. This seems difficult to
achieve at some other institutions. What are the ingre-
My general feeling is that computer music has seen a dients of this collaboration? Is it simply a matter of
very fine development in programming, software and survival for all of you, i.e., you have to work well
hardware development, but composers have a strong together, or, are there organizational techniques that
tendency to instrumental music and sometimes it make this easier to achieve?
seems that they avoid computers. Maybe they feel that
they wouldn't be able to express themselves the way GMK Ja, in the first place the Institute has grown step-by -
they are able to with an orchestra using human beings step. That means only people who were attracted to
and all the mechanical instruments of the musical past. our way of working joined the staff. In the second
stage of development we have tried to find an internal
CBR Are composers afraid of the concept of programmed organization which optimizes the collaboration of the
music? staff members and also everyone who comes from the
outside.
GMK
I think so.
CBR How is this actually implemented? Are there obliga-
CBR Why do you suppose that is? tions and responsibilities that each staff member has?
Must he or she fulfill a collaborative function within
GMK
In the first place, in most musical institutions, com- the group?
posers are not prepared to think of music in terms of
a set of given rules. I mean, programmed music is not a GMK
Yes. We have three main committees which are re-
concept taught in most music schools. That's one elected every year or every two years, concerned with
point. They're used to thinking in terms of what they scientific research, artistic creation, and technology,
can do with their own breath and hands, with mu- that's one committee, the second committee sees to
sicians and conductors, and all the given conditions of teaching, and third committee about public relations,
musical life. They have to fit in. That's the place where information and publications. In that way, everything
they will earn the money. that is going on in those fields is talked over in the
committee so that every subject can be examined from
Computer music is somethmg hard to sell to a larger different aspects-it's a good coordination that way.
audience. Composers who are not paid by the govern-
ment but working freelance have to consider this An Interactive Approach
condition.
CBR I notice one characteristic of the Utrecht studios seems
The Institute of Sonology to be that whenever you develop a synthesis technique
you seem to have someone from the hardware section
CBR I'd like to ask a few questions about the Institute. Can develop a device to carry out the synthesis. It seems
you tell us a little about some of the other work going that you optimize your computer resources towards
on at the Institute right now? interaction with composers, and leave the synthesis
computation to the hardware generators. That seems
GMK Yes. The Institute is busy in different ways. On the one to work well for you.
hand we have the development of studio and computer
facilities. Secondly there's an annual course, the sub- GMK Yes. We started with real-time synthesis without any
jects being signal processing, computer techniques, hardware oscillators, but soon we reached the limits
logic, and practical exercises in the studio. In the third of that technique. It is still applied to certain ap-
place, we have composers who make use of our facili- proaches, but on the other hand the computer is just
ties to produce music compositions. In the fourth too slow to compute in real-time complex sound
Page 14 Corrlputer Music Journal, Box E, Menlo Park, CA 94025 Volume I I Number 3
 structures, especially if they are also embedded in a GMK As long as only sound synthesis is involved I wouldn't
complex language structure. Hardware generators bother so much-analog means are fine, digital means
make it much easier t o compute these complex struc- are sometimes better, sometimes not. All the precision
tures. I want to point out that if there is a new tech- you get from digital apparatus is not what every
nical design, before it's carried out it's talked over with type of composer really wants. But still, then it
the artistic and scientific people at the Institute, and wouldn't make much difference to the composer
also the other way around. So nothing is done without whether he handles analog apparatus or digital appa-
an agreement from all sides. ratus. But if you want to be involved at the same time
in composing problems like programmed music or
CBR All of the systems here at the Institute are interactive. preconceived structural ideas, then I think the situation
Is this by design? Was there an agreement among all is slightly different because it affects the state
the people at the Institute? of the mind of the composer and not only the state
of the apparatus.
GMK A silent agreement I would say, and then at a certain
moment one talks about it, and recognizes its useful- In that respect I think we need many years to develop
ness. computer programs for not only making sounds
but composing structures, to develop things which are
CBR Almost all of the systems have immediate sound feed- really attractive to composers and useful to composers
back! so they can feel at home and express themselves.

GMK Yes. That's possible because we don't share the com- Today, the composer who wants t o work with precon-
puter with other institutions. We own our own com- ceived structural ideas mostly uses his desk for that
puters, and in that way we have the computer at our -not necessarily wanting sound immediately -but
disposal- twenty-four hours if necessary. Tinat speeds rather t o put certain features, certain characteristics,
it up; it's not necessary t o make elaborate preparations certain items in certain configurations. But I could
in terms of programming and input data, then carry it imagine a stage in which a computer would respond
to a computer center and wait for the output. not only with sound but also with structures with
such speed that the composer could have as many
For extensive experimentation the immediate sound compositions to consider in real-time as he's now able
response is necessary, otherwise we wouldn't be able to consider sound characteristics.
to carry out a large number of experiments which are
necessary to develop programs which serve composers.
Composer as Selector
Future Projects

CBR What ideas are on the horizon for the Institute? CBR So are you interested then in the idea of composer as
selector- the computer generates a number of
GMK One idea is t o build hybrid studios, in which analog examples of either a sound structure or a musical
means and digital means are combined, up to a certain structure and the composer selects from the possi-
degree. bilities generated by the con~puter?

