Audible Ecosystems and emergent sound structures in DiScipios music' Music philosophy helps musical

1
Audible Ecosystems and emergent sound structures
in Di Scipios music. Music philosophy helps
musical analysis

• Renaud Meric
• Makis Solomos
• (Rirra21, Université Paul Valéry-Montpellier 3)
• Fourth Conference for Interdisciplinary
  Musicology,
• ???????????, July 2008

2

• 1. Introduction

1.1. Musical analysis and music philosophy
Musical analysis and music philosophy two very
different paths, which try to understand music
-musical analysis works on the
(technical/technological) means -music
philosophy works on the meaning.
These two musical disciplines are complementary
normally, music philosophy starts where musical
analysis stops (or the opposite way).  
In some special cases, one can boost the other ?
one can force the other to change its perspective
or even its tools.
We are interested here to the cases where music
philosophy can boost musical analysis.
Is it sometimes the case when the musical work,
which is to be analysed, appeals to totally new
ways of musical behaviour, of music
understanding, of musical perception.
CIM08
3
1.2. Musical structure
Musical structure -it is often understood as
something static, and which underlies music (or
sound) -as something which is there, which has
an ob-jective quality (standing in front of us,
as an object).
We will discuss here a kind of music where there
is no such structure In this music, structure is
an emergent quality.
1.3. Our case study A. Di Scipios music
 
Agostino Di Scipios Audible Ecosystemics
Studies -the Audible Ecosystems implement an
audio system that interacts with the environment
(space) -in Audible Ecosystems, structure is
emergent it is unpredictable and unstable (it is
dependent on sound and the process of listening).
 
While trying to analyse this music, we will see
that we need a new approach, which can be
provided by music philosophy.
R. Meric, M. Solomos Audible Ecosystems...
CIM08
4

• 2. Agostino Di Scipios music

Italian composer, born in 1962. Compositions for
electronics, live electronics, electronics and
instruments, sound installations New theoretical
ideas cf. Di Scipio, 1994, 1999, 2003, 2005a,
2008
Two theoretical ideas
2.1. Emergent sound structures
Granular paradigm
A tradition starting with Xenakis and going to
recent music, through Horacio Vaggione, Curtis
Roads implementations
Xenakis -All sounds represent an integration
of corpuscles grains, of elementary acoustic
particles, of sound quanta. Every sound,
every even continuous variation of sound is to be
understood as an assembly of a sufficient number
of elementary particles being disposed adequately
within the time level (Xenakis, 1960
86) -Concret PH 1957-58, tape clouds of
grains crackling coal
CIM08
5

• -Xenakis Analogique A et B (1958-59, tape
  clouds of grains sinusoidal tones 9
  strings pizz., short arcos, col legno) ? higher
  order sonorities ? failure?

Ex.01. Concret PH
Emergent sound structures
Di Scipios analysis of Xenakis
experiments -Xenakis higher order sonorities
Di Scipio emergence of sound structures In
this case concerning Analogique B, the
distinction can hardly be made between a model of
musical articulation and a model of sound design,
insofar as the composers action is meant to let
the musical (macro-level) structure emerge from
sound itself and its internal organization
(micro-level) (Di Scipio, 1997
165). -Analogique A et Bs failure Xenakis
failure of grains fusion Di Scipio failure of
emergence.
Radicalisation of the granular paradigm
everything is emergent ? from the micro- to the
macrostructure no autonomous macrostructure
designs, no dramatic gestures (in Xenakis
granular compositions, there are macrostructure
designs).
CIM08
6

• 2.2. Audible Ecosystems

Audible Ecosystems
A series of live electronics pieces Audible
Ecosystemics
Triangular interaction -human agent -DSP
?Digital signal processing? computer -sonic
ambiance (of the performances room).
Fig.01. triangular recursive ecosystemic
connection.
CIM08
7

• System and circular causality

In these works, emergence is related to a
system organization ? close to living systems ?
emergence capacity of auto-organization The
passage of a system or process from a given
structural organization to a new state of order
which is recognized as a function of the
qualitative properties of the former, is what we
call here a phenomenon of emergence (Di
Scipio, 1994 206).
Auto-organization ? circular causality ?
feedback
Fig.02. Basic design of the Audible Ecosystemic
Interface.
CIM08
8

