Formalized Music

Formalized Music
THOUGHT AND MATHEMATICS IN COMPOSITION

Revised Edition
Iannis Xenakis
Additional material compiled and edited
by Sharon Kanach
HARMONOLOGIA SERIES No. 6
PENDRAGON PRESS
STUYVESANT NY
Formalized Music
THOUGHT AND MATHEMATICS IN COMPOSITION
Revised Edition
Iannis Xenakis
Additional material compiled and edited
by Sharon Kanach
HARMONOLOGIA SERIES No.6
PENDRAGON PRESS
STUYVESANT NY
Other Titles in the Harmonologia Series
No.1 Heinrich Schenker: Index to analysis by Larry
Laskowski (1978) ISBN 0-918728-06-1
No.2 Marpurg's Thoroughbass and Composition
Handbook: A narrative translation and critical study by
David A. Sheldon (1989) ISBN 0-918728-55-x
No.3 Between Modes and Keys: German Theory 1592-1802 by
Joel Lester (1990) ISBN 0-918728-77-0
No.4 Music Theory from Zarlino to Schenker: A Bibliography
and Guide by David Damschroder and David Russell
Williams (1991) ISBN 0-918728-99-1
No.5 Musical Time: The Sense of Order by Barbara R. Barry
(1990) ISBN 0-945193-01-7
Chapters I-VIII of this book were originally published in French.
Portions of it appeared in Gravesaner Blatter, nos. 1, 6, 9, 11/12, 18-
22, and 29 (1955-65).
Chapters I-VI appeared originally as the book Musiques Formeiles,
copyright 1963, by Editions Richard-Masse, 7, place Saint-Sulpice,
Paris. Chapter VII was first published in La NeJ, no. 29 (1967); the
English translation appeared in Tempo, no. 93 (1970). Chapter VII
was originally published in Revue d'Esthtftique, Tome XXI (1968).
Chapters IX and Appendices I and II Were added for the
edition by Indiana University Press, Bloomington
1971.
Chapters X, XI, XII, XIV, and AppendiJC III were added for this
edition, and all lists were updated to 1991.
Library of Congress Cataloging-Publication Data
Xenakis, Iannis, 1922-
Formalized music: thought and mathematics in
composition / lannis Xenakis. .
p. c.m. __ (Harmonologia senes ; no. 6)
"New expanded edition"--PreE
Includes bibliographical references and index.
ISBN 0-945193-24-6
1. Music--20th ccntury--Philosophy and aesthetics. 2.
Co
'f.J.' (MusiC) 3. Music--Theory--20th century. 4. mposl on
. . . . I
Music __ 20thcentury--History and cntlClsm. I. TItle. 1.
Series.
ML3800.X4 1990
781.3--dc20
90-41022
RP
L'P
Contents
Preface
Preface to Musiques formelles
Preface to the Pendragon Edition
I Free Stochastic Music
II Markovian Stochastic Music-Theory
III Markovian Stochastic Music-Applications
IV Musical Strategy
V Free Stochastic Music by Computer
VI Symbolic Music
Conclusions and Extensions for Chapters I-VI
VII Towards a Metamusic
VIn Towards a Philosophy of Music
IX New Proposals in Microsound Structure
X Concerning Time, Space and Music
XI Sieves
XII Sieves: A User's Guide
XIn Dynamic Stochastic Synthesis
XIV More Thorough Stochastic Music
Appendices
I & II Two Laws of Continuous Probability
III The New UPIC System
Bibliography
Discography
Biography: Degrees and Honors
Notes
Index
v
vii
ix
xi
1
43
79
110
131
155
178
180
201
242
255
268
277
289
295
323,32
7
329
335
365
371
373
383
Preface
The formalization that I attempted in trying to reconstruct part of the
musical edifice ex nihilo has not used, for want of time or of capacity, the
most advanced aspects of philosophical and scientific thought. But the
escalade is started an d others will certainly enlarge and extend the new thesis.
This book is addressed to a hybrid public, but interdisciplinary hybridiza-
tion frequently produces superb specimens.
I could sum up twenty years of personal efforts by the progressive
filling in of the following Table of Cohercnces. My musical, architectural,
and visual works are the chips of this mosaic. It is like a net whose variable
lattices capture fugitive virtualities and entwine them in a m u l t i t u ~ e of
ways. This table, in fact, sums up the true coherences of the successive
chronological chapters of this book. The chapters stemmed from mono-
graphs, which tried as much as possible to avoid overlapping.
But the profound lesson of such a table of coherences is that any
theory or solution given on one level can be assigned to the solution of
problems on another level. Thus the solutions in macrocomposition on the
Families level (programmed stochastic mechanisms) can engender simpler
and more powerful new perspectives in the shaping of micro sounds than the
usual trigonometric (periodic) functions can. Therefore, in considering
clouds of points and their distribution over a pressure-time plane, we can
bypass the heavy harmonic analyses and syntheses and create sounds that
have never bc.:fore existed. Only then will sound synthesis by computers and
digital-to-analogue converters find its true position, free of the rooted but
ineffectual tradition of electronic, concrete, and instrumental music that
makes use of Fourier synthesis despite the failure of this theory. Hence, in
this book, questions having to do mainly with orchestral sounds (which are
more diversified and more manageable) find a rich and immediate applica-
tion as soon as they are transferred to the Microsound level in the pressure-
time space. All music is thus automatically homogenized and unified.
vii
viii
Preface to the Second Edition
"Everything is everywhere'" h '
Coherences' Herakl 't Id
IS
t e word of thIs book and its Table of
, el os wou say th t tl
The French ed'( u' a le ways up and down are one,
1 IOn, ivmslques Fonnelles d d
Richard director of La R Lf"' ' was pro uce thanks to Albert
, evue iVluslcale The En r h d"
completed version res It r h' .,.. g IS e HIOn, a corrected and
, u s lrom t e InItIatIve of M CI' I
who translated the fi t' hr. lflStop ler Butchers,
Hopkins and M c apters, My thanks also go to Mr. G, W.
and VIII t rMs, Challifour, who translated Chapters VII
.' ,0 r. Ichael Aro d M B
IndIana University Pre } d' nson an r. ernard Perry of
ss, w 10 eCIded t bl' h .
Natalie Wrubel who d't d h' . 0 pu IS It; and finally to Mrs.
, e 1 e t IS dIfficult bo k . h . fi .
correcting and rephra . 0 WIt III Illite patience,
SIng many obscure passages.
1970 1. X,
TABLE (MOSAIC) OF COHERENCES
Philos(JphU (in the etymological sense)
Thrust towards truth, revelation Qu t'
creativity. . es In interrogation harsh c t' . .
elz . J rl IClsm, active knowledge through
apltrs <,m sense of the methods followed)
PartJaIly Inferential and experimental
Entire1y inferential and experimental
Other methods
ARTS (VISUAL, SONIC, MIXED )
SCIENCES (OF MAN, NATURAL) to con;e
This is Why the arts are li d PHYSICS, MATHEMATICS, LOGIC
o feer, an can therefore guide the sciences hO o.
Categolles of Queslrons (fragmentation of th dO. 0 ,W lch are entirely mferential and experimental.
e Irectlons leadIng to creativ k J
REALITY (EXISTENTlALJT'V)' CAU C now edge, to philosophy)
AS A CONSEQUENCE OF ' SAUTY j INFERENCE: CONNEXITY j COMPACTNE
NEW MENTAL STRUCTURES j INDETERMINISM AND SPATIAL UBIQUITY
FamUies of Solutions OT Proced:reJ (of th b . / . . . bl-po!e DETERMINISM j
e a OYe categones) ...
FREE STOCHASTIC MARKOVIAN GAMES

,eees (examples of particular realization)
materialized by
a computer pro-
gram
ACHORRIPSIS
'T/lO-I, 080262
'T/4B-l,240162
ATRBES
Clas if" S . EI MORSIMA'AMDRSiMA
ANALOGlQUE A DUEL
ANALOGIQ.UE B STRATEGIE
SYRMQS
GROUPS
AKRATA
NOMOS ALPHA
NOMOS GAMMA
us 0 onte ements (sounds that are heard and recog d
ELECTRONIC (produced by analogue as a whole, classified with respect to their Sources
M
o wIth computers and digital-to-analogue converters) S)J CONCRETE (microphone collected), DIGITAL
rcrosounds J
Forms and structur . th .
es In e pressure-time space recognition of h
microstructures produce ,t e classes to which microsounds belong h'
M' orw Ich
Icrosound types result from questions and solutions that
levels. were adopted at the CATEGORIES, FAMILIES, and PIEces
Preface to Musiques Formelles
This book is a collection of explorations in musical composition pursued in
several directions. The effort to reduce certain sound sensations, to under-
stand their logical causes, to dominate them, and then to use them in wanted
constructions; the effort to materialize movements of thought through sounds,
then to test them in compositions; the effort to understand better the pieces
of the past, by searching for an underlying unit which would be identical
with that of the scientific thought of our time; the effort to make "art"
while" geometrizing," that is, by giving it a reasoned support less perishable
than the impulse of the moment, and hence more serious, more worthy of
the fierce fight which the human intelligence wages in all the other domains
-all these efforts have led to a sort of abstraction and formalization of the
musical compositional act. This abstraction and formalization has found,
as have so many other sciences, an unexpected and, I think, fertile support
in certain areas of mathematics. It is not so much the inevitable use of
mathematics that characterizes the attitude of these experiments, as the
overriding need to consider sound and music as a vast potential reservoir in
which a knowledge of the laws of thought and the structured creations of
thought may find a completely new medium of materialization, i.e., of
communication,
For this purpose the qualification" beautiful" or "ugly" makes no
sense for sound, nor for the music that derives from it; the quantity of
intelligence carried by the sounds must be the true criterion of the validity
of a particular music.
This does not prevent the utilization of sounds defined as pleasant or
beautiful according to the fashion of the moment, nor even their study in
their own right, which may enrich symbolization and algebration, Efficacy
is in itself a sign of intelligence. Weare so convinced of the historical
necessity of this step, that we should like to see the visual arts take an
ix
x
Preface to Musiques Formelles
analogous path-unless that is "art' "
done it in laboratories 'I It d' r' IStS. of a new type have not already
T ' S Ie ere lrom 1l00SY publicit '
hese studies have alwa S b . ),
out thc various stages M Y C : I ~ matched by actual works which mark
. . Y COmpOSHlOns Cons!'t t I . .
of tillS undertaking I lb" 1 U e t Ie Cxpcnmcntal dOSSIer
, n tIe cgmnmg m ..
recognized and puiJI' h d I Y compOSItIons and research were
IS e , t lanks to tl fi' d h'
Support of Prof Herma S I h IC nen S Ip and moral and material
, nn Clerc en Ce t'} .
reflect the results of the t I' . ~ am c lapte
r
s m the present work
eac llng of certaIn
and Olivier Messiaen in . d masters, such as H. Scherchen
h
musIc, an Prof: G TI G 'lb .
w 0, through the virtuosit d 1'1 ." 1. UI aud 111 mathematics,
I
Y an I Jerality of h' th I .
c earer VIew of the I I' IS oug It, has gIven me a
age )ras w}uch const't t I '
devoted to Symbolic Music, 1 u e tIe fabnc of the chapter
1962
I. X.
Preface to the Pendragon Edition
Here is a new expanded edition of Formalized Music. It invites two fundamen-
tal questions:
Have the theoretical propositions which I have made over the past
thirty-five years
a) survived in my music?
b) been aestJIetically efficient?
To the first question, I will answer a general "yes." The theories which I
have presented in the various chapters preceding this new edition have
always been present in my music, even if some theories have been mingled
with others in a same work. The exploration of the conceptual and sound
world in which I have been involved necessitated an harmonious or even
conflicting synthesis of earlier theses. It necessitated a more global
architectural view than a mere comparative confrontation of the various
procedures. But the supreme criterion always remained the validation, tJIe
aesthetic efficiency of the music which resulted.
Naturally, it was up to me and to me alone to determine the aesthetic
criteria, consciously or not, in virtue of the first principle which one can not
get around. The artist (man) has the duty and the privilege to decide, radically
alone, his choices and the value of the results. By no means should he choose
any ot11er means; those of power, glory, money, ...
Each time, he must throw himself and his chosen criteria into question
all while striving to start from scratch yet not forget. We should not
"monkey" ourselves by virtue of the habits we so easily acquire due to our
own "echolalic" properties. But to be reborn at each and every instant, like a
child with a new and "independent" view of things.
All of this is part of a second principle: It is absolutely necessary to free
oneself, as much as possible, from any and all contingencies.
xi
xu
Preface
. This may be considered man's destiny in particular, and the universe's
In general. the Being's constant dislocations, be they continuous or
not, or chaotic (or both simultaneously) are manifestations of
the vItal and Incessant drive towards change, towards freedom without
return.
An artist can not remain isolated in the universal ocean of forms and
their changes. His interest lies in embracing the most vast horizon of
knowledge and problema tics, all in accordance with the two principles
presented above. From hence comes tlle new chapter in this edition entitled
"Concerning Time, Space and Music."
Finally, to finish wit11 t11e first question, I have all along continued to
develop certain theses and to open up some new ones. The new chapter on
"Sieves" is a am I ftl' I .L
. n ex. p e 0 liS a ong WIUl the computer program presented in
AppendIX III whIch represents a long aesthetic and theoretical search. This
research was developed as well as its application in sound synt11esis on
UPIC.*
Anotiler approach to the mystery of sounds is t11e use of cellular
automata which I employed in several instrumental compositions these
past few years. ThIS can be explained by an observation which I made: scales
of pitch (sieves) automatically establish a kind of global musical style. a sort of
"syntilcsis" of musical works, much like a "spectrum of
or ite.rations," of the physics of particles. Intemal symmetries or
d.1Ssymmetnes are the reason behind this. Therefore, tllrough a
of "non-octave" scales, we can obtain very
nch slmultaneltles (chords) or lmear successions which revive and generalize
tonal, modal or serial aspects. It is on t1lis basis of sieves that cellular
automata can be useful in harmonic progressions which create new and rich
timbric fusions with orchestral instruments. Examples of this can be found in
works of mine such asAta, Roros, etc.
Today, there is a whole new field of investigation called "Experimental
Mathematics," that gives fascinating insights especially in automatic dynamic
systems, by the use of math and computer graphics. Thus, many structures
such as t1le already- mentioned cellular automata or those which possess se1f-
*UPIC-l!nite Polygogique du CEMAMu. A sort of musical drawing
which, the digitalizatIon of a drawing, enables one to compose
mUSIC, teach acoustiCS, engage in musical pedagogy at any age. This machine was
dev:loped at the Centre d'Etudes de Mathematiqucs et Automatiques Musicales de
Pans.
Preface xiii
similarities such as Julia or Mandeibrot sets, are studied and visualized. These
studies lead one right into the frontiers of determinism and indeterminism.
Chaos to symmetry and the reverse orientation are once again being studied
and are even quite fashionable! They open up new horizons, although for
me, the results are novel aspects of the equivalent compositional problems I
started dealing with about thirty-five years ago. The theses presented in the
earlier editions of this book bear witness to tJlis fact although the dynamic of
musical works depends on several levels simultaneously and not only on tile
calculus level.
An important task of tJle research program at CEMAMu is to develop
synthesis through quantified sounds but with up-to-date tools capable of
involving autosimilitudes, symmetries or deterministic chaos, or stochastics
within a dynamic evolution of amplitude frequency frames where each pixel
corresponds to a sound quantum or "phonon," as already imagined by
Einstein in the 19105. This research, which I started in 1958 and wrongly
attributed to Gabor, can now be pursued with much more powerful and
modern means. Some surprises can be expected!
In Appendix IV of this edition, a new, more precise formulation of
stochastic sound synthesis can be found as a follow-up of tile last chapter of
the preceding edition of Formalized Music (presented here as Chapter IX). In
the interim, this approach has been tested and used in my work La Ugerule
d'Eer for seven-track tape. This approach was developed at t11e CEMAMu in
Paris and worked out at the WDR, tile West-German National Radio studio
in Cologne. This work was part of the Diatope which was installed for t11e
inauguration of the Pompidou/Beaubourg Center in Paris. The event was
entirely automated with a complete laser installation and 1600 electronic
flashes. This synt1lesis is part of CEMAMu's permanent research program.
In t11is same spirit, random walks or Brownian movements have been
the basis for several of my works, especially instrumental pieces such as
N'Shirna, which means "breath" or "spirit" in Hebrew; for 2 female voices, 2
French Horns, 2 trombones and 1 cello. This piece was written at the request
of Recha Freier, founder of the "Aliya movement" and premiered at t11e
Testimonium Festival in Jerusalem.
The answer to t11e second question posed at the beginning of this
Preface is not up to me. In absolute terms, the artisan musician (not to say
creator) must remain doubtful of the decisions he has made, doubtful.
however subtly, of the result. The percentage of doubt should not exist in
virtue ofthe principles elaborated above. But in relative terms, the public, or
connoisseurs (either synchronic or diachronic), alone decide upon a work's
XlV Preface
efficiency. However, any culture's validation follows "seasonal" rules, varying
between periods of a few years to centuries or even millennia. We must never
forget the nearly-total lack of consideration Egyptian art suffered for over
2000 years, or Meso-American art.
One can assimilate a work of art, or, let us say, just a work, to the
information we can put on a document, seal in a bottle which we will throw
into the middle of the ocean. Will it ever be found? When and by whom and
how will it be read, interpreted?
My gratitude and thanks go to Sharon Kanach, who translated and
supervised the new material in this updated edition of Formalized Music and to
Robert Kessler, the courageous publisher.
Formalized Music
fo
f1
-".
-'!it-
-,,4
Preliminary sketch Analogique B (1959). See Chapter III,
pp.l03-99.
.2
-:I.Z..
(MrS
..., _h Jil
Preliminary sketch Analogique B (1959). See Chapter III, pp.
103-09.
Chapter I
Free Stochastic Music
Art, and above all, music has a fundamental function, which is to catalyze
the sublimation that it can bring about through all means of expression. It
must aim through fixations which are landmarks to draw towards a total
exaltation in which the individual mingles, losing his consciousness in a
truth immediate, rare, enormous, and perfect. If a work of art succeeds in
this undertaking even for a single moment, it attains its goal. This tremen-
dous truth is not made of objects, emotions, or sensations; it is beyond these,
as Beethoven's Seventh Symphony is beyond music. This is why art can lead
to realms that religion still occupies for some people.
But this transmutation of every-day artistic material which transforms
trivial products into meta-art is a secret. The;" possessed" reach it without
knowing its" mechanisms." The others struggle in the ideological and tech-
nical mainstream of their epoch which constitutes the perishable "climate"
and the stylistic fashion. Keeping our eyes fixed on this supreme meta-artistic
goal, we shall attempt to define in a more modest manner the paths which
can lead to it from our point of departure, which is the magma of contra-
dictions in present music.
There exists a historical parallel between European music and the
successive attempts to explain the world by reason. The music of antiquity,
causal and deterministic, was already strongly influenced by the schools of
Pythagoras and Plato. Plato insisted on the principle of causality, "for it is
impossible for anything, to come into being without cause" (Timaeus).
Strict causality lasted until the nineteenth century when it underwent a
The English translation of Chaps. I-VI is by Christopher A. Butchers.
YI
J .,
Fig. 1-1. Score of Metastasis, 1953/54, Bars 309-17
__'0-.
:.'
__ 'I!"- __ __ -0
"' - '."
. __

h)

=
>
..
.:
]cc
>U
'C .,
" .. o ..
u '"

N "
<5
>
" t:
::I
o
" <0
;!
.....

NUl

(P
I !
I
I, "-
\ \:
\ 1
\'\7 L.
I
"!? I

I
I
;;.- - .. - T
- .. -I
'"
i
;
.
... -:- ,,':
\!
\ '/
)1'
\::1

"'I!!0 "", .. ,:-
\r"
I

:f

'Il'
.r
..
,;
i-

\
tt
I
.\
\\,,\0

'"
..,

::!:
'0
'<t
J;
0
I
M
!!!
I
co
CD
,,' '5
I
c:
'"
'"
.!!!
4-
c:;
Cl

c:
.;::
1 ;.
Iii
N
I
J.
-1
til
u:
I
I'
il
'i
II
I
4 Formalized Music
brutal and fertile transformation as a result of statistical theories in physics.
Since antiquity the concepts of chance (tyche) , disorder (ataxia), and dis-
organization were considered as the opposite and negation of reason (logos),
order (taxis), and organization (systasis). It is only recently that knowledge
has been able to penetrate chance and has discovered how to separate its
degrees-in other words to rationalize it progressively, without, however,
succeeding in a definitive and total explanation of the problcm of "pure
chance."
After a time lag of several decades, atonal music broke up the tonal
function and opened up a new path parallel to that of the physical sciences,
but at the same time constricted by the virtually absolute determinism of
serial music.
It is therefore not surprising that the presence or absence of the principle
of causality, first in philosophy and then in the sciences, might influence
musical composition. It caused it to follow paths that appeared to be diver-
gent, but which, in fact, coalesced in probability theory and finally in
polyvalent logic, which arc kinds of generalization and enrichments of the
principle of causality. The explanation of the world, and consequently of the
sonic phenomena which surround us or which may be created, necessitated
and profited from the enlargement of the principle of causality, the basis of
which enlargement is formed by the law of large numbers. This law implies
an asymptotic evolution towards a stable state, towards a kind of goal, of
stach as, whence comes the adjective" stochastic."
But everything in pure determinism or in less pure indeterminism is
subjected to the fundamental operational laws of logic, which were disen-
tangled by mathematical thought under the title of general algebra.
These laws operate on isolated states or on sets of elements with the aid of
operations, the most primitive of which arc the union, notated U, the
intersection, notated n, and the negation. Equivalence, implication, and
quantifications are elementary relations from which all current science can
be constructed.
Music, then, may be defined as an organization of these elementary
operations and relations between sonic entities or between functions of
sonic entities. We understand the first-rate position which is occupied by
set theory, not only for the construction of new works, but also for analysis
and better comprehension of the works of the past. In the same way a
stochastic construction or an investigation of history with the help of
stochastlcs cannot be carried through without the help of logic-the queen
of the sciences, and I would even venture to suggest, of the arts-or its
mathematical form algebra. For everything that is said here on the subject
Free Stochastic Music 5
is also valid for all forms of art (painting, sculpture, architecture, films,
etc.).
From this very general, fundamental point of view, from which we wish
to examine and make music, primary time appears as a wax or clay on which
operations and relations can be inscribed and engraved, first for the purposes
of work, and then for communication with a third person. On this level, the
asymmetric, noncommutative character of time is use (B after A of- A after
B, i.e., lexicographic order). Commutative, metric time (symmetrical) is
subjected to the same logical laws and can therefore also aid organizational
speculations. What is remarkable is that these fundamental notions, which
are necessary for construction, are found in man from his tenderest age, and
it is fascinating to follow their evolution as Jean Piaget
1
has done.
After this short preamble on generalities we shall enter into the details
of an approach to musical composition which I have developed over several
years. I call it "stochastic," in honor of probability theory, which has served
as a logical framework and as a method of resolving the conflicts and knots
encountered.
The first task is to construct an abstraction from all inherited conven-
tions and to exercise a fundamental critique of acts of thought and their
materialization. What, in fact, does a musical composition offer strictly on
the construction level? I t offers a collection of sequences which it wishes to
be causal. When, for simplification, the major scale implied its hierarchy of
tonal functions-tonics, dominants, and subdominants-around which the
other notes gravitated, it constructed, in a highly deterministic manner,
linear processes, or melodies on the one hand, and simultaneous events, or
chords, on the other. Then the serialists of the Vienna school, not having
known how to master logically the indeterminism of atonality, returned to
an organization which was extremely causal in the strictest sense, more ab-
stract than that of tonality; however, this abstraction was their great con-
tribution. Messiaen generalized this process and took a great step in sys-
tematizing the abstraction of all the variables of instrumental music. What
is paradoxical is that he did this in the modal field. He created a multimodal
music which immediately found imitators in serial music. At the outset
Messiaen's abstract systematization found its most justifiable embodiment
in a multiserial music. It is from here that the postwar neo-serialists have
drawn their inspiration. They could now, following the Vienna school and
Messiaen, with some occasional borrowing from Stravinsky and Debussy,
walk on with ears shut and proclaim a truth greater than the others. Other
movements were growing stronger; chief among them was the systematic
exploration of sonic entities, new instruments, and" noises." Varese was the
151 Peak
1st Peak
A. Ground profile of the left half of
the "stomach." The intention was to
build a shell, composed of as few
ruled surfaces as possible, over the
ground plan. A conoid (e) is con-
structed through the ground profile
curve; this wall is bounded by two
straight lines: the straight directrix
(rising from the left extremity of the
ground profile), and the outermost
generatrix (passing through the right
extremity of the ground profile). This
produces the first "peak" of the
pavilion.
B. A ruled surface consisting of two
con aids, a and d, is laid through the
curve bounding the right half of the
"stomach." The straight directrix of d
passes through the first peak, and the
outermost generatrix at this side forms
a triangular exit with the generatrix of
e. The straight directrix of a passes
through a second peak and is joined
by an arc to the directrix of d.
This basic form is the one used in
the first design and was retained, with
some modifications, in the final
structure. The main problem of the
design was to establish an aesthetic
balance between the two peaks.
C. Attempt to close the space between
the two ruled surfaces of the first
design by flat surfaces (which might
serve as projection w.alls).
Fig. 1-3. Stages in the Development of the First Design of the
Philips Pavilion
D. Another attempt. Above Ihe
entrance channel a small triangular
opening is formed. flanked by two
hyperbolic paraboloids (g and k), and
the whole is covered with a horizontal
top surface.
E. Elaboration of D. The third peak
begins to take shape (shyly).
3rd Peak
F. The first design completed (see
also the first model. Fig. 1-4). There
are no longer any flat surfaces. The
third peak is fully developed and
creates, with its opposing sweep, a
counterbalance for the first two peaks.
The heights of the three peaks have
been established. The third peak and
the small arc connecting the straight
directrixes of conolds a and d (see B.)
form, respectively, the apex and the
base of a part of a cone I.
151 Peak
2nd Peak
1st Peak
2nd Peak
1 st Peak
2nd Peak
8 Formalized Music
pioneer in this field, and electromagnetic music has been the beneficiary
(electronic music being a branch of instrumental music). However, in
electromagnetic music, problems of construction and of morphology were not
faced conscientiously. Multiserial music, a fusion of the multimodality of
Messiaen and the Viennese school, remained, nevertheless, at the heart of
the fundamental problem of music.
But by 1954 it was already in the process of deflation, for the completely
deterministic complexity of the operations of composition and of the works
themselves produced an auditory and ideological nonsense. I described the
inevitable conclusion in "The Crisis of Serial Music":
Linear polyphony destroys itselfby its very complexity; ll'hat one
hears is in reality nothing but a mass of notes in various registers. The
enormous complexity prevents the audience from following the inter-
twining of the lines and has as its macroscopic effect an irrational
and fortuitous dispersion of sounds over the whole extent of the sonic
spectrum. There is consequently a contradiction between the poly-
phonic linear system and the heard result, which is surface or mass.
This contradiction inherent in polyphony will disappear when the
independence of sounds is total. In fact, when linear combinations
and their polyphonic superpositions no longer operate, what will
count will be the statistical mean of isolated states and of transform a-
tions of sonic components at a given moment. The macroscopic effect
can then be controlled by the mean of the movements of elements
which we select. The result is the introduction of the notion ofproba-
bility, which implies, in this particular case, combinatory calculus.
Here, in a few words, is the possible escape route from the "linear
category" in musical thought.
2
This article served as a bridge to my introduction of mathematics in
music. For if, thanks to complexity, the strict, deterministic causality which
the neo-serialists postulated was lost, then it was necessary to replace it by
a more general causality, by a probabilistic logic which would contain strict
serial causality as a particular case. This is the function of stochastic science.
"Stochastics" studies and formulates the law of large numbers, which has
already been mentioned, the laws of rare events, the different aleatory
procedures, etc. As a result of the impasse in serial music, as well as other
causes, I originated in 1954 a music constructed from the principle of
indeterminism; two years later I named it "Stochastic Music." The laws
of the calculus of probabilities entered composition through musical
necessity.
But other paths also led to the same stochastic crossroads-first of all,
natural events such as the collision of hail or rain with hard surfaces, or the
song of cicadas in a summer field. These sonic events are made out of thou-
sands of isolated sounds; this multitude of sounds, seen as a totality, is a new
sonic event. This mass event is articulated and forms a plastic mold of time,
which itself follows aleatory and stochastic laws. If one then wishes to form a
large mass of point-notes, such as string pizzicati, one must know these
mathematical laws, which, in any case, are no more than a tight and concise
expression of chain of logical reasoning. Everyone has observed the sonic
phenomena of a political crowd of dozens or hundreds of thousands of
people. The human river shouts a slogan in a uniform rhythm. Then another
slogan springs from the head of the demonstration; it spreads towards the
tail, replacing the first. A wave of transition thus passes from the head to the
tail. The clamor fills the city, and the inhibiting force of voice and rhythm
reaches a climax. It is an event of great power and beauty in its ferocity.
Then the impact between the demonstrators and the enemy occurs. The
perfect rhythm of the last slogan breaks up in a huge cluster of chaotic
shouts, which also sprcads to the tail. Imagine, in addition, the reports of
dozens of machine guns and the whistle of bullets addipg their punctuations
to this total disorder. The crowd is then rapidly dispersed, and after sonic
and visual hell follows a detonating calm, full of despair, dust, and death.
The statistical laws of these events, separated from their political or moral
contcxt, are the same as those ofthc cicadas or the rain. Thcy are the laws of
the passage from complete order to total disorder in a continuous or explo-
sive manner. They are stochastic laws.
Here we touch on one of the great problems that have haunted human
intelligence since antiquity: continuous or discontinuous transformation.
The sophisms of movement (e.g., Achilles and the tortoise) or of definition
(e.g., baldness), especially the latter, are solved by statistical definition; that
is to say, by stochastics. One may produce continuity with either continuous
or discontinuous elements. A multitude of short glissandi on strings can give
the impression of continuity, and so can a multitude of pizzicati. Passages
from a discontinuous state to a continuous state are controllable with the
aid of probability theory. For some time now I have been conducting these
fascinating experiments in instrumental works; but the mathematical char-
acter of this music has frightened musicians and has made the approach
especially difficult.
Here is another direction that converges on indeterminism. The study
of the variation of rhythm poses the problem of knowing what the limit of
total asymmetry is, and of the consequent complete disruption of causality
among durations. The sounds of a Geiger counter in the proximity of a
lO Formalized Music
radioactive source give an impressive idea of this. Stochastics provides the
necessary laws.
Before ending this short inspection tour of events rich in the new logic,
which were closed to the understanding until recently, I would like to in-
clude a short parenthesis, If glissandi are long and sufficiently interlaced,
we obtain sonic spaces of continuous evolution, It is possible to produce
ruled surfaces by drawing the glissandi as straight lines. I performed this
experiment with Metastasis (this work had its premiere in 1955 at Donau-
eschingen). Several years later, when the architect Le COI'busier, whose
collaborator I was, asked me to suggest a design for the architecture of the
Philips Pavilion in Brussels, my inspiration was pin-pointed by the experi-
ment with Metastasis. Thus I believe that on this occasion music and archi-
tecture found an intimate connection.
3
Figs. 1-1-5 indicate the causal chain
of ideas which led me to formulate the architectlJrc of the Philips Pavilion
from the score of Metastasis.
Fig. 1-4. First Model of Philips Pavilion
Fig. 1-5. Philips Pavilion, Brussels World's Fair, 1958
12 Formalized Music
STOCHASTIC LAWS AND INCARNATIONS
I shall give quickly some of the stochastic laws which I introduced into
composition several years ago. We shall examine one by one the independent
components of an instrumental sound.
DURATIONS
Time (metrical) is considered as a straight line on which the points
corresponding to the variations of the other components are marked. The
interval between two points is identical with the duration. Among all the
possible sequences of points, which shall we choose? Put thus, the question
makes no sense.
If a mean number of points is designated on a given length the question
becomes: Given this mean, what is the number of segments equal to a length
fixed in advance?
The following formula, which derives from the principles of continuous
probability, gives the probabilities for all possible lengths when one knows
the mean number of points placed at random on a straight line.
(See Appendix 1.)
in which (j is the linear density of points, and x the length of any segment.
If we now choose some points and compare them to a theoretical
distribution obeying the above law or any other distribution, we can deduce
the amount of chance included in our choice, or the more or less rigorous
adaptation of our choice to the law of distribution, which can even be
absolutely functional. The comparison can be made with the aid of tests,
of which the most widely used is the X
2
criterion of Pearson. In our case,
where all the components of sound can be measured to a first approxima-
tion, we shall use in addition the correlation coefficient. It is known that if
two populations are in a linear functional relationship, the correlation
coefficient is one. If the two populations are independent, the coefficient is
zero. All intermediate degrees of relationship are possible.
Clouds of Sounds
Assume a given duration and a set of sound-points defined in the
intensity-pitch space realized during this duration. Given the mean super-
ficial density of this tone cluster, what is the probability of a particular
density occurring in a given region of the intensity-pitch space? Poisson's
Law answers this question:
" p = fLo e-#o
" , ' p..
where fLo is the mean density and fL is any particular density. As with
durations, comparisons with other distributions of sound-points can fashion
the law which we wish our cluster to obey.
INTERVALS OF INTENSITY, PITCH, ETC.
For these variables the simplest law is
ely) dy = ~ (I - ~ ) dy, (See Appendix 1.)
which gives the probability that a segment (interval of intensity, pitch, etc.)
within a segment of length a, will have a length included within y and
y + dy, for O:=;y:=;a.
SPEEDS
We have been speaking of sound-points, or granular sounds, which are
in reality a particular case of sounds of continuous variation. Among these
let us consider glissandi. Of all the possible forms that a glissando sound can
take, we shall choose the simplest-the uniformly continuous glissando. This
glissando can be assimilated sensorially and physically into the mathematical
concept of speed. In a one-dimensional vectorial representation, the scalar
size of the vector can be given by the hypotenuse of the right triangle in
which the duration and the melodic interval covered form the other two
sides. Certain mathematical operations on the continuously variable sounds
thus defined are then permitted. The traditional sounds of wind instruments
are, for example, particular cases where the speed is zero. A glissando
towards higher frequencies can be defined as positive, towards lower fre-
quencies as negative.
We shall demonstrate the simplest logical hypotheses which lead us
to a mathematical formula for the distribution of speeds. The arguments
which follow are in reality one of those" logical poems" which the human
intelligence creates in order to trap the superficial incoherencies of physical
phenomena, and which can serve, on the rebound, as a point of departure
for building abstract entities, and then incarnations of these entities In
sound or light. It is for these reasons that I offer them as examples:
Homogeneity hypotheses [11] *
1. The density of speed-animated sounds is constant; i.e., two regions
of equal extent on the pitch range contain the same average number of
mobile sounds (glissandi).
* The numbers in brackets correspond to the numbers in the Bibliography at the
end of the book.
14 Formalized Music
2. The absolute value of speeds (ascending or descending glissandi) is
spread uniformly; i.e., the mean quadratic speed of mobile sounds is the
same in different registers.
3. There is isotropy; that is, there is no privileged direction for the
movements of mobile sounds in any register. There is an equal number of
sounds ascending and descending.
From these three hypotheses of symmetry, we can define the function
f(v) of the probability of the absolute speed v. (f(v) is the relative frequency
of occurrence of v.)
Let n be the number of glissandi per unit of the pitch range (density
of mobile sounds), and r any portion taken from the range. Then the number
of speed-animated sounds between v and v + du and positive, is, from
hypotheses 1 and 3:
n r1f(u) dv (the probability that the sign is + is
From hypothesis 2 the number of animated sounds with speed of
absolute value Iv I is a function which depends on v
2
only. Let ulis function
be g(v
2
). We then have the equation
n r tf(u) du = n r g(v
2
) duo
Moreover if x = D, the probability function g(v
2
) will be equal to the law
of probability HoLt, when.ce g(v
2
) = H(x), or log g(v
2
) = hex).
In order that hex) may depend only on x
2
= v
2
, it is necessary and
sufficient that the differentials d log g( v
2
) = h' (x) dx and u dv = x dx have a
constant ratio:
dlog g(v
2
) = h'(x
d
) dx = constant = _2j,
v dv x x
whence h'(x) = - 2jx, hex) = - jx
2
+ c, and H(x) = ke-
jx2

But H(x) is a function of elementary probabilities; therefore its integral
from -00 to +00 must be equal to 1. j is positive and k = vJ/V7T. If
J = I Ja
2
, it follows that
I _ v'la2
ti(v) = g(v
2
) = H(x) = aV
7T
e
and
i( )
2 _v2 1a2
V = --e
aV7T
for v = x, which is a Gaussian distribution.
This chain of reasoning borrowed from Paul Levy was established after
Maxwell, who, with Boltzmann, was responsible for the kinetic theory of
gases. The function i( v) gives the probability of the speed v; the constant a
defines the" temperature" of this sonic atmosphere. The arithmetic mean
of v is equal to a/"/7T, and the standard deviation is a/,/2.
vVe offer as an example several bars from the work Pithoprakta for string
orchestra (Fig. 1-6), written in 1955-56, and performed by Prof. Hermann
Scherchen in Munich in March 1957.
4
The graph (Fig. 1-7) represents a
set of speeds of temperature proportional to a = 35. The abscissa represents
time in units of 5 cm = 26 MM (Malzel Metronome). This unit is sub-
divided into three, four, and five equal parts, which allow very slight
differences of duration. The pitches are drawn as the ordinates, with the
unit 1 semi tone = 0.25 em. I cm on the vertical scale corresponds to a major
third. There are 46 stringed instruments, each represented by a jagged line.
Each of the lines represents a speed taken from the table of probabilities
calculated with the formula
i( )
- _2_ -v
2
,a'
V - e
aV7T
A total of 1148 speeds, distributed in 58 distinct values according to Gauss's
law, have been calculated and traced for this passage (measures 52-60, with
a duration of 18.5 sec.). The distribution being Gaussian, the macroscopic
configuration is a plastic modulation of the sonic material. The same passage
was transcribed into traditional notation. To sum up we have a sonic
compound in which:
I. The durations do not vary.
2. The mass of pitches is freely modulated.
3. The density of sounds at each moment is constant.
4. The dynamic isffwithout variation.
5. The timbre is constant.
6. The speeds determine a "temperature" which is subject to local
fluctuations. Their distribution is Gaussian.
As we have already had occasion to remark, we can establish more or
less strict relationships between the component parts of sounds.
5
The most
useful coefficient which measures the degree of correlation between two
variables x and y is
2: (x - x)(y - y)
r = )-;'-2v'--':':=2::-;(-=Y-'-_'"""'y""") 2 '
16 Formalized Music
where x and fi are the arithmetic means of the two variables.
Here then, is the technical aspect of the starting point for a utilization
of the theory and calculus of probabilities in musical composition. With
the above, we already know that:
l. We can control continuous transformations oflarge sets of granular
and/or continuous sounds. In fact, densities, durations, registers, speeds,
etc., can all be subjected to the law of large numbers with the necessary
approximations. We can therefore with the aid of means and deviations
shape these sets and make them evolve in different directions. The best
known is that which goes from order to disorder, or vice versa, and which
introduces the concept of entropy. We can conceive of Qther continuous
transformations: for example, a set of plucked sounds transforming con-
tinuously into a set of arco sounds, or in electromagnetic music, the passage
from one sonic substance to another, assuring thus an organic connection
between the two substances. To illustrate this idea, I recall the Greek
sophism about baldness: "How many hairs must one remove from a hairy
skull in order to make it bald?" It is a problem resolved by the theory of
probability with the standard deviation, and known by the term statistical
definition.
2. A transformation may be explosive when deviations from the mean
suddenly become exceptional.
3. We can likewise confront highly improbable events with average
events.
4. Very rarified sonic atmospheres may be fashioned and controlled
with the aid of formulae such as Poisson's. Thus, even music for a solo
instrument can be composed with stochastic methods.
These laws, which we have met before in a multitude of fields, are
veritable diamonds of contemporary thought. They govern the laws of the
advent of being and becoming. However, it must be well understood that
they are not an end in themselves, but marvelous tools of construction and
logical lifelines. Here a backfire is to be found. This time it is these stochastic
tools that pose a fundamental question: "What is the minimum oflogical
constraints necessary for the construction ofa musical process?" But before
answering this we shall sketch briefly the basic phases in the construction of
a musical work.
W.81.
[EJ
I ----I
V.I
v.n
r--- S ---:X' 5J r,-.

A.
tmi
v,.
\0)
..

8, lH. 19'56'
Fig. 1-6. Bars 52-57 of Pithoprakta
18 Formalized Music
Free Stochastic rvIusic
19
Fig. 1-7. Graph of Bars 52-57 of Pithoprakta
20 Formalized Music Free Stochastic Music
21
Fig. 1-7 (continued)
22
Formalized Music
FUNDAMENTAL PHASES OF A MUSICAL WORK
l. Initial conceptions (intuitions, provisional or definitive data) ;
2. Difinition of the sonic entities and of their symbolism communicable
with the limits of possible means (sounds of musical instruments, electronic
sounds, noises, sets of ordered sonic elements, granular or continuous
formations, etc.);
3. Difinition of the transformations which these sonic entities must undergo
in the course of the composition (macrocomposition: general choice of
logical fi-amework, i.e., of the elementary algebraic operations and the set-
ting up of relations between entities, sets, and their symbols as defined in
2.); and the arrangement of these operations in lexicographic time with the
aid of succession and simultaneity) ;
4. Microcomposition (choice and detailed fixing of the functional or
stochastic relations of the elements of 2.), i.e., algebra outside-time, and
algebra in-time;
5. Sequential programming of 3. and 4. (the schema and pattern of the
work in its entirety);
6. Implementation of calculations, verifications, feedbacks, and definitive
modifications of the sequential program;
7. Final symbolic result of the programming (setting out the music on
paper in traditional notation, numerical expressions, graphs, or other means
of solfeggio) ;
8. Sonic realization of the program (direct orchestral performance,
manipulations of the type of electromagnetic music, computerized construc-
tion of the sonic entities and their transformations).
The order of this list is not really rigid. Permutations are possible in
the course of the working out of a composition. Most of the time these
phases are unconscious and defective. However, this list does establish ideas
and allows speculation about the future. In fact, computers can take in hand
phases 6. and 7., and even 8. But as a first approach, it seems that only
phases 6. and 7. are immediately accessible. That is to say, that the final
symbolic result, at least in France, may be realized only by an orchestra or
by manipulations of electroacoustic music on tape recorders, emitted by the
existing electroacoustic channels; and not, as would be desirable in the very
ncar future, by an elaborate mechanization which would omit orchestral or
tape interpreters, and which would assume the computerized fabrication of
the sonic entities and of their transformations.
Here now is an answer to the question put above, an answer that is
true for instrumental music, but which can be applied as well to all kinds of
sound production. For this we shall again take up the phases described:
2. Definition of sonic entities. The sonic entities of the classical orches-
tra can be represented in a first approximation by vectors of four usually
independent variables, Er(c, h, g, u) :
e
a
= timbre or instrumental family
hi = pitch of the sound
gj = intensity of the sound, or dynamic form
UJc = duration of the sound.
The vector Er defines a point M in the multidimensional space provided
by a base (c, h, g, u). This point M will have as coordinates the numbers
C
a
, hi, gj, Uk' For example: C
3
played areo and forte on a violin, one eighth
note in length, at one eighth note = 240 MM, can be represented as
(viOl.aTeo, h39 (= c
3
), g4 (= forte), U
5
(= -l- sec.). Suppose that these points
.tv! are plotted on an axis which we shall call En and that through its origin
we draw another axis t, at right angles to axis E
r
We shall represent on this
axis, called the axis of lexicographic time, the lexicographic-temporal succession
of the points M. Thus we have defined and conveniently represented a
two-dimensional space (En t). This will allow us to pass to phase 3., defini-
tion of transformation, and 4., microcomposition, which must contain the
answer to the problem posed concerning the minimum of constraints.
To this end, suppose that the points M defined above can appear with
no necessary condition other than that of obeying an aleatory law without
memory. This hypothesis is equivalent to saying that we admit a stochastic
distribution of the events Er in the space (En t). Admitting a sufficiently
weak superficial distribution n, we enter a region where the law of Poisson
is applicable:
Incidentally we can consider this problem as a synthesis of several
conveniently chosen linear stochastic processes (law of radiation from radio-
active bodies). (The second method is perhaps more favorable for a mecha-
nization of the transformations.)
A suIIiciently long fragment of this distribution constitutes the musical
work. The basic law defined above generates a whole family of compositions
as a function of the superficial density. So we have a formal archetype of
composition in which the basic aim is to attain the greatest possible asym-
metry (in the etymological sense) and the minimum oj constraints, causalities,
and rules. We think that from this archetype, which is perhaps the most
24
Formalized Music
general one, we can redescend the ladder of forms by introducing progres-
sively more numerous constraints, i.e., choices, restrictions, and negations.
In the analysis in several linear processes we can also introduce other pro-
cesses: those of Wiener-Levy, P. Levy's infinitely divisibles, Markov
chains, etc., or mixtures of several. It is this which makes this second
method the more fertile.
The exploration of the limits a and b of this archetype a :<; n :<; b is
equally interesting, but on another level-that of the mutual comparison of
samples. This implies, in effect, a gradation of the increments of n in order
that the differences between the families n
f
may be recognizable. Analogous
remarks arc valid in the case of other linear processes.
If we opt for a Poisson process, there are two necessary hypotheses
which answer the Ruestion of the minimum of constraints: 1. there exists in
a given space musical instruments and men; and 2. there exist means of
contact between these men and these instruments which permit the emission
of rare sonic events.
This is the only hypothesis (cf. the ekklisis of Epicurus). From these
two constraints and with the aid ofstochastics, I built an entire composition
without admitting any other restrictions. Achorripsis for 21 instruments was
composed in 1956-57, and had its first performance in Buenos Aires in 1958
under Prof. Hermann Scherchen. (Sec Fig. 1-8.)
At that time I wrote:*
\ \ , \ ... ." \ 1"
TO yap aUTO voeu' unv Tt Kat eLvaL
T6 yap aUTO eivat UT{V Te Kat aUK efVatt
ONTOLOGY
In a universe of nothingness. A brief train of waves so brief that its end
and beginning coincide (negative time) disengaging endlessly.
Nothingness resorbs, creates.
It engenders being.
Time, Causality.
These rare sonic events can be something more than isolated sounds.
They can be melodic figures, cell structures, or agglomerations whose
.. The following excerpt (through p. 37) is from" In Search of a Stochastic Music"
Gravesaner BlaUer, no. 11/12. '
" it is the same to think as to be" (Poem by Parmenidcs); and my paraphrase,
For It IS the same to be as not to be."
characteristics are also ruled by the laws of chance, for example, clouds of
sound-points or speed-temperatures.
6
In each case they form a sample of a
succession of rare sonic events.
This sample may be represented by either a simple table of probabilities
or a double-entry table, a matrix, in which the cells are filled by the fre-
quencies of events. The rows represent the particular qualifications of the
events, and the columns the dates (see Matrix M, Fig. 1-9). The frequencies
in this matrix are distributed according to Poisson's formula, which is the
law for the appearances of rare random events.
We should further define the sense of such a distribution and the manner
in which we realize it. There is an advantage in defining chance as an
aesthetic law, as a normal philosophy. Chance is the limit of the notion of
evolving symmetry. Symmetry tends to asymmetry, which in this sense is
equivalent to the negation of traditionally inherited behavioral frameworks.
This negation not only operates on details, but most importantly on the
composition of structures, hence tendencies in painting, sculpture, architec-
ture, and other realms of thought. For example, in architecture, plans worked
out with the aid of regulating diagrams are rendered more complex and
dynamic by exceptional events. Everything happens as if there were one-to-
one oscillations between symmetry, order, rationality, and asymmetry,
disorder, irrationality in the reactions between the epochs of civilizations.
At the beginning of a transformation towards asymmetry, exceptional
events are introduced into symmetry and act as aesthetic stimuli. When
these exceptional events multiply and become the general case, a jump to a
higher level occurs. The level is one of disorder, which, at least in the arts
and in the expressions of artists, proclaims itself as engendered by the com-
plex, vast, and rich vision of the brutal encounters of modern life. Forms
such as abstract and decorative art and action painting bear witness to this
fact. Consequently chance, by whose side we walk in all our daily occupa-
tions, is nothing but an extreme case of this controlled disorder (that which
signifies the richness or poverty ofthe connections between events and which
engenders the dependence or independence of transformations); and by
virtue of the negation, it conversely enjoys all the benevolent characteristics
of an artistic regulator. It is a regulator also of sonic events, their appearance,
and their life. But it is here that the iron logic of the laws of chance inter-
venes; this ehance cannot be created without total submission to its own
laws. On this condition, chance checked by its own force becomes a hydro-
electric torrent.

I. BafJk
B
. (
'Xyf
H.-S I.
gr.1i r.
1
VI. 2
3
1
Vcl.2
3
1
Kb.2.
a

PI

r-5"--, ..--.......
."."""/l
J
JI
..-3--,
rt5"--r
r-s--,
...
if -;--.. '
-:r:- 5--. r-- 3--,
r-- 5"--, r--3--.
...

... .:!"
'--"
.-
A

:t; =f!:: . .:t: :f!:
..
J'
-L
r-
5
--, .-5'--.
.. t Of. .. ., ..
.Op ..
,Pr I r I I" I' I' J I
-"-
,;-3-,

1'''
II
+ .. I
14--1,'---- .----" 8"';
16-- - _
16.: t:\
...

r:t'
..
::f!;

'--5'--'
;f
'--5---'
l-5'---.I
A
r-S'--,
oJ ...
if "

8 1' B
-------; 11. Ji
/ ...... _bif
oJ
'--5--.1 I
,

r"5?
...
-.,..
,,:j,:. v
"
!'1"!

.
A 7-'
r--'-.
-k x.
oJ

'---!i--.J I "-- 5---1

r,-5'-:;t)
-}( Jf
- r-;::;!>--' /1"=1
lr

tI
1l> / I
r-
3
-,
. ,
,3-, r-,5'--,
I I"'"
r-5-.
,,*,
.If
.... ....
Pia;.
Ob.
Klar;
Es
1/alJk
B
7fpt.
I.
1
Xyl.
I. #.'8
1
YI. 2.
3
1
YcI.2
3
J
'll._.J.l.-.f*.:. . --n .-.-.fJ,

..
1"":0 5--,
1\
oJ

II (q r;,,- - :';:;:-3--,
,01

- - L-5--'
0'\ ..-5----,

"
.----. 5---,
i

-
..1'1
,.{I.-
,"' r-- 3-, ...-3:-0 ..-5---,r-- &---, r--5'---, J;-.!l...
'fI'*:t:'$:
"

r- 3--, ;-3--0
"'
.. .. .. Of . .....
II r
I
I' L-J !'
I L...I,r"l "11
;-5'--, r--S'---,
., .... .. r;-.. 5:-, /I .1
" \' V I l) LLJ ..
I
r I (\
8r.;:
{boO'
1'1
r;1 1:'1'
r;I
oJ .

'---5--' '---3---'

r-3---! r- 5 --,
.--6'--,
" 8
1(,----- ....

.....
tl:f!.-51fr
jI' Il .....
.e-r-r-llf
,II 5---,
i
,-57"-5'-V
Sv:
yu
r-vs--,
II '-"" v
v
" ...
V
tI
"" '-' ':' P'!'- I....- '--5'--' '::
1 uX r-S--'
r-s----r;-[-:-,

'--....

.........,..
, t:.
.-3--,
iI---' ...
-
'"
.--5'--0

AM n n M
01
.

