Content uploaded by Phivos-Angelos Kollias
Author content
All content in this area was uploaded by Phivos-Angelos Kollias
Content may be subject to copyright.
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
59
Application of Systemic Principles in Music Composition
Phivos-Angelos Kollias
Centre de recherche Informatique et Création Musicale
Université de Paris VIII
France
soklamon@yahoo.gr
http://phivos-angelos-kollias.com
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
Abstract !
In this paper, I will present the practical results of my research on the interdisciplinary
scientific field of Systemics. I will include two experimental approaches to
composition based on the application of systemic principles.
Systemics consists of a number of interdisciplinary theories based on organizational
approach to problems. From a systemic viewpoint, everything is considered as a
system, i.e. as a complex of interacting elements.
In the first part of the paper, I will show how I have applied the theory in instrumental
composition. In this approach, I have attempted to develop an experimental
compositional model based on a model of live interactive music from a systemic
viewpoint. In the ‘Systemic Model of Symbolic Music’, we are interested on the
information’s flow through ‘symbolic’ means, i.e. through music notation. In addition,
the approach treats ‘systemically’ the compositional work, applying notions found in
Systemics through the help of the Cognitive Sciences.
In the second part, I will show an alternative approach to interactive electroacoustic
composition, also based on concepts of Systemics. In this approach, the musical
work it appears in time like a ‘living music organism’, a musical work able to adapt in
any given situation but always maintaining a stable and recognisable structural form.
This ‘organism’ results from a live algorithm, a software, installed on a computer. The
organism has the ability to ‘listen’ through the microphones and to ‘express’ itself
through the loudspeakers. In this way, the organism is a self-organised system, in
other terms it is capable of influencing its own organisation. Here, I will demonstrate
the structure of the ‘organism’ and I will explain the basic principles of its creation.
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
60
Introduction
Systemics or Systems Thinking is a
general term referring to a number of
interdisciplinary theories that have in
common the concept of organisation.
Among others, Cybernetics, General
Systems Theory and Complexity
Science, observe and describe
everything through models using
concepts like feedback, interaction, self-
organisation and emergence1.
Systemics emerged after the Second
World War and over the years that
followed influenced significantly the
scientific thinking but also the common
thinking. In addition, composers based
their musical ideas on systemic theories,
like Xenakis’ model of Markovian
Stochastic Music or Di Scipio’s model of
Audible Ecosystemics.
In this paper, I will discuss briefly two
experimental approaches to composition
based on the application of systemic
principles. The first approach I call it
Systemic Model of Symbolic Music and
it principally concerns instrumental
composition. The second approach I call
it Self-Organised Electroacoustic Music
and it concerns electroacoustic
composition2.
The Cybernetic Model
Both approaches I am going to present
are based on an interpretation of the
Cybernetic model. In the simplest
expression of the model, a feedback
system is formed from a receptor, a
control apparatus and an effector
(Bertalanffy 1968, pp. 42-43) (Figure 1).
The system receives stimuli in the
‘receptor’, which is the system’s input, a
kind of sensory organ for the system.
These messages are sent to the ‘control
apparatus’ where they are processed.
Then the result is transmitted to the
‘effector’, the system’s output. The
effector responds to the messages with
an action. The transmission of the
message from the input to the output
takes some time depending on many
factors perceived as a delay of the
effect. Finally, the receptor’s function is
‘fed back’ to the receptor that makes the
system able to regulate its own action,
i.e. a self-regulating system.
Applications of the cybernetic model can
be found everywhere, as simple as the
thermostat of a boiler or as complex as
the navigating system of living
organisms.
Figure 1. The cybernetic model (Bertalanffy 1968, p.42)
Systemic Model of Symbolic
Music
My first attempt to apply systemic
principles to music started in
instrumental composition, with the
Systemic Model of Symbolic Music
(Kollias 2007, chapter 4; 2008a). In this
approach, we are interested in the
information’s flow through ‘symbolic’
means, i.e. through music notation3. In
addition, the approach treats
‘systemically’ the compositional work,
applying notions found in Systemics.
Central role of the model has the
Creative System, a kind of regulating
system controlled by the composer for
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
61
the production of the music score. The
approach implies two creative stages:
system’s architecture and system’s
control.
In the stage of system’s architecture, the
composer designs a Creative System,
which will use for a part of the
composition. The designing of the
Creative System depends on the
elements of his choice and the
interrelations he creates among them
according to what result he is expecting
to get.
Figure 2. Systemic Model of Symbolic Music
Figure 3. System based on the relations of violin, viola, cello according to their lower pitches
In the stage of system’s control, the
composer puts into action his Creative
System.
He introduces symbolic information
to the system through the input.
The input information is connected
with the output through a mapping
function.
the result derives from the output
after certain delay, gradually forming
the emerging score.