• Composed interactions

The idea of interaction is central.
Interaction -not in the usual sense (as an
information flow transformation of a sound
source) ? there is not a (primary) source -it
means the capacity of auto-organization, the
emergence of a structure a principal aim would
be to create a dynamical system exhibiting an
adaptive behavior to the surrounding external
conditions, and capable to interfere with the
external conditions themselves. A kind of
self-organization is thus achieved . Here,
interaction is a structural element for
something like a system to emerge (Di
Scipio, 2003 271) -the process is more
important than the result This is a substantial
move from interactive music composing to
composing musical interactions, and perhaps more
precisely it should be described as a shift from
creating wanted sounds via interactive means,
towards creating wanted interactions having
audible traces (Di Scipio, 2003 271).
CIM08
9

• 3. About musical analysis

3.2. What is music?
Emergent sound structures, processes, composed
interactions, ecosystems ? what is music?
Is it -a sonic result? ? No the process is
more important than the result -a voluntary
gesture (from one or more human agents composer,
performer, listener)? ? Not only the environment
is part of it -a language? ? No there is not
a dichotomy between material and meaning, there
are no symbols
I am interested in composing desirable
interactions among available elements or
components, such that the music is heard as the
empirical epiphenomenon of that network of
interactions, not as an abstract discourse
written by me and diligently spoken by others
(Di Scipio, 2005a 385).
CIM08
10
3.2. How should we analyse it?
How should we analyse Agostino Di Scipios music?
Usually, music analysis works on the level of
music itself -it starts from the result (of
the generating process) the sound, the
work -music is supposed to be a determinate
and finished world -it is a fixed entity, lying
there, ready to be analysed -in the case of
electronic music the result, the world, the
entity, the musical structure which is
there is sound a sound world, a sound entity,
a sound structure -the analyst deals with sound
objects, with objective structures -i.e.
objective entities  supposed to be there ?
independently of the listener, of the environment
There are no objective entities in Di Scipios
music
CIM08
11
3.2. Example Audible Ecosystemics 3b
Audible Ecosystemics 3b. Background noise
study, with mouth performer (2005) -one
performer -two microphones (one in the room
one used in the mouth performer) -a DSP unit
(Kyma workstation or Pure Data) -eight
loudspeakers in the room, turned backwards,
facing the walls.
Only source of sound background noise the
mouths performer.
The score instructions for the performer
CIM08
12

• Fig.03. Agostino Di Scipio Audible Ecosystemics
  3b. Background noise study, with mouth performer
  score

Lets listen till the end of the first half of
the performance
Ex.02. Agostino Di Scipio Audible Ecosystemics
3b. Background noise study, with mouth performer
0000-450 (CD hörbare ökosysteme
live-elektronische kompositionen 1993-2005, rz
edition, 2005, performer Natalia
Pschenitschnikova)
CIM08
13
To find a way how to analyse this music, we have
to adopt a philosophical approach of music.
CIM08
14