"
r-E--. ;-3--,'
, I'
, ...
( S"--, .--5--,
.--5-,
...
11
.. . r
'"
'"
OJ
:;
0
c
.0
u: 0 cii
eli
.!:2 e
'"
0,
C-
N
N
o:::
'"
'" ~
al
OJ
c 0
. .: u
ci) Co
'"
E
t=
-
c
'" >
'"
'" 0.
::J
-0
CO
::J
d
-
c
Q)
>
Q)
'"
:c
:>
o
o
1111 II]
'" c
Q)
:>
'" o
Z
]
c
'"
:>
'" Q)
Ci.
~
c
'" >
'"
'" Cl
c
Vi
However, we are not speaking here of cases where one merely plays
heads and tails in order to choose a particular alternative in some trivial
circumstance. The problem is much more serious than that. It is a matter
here of a philosophic and aesthetic concept ruled by the laws of probability
and by the mathematical functions that formulate that theory, ofa coherent
concept in a new region of coherence.
The analysis that follows is taken from Achorripsis.
For convenience in calculation we shall choose a priori a mean density
of events
A = 0.6 events/unit.
Applying Poisson's formula,
we obtain the table of probabilities:
Po = 0.5488
Pi = 0.3293
P2 = 0.0988
P
3
= 0.0198
P
4
= 0.0030
P
5
= 0.0004.
(1)
Pi is the probability that the event will occur i times in the unit of
volume, time, etc. In choosing a priori 196 units or cells, the distribution of
the frequencies among the cells is obtained by multiplying the values of
Pj by 196.
N urn ber of cells
196 PI
0 107
1 65
2 19 (2)
3 4
4 1
The 196 cells may be arranged in one or several groups of cells, quali-
fied as to timbre and time, so that the number of groups of timbres times
the number of groups of durations = 196 cells. Let there be 7 distinct
timbres; then 196/7 = 28 units of time. Thus the 196 cells arc distributed
over a two-dimensional space as shown in (3).
30
Formalized Music
Timbre
Flute
Oboe
String gliss.
Percussion
Pizzicato
(3)
Brass
String arco
po
o 2 3 ........... 28 ;;rime
I.f the musical sample is to last 7 minutes (a subjective choice) the unit of
tIme U
t
will equal 15 sec., and each U
t
will contain 6.5 measures at
MM = 26.
How shall we distribute the frequencies of zero, single, double, triple,
and events per cell in the two-dimensional space of Matrix (3) ?
C?nslder the 28 columns as cells and distribute the zero, single, double,
trIple, and quadruple events from table (2) in these 28 new cells. Take as an
example single event; fr?m. table (2) it must occur 65 times. Everything
happens as If one to dIstrIbute events in the cells with a mean density
A = 65/28 = 2.32 smgle events per cell (here cell = column).
In applying anew Poisson's formula with the mean density i\ = 2.32
(2.32 30) we obtain table (4).
Poisson Distribution
Arbitrary Distribution
Frequency
No. of Product
Frequency No. of Product
K
Columns col X K
K Columns col x K
0
3 0
0 10 0
1
6 6
I 3 3
2 8 16
2 0 0
3
5 15
3 9 27
4
3
12
(4)
4 0 0
(5)
5
2 lO
5
5
6
1
6
6
5 30
7
0 0
7 0 0
Totals
28
65
Totals
28 65
One could choose any other distribution on condition that the sum of
single events equals 65. Table (5) shows such a distribution.
But in this axiomatic research, where chance must bathe all of sonic
space, we must reject every distribution which departs from Poisson's
law. And the Poisson distribution must be effective not only for the columns
but also for the rows of the matrix. The same reasoning holds true for the
diagonals, etc.
Contenting ourselves just with rows and columns, we obtain a homo-
geneous distribution which follows Poisson. It was in this way that the
distributions in rows and columns of Matrix (M) (Fig. 1-9) were calculated.
So a unique law of chance, the law of Poisson (for rare events) through
the medium of the arbitrary mean ,\ is capable of conditioning, on the one
hand, a whole sample matrix, and on the other, the partial distributions
following the rows and columns. The a priori, arbitrary choice admitted at
the beginning therefore concerns the variables of the" vector-matrix."
Variables or entries of the "vector-matrix"
1. Poisson's Law
2. The mean i\
3. The number of cells, rows, and columns
The distributions entered in this matrix are not always rigorously
defined. They really depend, for a given t\, on the number of rows or col-
umns. The greater the number of rows or columns, the more rigorous is the
definition. This is the law of large numbers. But this indeterminism allows
free will if the artistic inspiration wishes it. It is a second door that is open
to the subjectivism of the composer, the first being the" state of entry" of
the "Vector-Matrix" defined above.
Now we must specify the unit-events, whose frequencies were adjusted
in the standard matrix (M). We shall take as a single event a cloud of sounds
with linear density 1) sounds/sec. Ten sounds/sec is about the limit that a
normal orchestra can play. We shall choose 0 = 5 sounds/measure at MM
26, so that 1) = 2.2 sounds/sec 10/4).
We shall now set out the following correspondence:
32
Formalized Music
Cloud of density 0 =
Mean number of sounds/cell
Event
Sounds/ I Sounds/
measure 26MM sec
(15 sec)
zero 0 0 0
single 5 2.2 32.5
double 10 4.4 65
triple 15 6.6 97.5
quadruple 20 8.8 130
The hatchings in matrix (M) show a Poisson distribution offrequencies,
homogeneous and verified in terms of rows and columns. We notice that
the rows are interchangeable (= interchangeablc timbres). So are the
columns. This leads us to admit that the determinism of this matrix is weak
in part, and that it serves chiefly as a basis for thought-for thought which
manipulates frequencies of events of all kinds. The true work of molding
sound consists of distributing the clouds in the two-dimensional space of the
matrix, and of anticipating a priori all the sonic encounters before the
calculation of details, eliminating prejudicial positions. It is a work of
patient research which exploits all the creative faculties instantaneously.
This matrix is like a game of chess for a single player who must follow certain
rules of the game for a prize for which he himself is the judge. This game
matrix has no unique strategy. It is not even possible to disentangle any
balanced goals. It is very general and incalculable by pure reason.
Up to this point we have placed the cloud densities in the matrix. Now
with the aid of calculation we must proceed to the coordination of the
aleatory sonic elements.
HYPOTHESES OF CALCULATION
Let us analyze as an example cell III, tZ of the matrix: third row,
sounds of continuous variation (string glissandi), seventeenth unit of time
(measures 103-11). The density of the sounds is 4.5 sounds/measure at
MM 26 (0 = 4.5); so that 4.5 sounds/measure times 6.5 measures = 29
sounds for this cell. How shall we place the 29 glissando sounds in this cell ?
Hypothesis 1. The acoustic characteristic of the glissando sound is
assimilated to the speed v = df/dt of a uniformly continuous movement.
(See Fig. 1-10.)
Hypothesis 2. The quadratic mean ct of all the possible values of v is
proportional to the sonic density 8. In this ease IX = 3.38 (temperature).
Hypothesis 3. The values of these speeds are distributed according to the
most complete asymmetry (chance). This distribution follows the law of
/vor;::s
Fig. 1-10
: dt
I
I
1
I
1
I
I
33
/IHE
Gauss. The probability f(v) for the existence of the speed v is given by the
function
f( )
- 2 -v'la' .
V - --e
aV7r '
and the probability P (,\) that v will lie between V1 and V2, by the function
8('\) = ~ 7 r {' e-
7I2
d'\ (normal distribution).
Hypothesis 4. A glissando sound is essentially characterized by a. the
moment of its departure; b. its speed Vrn = cif/dt, (V1 < Vrn < v
2
); and c. its
register.
Hypothesis 5. Assimilate time to a line and make each moment of
departure a point on that line. It is as if one were to distribute a number of
points on a line with a linear density 0 = 4.5 points at MM = 26. This,
then, is a problem of continuous probabilities. These points define segments
and the probability that the i-th segment will have a length XI between
x and X + dx is
Hypothesis 6. The moment of departure corresponds to a sound. We
shall attempt to define its pitch. The strings have a range of about 80 semi-
tones, which may be represented by a line oflength a = 80 semitones. Since
between two successive or simultaneous glissandi there exists an interval
between the pitches at the moments of departure, we can define not only
the note of attack for the first glissando, but also the melodic interval which
separates the two origins.
34
Formalized Music
Put thus, the problem consists of finding the probability that a segment
s within a line segment of length a will have a length between) and) + dj
(0 :5, j :5, a). This probability is given by the formula
B(j) dj = (1 - dj.
(See Appendix I.)
Hypothesis 7. The three essential characteristics of the glissando sound
defined in Hypothesis 4 are independent.
From these hypotheses we can draw up the three tables of probability:
a table of durations, a table of speeds, and a table of intervals.
All these tables furnish us with the elements which materialize in
cell III, IZ. The reader is encouraged to examine the to see how the
results of the calculations have been used. Here also, may we emphasize,
a great liberty of choice is given the composer. The restrictions are more of
a general canalizing kind, rather than peremptory. The theory and the
calculation define the tendencies of the sonic entity, but they do not con-
stitute a slavery. Mathematical formulae arc thus tamed and subjugated
by musical thought. We have given this example of glissando sounds
because it contains all the problems of stochastic music, controlled, up to a
certain point, by calculation.
x
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
Table of Durations
o = 4.5 sounds/measure at MM 26
Unit x = 0.10 of the measure at 26 MM
4.5 . 6.5 = 29 sounds/cell, i.e., 28 duraLions
ox e-
ox
De-OX dx
0.00 1.000 4.500 0.362
0.45 0.638 2.870 0.231
0.90 0.407 1.830 0.148
1.35 0.259 1.165 0.094
1.80 0.165 0.743 0.060
2.25 0.105 0.473 0.038
2.70 0.067 0.302 0.024
3.15 0.043 0.194 0.016
Totals 12.415 0.973
10
7
4
3
2
o
28
35
An approximation is made by considering dx as a constant factor.
00
L dx = 1.
0
Therefore
co
dx = l/L oe-
6x
.
0
In this case dx = 1/12.415 = 0.805.
Table of Speeds
D = 4.5 glissando sounds/measure at 26 MM
ex = 3.88, quadratic mean of the speeds
v is expressed in scmitones/measure at 26 MM
Vrn is the mean speed (VI + v2)/2
4.56.5 = 29 glissando sounds/cell.
v A = via B(A) P(A) = B(A
2
) - B(A
I
) 29 P(A) Vrn
0 0.000 0.0000
0.2869 9 0.5
0.258 0.2869
0.2510 7 1.5
2 0.516 0.5379
0.1859 5 2.5
3 0.773 0.7238
0.1310 4 3.5
4 1.032 0.8548
0.0771 2 4.5
5 1.228 0.9319
0.0397 5.5
6 1.545 0.9716
0.0179 6.5
7 1.805 0.9895
0.0071 0 7.5
36
Table of Intervals
8 = 4.5 glissandi/measure at 26 MM.
Formalized Music
a = 80 semitones, or 18 times the arbitrary unit of 4.5 semi tones.
} is expressed in multiples of 4.5 semitones.
dj is considered to be constant. Therefore dj = 1/'iJJ(j) or dj = a/(m + 1),
and we obtain a step function. For} = 0, 8(j)c[j = 2/(m + 1) = 0.105; for
} = 18, a(j)d} = O.
4.5 . 6.5 = 29 glissando sounds per cell.
We can construct the table of probabilities by means of a straight line.
J
8(j) dj = P(j) 29 P(j)
0 .. 0.105 3
1 3
2 3
3 3
4 2
5 2
6 2
7 2
8 2
9 2
10 1
11 1
12 1
13 1
14
15 0
16 0
17 0
18 0
We shall not speak of the means of verification ofliaisons and correla-
tions between the various values used. It would be too long, complex, and
tedious. For the moment let us affirm that the basic matrix was verified by
the two formulae:
2: (x - x) (y - y)
r =
and
1 + r
z = .1log --
2 1 - r
37
Let us now imagine music composed with the aid of matrix (M). An
observer who perceived the frequencies of events of the musical sample
would deduce a distribution due to chance and following the laws of
probability. Now the question is, when heard a number of times, will this
music keep its surprise effect? Will it not change into a set of foreseeable
phenomena through the existence of memory, despite the fact that the law
of frequencies has been derived from the laws of chance?
In fact, the data will appear aleatory only at the first hearing. Then,
during successive rehearings the relations between the events of the sample
ordained by "chance" will form a network, which will take on a definite
meaning in the mind of the listener, and will initiate a special" logic," a
new cohesion capable of satisfying his intellect as well as his aesthetic sense;
that is, if the artist has a certain flair.
If, on the other hand, we wish the sample to be unforeseeable at all
times, it is possible to conceive that at each repetition certain data might
be transformed in such a way that their deviations from theoretical fre-
quencies would not be significant. Perhaps a programming useful for a first,
second, third, etc., performance will give aleatory samples that are not identi-
cal in an absolute sense, whose deviations will also be distributed by chance.
Or again a system with electronic computers might permit variations
of the parameters of entrance to the matrix and of the clouds, under certain
conditions. There would thus arise a music which can be distorted in the
course of time, giving the same observer n results apparently due to chance
for n performances. In the long run the music will follow the laws of proba-
bility and the performances will be statistically identical with each other, the
identity being defined once for all by the" vector-matrix."
The sonic scheme defined under this form of vector-matrix is conse-
quently capable of establishing a more or less self-determined regulation of
the rare sonic events contained in a musical composition sample. It repre-
sents a compositional attitude, a fundamentally stochastic behavior, a unity
of superior order. [1956-57].
If the first steps may be summarized by the process vision --+ rules --+
works afart, the question concerning the minimum has produced an inverse
38 Formalized Music
path: rules ---;.. vision. In fact stochastics permits a philosophic vision, as the
example of Achorripsis bears witness.
CHANCE-1M PROVISATION
Before generalizing further on the essence of musical composition, we
must speak of the principle of improvisation which caused a furore among
the nco-serialists, and which gives them the right, or so they think, to speak
of chance, of the aleatory, which they thus introduce into music. They
write scores in which certain combinations of sounds may be freely chosen
by the interpreter. It is evident that these composers consider the various
possible circuits as equivalent. Two logical infirmities are apparent which
deny them the right to speak of chance on the one hand and" composition"
on the other (composition in the broad sense, that is) :
1. The interpreter is a highly conditioned being, so that it is not possible
to accept the thesis of unconditioned choice, of an interpreter acting like a
roulette game. The martingale betting at Monte Carlo and the procession
of suicides should convince anyone of this. We shall return to this.
2. The composer commits an act of resignation when he admits several
possible and equivalent circuits. In the name of a" scheme" the problem of
choice is betrayed, and it is the interpreter who is promoted to the rank of
composer by the composer himself. There is thus a substitution of authors.
The extremist extension of this attitude is one which uses graphical
signs on a piece of paper which the interpreter reads while improvising the
whole. The two infirmities mentioned above are terribly aggravated here.
I would like to pose a question: Hthis sheet of paper is put before an inter-
preter who is an incomparable expert on Chopin, will the result not be
modulated by the style and writing of Chopin in the same way that a per-
former who is immersed in this style might improvise a Chopin-like cadenza
to another composer's concerto? From the point of view of the composer
there is no interest.
On the contrary, two conclusions may be drawn: first, that serial
has become so banal that it can be improvised like Chopin's,
whIch confirms the general impression; and second, that the composer
resigns his function altogether, that he has nothing to say, and that his
function can be taken over by paintings or by cuneiform glyphs.
Chance needs to be calculated
To finish with the thesis of the roulette-musician, I shall add this:
Chance is a rare thing and a snare. It can be constructed up to a certain
39
point with great difficulty, by means of complex reasoning which is sum-
marized in mathematical formulae; it can be constructed a little, but never
improvised or intellectually imitated. I refer to the demonstration of the
impossibility of imitating chance which was made by the great mathemati-
cian Emile Borel, who was one of the specialists in the calculus ofprobabil-
ities. In any case-to play with sounds like dice-what a truly simplistic
activity! But once one has emcrged from this primary field of chancc worth-
less to a musician, the calculation of the aleatory, that is to say stochastics,
guarantees first that in a region of precise definition slips will not be made,
and then furnishes a powerful method of reasoning and enrichment of sonic
processes.
STOCHASTIC PAINTING?
In line with these ideas, Michel Philippot introduced the calculus of
probabilities into his painting several years ago, thus opening new direc-
tions for investigation in this artistic realm. In music he recently endeavored
to analyze the act of composition in the form of a flow chart for an imaginary
machine. It is a fundamental analysis of voluntary choice, which leads to a
chain of aleatory or deterministic events, and is based on the work Composi-
tion pour double orchestre (1960). The term imaginary machine means that the
composer may rigorously define the entities and operating methods, just as
on an electronic computer. In 1960 Philippot commented on his Composition
pour double orchestre ..
If, in connection with this work, I happened to use the term
"experimental music," I should specify in what sense it was meant in
this particular case. It has nothing to do with concrete or electronic
music, but with a very banal score written on the usual ruled paper
and requiring none but the most traditional orchestral instruments.
However, the experiment ofwhieh this composition was in some sense
a by-product does exist (and one can think of many industries that
survive only through the exploitation of their by-products).
The end sought was merely to effect, in the context of a work
which I would have written independent of all experimental ambi-
tions, an exploration of the processes followed by my own cerebral
mechanism as it arranged the sonic elements. I therefore devised the
following steps:
I. Make the most complete inventory possible of the set of my
gestures, ideas, mannerisms, decisions, and choices, etc., which were
mine when I wrote the music.
40
Formalized Music
2. Reduce this set to a succession of simple decisions, binary, if
possible; i.e., accept or refuse a particular note, duration, or silence in
a situation determined and defined by the context on one hand, and
by the conditioning to which I had been subjected and my personal
tastes on the other.
3. Establish, if possible, from this sequence of simple decisions, a
scheme ordered according to the following two considerations (which
were sometimes contradictory): the manner in which these decisions
emerged from my imagination in the course of the work, and the
manner in which they would have to emerge in order to be most
useful.
4. Present this scheme in the form of a flow chart containing the
logical chain of these decisions, the operation of w h i ~ h could easily
be controlled.
5. Set in motion a mechanism of simulation respecting the rules
of the game in the flow chart and note the resull.
6. Compare this result with my musical intentions.
7. Check the differences between result and intentions, detect their
causes, and correct the operating rules.
8. Refer these corrections back to the sequence of experimental
phases, i.e., start again at l. until a satisfactory result has been ob-
tained.
If we confine ourselves to the most general considerations, it
would simply be a matter of proceeding to an analysis of the complex-
ity, considered as an accumulation, in a certain order, of single events,
and then of reconstructing this complexity, at the same time verifying
the nature of the elements and their rules ofeombination. A cursory
look at the flow chart of the first movement specifies quite well by a
mere glance the method I used. But to confine oneself to this first
movement would be to misunderstand the essentials of musical
composition.
In fact the "preludial" character which emerges from this
combination of notes (elementary constituents of the orchestra)
should remind us of the fact that composition in its ultimate stage is
also an assembly of groups of notes, motifs, or themes and their
transformations. Consequently the task revealed by the flow charts
of the following movements ought to make conspicuous a grouping of
a higher order, in which the data of the first movement were used as
a sort of "prefabricated" material. Thus appeared the phenomenon,
a rather banal one, of autogeneration of complexity by juxtaposition
and combination of a large number of single events and operations.
At the end of this experiment I possessed at most some insight into
my own musical tastes, but to me, the obviously interesting aspect of
c-
o
E
Q)
E
.!:
+
~
c-
o '" N ()
t:
I ~
;fI
.,
0-
o
cii
~
\ - (!) "'
\ ~ / -
o
c
a;
"
><
Q)
c
'iii
cii
o
c
I
IJ)
to
IJ)
...
:I
Cl
u:::
42
Formalized Music
it (as long as there is no error of omission!) was the analysis of the
composcr, his mental processes, and a ccrtain liberation of the
imagination.
The biggest difficulty encountered was that of a conscious and
voluntary split in personality. On one hand, was the composer who
already had a clear idea and a precisc audition of the work he wished
to obtain; and on the other was the experimenter who had to maintain
a lucidity which rapidly became burdensome in these conditions-a
lucidity with respcct to his own gestures and decisions. Wc must not
ignore the fact that such experiments must be examincd with the
greatest prudcnce, for everyone knows that no observation of a
phenomenon exists whieh does not disturb that phenomenon, and I
fear that the resulting disturbance might be particularly strong when
it concerns such an ill-defined domain and such a delicate activity.
Moreover, in this particular case, I fear that observation might pro-
voke its own disturbance. If I accepted this risk, I did not under-
estimate its extent. At most, my ambition confined itself to the attempt
to project on a marvelous unknown, that of aesthetic creation, the
timid light of a dark lantern. (The dark lantern had the reputation
of being used especially by housebreakers. On several occasions I have
been able to verify how much my thirst for investigation has made me
appear in the eyes of the majority as a dangerous housebreaker of
inspiration. )
Chapter II
Markovian Stochastic Music-Theory
Now we can rapidly generalize the study of musical composition with the
aid of stochastics.
The first thesis is that stochastics is valuable not only in instrumental
music, but also in electromagnctic music. Wc have demonstrated this with
several works: Diamorphoses 1957-58 (B.A.M. Paris), Concret PH (in the
Philips Pavilion at the Brussels Exhibition, 1958); and Orient-Occident, m u s ~ c
for the film of the same name by E. Fulchignoni, produced by UNESCO III
1960.
The second thesis is that stochastics can lead to the creation of new
sonic materials and to new forms. For this purpose we must as a preamble
put forward a temporary hypothesis which concerns the nature of sound, of
all sound [19].
BASIC TEMPORARY HYPOTHESIS (lemma) AND DEFINITIONS
All sound is an integration of grains, of elementary sonic particles, of
sonic quanta. Each of these elementary grains has a threefold nature:
duration, frequency, and intensity.1 All sound, even all continuous sonic
variation, is conceived as an assemblage of a large number of elementary
grains adequately disposed in time. So every sonic complex can be analyzed
as a series of pure sinusoidal sounds even if the variations of these sinusoidal
sounds are infinitely close, short, and complex. In thc attack, body, and
decline of a complex sound, thousands of pure sounds appear in a more or
less short interval of time, tlt. Hecatombs of pure sounds are necessary for
the creation of a complex sound. A complex sound may be imagined as a
multi-colored firework in which each point of light appears and instan-
43
44 Formalized Music
taneously disappears against a black sky. But in this firework there would
be such a quantity of points oflight organized in such a way that their rapid
and teeming succession would create forms and spirals, slowly unfolding, or
conversely, brief explosions setting the whole sky aflame. A line oflight would
be created by a sufficiently large multitude of points appearing and dis-
appearing instantaneously.
Ifwe consider the duration ~ t of the grain as quite small but invariable,
we can ignore it in what follows and consider frequency and intensity only.
The two physical substances of a sound are frequency and intensity in
association. They constitute two sets, F and G, independent by their nature.
They have a set product F x G, which is the elementary grain of sound. SetF
can be put in any kind of correspondence with G: many-valued, single-
valued, one-to-one mapping, .... The correspondence can be given by an
extensive representation, a matrix representation, or a canonical represen-
tation.
EXAMPLES OF REPRESENTATIONS
Extensive (term by term) :
Frequencies
111
12 la
Intensi ties g3 gn ga
Matrix (in the form of a table):
t 11 12 13 14 15 16
gl +
0
+
0 0 0
g2 0
+
0 0 0
+
g3 0 0 0
+ +
0
Canonical (in the form of a function):
vl=Kg
1 = frequency
g = intensity
K = coefficient.
14
g"
17
+
0
0
The correspondence may also be indeterminate (stochastic), and here
the most convenient representation is the matrical one, which gives the
transition probabilities.
Markovian Stochastic Music-Theory 45
Example:
t 11 12 13 14
gl 0.5 0 0.2 0
g2 0 0.3 0.3
ga 0.5 0.7 0.5 0
The table should be interpreted as follows: for each value ~ of 1 there are
one or several corresponding intensity values g" defined by a probability.
For example, the two intensities g2 and g3 correspond to the frequency 12,
with 30% and 70/
0
chance of occurrence, respectively. On the other hand,
each of the two sets F and G can be furnished with a structure-that is to
say, internal relations and laws of composition.
Time t is considered as a totally ordered set mapped onto F or G in a
lexicographic form.
Examples:
al1 12 13
t = 1,2,
c.
11 11 12 11 12 12 In la
b. 10.5 13 1';11 Ix
t = 0.5, 3, Vll, x,
t= ABC DE .. ., ....
~ t ~ t ~ t ~ t ~ t ~ t ~ t
~ t = M
Example c. is thc most general since continuous evolution is sectioned
into slices of a single thickness ~ t , which transforms it in discontinuity; this
makes it much easier to isolate and examine under the magnifying glass.
GRAPHICAL REPRESENTATIONS
We can plot the values of pure frequencies in units of octaves or semi-
tones on the abscissa axis, and the intensity values in decibels on the ordinate
axis, using logarithmic scales (see Fig. II-I). This cloud of points is the
cylindrical projection on the plane (FC) of the grains contained in a thin
slice ~ t (see Fig. II-2). The graphical representations Figs. II-2 and II-3
make more tangible the abstract possibilities raised up to this point.
Psychophysiology
We are confronted with a cloud of evolving points. This cloud is the
product of the two setsFand C in the slice of time ~ t . What are the possible
46
G
(dB)
.' .
, .
Formalized Music
Elementary grain
considered as an
instantaneous associ-
ation of an intensity
9 and a frequency f
Frequencies in logarithmic
Fig. 11-1 '--____________________ u_n_it_s....:(_e. . . : : g ~ . semitones)
F
G
G
\
~ ~ - - - - - - - r - - - - - - - - - - - - - - - - - ~ ~ - - F
F;goll_2
M
~ ~ - - - - - - - - - - - - - - - __ F
F
Fig. 11-3
.---
t
..
'.
-<- Point i
Each point i
must be attributed
to an intensity gj
t
restrictive limits of human psychophysiology? What are the most general
manipulations which may be imposed on the clouds and their transforma-
tions within psychophysiological limits?
The basic abstract hypothesis, which is the granular construction of all
possible sounds, gives a very profound meaning to these two questions. In
fact within human limits, using all sorts of manipulations with thcse grain
clusters, we can hope to produce not only the sounds of classical instruments
and elastic bodies, and those sounds generally preferred in concrete music,
but also sonic perturbations with evolutions, unparalleled and unimaginable
until now. The basis of the timbre structures and transformations will have
nothing in common with what has been known until now.
We can even express a more general supposition. Suppose that each
point of these clusters represents not only a pure frequency and its satellite
intensity, but an already present structure of elementary grains, ordered a
priori. We believe that in this way a sonority of a second, third, or higher
order can be produced.
Recent work on hearing has given satisfactory answers to certain
problems of perception. The basic problems which concern us and which
we shall suppose to be resolved, even if some of the solutions are in part
lacking, are [2, 3]: 1. What is the minimum perceptible duration (in com-
fort) of a sinusoidal sound, as a function of its frequency and its intensity?
2. What are the minimum values of intensities in decibels compatible with
minimum frequencies and durations of sinusoidal sounds? 3. What are the
minimum melodic interval thresholds, as a function of register, intensity,
and duration? A good approximation is the Fletcher-Munson diagram of
equal loudness contours (see Fig. II-4).
The total number of elementary audible grains is about 340,000. The
car is more sensitive at the center of the audible area. At the extremities it
perceives less amplitude and fewer melodic intervals, so that ifone wished to
represent the audible area in a homogeneous manner using the coordinates
F and G, i.e., with each surface element !:l.F!:l.G containing the same density
of grains of perceptible sounds, one would obtain a sort of mappa mundi
(Fig. II-5).
In order to simplify the reasoning which will follow without altering it,
we shall base our argument on Fletcher's diagram and suppose that an
appropriate one-to-one transrormation applied to this group of coordinates
will change this curved space into an ordinary rectangle (Fig. II-6).
All the above experimental results were established in ideal conditions
and without reference to the actual complexity of the natural sounds of the
orchestra and of clastic bodies in general, not to mention thc more complex
G
(d B)
48
4 o ~ - - - - ~ - + - 4 - - ~ - + -
2(:1 ,30 40 6() So too 40() dOG G(I(J lao/)
Audible area
Fig. 11-4. Fletcher-Munson Diagram
Equal Loudness Contours
Fig. 11-5
-<- Threshold of
sound perception
F
(Semilones)
Formalized Music
G
F
2000 0000
~ O O O D J..I::r:
Markovian Stochastic Music--Theory 49
-
-.r
~
~
sounds of industry or of chaotic nature [4]. Theoretically [5] a complex
sound can only be exhaustively represented on a three-dimensional diagram
F, G, I, giving the instantaneous frequency and intensity as a function of
time. But in practice this boils down to saying that in order to represent a
momentary sound, such as a simple noise made by a car, months of calcula-
tions and graphs are necessary. This impasse is strikingly reminiscent of
classical mechanics, which claimed that, given sufficient time, it could
account for all physical and even biological phenomena using only a few
formulae. But just to describe the state of a gaseous mass of greatly reduced
volume at one instant t, even if simplifications are allowed at the beginning
of the calculation, would require several centuries of human work!
This was a false problem because it is useless; and as far as gaseous
masses are concerned, the Maxwell-Boltzmann kinetic theory of gases,
with its statistical method, has been very fruitful [6]. This method re-
established the value of scales of observation. For a macroscopic phenomenon
it is the massed total result which counts, and each time a phenomenon is
to be observed the scale relationship between observer and phenomenon
must first be established. Thus if we observe galactic masses, we must decide
whether it is the movement of the whole mass, the movement ofa single star,
or the molecular constitution of a minute region on a star that interests us.
The same thing holds true for complex as well as quite simple sounds.
It would be a waste of effort to attempt to account analytically or graphi-
cally for the characteristics of complex sounds when they are to be used in
an electromagnetic composition. For the manipulation of these sounds
macroscopic methods are necessary.
Inversely, and this is what particularly interests us here, to work like
architects on the sonic material in order to construct complex sounds and
t
50 Formalized Music
evolutions of these entities means that we must use macroscopic methods of
analysis and construction. Microsounds and elementary grains have no
importance on the scale which we have chosen. Only groups of grains and
the characteristics of these groups have any meaning. Naturally in very
particular cases, the single grain will be reestablished in all its glory. In a
Wilson chamber it is the elementary particle which carries theoretical and
experimental physics on its shoulders, while in the sun it is the mass of
particles and their compact interactions which constitute the solar object.
Our field of evolution is therefore the curved space described above,
but simplified to a rectilinear space by means of complete one-to-one
transformation, which safeguards the validity of the reasoning which we
shall pursue.
SCREENS
The graphical representation of a cloud of grains in a slice of time tJ.t
examined earlier brings a new concept, that of the density of grains per unit
of volume, t:.Ft:.Gt:.t (Fig. 11-7). Every possible sound may therefore be cut
up into a precise quantity of elements t:.F!:::.G!:::.t!:::.D in four dimensions,
distributed in this space and following certain rules defining this sound,
which are summarized by a function with four variables: s(F, G, D, t).
G
AF
/"
Fig. 11-7
Plane of
reference (FG)
at moment t
F
fj,D will be
the dimension
of the density
Markovian Stochastic M u s i c ~ Theory 51
The scale of the density will also be logarithmic with its base between
2 and 3.
2
To simplify the explanation we will make an abstraction of this
new coordinate of density. It will always be present in our mind but as an
entity associated with the three-dimensional element t:.Ft:.Gt:.t.
If time is considered as a procedure of lexicographic ordering, we can,
without loss, assume that the t:.t are equal constants and quite small. We can
thus reason on a two-dimensional space defined by the axes F and G, on
condition that we do not lose sight of the fact that the cloud of grains of
sound exists in the thickness of time t:.t and that the grains of sound are only
artificially flattened on the plane (FG).
Difinition oj the screen. The screen is the audible area (FG) fixed by a
sufficiently close and homogeneous grid as defined above, the cells of which
mayor may not be occupied by grains. In this way any sound and its
history may be described by means of a sufficiently large number of sheets
of paper carrying a given screen S. Thcse sheets are placed in a fixed lexi-
cographic order (see Fig. 11-8).
Fig. 11-8
- Screen
Cell full of grains
A book of screens equals
the life of a complex sound
The clouds of grains drawn on the screens will differ from one screen
to another by their geographical or topographical position and by their
surface density (see Fig. 11-9). Screen A contains a small elemental rectan-
gle with a small cluster of density d of mean frequency Jand mean amplitude
g. It is almost a pure sound. Screen B represents a more complex sound with
strong high and low areas but with a weak center. Screen C represents a
52 Formalized Music
"white" sound of weak density which may therefore be perceived as a sonic
sheen occupying the whole audible area.
What is important in all the statements made up to now is that nothing
has been said about the topographic fixity of the grains on the screens. All
natural or instrumental sounds arc composed of small surface elements filled
with grains which fluctuate around a mean frequency and intensity. The
same holds for the density. This statement is fundamental, and it is very
likely that the failure of electronic music to create new timbres, aside from
the inadequacy of the serial method, is largely due to the fixity of the grains,
which form structures like packets of spaghetti (Fig. II-lO).
Topographic fixity of the grains is a very particular case, the most
general case being mobility and the statistical distributiOl' of grains around
positions of equilibrium. Consequently in the majority of cases real sounds
ean be analyzed as quite small rectangles, b.Fb.G, in which the topographic
positions and the densities vary from one screen to another following more
or less well-defined laws.
Thus the sound of example D at this precise instant is formed by the
collection of rectangles (f2g4), (f2gS), (f4g2), (f4ga), (fSgl)' (fSg2), (f6g1),
(f6g2), (f6gS), (f7g2), (f7g3), (f7g4), (f7g5), (fSg3), (fSg4), (fsgs), and in each
of the rectangles the grains are disposed in an asymmetric and homogeneous
manner (see Fig. II-ll).
CONSTRUCTION OF THE ELEMENTS b.Fb.G OF THE SCREENS
1. By calculation. We shall examine the means of calculating the elements
b.F b.Gb.lb.D.
How should the grains be distributed in an elemental volume? If we
fix the mean density of the grains (= number of grains per unit of volume)
we have to resolve a problem of probability in a four-dimensional space. A
simpler method would be to consider and then calculate the four coordinates
independently.
For the coordinate I the law of distribution of grains on the axis oftime
IS:
P
x
= c e-
CX
dx or P
x
, = e-
clv
(r) (See Appendix 1.)
For the coordinates G, F, D the stochastic law will be:
f(j) dj = (1 - dj
(r')
or 2 ( i)
Pi = 2. lOn I - 2. Ion - I .
(See Appendix 1.)
G
G G
. - .
, ------ -.
J F
B
F
A
Fig. 11-9
G
.. '
Fig. 11-10

F
G
(dB)
t,
. "
. .
........ : ... ',
(*
.' ..... .
... : ...
t3
t'L
:-:; .,.. ;" ; ."
rr
11
h h
Fig. 11-11
":::- ".' ....
....
. .
.
...
- '.
,': .... :.: "
c
(Semitones)
53
F
54 Formalized Music
From these formulae we can draw up tables of the frequencies of the
values t, G, F, D (see the analogous problem in Chapter I). These formulae
are in our opinion privileged, for they arise from very simple reasoning,
probably the very simplest; and it is essential to start out with a minimUIll
of terms and constraints if we wish to keep to the principle of the tabula rasa
(1st and 3rd rules of Descartes's Discourse on Method).
Let there be one of these elemental volumes fo,.tfo,.Dfo,.Ffo,.G of the screen
at the moment t. This volume has a density D taken from the table derived
from formula (r'). Points on fo,.t are defined with a linear density D = c
according to the table defined by formula (r). To each point is attributed a
sonic grain of frequency J and intensity g, taken from within the rectangle
llFllG by means of the table of frequencies derived from Iprmula (r'). The
correspondences are made graphically or by random successive drawings
from urns composed according to the above tables.
2. Mechanically. a. On the tape recorder: The grains are realized from
sinusoidal sounds whose durations are constant, about 0.04 sec. These
grains must cover the selected elemental area 6.Ffo,.G. Unfolding in time is
accomplished by using the table of durations for a minimum density c = D.
By mixing sections of this tape with itself, we can obtain densities varying
geometrically with ratio 1: 2: 3 ... according to the number of tracks that
we use. b. On computers: The grains are realized from wave forms duly
programmed according to Gabor's signals, for a computer to which an
analogue converter has been coupled. A second program would provide for
the construction of the elemental volumes /}.tllDllF6.G from formulae (r)
and (r').
First General Comment
Take the cell llF6.G6.t. Although occupied in a homogeneous manner
by grains of sound, it varies in time by fluctuating around a mean density
d
m
We can apply another argument which is more synthetic, and admit that
these fluctuations will exist in the most general case anyhow (if the sound is
long enough), and will therefore obey the laws of chance. In this case, the
problem is put in the following manner:
Given a prismatic cloud of grains of density d
m
, of cross section 6.F6.G
and length :z: 6./, what is the probability that d grains will be found in an
elemental volume fo,.Ffo,.G6.t? If the number d
m
is small enough, the probabil-
ity is given by Poisson's formula:
t
For the definition of each grain we shall again use the methods de-
scribed above.
Second General Comment (Vector Space) [8]
We can construct elemental cells 6.F6.G of the screens not only with
points, but with elemental vectors associated with the grains (vector space).
The mean density of 0.04 sec/grain really implies a small vector. The partic-
ular case of the grain occurs when the vector is parallel to the axis of time,
when its projection on the plane (FG) is a point, and when the frequency of
the grain is constant. In general, the frequencies and intensities of the grains
can be variable and the grain a very short glissando (see Fig. 11-12).
~
AF
F
I
\
Fig. 11-12
In a vector space (FG) thus defined, the construction of screens would
perhaps be cumbersome, for it would be necessary to introduce the idea of
speed and the statistical distribution of its values, but the interest in the
undertaking would be enormous. We could imagine screens as the basis of
granular fields which are magnetized or completely neutral (disordered).
In the case of total disorder, we can calculate the probability J(v) of
the existence of a vector v on the plane (FG) using Maxwell's formula as
applied to two dimensions [11]:
J(
v) - 2v _vOla
2
- a2 e .
56 Formalized Music
For the mean value VI ~ Vrn ~ V
2
,
in which Al = vda and
for Al ~ A ~ 1..2 (normal Gaussian law) [12]. In any case, whether it is a
matter of a vector space or a scalar space does not modify the arguments
[13].
Summary of the Screens
I. A screen is described by a set of clouds that are themselves a set of
elemental rectangles t.Ft.G, and which mayor may not contain grains of
sound. These conditions exist at the moment t in a slice of time 6.t, as small
as desired.
2. Thc grains of sound create a density peculiar to each elemental
rectangle 6.F6.G and are generally distributed in the rectangles in an er-
godic manner. (The ergodic principle states that the capricious effect of an
operation that depends on chance is regularized more and more as the
operation is repeated. Here it is understood that a vcry large succession of
screens is being considered [14].)
3. The conception of the elemental volume t.Ft.Gt. T6.D is such that
no simultaneity of grains is generally admitted. Simultaneity occurs when
the density is high enough. Its frequency is bound up with the size of the
density. It is all a question of scale and this paragraph refers above all to
realization. The temporal dimension of the grain (vector) being of the order
of 0.04 sec., no systematic overlapping of two grains (vectors) will be accepted
when the elementary density is, for example, Do = 1.5 grains/sec. And as
the surface distribution of the grains is homogeneous, only chance can
create this overlapping.
4. The limit for a screen may be only one pure sound (sinusoidal),
or even no sound at all (empty screen).
ELEMENTARY OPERATIONS ON SCREENS
Let there be a complex sound. At an instant t of its life during a
thickness t.t it can be represented by one or several clouds of grains or
vectors on the plane (FG). This is the definition which we gave for the
57
screen. The junction of several of these screens in a given order describes
or prescribes the life of this sonic complex. It would be interesting to envisage
in all its generality the manner of combining and juxtaposing screens to
describe, and above all to construct, sonic evolutions, which may be con-
tinuous or discontinuous, with a -view to playing with them in a composition.
To this end we shall borrow the terminology and symbolism of modern
algebra, but in an elementary manner and as a form of introduction to a
further development which we shall not undertake at the moment.
Comment: It does not matter whether we place ourselves on the plane
of physical phenomena or of perception. In general, on the plane of per-
ception we consider arithmetically that which is geometrical on the physical
plane. This can be expressed in a more rigorous manner. Perception
constitutes an additive group which is almost isomorphic with a physical
excitation constituting a multiplicative group. The" almost" is necessary to
exorcise approximations.
Grains or vectors on the plane (FG) constitute a cloud. A screen can be
composed of no grain at all or of several clouds of grains or vectors (see
Fig. II-l3).
Screen 1
Fig. 11-13
Screen 2
@
B
Screen 3
To notate that a grain or vector a belongs to a cloud E, we write a E E
the contrary is written a rf: E. If all the grains of a cloud X are grains of
another cloud Y, it is said that X is included in Y or that X is a part or
sub-cloud of Y. This relation is notated X c Y (inclusion).
Consequently we have the following properties:
X c X for any X.
X c Y and Y c Z imply X c Z.
When X c Yand Y c X, the clouds X and Y consist of the same grains;
they are indistinguishable and the relation is written: X = Y (equation).
A cloud may contain as little as a single grain. A cloud X is said to be
empty when it contains no grain a, such that a E X. The empty cloud is
notated 0.
58 Formalized Music
ELEMENTARY OPERATIONS
These operations apply equally well to clouds and to screens. We can
therefore use the terms" screen" and" cloud" indiscriminately, with cloud
and grain as "constitutive elements."
The intersection of two screens A and B is the screen of clouds which
belong to both A and B. This is notated as A n B and read as "A inter B"
(Fig. II-14). When A n B = 0, A and B are said to be disjoint (Fig.
II -15). The union of two screens A and B is the set of clouds which belong
to both A and/or B (Fig. II -16). The complement of a screen A in relation to
a screen E containing A is the set of clouds in E which do not belong to
A. This is notated GEA when there is no possible uncertainty about E
(Fig. II-17). The diJference (A - B) of A and B is the set of clouds of A
which do not belong to B. The immediate consequence is A - B = A -
(A n B) = CA(A n B) (Fig. II-18).
We shall stop this borrowing here; however, it will afford a stronger,
more precise conception on the whole, better adapted for the manipu-
lations and arguments which follow.
DISTINCTIVE CHARACTERISTICS OF THE SCREENS
In our desire to create sonic complexes from the temporary accepted
primary matter of sound, sine waves (or their replacements of the Gabor
sort), and to create sonic complexes as rich as but more extraordinary
than natural sounds (using scientifically controlled evolutions on very
general abstract planes), we have implicitly recognized the importance of
three basic factors which seem to be able to dominate both the theoretical
construction of a sonic process and its sensory effectiveness: 1. the density
of the elementary elements, 2. the topographic situation of events on
the screens, and 3. the order or disorder of events.
At first sight then the density of grains or vectors, their topography,
and their degree of order are the indirect entities and aspects perceived by
our macroscopic ears. It is wonderful that the ear and the mind follow
objective reality and react directly in spite of gross inherent or cultural
imperfections. Measurement has been the foundation of the experimental
sciences. Man voluntarily treats himself as a sensory invalid, and it is for this
reason that he has armed himself, justifiably, with machines that measure
other machines. His cars and eyes do measure entities or physical phen-
omena, but they are transformed as if a distorting filter came between im-
mediate perception and consciousness. About a century ago the logarithmic
Fig. 11-15
.'
A
Fig. 11-16
E.
Fig. 11-17
A
Fig. 11-18
~ t
... ::
U
59
A nB " rp
" ' : ' ~ ' . : ~ ; . ( ~ : ~ ~
::;.;.i
AVS
A
B
(A - B)
60 Formalized Music
law of sensation was discovered; until now it has not been contradicted.
But as knowledge never stops in its advance, tomorrow's science will with-
out doubt find not only a greater flexibility and exactitude for this law,
but also the beginnings of an explanation of this distorting filter, which is so
astonishing.
This statistical, but none the less quasi-one-to-one transformation of
excitation into perception has up to now allowed us to argue about physical
entities, such as screens, all the while thinking "perceived events." A
reciprocity of the same kind between perception and its comprehension
permits us to pass from the screens to the consequent distinctive characteris-
tics. Thus the arguments which we shall pursue apply equally well to pure
concepts and to those resulting from perception or sensory events, and we
may take the attitude of the craftsman or the listener.
We have already remarked on the density and the topography of
grains and cells and we have acknowledged the concepts of order and dis-
order in the homogeneous superficial distribution or grains.
We shall examine closely the concept of order, for it is probably hidden
behind the other two. That is to say, density and topography are rather
palpably simplified embodiments of this fleeting and many-sided concept
of disorder.
When we speak of order or disorder we imply first of all "objects" or
"elements." Then, and this is already more complex, we define the very
"elements" which we wish to study and from which we wish to construct
order or disorder, and their scale in relation to ours. Finally we qualify
and endeavor to measure this order or disorder. We can even draw up a list
of all the orders and disorders of these entities on all scales, from all aspects,
for all measurements, even the characteristics of order or disorder of this
very list, and establish anew aspects and measurements.
Take the example of the gases mentioned above. On the molecular
scale (and we could have descended to the atomic level), the absolute values
of the speeds, directions, and distributions in space are of all sorts. We can
distinguish the" elements" which carry order or disorder. Thus if we could
theoretically isolate the element "directions" and assume that there is an
obligation to follow certain privileged directions and not all directions, we
could impose a certain degree of order which would be independent of the
other elements constituting the concept "gas." In the same way, given
enough time, the values of the speeds of a single molecule will be distributed
around a mean value and the size of the deviations will follow Gauss's law.
There we will have a certain order since these values are vastly more
numerous in the neighborhood of their mean than anywhere else, from in-
finitely small to infinitely large.
Let us take another example, more obvious and equally true. A crowd
of 500,000 persons is assembled in a town square. If we examine the group
displacement of this crowd we can prove that it does not budge. However,
each individual moves his limbs, his head, his eyes, and displaces his center
of gravity by a few centimeters in every direction. If the displacements of the
centers of gravity were very large the crowd would break up with yells of
terror because of the multiple collisions between individuals. The statistical
values of these displacements normally lie between very narrow limits which
vary with the density of the crowd, From the point of view of these values as
they affect immobility, the disorder is weak.
Another characteristic of the crowd is the orientation of the faces. If an
orator on a balcony were to speak with a calming effect, 499,000 faces would
look at the balcony and 998,000 ears would listen to the honeyed words. A
thousand or so faces and 2000 ears would be distracted for various reasons:
fatigue, annoyance, imagination, sexuality, contempt, theft, etc. We could
confirm, along with the mass media, without any possible dispute, that
crowd and speaker were in complete accord, that 500,001 people, in fact,
were unanimous. The degree of order that the speaker was after would attain
a maximum for a few minutes at least, and if unanimity were expressed
equally strongly at the conclusion of the meeting, the orator could be per-
suaded that the ideas were as well ordered in the heads of the crowd as in his
own.
We can establish from these two extreme examples that the concept of
order and disorder is basic to a very large number of phenomena, and that
even the definition of a phenomenon or an object is very often attributable to
this concept. On the other hand, we can establish that this concept is
founded on precise and distinct groups of elements; that the scale is impor-
tant in the choice of elements; and finally, that the concept of order or dis-
order implies the relationship between effective values over all possible
values that the elements of a group can possess. This introduces the concept
of probability in the quantitative estimate of order or disorder.
We shall call the number of distinct elements in a group its variety. Vve
shall call the degree of order or disorder definable in a group of elements its
entropy. Entropy is linked with the concept of variety, and for that very
reason, it is linked to the probability of an element in the group. These
concepts are those of the theory of communications, which itself borrows
from the second law of thermodynamics (Boltzmann's theorem H) [15 J.
62 Formalized Music
Variety is expressed as a pure number or as its logarithm to the base 2.
Thus human sex has two elements, male and female, and its variety is 2,
or I bit: I bit = log22.
Let there be a group of probabilities (a group of real numbers p,
positive or zero, whose sum is 1). The entropy H of this group is defined as
H = -K 2Pi log Pi'
If the logarithmic base is 2, the entropy is expressed in bits. Thus if we
have a sequence of heads and tails, the probability of each is to and the
entropy of this sequence, i.e., its uncertainty at each throw, will be I bit. If
both sides of the coin were heads, the uncertainty would b ~ removed and the
entropy H would be zero.
Let us suppose that the advent of a head or a tail is not controlled by
tossing the coin, but by a fixed, univocal law, e.g., heads at each even toss
and tails at each odd toss. Uncertainty or disorder is always absent and the
entropy is zero. If the law becomes very complex the appearance of heads or
tails will seem to a human observer to be ruled by the law of chance, and
disorder and uncertainty will be reestablished. What the observer could do
would be to count the appearances of heads and tails, add up their respective
frequencies, deduce their probabilities, and then calculate the entropy in
bits. If the frequency of heads is equal to that of tails the uncertainty will be
maximum and equal to I bit.
This typical example shows roughly the passage from order to disorder
and the means of calibrating this disorder so that it may be compared with
other states of disorder. It also shows the importance of scale. The intelli-
gence of the observer would assimilate a deterministic complexity up to a
certain limit. Beyond that, in his eyes, the complexity would swing over into
unforeseeability and would become chance or disorder; and the visible (or
macroscopic would slide into the invisible (or microscopic). Other methods
and points of view would be necessary to observe and control the pheno-
mena.
At the beginning of this chapter we admitted that the mind and especi-
ally the ear were very sensitive to the order or disorder of phenomena. The
laws of perception and judgment are probably in a geometrical or logarith-
mic relation to the laws ofcxcitation. Wc do not know much about this, and
we shall again confine ourselves to examining general entities and to tracing
an overall orientation of the poetic processes of a very general kind of music,
without giving figures, moduli, or determinisms. We are still optimistic
enough to think that the interdependent experiment and action of abstract
63
hypotheses can cut biologically into the living conflict between ignorance
and reality(if there is any reality).
Study of Ataxy (order or disorder) on the Plane of a
Cloud of Grains or Vectors
Axis of time: The degree of ataxy, or the entropy, is a function of the
simultaneity of the grains and of the distinct intervals of time between the
emission of each grain. If the variety of the durations of the emissions is weak,
the entropy is also weak. If, for example, in a given (j.t each grain is emitted
at regular intervals of time, the temporal variety will be I and the entropy
zero. The cloud will have zero ataxy and will be completely ordered. Con-
versely, if in a fairly long succession of /::"t the grains are emitted according
to the law P x = 8e-
6X
dx, the degree of ataxy will be much larger. The limit
of entropy is infinity, for we can imagine all possible values of time intervals
with an equal probability. Thus, if the variety is n -+ ro, the probability
for each time interval is PI = lin, and the entropy is
n
H = -K"LP;logPi
i=O
n I I I I 1
H = - K L - log - = - K n - log - = - K log - = K log n
1=0 n n n n n
for n --+ 00, H -+ 00.
This is less true in practice, for a (j.t will never offer a very great variety
of durations and its entropy will be weak. Furthermore a sonic composition
will rarely have more than 100,000 (j.t's, so that H ::s; log 100,000 and
H ::s; 16.6 bits.
Axis offrequencies (melodic): The same arguments are valid here but with
greater restriction on the variety of melodic intervals and on the absolute
frequencies because of the limits of the audible area.
Entropy is zero when the variety of frequencies of grains is I, i.e., when
the cloud contains only one pure sound.
Axis of intensity and density; The above observations are valid. There-
fore, if at the limit, the entropies following the three axes of an element
(j.F(j.G(j.t(j.D are zero, this element will only contain one pure sound of
constant intensity emitted at regular intervals.
In conclusion, a cloud may contain just one single pure sound emitted
at regular intervals of time (see Fig. II-19), in which case its mean entropy
(arithmetic mean of the three entropies) would be zero. It may contain
chaotically distributed grains, with maximum ataxy and maximum mean
64 Formalized Music
entropy (theoretically co). Between these two limits the grains may be
distributed in an infinite number of ways with mean entropies between 0
and the maximum and able to produce both the Marseillaise and a raw,
dodecaphonic series.
t -----------1'
I
I
I
i
Fig. 11-19
f
A single grain emitted at regular intervals of time
Parentheses
GENERAL OBSERVATIONS ON ATAXY
Taking this last possibility as a basis, we shall examine the very general
formal processes in all realms of thought, in all physical and psychic
To this end we shall imagine a "Primary Thing," malleable at will;
capable of deforming instantaneously, progressively, or step-by-step; extend-
ible or retractable; unique or plural; as simple as an electron (!) or as com-
plex as the universe (as compared to man, that is).
It will have a given mean entropy. At a defined time we will cause it to
undergo a transformation. From the point of view of alaxy this transforma-
tion can have one of three effects;
1. The degree of complexity (variety) does not change; the transforma-
tion is neutral; and the overall entropy does not change.
2. The degree of complexity increases and so does the entropy.
3. The transformation is a simplifying one, and the entropy dimin-
ishes.
Thus the neutral transformation may act on and transform: perfect
disorder into perfect disorder (fluctuations), partial disorder into partial
disorder, and perfect order into perfect order.
Multiplicative transformation transforms: perfect disorder into perfect
disorder, partial disorder into greater disorder, and perfect order into partial
disorder.
And simplifying transformation transforms; perfect disorder into partial
disorder, partial order into greater order, and partial order into perfect
order. Fig. II-20 shows these transformations in the form of a kinematic
diagram.
Perfect Degree of order
disorder Max.
Partial
disorder
Perfect
order
Fig. 11-20
entropy
Entropy
max.
Entropy 0
Time
STUDY OF ATAXY AT THE LEVEL OF SCREENS (SET OF CLOUDS)
From the above discussion, a screen which is composed of a set of
cells !:1F!:1G associated with densities during a slice of time !:1t, may be
dissociated according to the two characters of the grains, frequency and
amplitude, and affected by a mean entropy. Thus we can classify screens
according to the criterion of ataxy by means of two parameters of disorder:
the variety of the frequencies and the variety of the intensities. We shall
make an abstraction of the temporal distribution of the grains in !:1t and of
the density, which is implicitly bound up with the varieties of the two
fundamental sizes of the grain. In symbolic form:
Perfect disorder = 00
Partial disorder = n or m
Partial order = m or n
Perfect order = O.
From the point of view of ataxy a screen is formulated by a pair of
entropy values ascribed to a pair of frequencies and intensities of its grains.
Thus the pair (n, co) means a screen whose frequencies have quite a small
entropy (partial order or disorder) and whose intensities have maximum
entropy (more or less perfect disorder).
66 Formalized Music
CONSTRUCTION OF THE SCREENS
We shall quickly survey some of the screens in the entropy table in
Fig. II-21.
Perfect Partial
disorder disorder
F G F G
Perfect
order
F G
Symbol
rF ____ G-t-______ -+-____ __ ro, ro
F G 00, n
F G 00, a
G F n, ro
G F D. ro
F G n,m
F G n,O
G F O,n
F G 0,0
Fig. 11-21. Screen Entropy Table
SCREEN (00, (0)
Description
Unique screen
Infinite number
of screens
Unique screen
Infinite number
of screens
Unique screen,
pure sounds
Infinite number
of screens
Infinite number
of screens
Infinite number
of screens
Unique screen,
pure sound
Diagram Diagram
I
Let there be a very large number of grains distributed at random over
the whole range of the audible area and lasting an interval of time equal to
!1t. Let there also be a grid fine enough so that the average density will not
be more than 30 grains per cell. The distribution law is then given by
Poisson's formula
where d
m
is the mean density and P
k
the probability that there will be k
grains in a cell. If d
m
becomes greater than about 30, the distribution law
will become normal.
Fig. II-22 is an example of a Poisson distribution for a mean density
d
m
= 0.6 grains/cell in a grid of 196 cells for a screen (00, (0).
Thus we may construct the (00, (0) screens by hand, according to the
distributions for the rows and columns, or with suitable computer programs.
.. \ .
I--\--\--!--+-+--\---I--+. -,- - -+-\--\-+-+-1--IH-l--+-I--I-+-+--+-H
H--+-+-+-1- ,..--I-r-- . -+ ... - - - I-- -1- -.J--l-+--l--l--'I--I
I
. .
l-l-+-+-+- 1- _. --.... I-f- I-I-+-H-i-l--l--I--I-+-++, -. +-H-i
. .. I

Fig. 11-22
. .
.
F
67
For a very high mean density the screens in which disorder is perfect
(maximum) will give a very rich sound, almost a white sound, which will
never be identical throughout time. Hthe calculation is done by hand we can
construct a large number of (00, (0) screens from the first (00, (0) screen in
order to avoid work and numerical calculation for each separate screen. To
this end we permute the cells by column and row (see Fig. II-23).

I
Fig. 11-23. Example of Permutation by Columns
Discussion. It is obvious that for a high mean density, the greater the
number of celis, the more the distribution of grains in one region of the
screen tends to regularize (ergodism) and the weaker are the fluctuations
from one cell or cloud to another. But the absolute limits of the density in
the cells in the audible area will be a function of the technical means
available: slide rules, tables, calculating machines, computers, ruled paper,
orchestral instruments, tape recorders, scissors, programmed impulses of
pure sounds, automatic splicing devices, programmed recordings, analogue
converters, etc.
If each cell is considered as a symbol defined by the number of grains k,
the entropy of the screen (for a given fineness of grid) will naturally be
affected by the mean density ofthe grains per cell and will grow at the same
111111
68 Formalized Music
time. It is here that a whole series of statistical experiments will have to
circumscribe the perceptible limits of ataxy for these screens (00, (0) and
even express the color nuances of white sound. It is vcry possible that the
ear classifies in the same file a great number of screens whose entropies vary
tremendously. There would result from this an impoverishment and a
simplification of the communication: physical information ~ perception,
but at least there will be the advantage that the work involved in construct-
ing screens will be considerably reduced.
ALL SCREENS
Starting from a few screens and applying the elementary operations we
can construct all the screens of the entropy table. See Fig. 11-24 for a few
examples. In practice, frequency and intensity filters imitate these elemen-
tary operations perfectly.
~
.
.
I I I ~
1=
--
(00 (0) l n co)
::=======
~
(n'm'1
~
~
~
(n m)
Fig. 11-24
Markovian Stochastic Music-Thcory
69
LINKING THE SCREENS
Up to now we have admitted that any sound or music could be de-
scribed by a number of screens arranged in the lexicographic order of the
pages ofa book. Ifwe represent each screen by a specific symbol (one-to-one
coding), the sound or the music can be translated by a succession of symbols
called a protocol:
abgkab ... bg ...
each letter identifying screens and moments t for isochronous fj.t's.
Without seeking the causes of a particular succession of screens, i.e.,
without entering into either the physical structure of the sound or the logical
structure of the composition, we can disengage certain modes of succession
and species of protocols [16]. We shall quickly review the elementary
definitions.
Any matter or its unique symbol is called a term. Two successive terms
cause a transition to materialize. The second term is called the transform and
the change effected is represented by term A -+ term B, or A ~ B.
A transformation is a collection of transitions. The following example is
drawn from the above protocol:
b g k
g k a
another transformation with musical notes:
1
o
A transformation is said to be closed when the collection of transforms
contains only elements belonging to the collection of terms, for example:
the alphabet,
b c
c d
z
a
70
musical notes,
D E F
G C F B A
musical sounds,
1.
. . '
, . .
Formalized :Music
G A B
E
-1 1. Cloud of sound-points.
2.
glissandi in one
direction
3

3. Network of parallel
glissandi in two
directions .
an infinity of terms,
1
1 2 3
674
456
100 1 2
A transformation is univocal or single-valued (mapping) when each term
has a single transform, for example:
1
hac e
abc d
The following are examples of transformations that are not univocal:
1 h: C
h c
a.
d m, n,p
J
,--5'4---,
d
J))))
b.
, r
r'
r
V
,
L-4: 3--1
c. timbre change ofa group of values
brass
I
clarinets oboes strings timpani
Timbres timpani, timpani, brass oboes strings,
oboes strings bassoon

r
and d. concrete music characteriology [4, 5]
r
vibrated
" l'v1anner" cyclical or
irregular
trembled
trembled
nil or
irregular
cyclical
trembled
irregular
nil or
vibrated or
cyclical
A transformation is a one-to-one mapping when each term has a single
transform and when each transform is derived from a single term, for
example:
l
ab c d
bad c
MATRICAL REPRESENTATION
A transformation:
l
ab c
a c c
can be represented by a table as follows:
.J.
a b c
a
+
0 0
h 0 0 0 or
c 0 .+ +
.J.
a b c
a 1 0 0
b 0 0 0
c 0
This table is a matrix .of the transitions of the collection of terms to a
collection of transforms.
PRODUCT
Let there be two transformations T and U:
and U:
J: C d b
T:
j
ab c d
h d a h
bed
In certain cases we can apply to a term 71 of T a transformation T, then
a transformation U. This is written: U[T(n)], and is the product of the two
transformations T and U, on condition that the transforms of T are terms
of U. Thus, first T: a ---'? b, then U: b ---'? c, which is summarized as
V = UT:a-+c.
72
Formalized Music
To calculate the product applied to all the terms of T we shall use the
following matrical representation:
~
a b c d
t
a b c d
a 0 0 1 0 a 0 0 0 0
T: b 0 0 1 U: b 0 0 0 I
c 0 0 0 0 c 0 1 0 0
d 0 0 0 d 0 0
the total transformation V equals the product of the two matrices T and U
in the order U, T.
U T V
0 0 0 0 0 0 I a 0 0 0 0
0 0 0 I 0 0 I 0 1 0 0
x
0 1 0 0 0 0 0 0 1 0 0 1
0 0 0 I 0 0 0 0 1 0
KINEMATIC DIAGRAM
The kinematic or transition diagram is a graphical expression of
transformation. To draw it each term is connected to its transform by an
arrow pointed at the transform. The representative point of a kinematic dia-
gram is an imaginary point which moves in jumps from term to term
following the arrows of the diagram; for an example see Fig. II-25.
T:
Fig, 11-25
l
A C D I
D D I A
L N P A
NAN N
A transformation is really a mechanism and theoretically all the
mechanisms of the physical or biological universes can be represented by
transformations under five conditions of correspondence:
1. Each state of the mechanism (continuity is broken down into discrete
states as close together as is desired) is in a one-to-one correspondence with
a term of the transformation.
2. Each sequence of states crossed by the mechanism by reason of its
internal structure corresponds to an uninterrupted sequence of the terms of
the transformation.
3. If the mechanism reaches a state and remains there (absorbing or
stationary state), the term which corresponds to this state has no transform.
4. If the states of a mechanism reproduce themselves in the same man-
ner without end, the transformation has a kinematic diagram in closed
circuit.
5. A halt of the mechanism and its start from another state is repre-
sented in the diagram by a displacement of the representative point, which
is not due to an arrow but to an arbitrary action on the paper.
The mechanism is determined when the corresponding transformation
is univocal and closed. The mechanism is not determined when the corre-
sponding transformation is many-valued. In this case the transformation is
said to be stochastic. In a stochastic mechanism the numbers 0 and I in the
transformation matrix must be replaced by relative frequencies. These arc
the alternative probabilities of various transformations. The determined
mechanism is a particular case of the stochastic mechanism, in which the
probabilities of transition are 0 and l.
Example: All the harmonic or polyphonic rules of classical music could
be represented by mechanisms. The fugue is one of the most accomplished
and determined mechanisms. One could even generalize and say that the
avant-garde composer is not content with following the mechanisms of his
age but proposes new ones, for both detail and general form.
If these probabilities are constant over a long period of time, and if they
are independent of the states of origin, the stochastic sequence is called,
more particularly, a Markov chain.
Let there be two screens A and B and a protocol of 50 transitions:
ABABBBABAABABABABBBBABAABABBAABABBABAAABABBA
ABBABBA.
The real freq uencies of the transi tions are:
A -+ B 17 times
A - A 6 times
23 times
B _ II 17 times
B _ 11 10 times
27 limes
A
D
76 Formalized Music
a diminution of the entropy. If melodic or harmonic liaisons are effected
and perceived in the same distribution, unpredictability and entropy are
both diminished.
Rate of ataxy
B
/
c ~
Time
~
E
V
Fig. 11-26
A. The evolution is nil. B. The rate of disorder and the richness
increase. C. Ataxy decreases. D. Ataxy increases and then
decreases. E. Ataxy decreases and then increases. F. The
evolution of the ataxy is very complex, but it may be analyzed
from the first three diagrams.
Thus after the first unfolding of a series of twelve sounds of the tem-
pered scale, the unpredictability has fallen to zero, the constraint is maxi-
mum, the choice is nil, and the entropy is zero. Richness and hence interest
are displaced to other fields, such as harmonies, timbres, and durations, and
many other compositional wiles are aimed at reviving entropy. In fact sonic
discourse is nothing but a perpetual fluctuation of entropy in all its forms
[17].
However, human sensitivity does not necessarily follow the variation
in entropy even if it is logarithmic to an appropriate base. It is rather a
succession or a protocol of strains and relaxations of every degree that often
excites the listener in a direction contrary to that of entropy. Thus Ravel's
Bolero, in which the only variation is in the dynamics, has a virtually zero
entropy after the third or fourth repetition of the fundamental idea. How-
ever, the interest, or rather the psychological agitation, grows with time
through the very fact of this immobility and banality.
All incantatory manifestations aim at an effect of maximum tension
with minimum entropy. The inverse is equally true, and seen from a certain
angle, white noise with its maximum entropy is soon tiresome. It would
seem that there is no correspondence aesthetics ~ entropy. These two
entities are linked in quite an independent manner at each occasion. This
statement still leaves some respite for the free will of the composer even if
this free will is buried under the rubbish of culture and civilization and is
only a shadow, at the least a tendency, a simple stochasm.
The great obstacle to a too hasty generalization is chiefly one oflogical
order; for an object is only an object as a function of its definition, and there
is, especially in art, a near-infinity of definitions and hence a near-infinity
of entropies, for the notion of entropy is an epiphenomenon of the definition.
Which of these is valid? The ear, the eye, and the brain unravel sometimes
inextricable situations with what is called intuition, taste, and intelligence.
Two definitions with two different entropies can be perceived as identical,
but it is also true that the set of definitions of an object has its own degree
of disorder. We are not concerned here with investigating such a difficult,
complex, and unexplored situation, but simply with looking over the
possibilities that connected realms of contemporary thought promise, with a
view to action.
To conclude brief1y, since the applications which follow are more elo-
quent than explanatory texts, we shall accept that a collection or book of
screens can be expressed by matrices of transition probabilities having
parameters. They are affected by a degree of ataxy or entropy which is
calculable under certain conditions. However, in order to render the
analysis and then the synthesis of a sonic work within reach of understanding
and the slide rule, we shall establish three criteria for a screen:
I. TOPOGRAPHIC CRITERION
The position of the cells ~ F ~ G on the audible arca is qualitatively
important, and an enumeration of their possible combinations is capable of
creating a group of well defined terms to which we can apply the concept
of entropy and its calculation.
2. DENSITY CRITERION
The superficial density of the grains of a cell IlFIlG also conslitules a
quality which is immediately perceptible, and we could equalJy welJ define
terms to which the concept and calculation of entropy would be applicable.
3. CRITERION OF PURE ATAXY (defined in relation to the grains of a
screen)
A cell has three variables: mean frequency, mean amplitude, and mean
78
Formalized Music
d
't of the grains. For a screen we can therefore establish three indepen-
ens 1 y . , b bT .
dent or connected protocols, then three matrices of tranSition pr? a 1 ltIes
h
' I a not be coupled Each of the matrices will have Its entropy
w lC 1 mayor my. .
and the three coupled matrices will have a mean entr?py. In the processlOn
of sound we can establish several series of three matnces and hence several
series of mean entropies, their variations constituting the criterion of ataxy.
The first two criteria, which are general and on thc scale of scree?s or
cells will not concern us in what follows. But the third, more conventlOnal
crit:rion will be taken up in detail in the next chapter.
Chapter III
Markovian Stochastic Music-
Applications
In this chapter we will discuss two musical applications: Analogique A, for
string orchestra, and Analogique B, Lor sinusoidal sounds, both composed in
1958-59.
We shall confine ourselves to a simple case in which each of the com-
ponents G, F, D of the screen take only two values, following matrices of
transition probability which will be coupled by means of parameters. In
addition, the choice of probabilities in the matrices will be made in such a
way that we shall have only the regular case, conforming to the chain of
events theory as it has been defined in the work of Maurice Frcchet [14].
It is obvious that richer and more complex stochastic mechanisms are
highly interesting to construct and to put in work, but in view of the con-
siderable volume of calculations which they necessitate it would be useless
to undertake them by hand, but very desirable to program them for the
computer.
Nevertheless, despite the structural simplicity of what follows, the
stochastic mechanism which will emerge will be a model, a standard sub-
jacent to any others that are far more complex, and will serve to catalyze
further studies of greater elaboration. For although we confine ourselves
here to the study of screens as they have been defined in this study (sets of
elementary grains), it goes without saying that nothing prevents the gen-
eralization of this method of structuralization (composition) for definitions
of sonic e n t ~ t i e s of more than three dimensions. Thus, let us no longer suppose
screens, but criteria of definitions of a sonic entity, such that for the timbre,
degree of order, density, variation, and even the criteria of more or less
79
80
Formalized :Music
a.--
Fig. 111-1. Syrmos for 18 strings
Markovian Stochastic Music-Applications 81
complex elementary structures (e.g., melodic and temporal structures of
groups of sounds, and instrumental, spatial, and kinematic structures) the
same stochastic scheme is adaptable. It is enough to define the variations
well and to be able to classify them even in a rough manner.
The sonic result thus obtained is not guaranteed a priori by calculation.
Intuition and experience must always play their part in guiding, deciding,
and testing.
ANALYSIS
(definition of the scheme of a mechanism)
We shall define the scheme of a mechanism as the "analogue" of a
stochastic process. It will serve for the production of sonic entities and for
their transformations over time. These sonic entities will have screens which
will show the following characteristics freely chosen:
1. They will permit two distinct combinations of frequency regions
fo andfl (see Fig. III-2).
Fig. 111-2
o---< __ --______ ---___ - ~ .. F Halfaxis .
I
. .1 of frequencies
in semitones
Audible frequencies
.---------------------:F Halfaxis
I. J
I
of frequencies
in semitones
Audible frequencies
Syrmos, written in 1959, is built on stochastic transformations of
eight basic textures: parallel horizontal bowed notes, parallel
ascending bowed glissandi, parallel descending bowed glissandi,
crossed (ascending and descending) parallel bowed notes,
pizzicato clouds, atmospheres made up of col legno struck notes
with short col legno glissandi. geometric configurations of
convergent or divergent glissandi, and glissando configurations
treated as undevelopable ruled surfaces. The mathematical struc-
ture of this work is the same as that of Ana/ogique A and
Ana/ogique B.
82
Formalized Music
2. They will permit two distinct combinations of intensity regions
(see Fig. III-3).
G
Fig. 111-3
(Phones)
Ul
. ~
"iii
c
Q)
E
Q)
;e
i5
Q)
()
ID
a.
(Phones)
Ul
w
"iii
c
Q)
E
Q)
:0
'"5.
Q)
()
a;
a.
3. They will permit two distinct combinations of density regions
(see Fig. III-4).
(Terts' or sounds/sec) D
(Terts' or sounds/sec) D
Fig. 111-4
'Ternary logarithms
4. Each of these three variables will present a protocol which may be
summarized by two matrices of transition probabilities (MTP).
t
X Y t
X Y
X 0.2 0.8 X 0.85 0.4
(p)
(a)
0.15 0.6
Y 0.8 0.2
y
The letters (p) and (a) constitute the parameters of the (MTP).
MTPF (of frequencies)
t io i1
{-
io i1
io
0.2 0.8 io
0.85 0.4
(a)
(f3)
i1
0.8 0.2 j ~ 0.15 0.6
MTPG (of intensities)
t go gl t go gl
(1')
go
0.2 0.8
( e)
go 0.85 0.4
gl 0.8 0.2 gl
0.15 0.6
MTPD (of densities)
t
do d
1 t
do d
1
(.:\)
do 0.2 0.8 do 0.85 0.4
(p,)
d
1
0.8 0.2 d
1
0.15 0.6
5. The transformations of the variables are indeterminate at the
interior of each (MTP) (digram processes), but on the other hand their
(MTP) will be connected by means ofa determined coupling ofparamcters.
The coupling is given by the following transformations;
l
io .h do d1 go g1 go gl io il do d1
.:\ p, a fJ .:\ p, (3 a I' eye
By these rules we have described thc structure of a mechanism. It is
thus constituted by three pairs of (MTP); (MTPF), (MTPG), (MTPD),
and by the group (eo) of the six couplings of these (MTP).
Significance cif the coupling. Let fo be the state of the frequencies of the
screen at an instant t of the sonic evolution of the mechanism during a slice
of time D.t. Let gl and d
l
be the values of the other variables of the screen
at the moment t. At the next moment, t + D.t, the term io is bound to change,
for it obeys one of the two (MTPF), (a) or (f3). The choice of (a) or (fJ) is
conditioned by the values gi and d
l
of the moment t, conforming to the
transformation of the coupling. Thus!!1 proposes the parameter (a) and d
l
the parameter ((3) simultaneously. In other words the term io must either
remainio or yield its place toil according to mechanism (a) or mechanism
((3). Imagine the term io standing before two urns (a) and ((3), each con-
taining two colors of balls, red for io and blue for iI' in the following
proportions:
Urn (a)
red balls (fo), 0.2
blue balls (f1), 0.8
Urn ((3)
red balls (fo), 0.85
blue balls (fl), 0.15
The choice is free and the termfo can take its successor from either urn (a)
84 Formalized Music
or urn (f3) with a probability equal to 1- (total probabilities).
Once the urn has been chosen, the choice of a blue or a red ball will
have a probability equal to the proportion of colors in the chosen urn.
Applying the law of compound probabilities, the probability thatio from
moment twill remainio at the moment t + 6.t is (0.20 + 0.85)/2 = 0.525,
and the probability that it will change toil is (0.80 + 0.15)/2 = 0.475.
The five characteristics of the composition of the screens have estab-
lished a stochastic mechanism. Thus in each of the slices 6.t of the sonic
evolution of the created mechanism, the three variables ;;, gil d
t
follow a
round of unforeseeable combinations, always changing according to the
three (MTP) and the coupling which connects terms and parameters.
We have established this mechanism without consideration
any of the screen criteria. That is to say, we have implied a topographic
distribution of grain regions at the time of the choice ofio,]1 and go, gl, but
without specifying it. The same is true for the density distribution. We shall
give two examples of very different realizations in which these two criteria
will be effective. But before setting them out we shall pursue further the
study of the criterion of ataxy.
We shall neglect the entropies of the three variables at the grain level,
for what matters is the macroscopic mechanism at the screen leveL The
fundamental questions posed by these mechanisms are, "Where does the
transformation summarized by an (MTP) go? What is its destiny?"
Let us consider the (MTP):
.j, X Y
X 0.2 0.8
Y 0.8 0.2
and suppose one hundred mechanisms identified by the law of this single
(MTP). We shall allow them all to set out from X and evolve freely. The
preceding question then becomes, "Is there a general tendency for the states
of the hundred mechanisms, and if so, what is it?" (See Appendix 11.)
After the first stage the 100X will be transformed into 0.2 (IOOX)
20X, and 0.8 (I OOX) ---+ 80 Y. At the third stage 0.2 of the X's and 0.8 of the
Y's will become X's. Conversely 0.8 of the X's will become Y's and 0.2 of
the Y's will remain Y's. This general argument is true for all stages and can
be written:
X' = 0.2X + 0.8Y
Y' = 0.8X + 0.2 Y.
Markovian Stochastic Music-Applications
85
If this is to be applied to the 100 mechanisms X as above, we shall ha vc:
Mechanisms Mechanisms
Stage X Y
a 100 a
I 20 80
2 68 32
3 39 61
4 57 43
5 46 54
6 52 48
7 49 51
8 50 50
9 50 50
We notice oscillations that show a general tendency towards a station-
ary state at the 8th stage. We may conclude, then, that of the 100 mecha-
nisms that leave from X, the 8th stage will in all probability send 50 to X
and 50 to Y. The same stationary prob;:tbility distribution of the Markov
chain, or the fixed probability vector, is calculated in the following manner:
At equilibrium the two probability values X and Y remain unchanged
and the preceding system becomes
or
X = 0.2X + 0.8Y
Y = 0.8Y + 0.2Y
o = -0.8X + 0.8Y
o = + O.8X - 0.8Y.
Since the number of mechanisms is constant, in this case 100 (or I), one of
the two equations may be replaced at the stationary distribution by
I = X + Y. The system then becomes
a = 0.8X - 0.8Y
I=X+Y
and the stationary probability values X, Yare X = 0.50 and Y = 0.50.
The same method can be applied to the (MTP) (a), which will give us
stationary probabilities X = 0.73 and Y = 0.27.
86 Formalized Music
Another method, particularly interesting in the case of an (MTP) with
many terms, which forces us to resolve a large system oflinear equations in
order to find the stationary probabilities, is that which makes usc of matrix
calculus.
Thus the first stage may be considered as the matrix product of the
(MTP) with the unicolumn matrix I l ~ O I
X: 10.2
Y: 0.8
0.81 x 1
100
1 = 1
20
1.
0.2 0 80
The second stage will be
1
0.2 0.81 1
20
I I 4 +
64
1 1
68
1
0.8 0.2 x 80 = 16 + 16 = 32'
and the nth stage
Now that we know how to calculate the stationary probabilities of a
Markov chain we can easily calculate its mean entropy. The definition of
the entropy of a system is
The calculation of the entropy of an (MTP) is made first by columns
CL Pi = 1), the Pi being the probability of the transition for the (MTP);
then this result is weighted with the corresponding stationary probabilities.
Thus for the (MTP) (a) :
t X Y
X 0.85 0.4
Y 0.15 0.6
The entropy of the states of X will be -0.8510g 0.85 - 0.15 log 0.15 =
0.611 bits; the entropy of the states of Y, - 0.4 log 0.4 - 0.6 log 0.6 =
0.970 bits; the stationary probability of X = 0.73; the stationary proba-
bility of Y = 0.27; the mean entropy at the stationary stage is
Ha = 0.611(0.73) + 0.970(0.27) = 0.707 bits;
and the mean entropy of the (MTP)(p) at the stationary stage is
H p = 0.722 bits.
The two entropies do not differ by much, and this is to be expected,
for if we look at the respective (MTP) we observe that the great contrasts
of probabilities inside the matrix (p) arc compensated by an external equality
of stationary probabilities, and conversely in the (MTP)(a) the interior
quasi-equality, 0.4 and 0.6, succeeds in counteracting the interior contrast,
0.85 and 0.15, and the exterior contrast, 0.73 and 0.27.
At this level we may modify the (MTP) of the three variables};, gi' dj
in such a way as to obtain a new pair of entropies. As this operation is
repeatable we can form a protocol of pairs of entropies and therefore an
(MTP) of pairs of entropies. These speculations and investigations are no
doubt interesting, but we shall confine ourselves to the first calculation made
above and we shall pursue the investigation on an even more general plane.
MARKOV CHAIN EXTENDED SIMULTANEOUSLY FOR};, gb d
t
On p. 83 we analyzed the mechanism of transformation of fa to fa or fl
when the probabilities of the two variables gj and d
j
are given. We can apply
the same arguments for each of the three variables};, gi' dj when the two
others are given.
Example for gj. Let there be a screen at the moment t whose variables
have the values (fa, gl, d
1
). At the moment t + t!.t the value of gl will be
transformed into gl or go' Fromfo comes the parameter (y), and from d
1
comes the parameter (E:).
With (MTP)(y) the probability that gl will remain gi is 0.2. With
(MTP)(e) the probability that gl will remain gl is 0.6. Applying the rules
of compound probabilities and/or probabilities of mutually exclusive events
as on p. 83, we find that the probability thatg
1
will remain gl at the moment
t + tlt under the simultaneous effects offo and d
1
is equal to (0.2 + 0.6) /2 =
0.4. The same holds for the calculation of the transformation from gi into go
and for the transformations of d
1