The composer monitors the result
during the whole process.
I will demonstrate a simple example in
order to show how the model works.
Let’s design a Creative System based
on the relations of violin, viola and cello
in terms of their lower strings (Figure 3).
To make the result clear, I will use only
linear mapping between the parameters
of the same symbolic domain, i.e. pitch
mapped with pitch, duration mapped
with duration and so on. Moreover, each
relation between two instruments
defines all the parameters (‘mapping’
and delay) and the number of events in
each transfer of information stays
constant (Figure 4).
In the table of Figure 5 we can see all
the changes resulting to the relations
among the instruments. Here there are
no relations among dynamics. Finally, in
Figure 6 we can observe what happens
if we enter into the input the information
of the violin’s first measure.
Fig. 4 – Creative System of Symbolic Music
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
62
Figure 4. System based on the relations of violin, viola, cello according to their lower pitches
Figure 5. Table of mapping and delays
Figure 6. The result of the system’s output
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
63
Self-Organised Electroacoustic
Music
My second attempt to apply systemic
principles was in electroacoustic music
(Kollias 2008b). Because of the inherited
systemic nature of the electronic
programming, the applications are
easier to prepare and more direct to
control. Furthermore, the results are
more apparent.
Based again on the cybernetic model, in
this approach the electronic device is
programmed in a way to be able to
organise itself, in direct interaction with
the sonic environment of the
performance.
I have previously described (Kollias
2008b, p.140) with the term Self-
Organised Music, ‘the result of the
interactions between some predefined
structures and an occasional context of
performance, through a particular
interpretational model’. In the context of
electroacoustic music, the ‘predefined
structures’ are represented by the DSP’s
setting. Moreover, the ‘interpretational
model’ is the definition of the real time
control parameters, what Agostino Di
Scipio refers to as Control Signal
Processing or CSP (Di Scipio 2003).
As in the approach we saw above, the
cybernetic model is of central
significance here. This time the
approach is based on the model of
Second-Order Cybernetics (Heylighen
and Joslyn 2001), a more advance
model to represented self-organised
systems.
In this general model of self-organised
electroacoustic music, we consider the
music work as a self-organised system
(Fig. 7). Its goal is to control a number of
preferable variables. These variables
represent particular features of sound. At
the same time, there are unforeseen
sounds that destabilize the system’s
preferable variables, in other terms
noise, described as perturbations in
cybernetics.
To begin with, the system observes its
sonic environment. Here, the system’s
input is the microphones. In the process
of perception, sound is represented
digitally within the system. The
representation of sound is treated in two
different lines of processing: the DSP
and the CSP. Within the CSP setting,
combinations of values, representing
specific sonic features, influence the
values of the DSP through a mapping
function. In this way, the DSP’s
characteristics are regulated from the
CSP, at the same time with the DSP’s
processing. The result of the system’s
process is going to the output, which are
here the speakers. The speakers are
acting in the sonic environment by
diffusing sound. This sonic action has an
impact on the ‘dynamics’ of the sound
environment. At the same time, the
perturbations of the environment
influence sound’s dynamics and
indirectly destabilize the system. Finally,
the circle restarts with the whole sound
result in the performance space that
again is perceived from the system.
Based on this model, it is possible to
program a self-organised music system.
From the interactions between the
system and the environment, a music
organism will emerge. Notably, the
artistry here is not how to construct
interesting events in time. Instead, the
creative challenge is how to create a
network of interaction, i.e. the setting of
the CSP, which can give a satisfying
sound result in different circumstances.
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
64
Figure 7. A general model of self-organised electroacoustic music
In my approach, my intention is to
acquire spontaneity and unpredictability
of self-organisation in the creation of
sound. At the same time, to be able to
control the result from a higher
organisational level in order to develop
formal structure. I have shown that this
is possible by applying the systemic
principle of equifinality.
Equifinality
According to von Bertalanffy (1968, p.
39), in a closed system, i.e. a system
isolated from its environment, the initial
conditions determine a particular final
state. As a result, a change of the initial
conditions results to a different final
state. However, this is not the case in
open systems, that is to say systems like
living organisms. One of the properties
of open systems is to achieve the same
finals state upon different initial
conditions, what is called equifinality
(von Bertalanffy 1968, pp. 142-143). For
example, the property of organisms of
the same species is to reach a specific
final size even though they start from
different sizes and going through
different growth’s courses.