• 4. The philosophical approach

4.1. An analytic image
Sound result
Sound result can only be what we are listening.
The result is emergent ? Two particular
points -it is ephemeral it depends on
interactions between the musician, the DSP
computer and sonic ambiance ? no static structure
can be found -it is subjective ? close
interactions between the work and the listener ?
the listener and the work are in the same
space-domain and in the same time-domain.
Listeners are a very special kind of external
observer or hearer, because their mere physical
presence in the room acts as an element of
acoustical absorption. Hence there are rather an
internal component of the ecosystemic dynamics.
In the AESI Audible Eco-Systemic Interface
project, this is not considered as a problem, nor
an element irrelevant to the music changes in
the ambience will reveal peculiar changes in the
overall ecosystemic dynamics, and therefore in
the audible results themselves (Agostino Di
Scipio, 203 274).
CIM08
15
Image
These two points are problematic for musical
analysis -the first step of musical analysis is
to make a distinction between what is subjective
and what is objective -the analyst has to choose
a viewpoint on the specific time and the specific
space of the work.
In other words, the analyst translates the
musical work in a static image ? The musical work
can have its own structure.
Image -a static structure each element has a
specific place in time and space -we analyse the
relations between these static elements.  
CIM08
16
4.2. Prints, track, mark, trace, empreinte
Nothing is static
There are no specific (no source) sounds, no
instruments, samples or recordings. -Audible
Ecosystemics 3a. Background noise study starts
with nothing ? We usually do not pay attention
to barely audible sound events unconsciously, we
remove them from our auditory focus (Di Scipio,
2005b 20). - In Audible Ecosystemics 3b.
Background noise study, with mouth performer
The source is any small sound involuntarily
produced in the mouth and throat (Di Scipio,
2005b 20).
Inside the DSP unit, which processes the
signalsthe background noisecaught by two
microphones, there is no linearity -the signals
are routed through different ways they are then
routed thought different processing blocks -from
input to output, these ways are crossing
themselves several times (the signal is always
disrupted) and these ways are looped -every
signaleach part of it can be considered at the
same time as sonic data and as processing data
the signal is both the material and the gesture
that shapes it.
CIM08
17
Outside the DSP unit, the sounds, which are
emerging from the loudspeakers (the loudspeakers
are turned backward, facing the walls and close
to them) help to improve the background noise and
they are immediately caught by the two
microphones each element is both end and
beginning of the ecosystem.
CIM08
18
A complex listener's situation
The listener can be considered as an element of
the ecosystem -he is listening to the
background noise, which emerges from the
loudspeakers -he is making modifications to this
background noise, before it is caught by the
microphones.
The sounds, which emerge from the loudspeakers,
cannot be considered as an end and they are the
only step in the the ecosystems time and
space -the traditional identification of music
with the loudspeakers sound seems to be
inappropriate -we cannot determine a static
structure, which asks What are we listening to?
Empreinte
Therefore, what we are listening isnt a sound.
It is its own empreinte (French) ? in English
prints, track, mark, trace ? in Greek
ichnos.
The French philosopher of art Georges
Didi-Huberman says about the ichnologist (the
paleolontologist specialised in the study of
fossils tracks)
CIM08
19
The ichnologist has to recognize the
complexity of forms, he has to know that shapes
are processes, and not only processes results
that these processes dont have an end, that the
image, which is seen, is only the anachronistic
present of an uninterrupted play of
deformations, alterations, deletions, and
feedbacks revenances of different kind. The
ichnologist doesnt have to question what he is
looking at as the content iconography or the
expression of a mimetic desire. The likeness in
every empreinte prints, track, mark, trace is
of another order in it nothing can be
desentangled, for instance we can not separate
forms from matter. Here, shapes are substrates
there are the dialectical process of the
substrates modifications thanks to an ordinary
gesture. The ichnologist knows that what he sees
is not a historys single, intangible, point.
He knows that shapes are times at work,
contradictory times entangled in the same image
times of the earth, times of the foot, which for
one moment has set on it forever. (Georges
Didi-Huberman, 2008 310 our translation) Lich
nologue est obligé de reconnaître la complexité
des formes, il est obligé de savoir que les
formes sont des processus, et pas seulement le
résultat de processus  que ces processus, à
proprement parler, nont pas de fin, que limage
actuellement vue nest que le présent
anachronique dun jeu ininterrompu de
déformations, daltérations, deffacements et de
revenances de toutes sortes. Dautre part,
lichnologue na nul besoin de questionner ce
quil regarde comme liconographie dun contenu
ou lexpression dun désir mimétique. La
ressemblance offerte dans chaque empreinte est
dun tout autre ordre  en elle rien ne peut être
désintriqué, les formes de la matière par
exemple. Car ici, les formes sont des substrats,
ou plutôt le processus dialectique des
modifications du substrat par un geste
quelconque. Lichnologue, enfin, na pas la
naïveté de situer ce quil voit comme un point
unique, intangible de lhistoire. . Il sait
donc que les formes sont des temps à luvre, des
temps contradictoires intriqués dans la même
image  temps de la terre et temps du pied qui,
un instant sy est posé pour toujours).
CIM08
20
Georges Didi-Huberman explains the importance of
the ichnologist work for the art
theorist -shapes are processes (and not only
processes result) -shapes are
substrates -shapes are times at work,
contradictory times entangled in the same image.
CIM08
21
5. Back to musical analysis
5.1. What is moving
We can transpose the question what are we
listening to? to What is moving inside what we
are listening to? ? to immerse the listener into
the process.
Consequences -the listener is in the same time
and in the same space (same image) than the work,
than the sounds -sound is considered as a
process (empreinte) time and space are not
stopped -each movement that we catch is the
emergent tipthe moving empreinteof an
underlying moving structure interweaving,
matching and clashing others movements, which are