We shall now attempt to emerge from this jungle of probability com-
binations, which is impossible to manage, and look for a more general
viewpoint, if it exists.
In general, each screen is constituted by a triad of specific values of
the variables F, G, D so that we can enumerate the different screens emer-
ging from the mechanism that we are given (see Fig. III-5). The possible
88 Formalized Music
do do
/ /
go go
/
'"
/
'"
d, d,
fa 1,
~
do
~
do
/ /
g, g,
Fig. 111-5
'" '"
d, d,
combinations arc: (Jagoda), (Jo/!,odj ), (Jo/!'ld
o
), (fogjdj), (J,god
o
), (Jlgod
1
),
(flgjda), (hg1djl; i.e., eight different screens, which, with their protocols,
will make up the sonic evolution. At each moment 1 of the composition we
shall encounter one of these eight screens and no others.
What are the rules for the passage from one combination to another?
Can one construct a matrix of transition probabilities for these eight
screens?
Let there be a screen (iogjdj) at the moment t. Can one calculate the
probability that at the moment t + 8..1 this screen will be transformed into
(fjgjd
o
)? The above operations have enabled us to calculate the probability
thatia will be transformed intoij under the influence of gl and d, and that
gj will remain gl under the influence of ia and d
l
. These operations are
schematized in Fig. III-6, and the probability that screen (Jogld
1
) will be
transformed into (flgld
o
) is 0.114.
Screen at the moment t:
Parameters derived from the coupling
transformations:
Screen at the moment t + !':.t:
Values of probabilities taken from the (MTP) 0.80 0.6 0.4
corresponding to the coupling parameters: 0.15 0.2 0.8
Compound probabilities: 0.475 0.4 0.6
Compound probabilities for independent events: 0.475 . 0.4 . 0.6 = 0.114
Fig. 111-6
We can therefore extend the calculation to the eight screens and construct
the matrix of transition probabilities. It will be square and will have eight
rows and eight columns.
MTPZ
1
A B C 0 E F G H
(tagada) (tagod, ) (fag,do) (fag, d,) (f,gado) (f,god, ) (f,g,d
a
) (f,g,d,)
A (fogodo) 0.021 0.357 0.084 0.189 0.165 0.204 0.408 0.096
B(fogad, ) 0.084 0.089 0.076 0.126 0.150 0.136 0.072 0.144
C(fOg,do) 0.084 0.323 0.021 0.126 0.150 0.036 0.272 0.144
o (fag,d,) 0.336 0.081 0.019 0.084 0.135 0.024 0.048 0.216
E(f,gado) 0.019 0.063 0.336 0.171 0.110 0.306 0.102 0.064
F(f,yod, ) 0.076 0.016 0.304 0.114 0.100 0.204 0.Q18 0.096
G(f,g,d
o
) 0.076 0.057 0.084 0.114 0.100 0.054 0.068 0.096
H(/,g,d, ) 0.304 0.014 0.076 0.076 0.090 0.036 0.012 0.144
Does the matrix have a region of stability? Let there be 100 mecha-
nisms Z whose scheme is summarized by (MTPZ). At the moment t, d
A
mechanisms will have a screen A, dB a screen B, ... , dJi a screen H. At the
moment t + I"!..t all 100 mechanisms will produce screens according to the
probabilities written in (MTPZ). Thus,
0.021 d
A
will stay in A,
0.357 dB will be transformed to A,
0.084 de will be transformed to A,
0.096 d
H
will be transformed to A.
The d
A
screens at the moment t will become d ~ screens at the moment
t + I"!..t, and this number will be equal to the sum of all the screens that will
be produced by the remaining mechanisms, in accordance with the corre-
sponding probabilities.
Therefore:
{
d ~ ~ 0.021d
A
+ O.357d
B
+ 0.084d
e
+
dB = 0.084d
A
+ 0.089d
B
+ 0.076d
e
+
(e
l
) d ~ = 0.084d
A
+ 0.323d
B
+ 0.021d
e
+
dlI = 0.304d
A
+ 0.014d
B
+ 0.076d
e
+
+ 0.096d
H
+ 0.144d
H
+ O.l44d
H
+ 0.144dj{'
At the stationary state the frequency of the screens A, B, C, ... , H will
remain constant and the eight preceding equations will become:
90
{
o = -0.979d
A
+ 0.357d
B
+ 0.084d
e
+
o = 0.084d
A
- 0.911d
B
+ 0.076d
e
+
(e2) 0.084dA + 0.323dB - 0.979de +
o = 0.304d
A
+ 0.014d
B
+ 0.076d
e
+ ...
On the other hand
Formalized Music
+ 0.096d
H
+ 0.144d
H
+ 0.144d
H
- 0.856d
H
Ifwe replace one of the eight equations by the last, we obtain a system
of eight linear equations with eight unknowns. Solution by the classic
method of determinants gives the values:
(e
a
) {dA = 0.17, dB = 0.l3, de = 0.13, dn = 0.11, dE = 0.14, dF = 0.12,
d
G
= 0.10, d
H
= 0.10,
which are the probabilities of the screens at the stationary stage. This
method is very laborious, for the chance of error is very high (unless a
calculating machine is available).
The second method (see p. 85), which is more approximate but
adequate, consists in making all 100 mechanisms Z set out from a single
screen and letting them evolve by themselves. After several more or less
long oscillations, the stationary state, if it exists, will be attained and the
proportions of the screens will remain invariable.
We notice that the system of equations (el) may be broken down into:
1. Two vectors V' and V which may be represented by two unicolumn
matrices:
V'=
and V=
d'
A

d;;
Markovian Stochastic 91
2. A linear operator, the matrix of transition probabilities Z. Conse-
quently system (e
l
) can be summarized in a matrix equation:
To cause all 100 mechanisms Z to leave screen X and evolve "freely"
means allowing a linear operator:
0.021 0.357 0.084 0.189 0.165 0.204 0.408 0.096
0.084 0.089 0.076 0.126 0.150 0.136 0.072 0.144
0.084 0.323 0.021 0.126 0.150 0.036 0.272 0.144
Z=
0.336 0.081 0.019 0.084 0.135 0.024 0.048 0.216
0.019 0.063 0.336 0.171 0.110 0.306 0.102 0.064
0.076 0.016 0.304 0.1l4 0.100 0.204 0.018 0.096
0.076 0.057 0.084 0.114 0.100 0.054 0.068 0.096
0.304 0.014 0.076 0.076 0.090 0.036 0.012 0.144
to perform on the column vector
0
0
V= 100
0
0
in a continuous manner at each moment t. Since we have broken down
continuity into a discontinuous succession of thickness in time b.t, the equa-
tion (e
4
) will be applied to each stage I1t.
Thus at the beginning (moment t = 0) the population vector of the
mechanisms will be Va. After the first stage (moment 0 + b.t) it will be
V' = ZVo; after the second stage (moment 0 + 2b.t) , V" = ZV' = Z2VO;
and at the nth stage (moment nl1t) , V(n) = zn Va. In applying these data to
the vector
0
0
0
va -
0
H-
0
0
0
100
92 Formalized Music Markovian Stochastic Music-Applications 93
after the first stage at the after the second stage at the P
A PH
moment 6.t: moment 2fl.t:
100 2.1 0 35.7
9.6
18.941
0 8.4 100 8.9
14.4
10.934
0 8.4 0 32.3
14.4
14.472
VJ =
0
V ~ =
33.6
V2 =
0
V ~ =
8.1
Vi; = z V ~ =
21.6
V ~ = ZVi; =
11.146
0 1.9 0 6.3
6.4
15.164
0 7.6 0 l.6
9.6
11.954
0 7.6 0 5.7
9.6
8.416
0 30.4 0 1.4
14.4
8.966
Pc P
D
after the third stage at the and after the fourth stage at the
0 8.4 0 18.9
moment 3fl.t: moment 4fl./:
0 7.6 0 12.6
16.860 17.111
100 2.1 0 12.6
lO.867 11.069
vg =
0
V ~ =
1.9
va -
100
Vb =
8.4
13.118 13.792
0 33.6
D-
O 17.1
V;; = zv;, =
13.143
V;; = ZV;; =
12.942
0 30.4 0 1l.4
14.575 14.558
0 8.4 0 11.4
12.257 12.111
0 7.6 0 7.6
8.145 8.238
11.046 10.716 P
E
Pp
Thus after the fourth stage, an average of 17 out of the 100 mechanisms will 0 16.5 0 20.4
have screen A, 11 screen E, 14 screen C, ... , 11 screen H. 0 15.0 0 13.6
If we compare the components of the vector Villi with the values (e
3
) 0 15.0 0 3.6
we notice that by the fourth stage we have almost attained the stationary
Vi =
0
V ~ =
13.5
VO -
0
V ~ =
2.4
state. Consequently the mechanism we have built shows a very rapid abate- 100 11.0
p -
0 30.6
ment of the oscillations, and a very great convergence towards final stability, 0 10.0 100 20.4
the goal (stochos). The perturbation PI!, which was imposed on the mecha- 0 10.0 0 5.4
nism (MPTZ) when we considered that all the mechanisms (here 100) left 0 9.0 0 3.6
from a single screen, was one of the strongest we could create.
Po
Let us now calculate the state of the 100 mechanisms Z after the first
stage with the maximal perturbations P applied.
0 40.8
0 7.2
0 27.2
V8 =
0
V ~ =
4.8
0 10.2
0 1.8
100 6.8
0 l.2
94
Formalized Music
Recapitulation of the Analysis
Having arrivcd at this stagc of the analysis we must take our bearings.
On the level of the screcn cells we now have: ]. partial mechanisms of
transformation for frequency, intensity, and density rangcs, which are
expressed by the (MTPF), (MTPG), (MTPD); and 2. an interaction
between the three fundamental variables F, G, D of the screen (transfor-
mations of the coupling (eo)).
On the level of the screens we now have: 1. eight different screens,
A, B, C, D, E, F, G, H; 2. a general mechanism, the (MTPZ), which sum-
marizes all the partial mechanisms and their interactions; 3. a final state of
equilibrium (the goal, stochos) of the system Z towards which it tends quite
quickly, the stationary distribution; and 4. a procedure ofl:lisequilibrium in
systcm Z with the help of the pcrturbations P which are imposed on it.
SYNTHESIS
Mechanism Z which we have just constructed does not imply a real
evolution of the scrcens. It only establishes a dynamic situation and a
po.tential evolution. The natural process is that provoked by a perturbation
P Imposcd on the system Z and the advancement of this system towards its
goal, its stationary state, once the perturbation has ceased its action. We can
therefore act on this mechanism through the intermediary ofa perturbation
such as P, which is stronger or weaker as the casc may be. From this it is
a brief step to imagining a whole series of successive perturbations
would force the apparatus Z to be displaced towards exceptional
reglOns at odds with its behavior at equilibrium.
In effect the intrinsic value of the organism thus created lies in the fact
it must manifest itself, be. The perturbations which apparently change
Its structure represent so many negations of this existence. And if we create
a succession. of perturbations or negations, on the one hand, and stationary
states or eXIstences on the other, we are only ajJirming mechanism Z. In
words, a.t first we argue positively by proposing and offering as
the eXistence itself; and then we confirm it negatively by opposing
It WIth perturbatory states.
The bi-pole of being a thing and not being this thing creates the whole
-the object which we intcnded to construct at the beginning of Chapter
III.. A dual dialectics is thus at the basis of this compositional attitudc, a dia-
lectics that scts the pace to be followed. The" experimental" sciences are an
of this argument on an analogous plane. An experiment estab-
IIshcs a body of data, a web which it disentangles from the magma of
objective reality with the help of negations and transformations imposed
on this body. The repetition of thesc dual operations is a fundamental
condition on which the whole universe of knowledge rests. To state some-
thing once is not to define it; the causality is confounded with the repetition
of phenomena considered to be identical.
In conclusion, this dual dialectics with which we are armed in order
to compose within the framework of our mechanism is homothetic with that
of the experimental sciences; and we can extend the comparison to the
dialectics of biological beings or to nothing more than the dialectics of
being. This brings us back to the point of departure.
Thus an entity must be proposed and then a modification imposed on
it. It goes without saying that to propose the entity or its modification in
our particular case of musical composition is to give a human observer the
means to perceive the two propositions and to compare them. Then the
antitheses, entity and modification, are repeated enough times for the entity
to be identified.
What does identification mean in the case of our mechanism Z?
Parenthesis. We have supposed in the course of the analysis that 100
mechanisms Z were present simultaneously, and that we were following the
rules of the game of these mechanisms at each moment of an evolution
created by a displacement beyond the stationary zone. We were therefore
comparing the states of 100 mechanisms in a 6.t with the states of these
100 mechanisms in the next t, so that in comparing two successive stages of
the group of 100 simultaneous states, we enumerate 100 states twice. Enumera-
tion, that is, insofar as abstract action implies ordered operations, means to
observe the 100 mechanisms one by one, classify them, and test them; then
start again with 100 at the following stage, and finally compare the classes
number by number. And if the observation of each mechanism necessi-
tates a fraction of time x, it would take 200x of time to enumerate 200
mechanisms.
This argument therefore allows us to transpose abstractly a simultan-
eity into a lexicographic (temporal) succession without subtracting any-
thing, however little, from the definition of transformations engendered by
scheme Z. Thus to compare two successive stages of the 100 mechanisms Z
comes down to comparing 100 states produced in an interval of time 100x
with 100 others produced in an equal interval of time 100x (see Fig. III-7).
MATERIAL IDENTIFICATION OF MECHANISM Z
Identification of mechanism Z means essentially a comparison between
Ul
B
co
'iii
o
o
96
I
k4

I
I !

Formalized Music
Period of 1 stage Period of 1 stage
1- time = 1 OOx = 100x I
1 stage 1 stage
Fig. 111-7
all its possibilities of being: perturbed states compared to stationary states,
independent of order.
Identification will be established over equal periods of time 100x
following the diagram:
Phenomenon: P N ->- E ->- PM ->- E
Time: lOOx 100x lOOx lOOx
in which P N and PM represent any perturbations and E is the state of Z at
equilibrium (stationary state).
An alternation of P and E is a protocol in which IOOx is the unit of
time (IOOx = period of the stage), for example:
P
A
P
A
E E E PH P
a
P
a
E Pc
A new mechanism W may be constructed with an (MTP), etc., which
would control the identification and evolution of the composition over more
general time-sets. We shall not pursue the investigation along these lines for
it would lead us too far afield.
A realization which will follow will use a very simple kinematic diagram
of perturbations P and equilibrium E, conditioned on one hand by the
degrees of perturbation P, and on the other by a freely agreed selection.
Definition of State E and of the Perturbations P
From the above, the stationary state E will be expressed by a sequence
of screens such as:
Protocol E(Z)
ADFFEeBDBeFEFADGeHCCHBEDFEFFECFEHEFFFBC
HDEABADDBADADAHHEGADGAHDADGFBEBGABEEB . '.
To carry out this protocol we shall utilize eight urns [A], [E], [e], [D],
[E], [F], [e], [H], each containing balls of eight different colors, whose
proportions are given by the probabilities of (MTPZ). For example, urn
[G] will contain 40.8'70 red balls A, 7.2'70 orange balls B, 27.2'70 yellow balls
C, 4.8'70 maroon balls D, 10.2'70 green balls E, 1.8% blue balls F, 6.8%
white balls e, and 1.2'70 black balls H. The composition of the other seven
urns can be read from (MTPZ) in similar fashion.
We take a yellow ball e at random from urn [e]. We note the result
and return the ball to urn [G]. We take a green ball E at random from urn
[C]. We note the result and return the ball to urn [C]. We take a black ball
H at random from urn [E], note the result, and return the ball to urn [E].
From urn [H] we take .... The protocol so far is: eCEH . ...
Protocol is obviously
AAAA
Protocol Consider an urn [Y] in which the eight colors of balls
are in the following proportions: 2.1 % color A, 8.4'70 color B, 8.4'70 color
e, 33.6% color D, 1.9% color E, 7.6% color F, 7.6'70 color e, and 30.4%
color H. Aftcr each draw return the ball to urn Y. A likely protocol might
be the following:
GFFGHDDCEHGGHDDHBEHCDDDCeDDDDFDDHHHBF
FHDBHDHHCHHECHDBHHDHHFHDDGDAFHHHDFDG . '.
Protocol Pb The same method furnishes us with a protocol of P' :
EEGFGEFEEFADFEBECGEEAEFBFBEADEFAAEEFH
ABFECHFEBEFEEFHFAEBFFFEFEEAFHFBEFEEE . '.
Protocol PS (vg) :
CCCC .. .
Protocol :
BBBB.
98 Formalized Music
Protocol P6 (V6):
AAADCCECDAACEBAFGBCAAADGCDDCGCADGAAGEC
CAACAAHAACGCDAACDAABDCCCGACACAACACB ...
REALIZATION OF ANALOG/QUE A FOR ORCHESTRA
The instrumental composition follows the preceding exposition point
by point, within the limits of orchestral instruments and conventional
execution and notation. The mechanism which will be used is system Z,
which has already been treated numerically. The choice of variables for the
screens are shown in Figs. III-B, 9, 10.
J 1/
(1
0
)
Regions 0 I I
m lJZ
1r==J I c==J1 .. Frequencies
(semitones) 123
" 5 (;
c.
r,JI,.JJ1 R
(f .. ) Regions O_,.,.---l----+II--..:It.!. I I--:--'4-1 --;:--+---;:;(;:--1-1 Frequencies
7 .:2. 3 4 5 (semitones)
(An = 440 Hz)
Fig. 111-8. Frequencies
Nuances of intensity
Nuances of intensity

t<)

'l-. "l

....

Regions Regions
Fig. 111-9. Intensities
Markovian Stochaslic Music-Applications 99
9
Fig. /11-10. Densities
This choice gives us the partial screens FC (Fig. III-Il) and FD
(Fig. III-12), the partial screens CD being a consequence of FC and FD.
The Roman numerals are the liaison agents between all the cells of the
three planes of reference, FC, FD, and GD, so that the different combinations
(fi., gj, die) which are perceived theoretically are made possible.
For example, let there be a screen (fl, gl' do) and the sonic entity C
3
corresponding to frequency region no. 3. From the partial screens above,
this entity will bc the arithmetic sum in three dimensions of the grains of
cells I, II, and III, lying on frequency region no. 3. C
3
= I + II + III.
The dimensions of thc cell corresponding to I are: 6.F = region 3,
/1G = region 1, 6.D = region 2. The dimensions of the cell corresponding
to II are: 6.F = region 3, 6.G = region 2, 6.D = region 1. The dimensions
of the cell corresponding to III are: 6.F = region 3, 6.C = region 2,
/1D = region 1. Consequently in this sonic entity the grains will have fre-
quencies included in region 3, intensities included in regions 1 and 2, and they
will form densities included in regions I and 2, with the
set forth above.
"
I '
r';.,::/
100
2
G
,
2
1
:u
I
II
III
.lY
I
.u
III
.t 23 I( S-,;F
(;1/0)
Fig. 111-11. Partial Screens for FG
[)
3
f
f)
3
2.
I
II
,
P
1lI
J!l
1
D
/0
Fig. 111-12. Partial Screens for FD
2.
c;
3
2
1
D
3
2
;
P
3
f
1l
I
I
Il
Formalized Music
lJI
n
.1I
or
I II
JIf
..N
1
U
III
n.
101
The eight principal screens A, B, C, D, E, F, G, H which derive from
the combinations in Fig. 111-5 are shown in Fig. 111-13. The duration tJ..t
of each screen is 1.11 sec. (I half note = 54 MM). Within this duration the
densities of the occupied cells must be realized. The period of time necessary
for the exposition ofthe protocol of each stage (ofthe protocol at the station-
ary stage, and of the protocols for the perturbations) is 30tJ..t, which becomes
15 whole notes (I whole note = 27 MM).
Screen A fo'o de)
.,
f
#
J
I
f
I!
Screen C (I .. d.,)
#
f
II
oJ
I
" 1l
Screen E (It fa do)
.I
I
I')
J
I
;+{

9
.N
Screen G /1, (1 do)
I
I
I-P
;.,."
'+11
j
r

lJ[.
y
.lY
" /Jl
,
lP
5'
N
Eo El O2 DD3 c. B. As
Fig. 111-13
.II
f
PI
9
.I
.3
.ll
Screen D (ioi""dl
.3
1L
9
I
Screen F (f,l ..
II
f
fP
.9
r

'+1
.,
N
Screen H (If 11 tid
.I
f
N
.UJ
'f+JJl
.9
I
1
IY
NOTE: The numbers written in the cells are the mean densities in grains/sec.
-I
.0
oJ
./II
J
1ll
.,
B
--r--
102 Formalized Music
The linkage of the perturbations and the stationary state of (MTPZ) is
given by the following kinematic diagram, which was chosen for this
purpose;
Fig. 111-14. Bars 105-15 of AnaJogique A
103
Fig. III-14, bars 105-15 of the score of Analogique A, comprises a section
of perturbations p ~ and P ~ . The change of period occurs at bar 109. The
disposition of the screens is given in Fig. III-IS. For technical reasons
screens E, F, G, and H have been simplified slightly.
105 109 115
. .. I BB I BB I BS I BB I AA I GE I CC I AA I CA I AH I
End of the period of ->I+- Beginning of return to cqui-
perturbation P ~ librium (perturbation P ~ )
Fig. 111-15
Analogique A replaces elementary sinusoidal sounds by very ordered
clouds of elementary grains, restoring the string timbres. In any case a
realization with classical instruments could not produce screens having a
timbre other than that of strings because of the limits of human playing.
The hypothesis of a sonority of a second order cannot, therefore, be con-
firmed or invalidated under these conditions.
On the other hand, a realization using electromagnetic devices as
mighty as computers and adequate converters would enable one to prove
the existence of a second order sonority with elementary sinusoidal grains
or grains of the Gabor type as a base.
While anticipating some such technique, which has yet to be developed,
we shall demonstrate how more complex screens are realizable with the
resources of an ordinary electroacoustic studio equipped with several mag-
netic tapes or synchronous recorders, filters, and sine-wave generators.
ELECTROMAGNETIC MUSIC (sinusoidal sounds)-EXAMPLE
TAKEN FROM ANALOG/QUE B
We choose; 1. Two groups offrequency regionsfo,fl' as in Fig. III-16.
The protocols of these two groups will be such that they will obey the
preceding (MTP)'s;
t fo fl t
fo fl
fo 0.2 0.8 fo 0.85 0.4
(a)
fl 0.8 0.2
(f3)
fl 0.15 0.6
in which Ca) and (f3) are the parameters.
r JI OJ
B K 1'T

rr--""1, ,r--I, r--r
Hz
(12 til: 178 3fS ?M It'"
Regions
,
2 j S 1 8 9 Ie II
(/.)
I
I' f!JF-,
I r:::::=l
I I, r-----,I I
Bq
If I
!'e
26",
Z S 6 ,
8 9 ID 1/
"

(h)
Fig. 111-16
2. Two
groups of intensity regions go, gIl
as in Fig. III-l7. The
protocols of this group will again
obey the same (MTP),s with their
parameters (y) and (e):
t go gl
t go gl
go 0.2 0.8
go 0.85 0.4
(y)
0.8 0.2
(e)
gl gl 0.15 0.6
G
G

0
0
'"

'"

0
0

'Il

"l
"l
g. 0

....
..
0
1>4 ...

\g Q

...
"I-
"0' 0 Ei1'

'" <I> <I>
C
'"
<I> <I>
0 c c
'" 'c, 0 0 c
'c,
Q
'"

0:: "-
'" c: "-
(6"0 )

Fig. 111-17
.
F
3. Two groups of density regions do, d
1
, as in Fig. III-IS. The protocols
of this group will have the same (MTP)'s with parameters (A) and (p.):
(,Xl
Fig. 111-18
{. do d
1
{. do d
1
0.2
O.S
O.S
0.2
0.85
0.15
0.4
0.6
This choice gives us the principal screens A, B, C, D, E, F, G, H, as
shown in Fig. 1II-19. The duration I1t of each screen is about 0.5 sec.
The period of exposition of a perturbation or of a stationary state is about
15 sec.
We shall choose the same protocol of exchanges between perturba-
tions and stationary states of (MTPZ), that of Analogique A.
The screens of Analogique B calculated up to now constitute a special
choice. Later in the course of this composition other screens will be used
more particularly, but they will always obey the same rules of coupling
and the same (MTPZ). In fact, if we consider the combinations of regions
of the variable]; of a screen, we notice that without tampering with the
name of the variable]; its structure may be changed.
Screen A
po (0 do)
Screen C
(1.,#.,c4)
Screen D
(hI, el,)
Screen F
(.1,/0.1.,)
Screen G
(I.".,
Screen H
(Tf if tit)
Fig. 111-19
t;
i{
J
:/
f
ii

3
:I.
-I
G
4-

:/
-I
C
I
Ii
2.
Ul.
.J
I

l!

0
I
a
I
0
I
3
I
I
0
N.
of

.3
Ii
2.
lU
<>

..
11
Ii
2.
'fllIl
4
III
oJ
K
f
.!l
l' .r
Y fl/I.
:z.
Xl1i
.f'
.J
,If
r

0
]j[
oF
.FlO
1
.D
1-
(I oJ
.DZ LY
.2
YlII
.J
nz
.r
.& Yl
<>
X
.r
FJlL

Yll
0 of'
1'l N
oF
:x:
C!
X
f
II
i
YlL
,f'
Yl
:[
.;
IX.
f
Ii
"
Xl
oJ
YlI

0
X
f
R
Iw:A
of'
X
Thus forlo we may have the regions shown in Fig. 1II-20. The Roman
numerals establish the liaison with the regions of the other two variables.
I 11 81

Hz e/i
Regions ( Z J
Fig. 111-20
7'"
tJ 9
ft'"
'" "
But we could have chosen another combinationfo, as in Fig. III-21.
Hz 8'1
Regions f Z
Fig. 111-21
I
IfB
S
I/JSO 51"" III."
It. "3 ".5' ""
(10)
This prompts the question: "Given n divisions !1F (regions on F) what
is the total number of possible combinations of !1F regions?
1st case. None of the n areas is used. The screen corresponding to this
combination is silent. The number of these combinations will be
n!
(n _ OlIO! (=1).
2nd case. One of the n areas is occupied. The number of combinations
will be
n!
(n - 1)!l!
3rd case. Two of the n areas are occupied. The number of combinations
will be
n!
(n - 2)!2!
mth case. m of the n areas are occupied. The number of combinations
will be
n!
(n - m) !m!
FIG. 111-19: The Arabic numbers above the Roman numerals in the cells indicate the density in
logarithmic units. Thus cell (10.1) will have a donsityof [(t09 1.3/1093) + 5],orts. which is 315.9
grains/sec on the average .
107
F
F
nth case. n of the areas are occupied. The number of the combinations
will be
n!
(n-n)!n!
The total number of combinations will be equal to the sum of all the
preceding:
n! n! n!
(n - O)!O! + (n - I)!I! + (n - 2)!2! +
+ n! + n! = 2n
[n - (n - l)]!(n - I)! (n - n)!n!.
The same argument operates for the other two variables of the screen.
Thus for the intensity, if k is the number of available regions t1C, the total
number of variables gj will be 2\ and for the density, if T is the number of
available regions t1D, the total number of variables d
j
will be 2r.
Consequently the total number of possible screens will be
T = 2(n+"+r).
In the case of Analogique B we could obtain 2(16+4+7) = 227 =
134,217,728 different screens.
Important comment. At the start of this chapter we would have accepted
the richness of a musical evolution, an evolution based on the method of
stochastic protocols of the coupled screen variables, as a function of the
transformations of the entropies of these variables. From the preceding
calculation, we now see that without modifying the entropies of the (MTPF),
(MTPG), and (MTPD) we may obtain a supplementary subsidiary evo-
lution by utilizing the different combinations of regions (topographic
criterion).
Thus in Analogique B the (MTPF), (MTPG), and (MTPD) will not
vary. On the contrary, in time the};, gj, dj will have new structures, corol-
laries of the changing combinations of their regions.
Complementary Conclusions about Screens and Their
Transformations
1. Rule. To form a screen one may choose any combination of regions
on F, C, and D, the};, gj, d}c.
2. Fundamental Criterion. Each region of one of the variables F, G, D
must be associable with a region corresponding to the other two variables
in all the chosen couplings. (This is accomplished by the Roman numerals.)
3. The preceding association is arbitrary (free choice) for two pairs,
but obligatory for the third pair, a consequence of the first two. For example,
the associations of the Roman numerals of!; with those of g; and with those
of d" are both free; the association orthe Roman numerals of gj with those
of d" is obligatory, because of the first two associations.
4. The components!;, g;, d" of the screens generally have stochastic
protocols which correspond, stage by stage.
5. The (MTP) of these protocols will, in general, be coupled with the
help of parameters.
6. If F, C, D are the "variations" (number of components);, gj, dj ,
respectively) the 'maximum number of couplings between the components
and the parameters of (MTPF), (MTPG), (MTPD) is the sum of the
products CD + FC + FD. In an example from Analogique A or B:
F = 2 (fo and!l)
C = 2 (go and gl)
D = 2 (do and d
1
)
the parameters of the (MTP)'s are: a, f3
y, e
A, JL
and there are 12 couplings:
1
10 11 10 11
yeA JL
go gl go gl do d1 do d1
~ a A JL a f3 y e
Indeed, FC + FD + CD = 4 + 4 + 4 = 12.
7. IfF, C, D are the "variations" (number of components);, gj, db
respectively), the number of possible screens T is the product FCD. For
example, if F = 2 (fo and 11), C = 2 (go and g1), D = 2 (do and d
1
),
T = 2 x 2 x 2 = 8.
8. The protocol of the screens is stochastic (in the broad sense) and can
be summarized when the chain is ergodic (tending to regularity), by an
(MTPZ). This matrix will have FCD rows and FCD columns.
SPATIAL PROJECTION
No mention at all has been made in this chapter of the spatialization
of sound. The subject was confined to the fundamental concept of a sonic
complex and of its evolution in itself. However nothing would prevent
broadening of the technique set out in this chapter and "leaping" into
space. We can, for example, imagine protocols of screens attached to a
particular point in space, with transition probabilities, space-sound coup-
lings, etc. The method is ready and the general application is possible,
along with the reciprocal enrichments it can create.
Chapter IV
Musical Strategy-Strategy, Linear
Programming, and Musical
Composition
Before passing to the problem of the mechanization of stochastic music by
the use of we shall take a stroll in a more enjoyable realm, that
of games, theIr theory, and application in musical composition.
AUTONOMOUS MUSIC
The musical composcr establishes a scheme or pattern which the con-
ductor and the instrumentalists are called upon to follow more or less rigor-
ously. From the final details-attacks, notes, intensities, timbres, and styles of
performance-to the form of the whole work, virtually everything is written
And even in the case where the composer leaves a margin of
ImprovIsatIOn to the conductor, the instrumentalist, the machine, or to all
th.ree together, the unfolding of the sonic discourse follows an open line
WIthout loops. The score-model which is presented to thcm oncc and for
all does not givc risc to any conflict other than that betwecn a "good" per-
formancc in the technical sense, and its "musical expression" as desired or
suggcsted by the writer of the score. This opposition between the sonic
the symbolic schema which plots its course might be called
znternal conflict; and the role of the conductors, instrumentalists, and their
is to control the output by feedback and comparison with the
mput sIgnals, a role analogous to that of servo-mechanisms that reproduce
profiles by such means as grinding machines. In general we can state that
110
Strategy, Linear Programming, and Musical Composition
111
the nature of thc tcchnical oppositions (instrumental and conductorial) or
even those relating to the aesthetic logic of the musical discourse, is internal
to the works writtcn until now. Thc tensions arc shut up in the score even
when marc or less dcfincd stochastic processes are utilized. This traditional
class of internal conflict might be qualified as autonomous music.
Fig.IV-1
1. Conductor
2. Orchestra
3. Score
4. Audience
HETERONOMOUS MUSIC
It would be interesting and probably very fruitful to imagine another
class of musical discourse, which would introduce a concept of external
conflict between, for instance, two opposing orchestras or instrumentalists.
One party's move would influence and condition that ofthe other. The sonic
discourse would then be identified as a very strict, although often stochastic,
succession of sets of acts of sonic opposition. These acts would derive from
both the will of the two (or more) conductors as well as from the will of the
composer, all in a higher dialectical harmony.
Let us imagine a competitive situation between two orchestras, each
having one conductor. Each of the conductors directs sonic operations
against the operations of the other. Each operation represents a move or a
tactic and the encounter between two moves has a numerical and/or a
qualitative value which benefits one and harms the other. This value is
written in a grid or matrix at the intersection of the row corresponding to
move i of conductor A and the column corresponding to movej of conductor
B. This is the partial score ij, representing the paymcnt one conductor gives
the other. This game, a duel, is defined as a two-person zero-sum game.
The external conflict, or heteronomy, can take aU sorts of forms, but can
always be summarized by a matrix of payments ij, conforming to the mathe-
matical theory of games. The theory demonstrates that t ~ l e r e is an optimum
way of playing for A, which, in the long run, guarantees him a minimum
advantage or gain over B whatever B might do; and that conversely there
exists for B an optimum way of playing, which guarantees that his disad-
vantage or loss under A whatever A might do will not exceed a certain
maximum. A's minimum gain and B's maximum loss coincide in absolute
value; this is called the game value.
The introduction of an external conflict or heteronomy into music is not
entirely without precedent. In certain traditional folk music in Europe and
other continents there exist competitive forms of music in which two instru-
mentalists strive to confound one another. One takes the initiative and
attempts either rhythmically or melodically to uncouple their tandem
arrangement, all the while remaining within the musical context of the
tradition which permits this special kind of impro\'isation. This contra-
dictory virtuosity is particularly prevalent among the Indians, especially
among tabla and sarod (or sitar) players.
A musical heteronomy based on modern science is thus legitimate even to
the most conformist eye. But the problem is not the historical justification of
a new adventure; quite the contrary, it is the enrichment and the leap
forward that count. Just as stochastic processes brought a beautiful gen-
eralization to the complexity of linear polyphony and the deterministic
logic of musical discourse, and at the same time disclosed an unsuspected
opening on a totally asymmetric aesthetic form hitherto qualified as non-
sense; in the same way heteronomy introduces into stochastic music a comple-
ment of dialectical structure.
We could equally well imagine setting up conflicts between two or more
instrumentalists, between one player and what we agree to call natural
environment, or between an orchestra or several orchestras and the public.
But the fundamental characteristic of this situation is that there exists a gain
Strategy, Linear Programming, and Musical Composition 113
and a loss, a victory and a defeat, which may be expressed by a moral or
material reward such as a prize, medal, or cup for one side, and by a penalty
for the other.
A degenerate game is one in which the parties play arbitrarily following
a more or less improvised route, without any conditioning for conflict, and
therefore without any new compositional argument. This is a false game.
A gambling device with sound or lights would have a trivial sense if it
were made in a gratuitous way, like the usual slot machines and juke boxes,
that is, without a new competitive inner organization inspired by any
heteronomy. A sharp manufacturer might cash in on this idea and produce
new sound and light devices based on heteronomic principles. A less trivial
use would be an educational apparatus which would require children (or
adults) to react to sonic or luminous combinations. The aesthetic interest,
and hence the rules of the game and the payments, would be determined by
the players themselves by means of special input signals.
In short the fundamental interest set forth above lies in the mutual
conditioning of the two parties, a conditioning which respects the greater
diversity of the musical discourse and a certain liberty for the players, but
which involves a strong influence by a single composer. This point of view
may be generalized with the introduction of a spatial factor in music and
with the extension of the games to the art of light.
In the field of calculation the problem of games is rapidly becoming
difficult, and not all games have received adequate mathematical clar-
ification, for example, games for several players. We shall therefore con-
fine ourselves to a relativcly simple case, that of the two-person zero-sum
game.
ANALYSIS OF DUEL
This work for two conductors and two orchestras was composed in
1958-59. It appeals to relatively simple concepts: sonic constructions put into
mutual correspondence by the will of the conductors, who are themselves
conditioned by the composer. The following events can occur:
Event I: A eluster of sonic grains such as pizzicati, blows with the
wooden part of the bow, and very brief area sounds distributed stochastic-
ally.
Event II: Parallel sustained strings with fluctuations.
Event III: Networks of intertwined string glissandi.
Event IV: Stochastic percussion sounds.
114
Event V: Stochastic wind instrument sounds.
Event VI: Silence.
Formalized Music
Each of these events is written in the score in a very precise manner and
with sufficient length, so that at any moment, following his instantaneous
choice, the conductor is able to cut out a slice without destroying the iden-
tity of the event. We therefore imply an overall homogeneity in the writing
of each event, at the same time maintaining local fluctuations.
We can make up a list of couples of simultaneous events x, y issuing
from the two orchestras X and Y, with our subjective evaluations. We can
also write this list in the form of a qualitative matrix (M
I
).
Table of Evaluations
Couple
(x,y) = (y,x)
(I, I)
(I, II) = (II, I)
(I, III) = (III, I)
(I, IV) = (IV, I)
(I, V) = (V, I)
(II, II)
(II, III) = (III, II)
(II, IV) = (IV, II)
(II, V) = (V, II)
(III, III)
(III, IV) = (IV, III)
(III, V) = (V, III)
(IV, IV)
(IV, V) = (V, IV)
(V, V)
Evaluation
passable
good
good +
passable +
very good
passable
passable
good
passable +
passable
good +
good
passable
good
passable
(p)
(g)
(g+)
(p+)
(g+ +)
(p)
(p)
(g)
(p+)
(p)
(g+)
(g)
(p)
(g)
(p)
Conductor Y
I II
I p g
- --
II g
P
----
Conductor X III g+
p
I-----
IV p+ g
----
V g++ p+
Maximum per g+ + g
column
III IV V
g+ p+ g++
----I-
P
g p+
-----
p g+
g
----I-
g+
p g
----
I-
g g p
Minimum
per row
p
p
p
p
p
115
In (Ml) the largest minimum per row and the smallest maximum per
column do not coincide (g # p), and consequently the game has no saddle
point and no pure strategy. The introduction of the move of silence (VI)
modifies (M
l
), and matrix (M
2
) results.
Conductor Y
I II III IV V VI
I p g
g++ g+ g+ p p
- ---------
II g p p g p+ p p
- ---------
III
g++
p p g+
g p p
Conductor X -----------
IV g+ g g+
p g p p
----------:-
V g+ p+
g g p p p
----------I-
VI
P P P P P
p-
g++ g
This time the game has several saddle points. All tactics are possible,
but a closer study shows that the conflict is still too slack: Conductor Y is
interested in playing tactic VI only, whereas conductor X can choose freely
among I, II, III, IV, and V. It must not be forgotten that the rules of this
matrix were established for the benefit of conductor X and that the game in
this form is not fair. Moreover the rules arc too vague. In order to pursue
our study we shall attempt to specify the qualitative values by ordering them
on an axis and making them correspond to a rough numerical scale:
p- P
p+ g g+ g+ +
I I I I I I
o 1 2 3 4 5
If, in addition, we modify the value of the couple (VI, VI) the matrix
becomes (M3)'
Conductor Y
I II III IV V VI
I 1 3 5 4 4 1
- - - - - 1-
II 3 1 I 3 2 1
- - - - - -
III 5 1 1 4 3 1
Conductor X - - - - - -
IV 4 3 4 1 3 1
- - - - - 1-
V 4 2 3 3 1 1
- - - - - I-
VI I 1 1 1 1 3
5 3 544 3
(M3) has no saddle point and no recessive rows or columns. To find
the solution we apply an approximation method, which lends itself easily to
computer treatment but modifies the relative equilibrium of the entries as
little as possible. The purpose of this method is to find a mixed strategy;
that is to say, a weighted multiplicity of tactics of which none may be zero.
It is not possible to give all the calculations here [21], but the matrix that
results from this method is (M4l, with the two unique strategies for X and
for Y written in the margin of the matrix. Conductor X must therefore play
117
Conductor Y
I II III IV V VI
I 2 3 4 2 3 2 18
--'- - - --- -
II 3 2 2 3 3 2 4
- - - --- 1-
III 4 2 1 4 3 1 5
Conductor X - - - --- - (M4)
IV 2 4 4 2 2 2 5
I--- - - --- 1-
V 3 2 3 3 2 2 11
I--- - - --- 1-
VI 2 2 1 2 2 4 15
9 6 8 12 9 14 58 Total
tactics I, II, III, IV, V, VI in proportions 18/58, 4158, 5/58, 5/58, 11/58,
15/58, respectively; while conductor Y plays these six tactics in the propor-
tions 9/58,6/58,8/58, 12/58,9/58, 14/58, respectively. The game value from
this method is about 2.5 in favor of conductor X (game with zero-sum but
still not fair).
We notice immediately that the matrix is no longer symmetrical about
its diagonal, which means that the tactic couples are not commutative,
e.g., (IV, II = 4) f; (II, IV = 3). There is an orientation derived from
the adjustment of the calculation which is, in fact, an enrichment of the
game.
The following stage is the experimental control of the matrix.
Two methods are possible:
1. Simulate the game, i.e., mentally substitute oneself for the two
conductors, X and Y, by following the matrix entries stage by stage, without
memory and without bluff, in order to test the least interesting case.
Game value: 52/20 = 2.6 points in X's favor.
2. Choose tactics at random, but with frequencies proportional to the
marginal numbers in (M
4
).
Game value: 57121 = 2.7 points in X's favor.
We now establish that the experimental game values are very close to
the value calculated by approximation. The sonic pI"ocesses derived from
the two experiments are, moreover, satisfactory.
We may now apply a rigorous method for the definition of the optimum
strategies for X and Y and the value of the game by using methods oflinear
programming, in particular the simplex method [22]. This method is based
on two theses:
I. The fundamental theorem of game theory (the" minimax theorem")
is that the minimum score (maximin) corresponding to X's optimum
strategy is always equal to the maximum score (minimax) corresponding to
Y's optimum strategy.
2. The calculation of the maximin or minimax value, just as the
probabilities of the optimum strategies ofa two-person zero-sum game, comes
down to the resolution of a pair of dual problems of linear programming
(dual simplex method).
Here we shall simply state the system oflinear equations for the player
of the minimum, Y. LetYl,Y2,Ya,Y4,Ys,Ys be the probabilities corresponding
to tactics I, II, III, IV, V, VI of Y; Y7' Ys, Yg, YlO' Yll' Y12 be the" slack"
variables; and v be the game value which must be minimized. We then have
the following liaisons:
Yl + Y2 + Ys + Y4 + Ys + Ys = 1
2Yl + 3Y2 + 4Ya + 2Y4 + 3Y5 + 2Ye + Y7 = V
3Yl + 2Y2 + 2Ya + 2Y4 + 3ys + 2Y6 + Ys = V
2Yl + 4Y2 + 4Ya + 2Y4 + 2Y5 + 2YB + Y9 = V
3Yl + 2Y2 + 3Ya + 3Y4 + 2Y5 + 2Y6 + YIO = V
2Yl + 2Y2 + Ya + 2Y4 + 2Y5 + 4YB + Yll = V
4Yl + 2Y2 + Ya + 4Y4 + 3Y5 + Ye + Y12 = V.
To arrive at a unique strategy, the calculation leads to the modification
of the score (III, IV = 4) into (III, IV = 5). The solution gives the follow-
ing optimum strategies:
For X For Y
Tactics Probabilities Tactics Probabilities
I 2/17 I 5/17
II 6/17 II 2/17
III 0 III 2/17
IV 3/17 IV 1/17
V 2/17 V 2/17
VI 4/17 VI 5/17
and for the game value, v = 42/17 ~ 2.47. We have established that X
must completely abandon tactic III (probability of III = 0), and this we
must avoid.
Modifying score (II, IV = 3) to (II, IV = 2), we obtain the following
optimum strategies:
For X For Y
Tactics Probabilities Tactics Probabilities
I 14/56 I 19/56
II 6/56 II 7/56
III 6/56 III 6/56
IV 6/56 IV 1/56
V 8/56 V 7/56
VI 16/56 VI 16/56
and for the game value, v = 138/56 ~ 2.47 points.
Although the scores have been modified a little, the game value has,
in fact, not moved. But on the other hand the optimum strategies have
varied widely. A rigorous calculation is therefore necessary, and the final
matrix accompanied by its calculated strategies is (M5)'
Conductor Y
I II III IV V VI
I 2 3 4 2 3 2 14
i--- - - - 1-
II 3 2 2 2 3 2 6
i-- - - - - 1-
III 4 2 1 5 3 1 6
Conductor X i-- - - - - 1- (Ms)
IV 2 4 4 2 2 2 6
- - - - - 1-
V 3 2 3 3 2 2 8
,.--- - - - - 1-
VI 2 2 1 2 2 4 16
19 7 6 7 16 56 Total
By applying the elementary matrix operations to the rows and columns
in such a way as to make the game fair (game value = 0), we obtain the
equivalent matrix eM
6
) with a zero game value.
Conductor X
I
II
III
IV
V
VI
I
-13
--
15
--
43
--
-13
t--
15
I-
- 13
II
15
--
-13
--
-13
--
43
--
- 13
--
- 13
Conductor Y
III IV
43 - 13
----
- 13 -13
----
-41 71
----
43 -13
----
15 15
----
-41 -13
V VI
15 -13
--I-
15 -13
--I-
15 -41
--I-
-13 -13
--I-
- 13 -13
--I-
-13
-2_
56
43
lJi.
S 6
II
56
-l1._
S6
_.lL
56
lJi.
56
As this matrix is difficult to read, it is simplified by dividing all the
scores by + 13. It then becomes (M7) with a game value v = - 0.07, which
Conductor Y
I II III IV V VI
I -1 +1 +3 -1 +1 -1
14
56
\--- --------I-
II +1 -1 -1 -I +1 -1
s ~
I---------I-
+3 -1 -3 +5 +1 -3 III
6
56
Conductor X I---------I- (M7)
-I +3 +3 -1 -1 -1 IV
- ~
S6
----------I-
+1 -1 +1 +1 -I -1 V
_.lL
S6
I---------I-
-I -1 -3 -1 -I +3 VI
16
56
means that at the end of the game, at the final score, conductor Y should
give 0.07m points to conductor X, where m is the total number of moves.
If we convert the numerical matrix (M7) into a qualitative matrix
according to the correspondence:
-1 -3
T
p
+1
1
I
p+ g g+
+3 +5
T
we obtain (Ma), which is not very different from (lvf
2
), except for the silence
couple, VI, VI, which is the opposite of the first value. The calculation is
now finished.
P
p+ g+ +
P
g
P
p+
P P P
p+
P
---
g++
P P
g++ p+
P
P
g++ g++
P P P
---
p+
P
p+ p+
P P
P P P P P
g++
Mathematical manipulation has brought about a refinement of the
duel and the emergence of a paradox: the couple VI, VI, characterizing
total silence. Silence is to be avoided, but to do this it is necessary to aug-
ment its potentiality.
It is impossible to describe in these pages the fundamental role of the
mathematical treatment of this problem, or the subtle arguments we are
forced to make on the way. We must be vigilant at every moment and over
every part of the matrix area. It is an instance of the kind of work where
detail is dominated by the whole, and the whole is dominated by detail. It
was to show the value of this intellectual labor that we judged it useful to
set out the processes of calculation.
The conductors direct with their backs to each other, using finger or
light signals that are invisible to the opposing orchcst;a. If the conductors
use illuminated signals operated by buttons, the successive partial scores
can be announced automatically on lighted panels in the hall, the way the
score is displayed at football games. If the conductors just usc their fingers,
then a referee can count the points and put up the partial scores manually
so they are visible in thc hall. At the end of a certain number of exchanges
or minutes, as agreed upon by the conductors, one of the two is declared
the winner and is awarded a prize.
Now that the principle has been set out, we can envisage the interven-
tion of the public, who would be invited to evaluate the pairs of tactics of
conductors X and Y and vote immediately on the make-up of the game
matrix. The music would then be the result of thc conditioning of the
composer who established the musical score, conductors X and Y, and the
public who construct the matrix of points.
RULES OF THE WORK STRATiGJE
The two-headed flow chart of Duel is shown in Fig. IV-2. It is equally
valid for Strategie, composed in 1962. The two orchestras are placed on
either side of the stage, the conductors back-to-back (Fig. IV-3), or on
platforms on opposite sides of the auditorium. They may choose and play
one of six sonic constructions, numbered in the score from I to VI. We call
them tactics and they are of stochastic structure. They were calculated on
the IBM-7090 in Paris. In addition, each conductor can make his orchestra
play simultaneous combinations oftwo or three of these fundamental tactics.
The six fundamental tactics are:
I. Winds
II. Percussion
III. String sound-box struck with the hand
IV. String pointillistic effects
V. String glissandi
VI. Sustained string harmonics.
123
The following are 13 compatible and simultaneous combinations of these
tactics:
I & II = VII
I & III = VIII
I & IV = IX
I & V = X
I & VI = XI
II & III = XII
II & IV = XIII
II & V = XIV
II & VI = XV
I & II & III = XVI
I & II & IV = XVII
I & II & V = XVIII
I & II & VI = XIX
Thus there exist in all 19 tactics which each conductor can make his orches-
tra play, 361 (19 x 19) possible pairs that may be played simultaneously.
The Game
1. Choosing tactics. How will the conductors choose which tactics to
play?
a. A first solution consists of arbitrary choice. For example, conductor
X chooses tactic I. Conductor Y may then choose anyone of the 19 tactics
including I. Conductor X, acting on Y's choice, then chooses a new tactic
(see Rule 7 below). X's second choice is a function of both his taste and Y's
choice. In his turn, conductor Y, acting on X's choice and his own taste,
either chooses a new tactic or keeps on with the old one, and plays it for a
certain optional length of time. And so on. We thus obtain a continuous
succession of couplings of the 19 structures.
b. The conductors draw lots, choosing a new tactic by taking one card
from a pack of 19; or they might make a drawing from an urn containing
balls numbered from I to XIX in different proportions. These operations can
be carried out before the performance and the results of the successive draws
set down in the form of a sequential plan which each of the conductors will
have before him during the performance. .
c. The conductors get together in advance and choose a fixed succession
which they will direct.
d. Both orchestras are directed by a single conductor who establishes
the succession of tactics according to one of the above methods and sets
them down on a master plan, which he will follow during the performance.
124
Fig.IV-2
1. Game matrix (dynamostat, dual regulator)
2. Conductor A (device for comparison and
decision)
3. Conductor B (device for comparison and
decision)
4. Score A (symbolic excitation)
5. Score B (symbolic excitation)
6. Orchestra A (human or electronic trans-
forming device)
7. Orchestra B (human or electronic trans-
forming device)
8. Audience
Formalized Music
'"
.'"
'"
'"
ro
OJ
en
"'C
c
. ~
"'C
0
0
s:
III
"'C
c
. ~
"0
c
0
s:
(/)
CD
.,
.,
ro
OJ
Percussion
Marimbaphone
[
Percussion
Vibraphone
Double basses
Cellos
Violas
Violins
Violins
Violas
Cellos
Double basses
125
-
'" c
C
ro
E
. ~
<II
.,
co
"0
'" :5
.!::
"0
CD
Cl
C
~
.,
t
.!!!
~
. ~ 1;;
:<5--+
CD
.c
:J 0
D..
0
t .s::.
0
'"
'"
-0
'"
III
:J
~
'"
'"
'" Cl
<tI
1;;
0
~
:=
u.;
I-
0
Z
126
eOA
COli
Formalized Music
e. Actually all these ways constitute what one may call" degenerate"
competitive situations. The only worthwhile setup, which adds something
new in the case of more than one orchestra, is onc that introduces dual
conflict between the conductors. In this case the pairs of tactics are per-
formed simultaneously without interruption from one choice to the next
(sec Fig. IV-4), and the decisions made by the conductors are conditioned
by the winnings or losses contained in the game matrix.
ORX
GMNS 78 72 45
PICTles IX XVII! XIV XV
.36
VII
OR. Y
'iAINS .5"2 40 48
TACTICS VI! XIX XV v
Fig.IV-4
2. Limiting the game. The game may be limited in several ways: a. The
conductors agree to play to a certain number of points, and the first to reach
it is the winner. h. The conductors agree in advance to play n engagements.
The one with more points at the end of the nth engagement is the winner.
c. The conductors decide on the duration for the game, m seconds (or
minutes), for instance. The one with more points at the end of the mth
second (or minute) is the winner.
3. Awarding points.
a. One method is to have one or two referees counting the points in
two columns, one for conductor X and one for conductor Y, both in positive
numbers. The referees stop the game after the agreed limit and announce
the result to the public.
h. Another method has no referees, but uses an automatic system that
consists of an individual board for each conductor. The board has the
n x n cells of the game matrix used. Each cell has the corresponding partial
score and a push button. Suppose that the game matrix is the large one of
19 x 19 cells. If conductor X chooses tactic XV against Y's IV, he presses
the button at the intersection of row XV and column IV. Corresponding to
this intersection is the cell containing the partial score of 28 points for X and
the button that X must push. Each button is connected to a small adding
machine which totals up the results on an electric panel so that they can be
seen by the public as the game proceeds, just like the panels in the football
stadium, but on a smaller scale.
4. Assigning of rows or columns is made by the conductors tossing a coin.
5. Deciding who starts the game is determined by a second toss.
6. Reading the tactics. The orchestras perform the tactics cyclically on a
closed loop. Thus the cessation of a tactic is made instantaneously at a bar
line, at the discretion of the conductor. The subsequent eventual resumption
of this tactic can be made either by: a. reckoning from the bar line defined
above, or b. reckoning from a bar line identified by a particular letter. The
conductor will usually indicate the letter he wishes by displaying a large
card to the orchestra. If he has a pile of cards bearing the letters A through
U, he has available 22 different points of entry for each one of the tactics.
In the score the tactics have a duration of at least two minutes. When the
conductor reaches the end of a tactic he starts again at the beginning, hence
the" da capo" written on the score.
7. Duration oj the engagements. The duration of each engagement is
optional. It is a good idea, however, to fix a lower limit of about 10 seconds;
i.e., if a conductor engages in a tactic he must keep it up for at least 10
seconds. This limit may vary from concert to concert. It constitutes a wish
on the part of the composer rather than an obligation, and the conductors
have the right to decide the lower limit of duration for each engagement
before the game. There is no upper limit, for the game itself conditions
whether to maintain or to changc the tactic.
8. Result of the contest. To demonstrate the dual structure of this compo-
sition and to honor the conductor who more faithfully followed the con-
ditions imposed by the composer in the game matrix, at the end of the
combat one might a. proclaim a victor, or b. award a prize, bouquet of
flowers, cup, or medal, whatever the concert impresario might care to
donate.
9. Choice oj matrix. In Strategic there exist three matrices. The large one,
19 rows x 19 columns (Fig. IV-5), contains all the partial scores for pairs
of the fundamental tactics I to VI and their combinations. The two smaller
matrices,3 x 3, also contain these but in the following manner: Row I and
column I contain the fundamental tactics from I to VI without discrimina-
tion; row 2 and column 2 contain the two-by-two compatible combinations
of the fundamental tactics; and row 3 and column 3 contain the three-by-
three compatible combinations of these tactics. The choice between the
large 19 x 19 matrix and one of the 3 x 3 matrices depends on the ease
with which the conductors can read a matrix. The cells with positive scores
mean a gain for conductor X and automatically a symmetrical loss for
conductor Y. Conversely, the cells with negative scores mean a loss for
conductor X and automatically a symmctrical gain for conductor Y.
The two simpler, 3 x 3 matrices with different strategies are shown in
Fig. IV-5.
128
rn
;:
g
><;
0
0
:J
'0
C
0
U
-
:r 116
I

r -'0

!I
-/16
.,' JE 0
*I -110
= :iL
J!(

ru -,fl,
fn., -3<
-:r
"
-JG

-
..