CBR This is already under way? GMK Yes. It's hard to describe. I think he should not only
select but also have some influence on the agents
GMK This is already under way in one studio and I could which prepare the selection options. But then I am
imagine that we could follow that direction later on. asking myself whether the composer has not always
The thing is, if you offer analog and digital means to selected, through the centuries, among given items
composers, it's hard to know beforehand what they like instruments and instrumentalists.
would turn to. I mean, there are composers who
would prefer the analog audio studio even if you have One should either offer the composer a wide variety
the most modem computer facilities around, just of compositional means, in terms of programs and
because of their habit of music-making. programming languages, or on the other hand, should
educate him in such a way that he would be able not
Analog versus Digital only to select among means but also create his own
means.
CBR As far as sound synthesizers go, it seems to me that
some of the new digital synthesizers will have much But that's an educational problem. That depends not
the same composer interfaces as the analog equipment only on the psychology of the composer but also on
-only everything behind the knobs and dials will society in general, on the role music plays in society.
be digital. Some people, however, draw a great distinc- And that's changing.
tion betweeen analog and digital. What are your
thoughts? CBR How is it changing? (continued on p. 29)

Curtis Roads: An Interview with Gottfried Michael Koenig

takes is the combined result of the overlapping individual older romantic idea of the individual composer writing notes
activities of all the parts with the coordinating influence of in isolation of audience and other musicians.
the data exchanged between computers. The structure of a circular system satisfies the desire
Since mistaken concepts and "bugs" seem inevitable, and for a symmetrical interactive network where the flow of
plans of any complexity usually break down, it is heartening influence emanates evenly from each point in the system.
to note the mystery that where several errors intersect they Because musicians, since time immemorial, have been
very often make an interesting pattern. playing together, music has developed into a wide variety of
But how do you get three modern composers to work "naturally occurring" parallel processing systems.
together? Micros with their simple structures provide an We created an interesting creature and spent an evening,
answer. On the other hand large computer facilities and large in public, listening to it.
electronic music studios seemed t o be an extension of the

(continued from p. 15)

GMK
Faster sometimes than technology, I think. It takes common body of knowledge that everybody can draw
years to develop electronic music studios according t o from?
the demands of a given time and then its always too
late! The moment the studio is developed up to its GMK That's a very steady but slow-growing process; every-
original standards, the composers are miles away. body talks about better communication between
studios-one should come together more often, have
CBR What programs and technical devices need to be built? conferences, and the like. The computer music con-
Exactly what systems then do composers need today? ference is now established but it's in only one place in
the world, and for Europeans it's harder to get to the
GMK The composers of the past, by that I mean the last ten States. On the other hand, if it were held in Europe it
or twenty years, would not have been able to build would be the same problem. You can't telephone all
their own hardware, or design their own studios, or day to everybody, its too costly and you don't have
to write their own software. So they had to accept the time. So its a theoretical wish that people have-
anything that was prepared in their respective studios to get together more frequently to exchange ideas
and institutions. But in the future, composers will and knowledge. I would think that its not really that
be educated in a different way or at least will have necessary to make an effort to bring people together
other possibilities of being educated, in terms of math- more often, because certain ideas lie in the air, so to
ematics, sound synthesis techniques, linguistics, arti- speak. Some things are developed independently of
ficial intelligence, or what have you. Also the prices each other at different places. You discover that it
of digital hardware are decreasing. So I could imagine wasn't really necessary to consult because you know
a future situation in which a composer would not actually what the present-day problems are.
wait for the Institute of Sonology to make things,
but would rather buy his own computer, to design On the other hand the exchange is difficult. Even the
their own software in a much faster time period than exchange of computer programs seems to be difficult.
the past. Even the technology of designing programs Systems are different and approaches are different.
is progressing. So collaboration is a little hypothetical.

CBR Do you then see a situation of decentralization? The nice thlng about different institutes is that you can
have different approaches to the same problem. Sup-
GMK
I think so. pose we would try to imitate what the French are
doing in Paris, or they would try t o imitate what we
Collaboration Amongst Computer Music Centers are doing here. Then you would have the same institute
twice. On the other hand, if you are an individual
walking in your own direction then you want to know
CBR What about an intermingling among centers for com-
what's around you but you don't want to follow
puter music now? Do you forsee any of this? What are
the others. Exchange is sometimes necessary but it
the difficulties or advantages of collaboration?
doesn't always solve your own problems.
GMK But music is made by individuals. Composers compose
alone. They don't form teams. Places like IRCAM CBR You see more of a collaboration on the level of ideas?
or this Institute or others are more-or-less centered
around individuals who design the basic principles. GMK Yes, and there are periodicals in which ideas and
developments are described. If you really are in need
CBR And yet is not musical research-by the definition of of certain knowledge, then you can do as people have
what is scientific-socially communicable? Is it not for centuries-just travel to the place and study the
something that should feed together to create a problems there.
John Bischoff, Rich Gold, and Jim Horton: Music for an Interactive Netwc~ r o
k f Microcomputers Page 29