not emerging and which are not audible.
Little by little, and from movement to movement,
this soundas an entityvanishes, and we build up
an imaginary complex structure ? In this image,
the process of listening is not the result, the
aim or the end of musical analysis, but only its
beginning.
Lets make an attempt to analyse the piece,
starting from the genesis of an ordinary sound.
CIM08
22
5.1. Analysing Audible Ecosystemics 3b
A sample taken at random
Here is a sample of one second taken from this
piece ? an ordinary second taken at random
Ex.03. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0416475ms
This second is located in the end of the first
half of performance, where the performer has to
make a /u/ (no sound) and to wait until
electronics settle (score)
CIM08
23
Fig.03b. Agostino Di Scipio Audible Ecosystemics
3b. Background noise study, with mouth performer
score
CIM08
24
The difficulty to define and describe this sample
This sample of one second is not a precise and
well-delimited entity ? it is not a sound
object ? it is a complex interlacing (network)
of disparate and multi-scale movements -it is
made up of several and heterogeneous events (a
lot of grains, some impacts, some little clicks,
different whistles) -there are different
entangled time scales.
In the whole piece, sound is dispatched from 8
loudspeakers turned backwards, facing the wall ?
the sample is heard as a background noise, and it
emerges from it -it is impossible to define a
duration for each movement -spatial dynamics are
important and active (moving) but, they are
directionless.
CIM08
25
Empreinte  the origin
What is the origin (the root) of the different
movements we can listen to in the sample?
A part of it is emitted directly from the
ambient and barely audible background noise
(audience noise, air-conditioning sound).
An other part of it arises from the amplified
background noise and the amplified performers
vocal track noises, which were recorded 20
seconds earlier.
Example To understand this process, lets listen
to the beginning of the work -the 20 first
seconds there are only ambient background noise
Ex.04. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0000-0020
-the following 20 seconds, which are made up of
a) the ambient background noise b) the
amplification of the previous 20 seconds (we can
listen the particular reverberation colour from
the performers mouth, which makes an /u/ at the
end, we can hear the performer who iswho
wasbringing the microphone just before his
lips)
Ex.05. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0020-0040
CIM08
26
The last part of the sample arises from the DSP
processing (the recording background noise is
processed).
Example After 40 first seconds, we can listen to
the first electronic artefacts, which are
emerging little by little
Ex.06. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0040-0105
When the sample (at 416474ms) is listened,
all these parts are mixed together, compiled and
confronted.
CIM08
27
Mixing
Some movements of the sample arise certainly
from the ambient background noise, but we cannot
recognise and delimit them.
We know that a part of the movements is an
amplification of the previous 20 seconds (which
are itself an amplification of the previous 20
seconds).
Example Lets compare the sample with the sample
of the 20 previous seconds and the 40 previous
seconds
Ex.03. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0416475ms (the
random sample).
-in the 40 previous seconds, we can listen the
same whistles
Ex.07. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0336475ms (40
previous seconds).
-in the 20 previous seconds, we observe a) that
many clicks originate from vocal sound (this
origin is lost when we listen only our sample)
b) that some loud impacts have been pulverized
and decrease (the influence of the 8
loudspeakers spatialisation is here obvious).
Ex.08. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0356475ms (20
previous seconds)
As the DSP process is very active and complex
we have to focalise on the DSP processing.
CIM08
28
Focalisation on the DSP processing
What we are listening in the sample is
spatialised on 8 loudspeakers which are sharing
out 7 outputs
Fig.04. Speaker assignments schema.
Two outputs emit the amplification of the
previous 20 seconds.
Five other outputs are dependent on a switch,
which -is activated depending on the input
amplitude (based on erratic sound
wave) -controls different programmed triggers in
various ways ? thus we cannot foresee when each
output is activate.
CIM08
29
These five outputs arise from two different
processes 1. Three outputs send out the input
signal depending on different delays and
differently programmed triggers we may recognise
movements among other movements previously heard.
Example For instance, we can hear a similarity
between the sample and a moment heard
approximately 5 seconds before on the right, we
can hear the same melody
Ex.03. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0416475ms (the
random sample).
Ex.09. Audible Ecosystemics 3b. Background noise
study, in the vocal tract 0411041ms
2. Two outputs send the results from a granular
sampling -it reads the mixed signal from the
three previous outputs -its parameters (grain
duration, density, memory pointer, memory pointer
jitter) are dependent on different shifting
inputs (signal amplitude, switch) ? thus we
cannot foresee how it will react.
Example There is a lot of grains in this
example it is impossible to hear them
independently and they are combined as instable
movements.
CIM08
30
6. Conclusion
This short analysis shows -the complexity and
the multiplicity of a sample, of a sound -the
multiplicity of interleaved roots, origins.
The notion of movement allows -to link sound
(that we can listen to) and the DSP signals
(which are inaudible) together they form a
process -to reach and link the different time
scales we can listen to into a single sample -to
focus on the instability of the emergent
construction.
Of course, this is only an attempt, only a
beginning of analysis
Of course again, many other ways of analysing Di
Scipios music could exist
CIM08
31
As about the general notion of (musical)
structure Di Scipios music (Audible
Ecosystemics) is a good example for emergent
sound structures -structure is not something
given, it emerges (from the lowest levels) -it
is not ob-jective, standing in front of us,
there.
It is why it can not be analysed though
traditional means, where the analyst can isolate
separate, objective, entities.
The philosophical approach, which can clearly
states the problemfor the classical analyst,
Di Scipio music is only not-analysable it is
only sonic dustcan help musical analysis.
CIM08