It
..
<I, /6&
32

>It'
-10

IE
-
,88
3!J.
i lCQ'-36
-1 "" ..
-$2
1

r IT
10 N
9 6

9
6
-:;0 JII
-1""
-36

It
-J.? J.o
-8 -5ol
10 -/6
.fX -1.t
N U
'9'
-188
-10 -u
-1+ -If
,
II
B8 /09
9l. !'2
-1'; 8
-,2 -66
MATRIX OF THE GAME
liZ Jtr
-'18
"
-.12
-u -,I. '>2
% 0
81, 4-
"
101{ -8
-III -6 64-
II, 3. J"
- 8 1:1,
"
-3t t
"
"
-It
-'f
"
-M
-.19 -1+ -u
-1'0 -vt
"
-u -J"/1
.28 -/8
-28 to 16
-2B
I'
8
0 -J!
4 Ii -6
Conductor Y (columns)
-
= "'"
.x-
-6. _110 J;;

-fO (;( U
-f6 -.U -u
lot -M 8
".
:10 -IJ 4-
2r -8
:"
"
1t4
"

-"8
"
48 -I/l
-""
.>,z
""
-I,C -16 B -.3C
-IOf 2.
"
u; 3/1 .It
"
IV
-"
2Il
4& l,t -1/1 ;u;
0 .111 2 -f
-:J. -16 .20 -20
-'rl, -98 -32 0

1
8 -12 -4
-4 44 -66 -4
" ,
u
-8
6
9
-8
B
-:II(
0
8
.I
-1.
c.
-3:/
-If
u
-].I
_$0
-/6
.
H
<U
-30
-(;8
-rJ.
-u
-8
-1;0
-f6
-f.!
1(8
-II
68
'H
80
-.U
IV
-.16
-/6
36 -8
41; 12
IP
-u.
-.10
-flJ
-."
-38
-n
-36
-I;
-/K
B

1
6
f.l
n
'/6
-8
-20
2.9
"'"
c;
J!.W
.u
_10
-k.
y
-ott
-II'
-I.t
8
6,
3,/
.0
-f
-"
,.
-rt
-3&
-3.t
-21,

Fig. IV-5. Strategy
Two-person Game. Value of the Game = O.
...A Woodwinds
Normal percussion
H Strings striking sound-boxes
... Strings pizzicato
.:If Strings glissando
== Strings sustained
Combinations
of two and three
different tactics
Formalized Music
.,
..

-,.
-16
Jt
-16
:"
-,/0
U
,/ .
11;
-1::[
"
-54
-r.I'
11&
-{'''
JIJ
-16
16
,,'
'0
' .1
fJ8
-.tt -J6
-".
-16
-'0 -16
/to I!I
H
.,.
-"0 -4-"
_1(0
. U
-J"
Y -8
-: -.11
-u -'/8
+-" -I'
-H -30
-Jr;
I/>e -.11'
-jo 9G
-/9
-ItG
""
r ",,.
-!8 3;1
-Jo ..
20
P
11
-I"
H
to -?-t-
"
-.B ,
16 +-0
-It.
:I"
-28

20 -u
-.11 _3.1 /0
40
'6
,
1B

/I
-19 f.t
r
-3 II<
-33 60
.r:? -36 B
-NO 9
-32
I
Strategy
Conductor Y (columns) 1:':+;; for X
.I:II'<=I:!$==
.. .......... ....... .. ......... 1: .. ...
I :
>< "
I
(5 .,
g
-g
8
: ..
-1'.
III.
3
3
0 -1
2
n
-3 3 ti

0
,g
"
3 -1
0
>< '*'
III
.2- ,
UH'
-5:. ,
I:,/J<'
8111.
-5 -1 4
HO
: ..
-1'.
m_
HO,
".,
: ...
*. ,
III
-5 4 -3

: .. ,
" ,
3
1 -3
Strategy for Y 5
33
Fig.IV-6
ll.
33
Twoperson Zero sum Game .
Value of the Game = 1/11.
This game is not fair for Y .
"WooawinC5
Normal percussion
H Strings striking sound-boxes
:. Strings pizzicato
% Strings glissando
III Strings sustained
11
III. ,
Strategy for Y .!!.

Two-person Zero-sum Game.
Value of the Game = O.
This game is fair for both
conductors.
,. Combinations
of two dis-
: tinct tactics
OH
. :.
0*
III
, H Combinations
: : of three dis-
, III tinct tactics
Simplification of the 19 x 19 Matrix
To make first performances easier, the conductors might use an equiva-
lent 3 x 3 matrix derived from the 19 x 19 matrix in the following manner:
Let there be a fragment of the matrix containing row tactics
r + I, ... , r + m and column tactics.l + 1, ... , s + n with the respective
probabilities qr+h ..... , qr+m and ks + h " .. OJ ks+TI.'
k3 +1 ks+J k.+n
qr+l ar +1 ,B+l ar +1 .'+J ar +1s +n
qr+f G'+',a+l lZ,.+f,S+J Gr+f,s+n
qr+m ar+m,S+l a,+m,s+J Gr+m,s+n
This fragment can be replaced by the single score
A _ (ar+i.s+J)(qr+i)(ks+J)
r+m.s+n - .z:f qr+f 2.'J kS+J
and by the probabilities
m
Q = 2: qr+i

and
n
K = 2: ks+i'

Operating in this way with the 19 x 19 matrix we obtain the following
matrix (the tactics will be the same as in the matrices in Fig. IV-6):
7704 8296 592
25 x 25
-
49 x 25 25 x 26
25
14522 17610 3088
- -
25 x 45 49 x 45 45 x 26
45
6818 9314 2496
25 x 30
-
49 x 30 30 x 26
30
25 49 26
2465 -1354 182 25
or
- 2581 1597 -528 45
1818 -1267 640 30
25 49 26
/
Chapter V
Free Stochastic Music by Computer
After this interlude, we return to the treatment of composition by machines.
The theory put forward by Achorripsis had to wait four years before
being realized mechanically. This realization occurred thanks to M.
Fran<jois Genuys of IBM-France and to M. Jacques Barraud of the Regie
Autonome des Transports Parisiens.
THE PARADOX: MUSIC AND COMPUTERS
A STOCHASTIC WORK EXECUTED BY THE IBM-7090
The general public has a number of different reactions when faced by
the alliance of the machine with artistic creation. They fall into three
categories:
"It is impossible to obtain a work of art, since by definition it is a handi-
craft and requires moment-by-moment "creation" for each detail and for
the entire structure, while a machine is an inert thing and cannot invent."
"Yes, one may play games with a machine or use it for speculative
purposes, but the result will not be "finished": it will represent only an
experiment-interesting, perhaps, but no more."
The enthusiasts who at the outset accept without flinching the
whole frantic brouhaha of science fiction. "The moon? Well, yes, it's
within our reach. Prolonged life will also be with us tomorrow-why not a
creative machine?" These people are among the credulous, who, in their
idiosyncratic optimism, have replaced the myths of Icarus and the fairies,
which have decayed, by the scientific civilization of the twentieth century,
and science partly agrees with them. In reality, science is neither all
paradox nor all animism, for it progresses in limited stages that are not
foreseeable at too great a distance.
131
There exists in all the arts what we may call rationalism in the etymo-
logical sense: the search for proportion. The artist has always called upon it:
out of necessity. The rules of construction have varied widely over the cen-
turies, but there have always been rules in every epoch because of the
necessity of making oneself understood. Those who believe the first statement
above are the first to refuse to apply the qualification artistic to a product
which they do not understand at all.
Thus the musical scale is a convention which circumscribes the area of
potentiality and permits construction within those limits in its own particu-
lar symmetry. The rules of Christian hymnography, of harmony, and of
counterpoint in the various ages have allowed artists to construct and t:o
make themselves understood by those who adopted the same constraints-
through traditions, through collective taste or imitation, or through sym-
pathetic resonance. The rules of serialism, for instance, those that banned
the traditional octave doublings of tonality, imposed constraints which were
partly new but none the less real.
Now everything that is rule or repeated constraint is part of the mental
machine. A little "imaginary machine," Philippot would have said-a
choice, a set of decisions. A musical work can be analyzed as a multitude
of mental machines. A melodic theme in a symphony is a mold, a mental
machine, in the same way as its structure is. These mental machines are
something very restrictive and deterministic, and sometimes very vague and
indecisive. In the last few years we have seen that this idea of mechanism is
really a very general one. It flows through every area of human knowledge
and action, from strict logic to artistic manifestations.
Just as the wheel was once one of the greatest products of human
intelligence, a mechanism which allowed one to travel farther and faster
with more luggage, so is the computer, which today allows the transforma-
tion of man's ideas. Computers resolve logical problems by heuristic methods.
But computers are not really responsible for the introduction of mathematics
into music; rather it is mathematics that makes use of the computer in
composition. Yet if people's minds are in general ready to recognize the
usefulness of geometry in the plastic arts (architecture, painting, etc.), they
have only one more stream to cross to be able to conceive of using more
abstract, non-visual mathematics and machines as aids to musical composi-
tion, which is more abstract than the plastic arts.
To summarize:
1. The creative thought of man gives birth to mental mechanisms,
Free Stochastic Music by Computer 133
which, in the last analysis, are merely sets of constraints and choices. This
process takes place in all realms of thought, including the arts.
2. Some of these mechanisms can be expressed in mathematical terms.
3. Some of them are physically realizable: the wheel, motors, bombs,
digital computers, analogue computers, etc.
4. Certain mental mechanisms may correspond to certain mechanisms
ofnature.
5. Certain mechanizable aspects of artistic creation may be simulated
by certain physical mechanisms or machines which exist or may be created.
6. It happens that computers can be useful in certain ways.
Here then is the theoretical point of departure for a utilization of
electronic computers in musical composition.
We may further establish that the role of the living composer seems to
have evolved, on the one hand, to one of inventing schemes (previously
forms) and exploring the limits of these schemes, and on the other, to effect-
ing the scientific synthesis of the new methods of construction and of sound
emission. In a short while these methods must comprise all the ancient and
modern means of musical instrument making, whether acoustic or electronic,
with the help, for example, of digital-to-analogue converters; these have
already been used in communication studies by N. Guttman, J. R. Pierce,
and M. V. Mathews of Bell Telephone Laboratories in New Jersey. Now
these explorations necessitate impressive mathematical, logical, physical,
and psychological impedimenta, especially computers that accelerate the
mental processes necessary for clearing the way for new fields by providing
immediate experimental verifications at all stages of musical construction.
Music, by its very abstract nature, is the first of the arts to have at-
tempted the conciliation of artistic creation with scientific thought. Its
industrialization is inevitable and irreversible. Have we not already seen
attempts to industrialize serial and popular music by the Parisian team of
P. Barbaud, P. Blanchard, and Jeanine Charbonnier, as well as by the
musicological research of Hiller and Isaacson at the University of Illinois?
In the preceding chapters we demonstrated some new areas of musical
creation: Poisson, Markov processes, musical games, the thesis of the mini-
mum of constraints, etc. They are all based on mathematics and especially
on the theory of probability. They therefore lend themselves to being
treated and explored by computers. The simplest and most meaningful
scheme is one of minimum constraints in composition, as exemplified by
Achorripsis.
Thanks to my friend Georges Boudouris of the C.N.R.S. I made the
acquaintance of Jacques Barraud, Engineer of the Ecole des Mines, then
director of the Ensemble Electroniques de Gestion de la Societe des Petroles
Shell-Berre, and Gcnuys, agrcgc in mathematics, and head of the
Etudes Scientifiques Nouvelles at IBM-France. All three are scientists, yet
they consented to attempt an experiment which seemed at first far-fetched-
that of a marriage of music with one of the most powerful machines in the
world.
In most human relations it is rarely pure logical persuasion which is
important; usually the paramount consideration is material interest. Now
in this case it was not logic, much less self-interest, that arranged the be-
trothal, but purely experiment for experiment's sake, or game for game's
sake, that induced collaboration. Stochastically speaking, my venture should
have encountered failure. Yet the doors were opened, and at the end of a
year and a half of contacts and hard work "the most unusual event wit-
nessed by the firm or by this musical season [in Paris]" took place on 24
May 1962 at the headquarters of IBM-France. It was a live concert pre-
senting a work of stochastic instrumental music entitled STj 1 0-1, 080262,
which had been calculated on the IBM-7090. It was brilliantly performed
by the conductor C. Simonovic and his Ensemble de Musique Contem-
poraine de Paris. By its passage through the machine, this work made
tangible a stochastic method of composition, that of the minimum of
constraints and rules.
Position of the Problem
The first working phase was thc drawing up of the flow chart, i.e.,
writing down clearly and in order the stages of the operations of the scheme
of Achorripsis, 1 and adapting it to the machine structure. In the first chapter
we set out the entire synthetic method of this minimal structure. Since the
machine is an iterative apparatus and performs these iterations with extra-
ordinary speed, the thesis had to be broken down into a sequential series of
operations reiterated in loops. An excerpt from the first flow chart is shown
in Fig. V-I.
The statement of the thesis of Achorripsis receives its first machine_
oriented interpretation in the following manner:
1. The work consists of a succession of sequences or movements each a
j
seconds
long. Their durations are totally independent (asymmetric) but have a fixed
mean duration, which is introduced in the form of a parameter. These
durations and their stochastic succession are given by the formula
(See Appendix I.)
l'ree Stochastic Music by Computer
,l; SS8.KoLi. (vic,c)
YlL
EF
j
=

","=:
VIGil:
To YUl
Tc::>-ble. :
e(j\) (:\4 .. aiD"
21;0 ya/IAQS

;Q5'
- 8t-
Fig. V-1. Excerpt from the First Flow Chart of Achorripsis
135
I ...... II
YJIB
2. Dqinition of the mean density of the sounds during aj. During a sequence
sounds are emitted from several sonic sources. If the total number of these
sounds or points during a sequence is N
al
, the mean density of this point-
cluster is NaJa
t
sounds/sec. In general, for a given instrumental ensemble
this density has limits that depend on the number of instrumentalists, the
nature of their instruments, and the technical difficulties of performance.
For a large orchestra the upper limit is of the order of 150 sounds/sec. The
lower limit (V3) is arbitrary and positive. We choose (V3) = 0.11 sounds/sec.
Previous experiments led us to adopt a logarithmic progression for the
density sensation with a number between 2 and 3 as its base. We adopted
e = 2.71827. Thus the densities are included between (V3)eO and (V3)e
R
sounds/sec., which we can draw on alinegraduated logarithmically (basee).2
As our purpose is total independence, we attribute to each of the sequences
aj calculated in 1. a density represented by a point drawn at random from
the portion of the line mentioned above. However a certain concern for
continuity leads us to temper the independence of the densities among
sequences a
f
; to this end we introduce a certain" memory" from sequence
to sequence in the following manner:
Let a;_l be a sequence of duration a
t
-
1
, (DA)t_1 its density, and aj the
next sequence with duration af and density (DA)t. Density (DA)j will be
given by the formula:
in which x is a segment of line drawn at random from a line segment s of
length equal to (R - 0). The probability of x is given by
p = ~ (1 - ~ ) dx
x s s
(see Appendix J)
and finally,
3. Composition Q of the orchestra during sequence af' First the instruments
are divided into r classes of timbres, e.g., flutes and clarinets, oboes and
bassoons, brasses, bowed strings, pizzicati, collegno strokes, glissandi, wood,
skin, and metal percussion instruments, etc. (See the table for Atrees.) The
composition of the orchestra is stochastically conceived, i.e., the distribution
of the classes is not deterministic. Thus during a sequence of duration at it
may happen that we have 80/
0
pizzicati, 10% percussion, 7% keyboard,
and 3% flute class. Under actual conditions the determining factor which
would condition the composition of the orchestra is density. We therefore
Composition of the Orchestra for A t r t ~ e s (ST/10-3,060962)
Timbre classes and instruments as on present input data
Class Timbre
2
3
4
.')
6
7
8
9
10
II
Percussion
Horn
Flute
Clarinet
Glissando
Tremolo
or
flutter-
tongue
Plucked
strings
Struck
strings*
Vibraphone
Trumpet
Trombone
12 Bowed
strings
* collegno
Instrument
Temple-blocks
Tom-toms
Maracas
Susp. cymbal
Gong
French horn
Flute
Clarinet BD
Bass claro BD
Violin
Cello
Trombone
Flute
Clarinet Bp
Bass claro BD
French horn
Trumpet
Trombone a
Trombone b
(pedal notes)
Violin
Cello
Violin
Cello
Violin
Cello
Vibraphone
Trumpet
Trombone a
Trombone b
(pedal notes)
Violin
Cello
Instrument No.
1-5
6-9
10
II
12
I
2
1
2
3
1
2
3
4
.')
6
7
8
9
I
2
2
I
2
2
137
138
Formalized Music
connect the orchestral composItIOn with density by means of a special
diagram. An example from STfJO-l, 080262 is shown in Fig. V-2.
Fig. V-2 is expressed by the formula
QT = (n - x) (enT - en + 1.T) + en,T
inwhichr = thenumberofthec1ass,x = loge[(DA),J(V3)],n = 0,1,2, . ,R,
such that n :5 x :5 n + I, and Cn,T and e
n
+ l,T are the probabilities of class T
as a function of n. It goes without saying that the composition of this table
is a precise task of great complexity and delicacy, Once these preliminaries
have been completed, we can define, one after the other, the N
a
, sounds of
sequence at.
4. Definition of the moment of occurrence of the sound N within the sequence a;_
The mean density of the points or sounds to be distributed within at is
k = NaJat The formula which gives the intervals separating the sound
attacks is
(See Appendix I.)
5. Attribution to the above sound of an instrument belonging to orchestra Q,
which has already been calculated. First class r is drawn at random with proba-
bility qT from the orchestra ensemble calculated in 3. (Consider an urn with
balls of r colors in various proportions.) Then from within class r the number
of the instrument is drawn according to the probability Pn given by an
arbitrary table (urn with balls of n colors). Here also the distribution of
instruments within a class is delicate and complex.
6. Attribution of a pitch as a function qf the instrument. Taking as the zero
point the lowest E[, of the piano, we establish a chromatic scale in semitones
of about 85 degrees. The range s of each instrument is thus expressed by a
natural number (distance). But the pitch hu of a sound is expressed by a
decimal number of which the whole number part is related to a note of the
chromatic scale within the instrument's range.
Just as for the density in 2., we accept a certain memory of or dependence
on the preceding pitch played by the same instrument, so that we have
where z is given by the probability formula
p = 3 (I - !:) dz.
z s s
(See Appendix 1.)
Pz is the probability of the interval z taken at random from the range s, and
CI\
~ O : 2 .
,." w .po. 01 0>
-..J
:J !. '3.
'0 U ::T
CD N' Ib 0 0
~
in'
ri :;
3
en 3
N
:;' e;'
<::
'"
'"
0
til ~
~
:J
0'
~
til
V>
C.
0
0'
OT
0
Q)
:::I
-a
Fig. V-2. ST/l0-1, 080262, Composition of the Orchestra
Density = (DA)I = O.11e
u
, U = 10ge (DAjO,11)
139
012-
co
0
.
Ib
'"
:::I
0
140
Formalized Music
s is expressed as the difference betwcen the highest and lowest pitches that
can be played on the instrument.
7. Attribution qf a glissando speed if class r is characterized as ([ glissando. The
homogeneity hypotheses in Chap. I led us to the formula
f(v) = _2_ e-v'/a2
ayrr '
and by the transformation via = u to its homologue:
T(u) = Jrr f e-
u2
du,
for ,:,hic.h there are tables.f(v) is the probability of oct:urrence ofthe speed v
(whl.ch IS expressed in semitones/sec.); it has a parameter a, which is pro-
portlOnal to the standard deviations (a = sY2).
a defined as a function of the logarithm of the density of sequence a
by: an Inversely proportional function f
a = yrr( 30 - L[(DA)d(V3)]).
or a directly proportional function
a = Y7T(lO + L[(DA)d(V3)]),
or a function independent of density
a = 17.7 + 35k,
where k is a random number between 0 and I.
The constants of the preceding formulae derive from the limits of the
speeds that string glissandi may take.
Thus for (DA)j = 145 sounds/sec.
a = 53.2 semi tones/sec.
2s = 75 semitones/sec.,
and for (DAyj = 0.13 sounds/sec.
a = 17.7 semi tones/sec.
2s = 25 semitones/sec.
8. qf a duration x to the sounds emitted. To simplify we establish
a mean duratIon for each instrument, which is independent of tessitura and
Free Stochastic Music by Computer 141
nuance. Consequently we reserve the right to modify it when transcribing
into traditional notation. The following is the list of constraints that we take
into account for the establishment of duration x:
G, the maximum length of respiration
or dcsired duration
(DA)i, the density of the sequence
qr, the probability of class r
pn, the probability of the instrument n
Then if we define z as a parameter of a sound's duration, z could be
inversely proportional to the probability of the occurrence of the
instrument, so that
Z will be at its maximum when (DA)fPnqT is at its minimum, and in this case
we could choose Zu,ax = G.
Instead of letting Zmax = G, we shall establish a logarithmic law so as
to freeze the growth of z. This law applies for any given value of z.
z' = G In z/ln Zmax
Since we admit a total independence, the distribution of the durations
x will be Gaussian:
1
f(x) = ---e-(x-m)2/2s
2
,
sy27r
where m is the arithmetic mean of the durations, s the standard deviation,
and
m - 4.25.1 = 0
m + 4.255 = Z'
the linear system which furnishes us with the constants m and s. By assuming
u = (x - m)/5y2 we find the function T(u), for which we consult the
tables.
Finally, the duration x of the sound will be given by the relation
x = usy2 + m.
We do not take into account incompatibilities between instruments, for
this would needlessly burden the machine's program and calculation.
9. Attribution of dynamic forms to the sounds emitted. We define four zones
of mean intensities: ppp, p,1, if. Taken three at a time they yield 4
3
= 64
permutations, of which 44 are different (an urn with 44 colors); for example,
ppp-<j>-p.
10. The same operations are begun again for each sound of the cluster N
a
,.
11. Recalculations of the same sort are made for the other sequences.
An extract from the sequential statement was reproduced in Fig. V-I.
Now we must proceed to the transcription into Fortran IV, a language
"understood" by the machine (see Fig. V-3).
It is not our purpose to describe the transformation of the flow chart
into Fortran. However, it would be interesting to show an example of the
adaptation of a mathematical expression to machine methods.
Let us consider the elementary law of probability (density function)
f(x) dx = cr
ex
dx. [20]
How shall we proceed in order for the computer to give us lengths x with
the probability f(x) dx? The machine can only draw random numbers Yo
with equiprobability between 0 and 1. We shall "modulate" this proba-
bility: Assume some length Xo; then we have
rXo
prob. (0 x xo) = Jo f(x) dx = 1 - e-CXo = F(x
o
),
where F(xo) is the distribution function of x. But
F(xo) = prob. (0 Y Yo) = Yo
then
and
Xo =
In (1 - Yo)
c
for all Xo O.
Once the program is transcribed into language that the machine's
internal organization can assimilate, a process that can take several months,
we can proceed to punching the cards and setting up certain tests. Short
sections are run on the machine to detect errors of logic and orthography
and to determine the values of the entry parameters, which are introduced
in the form of variables. This is a very important phase, for it permits us to
explore all parts of the program and determine the modalities of its opera-
tion. The final phase is the decoding of the results into traditional notation,
unless an automatic transcriber is available.
143
Table of the 44 IntensitY Forms Derived from 4 Mean Intensity
Values, ppp, p, f. ff
iN IN
tl! I
1/1 -===.J ==-/'11
I II;

II! -==-1 ==-))1
I = III
III
1/1
f/=====--
Iii -==-=! =----=-1
I =-=-=-III-=J
I ==-111-==-1
ill-===- #-==- J
/L - /w-==}
f= 111-=#
f-===:- III
I J
P-= /11-==.1
I I
/--= I===-P
f ======--- f
f-=====--.(-
f --=== .7 -==-/'
;r= f
il! #-==-1

; --====.

I -===== #'
j> ----====::: ",:::==-I
1/ =:==-; --====-./
/--
---/
.I
;1 :=========-- /
# ;f
j' -====-)# -==::
j!:::===--j --=:::::..1
\
I
Conclusions
A large number of compositions of the same kind as ST/ 10-1, 080262
is possible for a large number of orchestral combinations. Other works have
already been written: ST/48-1, 240162, for large orchestra, commissioned
by RTF (France III) ; Atdes for ten soloists; and M orisma-Amorisima, for four
soloists.
Although this program gives a satisfactory solution to the minimal
structure, it is, however, necessary to jump to the stage of pure composition
by coupling a digital-to-analogue converter to the computer. The numerical
calculations would then be changed into sound, whose internal organization
had been conceived beforehand. At this point one IO:ould bring to fruition
and generalize the concepts described in the preceding chapters.
The following are several of the advantages of using electronic compu-
ters in musical composition:
I. The long laborious calculation made by hand is reduced to nothing.
The speed of a machine such as the IBM-7090 is tremendous-of the order
of 500,000 elementary operations/sec.
2. Freed from tedious calculations the composer is able to devote him-
self to the general problems that the new musical form poses and to explore
the nooks and crannies of this form while modifying the values of the input
data. For example, he may test all instrumental combinations from soloists
to chamber orchestras, to large orchestras. With the aid of electronic com-
puters the composer becomes a sort of pilot: he presses the buttons, intro-
duces coordinates, and supervises the controls of a cosmic vessel sailing in
the space of sound, across sonic constellations and galaxies that he could
formerly glimpse only as a distant dream. Now he can explore them at his
ease, seated in an armchair.
3. The program, i.e., the list of sequential operations that constitute
the new musical form, is an objective manifestation of this form. The
program may consequently be dispatched to any point on the earth that
possesses computers of the appropriate type, and may be exploited by any
composer pilot.
4. Because of certain uncertainties introduced in the program, the
composer-pilot can instill his own personality in the sonic result he obtains.
145
Fig. V-3. Stochastic Music Rewritten in Fortran IV
c
c
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
C
C
C
C
C
C
C
C
C
C
C
C
C
C
c
C
C
C
c
c
C
C
C
C
C
C
C
C
c
PROGRAM FREE STOCHASTIC MUSIC (FORTRAN )V)
GLOSSARY OF THE PRINC1PAL ABBREVIATIONS
XEN
XEN
XEN
XEN
A - DURATION OF EACH IN SECONDS
AIO,A20.A17,A3S,A30 - NUMBERS FOR GLISSANDO CALCULATION XEN
ALEA - PARAMETER USED TO ALTER THE RESULT OF A SECOND WITH THEXEN
SAME INPUT DATA XEN
ALFA(JJ - THREE EXPRESSIONS ENTERING INTO THE SPEED VALUES XEN
OF THE SLIDING TONES I GLISSANDI I XEN
ALIM - MAXIMUM LIMIT OF SEQUENCE DURATION A XEN
TABLE OF AN EXPRESSION ENTERING INTO THE XEN
CALCULATION OF THE NOTE LENGTH IN PART 8 XEN
8F - DYNAMIC FORM NUMBER. THE LIST IS ESTABLISHED INDEPENDENTLV XEN
OF THIS PROGRAM AND IS SUBJECT TO MODIFICATION XEN
DELTA _ THE RECIPROCAL OF THE MEAN DENSITY OF SOUND EVENTS DURING XEN
A SEQUENCE OF DURATION A XEN
(E(I.J),I=1.KTR.4=1.KTE) - PROBABILITIES OF THE KTR TIMBRE CLASSESXEN
INTRODUCED AS INPUT DATA, DEPENDING ON THE CLASS NUMBER I=KR ANO XEN
ON THE POWER ..J=1I OATAINcD FROM V34-EXPF(U)=DA XEN
EPsr - EPSILON FOP ACCURACY IN COLCULATING PN AND E(ltJ).WHtCH XEN
IT IS ADVISABLE TO XEN
- TABLE OF THE GIVEN LENGTH OF BREATH XEN
FOR EACH INSTRUMENT. DEPENDING ON CLASS I AND INSTRUMENT XEN
GTNA _ NUMBER OF NOTES IN THE OF DURATION A XEN
GTNS - NUMBER OF NOTES IN KW LOOPS XEN
CHAMIN(1 ..J)'HAMAX(I XEN
TABLE OF INSTRUMENT COMPASS LIMITS, DEPENDING ON CLASS 1 XEN
AND tNSTRUMENT J. TEST INSTRUCTION 480 IN PART 6 DETERMINES XEN
THE HA OR THE HB TABLE IS FOLLOWEO. THE NUMBER 7 IS XEN
ARBITRARY. XEN
JW - ORDINAL NUMBER OF THE SEQUENCE COMPUTED. XEN
KNL - NUMBER OF LINES PER PAGE OF THE PRINTED RESULT,KNL=50 XEN
- NuMBER IN THE CLASS KR=t USED FOR PERCUSSION OR INSTRUMENTS XEN
WITHOUT A DEFINITE PITCH, XEN
- POWER OF THE EXPONENTIAL COEFFICIENT E SUCH THAT XEN
DAfMAX1=V34 (E*4(KTE-I) XEN
KTR - NUMBER OF TIMBRE CLASSES XEN
K\.\' - MAXIMUM NUMBER OF ..J'W XEN
KTEST1.TAV1.ETC - USEFUL IN CALCULATING HOw LONG THE XEN
VARIOUS PARTS OF THE PROGRAM WILL RUN. XEN
KTt - ZERO IF THE PROGRAM IS 8EING RUN. NONZERO DURING DEBUGGING XEN
KT? - NUMBER OF LOOPS. EOUAL TO 15 BY ARBITRARY DEFINITION. XEN
AUXILIARY FUNCTION TO INTERPOLATE VALUES IN XEN
THF. TETA(256) TABLE (SEE PART 7) XEN
NA _ NUMBER OF SOUNDS CALCULATED FOR THE SEQuENCE A(NA=DA*A) XEN
(NTCI).I=l.KTR) NUMBER OF INSTRUMENTS ALLOCATED TO EACH OF THE XEN
KTR TIMBRE CLASSES. KEN
TABLE OF PROBAAILITYXEN
OF EACH INSTRUMENT OF THE CLASS I. XEN
(Otll,l=l,KTRI PROBABILITIES OF THE KTR TIMBRE CLASSES. CONSIOEREDXEN
AS LINEAR FUNCTIONS OF THE DENSITy DA. XEN
(S(I).I=l,KTR) SUM OF THE SUCCESSIVE all) PROBABILITIES. USED TO XEN
CHOOSE THE CLASS KR BY COMPARING IT TO A RANDOM NUMBER Xl (SEE XEN
PART 3. LOOP 380 AND PART 5, LOOP 430). XEN
SINA - SUM OF THE COMPUTED NOTES IN THE CLOUDS NA, ALWAYS LESS XEN
THAN GTNS ( SEE TEST IN PART 10 ). XEN
SQPt - saUARE ROOT OF PI ( 3.14159 ) XEN
TA - SOUND ATTACK TIMF. AACISSA. XEN
TETA(256) - TABLE OF THE 256 VALUES OF THE INTEGRAL OF THE NORMAL
DISTRIBUTION WHICH IS USEFUL IN CALCULATING GLISSANDO SPEED XEN
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
46
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64

c
C
c
C
C
C
C
C
C
C
C
C
C
C
c
c
C
AND SOUND EVENT DURATION. XEN
VIGL - GLISSANDO SPEED cvlTESSE GLISSANDO). WHICH CAN VARY AS. BE XEN
INDEPENDENT OF. OR VARY INVERSELY AS THE DENSITY OF THE SEQUENCE, XEN
THE ACTUAL MODE OF VARIATION EMPLOYED REMAINING THE SAME FOR THE XEN
ENT t RE SEQUENCE {SEF. PART 7]. XEN
VITLIM - MAXIMUM LIMITING GLISSANOO SPEED (IN SEMITONES/SEC), XEN
SURJECT TO MODIFICATION. XEN
V3 - MINIMUM CLOUD DENSITY DA XEN
(ZlIII.Z2(1 TABLE COMPLEMENTARY TO THE TETA TABLE. XEN
READ CONSTANTS AND TABLES
XEN
XEN
XEN
XEN
DIMF'NSI0N OC t 2) .S( 12). E( 12-}2) ,PNC 12.")0) .SPN( 12_50 I .NT! 12)' )(EN
*HAMI N ( 12.50 I. HAMAX ( 12.50)., HeMIN( 12' 50) ,H8MAX ( 12- 50) .GN ( 12.50). H{ 1. Z><EN
*.50).TETACZ561.VIGLC3"MODI(7).ZlC8).Z2CS).ALFA(31,AMAX(121 XEN
XEN
XEN
XEN
XEN
DO to tX=t,7
IX88-IX
MOtll (t><8)=1
10 1:.1+1
READ 20,ITETACI),J:.t.256)
20 FORMATCIZF6,6)
READ 30, (ZI (I) ,Z2( tl, 1=1 .8)
XE"N
XEN
XEN
XEN
XEN
XEN
XEN
XEN
30 XEN
PRINT 40.TETA,ZI.Z2 XEN
40 FORMATC*I THE TETA TABLE = *./.21[12FIO.6./1.4FI0.6./////. XEN
** THE Zt TABLE = THE 22 TABLE = *./.8F14,e,/XEN
*.lHl) XEN
READ 50.DELTA.V3tAIO,A20,AI7,A30,A35,BF.SQPI.EPSI,VITLIM.ALEA. AXEN
XEN *LIM
50 FORMAT(F3.0.F3.3,5F3.1 .F2.0.FB.7,F8.e.F4.2,F8.e.F?,Zl XEN
READ 60,KTI.KT2.KW,KNL.KTR,KTE,KRleGTNA.GTNS.(NTC I).I=I.KTR) XEN
60 FORMAT(5J3.212.2F6.0.t212) XEN
PRINT 70.DELTA,V3.AIO.A20.AI7.A30.A35,BF.SQPl.EPSI.VITLJM.ALEAe AXEN
-ItL IM,KT 1 .KT2.KW,KNL .KTE .KR 1. GTNA .GTNS. ( ( I ,NTC I ) ) 1=1 .KTR) XEN
70 FORMAT(*IOELTA :: *.F4.0./.* V3 = .,F6,3./,* AIO = *,F4.1./. XEN
** A20 = *.F4.t./.* At? = A30 = A35 = *.F4.I.XEN
*/.* SF = *.F3.0./.* SOPt =*.FI1.a./.* EPSI =*.F12.8./.* VITLIM = *XEN
*.F5.2./.* ALEA =*.F12.8./.* AL[M = *,F6.2./.* KTt = *.13./, XEN
** KT2 = 13./,* KW = *.13./,* KNL = *.13./.* KTR = *,13,/4 XEN
** = *.12./.* KRI = *.12./.* GTNA = *.F7.Q./.* GTNS = *.F7.0. XEN
*/.121. IN CLASS *.12,., THERE ARE *.12 lNSTRUMENTS.*./)} XEN
READ BO.KTEST3.KTESTl.KTEST2
80 FORMAT(513)
PRINT 90.KTEST3.KTESTI.KTEST2
90 FORMAT(* KTEST3 = *.13./.* KTESTl
IFCKTEST3.NE.O) 830
R=KTE-t
AIO",AtO*SQPI
A20=A20*saPI/R
A30=A30*SOPI
*"3)
IF ALEA IS NON-ZERO. THE RANOOM NUMBER IS GENERATED FROM THE TIME
WHEN THE FOLLOWING JNSTQUCTtON IS EXECUTED. IF ALEA IS NON-ZERO
EACH RUN OF THIS FROGQAM DIFFERENT OUTPUT DATA.
IF(ALEA.NE.O.O) CALL RANFSET(TIMEF(I))
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XE"N
XEN
XEN
XEN
66
67
68
69
70
71
72
73
74
75
76
78
77
79
80
61
62
83
64
65
66
67
66
B9
90
91
92
93
9"
95
96
97
98
99
100
114
115
126
127
128
129
130
131
132
133
134
135
136
141
142
143
14"
14S
146
147
146
149
150
151
152
153
147
c
c
c
c
c
PRINT 830
DO 130 l=t.KTR
y=o.o
KTS=NT( 11
READ
XEN
XEN
XEN

XEN
XEN
*PNC I.J)
100 FORMATC5(SF2.0.F"3.:Jll XEN
110. I .IJ.HAMIN( t .J) ,HAMAX( I ..J) .HBMIN( 1 .Jl,HBMA)(' ,..J) .GN( t .J.XEN
.,PNCI ..J) .,)=l.KTS)
110 FORMAT(////.* IN CLASS NUMBER *.12.(/.* FOR INSTRUMENT NO. *.12.
** HAMIN = *.F"3.0.*.HAMAX = *.F"3.0,4,HBMIN = *.F3.0, HBMAX = *.
* F3.0.*.GN = *.F3.0.*. AND PN : *.F6.31
00 120 ")=1.I(T5
120 SPNCt.J)=v
130 IF CABSF(V-l.O).CE.EPSI) CALL EXIT
DO t 50 t = 1 KTR
READ 140.(E(I.J).J=1.KTE)
140 FORMATt]2F2.2)
ISO 160.1.(J.( 1 .,).J=I.KTE)
160 FORMATC//////.* CLASS NUMBER *.12./.[* IN DENSITY LEVEL *.12.
*4 HAS A PROBABILITY OF *.F6.2,/)
DO 180 J=I,KTE
y=o.O
DO I 70 I = I KTR
170 Y=Y+E(I,J)
180 IF fA8SFfY-l.0}.GE.EPSIJ CALL EXIT
DO ?OO t=I.KTR
AM AX ( T) = 1 .. O/E ( 1 I )
DO ?OO J=2.KTE
AJ=J-l
AX=1.0/(E(t .,)1*EXPFCAJ
IF (KTI.NE.O) PRINT 190.AX
190 FORMATIIH .9EI2.8)
200 IF fAX.GT.AMAXII)l AMAX(11=AX
IF (KT1.NE.O) pnlNT 210,AMAX
210 FORMATe IH ,912.8)
JlIII
SINA=O.O
IF CKTESTI.NE.O) TAVl=TIMEF(I)
220 NLINE=SO
PARTS 1 AND 2, DEFINE SEQUENCE A SECONDS AND CLOUD NA DURING A
KNA=O
KI=O
230 Xl"'RANFC-l)
A=-DELTA * LOGFlxtl
IFfA.LE.ALIMI GO TO 250
IF (Kl.GE.KT2. GO TO 240
1<1=Kl +1
GO TO 230
240
XI=O.O
250 K2=O
260 (-1 )
IF GO TO 280
270 UX=R*X2
GO TO :UO
XEN
XEN
XEN
XE"N
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
"EN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XFN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
154
155
156
1!l7
15B
159
160
161
162
163
164
165
166
167
166
169
170
171
172
173
174
175
176
177
178
179
IBO
181
182
183
184
185
IB6
187
IBB
)89
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
20A
209
210
211
212
213
214
148
280 IF (RANF{-l).GE,O.Sl GO TO 290
R * (1.O-SQRTFeX2
GO TO 300
290 UX=UPR - R. ( 1.O-SORTF(X2
300 IF ICUX.GE.O.O).ANO,fUX.LE.R GO TO 310
IF tK2.GE.KT2) GO TO 270
K2=K2+1
GO TO 260
310 U=UX
DA=V3 * EXPF(U)
NA=XINTFfA OA + + 1
IF (GTNA.GT.FLOATFfNA) GO TO 330
IF CKNA.GE.KT21 GO TO 320
KNA=KNA+l
GO TO ?30
320 A=DEL TA
GO TO ?50
330 UPR=U
IF (KT1.EO.O) GO TO 360
PRINT
340 FbRMATCIHlt4IS.3X.4EtS.St3X.18)
NA;:;KTI
IF (KTEST3.NE.O) PRINT 350.JW,NA.A
350 FORMATCIHO.219,F10.2)
Formalized Music
c
C
C
PART 3. DEFINE CONSTITUTION OF ORCHESTRA DURING SFOUENCE A
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XE"N
XEN
XEN
XEN
XEN
XFN
XFN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
215
216
217
218
219
220
221
222
223
224
225
226
227
22B
229
230
231
232
233
234
235
236
237
238
23<>
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
26B
269
270
271
272
273
274
275
c
C
C
360 SINA=SINA FLOATF(NA)
XLOGOA=U
XALOG=A20 *xLOGDA
M;:XINTF{XLOGDA)
IF rCM+?).GT.KTE) M=KTE-2
SR=O.O
Ml=M+l
M2=M+2
Do 380 I;:; 1 KTR
ALFX=E( t .MI)
BETA=E( I.M2)
XM=M
4 CAF.TA-ALFXl + ALFX
IF (KT1.NE.O) PRINT 370.XM,ALFx.BETA
370
Q(I)=Qr<

3BO S(I)=Sr<
IF CKTt,NE.O, PRINT 390.eQel),I=I.KTR).eSeJ).I:cl.KTR)
390 FORMAT(lH .]2F9.4)
PART 4.DEFINE INSTANT TA OF EACH POINT IN A
IF eKTESTZ.NE,O) TAV2=TIMEF(1)
N=:!
T=O,O
TA=O.O
GO TO 410
400 N=N+l
X=RANF(-I)
T=-LOGFC Xl /OA
410 IF (KTI.NE.O. PRINT 420.N.X.T.TA
420 FORMAT(//.IB.3E20,S)
c
C
C
c
C
C
c
C
C
PART 5.DEFINE CLASS AND INSTRUMENT NUMBER TO EACH POINT OF A
XI =RANF(-II
004301=I.KTR
430 IF eXI.LE.sel)) GO TO 440
I=KTR
440 KTScNTCI)
KRcl
X2s:RANF(-1)
00 450 J=t .!(TS
SPIENIliiSPNCKR,Jl
INSTRM=J
450 IF (X2.LE.SPIEN) GO TO 460
tNSTRMcKTS
460
IF (KTt.NE.O) PRINT 470,XI,SCKR),KR,X2,SP1EN.JNSTRM
470 FORMATe IH .2E20.S.16.2E20.S,16 )
PART 6.DEFINE PITCH HN FOR EACH POINT OF SEQUENCE A
IF (KR.GT,I) GO TO 480
IF (INSTRM.GE.KR), GO TO 490
GO TO 560
480 IF (KR,LT.7) GO TO 490
HSUP:HBMAXCKR,IN5TRM)

GO TO 500
490 HSUPZHAMAXCKR.INSTRM)
HtNFaHAMINCKR'INSTRM)
500
HPR=HCKR,INSTRM)
K=O
IF CHPR.LE.O.Ol GO TO 520
510 X=RANFC-l)
IF eN,GT.l) GO TO 530
520 HX=HINF+HM*X RANFC-IJ
GO TO 560
530 1F (RANFr-ll,GE.O.S) GO TO 540
HX=HPR+HM * C t.O-SORTF(X)
GO TO 5S0
540 HX=HPR-HM * Ct,O-SORTFCX
550 IF(CHX.GE.HINF).ANO.(HX,LE.HSUP GO TO 560
IF CK.GE.KT21 GO TO 520
I<=K+l
GO TO 510
560 HCKR.INSTRM)=HX
IF eKT1.NE.O, PRtNT 570.K,X.HX
570 FOQMAT(lH ,16,2E20.A)
PART 7,OEFINE SPEED VIGL TO EACH POINT OF A
IF (KR.EO.Sl GO TO 580
VIGL( I )rO.O
VIGL(2)=0.0
VIGL(3)=O.0
Xl =0.0
X2=0.O
XLAMBDA=O.O
GO TO 740
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
XEN
XEN
"EN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XeN
XEN
XEN
XEN
XEN
XEN
XEN
XFN
"EN
149
276
277
27B
279
2BD
281
2B2
2B3
284
285
286
2B7
2BB
2B9
290
291
292
293
294
295
296
297
29B
299
300
301
302
303
304
305
306
307
30B
309
310
311
312
313
314
315
316
317
31B
319
320
321
322
323
324
325
326
327
32B
329
330
331
332
333
334
335
336
ISO
c
c
c
580 -:X= 1
590
IF (Xt-O.9QQ7)
6DO 1=12A
DO 630 IX=1.7
610.660.620
610 t =l+MOOI (IX,
GO TO 630
620 I'I-MODI( IX)
630 CONTINUE
tFfTETA(1)-Xl} 670.640.660
640 XLAMBOA=FLOATFCJ-ll/10n.n
GO TO (720.760). KX
650
GO TO C 720. 760 1 .KX
660 I; I-I
670 TXI.TETA(II
XLAMBOA=(FLOATF(I-ll+(XI-TXI1/(TETA{I+1)-TXI/lOO.O
GO TO 720.760 ). KX
6BO
690
700
710
720
730
740
750
DO 690 1=2.7
TXI=Z2(t)
IF(XI-TXl) 700.710.690
CONTINUE
l=A
T><I=I.0
TX2,=ZI (I)
XLAMBDA:TX2-(CTXt-Xl)/(TXl-Z211-tl)I*CTX2-Z1CI-l)
GO TO ( 7?O.760 ). KX
XLAM8DA=Z I ( r )
GO Toe 720.760 I. KX
ALFA(IJ=AIO+XALOG
ALFA(3'=A30-XALOG
X2;RANf"(-11
ALFA(21=A17+A35*X2
DO 730 J= 1.3
1)4XLAMBDA+O.S)
IF (VIGL(ll.LT.O.O) vIGLCi)=-VIGL(I)
IF eVIGLII) .GT.VITLIM) VIGL( I )=VITLIM
IF' (RANF'(-ll.LT.O.Sl VIGL( I '=-VIGL( I]
IFIKTI.NE.O' PRINT 750.Xl.X2,XLAMBDA,VIGL
FORMAT(lH ,6EI9.8)
PART e,DEFINE DURATION FOR EACH POINT OF A
IF KR.EQ.71.0R.IKR.FO.8 GO TO 780
ZMAx=AMAXeKR)/(V3*PIEN,
G-=GNC KR.l NSTRMl
RO-=G/LOGFIZMAX'
QPNOA=t.O/IQ(KR'*PIEN*OA)
GE=ABSFCRO*LOGFCQPNDA,)
XMU=GE/Z.O
SIGMA=GE/4.0
KX=?
CO TO 5gQ
760 TAU=SJGMA*XLAMBOA*I.4142
)
IF (X2.GE.O.S) GO TO 770

GO TO 790
770
IF (XQUR.GE.O.O) GO TO 790
Formalized Music
XEN
XEN
XEN
XEN
XEN
X"N
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XF.N
XEN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
3a5
386
387
388
389
390
391
392
393
394
395
396
397
c
c
c
780 XOUR=O.O
790 tF(KTl.NE.O)PRINT 800.ZMAX,XMU,SIGMA.Xl.XLAMBOA.X2.XDUR
800 FORMAT(IH .5EIS,8,Ell.4,EI5.B)
PART 9.0EFINE INTENSITY FORM TO EACH POINT OF A
IFORM=XINTF!RANF(-ll*BF+O.Sl
JF (KTt.EO.O, GO TO 840
IF (NLtNE.LT.KNL) GO TO 810
IF (NLtNE.EO.KNL) GO TO 820
NLINE'I
GO TO 900
BIO NLINE=NLINE+I
GO TO QOO
820 PRINT A30
830 {lHl)
NLlNE=O
GO TO 900
840 IF CNLINE.GE.KNL) GO TO esc
NLtNE=NLINE+l
"EN
XEN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
"EN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
XEN
GO TO B80
850 PRINT A60.JW,A,NA,(Q[ll.t:t.KTR) XEN
860 FORMATe*1
*1,//) XEN
PRINT 870 XEN
870 FORMAT (6X. *N*, 8X .*ST ART* .5X,*CLASS* ,4X t" I NSTRM*,4X .*PI TCH* ,6X, XEN
OYNAM*) XEN
NLtNE=t XEN
880 PRINT A90.N.TA,KR.INSTRM,HX,IVIGL(IJ,J=1.3).XDU
R
,IFORM
8QfI FOQMAT(lH .17tF12.2.I9.18,Fl1.l. F 13.1.2FlO.t,F14.2,tl1J
c
"EN
XEN
XEN
"EN
XEN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
XEN
"EN
XEN
XEN
XEN
XEN
XEN
xEN
"EN
XEN
"EN
C PART SAME FOR ALL POINTS OF A
c
c
c
c
C
C
C
C
900 IF [N.LT.NAl GO TO 400
910
'."120
Q3C
940
PART 11' REPEAT SEQUENCES A
IF (KTFST2.EO.0) GO TO 910
TAP2=TIMEFIl)-TAV2
TAP?=TAP2/FLOATFCNA)
PPINT 750,TAP2
IF CJW.GE.KWI GO TO 930
J\tI=JIlI+t
IF (GTNS.GT.SINAI GO TO 220
IF (KTFSTt.F.O.O) CALL FXIT

TAPt=TAPI/FLOATF(KW)
PRINT 7c:iO,TAPl
END
DATA IST/tO-3. 0609(2)
C

265700276300286900297400307900318300328600338900349100359300369400379400
389300399200409000418700428400438000447500456900466200475500484700493700

151
398
399
400
401
,,02
403
,,0.
405
.06
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
44B
152
Formalized Music

87330067
68
0088020088350088680089000089310089610089910090200090
48
00907600
910300913000915500918100920500922900925200927500929700931900934000936100
93810094000094
1
900
9
43800945700947300949000950700952300953800955400956900
9583009597009611009624009637009649009661 0 0967300968400969500970600971600

989100989500989900990300990700991100991500001800992200902500992800993100
993400993700993900994200994400994700994 900995100995300995500995700995900
99610099630099640099660099670C996900997000997200997300997
4
00997500997600
997700997900997960998050998140998230998320998400998480998550998620998680
998740998800998850998910998960999010999060999100999140999180999230999270
999300999340999370999400999440999470999500999530999560999580999600999630
9906509Q9670999690999700
25509997000026309998000027509999n000313099999000346099999900377099999990
4
0
60999999991 0 QE30 100000000
04005010020017730035563177245390100000071000000000012000
0000150500500120720001600025000120101020309020201010202
01010000100700101000010090010100001012001010000101100101000010090
010100001012001010000100B0010100001008001010000101200101000010080

175500001099"
3975000015999

348500001540015630000154001953000010200
397500001515029710000100901754000007090 17550000100903363000010090
19530000100701013000010200348 5000015200 1563000015020
00003467005000000154800500

0000326810990
0000336310999
0000195310A000000101307200
00003487155000000157215500
250 80408011309
08071602010110
030 30420010110
0205032'5010112

1'l210(J30210:'lCl07
020 20203150207
020 20202410207
0.3090317041609
03132003200509
020 52801030409
450 112020 20106
Frce Stochastic Music by Computer 153
JW= 1 A= 9.13 NA-" <;5
QIII=0.12/0.04/0.04/0.05/0.12/0.29/0.04/0.04/0.l4/0.06IO.06/C.031
/>; START CLASS INSTRM PITCH Gll S5 1 GLI S52 GLI 55) DURATION OYNAM
0.(10 7 1 34.0 0.0 0.0 0.0
C.I'() 3
1
".41
50
2 0.10
1('
1 43.2 0.0 (l.0 0.0
3 O.ll 6 8 81.3 0.0 1).0 0.0
().63 21
"
0.13 6 3 47.0 0.0 0.0 0.0 (l.lB 10
5 O.lB 1 4 o./) 0.0 0.0 0.0 1.90 29
f 0.25 9 1 48.7 0.0 0.0 0.0 0.51 35
7 /).33 6 7 11.4 0.0 0.0 0.0 0.37 42
n.34 'l
1 38.1 0.0 0.0 0.0
0.00 59
8
45
c; 0.41) 1 1 Il.O 0.0 0.0 0.0 2.20
10 0.41 I: 9 55.(1 1).0 0.0 0.0 1.01
()
o .4(J 1
1] (.11: 6 7 1l.5 0.0 0.0 0.0
12 (1.90 B 2 23.2 0.0 0.1) 0.0 n.oo 19
6
13 1.00 1 2 26.9 0.0 0.0 0.0 0.00.
1'< l.r. 9 1 '<6.2 0.1) n.O 0.0 0.32 51
15 1.(19 6 2 68.5 0.0 ('.0 0.0 0.71 25
if 1.23 6 3 46.9 0.0 0.0 o.n 0.64 32
17 1.42 6 1 0.0 C.O 0.0 0.44 1
18 1.51 10 1 36.2 1).0 (\.0 0.0 0.22 21
Ie; 1.65 4 2 32.5 0.0 0.0 0.0 1.09 13
20 1.18 6 B 12.6 0.0 0.0 0.0
CI.O(: 60
21 1.92 6 3 38.9 0.0 0.0 0.0 0.55 60
22 1.94 5 1 11,.6 71.0 -25.0 -71.0 0.80 62
23 2.18 4 1 32.6 0.0 0.0 0.0 1.50 50
2.18 6 6 50.9 0.0 0.0 0.0 0.61) 26
2.19 1 12 0.0 0.0 O.Q 0.0 '<.58 24
U: 2.20
"
1 49.3 0.(1 0.0 0.0 0.02 58
27 2.B 9 1 51.0 0.0 0.0 0.0 0.22 13
2E 2.32 1 1 36.9 0.0 1:'.0 0.0
(l.On 43
2.33 4 1 31.8 1).0 Q.O 0.0 1.38 56