Control over Self-Organised Music
In order to acquire control over self-
organised music, I have formulated the
following hypothesis:
‘[I]f we consider the music organism as an
open system, it is possible to create
certain conditions in which the organism
will show tendency for ‘equifinal’
behavioural states. […] I believe that we
can influence the system in order to pass
from a series of behavioural states, which
can be similar in any constitution of the
same organism under similar
circumstances.’ (Kollias 2008b, p. 144)
In this approach it is possible to control
the system in a basic level, by designing
its elementary structures. At the same
time, we can acquire control over a
higher organisational level, that of
macro-structural form, without
interrupting the system’s ability of self-
organisation. In other terms, we can let
the system constitute itself, showing
emerging properties over the different
organisational levels and by indirectly
influencing these properties we can
acquire a desirable result of distinctive
character. Notably, in this approach the
composer is designing in a micro-
structural level and at the same time,
through the role of the performer, he is
controlling the sound result from a
higher organisational level.
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
65
Ephemeron
In my work Ephemeron (2008) I applied
the above hypothesis achieving to
create a ‘live’ organism with a specific
formal constitution in time. The basic
concept of the work is that the composer
creates a DSP as a living music
organism. This organism is able to
‘adapt’ in a given concert’s space, the
organism’s environment. The sound
result depends solely on the organism’s
interactions with its environment and the
minimal influence of the user upon his
behavioural states. There is no pre-
recorded material used at any point
during the performance. The organism’s
adaptation is the result of the organism’s
properties causing changes to the
organism’s processes as a consequence
of its constant communication with the
given spatial properties.
Conclusions
In this paper, we have seen two
approaches to composition based on the
applications of Systemics. Both
approaches are based on the cybernetic
model of a self-organised system.
The first approach is applied to
instrumental music, the ‘Systemic Model
of Symbolic Music’. Here the composer
creates a system, which he then uses by
introducing symbolic information and
which results to a music score.
The second approach is applied to
electroacoustic music, what I call ‘Self-
Organised Electroacoustic Music’. The
composer designs an electronic
algorithm which interacts with the given
concert hall causing to emerge a music
organism.
In: Motje Wolf (Ed.) Proceedings of Sound, Sight, Space and Play 2009
Postgraduate Symposium for the Creative Sonic Arts
De Montfort University Leicester, United Kingdom, 6-8 May 2009
http://www.mti.dmu.ac.uk/events-conferences/sssp2009
66
Bibliography
DI SCIPIO, A. (2003) ‘Sound is the interface’: From interactive to ecosystemic signal
processing, Organised sound: An international journal of music technology, 8 (3), pp.
269-277.
HEYLIGHEN, F. and JOSLYN, C. (2001) Cybernetics and Second-Order
Cybernetics. In: Meyers, R.A. ed. Encyclopedia of Physical Science & Technology.
3rd ed. New York : Academic Press, pp. 155-170. [WWW] available from:
http://pespmc1.vub.ac.be/Papers/Cybernetics-EPST.pdf [accessed 24/07/2007].
KOLLIAS, Ph. A. (2007) La Pensée Systémique et la Musique: Les rapports de
Iannis Xenakis et d’Agostino Di Scipio à la pensée systémique. La proposition d’un
modèle systémique de la musique symbolique. Unpublished Dissertation (Master 2),
Université de Paris VIII. English version: Systems Thinking and Music: The
connections of Iannis Xenakis and Agostino Di Scipio with systems thinking. The
proposal of a systemic model of symbolic music. [WWW] available from:
http://phivos-angelos-kollias.com/ [accessed 12/12/08].
KOLLIAS, Ph. A. (2008) Music and Systems Thinking: Xenakis, Di Scipio and a
Systemic Model of Symbolic Music. In: Proceedings of the 5th Conference of
Electroacoustic Music Studies Network, Paris, June 2008. Paris: INA-GRM and
University Paris-Sorbonne. [WWW] available from: http://www.ems-
network.org/ems08/papers/kollias.pdf [accessed 05/05/09].
KOLLIAS, Ph. A. (2008) Ephemeron: Control over Self-Organised Music. In:
Proceedings of the 5th International Conference of Sound and Music Computing,
Berlin, August 2008. Berlin: Technische Universität Berlin, pp. 138-146. [WWW]
available from:
http://www.smc08.org/images/proceedings/session7_number4_paper38.pdf
[accessed 12/12/2008].
VON BERTALANFFY, L. (1968) General System Theory: Foundation, Development,
Applications, Revised edition, 1969; 15th paperback reprint, 2006. New York: George
Braziller.
XENAKIS, I. (1963) Musiques Formelles. Paris: Editions Richard-Masse. [WWW]
available from: http://www.iannis-xenakis.org/MF.htm [accessed 20/10/2007].
1 For a more detailed presentation of Systemics see: Kollias 2007, chapter 2.
2 The two approaches are briefly presented here in the context of the conference. For detailed
discussion of each approach, see the individual research results.
3 I use the term ‘symbolic music’ (music symbolique) originally introduced by Xenakis. He refers to
‘symbolic music’ as ‘a logical and algebraic draft of music composition’ (‘une esquisse logique et
algébrique de la composition musicale’; Xenakis 1963, p. 185).