?O 2.54 1 6 (l.O 1).0 0.0 0.0 0.28 14
2.51 II 2 12.2 0.(' c.o 0.0
1.6e; 40

2.71
<;
1 48.5 0.0 0.0 0.0 0.37 55

33 2.80 1 5 0.0 /).0 0.0 0.0 1.50 58
3.28 5 2 15.4 49.0 5.(' -31.(\ 0.52 21
3!: 3.33 1 7 0.1) 0.0 0.0 0.0 1.38 8
3E 3.38 5 2 41.3 -71.0 -17.0 46.0 1.05 4
31 3.55 H' 1 37.6
0." 0.0 0.0 0.14 24
38 3.56 1 9 0.0 0.1'1 0.0 0.0 1.30 0
39 3.tl)
<;
1 1:4.3 0.0 0.0 0.0 ('I.1'l 13
41) 3.64 12 2 52.2 0.(' 0.0 0.0 3.72 9
41 3.65 6 5 59.0 0.0 Q.O 1:'.0 0.83 28
"2
3.71 5 3 38.8 25.1'1 2.0 -15.0 0.0(\ 11
"3 3.80 6 8 15.6 0.(1 C.O 0.0 0.43 17
44 3.87 6 2 51.5 0.0 C.O 0.0
('1.17 57
'<5 3.89 f: 7 12.1 0.0 (l.O O.Q (1.3'> 2
'It 4.15 5 2 43.0 -71.0 24.0 71.0 1.16 2
'I'
4.15 5 1 eO.3 36.(' 4.0 22.0 0.85 50
'Ie 4.25 9 1 59.9 0.0 0.0 0.0 r.lC In
... 31 12 2 40.1 0.0 0.0 0.0 2.4<; 33
5(1 1 In n.o 0.0 c.o 0.0 ('1.41: 34
Fig. V--4. Provisional Results of One Phase of the Analysis
154
Formalized Music
Chapter VI
Symbolic Music
Here we shall attack the thorny problem of the logic underlying musical
composition. Logic, that queen of knowledge, monopolized by mathematics,
wavers between her own name, borne through two millennia, and the name
of algebra.
Let us leave the task of logically connecting the preceding chapters
for the moment. We shall confine ourselves to following a path which
may lead us to regions even more harmonious in the not too distant future.
A LOGICAL AND ALGEBRAIC SKETCH OF
MUSICAL COMPOSITION
In this chapter we shall begin by imagining that we are suffering from
a sudden amnesia. We shall thus be able to reascend to the fountain-head
of the mental operations used in composition and attempt to extricate the
general principles that are valid for all sorts of music. We shall not make a
psycho-physiological study of perception, but shall simply try to understand
more clearly the phenomenon of hearing and the thought-processes involved
when listening to music. In this way we hope to forge a tool for the better
comprehension of the works of the past and for the construction of new
music. We shall therefore be obliged to collect, cut up, and solder scattered
as well as organized entities and conceptions, while unraveling the thin
thread of a logic, which will certainly present lacunae, but which will at
least have the merit of existing.
CASE OF A SINGLE GENERIC ELEMENT
Let there be a sonic event which is not endless. It is seen as a whole, as
155
an entity, and this overall perception is sufficient for the moment. Because
of our amnesia, we decide that it is neuter-neither pleasant nor unpleasant.
Postulate. We shall systematically refuse a qualitative judgment on
every sonic event. What will count will be the abstract relations within the
event or between several events, and the logical operations which may be
imposed on them. The emission ofthe sonic event is thus a kind of statement,
inscription, or sonic symbol, which may be notated graphically by the
letter a.
Ifit is emitted once it means nothing more than a single existence which
appears and then disappears; we simply have a.
Ifit is emitted several times in succession, the events are compared and
we conclude that they are identical, and no more. Identity and tautology
are therefore implied by a repetition. But simultaneously another phe"-
nomenon, subjacent to the first, is created by reason of this very repetition:
modulation of time. If the event were a Morse sound, the temporal abscissa
would take a meaning external to the sound and independent of it. In addi-
tion to the deduction of tautology, then, repetition causes the appearance
of a new phenomenon, which is inscribed in time and which modulates time.
To summarize: Ifno account is taken of the temporal element, then a
single sonic event signifies only its statement. The sign, the symbol, the
generic element a have been stated. A sonic event actually or mentally
repeated signifies only an identity, a tautology:
a V a Y a Va Y a = a.
Y is an operator that means "put side by side without regard to time."
The = sign means that it is the same thing. This is all that can be done with
a single sonic event.
CASE OF TWO OR MORE GENERIC ELEMENTS
Let there be two sonic events a and b such that a is not identical with b,
and such that the two are distinct and easily recognizable, like the letters
a and b, for example, which are only confused by a near-sighted person or
when they are poorly written.
If no account is taken of the temporal element, then the two elements
are considered as a pair. Consequently emitting first a then h, or first h
then a, gives us no more information about these distinct events than when
they are heard in isolation after long intervals of silence. And since no
account is taken of the relation of similitude or of the time factor, we can
Symbolic Music 157
write for a # b
a V b = b Y a,
which means that a and b side by side do not create a new thing, having the
same meaning as before. Therefore a commutative law exists.
In the case of three distinguishable events, a, b, c, a combination of two
of these sonic symbols may be considered as forming another element, an
entity in relation to the third:
(a Y b) Y c.
But since this associational operation produces nothing more we may write
(a V b) Y c = a V (b V c).
This is an associative law.
The exclusion of the time factor leads therefore to two rules of compo-
sition outside-time-the commutative and thc associative. (These two rules
are extensible to the case of a single event.)
On the other hand, when the manifestations of the generic events
a, b, c are considered in time, then commutativity may no longer be accepted.
Thus
aTb#bTa,
T being the symbol of the law of composition which means" anterior to."
This asymmetry is the result of our traditional experience, of our cus-
tomary one-to-one correspondence between events and time instants. It
is raised when we consider time by itself without events, and the conse-
quent metric time which admits both the commutative and the associative
properties:
aTb=bTa
(a T b) T C = aT (b T c)
CONCEPT OF DISTANCE (INTERVAL)
commutative law
associative law.
The consideration of generic clcmcnts a, h, c, ... as entities does not
permit much of an advance. To exploit and clarify what has just been said,
we must penetrate the internal organization of the sonic symbols.
Every sonic event is perceived as a set of qualities that is modified dur-
ing its life. On a primary level we perceive pitch, duration, timbre, attack,
rugosity, etc. On another level we may distinguish complexities, degrees of
order, variabilities, densitics, homogeneities, fluctuations, thicknesses, etc.
Our study will not attempt to elucidate these questions, which are not only
158
Formalized Music
difficult but at this moment secondary. They arc also secondary because
many of the qualities may be graduated, even if only broadly, and may be
totally ordered. We shall therefore choose one quality and what will be
said about it will be extensible to others.
Let us, then, consider a series of events discernible solely by pitch, such
as is perceived by an observer who has lost his memory. Two clements, a, b,
are not enough for him to create the notion of distance or interval. We must
look for a third term, e, in order that the observer may, by successive com-
parisons and through his immediate sensations, form first, the concept of
relative size (b compared to a and c), which is a primary expression of rank-
ing; and then the notion of distance, of interval. This mental toil will end
in the totally ordered classification not only o[ pitches, but also of melodic
intervals. Given the set of pitch intervals
Ii = (ha> ho, he, ... )
and the binary relation S (greater than or equal to), we have
1. hSh [or all h E H, hence reflexivity;
2. haShb =f hbSha except [or ha = ho, hence antisymmetry;
3. haShb and hbShc entail haShe, hence transitivity.
Thus the different aspects o[ the sensations produced by sonic events
may eventually totally or partially constitute ordered sets according to the
unit interval adopted. For example, if we adopted as the unit interval of
pitch, not the relationship of the semitone 1.059) but a relationship of
1.00001, then the sets of pitches and intervals would be very vague and
would not be totally ordered because the differential sensitivity of the human
is inferior to this relationship. Generally for a sufficiently large unit
dIstance, many of the qualities of sonic events can be totally ordered.
To with a first-degree acoustic experience, we shall suppose
that the ultImate aspects of sonic events are frequency! (experienced as
pitch), intensity, and duration, and that every sonic event may be constructed
from these three when duly interwoven. In this case the number three is
irreducible. For other assumptions on the microstructure of sonic events see
the Preface and Chapter IX.
Structure of the Qualities of Sonic Events*
. From a naive musical practice we have defined the concept of interval
or dIstance. Now let us examine sets of intervals which are in fact isomorphic
to the equivalence classes of the N x N product set of natural numbers.
Symbolic Music
159
1. Let there be a set H of pitch intervals (melodic). The law of internal
composition states that to every couple (ha, ho E H) a third element may be
made to correspond. This is the composite of ha by hb' which we shall notate
h + h - h such that h E H For example, let there be three sounds
a b - c, c
characterized by the pitches I, II, Ill, and let h(I.Il)' h(II.III) be the intervals
in semi tones separating the couples (I, II) and (II, III), respectively. The
interval h(I.III) separating sound I and sound III will be equal to the sum of
the semi tones of the other two. We may therefore establish that the law of
internal composition for conjuncted intervals is addition.
2. The law is associative:
Ita + (Ito + he) = (ha + hb) + he = Ita + hb + he
3. There exists a neutral element ho such that for every Ita E H,
ho + ha = ha + ho = ha
For pitch the neutral element has a name, unison, or the zero interval; for
intensity the zero interval is nameless; and for duration it is simultaneity.
4. For every ha there exists a special element called the inverse, such
that
+ ha = ha + = ho = O.
Corresponding to an ascending melodic interval ha, there may be a descend-
ing interval which returns to the unison; to an increasing interval of
intensity (expressed in positive decibels) may be added another diminishing
interval (in negative db), such that it cancels the other's effect; correspond-
ing to a positive time interval there may be a negative duration, such that
the sum of the two is zero, or simultaneity.
5. The law is commutative:
* Following Peano, we may state an axiomatics of pitch and construct the chro-
matic or whole-tone scale by means of three primary terms-origin, note, and the
successor of. .. -and five primary propositions:
1. the origin is a note;
2. the successor of a note is a note;
3. notes having the same successor are identical;
4. the origin is not the successor of any note; and
5. if a property applies to the origin, and if when it applies to any note it also
applies to its successor, then it applies to all notes (principal of induction).
See also Chap. VII, p. 194.
I I
These five axioms have been established for pitch, outside-time. But
the examples have extended them to the two other fundamental factors of
sonic events, and we may state that the sets Ii (pitch intervals), G (intensity
intervals), and U (durations) are furnished with an Abelian additive group
structure.
To specify properly the difference and the relationship that exists
between the temporal set T and the other sets examined outside-time,
and in order not to confuse, for example, set U (durations characterizing a
sonic event) with the time intervals chronologically separating sonic
events belonging to set T, we shall summarize the successive stages of our
comprehension.
SUMMARY
Let there be three events a, b, c emitted successively.
First stage: Three events are distinguished, and that is all.
Second stage: A "temporal succession" is distinguished, i.e., a corre-
spondence between events and moments. There results from this
a before b - b before a (non-commutativity) .
Third stage: Three sonic events are distinguished which divide time into
two sections within the even ts. These two sections may be com pared and then
expressed in multiples of a unit. Time becomes metric and the sections
constitute generic elements of set T. They thus enjoy commutativity.
According to Piaget, the concept of time among children passes through
these three phases (see Bibliography for Chapter VI).
Fourth stage: Three sonic events are distinguished; the time intervals
are distinguished; and independence between the sonic events and the
time intervals is recognized. An algebra outside-time is thus admitted for sonic
events, and a secondary temporal algebra exists for temporal intervals; the
two algebras are otherwise identical. (It is useless to repeat the arguments
in order to show that the temporal intervals between the events constitute
a set T, which is furnished with an Abelian additive group structure.)
Finally, one-to-one correspondences are admitted between algebraic func-
tions outside-time and temporal algebraic functions. They may constitute
an algebra in-time.
In conclusion, most musical analysis and construction may be based on:
1. the study of an entity, the sonic event, which, according to our temporary
assumption groups three characteristics, pitch, intensity, and duration, and
which possesses a structure outside-time; 2. the study of another simpler entity,
Symbolic Music
161
time, which possesses a temporal structure; and 3. the correspondence between
the structure outside-time and the temporal structure: the structure in-time.
Vector Space
Sets H (melodic intervals), G (intensity intervals), U (time intervals),
and T (intervals of time separating the sonic events, and independent of
them) are totally ordered. We also assume that they may be isomorphic
under certain conditions with set R of the real numbers, and that an external
law of composition for each of them may be established with set R. For every
a E E (E is anyone of the above sets) and for every element A E R, there
exists an element b = Aa such that bEE. For another approach to vector
space, see the discussion of sets of intervals as a product of a group times a
field, Chap. VIII, p. 210.
Let X be a sequence of three numbers Xl, X2, X
3
, corresponding to the
elements of the sets H, G, U, respectively, and arranged in a certain order:
X = (Xl' X2, X3)' This sequence is a vector and Xl' X2, X3 are its components.
The particular case of the vector in which all the components are zero is a
zero vector, G. It may also be called thc origin of the coordinates, and by
analogy with elementary geometry, the vector with the numbers (Xl' X2, X3 )
as components will be called point M of coordinates (Xl' X2, x3). Two points
or vectors are said to be equal if thcy are defined by the same sequence:
Xi =YI'
The set of these sequences constitutes a vector space in three dimensions,
E
3
There exist two laws of composition relative to E3: 1. An internal law of
composition, addition: If X = (Xl> X2, X3) and Y = (YI' Y2, Y3), then
X + Y = (Xl + Yl, X2 + Y2' X3 + Y3)'
The following properties are verified: a. X + Y = Y + X (commutative);
b. X + (Y + Z) = (X + Y) + Z (associative); and e. Givcn two vectors
X and Y, there exists a single vector Z = (Zl' Z2, Z3) such that X = Y +
Z. We have Zj = Xi - Yj; Z is called the difference of X and Yand is nota-
ted Z = X - Y. In particular X + G = G + X = X; and each vector X
may be associated with the opposite vector( - X), with components (- Xl,
-X2' -x
3
), such that X + (-X) = G.
2. An external law of composition, multiplication by a scalar: If
pER and X E E, then
pX = (px
l
, PX2' PX3) E E3
The following properties are verified for (p, q) E R: a. 1 X = X; b. P(qX)
(pq)X(associative);andc.(p + q)X = pX + qXandp(X + Y) = pX + PY
(distributive).
BASIS AND REFERENT OF A VECTOR SPACE
Ifit is impossible to find a system of p numbers aI' a
2
, a
3
, .. , a
p
which
are not all zero, such that
alXI + a2X2 + ... + apXp = 0,
and on the condition that the p vectors Xl' X
2
, , Xp of the space En are
not zero, then we shall say that these vectors are linearly independent.
Suppose a vector of En, of which the ith component is I, and the others
are O. This vector it is the ith unit vector of En. There exist then 3 unit vectors
of E3, for example, Ii, g, ii, corresponding to the sets Ii, G, U, respectively;
and these three vectors are linearly independent, [or the relation
a/i + a
2
g + a
3
ii = 0
entails al = a2 = a3 = O. Moreover, every vector X = (Xl' X
2
, xs) of E
may be written
x = Xlii + x
2
g + x3ii.
It immediately results from this that there may not exist in E3 more than
3 linearly independent vectors. The set ti, g, ii, constitutes a basis of E. By
analogy with elementary geometry, we can say that Oh, Og, Ou, are axes
of coordinates, and that their set constitutes a riferent of E
3
In such a space,
all the referents have the same origin O.
Linear vectorial multiplicity. We say that a set V of vectors of En which is
non-empty constitutes a linear vectorial multiplicity if it possesses the following
properties:
1. If X is a vector of V, every vector pX belongs also to V whatever the
scalar p may be.
2. If X and Yare two vectors of V, X + Yalso belongs to V. From this
we deduce that: a. all linear vectorial multiplicity contains the vector
O(O.X = OJ; and b. every linear combination alX
1
+ a
2
X
2
+ ... + apXp
of p vectors o[ V is a vector of V.
REMARKS
1. Every sonic event may be expressed as a vectorial multiplicity.
2. There exists only one base, Ii, g, it. Every other quality of the sounds
and every other more complex component should be analyzed as a linear
combination of these three unit vectors. The dimension of V is therefore 3.
Symbolic Music 163
3. The scalars p, q, may not in practice take all values, [or we would
then move out o[ the audible area. But this restriction o[ a practical order
does not invalidate the generality of these arguments and their applications.
For example, let 0 be the origin of a trihedral of reference with Oh,
Og, Ou, as referent, and a base Ii, g, ii, with the following units:
[or Ii, 1 = semitone;
for g, 1 = 10 decibels;
for ii, 1 = second.
The origin 0 will be chosen arbitrarily on the "absolute" scales established
by tradition, in the manner of zero on the thermometer. Thus:
and the vectors
for Ii, 0 will be at C
3
;
[or g, 0 will be at 50 db;
for ii, 0 will be at 10 sec;
Xl = 51i - 3g + 5ii
X
2
= 71i + ig - iii
may be written in traditional notation for 1 sec ~ J.
11' ~ ( 50 - 50 = 20 ctB)
In the same way
(As = 440 Hz)
Xl + X
2
= (5 + 7)/i + (1 - 3)g + (5 - i)ii = i21i - 2g + 4ii.
= J
We may similarly pursue the verification of all the preceding propositions.
We have established, thanks to vectorial algebra, a working language
which may permit both analyses of the works of the past and new construc-
tions by setting up interacting functions of the components (combinations
of the sets H, G, UJ. Algebraic research in conjunction with experimental
research by computers coupled to analogue converters might give us
information on the linear relations of a vectorial multiplicity so as to obtain
the timbres of existing instruments or of other kinds of sonic events.
The following is an analysis of a fragment of Sonata, Op. 57 (Appas-
sionata), by Beethoven (see Fig. VI-I). We do not take the timbre into
account since the piano is considered to have only one timbre, homogeneous
over the register of this fragment.
A, r -------,
8 1
Fig. VI-1
A A
I
------
Assume as unit vectors: Ii, for which 1 ~ semi tone; g, for which
1 ~ 10 db; and ii, for which 1 ~ ). Assume for the origins
on the Ii axis,
ff = 60 db (invariable) on the g axis, and
5) on the ii axis.
ALGEBRA OUTSIDE-TIME (OPERATIONS AND RELATIONS IN SET A)
The vector Xo = 18/i + o.g + 5ii corresponds to G.
The vector Xl = (18 + 3)/i + Og + 417 corresponds to B ~ .
The vector X
2
= (18 + 6)/i + OJ + 317 corresponds to Db.
The vector X3 = (I8 + 9)/i + Og + 2ii corresponds to E.
The vector X
4
= (18 + 12)1i + Og + Iii corresponds to G.
The vector X5 = (18 + O)/i + OJ + 1U corresponds to G.
(See Fig. VI-2.)
Let us also admit the free vector v = 3/i + Og - Iil; then the vectors
XI (for i = 0, 1,2,3,4) are of the form XI = Xo + vi.
We notice that set A consists of two vector families, Xi and iv, combined
by means of addition.
Symbolic Music 165
SI.l
o
Fig. VI-2
A second law of composition exists in the set (i = 0, 1, 2, 3, 4); it is
an arithmetic progression.
Finally, the scalar i leads to an anti symmetric variation of the com-
ponents /i and ii of XI' the second j remaining invariant.
TEMPORAL ALGEBRA (IN SET T)
The sonic statement of the vectors Xi of set A is successive:
Xo T Xl T X
2
T ...
T being the operator "before."
This boils down to saying that the origin a ofthe base of A ~ E3 ~ V
is displaced on the axis of time, a shifting that has nothing to do with the
change of the base, which is in fact an operation within space E3 of base
Ii, g, ii. Thus in the case ofa simultaneity (a chord) of the attacks of the six
vectors described for set A, the displacement would be zero.
In Fig. VI-3 the segments designated on the axis of time by the origins
a of Xi are equal and obey the function 6.t
l
= 6.t
j
, which is an internal law
166
Formalized Music
of composition in set T; or consider an origin 0' on the axis of time and a
segment unit equal to f::..t; then I, = a + if::..t, for i = 1, 2, 3, 4, 5.
Fig. VI-3
ALGEBRA IN-TIME (RELATIONS BETWEEN SPACE E3 AND SET T)
We may say that the vectors Xl of A have components H, G, U, which
may be expressed as a function of a parameter ii' Here Ii = if::..t; the values
are lexicographically ordered and defined by the increasing order i = 1, 2,
3, 4, 5. This constitutes an association of each of the components with the
ordered set T. It is therefore an algebration of sonic events that is indepen-
dent of time (algebra outside-time), as well as an algebration of sonic events
as a function of time (algebra in-time).
In general we admit that a vector X is a function of the parameter of
time t if its components are also a function of t. This is written
X(t) = H(t)li + G(t)g + U(t)ii.
When these functions are continuous they have differentials. What is
the meaning of the variations of X as a function of time t? Suppose
dX dH dG dU
(jj = de fi + dt g + dt ii.
If we neglect the variation of the component G, we will have the following
conditions: For dH/dt = 0, H = Ch, and dU/dt = 0, U = c
u
, Hand U will
be independent of the variation of t; and for Cit and C
u
of. 0, the sonic event
will be of invariable pitch and duration. If Cit and C
u
= 0, there is no sound
(silence). (See Fig. VI-4.)
For dH/dt = 0, H = Cit, and dU/dt = c
u
, U = cut + k, if Cit and
C
u
of- 0, we have an infinity of vectors at the unison. If e
u
= 0, then we have
a single vector of constant pitch Cit and duration U = k. (See Fig. VI-5).
Symbolic Music
167
For dH/dt = 0, H = Cit, and dU/dt =1(t), U = F(t), we have an
infinite family of vectors at the unison.
For dH/dt = Cit, H = Cltt + k, and dU/dt = 0, U = cu, jf Cu < 8,
lim e = 0 we have a constant glissando of a single sound. If Cu > 0, then
we have ~ chord composed of an infinity of vectors of duration Cu (thick
constant glissando). (See Fig. VI.:...6.)
Fig. VI-4
Fig. VI-5
Fig. VI-6
C
h
-- ------- _ .. 1-1 """"""F===f,==::!::::=:::::L---l1 - - --
c
14
,t ... "
-'" -
t, LL
:t,a
168
Formalized Music
For dHfdt = Ch , H = Cht + k, and dU/dt = c
u
, U = cut + r, we have a
chord of an infinity of vectors of variable durations and pitches. (See
Fig. VI-7.)
u= H:
U .c",t ... k
Fig. VI-7
"',iA
For dHfdt = Ch , H = Cht + k, and dU/dt = /(t), U = F(t), we have a
chord of an infinity of vectors. (See Fig. VI -8.)
h

tJ; !=(t)
Fig. VI-8
i I li
For dH/dt = J(t), H = F(t), and dU/dt = 0, U = c
u
, if e
u
< e,
lim e = 0, we have a thin variable glissando. If C
u
> 0, then we have a
chord of an infinity of vectors of duration C
u
(thick variable glissando). (See
Fig. VI-9.)
U"Cu.
Fig, VI-9
flU
Symbolic Music 169
For dH/dt =/(t), H = F(t), and dU/dt = s(t), U = S(t), we have a
chord of an infinity of vectors. (See Fig. VI-IO.)
h
u:Slt)
Fig. VI-10
In the example drawn from Beethoven, set A of the vectors Xi is not a
continuous function of t. The correspondence may be written
1
Xo Xl X
2
X3 X
4
X5
4 4
Because of this correspondence the vectors are not commutable.
Set B is analogous to set A. The fundamental difference lies in the change
of base in space E3 relative to the base of A. But we shall not pursue the
analysis.
Remark
If our musical space has two dimensions, e.g., pitch-time, pitch-intensity,
pressure-time, etc., it is interesting to introduce complex variables. Let x be
the time and y the pitch, plotted on the i axis. Then z = x + yi is a sound of
pitch y with the attack at the instant x. Let there be a plane uv with the
following equalities: u = u (x, y), v = v (x, y), and w = u + vi. They define
a mapping which establishes a correspondence between points in the uv
and xy planes. In general any w is a transformation of z.
The four forms of a melodic line (or of a twelve-tone row) can be
represented by the following complex mappings:
w = z, with u = x and v = y, which corresponds to identity (original form)
w = \Z\2/Z, with u = x and v = -y, which corresponds to inversion
w = \ Z\2 / - z with u = - x and v = y, which corresponds to retrogradatiun
w = - z, with u = - x and v = - y, which corresponds to inverted retro-
gradation.
170
Formalized Music
These transformations form Ule Klein
Other transformations, as yet unknown, even to present-day musicians,
could be envisaged. They could be applied to any product of two sets of
soundcharacteristics.Forexample,w = (A Z2 + Bz + C)j(DZ2 + Ez + F),
which can be considered as a combination of two bilinear transformations
separated by a transformation of the type p = 0
2
Furthermore, for a
musical space of more than two dimensions we can introduce hypercomplex
systems such as the system of quaternions.
EXTENSION OF THE THREE ALGEBRAS TO SETS OF
SONIC EVENTS (an application)
We have noted in the above three kinds of algebras:
1. The algebra of the components of a sonic event with its vector
language, independent of the procession of time, an algebra
outside-time.
2. A temporal algebra, which the sonic events create on the axis of metric
time, and which is independent of the vector space.
3. An algebra in-time, issuing from the correspondences and functional
relations between the clements of the set of vectors X and of the set of metric
time, T, independent of the set of X.
All has been said about sonic events themselves, their components,
and about time Can be generalized for sets of sonic events X and for sets T.
In this chapter we have assumed that the reader is familiar with the
of set, and in particular with the concept of the class as it is
111 Boolean algebra. We shall adopt this specific algebra,
whIch IS Isomorphic with the theory of sets.
.To the exposition, we shall first take a concrete example by
e?nsIdermg the referential or universal set R, consisting of all the sounds of a
plano: We shall consider only the pitches; timbres, attacks, intensities, and
will be utilized in order to clarify the exposition of the logical
operatIOns and relations which we shall impose on the set of pitches.
. Suppose, then, a set A of keys that have a characteristic property.
wIll ?e set A, a subset of set R, which consists of all the keys of the
pIan.o. ThIS is chosen a priori and the characteristic property is the
particular chOIce of a certain number of keys.
For the amnesic observer this class may be presented by playing the
keys one after the other, with a period of silence in between. He will deduce
from this that he has heard a collection of sounds, or a listing of elements.
Symbolic Music 171
Another class, B, consisting of a certain number of keys, is chosen in the
same way. It is stated after class A by causing the elements of B to sound.
The observer hearing the two classes, A and B, will note the temporal
fact: A before B; A T B, (T = before). Next he begins to notice relation-
ships between the clements of the two classes. If certain elements or keys are
common to both classes the classes intersect. If none are common, they are
disjoint. If all the clements of B are common to one part of A he deduces that
B is a class included in A. If all the elements of B are found in A, and all the
elements of A are found in B, he deduces that the two classes are indistin-
guishable, that they are equal.
Let us choose A and B in such a way that they have some clements in
common. Let the observer hear first A, then B, then the common part. He
will deduce that: 1. there was a choice of keys, A; 2. there was a second choice
of keys, B; and 3. the part common to A and B was considered. The opera-
tion of intersection (conjunction) has therefore been used:
AB or BA.
This operation has therefore engendered a new class, which was symbolized
by the sonic enumeration of the part common to A and B.
If the observer, having heard A and B, hears a mixture of all the ele-
ments of A and B, he will deduce that a new class is being considered, and
that a logical summation has been performed on the first two classes. This
operation is the union (disjunction) and is written
A + B or B + A.
If class A has been symbolized or played to him and he is made to hear
all the sounds of R except those of A, he will deduce that the complement
of A with respect to R has been chosen. This is a new operation, negation,
which is written A.
Hitherto we have shown by an imaginary experiment that we can
define and state classes of sonic events (while taking precautions for clarity
in the symbolization); and effect three operations of fundamental impor-
tance: intersection, union, and negation .
On the other hand, an observer must undertake an intellectual task
in order to deduce from this both classes and operations. On our plane of
immediate comprehension, we replaced graphic signs by sonic events. We
consider these sonic events as symbols of abstract entities furnished with
abstract logical relations on which we may effect at least the fundamental
operations of the logic of classes. We have not allowed special symbols for
the statement of the classes; only the sonic enumeration of the generic
elements was allowed (though in certain cases, if the classes are already
known and if there is no ambiguity, shortcuts may be taken in the state-
ment to admit a sort of mnemotechnical or even psychophysiological
stenosym bolization).
We have not allowed special sonic symbols for the three operations
which are expressed graphically by " +, - ; only the classes resulting from
these operations are expressed, and the operations are consequently deduced
mentally by the observer. In the same way the observer must deduce the
relation of equality of the two classes, and the relation of implication based
on the concept of inclusion. The empty class, however, may be symbolized
by a duly presented silence. In sum, then, we can only state classes, not the
operations. The following is a list of correspondepces between the sonic
symbolization and the graphical symbolization as we have just defined it:
Graphic symbols
Classes A, B, C, ...
Intersection ( . )
Union (+)
Negation ( - )
Implication (-+)
Membership (E)
A
AB
A+B
A:::>B
A=B
Sonic symbols
Sonic enumeration of the generic
elements having the properties A, B,
C, . (with possible shortcuts)
Sonic enumeration of the elements of
R not included in A
AB
A+B
This table shows that we can reason by pinning down our thoughts by
means of sound. This is true even in the present case where, because of a
concern for economy of means, and in order to remain close to that immedi-
ate .intuition from which all sciences are built, we do not yet wish to propose
some conventions symbolizing the operations " +, -, and the relations
=, -+. Thus propositions of the form A, E, I, 0 may not he symbolized by
sounds, nor may theorems. Syllogisms and demonstrations of theorems may
only be inferred.
Symbolic Music 173
Besides these logical relations and operations outside-time, we have
seen that we may obtain temporal classes (T classes) issuing from the sonic
symbolization that defines distances or intervals on the axis of time. The role
of time is again defined in a new way. It serves primarily as a crucible, mold,
or space in which are inscribed the classes whose relations one must decipher.
Time is in some ways equivalent to the area of a sheet of paper or a black-
board. It is only in a secondary sense that it may be considered as carrying
generic elements (temporal distances) and relations or operations between
these elements (temporal algebra).
Relations and correspondences may be established between these
temporal classes and the outside-time classes, and we may recognize in-time
operations and relations on the class level.
After these general considerations, we shall give an example of musical
composition constructed with the aid of the algebra of classes. For this we
must search out a necessity, a knot of interest.
Construction
Every Boolean expression or function F (A, B, C), for example, of the
three classes A, B, C can be expressed in the form called disjunctive canonic:
where aj = 0; 1 and k
j
= A.BC, A.B.C, A.E.C, A.E.C, .if.BC, .if.BC,
ABC, .if.BC.
A Boolean function with n variables can always be written in such a way
as to bring in a maximum of operations +, ., -, equal to 3n2
n
-
2
- 1.
For n = 3 this number is 17, and is found in the function
F = ABC + AeC + ABC + AEC. (1)
For three classes, each of which intersects with the other two, function (I)
can be represented by the Venn diagram in Fig. VI-II. The flow chart of
the operations is shown in Fig. VI-12.
This same function F can be obtained with only ten operations:
F = (A. B + A E) . C + (A. B + A B) . C. (2)
Its flow chart is given in Fig. VI-13.
Ifwe compare the two expressions ofF, each of which defines a different
procedure in the composition of classes A, B, C, we notice a more elegant
174
Formalized Music
Fig. VI-11
Fig. VI-12
Symbolic Music 175
Fig. VI-13
symmetry in (1) than in (2). On the other hand (2) is more economical (ten
operations as against seventeen). It is this comparison that was chosen for
the realization of Herma, a work for piano. Fig. VI-14 shows the flow chart
that directs the operations of (1) and (2) on two parallel planes, and Fig.
VI-I5 shows the precise plan of the construction of Herma.
The three classes A, B, C result in an appropriate set of keys of the piano.
There exists a stochastic correspondence between the pitch components and
the moments of occurrence in set T, which themselves follow a stochastic
law. The intensities and densities (number of vectors/sec.), as well as the
silences, help clarify the levels of the composition. This work was composed
in 1960-61, and was first performed by the extraordinary Japanese pianist
Yuji Takahashi in Tokyo in February 1962.
176
Formalized Music
In conclusion we can say that our arguments are based on relatively
simple generic elements. With much more complex generic elements we
could still have described the same logical relations and operations. We
would simply have changed the level. An algebra on several parallel levels
is therefore possible with transverse operations and relations between the
various levels.
Fig. VI-14
Symbolic Music
.. I
o
<1\
....

...
r;

..
I

do I
1
1 I
"0 1
I
1
T
-..
v
x

1
1

:1
"I
a! I
vI
00( 1
1
.. 1
1
1
IHt' ...... 1
,'+- >t.i:
1
1
,:.r
1
I
I
I
I
I
1
1

I;;;

m
,
:

1
'" 1
"I
" 1
I
1
1-
4'_
1
,,,
1
;, I
?<g' 1 \oJ I

1
u:
I
IL
\
177
I
Ie;
'0-
)
0')
I .....
Ig
.J::.
IU

u.

12-
E

I
I

Conclusions and Extensions
for Chapters I-VI
I have sketched the general framework of an artistic attitude which, for the
first time, uses mathematics in three fundamental aspects: 1. as a philo-
sophical summary of the entity and its evolution, e.g., Poisson's law; 2. as a
qualitative foundation and mechanism of the Logos, e.g., symbolic logic, set
theory, theory of chain events, game theory; and 3. as an instrument of
mensuration which sharpens investigation, possible realizations, and per-
ception, e.g., entropy calculus, matrix calculus, vector calculus.
To make music means to express human intelligence by sonic means.
This is intelligence in its broadest sense, which includes not only the pere-
grinations of pure logic but also the" logic" of emotions and of intuition.
The technics set forth here, although often rigorous in their internal struc-
ture, leave many openings through which the most complex and mysterious
factors of the intelligence may penetrate. These technics carryon steadily
between two age-old poles, which are unified by modern science and
philosophy: determinism and fatality on the one hand, and free will and
unconditioned choice on the other. Between the two poles actual everyday
life goes on, partly fatalistic, partly modifiable, with the whole gamut of
interpenetrations and interpretations.
In reality formalization and axiomatization constitute a procedural
guide, better suited to modern thought. They permit, at the outset, the
placing of sonic art on a more universal plane. Once more it can be con-
sidered on the same level as the stars, the numbers, and the riches of the
human brain, as it was in the great periods of the ancient civilizations. The
17B
Conclusions and Extensions for Chapters I-VI 179
movements of sounds that cause movements in us in agreement with them
"procure a common pleasure for those who do not know how to reason; and
for those who do know, a reasoned joy through the imitation of the divine
harmony which they realize in perishable movements" (Plato, Timaeus).
The theses advocated in this exposition are an initial sketch, but they
have already been applied and extended. Imagine that all the hypotheses of
generalized stochastic composition as described in Chapter II were to be
applied to the phenomena of vision. Then, instead of acoustic grains, sup-
pose quanta of light, i.e., photons. The components in the atomic, quan-
tic hypothesis of sound-intensity, frequency, density, and lexicographic
time-are then adapted to the quanta of light.
A single source of photons, a photon gun, could theoretically reproduce
the acoustic screens described above through the emission of photons of a
particular choice offrequencies, energies, and densities. In this way we could
create a luminous flow analogous to that of music issuing from a sonic source.
If we then join to this the coordinates of space, we could obtain a spatial
music of light, a sort of space-light. It would only be necessary to activate
photon guns in combination at all corners in a gloriously illuminated area
of space. It is technically possible, but painters would have to emerge from
the lethargy of their craft and forsake their brushes and their hands, unless
a new type of visual artist were to lay hold of these new ideas, technics, and
needs.
A new and rich work of visual art could arise, whose evolution would
be ruled by huge computers (tools vital not only for the calculation of bombs
or price indexes, but also for the artisticlife of the future), a total audiovisual
manifestation ruled in its compositional intelligence by machines serving
other machines, which are, thanks to the scientific arts, directed by man.
Chapter VII
Towards a Metamusic
Today's technocrats and their followers treat music as a message which the
composer (source) sends to a listener (receiver). In this way they believe
that the solution to the problem of the nature of music and of the arts in
general lies in formulae taken from information theory. Drawing up an ac-
count of bits or quanta of information transmitted and received would thus
seem to provide them with "objective" and scientific criteria of aesthetic
value. Yet apart from elementary statistical recipes this theory-which
is valuable for technological communications-has proved incapable of
giving the characteristics of aesthetic value even for a simple melody of
J. S. Bach. Identifications of music with message, with communication, and
with language are schematizations whose tendency is towards absurdities
and desiccations. Certain African tom-toms cannot be included in this
criticism, but they are an exception. Hazy music cannot be forced into too
precise a theoretical mold. Perhaps, it will be possible later when present
theories have been refined and new ones invented.
The followers of information theory or of cybernetics represent one
extreme. At the other end there are the intuitionists, who may be broadly
divided into two groups:
l. The "graphists," who exalt the graphic symbol above the sound of
the music and make a kind of fetish of it. In this group it is the fashionable
thing not to write notes, but to create any sort of design. The" music" is
judged according to the beauty of the drawing. Related to this is the so-called
aleatory music, which is an abuse of language, for the true term should be
English translation of Chapter VII by G. W. Hopkins.
180
Towards a Metamusie 181
the "improvised" music our grandfathers knew. This group is ignorant of
the fact that graphical writing, whether it be symbolic, as in traditional
notation, geometric, or numerical, should be no more than an image that
is as faithful as possible to all the instructions the composer gives to the
orchestra or to the machine.
l
This group is taking music outside itself.
2. Those who add a spectacle in the form of extra-musical scenic action
to accompany the musical performance. Influenced by the" happenings"
which express the confusion of certain artists, these composers take refuge
in mime tics and disparate occurrences and thus betray their very limited
confidence in pure music. In fact they concede certain defeat for their
music in particular.
The two groups share a romantic attitude. They believe in immediate
action and are not much concerned about its control by the mind. But
since musical action, unless it is to risk falling into trivial improvisation,
imprecision, and irresponsibility, imperiously demands reflection, these
groups are in fact denying music and take it outside itsel(
linear Thought
I shall not say, like Aristotle, that the mean path is the best, for in
music-as in politics-the middle means compromise. Rather lucidity and
harshness of critical thought-in other words, action, reflection, and self-
transformation by the sounds themselves-is the path to follow. Thus when
scientific and mathematical thought serve music, or any human creative
activity, it should amalgamate dialectically with intuition. Man is one,
indivisible, and total. He thinks with his belly and feels with his mind. I
would like to propose what, to my mind, covers the term" music" :
1. It is a sort of comportment necessary for whoever thinks it and
makes it.
2. It is an individual pleroma, a realization.
3. It is a fixing in sound of imagined virtualities (cosmological,
philosophical, ... , arguments).
4. It is normative, that is, unconsciously it is a model for being or for
doing by sympathetic drive.
5. It is catalytic: its mere presence permits internal psychic or mental
transformations in the same way as the crystal ball of the hypnotist.
6. It is the gratuitous play of a child.
7. It is a mystical (but atheistic) asceticism. Consequently expressions
of sadness, joy, love, and dramatic situations are only very limited particular
instances.
182
Formalized Music
Musical syntax has undergone considerable upheaval and today it
seems that innumerable possibilities coexist in a state of chaos. We have an
abundance of theories, of (sometimes) individual styles, of more or less
ancient "schools." But how does one make music? What can be communi-
cated by oral teaching? (A burning question, if one is to reform musical
education-a reform that is necessary in the entire world.)
It cannot be said that the informationists or the cyberneticians-much
less the intuitionists-have posed the question of an ideological purge of the
dross accumulated over the centuries as well as by present-day develop-
ments. In general they all remain ignorant of the substratum on which they
found this theory or that action. Yet this substratum exists, and it will allow
us to establish for the first time an axiomatic system, and to bring forth a
formalization which will unify the ancient past, the present, and the future;
moreover it will do so on a planetary scale, comprising the still separate
universes of sound in Asia, Africa, etc.
In 1954
2
I denounced linear thought (polyphony), and demonstrated the
contradictions of serial music. In its place I proposed a world of sound-
masses, vast groups of sound-events, clouds, and galaxies governed by new
characteristics such as density, degree of order, and rate of change, which
required definitions and realizations using probability theory. Thus stochas_
tic music was born. In fact this new, mass-conception with large numbers
was more general than linear polyphony, for it could embrace it as a particu_
lar instance (by reducing the density of the clouds). General harmony?
No, not yet.
Today these ideas and the realizations which accompany them have
been around the world, and the exploration seems to be closed for all
intents and -purposes. However the tempered diatonic system-our musical
terra firma on which all our music is founded-seems not to have been
breached either by reflection or by music itself.
3
This is where the next stage
will come. The exploration and transformations of this system will herald
a new and immensely promising era. In order to understand its determina-
tive importance we must look at its pre-Christian origins and at its subse-
quent development. Thus I shall point out the structure of the music of
ancient Greece; and then that of Byzantine music, which has best preserved
it while developing it, and has done so with greater fidelity than its sister
the occidental plainchant. After demonstrating their abstract logical c o n ~
struction in a modern way, I shall try to express in a simple but universal
mathematical and logical language what was and what might be valid in
time (transverse musicology) and in space (comparative musicology).
Towards a Metamusic 183
In order to do this I propose to make a distinction in musical archi-
tectures or categories between outside-time,
4
in-time, and temporal. A given pitch
scale, for example, is an outside-time architecture, for no horizontal or
vertical combination of its elements can alter it. The event in itself, that is,
its actual occurrence, belongs to the temporal category. Finally, a melody
or a chord on a given scale is produced by relating the outside-time category
to the temporal category. Both are realizations in-time of outside-time con-
structions. I have dealt with this distinction already, but here I shall show
how ancient and Byzantine music can be analyzed with the aid of these cate-
gories. This approach is very general since it permits both a universal
axiomatization and a formalization of many of the aspects of the various
kinds of music of our planet.
Structure of Ancient Music
Originally the Gregorian chant was founded on the structure of ancient
music, pace Combarieu and the others who accused Hucbald of being behind
the times. The rapid evolution of the music of Western Europe after the
ninth century simplified and smoothed out the plainchant, and theory was
left behind by practice. But shreds of the ancient theory can still be found
in the secular music of the fifteenth and sixteenth centuries, witness the
Terminorum Musicae difJinitorium of Johann is Tinctoris.
5
To look at antiquity
scholars have been looking through the lens of the Gregorian chant and its
modes, which have long ceased to be understood. We are only beginning to
glimpse other directions in which the modes of the plainchant can be ex-
plained. Nowadays the specialists are saying that the modes are not in fact
proto-scales, but that they are rather characterized by melodic formulae.
To the best of my knowledge only Jacques Chailley6 has introduced other
concepts complementary to that of the scale, and he would seem to be
correct. I believe we can go further and affirm that ancient music, at least
up to the first centuries of Christianity, was not based at all on scales and
modes related to the octave, but on tetrachords and systems.
Experts on ancient music (with the above exception) have ignored this
fundamental reality, clouded as their minds have been by the tonal con-
struction of post-medieval music. However, this is what the Greeks used
in their music: a hierarchic structure whose complexity proceeded by succes-
sive "nesting," and by inclusions and intersections from the particular
to the general; we can trace its main outline if we follow the writings of
Aristoxenos :
7
A. The primary order consists of the tone and its subdivisions. The whole
184
Formalized Music
tone is defined as the amount by which the interval of a fifth (the penta-
chord, or dia pente) exceeds the interval of a fourth (the tetrachord, or dia
tessaron). The tone is divided into halves, called semitones; thirds, called
chromatic dieseis; and quarters, the extremely small enharmonic dieseis. No
interval smaller than the quarter-tone was used.
B. The secondary order consists of the tetrachord. It is bounded by the
interval of the dia !essaron, which is equal to two and a half tones, or thirty
twelfth-tones, which we shall call Aristoxenean segments. The two outer
notes always maintain the same interval, the fourth, while the two inner notes
are mobile. The positions of the inner notes determine the three genera of
the tetrachord (the intervals of the fifth and the octave play no part in it).
The position of the notes in the tetrachord are always counted from the
lowest note up:
1. The enharmonic genus contains two enharmonic dieseis, or
3 + 3 + 24 = 30 segments. If X equals the value of a tone, we can express
the enharmonic as Xl14. Xl14. X2 = XSI2.
2. The chromatic genus consists of three types: a. soft, containing two
chromatic dieseis, 4 + 4 + 22 = 30, or X1/3XlI3X(lf3+312) = X
5
12; h.
hemiolon (sesquialterus), containing two hemioloi dieseis,4.5 + 4.5 + 21
= 30 segments, or X(312)(11
4
). X(312)(114). X71
4
= X512; and c. "toniaion," con-
sisting of two semitones and a trihemitone, 6 + 6 + 18 = 30 segments,
or XlI2.X1I2.X
3
12 = X512.
3. The diatonic consists of: a. soft, containing a semi tone, then three
enharmonic dieseis, then five enharmonic dieseis, 6 + 9 + 15 = 30 seg-
ments, or Xl12. X314. X5/4 = X512; b. syntonon, containing a semitone, a
whole tone, and another whole tone, 6 + 12 + 12 = 30 segments, or
X1I2.X.X = X5/2.
C. The tertiary order, or the system, is essentially a combination of the
elements of the first two-tones and tetra chords either conjuncted or
separated by a tone. Thus we get the pentachord (outer interval the perfect
fifth) and the octochord (outer interval the octave, sometimes perfect). The
subdivisions of the system follow exactly those of the tetrachord. They are
also a function of connexity and of consonance.
D. The quaternary order consists of the tropes, the keys, or the modes,
which were probably just particularizations of the systems, derived by
means of cadential, melodic, dominant, registral, and other formulae, as in
Byzantine music, ragas, etc.
These orders account for the outside-time structure of Hellenic music.
After Aristoxenos all the ancient texts one can consult on this matter give
Towards a Metamusic
185
this same hierarchical procedure. Seemingly Aristoxenos was used as a
model. But later, traditions parallel to Aristoxenos, defective interpretations,
and sediments distorted this hierarchy, even in ancient times. Moreover, it
seems that theoreticians like Aristides Quintilianos and Claudios Ptolemaeos
had but little acquaintance with music.
This hierarchical" tree" was completed by transition algorithms-
the metabolae-from one genus to another, from one system to another, or
from one mode to another. This is a far cry from the simple modulations or
transpositions of post-medieval tonal music.
Pentachords are subdivided into the same genera as the tetrachord they
contain. They are derived from tetrachords, but nonetheless are used as
primary concepts, on the same footing as the tetrachord, in order to define
the interval of a tone. This vicious circle is accounted for by Aristoxenos'
determination to remain faithful to musical experience (on which he insists),
which alone defines the structure of tetrachords and of the entire harmonic
edifice which results combinatorially from them. His whole axiomatics
proceeds from there and his text is an example of a method to be followed.
Yet the absolute (physical) value of the interval dia tessaron is left undefined,
whereas the Pythagoreans defin,ed it by the ratio 3/4 of the lengths of the
strings. I believe this to be a sign of Aristoxenos' wisdom; the ratio 3/4
could in fact be a mean value.
Two languages
Attention must be drawn to the fact that he makes use of the additive
operation for the intervals, thus foreshadowing logarithms before their
time; this contrasts with the practice of the Pythagoreans, who used the
geometrical (exponential) language, which is multiplicative. Here, the
method of Aristoxenos is fundamental since: 1. it constitutes one of the two
ways in which musical theory has been expressed over the millennia; 2. by
using addition it institutes a means of" calculation" that is more economi-
cal, simpler, and better suited to music; and 3. it lays the foundation of the
tempered scale nearly twenty centuries before it was applied in Western
Europe.
Over the centuries the two languages-arithmetic (operating by
addition) and geometric (derived from the ratios of string lengths, and
operating by multiplication)-have always intermingled and interpene-
trated so as to create much useless confusion in the reckoning of intervals
and consonances, and consequently in theories. In fact they arc both ex-
pressions of group structure, having two non-identical operations; thus they
have a formal equivalence.
8
There is a hare-brained notion that has been sanctimoniously repeated
by musicologists in recent times. "The Greeks," they say, "had descending
scales instead of the ascending ones we have toda y." Yet there is no trace
of this in either Aristoxenos or his successors, including Quintilianos
9
and
Alypios, who give a new and fuller version of the steps of many of the tropes.
On the contrary, the ancient writers always begin their theoretical
tions and nomenclature of the steps from the bottom. Another bit of foolish-
ness is the supposed Aristoxenean scale, of which no trace is to be found in
his text.
10
Structure of Byzantine Music
Now we shall look at the structure of Byzantine music. It can contribute
to an infinitely better understanding of ancient music, occidental plain-
chant, non-European musical traditions, and the dialectics of recent Euro-
pean music, with its wrong turns and dead-ends. It can also serve to foresee
and construct the future from a view commanding the remote landscapes
of the past as well as the electronic future. Thus new directions of research
would acquire their full value. By contrast the deficiencies of serial music in
certain domains and the damage it has done to musical evolution by its
ignorant dogmatism will be indirectly exposed.
Byzantine music amalgamates the two means of calculation, the
Pythagorean and the Aristoxenean, the multiplicative and the additive.
ll
The fourth is expressed by the ratio 3/4 of the monochord, or by the 30
tempered segments (72 to the octave)Y It contains three kinds of tones:
major (9/8 or 12 segments), minor (10/9 or 10 segments), and minimal
(16(15 or 8 segments). But smaller and larger intervals are constructed and
the elementary units of the primary order are more complex than in
Aristoxenos. Byzantine music gives a preponderant role to the natural
diatonic scale (the supposed Aristoxenean scale) whose steps are in the
ing ratios to the first note: 1,9/8, 5/4,4(3,27/16, 15/8,2 (in segments 0,
12, 22, 30, 42, 54, 64, 72; or 0, 12, 23, 30, 42, 54, 65, 72). The degrees of
this scale bear the alphabetical names A, B, r, f:.., E, Z, and H. f:.. is the
lowest note and corresponds roughly to G
2
This scale was propounded at
least as far back as the first century by Didymos, and in the second century
by Ptolemy, who permuted one term and recorded the shift of the tetra_
chord (tone-tone-semitone), which has remained unchanged ever since.
l3
But apart from this dia pason (octave) attraction, the musical architecture is
hierarchical and" nested" as in Aristoxenos, as follows:
A. The primary order is based on the three tones 9/8, 10/9, 16/15, a
Towards a Metamusic
187
supermajor tone 7(6, the trihemitone 6(5, another major tone 15/14, the
semitone or leima 256/243, the apotome of the minor tone and
finally the comma 81(80. This complexity results from the mIxture of the
two means of calculation.
B. The secondary order consists of the tetrachords, as defined in Aristox-
enos, and similarly the pentachords and the octochords. The tetrachords
are divided into three genera:
l. Diatonic, subdivided into: first scheme, 12 + 11 + 7 = 30 seg-
ments, or (9/8)(10/9)(16/15) = 4/3, starting on f:.., H, ctc; scheme,
11 + 7 + 12 = 30 segments, or (10/9)(16/15)(9/8) = 4/3, startmg on E,
A etc-thirdscheme 7 + 12 + 11 = 30segments,or(16/15)(9/8)(10/9) =
4;3, s;arting on Z, Here we notice a developed combinatorial
that is not evident in Aristoxenos; only three of the six possible permutatIOns
of the three notes are used.
2. Chromatic, subdivided into:
14
a. soft chromatic, derived from the
diatonic tetrachords of the first scheme, 7 + 16 + 7 = 30 segments, or
(16/15)(7/6)(15/14) = 4(3, starting on f:.., H, ctc.; b. syntonon, or hard
chromatic derived from the diatonic tetrachords of the second scheme,
5 + 19 +'6 = 30 segments, or (256/243)(6/5)(135/128) = 4/3, starting on
E, A, etc. . I b'l
3. Enharmonic, derived from the diatonic by alteratIOn of t Ie mo I e
notes and subdivided into: first scheme, 12 + 12 + 6 = 30 segments, or
(9/8)(9/8)(256/243) = 4/3, starting on Z, H, r, etc.; second
12 + 6 + 12 = 30 segments, or (9/8)(256/243) (9/8) = 4/3, startmg on
f:.., H, A, etc.; third schcmc, 6 + 12 + 12 = 30 segments, or (256/243) (9/8)
(9/8) = 4/3, starting on E, A, B, etc.
PARENTHESIS
We can sec a phenomenon of absorption of the ancient cnharn:lOnic
by the diatonic. This must have taken place during th: first of
Christianity, as part of the Church fathers' struggle agamst pagamsm and
certain of its manifestations in the arts. The diatonic had always been
sidered sober, severe, and noble, unlike the other types. In fact the
genus, and especially the enharmonic, demanded a more advance? mUSical
culture as Aristoxenos and the other theoreticians had already pam ted out,
and a culture was even scarcer among the masses of the Roman period.
Consequently combinatorial speculations on one practical
usage on the other must have caused the chara:t:l:lstlcs of e.I,I-
harmonic to disappear in favor of the chromatIc, a subdlV1SIOll ofwlllch fell
away in Byzantine music, and of the syntonon diatonic. This phenomenon
of absorption is comparable to that of the scales (or modes) of the Renais_
sance by the major diatonic scale, which perpetuates the ancient syntonon
diatonic.
However, this simplification is curious and it would be interesting to
study the exact circumstances and causes. Apart from differences, or rather
variants of ancient intervals, Byzantine typology is built strictly on the
ancient. It builds up the next stage with tetrachords, using definitions which
singularly shed light on the theory of the Aristoxencan systems; this was
expounded in some detail by Ptolemy.15
THE SCALES
C. The tertiary order consists of the scales constructed with the help o:f
systems having the same ancient rules of consonance, dissonance, and asso-
nance (paraphonia). In Byzantine music the principle of iteration and
juxtaposition of the system leads very clearly to scales, a development which
is still fairly obscure in Aristoxenos and his successors, except for Ptolemy.
Aristoxenos seems to have seen the system as a category and end in itself,
and the concept of the scalc did not emerge independently from the method
which gave rise to it. In Byzantine music, on the other hand, the system was
called a method of constructing scales. It is a sort of iterative operator, which
starts from the lower category of tetrachords and their derivatives, the
pentachord and the octochord, and builds up a chain of more complex
organisms, in the same manner as chromosomes based on genes. From this
point of view, system-scale coupling reached a stage of fulfillment that had
been unknown in ancient times. The Byzantines defined the system as the
simple or multiple repetition of two, several, or all the notes of a scale.
"Scale" here means a succession of notes that is already organized, such
as the tetrachord or its derivatives. Three systems are used in Byzantine
music:
the octachord or dia pason
the pentaehord or wheel (trochos)
the tetrachord or triphony.
The system can unite elements by conjunct (synimenon) or disjunct
(diazeugmenon) juxtaposition. The disjunct juxtaposition of two tetra-
chords one tone apart form the dia pason scale spanning a perfect octave.
The cOrUunctjuxtaposition of several ofthese perfect octave dia pason leads
to the scales and modes with which we are familiar. The conjunct juxta-
position of several tetrachords (triphony) produces a scale in which the
octave is no longer a fixed sound in the tetrachord but one of its mobile
sounds. The same applies to the conjunct juxtaposition of several pentachords
(trochos).
The system can be applied to the three genera of tetrachords and to
each of their subdivisions, thus creating a very rich collection of scales.
Finally one may even mix the genera of tetrachords in the same scale (as in
the sclidia of Ptolemy), which will result in a vast variety. Thus the scale
order is the product of a combinatorial method-indeed, of a gigantic
montage (harmony)-by iterative juxtapositions of organisms that are
already strongly differentiated, the tetrachords and their derivatives. The
scale as it is defined here is a richer and more universal conception than all
the impoverished conceptions of medieval and modern times. From this
point of view, it is not the tempered scale so much as the absorption by the
diatonic tetrachord (and its corresponding scale) of all the other com binations
or montages (harmonies) of the other tetrachords that represents a vast loss
of potential. (The diatonic scale is derived from a disjunct system of two
diatonic tetrachords separated by a whole tone, and is represented by the
white keys on the piano.) It is this potential, as much sensorial as abstract,
that we are seeking here to reinstate, albeit in a modern way, as will be
seen.
The following are examples of scales in segments of Byzantine tem-
pering (or Aristoxenean, since the perfect fourth is equal to 30 segments):
Diatonic scales. Diatonic tetraehords: system by disjunct terrachords,
12,11,7; 12; 11,7,12, starting on the lower ()., 12,11,7; 12; 12,11,7,
starting on the lower H or A; system by tetrachord and pentachord, 7,12,
11; 7, 12, 12, 11, starting on the lower Z; wheel system (trochos), 11, 7, 12,
12; 11, 7, 12, 12; 11, 7, 12, 12; etc.
Chromatic scales. Soft chromatic tetrachords: wheel system starting on
H,7, 16, 7, 12; 7, 16,7, 12; 7, 16,7, 12; etc.
Enharmonic scales. Enharmonic tetrachords, second scheme: system by
disjunct tetrachords, starting on Ll, 12, 6, 12; 12; 12, 6, 12, corresponding
to the mode produced by all the white keys starting with D. The enharmonic
scales produced by the disjunct system form all the ecclesiastical scales or
modes of the West, and others, for example: chromatic tetraehord, first
scheme, by the triphonie system, starting on low H: 12, 12, 6; 12, 12, 6; 12,
12,6; 12, 12,6.
Mixed scales. Diatonic tetrachords, first scheme + soft chromatic;
disjunct system, starting on low H, 12,11,7; 12; 7,16,7. Hard chromatic
tetrachord + soft chromatic; disjunct system, starting on low If, 5, 19, 6;
12; 7, 16, 7; etc. All the montages arc not used, and one can observe the
190
Formalized Music
of the absorption of imperfect octaves by the perfect octave
y VIr ue of the basic rules of consonance TI 's . I'" ..
D TI . 11 IS a ImItmg condItIOn
. d . 1e quaternary order consists of the tropes or echoi (ichi) Th . h
IS efined by; . e ec os
the genera of (or derivatives) constituting it
the system of Juxtaposition
the attractions
the bases or fundamental notes
the dominant notes
the or cadences (katalixis)
the apIchIma .or melodies introducing the mode
the ethos, wl1lch follows ancient definitions.
We shall not concern ourselves with the details of
order. this quaternary
Thus we have succ' tl d
struct fB . mc. y expoun ed our analysis of the outside-time
ure 0 yzantme mUSIC.
THE METABOLAE
this .outside-time structure could not be satisfied w't!
mentahzed hIerarch It I 1 a compart-
notes and the' 1 ; ... was necessary to have free circulation between the
1
Ir su J IVlSlOns, between the kinds of tetrachords between the
genera, )etween the s st db'
sketch of tI . . y ems, an etween the echoi-hence the need for a
. Ie 1I1-l1me structure, which we will now look at brI'efty Th
eXIst operati . h' h . ere
and other modulations,
the h . f . lese sIgns are the phthorai and
c roal 0 notes, tetrachords, systems ( or scales) and ech .
Note metabolae ' Ol.
The metathesis; transition from a tetrach rd f
fourth) to another tetrachord of30 0 0 30 segments (perfect
segments. .
segm!:t: 10f the .intervall to the 30
. . arger mterva and VIce ve s . .
transcltlOn bon Ie distorted tetrachord to another distorted
enus JVIela 0 ae .
Phthora characteristic of the genus not cha .
Changing note names ,ngmg note names
Using the parachordi
Using the chroai.
System melabolae
Transition from 0
ne system to another using the above metabolae.
Towards a Metamusic
191
Echos metabolae using special signs, thc martyrikai phthorai or altera-
tions of the mode initialization.
Because of the complexity of the metabolae, pedal notes (isokratima)
cannot be "trusted to the ignorant." Isokratima constitutes an art in itself,
for its function is to emphasize and pick out all the in-time fluctuations of the
outside-time structure that marks the music.
First Comments
It can easily be seen that the consummation of this outside-time strue-
ture is the most complex and most refined thing that could be invented by
monody. What could not be developed in polyphony has been brought to
such luxuriant fruition that to become familiar with it requires many years
of practical studies, such as those followed by the vocalists and instrumenta-
lists of the high cultures of Asia. It seems, however, that none of the special-
ists in Byzantine music recognize the importance of this structure. It would
appear that interpreting ancient systems of notation has claimed their
attention to such an extent that they have ignored the living tradition of the
Byzantine Church and have put their names to incorrect assertions. Thus
it was only a few years ago that one ofthem
16
took the line of the Gregorian
specialists in attributing to the echoi characteristics other than those of the
oriental scales which had been taught them in the conformist schools. They
have finally discovered that the echoi contained certain characteristic mel-
odic formulae, though of a sedimentary nature. But they have not been able
or willing to go further and abandon their soft refuge among the manuscripts.
Lack of understanding of ancient music,17 of both Byzantine and Greg-
orian origin, is doubtless caused by the blindness resulting from the growth
of polyphony, a highly original invention of the barbarous and uncultivated
Occident following the schism of the churches. The passing of centuries and
the disappearance of the Byzantine state have sanctioned this neglect and
this severance. Thus the effort to feel a "harmonic" language that is much
more refined and complex than that of the syntonon diatonic and its scales
in octaves is perhaps beyond the usual ability of a Western music specialist,
even though the music of our own day may have been able to liberate him
partly from the overwhelming dominance of diatonic thinking. The only
exceptions are the specialists in the music of the Far East,18 who have always
remained in close contact with musical practice and, dealing as they were
with living music, have been able to look for a harmony other than the tonal
harmony with twelve semitones. The height of error is to be found in the
transcriptions of Byzantine melodies
19
into Western notation using the
tempered system. Thus, thousands of transcribed melodies are completely
wrong! But the real criticism one must level at the Byzantinists is that in
remaining alooffrom the great musical tradition of the eastern church, they
have ignored the existence of this abstract and sensual architecture, both
complex and remarkably interlocking (harmonious), this developed remnan t
and genuine achievement of the Hellenic tradition. In this way they have
retarded the progress of musicological rescarch in the areas of:
antiquity
plainchant
folk music of European lands, notably in the East
20
musical cultures of the civilizations of other continents
better understanding of the musical evolution of Western Europe from
the middle ages up to the modern period
the syntactical prospects for tomorrow's music, its enrichment, and its
survival.
Second Comments
I am motivated to present this architecture, which is linked to antiquity
and doubtless to other cultures, because it is an elegant and lively witness
to what I have tried to define as an outside-time category, algebra, or struc-
ture of music, as opposed to its other two categories, in-time and temporal.
It has often been said (by Stravinsky, Messiaen, and others) that in music
time is everything. Those who express this view forget the basic structures on
which personal languages, such as "pre- or post-Webernian" serial music,
rest, however simplified they may be. In order to understand the universal
past and present, as well as prepare the future, it is necessary to distinguish
structures, architectures, and sound organisms from their temporal manifes-
tations. It is therefore necessary to take "snapshots," to make a series of
veritable tomographies over time, to compare them and bring to light their
relations and architectures, and vice versa. In addition, thanks to the
metrical nature of time, one can furnish it too with an outside-time structure,
leaving its true, unadorned nature, that of immediate reality, of instan-
taneous becoming, in the final analysis, to the temporal category alone.
In this way, time could be considered as a blank blackboard, on which
symbols and relationships, architectures and abstract organisms arc in-
scribed. The clash between organisms and architectures and instantan-
eous immediate reality gives rise to the primordial quality of the living
conSCIousness.
The architectures of Greece and Byzantium are concerned with the
pitches (the dominant character of the simple sound) of sound entities.
Here rhythms are also subjected to an organization, but a much simpler
one. Thercfore we shall not refer to it. Certainly these ancient and Byzantine
models cannot serve as examples to be imitated or copied, hut rather to
exhibit a fundamental outside-time architecture which has been thwarted
by the temporal architectures of modern (post-medieval) polyphonic music.
These systems, induding those of serial music, are still a somewhat confused
magma of temporal and outside-time structures, for no one has yet thought
of unravelling them. However we cannot do this here.
Progressive Degradation of Outside-Time Structures
The tonal organization that has resulted from venturing into polyphony
and neglecting the ancients has leaned strongly, by virture of its very nature,
on the temporal category, and defincd the hierarchies of its harmonic
functions as the in-time category. Outside-time is appreciably poorer, its
"harmonics" being reduced to a single octave scale (C major on the two
bases C and A), corresponding to the syntonon diatonic of the Pythagorean
tradition or to the Byzantine enharmonic scales based on two disjunct
tetra chords of the first scheme (for C) and on two disjunct tetrachords of the
second and third scheme (for A). Two metabolae have been preserved: that
of transposition (shifting of the scale) and that of modulation, which consists
of transferring the base onto steps of the same scale. Another loss occurred
with the adoption of the crude tempering of the semitone, the twelfth root
of two. The consonances have been enriched by the interval of the third,
which, until Debussy, had nearly ousted the traditional perfect fourths and
fifths. The final stage of the evolution, atonalism, prepared by the theory
and music of the romantics at the end of the nineteenth and the beginning
of the twentieth centuries, practically abandoned all outside-time structure.
This was endorsed by the dogmatic suppression of the Viennese school, who
accepted only the ultimate total time ordering of the tempered chromatic
scale. Of the four forms of the series, only the inversion of the intervals is
related to an outside-time structure. Naturally the loss was felt, consciously
or not, and symmetric relations between intervals were grafted onto the
chromatic total in the choice of the notes of the series, but these always
remained in the in-time category. Since then the situation has barely
changed in the music of the post-Webernians. This degradation of the
outside-time structures of music since late medieval times is perhaps the
most characteristic fact about the evolution of Western European music, and
it has led to an unparalleled excrescence of temporal and in-time structures.
In this lies its originality and its contribution to the universal culture. But
herein also lies its impoverishment, its loss of vitality, and also an apparent
risk of reaching an impasse. For as it has thus far developed, European music
is ill-suited to providing the world with a field of expression on a planetary
scale, as a universality, and risks isolating and severing itself from historical
necessities. We must open our eyes and try to build bridges towards other
cultures, as well as towards the immediate future of musical thought, before
we perish suffocating from electronic technology, either at the instrumental
level or at the level of composition by computers.
Reintroduction of the Outside-Time Structure by Stochastics
By the introduction of the calculation of probability (stochastic music)
the present small horizon of outside-time structures and asymmetries was
completely explored and enclosed. But by the very f a ~ t of its introduction,
stochastics gave an impetus to musical thought that carried it over this
enclosure towards the clouds of sound events and towards the plasticity of
large numbers articulated statistically. There was no longer any distinction
between the vertical and the horizontal, and the indeterminism of in-time
structures made a dignified entry into the musical edifice. And, to crown the
Herakleitean dialectic, indeterminism, by means of particular stochastic
functions, took on color and structure, giving rise to generous possibilities
of organization. It was able to include in its scope determinism and, still
somewhat vaguely, the outside-time structures of the past. The categories
outside-time, in-time, and temporal, unequally amalgamated in the history
of music, have suddenly taken on all their fundamental significance and for
the first time can build a coherent and universal synthesis in the past,
present, and future. This is, I insist, not only a possibility, but even a direc-
tion having priority. But as yet we have not managed to proceed beyond
this stage. To do so we must add to our arsenal sharper tools, trenchant
axiomatics and formalization.
SIEVE THEORY
It is necessary to give an axiomatization for the totally ordered struc-
ture (additive group structure = additive Aristoxenean structure) of the
tempered chromatic scale.
21
The axiomatics of the tempered chromatic
scale is based on Peano's axiomatics of numbers:
Preliminary terms. 0 = the stop at the origin; n = a stop; n' = a stop
resul ting from elementary displacement of n; D = the set of values of the
particular sound characteristic (pitch, density, intensity, instant, speed,
disorder ... ). The values are identical with the stops of the displacements.
First propositions (axioms).
L Stop 0 is an element of D.
2. Ifstop n is an element of D then the new stop n' is an element of D.
3. If stops nand m are elements of D then the new stops n' and m' are
identical if, and only if, stops nand m are identical.
4. If stop n is an element of D, it wj]] be different from stop 0 at the
origin.
5. If elements belonging to D have a special property P, such that
stop 0 also has it, and if, for every clement n of D having this property the
element n' has it also, all the clements of D will have the property P.
We have just defined axiomatically a tempered chromatic scale not
only of pitch, but also of all the sound properties or characteristics referred
to above in D (density, intensity ... ). Moreover, this abstract scale, as
Bertrand Russell has rightly observed, a propos the axiomatics of numbers
of Pea no, has no unitary displacement that is either predetermined or related
to an absolute size. Thus it may be constructed with tempered semitones,
with Aristoxenean segments (twelfth-tones), with the commas of Didymos
(81/80), with quarter-tones, with whole tones, thirds, fourths, fifths, octaves,
etc. or with any other unit that is not a factor of a perfect octave.
Now let us define another equivalent scale based on this one but having
a unitary displacement which i ~ a multiple of the first. It can be expressed
by the concept of congruence modulo m.
Difinition. Two integers x and n are said to be congruent modulo m when
m is a factor of x - n. It may be expressed as follows: x == n (mod m). Thus,
two integers are congruent modulo m when and only when they differ by
an exact (positive or negative) multiple ofm; e.g., 4 == 19 (mod 5), 3 == 13
(mod 8), 14 == (mod 7).
Consequently, every integer is congruent modulo m with one and with
only one value of n:
n = (0, 1, 2, ... , m - 2, m - 1).
Of each of these numbers it is said that it forms a residual class modulo
m; they are, in fact, the smallest non-negative residues modulo m. x ==
n(mod m) is thus equivalent to x = n + km, where k is an integer.
k E Z = {O, 1, 2, 3, ... }.
For a given n and for any k E Z, the numbers x will belong by definition
to the residual class n modulo m. This class can be denoted m
n

In order to grasp these ideas in terms of music, let us take the tempered
semitone of our present-day scale as the unit of displacement. To this we
shall again apply the above axiomatics, with say a value of 4 semi tones
(major third) as the elementary displacement.
22
We shall define a new
chromatic scale. Hthe stop at the origin of the first scale is a D ~ , the second
scale will give us all the multiples of 4 semi tones, in other words a "scale"
of major thirds: D ~ , G, B, D ' ~ , G', B'; these are the notes of the first scale
whose order numbers are congruent with 0 modulo 4. They all belong to the
residual class 0 modulo 4. The residual classes I, 2, and 3 modulo 4 will use
up all the notes of this chromatic total. These classes may be represented in
the following manner:
residual class 0 modulo 4: 40
residual class 1 modulo 4:4
1
residual class 2 modulo 4:4
2
residual class 3 modulo 4: 43
resid ual class 4 modulo 4: 4
0
, etc.
Since we are dealing with a sieving of the basic scale (elementary dis-
placement by one semitone), each residual class forms a sieve allowing
certain elements of the chromatic continuity to pass through. By extension
the chromatic total will be represented as sieve 1
0
, The scale offourths will
be given by sieve 5n> in which n = 0, I, 2, 3, 4. Every change of the index n
will entail a transposition of this gamut. Thus the Debussian whole-tone
scale, 2n with n = 0, 1, has two transpositions:
20 -7 C, D, E, F ~ , G#, Alt, C .. ..
21 -+ C ~ , D ~ , F, G, A, E, C ~ ... ,
Starting from these elementary sieves we can build more complex
scales-all the scales we can imagine-with the help of the three operations
of the Logic of Classes: union (disjunction) expressed as V, intersection
(conjunction) expressed as II, and complementa60n (negation) expressed
as a bar inscribed over the modulo of the sieve. Thus
20 V 21 = chromatic total (also expressible as 1
0
)
20 /I 21 = no notes, or empty sieve, expressed as 0
20 = 21 and 21 = 2
0
,
The major scale can be written as follows:
By definition, this notation does not distinguish between all the modes
on the white keys of the piano, for what we are defining here is the scale;
modes are the architectures founded on these scales. Thus the white-key
mode D, starting on D, will have the same notation as the C mode. But in
order to distinguish the modes it would be possible to introduce non-
commutativity in the logical expressions. On the other hand each of the
12 transpositions of this scale will be a combination of the cyclic permuta-
tions of the indices of sieves modulo 3 and 4. Thus the major scale transposed
a semitone higher (shift to the right) will be written
and in general
(3
n
+
2
/I 4
n
) V (3
n
+
1
/\ 4
n
+
1
) V (3
n
+
2
/I 4
n
+2 ) V (3n /I 4n + 3 ),
where n can assume any value from 0 to 11, but reduced after the addition
of the constan t index of each of the sieves (moduli), modulo the correspond-
ing sieve. The scale of D transposed onto C is written
(3n /\ 4
n
) V (3
n
+
1
/I 4
n
+
1
) V (3
n
/I 4
n
+ 2) V (3n + 2 /I 4n +3 )
Musicology
Now let us change the basic unit (elementary displacement ELD) of
the sieves and use the quarter-tone. The major scale will be written
(8
n
/\ 3
n
+
1
) V (8
n
+
2
/I 3
n
+
2
) V (8
n
+
4
/\ 3n + 1 ) V (8n +6 /\ 3n),
with n = 0, I, 2, ... , 23 (modulo 3 or 8). The same scale with still finer
sieving (one octave = 72 Aristoxenean segments) will be written
(8
n
/I (9
n
V 9
n
+
6
)) V (8
n
+
2
/I (9
n
+
3
V 9n +6 )) V (8n +4 /I 9n +3 )
V (8
n
+
6
/\ (9
n
V 9
n
+3 )),
with n = 0, 1,2, ... , 71 (modulo 8 or 9).
One of the mixed Byzantine scales, a disjunct system consisting of a
chromatic tetrachord and a diatonic tetrachord, second scheme, separated
by a major tone, is notated in Aristoxenean segments as 5, 19, 6; 12; II, 7,
12, and will be transcribed logically as
(8
n
/I (9
n
V 9
n
+
6
)) V (9
n
+
6
/I (8n +2 V 8n +4))
V (8
n
+
5
/\ (9
n
+
5
V 9
n
+s)) V (8n +6 V 9n +3 ),
with n = 0, 1, 2, ... , 71 (modulo 8 or 9).
The Raga Bhairavi of the Andara-Sampurna type (pentatonic as-
cending, heptatonic descending),23 expressed in terms of an Aristoxenean
basic sieve (comprising an octave, periodicity 72), will be written as:
Pentatonic scale:
(8n 1\ (9n V 9n+3 ) V (8
n
+
2
A (9
n
V 9
nt6
V (8nt6 A 9n +3 )
Heptatonic scale:
(8n 1\ (9n V 9n +3 V (8
n
+
2
1\ (9
n
V 9
n
+
6
V (8
n
+4 1\ (9
n
+4 V 9n +6
V (8n+6 A (9
n
+
3
V 9
n
+6 ))
with n = 0, 1,2, ... , 71 (modulo 8 or 9).
These two scales expressed in terms of a sieve having as its elementary
displacement, ELD, the comma of Didymos, ELD = 81(BO (Bl(SO to the
power 55.B = 2), thus having an octave periodicity of 56, will be written as:
Pentatonic scale:
(7n A (8n V 8n +6) V (7n+2 A (Bnt5 V 8
n
+
7
) V (7
n
+
5
A 8n +1 )
Heptatonic scale:
(7n A (8n V 8n +6 V (7n+2 A (8
n
+
5
V 8
n
+
7
V (7n+3 A 8n +3 )
V (7n+4 A (8n +4 V 8n +6 ) V (7n +5 A 8n +1 )
for n = 0, 1,2, ... , 55 (modulo 7 or 8).
We have just seen how the sieve theory allows us to express-any scale
in terms oflogical (hence mechanizable) functions, and thus unify our study
of the structures of superior range with that of the total order. It can be
useful in entirely new constructions. To this end let us imagine complex,
non-octave-forming sieves.
24
Let us take as our sieve unit a tempered
quarter-tone. An octave contains 24 quarter-tones. Thus we have to con-
struct a compound sieve with a periodicity other than 24 or a multiple of
24, thus a periodicity non-congruent with k24 modulo 24 (for k = 0, 1,
2, ... ). An example would be any logical function of the sieve of moduli
11 and 7 (periodicity II x 7 = 77 1= k24), (lIn V 1I
n
+
1
) A 7
n
+
6
This
establishes an asymmetric distribution of the steps of the chromatic quarter-
tone scaJe. One can even use a compound sieve which throws periodicity
outside the limits of the audible area; for example, any logical function of
modules 17 and 18 (f[17, 18J), for 17 x 18 = 306> (11 x 24).
Suprastructures
One can apply a stricter structure to a compound sieve or simply leave
the choice of elements to a stochastic function. We shall obtain a statistical
coloration of the chromatic total which has a higher level of complexity.
Using metabolae. We know that at every cyclic combination of the sieve
indices (transpositions) and at every change in the module or moduli of the
sieve (modulation) we obtain a metabola. As examples of metabolic trans-
formations let us take the smallest residues that are prime to a positive
number r. They will form an Abelian (commutative) group when the
composition law for these residues is defined as multiplication with reduc-
tion to the least positive residue with regard to r. For a numerical example
let r = 18; the residues 1,5,7, II, 13,17 arc primes to it, and their products
after reduction modulo 18 will remain within this group (closure). The
finite commutative group they form can be exemplified by the following
fragment:
5 x 7 = 35; 35 - IB = 17;
11 x II = 121; 121 - (6 x IS) = 13; etc.
Modules 1, 7, 13 form a cyclic sub-group of order 3. The following is a
logical expression of the two sieves having modules 5 and 13:
L(5, 13) = (13
n
+
4
V 13
n
+
5
V l3
n
+
7
V 13
n
+
9
)
A 5
n
+1 V (5
n
+
2
V 5n +4) A 13n +9 V 13n +6
One can imagine a transformation of modules in pairs, starting from the
Abelian group defined above. Thus the cinematic diagram (in-time) will be
L(5, 13) 17) 11) 1) 5) ... --->-L(5, 13)
so as to return to the initial term (closure).25
This sieve theory can be put into many kinds of architecture, so as to
create included or successively intersecting classes, thus stages of increasing
complexity; in other words, orientations towards increased determinisms
in selection, and in topological textures of neighborhood.
Subsequently we can put into in-time practice this veritable histology
of outside-time music by means of temporal functions, for instance by giving
functions of change-of indices, moduli, or unitary displacement-in other
words, encased logical functions parametric with time.
Sieve theory is very general and consequently is applicable to any other
sound characteristics that may be provided with a totally ordered structure,
such as intensity, instants, density, degrees of order, speed, etc. I have al-
ready said this elsewhere, as in the axiomatics of sieves. But this method can
be applied equally to visual scales and to the optical arts of the future.
Moreover, in the immediate future we shall witness
e
:OCl '1'('1>

!;
'" .'
this theory and its widespread use with the help of computers, for it is
entirely mechanizable. Then, in a subsequent stage, there will be a study
of partially ordered structures, such as are to be found in the classification
of timbres, for example, by means of lattice or graph techniques.
Conclusion
I believe that music today could surpass itself by research into the out-
side-time category, which has been atrophied and dominated by the
temporal category. Moreover this method can unify the expression of
structures of all Asian, African, and European music. It has a
conSiderable advantage: its mechanization-hence tests and models of
all sorts can be fed into computers, which will effect great progress in the
musical sciences.
In fact, what we are witnessing is an industrialization of music which
has already started, whether we like it or not. It already floods our ears in
many public places, shops, radio, TV, and airlines, the world over. It
permits a consumption of music on a fantastic scale, never before approached.
But this music is of the lowest kind, made from a collection of outdated
the dregs of the musical mind. Now it is not a matter of stopping
thiS which, after all, increases participation in music, even if only
passively. It is rather a question of effecting a qualitative conversion of this
by exercising a radical but constructive critique of our ways of think-
mg and of making music. Only in this way, as I have tried to show in the
pr:sent. study, will the musician succeed in dominating and transforming
th.ls pOlson that is discharged into our ears, and only if he sets about it
WIthout further ado. But one must also envisage and in the same way a
d' "
ra Ical conversion of musical education, from primary studies onwards,
the entire world (all national councils for music take note).
systems and the logic of classes are already taught in certain
countries, so why not their application to a new musical theory, such as is
sketched out here?
Chapter VIII
Towards a Philosophy of Music
PRELIMINARIES
We are going to attempt briefly: 1. an "unveiling of the historical
tradition" of music,! and 2. to construct a music.
"Reasoning" about phenomena and their explanation was the greatest
step accomplished by man in the course of his liberation and growth. This
is why the Ionian pioneers-Thales, Anaximander, Anaximenes-must be
.considered as the starting point of our truest culture, that of "reason."
When I say" reason," it is not in the sense of a logical sequence of arguments,
syllogisms, or logico-technical mechanisms, but that very extraordinary
quality of feeling an uneasiness, a curiosity, then of applying the question,
EAEYX0<;' It is, in fact, impossible to imagine this advance, which, in Ionia,
created cosmology from nothing, in spite of religions and powerful mystiques,
which were early forms of "reasoning." For example, Orphism, which so
influenced Pythagorism, taught that the human soul is a fallen god, that
only ek-stasis, the departure from self, can reveal its true nature, and that
with the aid of purifications (KcxBapflo{) and sacraments (oPYta) it can regain
its lost position and escape the Wheel afBirth (Tpoxor; YJ.IUw<;, bhavachakra)
that is to say, the fate of reincarnations as an animal or vegetable. I am citing
this mystique because it seems to be a very old and widespread form of
thought, which existed independently about the same time in the Hinduism
of India.
2
Above all, we must note that the opening taken by the Ionians has
finally surpassed all mystiques and all religions, including Christianity.
English translation of Chapter VIII by John and Amber Challifour.
201
202
Formalized Music
Never has the spirit of this philosophy been as universal as today: The
U.S., China, U.S.S.R., and Europe, the present principal protagonists,
restate it with a homogeneity and a uniformity that I would even dare to
qualify as disturbing.
Having been established, the question (E''\EYX0s") embodied a Wheel of
Birth sui generis, and the various pre-Socratic schools flourished by con-
ditioning all further developmcn t of philosophy until our time. Two are in
my opinion the high points of this period: the Pythagorean concept of
numbers and the Parmenidean dialectics-both unique expressions of the
same preoccupation.
As it went through its phases of adaptation, up to the fourth century
B.C., the Pythagorean concept of numbers affirmed that things are numbers,
or that all things are furnished with numbers, or that things arc similar to
numbers. This thesis developed (and this in particular interests the musician)
from the study of musical intervals in order to obtain the orphic catharsis,
for according to Aristoxenos, the Pythagoreans used music to cleanse. the
soul as they used medicine to cleanse the body. This method is found in
other orgia, like that of Korybantes, as confirmed by Plato in the Laws. In
every way, Pythagorism has permeated all occidental thought, first of all,
Greek, then Byzantine, which transmitted it to Western Europe and to the
Arabs.
All musical theorists, from Aristoxenos to Hucbald, Zarlino, and
Rameau, have returned to the same theses colored by expressions of the
moment. But the most incredible is that all intellectual activity, including
the arts, is actually immersed in the world of numbers (I am omitting the
few backward-looking or obscurantist movements). We are not far from the
day when genetics, thanks to the geometric and combinatorial structure of
DNA, will be able to metamorphise the Wheel of Birth at will, as we wish
it, and as preconceived by Pythagoras. It will not be the ek-stasis (Orphic,
Hindu, or Taoist) that will have arrived at one of the supreme goals of all
time, that of controlling the quality of reincarnations (hereditary rebirths
7raALyyeveala) but the very force of the" theory," of the question, which is
the essence of human action, and whose most striking expression is Pythag-
orism. We are all Pythagoreans.
3
On the other hand, Parmenides was able to go to the heart of the q ues-
tion of change by denying it, in contrast to Herakleitos. He discovered the
principle of the excluded middle and logical tautology, and this created
such a dazzlement that he used them as a means of cutting out, in the
evanescent change of senses, the notion of Being, of that which is, one,
motionless, filling the universe, without birth and indestructible; the
Towards a Philosophy of Music 203
not-Being, not existing, circumscribed, and spherical (which Melissos had
not understood).
[F]or it will be forever impossible to prove that things that are not
are; but restrain your thought from this route of inquiry .... Only one
way remains for us to speak of, namely, that it is; on this route there
are many signs indicating that it is uncreated and indestructible, for
it is complete, undisturbed, and without end; it never was, nor will
it be, for now it is all at once complete, one, continuous; for what
kind of birth are you seeking for it? How and from where could it
grow? I will neither let you say nor think that it came from what is
not for it is unutterable and unthinkable that a thing is not.
And what need would have led it to be created sooner or later if
it came from nothing? Therefore it must be, absolutely, or not at
all.
-Fragments 7 and 8 of Poem, by Parmenides
4
Besides the abrupt and compact style of the thought, the method of the
question is absolute. It leads to denial of the sensible world, which is only
made of contradictory appearances that "two-faced" mortals accept as
valid without turning a hair, and to stating that the only truth is the notion
of reality itself. But this notion, substantiated with the help of abstract
logical rules, needs no other concept than that of its opposite, the not-
Being, the nothing that is immediately rendered impossible to formulate and
to conceive.
This concision and this axiomatics, which surpasses the deities and
cosmogonies fundamental to the first elements,
5
had a tremendous influence
on Parmenides' contemporaries. This was the first absolute and complete
materialism. Immediate repercussions were, in the main, the continuity of
Anaxagoras and the atomic discontinuity ofLeukippos. Thus, all intellectual
action until our time has been profoundly imbued with this strict axiomatics.
The principle of the conservation of energy in physics is remarkable. En-
ergy is that which fills the universe in electromagnetic, kinetic, or material
form by virtue of the equivalence matter-energy. It has become that which
is "par excellence." Conservation implies that it docs not vary by a single
photon in the entire universe and that it has been thus throughout eternity.
On the other hand, by the same reasoning, the logical truth is tautological:
All that which is affirmed is a truth to which no alternative is conceivable
(Wittgenstein). Modern knowledge accepts the void, but is it truly a non-
Being? Or simply the designation of an unc1arified complement?
After the failures of the nineteenth century, scientific thought became
rather skeptical and pragmatic. It is this fact that has allowed it to adapt
and develop to the utmost. "All happens as if ... " implies this doubt, which
is positive and optimistic. We place a provisional confidence in new theories,
but we abandon them readily for more efficacious ones provided that the
procedures of action have a suitable explanation which agrees with the
whole. In fact, this attitude represents a retreat, a sort of fatalism. This is
why today's Pythagorism is relative (exactly like the Parmenidean axio-
matics) in all areas, including the arts.
Throughout the centuries, the arts have undergone transformations
that paralleled two essential creations of human thought: the hierarchical
principle and the principle of numbers. In fact, these principles have domi-
nated music, particularly since the Renaissance, down to present-day pro-
cedures of composition. In school we emphasize unity and recommend the
unity of themes and of their deVelopment; but the serial system imposes
another hierarchy, with its own tautological unity embodied in the tone row
and in the principle of perpetual variation, which is founded on this
tautology ... -in short, all these axiomatic principles that mark our lives
agree perfectly with the inquiry of Being introduced twenty-five centuries
ago by Parmenides.
It is not my intention to show that everything has already been dis-
covered and that we are only plagiarists. This would be obvious nonsense.
There is never repetition, but a sort of tautological identity throughout the
vicissitudes of Being that might have mounted the Wheel of Birth. It would
seem that some areas are less mutable than others, and that some regions of
the world change very slowly indeed.
The Poem of Par men ides implicitly admits that necessity, need, causality,
and justice identify with logic; since Being is born from this logic, pure
chance is as impossible as not-Being. This is particularly clear in the phrase,
" And what need would have led it to be born sooner or later, if it came from
nothing?" This contradiction has dominated thought throughout the
millennia. Here we approach another aspect of the dialectics, perhaps the
most important in the practical plan of action-determinism. Iflogic indeed
implies the absence of chance, then one can know all and even construct
everything with logic. The problem of choice, of decision, and of the future
is resolved. '
We know, moreover, that ifan element of chance enters a deterministic
construction all is undone. This is why religions and philosophies every-
where have always driven chance back to the limits of the universe. And
what they utilized of chance in divination practices was absolutely not con-
sidered as such but as a mysterious web of signs, sent by the divinities (who
were often contradictory but who knew well what they wanted), and which
could be read by elect soothsayers. This web of signs can take many forms-
the Chinese system ofI-Ching, auguries predicting the future from the flight
of birds and the entrails of sacrificed animals, even tclling fortunes from tea
leaves. This inability to admit pure chance has even persisted in modern
mathematical probability theory, which has succeeded in incorporating it
into some deterministic logical laws, so that pure chance and pure determinism
are only two facets of one entity, as I shall soon demonstrate with an example.
To my knowledge, there is only one "unveiling" of pure chance in all
of the history of thought, and it was Epicurus who dared to do it. Epicurus
struggled against the deterministic networks of the atomists, Platonists,
Aristoteleans, and Stoics, who finally arrived at the negation offree will and
believed that man is subject to nature's will. For if all is logically ordered in
the universe as well as in our bodies, which are products of it, then our will
is subject to this logic and our freedom is nil. The Stoics admitted, for ex-
ample, that no matter how small, every action on earth had a repercussion
on the most distant star in the universe; today we would say that the network
of connections is compact, sensitive, and without loss of information.
This period is unjustly slighted, for it was in this time that all kinds of
sophisms were debated, beginning with the logical calculus of the Megarians,
and it was the time in which the Stoics created the logic called modal, which
was distinct from the Aristotelian logic of classes. Moreover, Stoicism, by its
moral thesis, its fullness, and its scope, is without doubt basic to the forma-
tion of Christianity, to which it has yielded its place, thanks to the substitu-
tion of punishment in the person of Christ and to the myth of eternal reward
at the Last Judgment-regal solace for mortals.
In order to give an axiomatic and cosmogonical foundation to the
proposition of man's free will, Epicurus started with the atomic hypothesis
and admitted that" in the straight line fall that transports the atoms across
the void, ... at an undetermined moment the atoms deviate ever so little
from the vertical ... but the deviation is so slight, the least possible, that we
could not conceive of even seemingly oblique movements." 6 This is the
theory of ekklisis (Lat. clinamen) set forth by Lucretius. A senseless principle
is introduced into the grand deterministic atomic structure. Epicurus thus
based the structure ofthc universe on determinism (the inexorable and paral-
lel fall of atome) and, at the same time, on indeterminism (ekklisis). It is
striking to compare his thcory with the kinetic theory of gases first proposed
by Daniel Bernoulli. It is founded on the corpuscular nature of matter
and, at the same time, on determinism and indeterminism. No one but
Epicurus had ever thought of utilizing chance as a principle or as a type of
behavior.
It was not until 1654 that a doctrine on the use and understanding of
chance appeared. Pascal, and especially Fermat, formulated it by studying
"games of chance" -dice, cards, etc. Fermat stated the two primary rules
of probabilities using multiplication and addition. In 1713 Ars Conjectandi
by Jacques Bernoulli was published.
7
In this fundamental work Bernoulli
enunciated a universal law, that of Large Numbers. Here it is as stated by
E. Borel: "Let p be the probability of the favorable outcome and q the
probability of the unfavorable outcome, and let e be a small positive num-
ber. The probability that the difference between the observed ratio of
favorable events to unfavorable events and the theoretical ratio P/g is larger
in absolute value than e will approach zero when the number of trials n
becomes infinitely large." B Consider the example of the game of heads and
tails. If the coin is perfectly symmetric, that is to say, absolutely true, we
know that the probability p of heads (favorable outcome) and the probability
q of tails (unfavorable outcome) are each equal to 1/2, and the ratio p/q to 1.
Ifwe toss the coin n times, we will get heads P times and tails Q times, and
the ratio P/Q will generally be different from 1. The Law of Large Numbers
states that the more we play, that is to say the larger the number n becomes,
the closer the ratio P/Q will approach 1.
Thus, Epicurus, who admits the necessity of birth at an undetermined moment,
in exact contradiction to all thought, even modern, remains an isolated case;*
for the aleatory, and truly stochastic event, is the result of an accepted
ignorance, as H. Poincare has perfectly defined it. If probability theory ad-
mits an uncertainty about the outcome of each toss, it encompasses this
uncertainty in two ways. The first is hypothetical: ignorance of the tra-
jectory produces the uncertainty; the other is deterministic: the Law of
Large Numbers removes the uncertainty with the help oftime (or of space).
However, by examining the coin tossing closely, we will sec how the sym-
metry is strictly bound to the unpredictability. If the coin is perfectly
symmetrical, that is, perfectly homogeneous and with its mass uniformly
distributed, then the uncertainty9 at each toss will be a maximum and the
probability for each side will be 1/2. Ifwe now alter the coin by redistribu-
ting the matter unsymmetrically, or by replacing a little aluminum with
platinum, which has a specific weight eight times that of aluminum, the
coin will tend to land with the heavier side down. The uncertainty will
decrease and the probabilities for the two faces will be unequal. When the
substitution of material is pushed to the limit, for example, if the aluminum
is replaced with a slip of paper and the other side is entirely of platinum,
then the uncertainty will approach zero, that is, towards the certainty that
* Except perhaps for Heisenberg.
the coin will land with the lighter side up. Here we have shown the inverse
relation between uncertainty and symmetry. This remark seems to be a
tautology, but it is nothing more than the mathematical definition ofprob-
ability: probability is the ratio of the number of favorable outcomes to the
number of possible outcomes when all outcomes are regarded as equally
likely. Today, the axiomatic definition of probability does not remove this
difficulty, it circumvents it.
MUSICAL STRUCTURES EX NIHILO
Thus we are, at this point in the exposition, still immersed in the lines
offoree introduced twenty-five centuries ago and which continue to regulate
the basis of human activity with the greatest efficacy, or so it seems. It is the
source or those problems about which we, in the darkness of our ignorance,
concern ourselves: determinism or chance,10 unity of style or eclecticism,
calculated or not, intuition or constructivism, a priori or not, a metaphysics
of music or music simply as a means of entertainment.
Actually, these are the questions that we should ask ourselves: 1. What
consequence does the awareness of the Pythagorean-Parmenidean field have
for musical composition? 2. In what ways? To which the answers are:
1. Reflection on that which is leads us directly to the reconstruction, as much
as possible ex nihilo, of the ideas basic to musical composition, and above all
to the rejection of every idea that docs not undergo the inquiry (if;\.eyXo"
8t'TJULS). 2. This reconstruction will be prompted by modern axiomatic
methods.
Starting from certain premises we should be able to construct the
most general musical edifice in which the utterances of Bach, Beethoven,
or Schonberg, for example, would be unique realizations of a gigantic
virtuality, rendered possible by this axiomatic removal and reconstruc-
tion.
It is necessary to divide musical construction into two parts (see
Chapters VI and VII): l. that which pertains to time, a mapping of entities
or structures onto the ordered structure of time; and 2. that which is inde-
pendent of temporal becomingness. There are, therefore, two categories:
in-time and outside-time. Included in the category outside-time are the dura-
tions and constructions (relations and operations) that refer to elements
(points, distances, functions) that belong to and that can be expressed
on the time axis. The temporal is then reserved to the instantaneous
creation.
In Chapter VII I made a survey of the structure of monophonic music,
with its rich outside-time combinatory capability, based on the original
texts of Aristoxenos of Tarentum and the manuals of actual Byzantine mu-
sic. This structure illustrates in a remarkable way that which I understand
by the category outside-time.
Polyphony has driven this category back into the subconscious of
musicians of the European occident, but has not completely removed it;
that would have been impossible. For about three centuries after Monte-
verdi, in-time architectures, expressed chiefly by the tonal (or modal)
functions, dominated everywhere in central and occidental Europe. How-
ever, it is in France that the rebirth of outside-time preoccupations occurred,
with Debussy and his invention of the whole-tone scale. Contact with three
of the more conservative traditions of the Orientals was the cause of it: the
plainchant, which had vanished, but which had been rediscovered by the
abbots at Solesmes; one of the Byzantine traditions, experienced through
Moussorgsky; and the Far East.
This rebirth continues magnificently through Messiaen, with his
"modes of limited transpositions" and" non-retrogradable rhythms," but
it never imposes itself as a general necessity and never goes beyond the
framework of the scales. However Messiaen himself abandoned this vein,
yielding to the pressure of serial music.
In order to put things in their proper historical perspective, it is
necessary to prevail upon more powerful tools such as mathematics and
logic and go to the bottom of things, to thc structure of musical thought and
composition. This is what I have tried to do in Chapters VI and VII
and what I am going to develop in the analysis of Nomos alpha.
Here, however, I wish to emphasize the fact that it was Debussy and
Messiaen
ll
in France who reintroduced the category outside-time in the
face of the general evolution that resulted in its own atrophy, to the advan-
tage of structures in-time.
12
In effect, atonality does away with scales and
accepts the outside-time neutrality of the half-tone scale.
13
(This situation,
furthermore, has scarcely changed for fifty years.) The introduction of
in-time order by Schonberg made up for this impoverishment. Later, with
the stochastic processes that I introduced into musical composition, the
hypertrophy of the category in-time became overwhelming and arrived at
a dead end. It is in this cuI-de-sac that music, abusively called aleatory,
improvised, or graphic, is still stirring today.
Questions of choice in the category outside-time are disregarded by
musicians as though they were unable to hear, and especially unable to
think. In fact, they drift along unconscious, carried away by the agitations
of superficial musical fashions which they undergo heedlessly. In depth,
however, the outside-time structures do exist and it is the privilege of man
not only to sustain them, but to construct them and to go beyond them.
Sustain them? Certainly; there are basic evidences of this order which
will permit us to inscribe our names in the Pythagorean-Parmenidean field
and to lay the platform from which our ideas will build bridges of under-
standing and insight into the past (we are after all products of millions of
years of the past), into the future (we are equally products of the future),
and into other sonic civilizations, so badly explained by the present-day
musicologies, for want of the original tools that we so graciously set up for
them.
Two axiomatics will open new doors, as we shall see in the analysis of
Nomos alpha. We shall start from a naive position concerning the perception
of sounds, naive in Europe as well as in Africa, Asia, or America. The
inhabitants of all these countries learned tens or hundreds of thousands of
years ago to distinguish (if the sounds were neither too long nor too short)
such characteristics as pitch, instants, loudness, roughness, rate of change,
color, timbre. They are even able to speak of the first three characteristics
in terms of in tervals.
The first axiomatics leads us to the construction of all possible scales.
We will speak of pitch since it is more familiar, but the following arguments
will relate to all characteristics which are of the same nature (instants,
loudness, roughness, density, degree of disorder, rate of change).
We will start from the obvious assumption that within certain limits
men are able to recognize whether two modifications or displacements of
pitch are identical. For example, going from C to D is the same as going
from F to G. We will call this modification elementary displacement, ELD.
(It can be a comma, a halftone, an octave, etc.) It permits us to define any
Equally Tempered Chromatic Gamut as an ETCHG sieve.14 By modifying the
displacement step ELD, we engender a new ETCHG sieve with the same
axiomatics. With this material we can go no farther. Here we introduce the
three logical operations (Aristotelean logic as seen by Boole) of conjunction
("and," intersection, notated 1\), ("or," union, notated V), and
negation ("no," complement, notated -), and use them to create classes of
pitch (various ETCHG sieves).
The following is the logical expression with the conventions as indicated
in Chapter VII:
The major scale (ELD = :!- tone):
(8
n
1\ 3
n
+
1
) V (8
n
+
2
1\ 3n + 2 ) V (8
n
+4 1\ 3
n
+ 1 ) V (8
n
+ 6 1\ 3
n
)
where n = 0, 1,2, ... , 23, modulo 3 or 8.
(It is possible to modify the step ELD by a "rational metabola." Thus the
logical function of the major scale with an ELD equal to a quarter-tone can
be based on an ELD = 1/3 tone or on any other portion of a tone. These
two sieves, in turn, could be combined with the three logical operations to
provide more complex scales. Finally, "irrational metabolae" ofELD may
be introduced, which can only be applied in non-instrumental music.
Accordingly, the ELD can be taken from the field of real numbers).
The scale of limited transposition n 4 of Olivier Messiaen 15 (ELD =
1/2 tone):
3n A (4n+l V 4n +3 ) V 3n +1 A (4n V 4n +2 )
4n+l V 4
n
+3 V 3n+1 /\ (4n V 4n +2 )
where n = 0, 1, ... , modulo 3 or 4.
The second axiomatics leads us to vector spaces and graphic and
numerical representations.
16
Two conjunct intervals a and b can be combined by a musical operation
to produce a new interval c. This operation is called addition. To either an
ascending or a descending interval we may add a second conjunct interval
such that the result will be a unison; this second interval is the symmetric
interval of the first. Unison is a neutral interval; that is, when it is added to
any other interval, it does not modify it. We may also create intervals by
association without changing the result. Finally, in composing intervals we
can invert the orders of the intervals without changing the result. We have
just shown that the naive experience of musicians since antiquity (cf.
Aristoxenos) all over the earth attributes the structure of a commutative
group to intervals.
Now we are able to combine this group with a field structure. At least
two fields are possible: the set of real numbers, R, and the isomorphic set of
points on a straight line. It is morcover possible to combine the Abelian
group of intervals with the field C of complex numbers or with a field of
characteristic P. By definition the combination of the group of intervals
with a field forms a vector space in the following manner: As we have just
said, interval group G possesses an internal law of composition, addition.
Let a and b be two elements of the group. Thus we have:
1. a + b = c, c E G
2. a + b + c = (a + b) + c = a + (b + c) assocIatIVIty
3. a + 0 = 0 + a, with 0 E G the neutral element (unison)
4. a + a' = 0, with a' = - a = the symmetric interval of a
5. a + b = b + a commutativity
We notate the external composition of elements in G with those in the
field C by a dot. If A, fL E C (where C = the field of real numbers) then we
have the following properties:
6. Aa, f-La E G
7. la = al = a (I IS the neutral element m C with respect to
multiplication)
8. A (fLa) = (Af-L)a
9. (A + fL)a = Aa + /L.a}
A.(a + b) = Aa + Ab
associativity of I\, fL
distributivity
MUSICAL NOTATIONS AND ENCODINGS
The vector space structure of intervals of certain sound characteristics
permits us to treat their elements mathematically and to express them by
the set of numbers, which is indispensable for dialogue with computers, or
by the .set of points on a straight line, graphic expression often being very
convement.
The two preceding axiomatics may be applied to all sound charac-
teristics that possess the same structure. For example, at the moment it
would not make sense to speak of a scale of timbre which might be univer-
sally accepted as the scales of pitch, instants, and intensity arc. On the other
hand,. time, intensity, density (number of events per unit of time), the
quantIty of order or disorder (measured by entropy), etc., could be put into
one-to-one correspondence with the set of real numbers R and the set of
points on a straight line. (See Fig. VIII-I.)
Fig. VIII-1 Pitches Instants Intensities Densities Disorder
Moreover, the phenomenon of sound is a correspondence of sound
characteristics and therefore a correspondence of these axes. The simplest
212
Formalized Music
correspondence may be shown by Cartesian coordinates; for example, the
two axes in Fig. VIII-2. The unique point (H, T) corresponds to the sound
that has a pitch H at the instant T.
Fig. VIII-2
t
r .. --.. -.---------------1{1-1, T)
I
I
I
I
I
I
H
I must insist here on some facts that trouble many people and that are
used by others as false guides. We are all acquainted with the traditional
notation, perfected by thousands of years of effort, and which goes back to
Ancient Greece. Here we have just represented sounds by two new methods;
algebraically by a collection of numbers, and geometrically (or graphically
by sketches).
These three types of notation are nothing more than three codes, and
indeed there is no more reason to be dismayed by a page of figures than by a
full musical score, just as there is no reason to be totemically amazed by a
nicely elaborated graph. Each code has its advantages and disadvantages,
and the code of classical musical notation is very refined and precise, a
synthesis of the other two. It is absurd to think of giving an instrumentalist
who knows only notes a diagram to decipher (I am neglecting here certain
forms of regression-pseudomystics and mystifiers) or pages covered with
numerical notation delivered directly by a computer (unless a special coder
is added to it, which would translate the binary results into musical nota-
tion). But theoretically all music can be transcribed into these three codes
at the same time. The graph and table in Fig. VIII-3 are an example of
this correspondence: We must not lose sight of the fact that these three codes
are only visual symbols of an auditory reality, itself considered as a symbol.
Graphical Encoding for Macrostructures
At this point of this exposition, the unveiling of history as well as the
axiomatic reconstruction have been realized in part, and it would be useless
to continue. However, before concluding, I would like to give an example
of the advantage of a diagram in studying cases of great complexity.
i
A-440Cfs-
f t ; m e ~
T
Fig. VIII-3
/ ~
- t
I 7.. I
~ I sec =J P? P
-
f If
N T H V D
,
1 1.00 1 0 0.66 3
2 1.66 6 0 0.33 5
3 2.00 6 +17.5 0.80 6
4 2.80 13 0 ? 5
N = note number
H = pitch in half tones with + 1 0 h A 2 440 Hz
V = slope of glissando (if it exists) in semitones/sec,
positive if ascending, negative if descending
D = duration in seconds
I = number corresponding to a list of intensity
forms
213
Let us imagine some forms constructed with straight lines, using string
glissandi, for examplc.
17
Is it possible to distinguish some elementary forms?
Several of these elementary ruled fields are shown in Fig. VIn -4. In fact,
they can constitute elements incorporated into larger configurations.
Moreover it would be interesting to define and use in sequence the inter-
mediary steps (continuous or discontinuous) from one element to another,
especially to pass from the first to the last element in a more or less violent
way. If one observes these sonic fields well, one can distinguish the following
general qualities, variations of which can combine with these basic general
forms:
1. Registers (medium, shrill, etc.)
2. Overall density (large orchestra, small ensemble, etc.)
3. Overall intensity
4. Variation of timbre (area, sui pantieella, tremolo, etc.)
5. Fluctuations (local variations of 1., 2., 3., 4. above)
6. General progress of the form (transformation into other elementary
forms)
7. Degree of order. (Total disorder can only make sense ifit is calculated
according to the kinetic theory of gases. Graphic representation is the most
convenient for this study.)
214
Formalized Music

Fig. VIII-4
Let us now suppose the inverse, forms constructed by means of dis-
continuity, by sound-points; for example, string pizzicati. Our previous
remarks about continuity can be transferred to this case (see Fig. VIII-5).
Points 1.-7. are identical, so very broad is the abstraction. Besides, a mixture
of discontinuity and continuity gives us a new dimension.
Fig. VIII-5
GENERAL CASE
Organization Outside-Time
215
Consider a set U and a comparison of U by U (a product U x U)
denoted ",(U,f). Thenrp(U,j) c U x Uand for all pairs (u, u
f
) E U X U
such that u, u
f
E U, either (u, u
f
) E rp( U,j), or (u, u
f
) rp( U,j). It is reflexive
and (u uf) =:> (uf u); (u uf and Uf u') =:> U '" u' for u, u', Uf E
rp( U,f).
Thus rp( U,j) is an equivalence class. In particular if U is isomorphic
to the set Q of rational number.s, then U '" u
r
if lu - Uri ::;; !luI for arbitrary
!luf'
Now we define rp( U,j) as the set of weak values of U, rp( U, m) as the set
of average values, and rp(U, p) as the strong values. We then have
rp = rp(U,j) u tjJ(U, m) u rp(U,p) S;; U x U
where rp is the quotient set of U by rp. The subsets of tjJ may intersect or be
disjoint, and mayor may not form a partition of U x U. Here
rp(U,j) --3 tjJ(U, m) -3 rp(U,p)
are ordered by the relation -3 in such a way that the elements of rp( U,j) are
smaller than those of rp( U, m) and those of rp( U, m) are smaller than those of
rp(U, p). Then
rp(U,j) n rp(U, m) = 0, ",(U, m) n rp(U,p) = 0.
In each of these subsets we define four new equivalence relations and
therefore four sub-classes:
if and only if
Iu} - (u})'1 ::;; !lu} with u}, E rp(U,j)
for i = 1, 2, 3, 4, with rpl(U, f) c rp(U, f) and rpl(U,j) -3 rp2(U, f) -3
rp3( UJ) -3 rp4( UJ) ordered by the same relation -3. The same equivalence
relations and sub-classes are defined for rp( U, m) and rp( U, p).
For simplification we write
U{={U: UEtjJl(U,j)},
and the same for uj and
In the same way, equivalence sub-classes are created in two other sets,
G and D. Here U represents the set of time values, G the set of intensity
values, and D the set of density values with
U = {u{, u7, un
G - {gf gm gl'}
- U j' k
D = {dr, dr, dk}
for i,j, k = 1,2,3,4.
Take part of the triple product U x G x D composed of the points
gf, dj). Consider the paths VI: {ul, gj', d{}, V2: {u{, gi, di}, ... , VS:
{(ui, uV, (gf, g&, gD, (d'{', d.\")} for i = 1,2,3,4. VS will be
a subset of the triple product U x G x D split into 4
3
= 64 different points.
In each of these subsets choose a new subset K; defined by the n points
K; (j = 1,2, ... , nand ,\ = VI, V2, ... , VS). These n points are considered
as the n vertices ofa regular polyhedron. Consider the transformations which
leave the polyhedron unchanged, that is, its corresponding group.
To sum up, we have the following chain of inclusions:
w E C KI\ C ,\ U x G x D.
J
element vertex of set of path ,\
of the poly- vertices (subset of
UxGxD hedron K
j
of the U x G x D)
polyhedron
Consider the two other sets H (pitch) and X (sonic material, way of
playing, etc.). Form the product H x X x C in which C is the set of n
forms or complexes or sound types C
I
(i = 1, 2, ... , n); for example, a
cloud of sound-points or a cloud of glissandi. Map the product H x X x C
onto the vertices of the polyhedron Ki.
I. The complexes C
i
traverse the fixed vertices and thus produce group
transformations; we call this operation 8
0
,
2. The complexes C( are attached to corresponding vertices which
remain fixed, but the H x X traverse the vertices, also producing group
transformations; this operation is called 0I'
3. The product H x X x C traverses the vertices thus producing the
group transformations of the polyhedron; we call this operation OJl because
the product can change definition at each transformation of the polyhedron.
Organization In-Time
The last mapping will be inscribed in time in two possible ways in order
to manifest the peculiarities of this polyhedral group or the symmetric group
to which it is isomorphic: operation to-the vertices of the polyhedron are
expressed successively (model of the symmetric group); operation tl-the
vertices are expressed simultaneously (n simultaneous voices).
Product to x B 0 :
The vertices Kt are expressed successively with:
J. only one sonic complex C
n
always the same one, for example, a
cloud of sound-points only,
2. several sonic complexes, at most 11, in one-to-one attachment with
indices of vertices Kt,
3. several sonic complexes whose successive appearances express the
operations of the polyhedral group, the vertices i (defined by U x G x D)
always appearing in the same order,
4. several sonic complexes always in the same order while the order of
the vertices i reproduces the group transformations,
5. several sonic complexes transforming independently from the
vertices of the polyhedron.
Product to x 8
1
:
The list which this product generates may be obtained from the pre-
ceding one by substituting H x X in place of C
I

Product to x B:
This list may be readily established.
Case tl and OJ is obtained from the preceding ones by analogy.;-
To these in-time operational products one ought to be able to add
in-space operations when, for example, the sonic sources are distributed in
space in significant manner, as in Terretektorh or Nomos gamma.
218
Formalized Music
The three sets, D (densities), G (intensities), U (durations), are mapped
onto three vector spaces or onto a single three-dimensional vector space.
The following selection (subset) of equivalence classes, called path VI, is
made: D (densities) strong, G (intensities) strong, U (durations) weak.
Precise and ordered values have been given to these classes:
SetD a b c Set G Set U sec
(Elements/sec)
d
1
1.0 0.5 1 gl mf U
l
2
d
2
1.5 1.08 2 g2 f
112 3
d
a
2.0 2.32 3 ga ff
113 4
d
4
2.5 5.00 4
g4 Ilf 114 5
A second selection (subset), called path V2, is formed in the following
manner: D strong, G average, U strong, with ordered and precise values:
SetD Elements/sec Set G Set U sec
d
1
0.5 gl P
U
l
10
d
2
1
g2
mp U
2
17
d
a 2 g3 ml U3 21
d
4
3
g4 1
U
4
30
Eight" points" of the triple product D x G x U are selected.
For path VI:
Ki = dlglUl ; K ~ = dl g 4U4 ; K& = d4 g 4a4 ; K ~ = d 4g
1
Ul;
K ~ = d2 g2U2 ; K ~ = d2 g3U3 ; K ~ = d3 g
3
U3; K ~ = d
3
g
2
112'
r is the column (sub-class) of the table of set D. (r = a, b, c.)
For path V2:
Kl = d4 g aU2 ; K2 = d
a
g
2
U
l
; K3 = d
2
g
4
U
4
; K4 = d
l
g
2
u
a
;
K5 = d 4 g l U4 ; Ks = d
a
g
2
u
a
; K7 = d
2
g
a
U
2
; Ks = d
l
g
4
U
l

I. These eight points are regarded as solidly connected to each other
so as to form a cube (a mapping of these eight points onto the vertices of a
cube). The group formed by substitutions among these eight points, iso-
morphic to the symmetric group P4' is taken as the organizer principle. (See
Fig. VIII-6.)
ANALYSIS OF NOMOS ALPHA
I. The symmetry transformations of a cube given by the clements Kr
form the hexahedral group isomorphic to the symmetric group P4' The
rules for in-time setting are: 1. The vertices of the cube are sounded suc-
cessively at each transformation thanks to a one-to-one correspondence. 2.
The transformations are themselves successive (for a larger ensemble of
instruments one could choose one of the possible simultaneities as in Nomos
gamma). They follow various graphs (kinematic diagrams) inherent in the
internal structure of this particular group. (Sec Figs. VIII-G, 7, 8.)
220
(JJj
"'" '<),-
<0 i '"
....... < r;
/ '-. ""
Q:)
CJlJ
"t IQ
t-.: '
, '-
.-&- ()
"l - : ...
....... ",. (J
C\f.... , .. \,.,.,.,.
'tl
7J/b
'" '"..,
en : ""
v , f
\0 , : ..,
...... .. i.,i""
.... I ". '"
VJl:J
""
... \,0 '!l
co .:' "> i ....
-...., : /. <Q
r5
",
'"
.,
J;f]J
"> :' .... -
-;--
....... - : oa",J
-, ........... ...
, I\) '.
: '"
"" "/
t..; : to
/:
-.. ,-'!
/ .... ":::...... , ..
'b
lNJ
I 'b
\0
\.' "-
\
! N
: ....
- ...... <0
'.
....
\ '"
cCJJJ
"" \D
,a.-; :
: :.;j-
,:
':i>.. :::
............ -::::. '- G-
'" ..... .
"
f(Jj
"'" .' t--.
.... :
'" .. IX)
'"
Formalized Music

- <\I -

- IT) W ...:r--.
kJ1
- ('\I I"C1

..... I"\t -
ldf1rJ
!T>
: c..
: :::l
: 0
......... -... "'! .....
.....
". -..
" . -

N i
, ca
. "- -
: 0
\.j .... e.
...... ..." <::::l _
H') ..... '
._ '. "0
"1- Q)

: '" :t:
:: cD
:\ I
........ "< =
<>j ...... -:- :>
'<t-
'\
LL
221
+
I ABc /) D" F Fe Q Q'2. L
L L Q... lP] (jJ, f.r . 4?P (j)., 4l-o f/J;( . tPt..
1 1 A B c P
/)2
E'

,2
'-
L'
<r1
Qz
% I>!lr
4>"
416
10
(?( t?, jy;>", @II tBJ
4
A I (' 13 Ii L C;' L'2.
P F D' E' <i'l 4l q.r Yo. Q3 cr/J. /(>I QID
41",
.J>,r 4'9 tp4
B B C I A
LL E /)2
Ci
' L f/2 )
4',; 419 'ild' tp",
<;>'"
ty.(
!{Il CPJ l?.z fir Qy 4',..
(' ('.
8 A I
,2... t$J L () .L." 'If E t; 'I'll /fl! I?/o 419 <?.r
4'"
<?.r /fly t?3 4>,,- 4'1
D D L2 E1

b'-
I C L E -4 8
.-;<
03 I?, 4lr 4'.(
"'11'
frIo C?8
ol1. 4J.s-
l
/)1 <$'- L
E X /)
EI B c .(' el.
A
W+
flO /(J <I'll
Q:t,
tf'f Qrl416 9 ...
( E
<-
1$1
l 8
Et :z
/I j) G C
Wi'
IYI' 41; 4-t
{h
4'9
t;..,
(?IJ.. t/lJi 4' ... ,p/D 41
9
I
I. / D
[.l gt c Z E L 8 /I
[).
4"'0
1Y
Il, tp3 fYS' GAt ttl tj), {(s
9 G
/ L2 D L A 8
/)"-
9
2
1: C l= ifi.r flit- tp?- li?P <I'd' ((10 till
(j).(
i?9 i?6
93
9'-.!i.z
pL.
E L C
(;2.
1.
2
.4
I
Ii
1>
!3 (j>9 0 0l 4"0
Qj
4># ,py
4>, ct,
0 4 W.z
L l.. E
G
L
A G b C 8
1.2. 1:
tt2 ({R
tf9 41r
<96 I
tP"l
4J1 4'",,;4'/J P;

Lt L2 pie
['t
Fe 4 (-
p'l.
C r L
tP/ tPl gJl/ 116
9y
tP.r 0 tPz tP",14'9 C?.;z 14"0;

Q)'l !f6 ((9 C
r
r;, 92-
61"
<rA 473 <?v A
L'- Ii Ii'l.. L A 1:
(i E' /) c'
lP1G 1/1, lfiq r;.,
l/ilo
<r' tY"
{(s l/i3
qI;>
4J.l l= ir 16'
C' L2. II L FE D ,4
t;'t
Q31<?J' 147.r 1 qto 4J,. tPll C?'l ((6 t?f2I!'g
lY .. I?.,
1.2. 62.!Z L 8
2. /)
A
lilll 142 t18 lL?l t}. rr.r 4'6 (fJ",
16
2
1,.1 [) 8 L
p'L
L>-
<f'J' It!} l(>l,o (il] @d' t, try 1?/1 4?

E 4 /,<, (' L'- if ,
, 1((
tP{ fV2. (PIc
'P3
(/J9 (J.r tfJ' Itp.l{ ( D E 6 6
2
I 8 L
rrJ
lVt (J?{ Q, q>1( ti>3 qlo fPJ
93 rr :r
FL L Ll.. p"- c:
II
;;
ips 4JcP {j),o tY.r
{jJy
tP. I@> lif# f{6 IP L B IG'l. .I
Ii (2..
C
If', .til lA
6)2. ,y",
19.r @,
f6 If ... tP9- .. ,/- 8
L
1..4- /) ' 67,.

q ...
4>6 fi 109., tW 47" fiJp
(} p ... c
E .4 lJ 1/ &

I QQ
t!Y"l CP.r /fC Iq.q L C L2 Z t:i (/ E
b
Yl-f.l
li1
6), 4',
fi)i' 4f tPJ" tYtl
fPr

9
l> E ,If c d-
Fig. VIII-7
Example: DA = G on 0 the transformation of A. (Columns rows)
Fig. VIII-6. Symmetric Group P4: (1,2,3,4)
I 12345678 G2 32417685 Os 68572413
A 21436587 G 42138657 0
6
65782134
8 34127856 L2 13425786 0
1
87564312
C 43218765 L 14235867 Os 75863142
D2 23146758 0
7
78653421 0
9
58761432
D 31247568 O2 76583214 010 57681324
2 24316875 0 3 86754231 0
4
85674123
._ E 41328576 0
11
67852341 0
12
56871243
The numbers in roman type
also correspond to Group P4 = 4!
F jC Ii-
fi
c: I Ci' I l=
9"'-1.8
'-
j)
[:'- ./J'- il...
.4
JJ Ci' g
D'- A E
[;.2
Z
.c;2
C L..
L I
1/
?
L2.
4-

8

L.
.z
222
Formalized Music
II. Eight elements from the macroscopic sound complexes are mapped
onto the letters C
f
in t12rce ways, a, {3, y:
a
C
1
C
7
C
3
C
5
C
s
C
2
C
a
C
4
{3
C
1
C
2
C
3
C
5
C
a
C
7
C
a
C
4
Y
C
1
= ataxic cloud of sound-points
C
5
= relatively ordered ascending or descending cloud of sound-
points
C
a
= relatively ordered cloud of sound-points, neither ascending
nor descending
C
2
= ataxic field of sliding sounds
C
a
= relatively ordered ascending or field of sliding
sounds
C
4
= relatively ordered field of sliding sounds, neither ascending
nor descending
C
a
= atom represented on a cello by interferences of a quasi-
UJ1lson
C
7
= ionized atom represented on a cello by interferences,
accompanied by pizzicati
III. These letters are mapped one-to-one onto the eight vertices of a
second cube. Thus a second hexahedral group is taken as the organizer
principle.
223
II. The mapping of the eight forms onto the letters C
f
change cyclically
in the order a, {3, y, a, ... after each three substitutions of the cube.
III. The same is true for the cube of the letters C
f

o
V,
Fig. VIII-8
V, V2,. v} \ I.j "J u.
1/, v, V. v. Y" VI" vi
V, Vz v, V, Vb V. Vr
Vl Y, V, Y, v .. V, 1/4
y VI 1/;; V. V, v. V,
Vr vr Vb 1/4 V3 v, v.
'h Vi, Yv 'k V2. Vl V,
.,
"'0'
VL II,
,
v,
V,=Z+A-t-8f-C
'= P + F-'Lf- 9 f-
1'3 p>- f- !:- f- fi'- -f- t...
[
Vlt "'- 'f6 f- f /:1.. i- 0- i- 4'.
V, ==. flo -f if.! f- Qj7'-ip s-
y'G := ej).z 1- 6(4 +tf.l/1- 'P,?
IV. Take the products K; x C
j
and K/ X Cm. Then take the product
set H x X. Set H is the vector space of pitch, while set X is the set of ways
of playing the C
i
. This product is given by a table of dou ble entries:
Extremely
High
Medium
High
Medium
Low
Extremely
Low
biJ
N N
N
U C
N
C
'5.. '5.. w oj ro
I I
C
l
, C
2
, C
3
pizzo = plzzlcati
f.c.!. = struck with the wood of the
bow
an = normal arco
pizzo gl. = pizzicato-glissando
a trem. = normal arco with tremolo
harm. = harmonic sound
s
w
S
E
<1.l
e
.... ....
Q.)
""
Q.)
11)
t::
.....
.... ....
ci. ci. .S
""
oj .... c
'" '"
oj ..c: ..c: oj oj o:s ~
LJ
C
4
, C
5
, C
6
C
7
, C
a
hr trem. = harmonic sound with
tremolo
asp = arco suI ponticello
asp trem. = arco sui ponti cello
tremolo
a interf. = arco with interferences
Various methods of playing are attributed to the forms C
l
, " C
a
, as
indicated in the table. The first and fourth rows, extremely high and
extremely low pitches, are reserved for path V2. A sub-space of H' is
attributed to path VI. It consists of the second and third rows of the pre-
ceding table, each divided into two. These four parts are defined in terms
of the playing range of the corresponding column.
V. The mapping of C
1
onto the product set H x X is relatively in-
dependent and will be determined by a kinematic diagram of operations
at the moment of the in-time setting.
IV. The products K[ x C
j
and K/ X C
m
are the result of the product
of two graphs of closed transformations of the cube in itself. The mapping
of the graphs is one-to-one and sounded successively; for example:
(See Figs. VIII-9, 10.)
v. "
~
"
' ~
,
" "
"0"
" 110
,
Q'Oq,
fl. ~ 1 ~ Q ~ Q, lJ G ~ Q, ,L Q ~
GO'
Q'OG,
G'O"
V" v, I'it,Q, Q E. <:IS' ~ q, Q t I f Q, Q.. 0, Q, Qt III Q, l tl,v
,q, q"
.'
,
"
,
, <'
"0''-Y,
Vr VII
Fig. VIII-9
V. Each Cj is mapped onto one of the cells of H x X according to
two principles: maximum expansion (minimum repetition), and maximum
contrast or maximum resemblance. (See Fig. VIII-II.)
"
I
J
226
Fig. VIII-10
Formalized Music
d.,: /.0
d.: I.f I"
d
J
, 2.0 2-32
c4': 2S S"
P4 TH V1
lw-) /PI./ "S ... 0,;
2M QJ k K6 k,
';.- / -<.
6'1!1 3.7.2 7j8 2Sa IV': 2.;lf :&!s
l;f;! r "'/$ $
i&-) q/9 '" "7 "J' ... '3 .. r,
i{i<} ;7 /.0 /.7 f, K, k. k3 1</ k&
iWi E
l{t} L
r
A (13,11)
L (13,11)
/' _ -< -;,,/:
6.()i 79!1 J>.83 37.2
I/!/ ;!
CF,i Di
/r/ r,
\;;1 =
Gj U;
k. k.
-::/ .::,:
-"'4> U.s. 1.00
"'I ,#' "'I ;fF
u-8 r,. C1j-
h K? ka k:.
-< -< ,::::'
2. 8 4.0 S..lt ToN
"'/ $ .. ; /I' .I " ./ #
i&-)c(8 tr7 "i' t18 I1t "3 0; oq 0'"'.2,
i{i<) /('" itS' k. .t:7 It. ,1:2 .1:3
r/ ....
7.8(; 7f 6R!' /J
11.f#",1
or
k. k,
-< -9-
6,88 S:2v 1M2 786 4, 2. 1+ 8
/I f ! I ;I "'I #I "'I
"Of" Pad VI
k'j'd. "'/.2
k ..
Ol
= /. HY." . ..r
kJ' = 2.f. #,'''5
kg<
K; =/"'/'U2'
k: = ;.S'..f 3.N
k7"" =).,: . .f. !He
k.1 =.2.,: . .f. 2.0
6"i D3
= 2,./

1I-'1J#,I
= ;}.,I

=5".IS'#
,fjff
=,f'2i1'
t;-- = (J.J-ntf-t.
):'/ = u.ff . 40$
4: : :5'0,J!/' 'N'
it
kf =/.01/.2. 62
k/
kJ =.<.111',,4.
k/ =2.321?t:2
K+- k2 KJ kr k .1:'4; i:"
:?
if:) t;!. 116 ffJ !rf. If'J rr; cr.;..
i(K} IPS' t, 1:8 h k? Ky .k, *3
-<:;'" J .,<; ...... . "
Sir J.:/fo bM 95 S' :( 1/.2J'
,f //JZ.I!o/V!f
j{v-) cPf D'4; U; '8 U; t7i 0; U; U;
j M i?". "'8 k, k k" ,<J K, k.
Jf" -< -::-?, - -;i;:;
'Ii! !!J1 fJ, ;} 4'1:r':< f"
f ,! ;" # 0/ J!' "'/ II
i. (trJ P7. OJ U; d"'i cry a; fJ- "
1 11:) If 1:, t, kr k.
f ...
,1 (s; 7) .2.lr / I.?-<r 10 7J<! ,z8] J72 . 08
Uq IJI "" >? "'I ff /- /! ./
(3 "8 CJ.; us- "i Ci 'J U;
i(k) ({; k.r .If /(7 k. kl k, kJ K.
-< I' -< ,;::.:. .;;:'
?SJ {f5' 1Je 5J1 ( 10 em' 'n,
I I ;f ; "'I "'I )'I $
Q
i(!.) <1'1 '" kc kr kJ t, k, Iy
-< I' =.,.:.,; .:'.',: .,;,/.
7' . '11' III 2.2" 10 ,22, !
II /;f r ,F .. f#''''f
-/ . .,./ K{ = 1.0",f2. 0
=:;.2S'$ kl 01f 4 . .>
",if' =4h/f4.r
=: {o[J.1 k!
=NJ.f ;1
72''# k/
.. :F k/
. 8f '>l =3.0'/':;, oJ
=2.0.";:
-".5.1f
=$ .. "
=8.0nf
=J:.!'tf
=01.#
=/tJ. 31'.#
'" 7.l!6f
227
-<. ...... -<
Vo "" :2. "''It ,7.86 5.'t
J!' "'/ "'I / ,.; / /
i{rr)c?7 ere '? or '7 VI; r,- tTi.
K", h K? f:, k, k, Jj
-< co:' =- ,f' ..:,:
G;$' TBG .5'21 /1J.3,Z B .z Jy
j'/;"-f?/IL,,,/V'

k h
-i. ;"" -< ;f'
C:fi; 2y 10.11' 78 4.r.2 I..
,,;",f /Jf v4''''r
4J' 2 //.2; f"r SO" g9J (.ffB
!! .. , /;f ./ ;f / :I'
U; U; '4 5C ':t U; o-v "i.
h
,:.,::" fl..::. -< / ",. ":': ..q,
S:;f 3 !)r CH; of' .2 . 4.J" INJ'
I I /I ,f "'/ "I # #'
tfr)tPtr rr; u; a; r,. tr;
i k;. .1:, I:r I::J /.;
-< r:;t'.f Y ... :'., -. :,':;
G& 5:1f 5:/. :11J II .. , of'
# .r # ." .#' "'I .If" ",r
228
" ,,'
............ "
...........
" ..... :::

/,z ,'to" //

pz an flI1
fJz. = piZ"Licat;'
Fig. VIII-11. Products H' x X in the Set of the Cj
Formalized Music
an. beats
f' VA/ln),>.>'" f< v"vn
A t v -\A ft f' r Y v n n y
f<rVVY)'MrJf'f'
OIn 1-
.a .. /o
d j> t

229
VI. The products K[ x C
i
x H' x X and K
j
x G
I
X Hextremes x X
are formed.
VII. The set of logical functions (a) is used in this piece. Its moduli
are taken from the subset formed by the prime residual classes modulo
18, with multiplication, and reduction modulo 18.
L(m, n) = (n! V nj V n
k
V nl) II mp V (mq V m
r
) II ns V (n
t
V nu V n,J (a)
Its clements are developed:
1. From a departure function:
L(II, 13) = (13
3
v 13
5
V 13
7
v 13
9
) II 112 v (114 V lIB)
II 13
9
V (13
0
V 13
1
V 13
0
)
2. From a "metabola" of moduli which is identical here to the graph
coupling the elements of the preceding subset. This metabola givcs the
following functions: L(II, 13), L(I7,5), L(I3, II), L(17,7), L(l1,5),
L(I, 5), L(5, 7), L(I7, 11), L(7, 5), L(17, 13), L(5, 11), L(l, 11). (See Fig.
VIII-12, and Table of the Sieve Functions and Their Metabolae.)
3. From three substitution rules for indices (residual classes) :
Rule a: ma --'>- n
O
+
1
Rule b: If all indices within a set of parentheses are equal, the next
function L(m, n) puts them in arithmetic progression modulo the cor-
responding sieve.
Rule c; Conversion of indices as a consequence of moduli metabolae
(see Rule c. Table) :
for example, 74 --'>- llx, x = 4( 11 /7) '" 6.
4. From a metabola of ELD (elementary displacement: one quarter-
tone for path VI, three-quarters of a tone for path V2).
The two types ofmctabolae which generate the elements of set L(m, n)
can be used outside-time or inscribed in-time. In the first case, they give us
the totality of thc clements; in the second case, these elements appear in a
temporal order. Nevertheless a structure of temporal order is subjacent even
in thc first case.
5. From a special metabola that would simultaneously attribute differ-
ent notes to the origins of the sieves constituting the function L(m, n).
VI. The elements of the product K[ x G
j
x H' x X of the path VI
are sounded successively, except for interpolation of elements of the product
K
j
x G
j
x Hcxtremes x X from path V2, which are sounded intermittently.
VII. Each of the three substitutions of the two cubes Ki and Gil the
logical function L(m, n) (see Fig. VIII-ll), changes following its kinematic
diagram, developed from the group: multiplication by pairs of residual
classes and reduction modulo 18. (See Fig. VIII-lO.)
Table of the Sieve Functions and Their Metabolae
L(11, 13) = (13
3
+ 13
5
+ 13
7
+ 13
9
)11
2
+ (114 + l1s)13
9
+ 13
0
+ 13
1
+ 13
6
L(17, 5) = (51 + 52 + 53 + 5
4
)17
1
+ (177 + 17d54 + 51 + 50 + 52
L(]3, 11) = (112 + 114 + 117 + 11
9
)]3
0
+ (13
5
+ 13
10
)11
9
+112+111+114
L(17,7) = (71 + 73 + 75 + 76)171 + (176 + 17d76 + 71 + 70 + 73
L(II,5) = (50 + 52 + 53 + 5
4
)110 + (114 + 118)54 + 50 + 51 + 52
L(I,5) = (51 + 52 + 53 + 5
4
)1
1
+ (11 + 1
1
)5
4
+ 51 + 52 + 53
L(5, 7) = (71 + 73 + 74 + 76)5
0
+ (50 + 5
1
)76 + 71 + 73 + 74
L( I 7, 11) = (I 12 + I I 5 + 116 + 11
9
) 111 + (17 1 + 1 7
3
) 119
L(7, 5)
L(17, 13)
L(S, 11)
L( I, 11)
n
-'
m
+ lIz + 115 + 116
(51 + 52 + 53 + 5
4
)7
0
+ (70 + 71)54 + 51 + 52 + 53
(13
3
+ 13
5
+ 13
s
+ 13
10
)17
1
+ (171 + 17
2
)13
10
+ 13
3
+ 13
5
+ 13
8
(113 + 114 + 117 + 11s)5
0
+ (50 + 5
1
)l1s + 113 + 114 + 117
(113 + 114 + 117 + lIs) 11 + (11 + 10) 118 + 113 + 114 + 117
Rule c. Table
5
5
7
1.4
7
:5 "7
!.! = 2.2
11
1.57
II
5 7
rr=
I; = 2.6
13
1.85
13
1.2
13
7= rr=
13
1; := 3.4 1; = 2.43
17
1.54
17
1.3
rr=
13
232
Formalized Music
Group and sub-group of residual classes obtained by ordinary multiplication
followed by reduction relative to the modulus 18
1
5
7
II
13
17
I
5
7
11
13
17
5 7 11 13 17 7 13
5 7 11 13 17 1 1 7 13
7 17 I 11 13 7 7 13 1
17 13 5 1 11 13 13 1 7
1 5 13 17 7
11 1 17 7 5
13 11 7 5 1
DETAILED AN AL YS IS OF THE BEGINNING OF THE SCORE (L (11, 13) )18
Thanks to the metabola in 5. of the outside-time organization, the
origins of the partial sieves (13
a
V 13
5
V 13
7
V 13
s
) A 112 V (114 V 11
8
) A
13
s
and 13
0
V 13
1
V 13
6
correspond to Aa# and A
a
, respectively, for
A3 = 440 Hz. Hence the sieve L(l1, 13) will produce the following pitches:
... C;, C2#, D2, D2t F
2
, F.,J/t, G
2
, G
2
#f, A
2
, B2t C
a
, CN, D
3
#, DJjf, Fat,
Fa#, Ga#, Aat Aa#, Ba, C4 t D'l +. E4, E4 -f. G
4
, A
4
, A4 #, A4=/11 ,
The order applied to the sonic complexes (Sn) and to the density,
intensity, and duration combinations (K,,) are for transformation fJ:
............ :.:.
....
....
S
....

mf
K2 = 2.25 fff
Ka = 22.5 fff

K5 = 2.83 f

Ke = 3.72 ff
S7===
K7 = 7.98 ff
Sa =-==:::: Ka = 6.08 f
(In this text C
n
is replaced by Sn.)
First sequence (see Fig. VIII-13):
I 2 3 4 5 6 7 8
.j.
t
.j. .j. .j.
t
.j.
t
D(Sn) = S2 Sa Sl S4 S6 S7 S5 Se
D(Kn) = K2 Ka KI K4 Ke K7 K5 Ka
2.25 22.5 I 10 3.72 7.98 2.83 6.08
fff fff mf mf ff ff f f
This part begins with a pizzicato glide on the note C, fff (the
sliding starts PPP). The slope of the glide is zero at first and then very
weak (1/4 tone per 2.5 seconds).
Sa consists of c-f C# D struck col legno, fff (with p in the
middle). In Sa there is an introduction of beats obtained by raising
Gilt towards A.
Second sequence, beginning at QI2/Qa:
I 2 3 4 5 6 7 8
.j. .j. .j. .j. .j. .j. .j. .j.
Q12(S,,) = Ss S6 Sa S7 Sl S2 S4 Sa
Qa(Kn) = Ka Ke K7 Ks K4 K2 Ka KI
6.08 3.72 7.98 2.83 10 2.25 22.5 1.0
f ff ff f mf fff fff mf
Note, as in the preceding part, the previously calculated con-
traction of the values of duration .
Sl is ataxic, lasting more than a second .
Third sequence, beginning at Q4/Q7:
1 2 3 4 5 6 7 8
t t t
.j. .j. .j.
t
.j.
Q4(S,,) = Se S7 Sa S5 S2 Sa S4 S1
Q7(K .. ) = Ka K7 Ks Ks K4 Ka KI K2
6.08 7.98 2.83 3.72 lO 22.5 1.0 2.25
f ff f ff mf fff mf fff
233
234
Formalized Music Towards a Philosophy of Music 235
I
t,

,'11,1
'"
li
\
t 111
U, """I
r= 'r

cl) >< ..
L
'Ij,
fL
I I

:;
'"
r ,:.
I
i
r I

I,
i
lill
u,"i: iI:
'"'
II

l:
,

l
.

l
a;
'1\
U

0
II
-
I
'"
-t::
I .s.
I:
'"
'"
r "
c
E
.....

I
I) -

f
0
" .,.
l' f!!
lM
r
II) "< <II
(1)""
L'-
{f
m

"- OJ

11\
c: ,
'2
r
LM
0-
lD
JG
0-
r
,.
'/'
0
Pi
..
lL ..
::!:
c;-::2

f'
:>

I
0,
Fig. VIII-12. Nomos alpha Sieves
,
I
I
(
it
FO""-;n>

In Sa the slopes of the glissandi in opposite directions cancel each
other. The enlargement in S4 is produced by displacement of the
lower line and the inducement of beats. The cloud is introduced by a
pizzicato on the C string; the index finger of the left hand is placed
on the string at the place where one would play thc note in square
brackets; then by plucking that part of the string between thc nut
and the index finger with the left thumb, the sound that results will
be the note in parentheses.
NOMOS GAMMA-A GENERALIZATION OF NOMOS ALPHA
The finite combinatorial construction expressed by finite groups and
performed on one cello in Nomos alpha is transposod to full orchestra in
Nomos gamma (1967/68). The ninety-eight musicians are scattered in the
audience; this scattering allows the amplification of Nomos alpha's structure.
Terretektorh (1965/66), which preceded Nomos gamma, innovated the
scattering of the orchestra and proposed two fundamental changes:
fl. The quasi-stochastic sprinkling of the orchestral musicians
among the audience. The orchestra is in the audience and the audience
is in the orchestra. The public should be free to move or to sit on
camp-stools given out at the entrance to the hall. Each musician of
the orchestra should be seated on an individual, but un resonant, daiS
with his desk and instruments. The hall where the piece is (0 be per-
formed should be cleared of every movable object that might cause
aural or visual obstruction (seats, stage, etc.) A large ball-room having
(if it were circular) a minimum diameter of 45 yards would scrve in
default of a new kind of architecture which will have to be devised for
all types of present-day music, for neither amphitheatres, and still less
normal theatres or concert-halls, arc suitable.
The scattering of the musicians brings in a radically new kinetic
conception of music which no modern electro-acoustical mcans could
match.
I9
For ifit is not possible to imagine 90 magnetic tapc tracks re-
laying to 90 loud speakers disseminated all over the auditorium, on
the contrary it is quite possible to achieve this with a classical orchestra
of 90 musicians. The musical composition will thereby be entirely
enriched throughout the hall both in spatial dimension and in move-
ment. The speeds and accelerations of the movement of the sounds
will bc realized, and new and powerful functions will be able (0 be
made use of, such as logarithmic or Archimedean spirals, in-timc and
geometrically. Ordered or disordercd sonorous masses, rolling one
against thc other like waves ... etc., will bc possible.
Terrelektorh is thus a "50notron": an accelerator of sonorous
particles, a disintegrator of sonorous masses, a synthesizer. It puts the
sound and the music all around the listener and close up to him. It
tears down the psychological and auditive curtain that separates him
from the players when positioned far off on a pedestal, itself frequently
enough placed inside a box. The orchestral musician rediscovers his
responsibility as an artist, as an individual.
b. The orchestral colour is moved towards the spectrum of dry
sounds, full of noise, in order to broaden the sound-palette of the
orchestra and to give maximum effect to the scattering mentioned
above. For this effect, each of the 90 musicians has, besides his
normal string or wind instrument, three percussion instruments, viz.
Wood-block, Maracas, and Whip as well as small Siren-whistles,
which are of three registers and give sounds resembling flames. So if
necessary, a shower of hail or even a murmuring of pine-forests can
encompass each listener, or in fact any other atmosphere or linear
concept either static or in motion. Finally the listener, each one
individually, will find himself either perched on top of a mountain in
the middle of a storm which attacks him from all sides, or in a frail
barque tossing on the open sea, or again in a universe dotted about
with little stars of sound, moving in compact nebulae or isolated.
20
237
Now the crux or thesis of Nomos gamma is a combinatorial organization
of correspondences, finite and outside the time of the sets of sound
characteristics. Various groups are exploited; their inner structure and
their interdependency are put in relief musically: cyclic group of order 6,
groups of the rectangle (Klein), the triangle, the square, the pentagon, the
hexagon, the tetrahedron, and the hexahedron.
The isomorphisms are established in many ways, that is, each one of
the preceding groups is expressed by different sets and correspondences,
thus obtaining structures set up on several interrelated levels. Various
groups are interlocked, intermingled, and interwoven. Thus a vast sonic
tapestry of non-temporal essence is formed (which incidentally includes the
organization of time and durations). The space also contributes, and is
organically treated, in the same manner as the more abstract sets of sound
elements.
A powerful deterministic and finite machinery is thus promulgated. Is it
symmetrical to the probabilistic and stochastic machineries already proposed?
The two poles, one of pure chance, the other of pure determinacy, are
dialectically blended in man's mind (and perhaps in nature as well, as
Epicurus or Heisenberg wished it). The mind of man should be able to
travel back and forth constantly, with ease and elegance, through the
fantastic wall, of disarray caused by irrationality, that separates determinacy
from indeterminacy.
We will now consider some examples. It goes without saying that
Nomos gamma is not entirely defined by group transformations. Arbitrary
ranges of decisions are disseminated into the piece, as in all my works
except for those originated by the stochastic program in Chap. V. However
Nomos gamma represents a stage in the method of mechanization by com-
puters for this category of problem.
Measures 1-16 (three oboes. then three clarinets)
OUTSIDE-TIME STRUCTURE
Set of pitches: H = {H1' H
2
, H
3
, H
4
, H5}' Origins: D
3
, G1b D
4
, G#4,
D
5
, respectively, with range 3 semitones. ,.,. "
Set of durations: V = {V1' V
2
, Va, V
4
}. Origins: d,-,-' v, v ___ J-.J' ,
~ J J, respectively, with range one sixteenth-note and a half note
~ 1 sec.
Set of intensities: G = {G
1
,G
2
,G
3
,G
4
}. G
1
= {PPp,p}P,pp,pp,lp,P},
"> > > > >
G
2
= {p, p, mp, ~ p , ~ p , mf, mf}, G
3
= {mf,f,],j; if,jJ,ff}, G
4
= {it, sfjJff,
> > >
fff,fff, sirf, Jif}. Origins: pp, ~ p J iff, respectively.
Product sets: K = H x V x G. Each one of the points of the product
set is defined by a sieve modulo n considered as an element of an additive
group (e.g., n = 3, ... ,3
0
-+ 3
1
-)- 3
1
-+ 3
2
--+ 3
0
--+ 3
2
-+ 3
2
--+ 3
1
--+ ... )
and by its unit, that is, the elementary displacement ELD:
Kl = H4 X G
2
X VI K2 = H4 X G
3
X V
2 K3 = H4 X G
1
X VI
Moduli: 2 2 2 2 2 2 2 2 2
ELD: t tone t sec t tone t sec t tone i see
K4 = H4 X G
3
X VI K5 = H4 X G
z
X V
2
K6 = H4 X G
3
X U
3
Moduli: 3 2 3 3 2 3 3 2 3
ELD: t tone t sec t tone t sec t tone i sec
In addition, K2 and K3 are deformed by translations and homothetic trans-
formations of the H values.
Let us now consider the three points K
1
, K
2
, K3 of the product H x
G x V, and map them one-to-one onto three successive moments of time.
We thus define the triangle group with the following clements:
{I, A, A2, B, BA, BA2} +--+ {l23, 312, 231, 132,213, 321}
IN-TIME STRUCTURE
For each transformation of the triangle the vertices are stated by K1,
K
2
, Ka, which arc played successively by the oboes and the clarinets,
according to the above permutation group and to the following circuit:
BA, BA2, A, B, BA2, A2.
Measures 16-22 (three oboes and three clarinets)
OUTSIDE-TIME STRUCTURE
Form the product K
f
x C
j
: Kl x C
1
, K2 X C
2
, Ka x C
3
, in which the
C
j
are the ways of playing. C
1
= smooth sound without vibrato, C
2
= flutter
tongue, C
3
= quilisma (irregular oscillations of pitch).
Consider now two triangles whose respective vertices are the three
oboes and the three clarinets. The K
j
x C
j
values are the names of the
vertices. All the one-to-one mappings of the Ki x Cj names onto the three
space positions of the three oboes or of the three clarinets form one triangle
group.
IN-TIME STRUCTURE
To each group transformation the names Kj x C
j
are stated simul-
taneously by the three oboes, which alternate with the three clarinets. The
circuits are chosen to be 1, BA, BA, 1, A2, B, BA, A, BA2 and 1, B, B.
Measures 404-42-A Sound Tapestry
The string orchestra (sixteen first violins, fourteen second violins,
twelve violas, ten cellos, and eight double basses) is divided into two times
three teams of eight instruments each: rp1' rp2' <P3' .p1' .p2, .p3' The remaining
twelve strings duplicate the ones sitting nearest them. In the text that
follows the <Pi and .pj are considered equivalent in pairs (rpl ~ if'i)' Therefore
we shall only deal with the <Pt.
LEVEL I -OUTSIDE-TIME STRUCTURE
The eight positions of the instruments of each <Pi are purposely taken
into consideration. Onto these positions (instruments) we map one-to-one
eight ways of playing drawn from set X = {on the bridge tremolo, on the
bridge tremolo and trill, sui ponticello smooth, sui ponticello tremolo,
smooth natural harmonic notes, irregular dense strokes with the wood of
the bow, normal arco with tremolo, pizzicato-glissando ascending or
descending}. We have thus formed a cube: KVBOS 1.
Onto these same eight positions (instruments) of <Pt we map one-to-one
eight dynamic forms of intensity taken from the following sets: gA =
{ppp crescendo, ppp diminuendo, pp crcsc, pp dim, p cresc, }j dim, mp cresc,
mp dim}, g ~ = {mf cresc, mf dim, f cresc, j dim, ff cresc,J! dim, fff cresc,
iff dim}, & = {p dim, }j cresc, mp dim, rfip cresc, mf dim, mf cresc, J dim,
!cresc}. We have thus defined a second cube: KVBOS 2.
LEVEL I-IN-TIME STRUCTURE
Each one of these cubes is transformed into itsclffollowing the kinematic
diagrams of the hexahedral group (cf. Nomos alpha, p. 225); for example,
KVBOS I following D2Q12 . .. and KVBOS 2 following Ql1Q7.
LEVEL 2-0UTSIDE-TIME STRUCTURE
The three partitions 4>1' 4>2' 4>3 are now considen!d as a triplet of points
in space_ We map onto them, one-to-one, three distinct pitch ranges H
a
,
He, Hy in which the instrumentalists of the preceding cubes will play. We
have thus formed a triangle TRIA I.
Onto these same three points we map one-to-one three elements drawn
from the product (durations x intensities), U x G = {2.5 sec gA' 0.5 sec gil'
1.5 sec gd. We have thus defined a second triangle TRIA 2.
LEVEL 2-IN-TIME STRUCTURE
When the two cubes playa Level 1 transformation, the two triangles
simultaneously perform a transformation of the triangle group. If I, A,
A2, B, BA, BA2 are the group elements, then TRIA I proceeds according
to the kinematic diagram A, B, BA2, A2, BA, BA2, and TRIA 2 proceeds
simultaneously according to A, BA2, BA, A2, B, AB.
Form the product C( x M
j
with three macroscopic types: C
1
=
clouds of webs of pitch glissandi, C
z
= clouds of sound-points, and C
3
=
clouds of sounds with quilisma. Three sieves with modulus AI = 3 are taken:
3
0
, 3
1
, 3
2
From this product we select five elements: C
1
x 3
0
= I, C
1
X
3
1
= A, C
1
X 3
2
= A3, C
2
X 3
0
= A4, C
3
X 3
1
= A5, which could belong
to the cyclic group of order 6.
The nested transformation of Levels 1 and 2 are plunged into the
product C( x M
j
, which traverses successivcly C
1
x 3
2
, C
2
X 3
0
, C
1
X 3
1
,
C
3
X 3
1
, C
1
X 3
0
o ( ~ A3, A\ A, A5, I, during the corresponding arbitrary
durations of 20 sec, 7.5 sec, 12.5 sec, 12.5 sec, 7.5 sec.
The partition of the string orchestra into teams q,(, .pj is done in two
modes: compact and dispersed. The compact mode is itself divided into two
cases: Compact I and Compact II. For example,
in Compact I, 4>1 = {VI
I3
, VIII, VII
2
, VII
14
, A
7
, VC
2
, VC
6
, CB4}
in Compact II, <PI = {VII, VI
7
, VIs, VI g , VI
1o
, As, VC
3
, CB2}
in the dispersed mode, 4>1 = {VI2' VI
3
, VI
6
, VIII' VII6, VIllI> CB3 , CB7 }
(VIi = ith first violin, VIIi = ith second violin, Ai = ith viola, VCi = ith
cello, CB
i
= ith double bass.) These partitions cannot occur simultaneously.
All the mechanisms that sprang from Levels I, 2, 3 are in turn plunged
into the various above definitions of the <Pi and .pj teams, and successively
into Compact I during the 27.5 sec duration, into the dispersed mode during
the 17.5 sec duration, into Compact II during 5 sec, into the dispersed mode
during 5 sec, and into Compact I during 5 sec.
DESTINY'S INDICATORS
Thus the inquiry applied to music leads us to the innermost parts of our
mind. Modern axiomatics disentangle once more, in a more precise manner
now, the significant grooves that the past has etched on the rock of our
being. These mental premises confirm and justify the billions of years of
accumulation and destruction of signs. But awareness of their limitation,
their closure, forces us to destroy them.
All of a sudden it is unthinkable that the human mind forges its con-
ception of time and space in childhood and never alters it.
21
Thus the
bottom of the cave would not reflect the beings who are behind us, but
would be a filtering glass that would allow us to guess at what is at the very
heart of the universe. It is this bottom that must be broken up.
Consequences: 1. It would be necessary to change the ordered structures of
time and space, those of logic, ... 2. Art, and sciences annexed to it, should
realize this mutation.
Let us resolve the duality mortal-eternal: the future is in the past and
vice-versa; the evanescence of the present is abolished, it is everywhere at
the same time; the here is also two billion light-years away ....
The space ships that ambitious technology have produced may not
carry us as far as liberation from our mental shackles could. This is the
fantastic perspective that art-science opens to us in the Pythagorean-
Parmenidean field.
Chapter IX
New Proposals in Microsound
Structure
FOURIER SERIES-BASIC IMPORTANCE AND INADEQUACY
The physico-mathematical apparatus of acoustics [2, 23] is plunged into the
theories of energy propagation in an elastic medium, in which harmonic
analysis is the cornerstone.
The same apparatus finds in the units of electronic circuit design the
practical medium where it is realized and checked.
The prodigious development of radio and TV transmissions has expanded
the Fourier harmonic analysis to very broad and heterogeneous domains.
Other theories, quite far apart, e.g., servomechanisms and probability, find
necessary backing in Fourier series.
In music ancient traditions of scales, as well as those of string and pipe
resonances, also lead to circular functions and their linear combinations [24].
In consequence, any attempt to produce a sound artificially could not be
conceived outside the framework of the above physico-mathematical and
electronic apparatus, which relics on Fourier series.
Indeed the long route traversed by the acousmatics of the Pythagoreans
seemed to have found its natural bed. Musical theoreticians did base their
theories on Fourier, more or less directly, in order to support the argument
about the natural harmony of tonality. Moreover, in defining tonality, the
20th-century deprecators of the new musical languages based their argu-
ments on the theory of vibration of elastic bodies and media, that is, in the
end, on Fourier analysis. But they were thus creating a paradox, for al-
242
New Proposals in Microsound Structure 243
though they wanted to keep music in the intuitive and instinctive domain,
in order to legitimatize the tonal universe they made use of physico-
mathematical arguments!
The Impasse of Harmonic Analysis and Some Reasons
Two major difficulties compel us to think in another way:
1. The defeat by the thrust of the new languages of the theory accord-
ing to which harmony, counterpoint, etc., must stem, just from the basis
formed by circular functions. E.g., how can we justify such harmonic con-
figurations of recent instrumental or electro-acoustic music as a cloud of
gliding sounds? Thus, harmonic analysis has been short-circuited in
spite of touching attempts like Hindemith's explanation of Schonberg's
system [25]. Life and sound adventures jostle the traditional thescs, which
are nevertheless still being taught in the conservatories (rudimentally, of
course). It is therefore natural to think that the disruptions in music in the
last 60 years tend to prove once again that music and its "rules" are socio-
cultural and historical conditionings, and hence modifiable. These conditions
seem to be based roughly on a. the absolute limits of our senses and their
deforming power (c.g., Fletcher contours); b. our canvass of mental struc-
tures, some of which were treated in the preceding chapters (ordering,
groups, etc.); c. the means of sound production (orchestral instruments,
electro-acoustic sound synthesis, storage and transformation analogue
systems, digital sound synthesis with computers and digital to analogue
converters). If we modify anyone of these three points, our socio-cultural
conditioning will also tend to change in spite of an obvious inertia inherent
in a sort of "en tropy" of the social facts.
2. The obvious failure, since the birth of oscillating circuits in elec-
tronics, to reconstitute any sound, even the simple sounds of some orchestral
instruments! a. The Trautoniums, Thercmins, and lV[ artenots, all pre-
World War II attempts, prove it. b. Since the war, all "electronic" music
has also failed, in spite of the big hopes of the fifties, to pull electro-acoustie
music out of its cradle of the so-called electronic pure sounds produced by
frequency generators. Any electronic music based Oil such sounds only, is
marked by their simplistic sonority, which resembles radio atmospherics or
heterodyning. The serial system, which has becn used so much by electronic
music composers, could neit by any means improve the result, since it itself
is much too elementary. Only when the "pure" electronic sounds were
framed by other "concrete" sounds, which were much richer and much
more interesting (thanks to E. Varese, Pierrc Schaeffer, and Pierre Henry),
could electronic music become really powerful. c. The most recent attempts
to usc the flower of modern technology, computers coupled to converters,
have shown that in spite of some relative successes [26], the sonorous results
are even less interesting than those made ten years ago in the classic
electro-acoustic studios by means of frequency generators, filters, modula-
tors, and reverberation units.
In line with these critiques, what are the causes of these failures? In
my opinion the following are some of them:
1. Meyer-Eppler's studies [I] have shown that the spectral analysis
of even the simplest orchestral sounds (they will form a reference system for
a long time to come) presents variations of spectral lines in frequency as
well as in amplitude. But these tiny (second orderj variations are among
those that make the difference between a lifeless sound made up of a sum of
harmonics produced by a frequency generator and a sound of the same sum
of harmonics played on an orchestral instrument. These tiny variations,
which take place in the permanent, stationary part of a sound, would
certainly require new theories of approach, using another functional basis
and a harmonic analysis on a higher level, e.g., stochastic processes, r>.1arkov
chains, correlated or autocorrelated relations, or theses of pattern and form
recognition. Even so, analysis theories of orchestral sounds [27] would
result in very long and complex calculations, so that if we had to simulate
such an orchestral sound from a computer and from harmonic analysis on a
first level, we would need a tremendous amount of computer time, which is
impossible for the moment.
2. It seems that the transient part of the sound is far more important
than the permanent part in timbre recognition and in music in general [28].
N ow, the more the music moves toward complex sonorities close to "noise,"
the more numerous and complicated the transients become, and the more
their synthesis from trigonometric functions becomes a mountain of diffi-
culties, even more unacceptable to a computer than the permanent states.
It is as though we wanted to express a sinuous mountain silhouette by using
portions of circles. In fact, it is thousands of times more complicated. The
intelligent ear is infinitely demanding, and its voracity for information is
far from having been satisfied. This problem of a considerablc amount of
calculation is comparable to the 19th-century classical mechanics problem
that led to the kinetic gas theory.
3. There is no pattern and form recognition theory, dependent on
harmonic analysis or not, that would enable us to translate curves synthe-
sized by means of trigonometric functions in the perception of forms or
configurations. For instance, it is impossible for us to define equivalence
classes of very divcrsified oscilloseope curves, which the ear throws into the
same bag. Furthermore, the ear makes no distinction between things that
actual acoustic theories differentiate (e.g., phase differences, differential
sensitivity ability), and vice versa.
The Wrong Concept of Juxtaposing Finite Elements
Perhaps the ultimate reason for such difficulties lies in the improvised
entang-Iement of notions of finity and infinity. For example, in sinusoidal
osciIl;tion therc is a unit clement, the variation included in 27T. Then this
finite variation is repeated endlessly. Seen as an economy of mcans, this
procedure can be onc of thc possible optimizations. \'Ve labor during a
limited span oftimc (one period), then repeat the product indefinitely with
almost no additional labor. Basically, therefore, we have a mechanism
(e.g., the sine function) engendering a finite temporal object, which is
repeated for as long as we wish. This long object is now considcred as a new
clcment, to which wc juxtapose similar ones. The odds are that one can
draw any variation of one variable (e.g., atmospheric pressure) as a function
of time by mcans of a finite supcrposition (sum) of the preceding clements.
In doing this we expect to obtain an irregular curve, with increasing irregu-
larity as we approach "noises." On the oscilloscopc such a curve would
look quite complex. If we ask the eye to recognize particular forms or
symmetries on this curve it would almost certainly be unable to make any
judgment from samples lasting say 10 microseconds because it would have to
follow them too fast or too slowly: too fast for the everyday limits of visual
attention, and too slow for the TV limits, which plunge the instantaneous
judgment into the level of global perception of forms and colors. On the
other hand, for the same sample duration, the ear is made to recognize
forms and patterns, and therefore senses the correlations between fragments
of the pressure curve at various levels of understanding. \Ve ignore the
laws and rules of this ability of the car in the more complex and general
cases that we arc interested in. However, in the case in which we superpose
sine curves, we know that below a certain degree of complexity the car
disentangles the constituents, and that above it the sensation is transformed
into timbre, color, power, movement, roughness, and degree or disorder;
and this brings us into a tunnel of ignorance. To we expect that
by judiciously piling up simple clements (pure sounds, sine fi.tl1ctions) we
will create any desired sounds (pressure curve), cven those that come close
to very strong irregularities-almost stochastic ones. This sallle statemcnt
holds even when the unit clement of the iteration is taken li'om a function
246
Formalized l\1usic
other than the sine. In general, and regardless of the specific function of the
unit element, this procedure can be called synthesis by finite Juxtaposed elements.
In my opinion it is from here that the deep contradictions stem that should
prevent us from using it. *
NEW PROPOSAL IN MICROCOMPOSITION BASED ON
PROBABILITY DISTRIBUTIONS
vVe shall raise the contradiction, and by doing so we hope to open a
new path in microsound synthesis that without pretending
to be able to simulate already known sounds, will nevertheless launch
music, its psychophysiology, and acoustics in a direction that is quite inter-
esting and unexpected.
Instead of starting from the unit clemen t concept and its tircless iteration
and from the increasing irregular superposition of such iterated unit ele-
ments, we can start from a disorder concept and then introduce means that
would increase or reduce it. This is like saying that we take the inverse road:
We do not wish to construct a complex sound edifice by using discontinuous
unit clements (bricks = sine or other functions); we wish to construct
sounds with continuous variations that are not made out of unit elements.
This method would use stochastic variations of the sound pressure directly.
We :an imagine the pressure variations produced by a particle capriciously
movmg around equilibrium positions along the pressure ordinate in a non-
deterministic way. Therefore we can imagine the use of any "random
walk" or multiple combinations of them.
Method 1. Every probability function is a particular stochastic varia-
tion, which has its own personality (personal behavior of the particle). We
thcn use any o.ne of them. They can be discontinuous or continuous; e.g.,
POlss.on, exponential (ee - ex), normal, uniform, Cauchy (t [77(1 2 + x2)] -1),
arcs1l1 (7T-
1
[x(l-x)]-1/2),logistic [(ae-ax-B)(l +e-aX-p)-l] distributions.
Method 2. Combinations of a random variable X with itself can be
established. Example: If f(x) is thc probability function of X we can form
Sn. = .Xl + X2 + ... + Xn (by means of the n-fold convolution of f(x)
WIth Itself) or P K = Xl' X2 X
K
, or any linear, polynomial, ... , function
of the variable X.
* In spite of this .criticism I would like to draw attention to the magnificent manipu-
latory language MUSIC V of Max V. Mathews, which achieves the final step in this
procedure and automates it [29]. This language certainly represents the realization of
Ihe dream of an electronic music composer in the fifties.
Method 3. The random variables (pressure, time) can be functions of
other ,ariables (clastic forces), even of random variables. Example: The
pressure variable x is under the influence of a centrifugal or centripetal
force rf;(x, I). For instance, if the particle (pressure) is influenced by a force
wx (w being a constant) and also obeys a Wiener-Levy process, then its
density will be
where x and yare the values of the variable at the instants 0 and I,
respectively. (This is also known as the Ornstein-Uhlenbcck process.)
Alethod 4. The random variable moves bctween two reflecting (elastic)
barriers. Example: If we again have a Wiener-Levy process with two
reflecting barriers at a > 0 and zero, then the density of this random
walk will be
oo
(211"t)-1/2 I (exp [-(y x + 2ka)2f2tJ
k=O
+ exp [ - (y + x + 2ka)2f2t]),
where x and yare the values of the variables at the instants 0 and t,
respectively, and k = 0, I, 2, ....
lviethod 5. The parameters of a probability function can be considered
as variables of other probability functions (randomization, mixtures) [30].
Examples:
a. 1 is the parameter of a Poisson distributionf(k) = (at)k(k!) -Ie-at,
and the random variable of the exponential density g(t) = /3e-
Bt
The
combination is
which is a geometric distribution.
b. p and q are the probabilities of a random walk with jumps I
(Bernoulli distribution). Thc time intervals between successivc jumps arc
random variables with common density e - t (Poisson distribution). Then the
probability of the position n at instant t will beln(t) = In(2Iv'pq)e-t(pfq)n/2,
where
00
In(x) = I + n + I)] -1(x}2)2k+n
k=O
is the modified Bessel function of the first kind of order II.
248
Formalized
,:1ethod 6. Line.ar, polynomial, ... , combinations of probability
functlOns.A are consIdered as well as composite functions (mixtures of a
family of distributions, transformations in Banach space, subordination,
etc.).
a. If A and B are any pair of inten"als on the line, and Q(A, B) =
pr?b {X E A, Y E B} with q!x, B) = prob {X = x, Y E B} (q, under appro-
pnate regularity conditions being a probabilitv distribution in B for a
given x and a continuous function in x for a fixed B; that is, a conditional
probability of the event {Y E B}, gi\"Cn that X = x), and ,u.{A} is a probability
distribution of x-t=' A, then Q(A, B) = L q(x, B),u.{dx} represents a mixture
the of distributions q(X, B), which depcnds on the parameter x,
WIth ,u. servll1g as the distribution of the randomized parameter [30].
h. Interlocking probability distributions (modulation). If 11,12, ... ,
1n are the probability distributions of the random \"ariables Xl X2 "
xn, respectively, then we can form ' , "
n
and sn( 2:
i =1
or
n
and pn(n
p
\) =Pil. P;2"' P;,,,
k=l
or any combination "(functional or stochastic) of these sums and products.
the" al yk could be generated by either independent
fun.ctlOns, mclependent stochastic processes, or interrelated
determmed or mdetermincd processes. In some of these cases we would
the of renewal processes, if, for instance, the ai were considered
waIting times Ti. From another point of view, some of these cases would
also to the time series analysis of statistics. In reality, the ear
seems to reahze such an analysis when in a given sound it recognizes th
tone pitch togethe,r with timbre, fluctuation, or
of that sound! In fact, time series analysis should ha\"e been
IIlvented by composers, if they had-.
.c. Subordination [30). Suppose {X( t)}, a i\Iarko\"ian process with
continuous transition probabilities
Qt(x, r) = prob {X(T(t + 5)) E rIX(T!s)) = x}
kernel independent of 5), and {T!t)}, a process with non-
negatIve lI1dependent increments. Then {X( T( t))} IS a process
New Proposals in Structure 249
with transition probabilities
where U
1
is the infinitely di\"isible distribution of T(t). This P
t
is said to be
subordinated to {X (t)}, using the operational time T (t) as the directing process.
.tIethod 7. The probability functions can be filed into classes, that is,
into parent cun"e configurations. These classes are then considcred as
elements of higher order sets. The classification is obtained through at
least three kinds of criteria, which can be interrelated: a. analytical source
of derived probability distribution; gamma, beta, ... , and related densi-
ties, such as the density of X
2
with n degrees of freedom (Pearson) ; Student's
I density; density; b. other mathematical criteria, such as
stability, infinite di\"isibility; and c. characteristic features of the curve
designs: at le\"C1 0, where the \"alues of the random \"ariable are accepted
as such; at le\"el 1, whcre their values are accumulated, etc.
Macrocomposition
,\fethod 8. Further manipulations with classes of distributions envisaged
by 7 introduce us to the domain of macrocomposition. But we
will not continue these speculations since many things that ha\"e been ex-
posed in the preceding chapters could be used fruitfully in ob\"ious ways.
For example, sound molecules produced by the above methods could be
injected into the ST(ochastic) program of Chap. V, the program forming
the macrostructure. The same could be said about Chaps. II and III
processes at a macrob:el)' As for Chaps. VI and VIII (sym-
bolic music and group organization) establishing a complex microprogram
is not as easy, but it is full of rich and unexpccted possibilities.
All of the abo\"e new proposals are being investigated at the Centers
for :'fathematical and Automated l'.Iusic at both the School of
of Indiana U ni\"ersity, Bloomington, Indiana, and the Nuclear
Research Center of thc College dc France, in Paris. Digital to analogue
conYerters with 16 bits resolution at a rate of 0.5 10
5
samples per second
are available in both places.
Figs. IX, 1-8 were calculated and plotted at the Research Computing
Center of Indiana Uni\"Crsity under the supen"ision of Cornelia Colyer.
These graphs could correspond to a sound duration of 8 milliseconds, the
ordinates being the sound pressures.
250
Formalized Music
New Proposals in Microsound
251
-=:=:;: '
I '
J I
.7;'-' I
J. . ~ I
r I
-ff-- I
- ~ ~ ~ : > - . - I
iEy .
\ I
-J +
Fig. IX-2. Exponential x Cauchy Densities with Barriers and Randomized
Time
t
;
f

i
IX-3
IX-4 I
IX-5
-
I
:;:

-C
I
!
IX-6
44i ';

1
2 - """""-

-.2
5-

7
L>
"-----=

--:s=
L

,----=
T I

"

c:=
1-
"?-

-::jl;
1L

=-
r
Fig.IX-3.
Exponential x Cauchy Densities with Barriers and Randomized
Fig. IX-5. Hyperbolic Cosine x Exponential Densities with Barriers and
Time
Determined Time
Fig.IX-4.
Exponential x Cauchy Densities with Barriers and Randomized
Fig. IX-6. Hyperbolic Cosine x Exponential Densities with Barriers and
Randomized Time
Time
IX-7

Fi
9
d
D
IX
-
7
.. Hyperbolic Cosine x Exponential x Cauchy with Barriers
an etermlned Time
IX-S. Logistic x Exponential Densities with Barriers and Randomized
Chapter X
Concerning Time, Space and Music*
WHAT IS A COMPOSER?
A thinker and plastic artist who expresses himself through sound beings.
These two realms probably cover his entire being.
A few points of convergence in relation to time and space between the
sciences and music:
First point:
In 1954, I introduced probability theory and calculus in musical
composition in order to control sound masses both in their invention and in
their evolution. This inaugurated an entirely new path in music, more global
than polyphony, serialism or, in general, "discrete" music. From hence came
stochastic music. I will come back to that. But the notion of entropy, as
formulated by Boltzmann or Shannon,] became fundamental. Indeed, much
like a god, a composer may create the reversibility of the phenomena of
masses, and apparently, invert Eddington's "arrow of time."2 Today, I use
probability disu'ibutions either in computer generated sound synt]lesis on a
micro or macroscopic scale, or in instmmental compositions. But the laws of
probability tllat I use are often nested and vary with time which creates a
*Excerpts of Chapter X originally appeared in English in Perspectives of New Music,
Vol. 27, N 1. Those excepts appeared originally in French in RedecOllV1il' Ie Tem/)s,
Editions de l'Universite de Bruxelles, 1988, Vol. 1-2.
255
stochastic dynamics which is aesthetically interesting. This procedure is akin
to the mathematical analysis of Liouville's equation on non-unitary
transformations proposed essentially by I. Prigogine;3 namely, if the
microscopic entropy M exists, then M = A2, where A acts on the distribution
function or the density matrix. A is non-unitary which means that it does not
maintain the size of probabilities of the states considered during the evolution
of the dynamic system, although it does maintain the average values of those
which can be observed. This implies the irreversibility of the system to the
equilibrium state; that is, it implies the irreversibility of time.
Second point:
This point has no obvious relationship to music, except that we could
make use of Lorentz-Fitzgerald and Einstein transformations in the
macroscopic composition of music.
4
I would nevertheless like to make some
comments related to these transformations.
We all know of the special theory of relativity and the equations of
Lorentz-Fitzgerald and Einstein, which link space and time because of the
finite velocity of light. From this it follows that time is not absolute. Yet time
is always there. It "takes time" to go from one point to another in space, even
if that time depends on moving reference frames relative to the observer.
There is no instantaneous jump from one point to another in space, much
less "spatial ubiquity"-that is, simultaneous presence of an event or an
object in two sites in space. On the contrary, one posits the notion of
displacement. Within a local reference frame, what then does displacement
signify? If the notion of displacement were more fundamental than that of
time, one could undoubtedly reduce all macro and microcosmic
transformations to extremely short chains of displacement. Consequently
(and this is an hypothesis that I freely advance), if we were to adhere to
quantum mechanics and its implications accepted now for decades, we would
perhaps be forced to admit the notion of quantified space and its corollary,
quantified time. But then, what could a quantified time and space signify, a
time and space in which contiguity would be abolished? What would tile
pavement of the universe be if there were gaps between the paving stones,
inaccessible and filled with nothing? Time has already been proposed as
having a quantic stmcture by T. D. Lee of Columbia University.
Let us return to the notion of time considered as duration. Even after
the experimental demonstration of Yang and Lee which has abolished tile
parity symmetry P,5 it seems that the CPT theorem still holds for tile
symmetries of the electron (C) and of time (T), symmetries that have not yet
Concerning Time, Space and Music 257
been completely annulled. This remains so even if the "arrow of time"
appears to be nonreversible in certain weak interactions of particles. We
might also consider the poetic interpretation of Feynman,6 who holds that
when a positron (a positively charged particle created simultaneously with an
electron) collides with an electron, there is, in reality, only one electron
rather than three elementary particles, the positron being nothing but the
temporal retrogression of the first electron. Let us also not forget the theory
of retrograde time found in Plato's Politicos-or in the future contraction of
the universe. Extraordinary visions!
Quantum physics will have difficulty discovering the reversibility of
time, a theory not to be confused with the reversibility of Boltzmann's "arrow
of entropy." This difficulty is reflected in the explanations that certain
physicists are attempting to give even today for the phenomenon called the
"delayed choice" of the two states--corpuscular or wave -of a photon. It has
been proven on many occasions that the states depend entirely on
observation, in compliance with the theses of quantum mechanics. These
explanations hint at the idea of an "intervention of the present into the past,"
contrary to the fact that casuality in quantum mechanics cannot be inverted.
For, if the conditions of observation are established to detect the particle,
then one obtains the corpuscular state and never the wave state, and vice
versa. A similar discussion on non-temporality and the irreversibility of the
notion of causality was undertaken some time ago by Hans Reichenbach.'
Another fundamental experiment has to do with the correlation of the
movement of two photons emitted in opposite directions by a single atom.
How can one explain that both either pass through two polarizing films, or
that both are blocked? It is as if each photon "knew" what the other was doing
and instantaneously so, which is contrary to the special theory of relativity.
Now, this experiment could be a starting point for the investigation of
more deeply seated properties of space, freed from the hltelage of time. In
this case, could the "nonlocaIity" of quantum mechanics perhaps be explained
not by the hypothesis of "hidden variables" in which time still intervenes, but
rather by the unsuspected and extravagent properties of nontemporal space,
such as "spatial ubiquity," for example?
Let us take yet one more step. As space is perceptible only across the
infinity of chains of energy transformations, it could very well be nothing but
an appearance of these chains. In fact, let us consider the movement of a
photon. Movement means displacement. Now, could this displacement be
considered an autogenesis of the photon by itself at each step of its trajectory
258
Formalized Music
(continuous or quantized)? This continuous auto- creation of the photon,
could it not, in fuct, be space?
Third pOint: Case of creating something from nothing
In musical composition, construction must stem from originality which
can be defined in extreme (perhaps inhuman) cases as the creation of new
rules or laws, as far as that is possible; as fur as possible meaning original, not
yet known or even forseeable. Construct laws therefore from nothing, since
without any causality.
But a construction from nothing, therefore totally engendered, totally
original, would necessarily call upon an infinite mass of rules duly entangled.
Such a mass would have to cover the laws of a universe different from our
own. For example: rules for a tonal composition have been constructed. Such
a composition therefore includes, a priori, the "tonal functions." It also
includes a combinatory conception since it acts on entities, sounds, as defined
by the instruments. In order to go beyond this slight degree of originality,
other functions would have to be invented, or no functions should exist at all.
One is therefore obliged to conceive of forms from thoughts bearing no
relation to the preceeding ones, thoughts without limits of shapes and
without end. Here, we are obliged to progressively weave an unlimited web
of entangled rules-and that alone in the combinatory realm which itself
excludes, by definition, any possible continuums of sound. However, the
insertion of continuity will consequently augment the spread of this web and
its compacity. Furthermore, if one cared to engender the unengenderable in
the realm of sound, then it would be necessary to provide rules other than
those for sound machines such as pipes, strings, skins, etc. which is possible
today thanks to computers and corresponding technologies. But technology is
both but a semblance of thought and its materialisation. It is therefore but an
epiphenomenon in this discussion. Actually, rules of sound synthesis such as
those stemming from Fourier series should not be used any more as the basis
of construction. Others, different ones, must be formulated.
Arwther perspective: We have seen how construction stems from an
originality which is defined by the creation of rules and laws outside of an
individual's or even the human species' memory. However, we have left aside
the notion of rules or laws. Now the time has come to discuss this notion. A
rule or law signifies a finite or infinite procedure, always the same, applied to
continuous or "discrete" elements. This definition implies the notion of
repetition, of recurrence in time, or symmetry in realms outside time (hors
temps). Therefore, in order for a rule to exist, it must be applicable several
times in eternity's space and time. If a rule were to exist but once, it would be
swallowed up in this immensity and reduced to a single point, therefore
unobservable. In order for it to be observable, it must be repeatable an
infinite number of times.
Subsidiary question: Can one repeat a phenomenon? (cf. Herakleitos: "It
is impossible to step twice into the same river," and Kratylos: "not even
once.")
But the fact remains that the universe:
a) seems, for the time being, to be made up of rules-procedures;
b) that these rules-procedures are recurrent.
It is as though the Being (in disagreement with Parmenides), in order to con-
tinue existing, is obliged to die; and once dead, is obliged to start his cycle
again. Existence, therefore, is a dotted line.
Can one, at last, imagine an infinitesmal microscopic rule that is
engendered from nothing? Even if physics has yet to discover anything
resembling this, despite "Lamb's shift" (which sees each point in space in our
universe as seething in virtual pairs of particles and anti-particles), we can
imagine such an eventuality which would, by the way, be of the same nature
as the fuct of pure chance, detached from any causality.
It is necessary to depend on such a conclusion of a Universe open to the
unprecedented which relentlessly would be formed or would disappear in a
truly creative whirlwind, beginning from nothingness and disappearing into
nothing. The same goes for the basis of art as well as for man's destiny.
Here, below, is the thesis of a few astrophysicians such as Edward
Tryon, Alexander Vilenkin, Alan Guth, Paul Steinhardt, adherents to the Big
Bang theory:
If grand unified theories are correct in their prediction that baryon
number is not conserved, there is no known conservation law that
prevents the observed universe from evolving out of nothing. The infla-
tionary model of the universe provides a possible mechanism by which
the observed universe could have evolved from an infinitesimal region.
It is then tempting to go one step further and speculate that the entire
Ulllverse evolved from literally nothing. (cf. ScientifIC American, May,
1984)
The multiplicity of such universes according to Linde
s
from Moscow is
also quite intriguing.
Here, below, is an alternative to the Big Bang scenario. These studies
have been followed by the physicists of the University of Bruxelles; namely R.
Brout, E. Giinzig, F. Englert and P. Spindel:
Rather than the Universe being born of an explosion, they propose that
it appeared ex-nihilo foIlowing an instability of the minkonskian quan-
tum void, meaning that space-time was devoid of any matter, therefore
flat or yet-without any curvature." (cf. Coveney, Peter V.,
"L'irreversibilite du temps," La Recherche, Paris, February, 1989).9
What is extraordinary is that both propositions, Big Bang or not, admit
a beginning, an origin from notlJing, or nearly nothing witlJ, however, cycles
of re-creation! With a most extreme modesty, I would like to compare,
especially tlJe last hypothesis, witlJ a scientific-musical vision I had made in
1958. At tllat time, I wanted to do away with all of tlJe inherited rules of
composition in order to create new ones. But ilie question that came to my
mind at iliat time was whether a music could still have meaning even ifit was
not built on rules of occurence. In oilier words, void of rules. Below are the
steps in this thought process:
"For it is the same thing to think and to be"
(The Poem, Parmenides)
and my paraphrase
"For it is the same thing not to be and to be"
Ontology:
In a Universe of Void. A brief train of waves whose beginning and end coin-
cide (nil Time), perpetually triggering off.
Nothingness resorbs, creates.
It is the generator of Being.
Time, Causality.
This text was first published in Gravesaner Bliitter, N 11/12, 1958, the
revue published by the great conductor, Hermann Scherchen. At tlJat time, I
had temporarily resolved tlJis problem in creating music uniquely tlJrough the
help of probability distributions. I say "temporarily" since each probability
function has its own finality and therefore is not a nothing. *
*Cf. also page 24 for a slightly different rendition of the same material (S.K.)
Concerning Time, Space and Music
261
Another question
The actual state of knowledge seems to be the manifestation of the
evolution of the universe since, let us say, some fifteen billion years.
IO
By that,
I mean that knowledge is a secretion of the history of humanity, produced by
this great lapse of time. Assuming this hypothesis, all that which our
individual or collective brain hatches as ideas, theories or know-how, is but
the output of its mental structures, formed by the history of the innumerable
movements of its cultures, in its anthropomorphic transformations, in the
evolution of the earth, in that of the solar system, in that of the universe. If
this is so, then we face a frightening, fundamental doubt as to the "true
objectivity" of our knowledge and know-how. For if, with bio-technologies
already developing, one were to transform these mental structures (our own)
and their heredity,therefore the rules for the functioning of the brain based
on certain premises today, on logic or systems of logic, and so on ... , if one
were to succeed in modifying them, one would gain, as ifby sort ofa miracle,
another vision of our universe, a vision which would be built upon theories
and knowledge which are beyond the realm of our present thought.
Let us pursue this thought. Humanity is, I believe, already on this paul.
Today, humanity, it seems to me, has already taken the first step in a new
phase of its evolution, in which not only the mutations of the brain, but also
the creation of a universe very different from that which presently surrounds
us, has begun. Humanity, or generalizing, the species which may follow it,
will accomplish this process.
Music is but a path among others for man, for his species, first to
imagine and then, after many, many generations, to entail this existing
universe into another one, one fully created by man. Indeed, if man, his
species, is the image of his universe, then man, by virtue of the principle of
creation from nothingness and disappearance into nothingness (which we are
forced to set), could redefine his universe in harmony with his creative
essence, such as an environment he could bestow upon himself.
IN MUSIC
In the following comments, the points of view on time are taken from
music in gestation or under observation. This is not to say that my preceding
comments do not concern tlJe musician. On the contrary, ifit is incumbent on
music to serve as a medium for the confrontation of philosophic or scientific
ideas on tlJe being, its evolution, and tlJeir appearances, it is essential that tlIe
composer at least give some serious thought to tlJese types of inquiry.
262
Formalized Music
Furthermore, I have deliberately not approached the psychological
apprehension of time from higher levels, for example, the effects of the
temporal dynamic experience while listening to a symphony or to electronic
music.
What is time for a musician? What is We flux of time which passes
invisibly and impalpable? In truth, we seize it only with the help of perceptive
reference-events, thus indirectly, and under the condition that these
reference-events be inscribed somewhere and do not disappear without
leaving a trace. It would suffice that they exist in our brain, OUT memory. It is
fundamental that the leave a trace in my memory, for
if not, they would not exist. Indeed, We underlying postulate is that time, in
the sense of an impalpable, Heraclitian flux, has signification only in relation
to the person who observes, to me. Otherwise, it would be meaningless. Even
assuming the hypothesis of an objective flux of time, independant from me,
its apprehension by a human subject, thus by me, must be subject to the
phenomena-reference of the flux, first perceived, then inscribed in my
memory. Moreover, this inscription must satisfy tile condition that it be in a
manner which is well circumscribed, well detached, individualized, without
possible confusion. But tllat docs not suffice to transform a phenomenon that
has left traces in me into a referential phenomenon. In order tlIat this
trace-image of the phenomenon become a reference mark, the notion of
anteriority is necessary. But this notion seems to be circular and as
impenetrable as the immediate notion of flux. It is a synonym. Let us alter
our point of view, if only slightly. When events or phenomena are
synchronic, or rather, if all imaginable events were synchronic, universal
time would be abolished, for anteriority would disappear. By the same token,
if events were absolutely smooth, without beginning or end, and even
without modifications or "perceptible" internal roughness, time would
likewise find itself abolished. It seems that the notion of separation, of
bypassing, of difference, of discontinuiity, which are strongly interrelated, are
prerequisite to the notion ofanteriority. In order for anteriority to exist, it is
necessary to be able to distinguish entities, which would then make it possible
to "go" from one to the other. A smooth continuum abolishes time, or rather
time, in a smooth continuum, is illegible, inapproachable. Continuum is thus
a unique whole filling both space and time. We are once again coming back
to Parmenides. Why is space included among those things that are illegible?
Well, because of its non-roughness. Without separability, tllere is no
extension, no distance. The space of the universe would find itself condensed
into a mathematical point without dimensions. Indeed, Parmenides' Being,
which fills all space and eternity, would be nothing but an absolutely smooth
"mathematical point."
Let us get back to the notion of separability, first in time. At the least,
separability means non-synchronisation. We discover once again the notion
of anteriority. It merges with the notion of temporal ordering. The ordering
anterioity admits no holes, no empty spaces. It is necessary for one separable
entity to be contiguous with the next, otherwise, one is subject to a confusion
of time. Two chains of contiguous events without a commmon link can be
indifferently synchronous or anterior in relation to each other; time is once
again abolished in the temporal relation of each of the universes represented
by the two chains. On the contrary, local clocks serve as chains without gaps,
but only locally. Our biological beings have alSo developed local clocks but
they are not always effective. And memory is a spatial translation of the
temporal (causal) chains. We will come back to this.
I have spoken of chains without gaps. At the moment and to my
knowledge, local gaps have not yet been discovered in sub-atomic physics or
in astrophysics. And in his theory of the relativity of time, Einstein tacitly
accepts this postulate of time without gaps in local chains, but his theory also
constructs special chains without gaps between spatially separable localities.
Here, we are definitely not concerned with the reversibility of time which was
partially examined above in light of recent discoveries in sub-atomic physics,
for reversibility would not abolish time.
Let us examine the notion of separability, of discontinuity in space. Our
immediate consciousness (a mental category?) allows us to imagine separated
entities which, in turn, necessitate contiguity. A void is a unity in this sense,
contrarily to time, in which our inherited or acquired mental notions bar us
from conceiving the absence of time, its abolition, as an entity sharing time,
the primordial flux. Flux either is, or is not. We exist, therefore it is. For the
moment, one cannot conceive of the halting of time. All this is not a
paraphrase of Descartes or better yet, of Parmenides: it is a presently
impassable frontier. (But certainly, by using Parmenides once more, passable:
"TO rAP AYrO NOEIN TE KAI EINAI").
To get back to space, the void can be imagined as a dwindling of tllC
entity (phenomenon) down to an infinitesimal tenuousness, having no
density whatsoever. On the other hand, to travel from one entity to anotller is
a result of scale. If a person who voyaged were small, tlle person would not
encompass the totality of entities, the universe at once. But if Illis person's
scale were colossal, Wen yes. The universe would offer itself in one stroke,
with hardly a scan, as when one examines Ille sun from afar.
264
Formalized Music
The entities would appear, as in a snapshot, reunited in a dense
network of nontemporal contiguities, uninterrupted, extending through the
entire universe. I said, in a snapshot. This is to say that in the snapshot, the
spatial relations of the entities, the forms tllat their contiguities assume, the
structures, are essentially outside time (hors-temps). The flux of time does
not intervene in any way. 'Tbat is exactly what happens with the traces that
the phenomenal entities have left in our memory. Their geographical map is
outside time.
Music participates both in space outside time and in the temporal flux.
Thus, the scales of pitch; the scales of the church modes; the morphologies of
higher levels; structures, fugal architectures, mathematical formulae
engendering sounds or pieces of music, these are outside time, whether on
paper or in our memory. The necessity to cling against the current of the
river of time is so strong that certain aspects of time are even hauled out of it,
such as the durations which become commutable. One could say that every
temporal schema, pre-conceived or post-conceived, is a representation
outside time of the temporal flux in which the phenomena, the entities, are
inscribed.
Due to the principle of anteriority, the flux of time is locally equipped
with a structure of total order in a mathematical sense. That is to say that its
image in our brain, an image constituted by the chain of successive events,
can be placed in a one-to-one correspondance with the integers and even,
with the aid of a useful generalization, with real numbers (rational and
irrational). Thus, it can be counted. This is what the sciences in general do,
and music as well, by using its own clock, the metronome. By virtue of this
same structure of total order, time can be placed in a one-to-one
correspondence with fue points of a line. It can fuus be drawn.
This is done in fuc sciences, but also in music. One can now design
temporal architectures-rhythms-in a modern sense. Here is a tentative
axiomatization of the temporal structures placed outside oftime:
1. We perceive temporal events.
2. Thanks to separability, these events can be assimilated to landmark
poinls in the flux oftime, points which are instantaneously
hauled up outside oftime because of their trace in our memory.
3. The comparison oftbe landmark points allows us to assign to them
distances, intervals, durations. A distance, translated spatially,
can be considered as the displacement, the step, the jump from
one point to another, a nontemporal jump, a spatial distance.
4. It is possible to repeat, to link together these steps in a chain.
5. There are two possible orientations, one by an accumulation of
steps, the other by a de-accumulation.
From here, we can construct an object which can be represented by points on
a line, evenly spaced and symbolized by the numeral 1 with index zero: 10 =
( ... , -3,-2,-1,0,1,2,3, ... ). This is the regular rhythm, corresponding to the whole
numbers. As the size of the step is not defined in the preceding propositions
(recalling Bertrand Russell's observation concerning Peano's axiomatic of
natural numbers
ll
), we can affix to the preceding object the following objects
which I call "sieves," by using solely proposition 4:
20 = { ... ,-4,-2,O,2,4,6, ... } or 2} = { ... ,-3,-1,1,3,5, ... } or
3
0
= { ... ,-3,O,3,6,9, ... } or 3} = { ... ,-5,-2,1,4,7, ... } or
3
2
= { ... ,-4,-1,2,5,8, ... } etc ...
From these objects and their modular nature, and with the help of these three
logical operations:
U union, disjunction ex. 20 U 2} = 10
n intersection, conjunction ex. 20 n 2} =
complementarity, negation ex. 20 = 2}
we can construct logical functions L-that is to say, very complex rhythmic
architectures which can even go as far as a random-like distribution of points
on a line-if the period is sufficiently long. The interplay between complexity
and simplicity is, on a higher level, another way of defining the landmark
points, which certainly plays a fundamental role in aesthetics, for this play is
juxtaposed with the pair release/tension.
Example of a logical function L:
L=(Mkn ~ nP1)U (N,U Q,U ... Ty)U ...
The upper-case letters designate moduli and the subscripts designate shifts in
relation to a zero point of reference.
Up to this point, we have examined time perceived by means of our
faculties of attention and conscious thought-time on the level of forms and
stTuctures of an order ranging from tens of minutes to approximately one
twenty-fifth of a second. A stroke of the bow is a refcrrential event that can
define durations of a fraction of a second. Now, there are some subliminal
events found on several even lower levels. Such an example is that of the
temporal segmentation produced by a very choppy amplitude envelope on
the sound of an unvarying sinusoidal wave form. If the duration of the note is
long (about one minute), we perceive the rhythms of the beats as pleasant.
moving vibratos. If the duration is relatively short (three seconds), the ear
and the brain interpret it as a timbre. That is to say that the result of
subliminal, unconscious counting is different in nature and is recognized as
timbre.
Let us take a brief moment to consider the mechanism of the internal
ear coupled with the brain which recognizes the wave form-that is to say,
the timbre-and the frequency of a sound. On the one hand, it seems that the
points of deformation of the basilar membrane playa fundamental role in
the recognition; but, on the other hand, a sort of temporal Morse code of
electrical discharges of neurons is taken statistically into account for the
detection of tone. A remarkably complex subliminal counting of time is
taking place. But knowledge of acoustics in this domain is still very limited.
On this subliminal level, here is another disconcerting phenomenon. It
is the result of a new theory on the synthesis of computer sounds which
circumvents the hanronic synthesis of Fourier, practiced everywhere today, a
theory which I introduced now more than fifteen years ago.
12
It is a question
of beginning with any form whatsoever of an elementary wave, and with each
repetition, of having it undergo small deformations according to certain
densities of probabilities (Gauss, Cauchy, logistic, ... ) appropriately chosen and
implemented in the form of an abstract black box. The result of these
deformations is perceptible on all levels, microstructure (= timbre),
ministructure (= note), mesostructure (= polyrhythm, melodic scales of
intensities), macrostructure (= global evolution on the order of some tens of
minutes).
If the rate of sampling had been 1,000,000 or 2,000,000 samples per
second instead of approximately 44,100 (commercial standard), one would
have had an effect of sounding fractals, with a sonorous effect which is
impossible to predict.
We see to what extent music is everywhere steeped in time: (a) time in
the form of an impalpable flux or (b) time in its frozen form, outside time,
made possible by memory. Time is the blackboard on which are inscribed
phenomena and their relations outside the time of the universe in which we
live. Relations imply architectural structures, rules. And, can one imagine a
rule without repetition? Certainly not. I have already treated this subject.
Besides, a single event in an absolute eternity of time and space would make
Concerning Time, Space and Music
267
no sense. And yet, each event, like each individual on earth, is unique. But
this uniqueness is the equivalent of death which lies in wait at every step, at
every moment. Now, the repetition of an event, its reproduction as faithfully
as possible, corresponds to this struggle against disappearance, against
nothingness. As if the entire universe fought desperately to hang on to
existence, to being, by its own tireless renewal at every instant, at every
death. The union of Parmenides and of Heraclitus. Living species are an
example of this struggle of life or death, in an inert Universe launched
perhaps by the Big Bang (is it really inert, that is, without any changes in its
laws?). This same principle of dialectical combat is present everywhere,
verifiable everywhere. Change-for there is no rest-the couple death and
birth lead the Universe, by duplication, the copy being more or less exact.
The "more or less" makes the difference between a pendular, cyclic Universe,
strictly determined (even a deterministic chaos), and a nondetermined
Universe, absolutely unpredictable and chaotic. Unpredictability in thought
obviously has no limits. On a first approach it would correspond to birth from
nothingness, but also to disappearance, death into nothingness. At dIe
moment, the Universe seems to be midway between these two chasms,
something which could be the subject of another study. This study would deal
with the profound necessity for musical composition to be perpetually
original-philosophiocally, technically, aesthetically.13
In what follows and as a consequence of the preceeding axioms, we will
study in greater detail the practical questions of how to create a sieve (=
series of points on a line), beginning from a logical function of moduli
(periods), or inversely, from a series of points on a line, how to create a
logical function of moduli which should be able to engender the given series.
This time, we shall use series of "pitches" taken from musical space.
Chapter XI
Sieves*
In music, the question of symmetries (spatial identities) or of periodicities
(identities in time) plays a fundamental role at all levels: from a sample in
sound synthesis by computers, to the architecture of a piece. It is thus neces-
sary to formulate a theory permitting the construction of symmetries which
are as complex as one might want, and inversely, to retrieve from a given
series of events or objects in space or time the symmetries that constitute the
series. We shall call these series "sieves."!
Everytl1ing that will be said here could be applied to any set of
of sound or of well-ordered sound structures, and especially, to
any group which entails an additive operation and whose elements are
multiples of a unity; that is to say that they belong to the set N of natural
numbers. For example: pitches, time-points, loudnesses, densities, degrees of
order, local timbres, etc. In the case of pitches, there must be a distinction
between sieve (scale) and mode. Indeed, tile white keys on a piano constitute
a unique sieve (scale) upon which are formed ilie "modes" of C major, D, E,
G, A (natural minor), etc. Just like Indian ragas or Olivier Messiaen's modes
"of limited transpositions," modes are defined by cadential, harmonic, etc.
formulas.
But every well-ordered set can be represented as points on a line, as
long as a reference point is given for the origin and a length u for the unit
distance, and this is a sieve. Historically, the invention of tile well-tempered
*This chapter is scheduled to appear in a future issue of Perspectives of New Music.
John Rahn's personal contribution to the following material is most appreciated
(LX.)
268
Sieves 269
chromatic scale, attributed to the Renaissance, is of upmost importance since
it provided a universal standarization of the realm of pitches, as fertile as that
which already existed for rhythm. However, it should be remembered that
the first theoretical attempt towards such an approach which opened the path
to number theory in music dates back to Aristoxenas of Tarent, during the
IVth century B.C., in his "Harmonics."2
CONSTRUCTION OF A SIEVE
Starting from symmetries (repetitions), let us construct a sieve (scale).
As a melodic example, we shall construct the diatonic scale formed by the
white keys of the piano.
With u = one semitone = one millimeter and a zero reference point
taken arbitrarily on a note, for example e3, we can notate the diatonic sieve
(scale) on graph paper scaled to the millimeter, by means of points to the left
and to the right of this zero reference point with successive intervals counted
from left to right of 2, 2, 1, 2, 2, 2, 1,2, 2, 1,2, 2, 2, 1, ... millimeters, or we
can write the sieve in a logical-arithmetic notation as L = 120 U 122 U 124 U
125 U 127 U 129 U 12]] where 12 is the modulus of the symmetry (period) of
the octave with u for the semitone. This notation gives all the Cs, all the Ds, ...
all the Bs, considering that the moduli 12 repeat on both sides of the zero
reference point. The indices, 0, 2, 4, 5, 7, 9, 11 of !lIe modulus 12 signify
shifts to the right of the zero of tl1e modulus 12. They also represent !lIe
residue classes of congruence mod. 12.
With a different unit distance u, for example, a quartertone, one would
have tile same structure as the diatonic scale but the period of the series
would no longer be an octave, but an augmented fourth.
In a similar fashion, a periodic rhythm, for example 3, 3, 2:
J. J. J I J. J J I J J. J \ etc.
can be notated as L = 70 U 73 U 7
5
, In botil of these examples, the sign U is a
logical union (and/or) of the points defined by ilie moduli and their shiftings.
The periodicity of the diatonic sieve (scale) is external to the sieve itself
and is based on the existence of me modulus 12 (tlle octave). Its internal
symmetry can be studied in tile indices ] (shiftings, residue classes) of Ule
terms 121" But it would be interesting to give, when it exists, a more hidden
symmetry derived from .the decomposition of the modulus 12 into simplier.
moduli (symmetries, periodicities), such as 3 and 4, a decomposition which
would have the advantage of allowing a comparison among different sieves in
order to study the degree of their difference and to be able to define a notion
of distance in this way.
Let us take the elementary sieves 3
0
and 4
0
, In taking the points 3
0
and/or the points 4
0
, we obtain a series HI = ( ... , 0,3.4.6.8.9. 12. 15. 16.
18, 20. 21, 24, 27, 28, ... ) = 3
0
U 4
0
, and if C is the zero and u = one
semitone, HI becomes ( ... C, D#, E, F#, G#, A. C, D#, E, ... ). But if we take
the points common to 3
0
and 4
0
, we obtain the series H2 = ( ... , 0, 12, 24, 36.
... ) = 3
0
n 40 where the sign n is the logical intersection (and) of the sets of
points defined by these moduli and their respective shiftings.
Hence, we observe that the series H2 can be defined by the modulus 12
= 3 * 4 and by the logical expression L = 120 which gives the octaves. The
number 12 is the smallest common multiple of 3 and 4. which are coprime.
meaning their largest common denominator is 1.4
Let us imagine now the elementary sieves 20 and 6
0
, Then G
I
= 20 U = (
... ,0,2,4,6,8,10, 12, ... ) and the common points in G
2
= 20 n 6
0
= ( ... ,0,6,
12, 18, ... ). But here, the series is no longer made into octaves as in the
preceding case.
To understand this, let us take another example with elementary
moduli Ml = 6 and M2 = 15 which have been adjusted to the original. We
then form the pairs 6
0
= (MI, II) and 15
0
:: (M2,I2) with II = and 12 = 0
as indices.
The series of the union (MI, 11) U (M2, 12) = Kl will be KI = { ... , 6,
15, 18,24,30.36,42,45, ... } and their common points (the intersection) will
form the series (Ml, 11) n (M2, 12) = K2 where K2 = { .... 0, 30, 60, ... }. The
period is clearly equal to 30 and the largest common denominator D of 6 and
15 is 3 (which is, by multiplication, the part congruent to M I and M2) and the
smaIlest common multiple is M3, equals 30. Now, 6 divided by the largest
common denominator DisCI, equals 2; and 15 divided by the largest
common denominator D is C2, equals 5. Generalizing, the period of the
points common to the two moduli MI and M2 will be the smallest common
multiple M3 of these two moduli. So, (Ml, II) n (M2, 12) = (M3, 13) with 13
= 0, if 11 = 12 = and M3 = D * C 1 * C2, where Cl = MIl D and C2 = M2
ID.
It will also be noted that the operation oflogical union, notated as U, of
the two elementary moduli Ml and M2 is cumulative in that it takes into
Sieves 271
account the periodic points of both moduli simultaneously. On the other
hand, the logical operation of intersection, notated as n, is reductive since we
take only the points common to both moduli.
When we mix the points of several moduli MI, M2, M3, M4 ... :
a) by union, we obtain a sieve which is dense and complex
depending on the elementary moduli;
PI = (MI, 11) U (M2, 12) U (M3, 13) U ...
b) by intersection, we obtain a sieve which is more rarified than that
of the elementary moduli, and there would even be some cases
in which the sieve would be empty of points when it lacks
coincidences;
P2 = (MI, II) n (M2, 12) n (M3,13) n ...
c) by simultaneous combinations of the two logical operations, we
obtain sieves which can be very complex;
(0) L = {(MIl, Ill) n (M12,Il2) n ... } U {(M21, 121) n (M22,122)
n ... } U {( ... )}
k. k(i)
2, (IT)
i= I 1
The intersection of each set of pallS between curly brackets should furnish a
single final pair, if it exists. The final pairs will be combined by their union,
which will provide the desired sieve.
Now let us exami!'le tile rigorous formulation of tile calculation of (Jle
intersection of tile two moduli (Ml, 11) and (M2, 12) where the periods Ml
and M2 start from some II and 12 respectively. First II and 12 are reduced by
taking their moduli in relation to Ml and M2, II = MOD(ll, Ml) and 12 =
MOD(12, M2).5
The first coincidence will eventually appear at a distance:
(1) S = II + A * Ml = 12 + 0 * M2
where A and 0 are elements ofN, and ifMl = D * Cl and M2 = D >I< C2 wit.h
D equal to the largest common denominator, Cl and C2 being coprime, then
the period M3 of the coincidences will be: M3 = D * C 1 * C2. From (1) tJlere
follows:
II - 12 = (0 * D >I< C2) - (), * D * Cl) and
(11 - J2) I D = (0 * C2) - (..1. * Cl).
272
Formalized Music
Now, since the expression on the right of the equal sign is a whole
number, the expression on the left of tlle equal sign should also be a whole
number. But, if 11 - J2 is not divisible by D (for some Il, 12), then, there are
no coincidences and the intersection (MI, II) (M2, 12) will be empty. Ifnot:
(2) (II - 12)/D = W:L Nand 11' = cr * C2 -A * CI, as well as:
11' + A >I< C I = cr >I< C2.
But following Bachet de Meziriac's theorem (1624), in order for x and y to be
two coprimes, it is necessary and sufficient that there exist two relative whole
numbers, and such that:
(3) I + * x = >I< Y or

where and , come from the recursive equations:
(4) * C2, Cl) = 1 and
6
(5) , * CI, C2) = 1
while letting and , run through the successive values 0, I, 2, 3, ... (except if
CI = I and C2 = 1).
But since Cl and C2 are coprime, there follows from (2) and (3):
A / a = a 11Jl A / (-'II) = and
al(-'I') = " and if (MI, II) n (M2, 12) = (M3, 13), then
(6) 13 = MODI2 + * (II - 12) >I< C2), M3) or
13 = MODII + * (12 - II) * Cl), M3)
with M3 = D * CI * C2.
Example I: Ml = 60,11 = 18, M2 = 42,12 = 48, D = 6, Cl = 10, C2
= 7, M3 = 6 >I< 10 * 7 = 420, with Cl and C2 coprime.
From (3) and (4) we get: , = 5.
From (6) we get: 13 = MOD(I8 + 5>1< (48 -18) >I< 10,420) = 258.
Example 2: MI = 6, II = 3, M2 = 8, 12 =3, D = 2, CI = 3, C2 = 4, M3
= 24, with Cl and C2 coprime.
From (4) we get: = L
And from (6) we get: 13 = MOD3 + 1 >I< (3 - 3) * 4), 24) = 3; tllat is, in
the case that II = 12, then 13 = Il = 12, and here M3 = 24 and 13 = 3.
Take the preceding example but with 11 = 13 and 12 = 4, so I1 is not
equal to 12. Since II I D = 1.5, which is not an element of N, tllere are no
coincidences and M3 = 0 and 13 = O. But, if II = 2 and 12 = 16, and since (11
- 12) I D = 7 :L N, we obtain from (4) = I and from (6) 13 = MOD(O + 1 * (2
- 0) >I< 4, 24) = 8 and (M3, 13) = (24, 8).
Sieves
Computation of the Intersection (M1,11)n (M2,12) = (M3,13)
Are given: Ml, M2, 11, 12, with Ii = MOD(Ii, Mi) 0
D = the largest common denominator of M I and M2
M3 = tlle smallest common multiple ofMl and M2
CI = MIl D, C2 = M2/ D, M3 = D * CI * C2
Figure 1.
273
274
Formalized Music
To compute several simultaneous intersections (coincidences) from an
expression between brackets in the equation (0) of L, it suffices to calculate
the pairs in that expression two by two. For example:
kO = 4 k(i)
L= L (IT)
i=1
= {(3, 2) n (4, 7) n (6,11) n (8, 7)} U {(6, 9) n (15, IS)} U
{(13,5) n (8,6) n (4, 2)} U {(6,9) n (15, 19)}, with kO = 4.
For the first expression between brackets, we first do (3, 2) n (4, 7) = (12,
11), then, after modular reduction of the indices, (12,11) n (6,5) = (12, 11),
then (12, 11) n (8,7) = (24,23). We go on to the following brackets, and so 011.
Finally,
L = (24, 23) U (30, 3) U (104, 70) U (0,0) for ko = 4, k(l)
= 4, k(2) = 2, k(3) = 3, k(4) = 2.
Through a convenient scanning, this logical expression will provide us with
the points of a sieve constructed in the following fashion:
H = { ... 3, 23, 33, 47, 63, 70, 71, 93, 95,119,123,143,153,
167, ... 479, .. J}
with a period of P = 1560. The zero of this sieve wiuun Ule set of pitches
can be arbitrarily taken to be c-2 = 8.25 Hz and at ten octaves, (2]0 * 8.25 =
16384 Hz) wiUl u equal to the semitone. It will give us the notes #D.
2
, B.l> Ao, Bb
#D
3
, # As, B
3
, etc.
For the same zero taken to be C.
2
and for u to be equal to a qua!i:ertone,
Ule series gives us the notes '/If C.
2
, + B_2' + E_1' + B_1 +G
o
, B
o
, f/lf.Al> + B],
+ B
2
, : j l f ~ C
3
, etc.
Inverse case
Let us start from a series of points eitl1er given or constructed
intuitively and deduce its symmetries; that is to say, the moduli and tl1eir
shiftings (Mj, Ij), and construct tl1e logical expression L describing this series
of points. The steps to follow are:
a) each point is considered as a point of departure ( = In) of a
modulus.
b) to find Ule modulus corresponding to this point of departure, we
begin by applying a modulus of value Q = 2 unities. If each
one of its multiples meets a point which has not already been
encountered and which belongs to the given sieve, then we
Sieves 275
keep tl1e modulus and it forms tl1e pair (Mn, In). But if anyone
of its multiples happens not to correspond to one of tl1e points
of tl1e series, we abandon it and pass on to Q + 1. We proceed
so until each one oftl1e points in tl1e given series has been
taken into account.
c) if for a given Q, we garner an its points (Q, Ik) under another pair
(M, I); tl1at is, if the set (Q, Ik) is included in (M, I), tl1en, we
ignore (Q, Ik) and pass on to the following point I
k
+].
d) similarly, we ignore all the (Q, I) which, while producing some of
tl1e not-yet-encountered points of the given series, also
produce, upstream of the index I, some parasitical points other
tl1an tl10se of Ule given series.
An example: from the preceding series H, we will select only the points
between 3 and 167 inclusive. Then, we could construct the following union:
L = (73,70) U (30, 3) U (24, 23),
with P = 8760 as its period. However, ifthe same series H
were limited between the points 3 and 479 inclusive, (this time
having 40 points), it would be generated by:
L = (30, 3) U (24, 23) U (104, 70),
tl1e modulus 30 covering 16 points, the modulus 24 covering 20 points, and
the modulus 104 covering 4 points. The function L is identical to iliat given
earlier. Its period is 1560.
In general, to find ilie period of a series of points derived from a logical
expression whose definitive form is tl1e union of moduli (Mj, Ij), it is enough
to compose ule intersection of the moduli within tl1e parentheses two by two.
For example: Ml = 12, M2 = 6, M3 = 8; Ml n M2 = D * Cl * C2 = 6 * 2
* I = 12 = M; M n M3 = D * Cl * C2 = 4 * 3 * 2 = 21. And the period P = 24.
In general, one should take into account as many points as possible in
order to secure a more precise logical expression L.
Metabolae of Sieves
Metabolae (transformations) of sieves can come about in various ways:
a) by a change of the indices oftl1e moduli. For example: L = (5,1)
U (3, 2) U (7, 3) of period P = 105 will give the series:
H = { ... , 2,3,4,5,8,9, 10, 11, 14, 17, 19,20,23,24,26,29,
31, ... }. But if a whole number n is added to the indices, the
expression L becomes for n = 7:
L' = (5, 11) U (3,9) U (7, 10) and after modular reduction of
the indices:
L' = (5,1) U (3,0) U (7,3), of the same period P = 105.
The series H' = { ... , 0,1,3,6,9,10, 11, 12, 15, 16, 17, 18, 21, 24, 26, 27.
30, ... } derived from this last expression L', having the same intervallic structure
as the H series and differing from it only by its initial point, which is given by the
smallest index of the expression L' and by a shifting n of the intervallic structure
of H. Indeed, if in the series H, the intervals start from 2, which is the index of
the smallest modulus ofM, then the same intervals are to be found starting from
2 + 7 = 9 within the series H'. This case is what musicians call "transposition"
upwards and is part of the technique of "variations." If, on tlle otller hand, we
add to each index any whole number n, then the intervallic structure of the sieve
changes while its period is maintained. For example: add 3,1, and -6 respectively
to the tllree indices ofL, which becomes after their modular reductions:
L = (5,2) U (3,0) U (7,4) of period P = 105, and which gives:
H = { ... , 0, 2, 3, 4, 6, 7, 9,11,12,15,17,18,21,22,24,25,
27,30,32, ... }.
b) by transformations of the logical operations in some manner,
using the laws oflogic and mathematics, or arbitrarily.
c) by the modification of its unity u. For example, sing the national
anthem, which is based on the diatonic scale (white keys), while
transforming the semitones into quartertones or into
eighthtones, etc. If this metabola is used rarely melodically or
harmonically, it does however occur in other characteristics of
sound such as time by changes in tempo, and this, as far as
history can remember.
Conclusion
In provisional conclusion, it will be said tl\at sieve theory is tlle study of
the internal symmetries of a series of points either constructed intuitively,
given by observation, or invented completely from moduli ofrcpetition.
In what has been demonstrated above, the examples have been taken
from instrumental music. But it is quite conceivable to apply this theory to
computer generated sound synthesis, imagining that the amplitude and/or
the time of the sound signal can be ruled by sieves. The subtle symmetries
thus created should open a new field for exploration.
Chapter XII
Sieves: a User's Guide
I would like to give credit and express my thanks to Gerard Marino, a
programmer who works willi me at CEMAMu. He has adapted my own pro-
gram which I originally wrote in "Basic" into "C."
The program is divided into two parts:
A Generation of points on a straight line from the logical formula of
ilie sieve.
B. Generation of the logical formula of tlle sieve from a series of points
on a straight line.
A GENERATION OF POINTS ON A STRAIGHT LINE
FROM THE LOGICAL FORMULA OF THE SIEVE
Example:
DEFINITION OF ASIEVE:
L = [0 * 0 * ... * OJ
+ [()*()* ... *()]
+
+ [()*()* ... *()]
In each parenthesis are given in order: modulus, starting point
(taken from the set of integers)
[J + [] is a union
0* () is an intersection
Given the formula ora sieve made out of unions and intersections of
moduli, the program reduces the number of intersections to one and
keeps only the given unions. The abscissa of the final points of the
sieve are computed from these unions and displayed.
NUMBER OF UNIONS? = 2
union 1: number of modules ? = 2 modulus I? = 3
start? = 2
modulus 2 ? = 4
start? = 7
277
278
union 2: number of modules ? = 2
FORMULA OF THE SIEVE:
L = [[ 3, 2) * ( 4, 7) 1
+ [ [ 6, 9) * ( 15, 18) 1
modulus I? = 6
start? = 9
modulus 2 ? = 15
start? = 18
REDUCTION OF THE INTERSECTIONS:
union 1
[ (3,2) * (4,7) ) 1 = (12,11)
decompression into prime modules?
(press 'y' for yes, any other key for no): y
(12,11) = 4,3) * (3,2)
union 2
[ (6,9) * (15,18) ) 1 = (30,3)
decompression into prime modules?
(press 'y' for yes, any other key for no): y
(30,2) = 2,1) * (3,0) >I< (5,3)
SIMPLIFIED FORMULA OF THE SIEVE:
L = L( 12, 11) + ( 30, 3)
Formalized Music
POINTS OF THE SIEVE CALCULATED WITH THIS FOPRMULA:
rank of first displayed point? = 0
press <enter> to get a series of 10 points
Rank
0 3 11 23 33 35 47 59 63 71 83
10 93 95 107 119 123 131 143 153 155 167
20 179 183 191 203 213 215 227 239 243 251
30 263 273 275 287 299 303 311 323 333 335
40 347 359 363 371 383 393 395 407 419 423
50 431 443 453 455 467 479 483 491 503 513
60 515 527 539 543 551 563 573 575 587 599
70 603 611 623 633 635 647 659 663 671 683
80 693 695 707 719 723 731 743 753 755 767
90 779 783 791 803 813 815 827 839 843 851
100 863 873 875 887 899 903 911 923 933 935
110 947 959 963 971 983 993 995 1007 1019 1023
120 1031 1043 1053 1055 1067 1079 1083 1091 1103 1113
130 1115 1127 1139 1143 1151 1163 1173 1175 1187 1199
Sieves: A User's Guide
Line# Source Line
1 #include <stdiuo.h >
2 #include <sLdlib.h>
3 #include <coniuo.h>
4
279
5 !* ----------------------------------- types definitions -------------- ------------------- */
6 typedef struct /* period (congruence class) */
7 {
8 short mod;
9 short ini;
10 } periode;
11 typedef struct
12 {
13 short clnb;
14 periode *cl;
15 periode elr;
16 unsigned long ptval;
17 } inter;
/* modulus of the period
/>1< starting point
/* intersection of several periods
/* number of terms in the intersection
/* terms in the intersection
/* resulting period
/* current point value
"/
*/
*/
*/
*/
*/
*/
18 /* ---------------------------------- function prototypes ------------------------------- */
19 periode ReducInter(short u); /* computation of the intersections */
20 short Euelide(short ml,short m2); /* computation of the LCD */
21 short Meziriac(short cl,short c2); /* computation of "dzeta" */
22 void Decompos(periode p r Y ~ decomposition into prime factors */
23
24
25
26
27
28
29
30
31
32
/* ---------------------------------------- variables ---------------------------------------- */
inter *tCrib; /* sieve formula */
short unb = 0; /* number of unions in the formula */
short uO, u1, u = 0; /* current union index */
short i = 0; /* current intersection index */
unsigned long lastval,nO,ptnb = OJ
peri ode CL_EMPTY = { 0, }!* empty period */
33 #define NONEMPTY 1
34 short flag = OJ
35 short decomp = 0;
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
! * = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = ~
void main(void)
{
printf("SIEVES: user's guide\n\n"
"A. GENERATION OF POINTS ON A STRAIGHT LINE FROM\n'
" THE LOGICAL FORMULA OF THE SIEVE\n\n"
"Example:\n'
"---------------------------------\n "
"DEFINITION OF A SIEVE:\n"
" L = [0 * 0 * ... * ()]\n'
+ [0 * 0 * ... * ()]\n"
+ ... \n"
+ [0 * () * ... * ()]\n\n"
"In each parenthesis are given in order: modulus, starling point\n"
"(taken [rom the set ofintegers)\n"
280
Line #
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
Formalized Music
Source Line
"[J + [] is a union\n'
"0 * () is an intersection\n\n');
prin tf(" ---------------------------------\n"
"Given the formula of a sieve made out of unions and\n"
"intersections of moduli, the program reduces the number of\n"
"intersections to one and keeps only the given unions.\n"
"Then, the abscissa of the final points of the sieve are\n"
"computed from these unions and displayed.\n\n"l;
/* ----------------------- get the formula ofthe sieve ------------------------- */
while (unb == 0)
{
printf("NUMBER OF UNIONS? = ");
scanf("%d",&unb);
}
fCrib = (inter *)(maIJoc (sizeof(inter) * unb;
if(fCrib == NULL)
{
printf("not enough memory\n");
exit(l);
}
p ri n tf(" ---------------------------------\n");
for (u = 0; u unb; u++)
{
printf("union %d: number of modules ? = ", u + 1);
scanf('%d" ,&fCrib[u].c1nb);
printf("\n ");
fCrib[u].c1 = (peri ode "')(malloc (sizeof(periode) '" fCrib[u].clnb;
if (fCrib[u].c1 = = NULL)
{
exit(l);
}
for (i = 0; i fCrib[u].clnb; i++)
{
printf("\n modulus %d ? = ", i + 1);
scanf("%d" ,&fCrib[ u].cl [i ].mod);
printf(" start? = ");
scanf("%d",&fCrib[u].cl[i].ini);
}
printf(" ---------------------------------\n");
}
1* ------------------ reduction of the formula --------------_________________ */
printf("FORMUIA OF THE SIEVE:\n\n"
" L = [ ");
for (u = 0; u unb; u++)
{
if(u 1= 0)
printf(" + [ ");
for (i = 0; i fCrib[u].c1nb; i++)
{
if(i!= 0)
{
Sieves: A User's Guide 281
Line# Source Line
104 if(i % 4 == 0)
105 printf("\n ");
106 printf("'" ");
107 }
108 printf("(%5d,%5d)", fCrib[u].c1[i].mod, fCrib[u].c1[i].ini);
109 }
11 0 printf("]\n");
III }
112 printf(" ---------------------------------\n ");
113 printf("REDUCTION OF THE INTERSECTIONS:\n\n");
114 for (u = 0; u unb; u++)
115 {
116 printf("union %d\n [",u + 1);
117 for (i = 0; i fCrib[u].clnb; i++)
118 {
119 printf("(%d,%d) ", fCrib[u].cl[i].mod, fCrib[u].cl[i].ini);
120 if (i != fCrib[u].clnb - 1)
121 printf("* ");
122 }
123 fCrib[u].c1r = ReducInter(u); /* reduction of an intersection */
124 printf("] = (%d,%d)\n\n", fCrib[u].c1r.mod, fCrib[u].clr.ini);
125 printf(" decomposition into prime modules ?\n"
126 "(press 'y' for yes, any other key for no): ");
127 if (getcheO == 'y')
128 {
129 printf("\n\n (%d,%d)", fCrib[u].clr.mod, fCrib[u].clr.ini);
130 Decompos(fCrib[u].clr);
131 }
132 else
133 printf("\n\n");
134 }
135 prin tf(" ---------------------------------\n");
136 /* ------------------ display the simplified formula ------------------------ */
137 printf("SIMPLIFIED FORMUIA OF THE SIEVE:\n\n");
138 printf(" L = ");
139 for (u = 0; u unb; u++)
140 {
141 if(u!= 0)
142 {
143 if(u % 4 == 0)
144 printf("\n ");
145 printf("+ ");
146 }
147 printf("(%5d,%5d) ", fCrib[u].clr.mod, fCrib[u].clr.ini);
148 }
149 printf("\n---------------------------------\n");
150 /* --------------------- points ofthe sieve ------------------------------------- */
151 printf("POINTS OF THE SIEVE CALCULATED WITH TI-IIS
FORMULA:\n");
152 printf("rank of first displayed point? = ");
153 scanf("%lu",&nO);
154 nO=nO-nO%10;
Line# Source Line
155 printf("\npress <enter> to get a series of 10 points\n\n"
156 "Rank I");
157 for (u = 0; u unb; u++)
158 {
159
160
161
162
163
161
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
if (fCrib[u].c1r.mod != 0 II fCrib[u].c1r.ini 1=0)
{
fCrib[uJ.ptval = fCrib[uJ.e1r.ini;
flag = NONEMPTY;
}
else
fCrib[uJ.ptval = OxFFFFFFFF;
}
if (flag! = NONEMPTY)
return;
uO = ul = 0;
lastval = OxFFFFFFFF;
while (1)
{
for (u = (uO + 1) % unb; u!= uO; u = (u + 1) % unb)
{
}
if (fCrib[ u]. ptval fCrib[ u 1]. ptval)
u1 = U;
if (fCrib[u1J.ptval != lastval) /* new point */
{
lastval = fCrib[u1].ptva1;
if (ptnb = nO)
{
if (ptnb% 10 == 0)
ietchO; /* get a character from the keyboard */
printf("\n%71u I", ptnb);
}
printf("%6Iu ", fCrib[ul].ptval);
}
ptnb++;
}
fCrib[u1].ptval += fCrib[u1].elr.mod;
uO = ul;
}
/* === ===== reduction of an intersection ======== */
periode ReducInter(short u)
{
periode c1,el1,cl2,cl3;
short pged,T,n;
long c1,c2;
cl3 = fCrib[u].c1[O];
for (n = 1; n fCrib[u].clnb; n++)
{
cll = cl3;
Line#
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
Source Line
}
c12 = fCrib[uJ.e1[n];
if (cll.mod c12.mod)
{
cl = ell;
ell = c12;
cl2 = el;
}
if(c1l.mod!= 0 && cl2.mod!= 0)
{
ell.ini %= c1l.mod;
cl2.ini %= cl2.mod;
}
else
return CL EMPTY;
/* module from the intersection of2 modules */
pgcd = Euclide(e1l.mod, c12.mod);
c1 = e11.mod / pgcd;
e2 = c12.mod / pged;
if(pgcd!= 1
&& ( (cll.ini - cl2.ini) % pgcd ! = 0
return CL EMPTY;
if (pged != 1
&& ell.ini - c12.ini) % pgcd == 0)
&& (elUni != cl2.ini) && (cl == c2) )
{
cl3.mod = pgcd;
el3.ini = c1l.ini;
continue;
}
T = Meziriacshort) el, (short) c2);
cl3.mod = (short) (el * c2 * pgcd);
cl3.ini = (short) cll.ini
+ T * (el2.ini - cll.ini) * el) % el3.mod);
while (e13.ini ell.ini II el3.ini c!2.ini)
cl3.ini += el3.mod;
}
return cl3;
/*=== === decomposition into an intersection ======== "/
/" of prime modules */
void Decompos (periode prj
{
periode pf;
short fct;
if (pr.mod = = 0)
{
printfr = (%d,%d)\n", pr.mod, pr.ini);
return;
}
printf(" =");
for (i = 0, fct = 2; pr.mod!= 1; fct++)
283
284
Line#
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
Source Line
}
{
pf.mod = 1;
while (pr.mod % fet == 0 && pr.mod 1= 1)
{
pf.mod "'= fet;
pr.mod /= fct;
}
if (pf.mod1= 1)
{
}
pf.ini = pr.ini % pf.mod;
pr.ini %= pr.mod;
if(i 1= 0)
printf(" *");
printf(" (%d,%d)", pf.mod, pf.ini);
i++j
}
printf("\n");
Formalized Music
I'" ========== Euclide'salgorithm======== */
short Euclide (aI, a2) 1* al = a2 0 *1
short al;
short a2;
{
}
short tmp;
while ((tmp = al % a2) 1= 0)
{
al = a2;
a2 = tmp;
}
return a2;
1* ======== De Meziriac's theorem ========= */
short Meziriac (el, c2) /* el = e2 a */
short el;
short c2;
{
short T = 0;
if(c2 == 1)
T= 1;
else
while (++T * el) % c2) 1= 1)
,
return T;
B. GENERATION OF THE LOGICAL FORMULA OF
THE SIEVE FROM A SERIES OF POINTS ON A
STRAIGHT LINE
Example:
Given a series of points, find the starting points with their moduli
(periods).
NUMBER OF POINTS? = 12
abscissa ofthe points
point 1 = 59
point 5 = 63
point 9 = 95
point 2 = 93
point 6 = 11
point 10 = 71
point 3 = 47
point 7 = 23
point 11 = 35
point 4 = 3
point 8 = 33
point 12 = 83
POINTS OF THE SIEVE (ordered by their increasing abscissa):
Rank 0 3 11 23 33 35 47 59 63 71 83
10 93 95
FORMULA OF THE SIEVE:
In each parenthesis are given in order:
(modulus, starting point, number of covered points)
L = (30, 3, 1}+( 12, 11, 8)
period of the sieve: P = 60
Line #
1
Source Line
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#includc <string.h>
2
3
4
5
6
7
8
9
/* ------------------------------ types definitions --------------------------------------- */
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
typedef struet 1* period ( congruence class) */
{
short mod;
short ini;
short couv;
} periode;
/* modulus of the period
/* starting point
I'" number of covered points
*/
*/
1* ---------------------------- function prototypes ------------------------------------- */
unsigned long Euclide(unsigned long mI,
unsigned long m2); /* computation of the LCD */
1* -------------------------- variables and constants --------------------------------- *1
periode"perCrib; 1* periods of the sieve */
short perTotNb = 0; 1* number of periods in the formula .. /
long *ptCrib; 1* points ofthe erible .. /
long *ptReste; /* points outside the periods .. /
short ptTotNb = 0; 1* number of points in the sieve *1
short p,ptnb;
long ptval;
unsigned long percrib;
285
286
Line #
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
Formalized Music
Source Line
periodEper;
#define NON REDUNDANT 0
#define REDUNDANT I
#define COVERED -lL
short flag;
/* =================================== */
void main(void)
{
printf("B. GENERATION OF TI-IE LOGICAL FORMULA OF THE
SIEVE FROM\n"
" A SERIES OF POINTS ON A STRAIGHT LINE\n\n"
"Example:\n"
"---------------------------------\n "
"Given a series of points, find the starting points\n"
"with their moduli (pcriods).\n\n");
/* --------- entry of the points of the sieve and their sorting -------*/
while (ptTotNb = = 0)
{
printf ("NUMBER OF POINTS? = ");
scanf("%d",&ptTotNb);
}
ptCrib = (long *)(malloc (ptTotNb * sizeof(long);
ptReste = (long *)(malloc (ptTotNb * sizeof(long);
= (periode *)(malloc (ptTotNb * sizeof(periode)));
If (ptCnb == NULL II ptRcstc == NULL II perCrib == NULL)
{
exit(1);
}
prin tf(" ---------------------------------\n"
"abscissa of the points:\n");
for (p = 0; P ptTotNb; p++)
{
if(p%4 == 0)
printf("\n to);
printf("point %2d = ", P + I);
scanf("%ld", &ptval);
for (ptnb = 0;
,
ptnb p && ptval ptCrib[ptnb];
ptnb++)
if(ptnb p)
{
if (ptval ptCrib[ptnb]) /* new point
+ 1], &ptCrib[ptnb),
slzeof(long) * (p - ptnb;
else /* point already exist */
{
p--;
*/
Sieves: A User's Guide 287
Line#
76
77
78
79
80
81
82
83
Source Line
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
}
}
ptTotNb--;
}
ptCrib[ptnb] = ptval;
prin tf("\n ---------------------------------\n ");
/* ______________________ points of the sieve -----------------.---:---------.--- */
printf("POINTS OF THE SIEVE (ordered by their mcreasmg
abscissa):\n\n'
"Rank I');
{
if(p% 10 == 0)
printf("\n%7d I", p);
printf("%6ld ", ptCrib[p));
}
p rin tf("\n \n ---------------------------------\n");
/* ________________ compute the periods of the sieve ------------------- */
memcpy(ptReste, ptCrib, ptTotNb * sizeof(long);
{
if ( ptRcste[p] = = COVERED)
continue;
/* ________ compute a period starting at current point --------- */
per.mod = 0;
do
{
per.mod++;
per.ini = (short) (ptCrib[pJ % (long)per.mod)j
per.collv = 0;
for (ptnb = 0, ptval = per.ini; .
ptnb ptTotNb && ptval = ptCnb[ptnb);
ptnb++)
}
}
{
if (ptval == ptCrib[ptnb])
{
per.collv++;
ptval += per.mod;
}
while (ptnb ptTotNb);
/* ----------- check the redundancy afthe period ------------------ */
for (ptnb = 0, ptval = per.ini, flag = REDUNDANT;
ptnb ptTotNb;
ptnb++)
{
if (ptval = = ptCrib[ptnb])
{
if (ptval = = ptReste[ptnb])
{
288
Line #
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
Formalized Music
Source Line
ptRcste[ptnb] = COVERED;
flag = NON_REDUNDANT;
}
ptval += per.mod;
}
}
if(Oag == NON_REDUNDANT)
perCrib[perTotNb+ +] = per;
}
/* ----------------- compute the period of the sieve ---------------------- *1
percrib = perCrib[O].mod;
for (p = 1; P perTotNb; p++)
{
if long) perCrib[p].mod = pcrcrib)
percrib *= (long) perCrib[p].mod 1 Euclidelong)perCrib[p].mod,
percrib);
else
percrib *= (long) perCrib[p].mod 1 Euclide(percrib,
(long)perCrib[p ].mod);
}
/'" ----------------- display the formula of the sieve ----------------------- */
printf('FORMULA OF THE SIEVE:\n"-
"In each parenthesis are given in order:\n"
"(modulus, starting point, number of covered points)\n\n")j
printfr L = ");
for (p = OJ P perTotNbj p + +)
{
if(p != 0)
{
if(p %3 == 0)
printf("\n ");
printf('+ ;
}
printf("(%5d,%5d,%5d) ", perCrib[p ].mod,perCrib[p ].ini,
perCrib[p ],couv)j
}
printf("\n\n period of the sieve: P = %Iu\n", percrib)j
I'" == ========= Euclide's algorithm ============= */
unsigned long Euclide (aI, a2) /* al = a2 0 wi
unsigned long al;
unsigned long a2;
{
unsigned long tmp;
while tmp = a1 % a2) 1= 0)
{
a1 = a2;
a2 = tmp;
}
return a2;
Chapter XIII
Dynamic Stochastic Synthesis
What is the most economical way to create a plane wave in an amplitude-
time space (atmospheric pressure-time), encompassing all possible forms
from a square wave to white noise? From an informatics point of view, a
square wave is quite simple with only two amplitudes, a over n of fixed
samplings. White noise is also quite simple and generated by a compound of
stochastic functions whose samplings are dovetailed, nested, or not.
But what about waves representing melodies, symphonies, natural
sounds .. .?
The foundation of their nature and therefore of their human
intelligibility is temporal periodicity and the symmetry of the curves. The
brain can marvelously detect, with a fantastic precision, melodies, timbres,
dynamics, polyphonies, as well as their complex transformations in the form
of a curve, unlike the eye which has difficulty perceiving a curve with such a
fast mobility.
An attempt at musical synthesis according to this orientation is to begin
from a probabilistic wave form (random walk or Brownian movement)
constructed from varied distributions in the two dimensions, amplitude and
time (a, t), all while injecting periodicities in t and symmetries in a. If the
symmetries and periodicities are weak or infrequent, we will obtain
something close to white noise. On the oUler hand, the more numerous and
complex (rich) the symmetries and periodicities are, the closer the resulting
music will resemble a simple held note. Following these principles, the whole
gamut of music past and to come can be approached. Furthermore, t11e
relationship between the macroscopic or microscopic levels of these injections
plays a fundamental role.
Below, is a first approach to constructing such a wave.
Procedure
AI. Following Ule absciss oft, we begin with a length (period) T where
T = l/f seconds and f is a freely chosen frequency. At the start, this period T
is subdivided into n equal segments; for example, n = 12 (this is one
289
macroscopic level). Every time T is repeated, each segment trli.l of (i = 0, 1,
2, 3, ... ,n-1) undergoes a stochastic alteration which increases or reduces it
within certain limits imposed, for example, by clastic barriers.
Bl. Following the amplitude axis, a value is given to each extremity of
the 12 preceding segments. These values form a polygon inscribed or
enveloping a sine wave, or a rectangular form, or a form born of a stochastic
function such as that of Cauchy, or even a polygon flattened at the zero level.
The Ei ordinates of these n summits undergo a stochastic alteration at each
repetition which is sufficiently weak and even more, compressed between two
adequate elastic barriers.
Cl. The E ordinates of dIe samplings found between the two
extremities of a segment T will be calculated by a linear interpolation of the
ordinates Ei_1 and Ej of these extremities.
A2. Abscissa of the polygon's summits
0 I i-, 0, i-I
I
t
i
_
1
t
ti
I
(-1
t'
tl.
I I
I
T I ..
' I
1
7'---1
Figure 1.
() prec(eding) = t; - 'i'1 ; e pres(ent) = t'l - t'i'1
Procedures
Construction ofe present from "logistic" distribution:
and its distribution function,
= _ -t = a'-(3)
we obtain - (/3 + In(.!..=1. ) ) / a with y coming from dIe
y
uniform distribution:
O:5y:5l.
(2)
take: pres = prec +
(3)
Pass this pres into local elastic barriers
100 taken from /3/2, to obtain
(4)
Then do:
(} pres = e prec + S-'*Rdct
where Rdct is a reduction factor.
291
(5)
Finally pass e pres into general clastic barriers fimin and e max
obtained as follows:
a) the minimum frequency is, say 3 HZ. Then the maximum
period is T = i sec and each of the 12 segments will have a mean
1
length of 8
max
= 3 * 12 sec.
b) The maximum frequency could be HZ where
SAMP is the sampling rate, say 44100 HZ. Therefore each of the
12 segments could have a minimum length of the period
T 1
12 = SAMP = emin.
(6)
Repeat the above procedures for each of the n = 12 segments.
292
Formalized Music
B2. Ordinates of the Polygon's summits
- - - - - - o ~ - - - - - - - - - - ~ ~ - - - - - - - - - - - -
i - 1 l
Figure 2.
The i
th
present ordinate is obtained from tJle i
th
preceding ordinate in the
following manner:
Construction of the Ei pres:
(1) Take a probability distribution W(a); then its distribution
<J
function Q(W) = J W (a)da. We obtain a = V(Q, y) with
- 00
o ~ y ~ 1 (the uniform distribution) and W(a) any distribution.
(2) Pass a through local barriers ( 0.2)
(3) Add this a to the Ei .
pree.
Ei pres = Ei prec + a
(4) Pass Ej pres through limitative barriers 8 bits ( 32768) and
this is the final Ej pres.
(5) Do this for each of the 10 summits wjtJlin the two boundary
summits oftJle polygon.
(6) The last boundary summit will be taken as tJle first boundary
summit of the next period.
C2. Construct the Ei ordinate ofthe sampling point t which can be found on
the segment lj - lj-l between the ordinates Ei-l and Ej through a linear inter-
polation.
(Ei - Ei- 1) (t - ti- 1)
Et= ~ - - ~ - - - 7 ~ - - - - ~
ti - ti- 1
Therefore, a microscopic construction.
293
General comment: the distribution functions U ( ~ ) and W(a) can be
either simple, for example, sine, Cauchy, logistic, ... ; or more complex,
through nesting, etc.
The data given above is naturally an arbitrary starting point which I
used in La Ligende d'Eer.
This approach to sound synthesis represents a non-linear dynamic
stochastic evolution which bypasses the habitual analyses and harmonic
syntheses of Fourier since it is applied to the f(t) part on the left of the equal
sign of Fourier's transformation. This approach can be compared to current
research on dynamic systems, deterministic chaoses or fractals. Therefore, we
can say that it bears the seed of future exploration.
Chapter XIV
More Thorough Stochastic Music
Introduction
This chapter deals with a generalisation of sound synthesis by using not
periodic functions, but quite the opposite, non-recurring, non-linear
functions. The sound space in question is one which will produce a likeness
of live sounds or music, unpredictable in the short or long run, but, for
example, being able to vary their timbre from pure "sine - wave" sound to
noise.
Indeed, the challenge is to create music, starting, in so far as it is
possible, from a minimum number of premises but which would be
"interesting" from a contemporary aesthetical sensitivity, without
borrowing or getting trapped in known paths.
The ontological ideas behind this subject have already been exposed
in the chapters treating ACHORRIPSIS (cf. chapters I and V) some 33
years ago, and still form the background canvas to this new, somewhat
more thorough scope, which should result in more radical experimental
solutions.
If, at that time, the "waves" in the "black universe" were still produced
by musical instruments and human beings, today, these "waves" would be
produced mainly by probability distributions (adorned with some
restrictions) and by computers.
Therefore, we find ourselves in front of an attempt, as objective as
possible, of creating an automated art, without any human interference
except at the start, only in order to give the initial impulse and a few
premises, like in the case of the Demiourgos in Plato's Politicos, or of
Yahweh in the Old Testament, or even of Nothingness in the Big Bang
Theory.
295
Microstructure
The fundamental ingredients used are (almost like in the case of La
Legende d'Eer) four in number:
a) A temporal ficticious length divided into a given number of
segments, at whose ends we draw amplitudes in order to form a stochastic
polygonal wave-form (PWF);
b) As a matter of fact, this polygone is built continuously and endlessy
through the help of probability distributions by cumulatively varying
temporal lengths as well as the amplitudes of the vertices;
c) In order to avoid excessive cumulated values, elastic barriers are
imposed;
d) A linear interpolation joins the vertices.
Under certain conditions, this procedure, although chaotic and
undeterministic, produces a relatively stable sound.
The computation of the stochastic polygonal uses one
stochastic law that governs the amplitudes and another one that governs
the durations of the time-segments. The user chooses among several disc-
tinct stochastic laws (Bernouilli, Cauchy, Poisson, Exponential. .. ). The
sizes of the elastic-mirrors that are applied to the amplitudes and the dura-
tions can be chosen too.
Macrostructure
A) The preceeding procedure therefore produces a sound of a certain
duration;
B) A sequence (PARAG(psi% results from a simultaneous and temporal
multiplicity of such sounds. This sequence is equally constructed through
decisions governed by probability distributions;
C) An arbitrary chain of such sequences could produce an interesting
musical composition.
DATA
of the sequence
PARAG(psi%)
The end-figures of the dyn%-routes arc given by dynMIN%:$ dynMAX%
(here, up to 16 arbitrary routes)
For each dyn%-route are defined :
I) The number Imax% of segments for the polygonal wave-form (PWF),
2) The number of sound fields per dyn%-route,
297
3) The coefficient of the exponential distribution which stochastically
governs the sound or silence fields of this dyn%-route,
4) The probability (Bernoulli distribution) by which a field becomes a
sound field,
5) Various digital filters,
6) Two stochastic laws that govern the amplitudes (ordinates) and the
intervals (durations) of the vertices of the successive polygonal wave-forms,
(at least six distinct stochastic laws are introduced),
7) If needed, two numerical coefficients for each of the previous stochastic
laws,
8) a) The sizes of the first two elastic-mirrors that are used for the amplitudes
( ordinates),
b) The sizes of the first two elastic-mirrors that are used for the abscissa
(time)
, ., tl d fi r the
c) The sizes of the second two elasuc-mIrrors lat are use 0
amplitudes (ordinates),
d) The sizes of the second two elastic-mirrors that are used for the
abscissa (time),
9) Proportional corrections of the mirror-sizes ill order to avoid an
overflow (> 16 bits) per sample,
10) For all the of this PARAG(psi%) sequence, a stoc!lastic
computation (through exponential distribution) of the sound or
fields is carried out, determining namely tlleir starting points and their
durations.
I
I
298
dyn%<=dynMin%
DYNAS(dyn%)
..
Formalized Music
l
'----iABLE of arbitrary sUosion of PARAG(psi%)
with: ysp% = ordinal numbc=r of the ;
psi% = specific number ora
- ------
I I
J
DATAofscqu_PARAG(psi%)1
CONSTRUCTION ofa soundoSeCjuence thanks
to 3 and temporal stochastic multiplicity made up
with the contribution of5t':veral given
d)fl%=dJ1lMin%+l
CONTRIBUTION of the dyn%--route by means of the
sub-routine DYNAS(dyn%). that is, slOl'hau.ic construction of
succenive polygonal wave-forms (PWF)
AMPLITUDE (ordinate)
oCa
1. Choose, according to
PARAG(psi%) the stochastic
distribution fey).
2. From the uniform distrib. we
drawOSZl:s 1. Then from the
distrib. function
F(y) = ZI
I
we draw amplitude (ordi-
nate) increase: of thi3 vertex.
3. This Yo is then raken through
a pair of elastic mirron.
4. The result y' is added to the
amplitude (ordinate) ofthe
same vertex of the preceding
PWF_
5. This new value y (after it has)
been taken through a 2d pair
of elastic mirrors) gives the
amplitude (ordinate) oCthe
present vertex of the actual
PWF.
ABSCISSA (ume)
of a vertc=:x:
1. Choose, according to
PARAG(psi%) the slOchastic
distribution g(x)
2. From !.he uniform distr. we
draw 0" :S Z2 S 1. The from the
distribUtion function.
F(x) jO g(x)dx _ Z2
we draw i-;;crease Xo of the
interval (duration) that sepa-
rares this present from
its predccCMOr.
3. This XO is taken through a
pair of elastic mirrors.
4. The result x' is added to the
interval (duration) that was
separating same vertices in the
preceding PWF.
5. This new value x it has
been taken through a 2d pair
of elastic mirrors) gives the in-
terval (duration) that separates
the present vertex from the
preceding one of the aClual
PWFo
sample by sample linear interpolations of the PWF amplitudes
(ordinate) that are separated by this x interv31 (duration) : 44100
samples/sec.
arbitrary digital filtering of the amplitudes (ordinateS) and/or of
the durationlOi.
Having used all contributions of the dyn%-routes of this sequence PARAG(psi%) and as soon as the above computation for just one
sample is ended, thcn :
a. If the of this PARAG(psi%)-scquence arc not ended, we compute anew the contribution of each one of the
dyn'ro-routes for the next sample, by repeating the abo"'e procedure.
b. If the armis are ended,.hen are introduced in the main programme the DATA of the
(m agnxmcnt with the TABLE afthe.PARAG(psi%)-sequences) and we repeat the previous
c. If the last PARAG{psi%)--sequen<:e or the TABLE has been computed then the task and the mw.ic are terminated .
.t"or many-ehannd stereo music: a. compute from the start thc same main programme ali many times as there are channels; b. use
separate random-generator for !".ach channel for the am._op_l_it_ud_es_an __di_o_,_fo_,_th_e_a_bsci_-""'_-__________ -'
299
---
... ... ...
dytl%=dynMax%-2
I

I

DYNAS(dy"%) DYNAS(dyn%) DYNAS(dyn%)
...i. ... ... ... i ....i:.
]
.. ... -0,.
I
... ... ...
--- --*
,.
.-Y
300
, PROGRAMME*
'PARAG3.BAS
Formalized Music
'AUTOMATED COMPUTATION of the SOUND-PATCHES for GENDY1.BAS
,
---------------------------------------------------------------------------------
, do RANDOMIZE n
with -32768 < n < 32767
ex. n = 4000
then RANDOMIZE n ' Uniform distrib.
,
---------------.......... _-----_ ...._----------......... _--_ ...._ .. __.......... _-_.. _--... ----_ ......----_ ........'
R$ = LTRIMS(STR$(psi%))
n = 4300: RANDOMIZE n
'++++++++++++++++++++++
pSi% = 3 'index of this data
'programme.
'******* .. *
prtS = "prt" + R$: prtS = prtS + ".DAT"
'file for soundpatches
= "ARAGOO" + R$: QO$ = QO$ + '.DAT"
'file for general data
'data file for the 13th dyn%-field:
MO$ = "ARAG130' + R$: MO$ = MO$ + ".DAT"
Ml$ = "ARAG131" + R$: Ml$ = MIS + ".DAT"
= "ARAG132" + R$: M2$ = M2$ + ".DAT"
'################################################
dyn% = horiz% = 1: e% = 2: ecrvrt% = 3: convrt% = 1: mkr = 1.2
DIM DEBmax&(O TO 20)
DIM D(O TO 20)
DIM p(O TO 20)
DIM ralono/c(1 TO 20)
DIM U2&(0 TO 20)
'last sound-patch of this dyn%-field
'coefficient for the exponential distribution
'probability for the Bernoulli distribution: ,0 :;; p :;; 1
'extention of the time-interval (abscissa)
'size of the upper second-elastic-mirror
"'Th'
IS programme has been technically realized with the help of
Marie-Helene Serra (I.X.).
More Thorough Stochastic Music 301
DIM V2&(0 TO 20)
DIM filter%(O TO 20, TO 10)
'size of the lower second-elastic-mirror
'there are ten possible filters per dyn%-field
OPEN QO$ FOR OUTPUT AS # 1 'general data for the sequence
vertec% = 1: vertcon% = 2 'indexes of the ampl.-ordinate for the screen
'and the converter.
Nmax& = 10000000
dynMin% = 1
dynMax% = 16
flrt%(vertec%) = 0 'vert.screen-filter for GENDYl.BAS
flrt%(vertcon%) = 1 'vert.convert.-filter for GENDY1.BAS
WRITE #1, Nmax&, dynMin%, dynMax%, flrt%(vertec%), flrt%(vertcon%)
CLOSE #1
'**************************************************************************
OPEN MO$ FOR OUTPUT AS #1 'as an example,this is the 13th dyn%-field
dyn% = 13
Il3max% = 13
DEBmax&(dyn%) = 25
'number of divisions of the waveform
'max.number of sound or silence sound-patches.
'proportionality factor and coefficient for
'the exponential distribution:
D(dyn%) = mkr * .45/ (1.75" 1.25)
p(dyn%) = .35 'the BERNOULLI distribution.
ralon%(dyn%) = 9 'minimal time interval extention
filter%(dyn%, horiz%) == 1
filter%(dyn%, e%) = 1
filter%(dyn%, ecrvrt%) = 1
filter%(dyn%, convrt%) = 1
WRITE #1, dyn%, Il3max%, DEBmax&(dyn%), D(dyn%), p(dyn%),
ralon%(dyn%), filter%(dyn%, horiz%), filter%(dyn%, e%), filter%(dyn%, ecrvrt%},
filter%(dyn%, convrt%)
CLOSE #1
OPEN Ml$ FOR OUTPUT AS #1
A13 = .01: B13 = 5: U131& = 1: V131& = -1: U2&(dyn%) = 7:
V2&(dyn%) = -7: Rdctl3 = 1: distrPC13 = 1
WRITE #1, A13, B13, U131&, V131&, U2&(dyn%), V2&(dyn%), Rdctl3, distrPC13
CLOSE #1
OPEN M2$ FOR OUTPUT AS #1
Ad13 = 1: Bd13 = 6: Ud131& = 2: Vd131& = -2: Ud132& = 20:
Vd132& = 0: Rdcd13 = 1: distrPDl3 == 2
WRITE #1,AdI3, Bd13, Ud131&, VdI31&, UdI32&, VdI32&, Rdcdl3, distrPD13
CLOSE #1
'*************************************************************************
302
Formalized Music
'################################################
DIM TH&(O TO 20, 0 TO 100) 'slarting point (sample) ora sound/silence patch
DIM DUR&(O TO 20, TO 100) 'duration of that patch
DIM THpr&(O TO 20, 0 TO 100) 'present slarting point
DIM BED&(O TO 20, 0 TO 100) 'variable for the compulation of the patches
DIM sTHend&(O TO 20) 'last sample
'@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
OPEN prt.$ FOR OUTPUT AS #1 'COMPUTING the sound or silence patches .
Gpl:
FOR dyn% == dynMin% TO dynMax%
'n == 4000 + 100" psi% + 10" dyn%: RANDOMIZE n
DEB&==O: IFp(clyn%) <==OTHEN
ENDIF
'ignore this dyn%field
GOTO Gp2
DEB& == DEB& + 1: yl == RND: y2 == RND
DR == -(LOG(1 - y2 / D(dyn%) 'patch-duration==EXPON.
'distrib/sec.
DUR&(dyn%, DEB&) == DR * 44100 'same in samples.
THpr&(dyn%, DEB&) == THpr&(dyn% DEB& - 1) + DUR&(dyn%
DEB&) , ,
IF yl < == p(dyn%) THEN 'the sound is in this patch!
END IF
TII&(dyn%, DEB&) == THpr&(dyn%, DEB& - 1)
TIIDUR == TIIDUR + DR
BED&(dyn%, DEB&) == BED&(dyn%, DEB&) + 1
DBE& == DBE& + 1
IFDEB& < DEBmax&(dyn%)TIIEN
GOTO GpJ
ELSE
FOR xi% == I TO DEBmax&(dyn%)
THend& == TH&(dyn%, xi%) + DUR&(dyn%, xi%): TELOS& ==
TELOS& + DUR&(dyn%, xi%)
WRITE #1, BED&(dyn%, xi%), TH&(dyn%, xi%), DUR&(dyn%, xi%),
THend&, TH&(dyn%, xi%) / 44100, DUR&(dyn%, xi%) / 44100,
THend& / 44100
NEXTxi%
of this
IF TIIend& > == sTIIend&(dyn%) TI-IEN
sTHend&(dyn%) == THend&
END IF
WRITE # 1, THDUR, THDUR / (TELOS& / 44100), sTHend&(dyn%)
DURsec == (sTHcnd&(dyn%)) / 44100
END IF
Gp2:
'against the overflow
,
--------
sU2& == sU2& + U2&(dyn%)
sV2& == sV2& + V2&(dyn%)
TIIDUR == 0: TELOS& == 0: DBE& == 0
sTHend&(dyn%) == 0
303
NEXTdyn%
'Proportionality for less than 16 bits amplitudes (upper mirrors)
,++++++++++++++++++++++++++++++++++++++++++++
IF p(dyn%) > THEN
IF dyn% == 1 THEN
OPEN Al$ FOR OUTPUT AS # 1: U2&(dyn%) == (98/ s U2&) * U2&(dyn%)
WRITE #1,Al, Bl, Ull&, Vll&, U2&(dyn%), V2&(dyn%), Rdctl, distrPCl
CLOSE #1
ELSEIF dyn% == 2 THEN
OPEN Ml$ FOR OUTPUT AS #1: U2&(dyn%) == (98/ sU2&)" U2&(dyn%)
WRITE #1, A13, B13, U131&, VI31&, U2&(dyn%), V2&(dyn%), Rdct13, distrPC13
CLOSE #1
END IF
END IF
NEXTdyn%
'Proportionality for less than 16 bits amplitudes(lower mirrors)
'++++++++++++++++++++++++++++++++++++++++++++
IF p(dyn%) > 0 THEN
IF dyn% == 1 THEN
OPEN Al$ FOR OUTPUT AS #1: V2&(dyn%) == (-98/ sV2&) * V2&(dyn%)
WRITE #1, AI, Bl, Ull&, Vll&, U2&(dyn%), V2&(dyn%), RdCll, distrPCl
CLOSE #1
ELSEIF dyn% = 13 THEN
OPEN Ml$ FOR OUTPUT AS #1: V2&(dyn%) = (-98/ sV2&) * V2&(dyn%)
WRITE #1,A13, B13, U131&, V131&, U2&(dyn%), V2&(dyn%), Rdct13, distrPC13
CLOSE #1
END IF
END IF
NEXTdyn%
'( (
END
'G END Y 1 . BAS
'= = = = =
------------
------------
- ---
------
'This programme controls several stochastic-dynamic sound-fields.
'A stochastic-dynamic sound-field is made out of a wave-length Tl
'divided in Imax% segments (durations). Each one of these segments
'is stochastically varied by a cumulated probability-distribution.
'At the ends of each one of these segments are computed the amplitudes
'(ordinates) that will form the waveform polygone. Are defined:
'for the duration abscissa a probability distribution and 2 times 2
'elastic mirrors; for the amplitude ordinates a probability distri-
'bution and 2 times 2 elastic mirrors. In between the vertices a linear
'interpolation of points completes the waveform polygone.
'@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
'1st field:
'compute one sound-sample:
DECLARE SUB DYNASI (Ilmax%, SMP&, Cll&, C12&, t11&, t12&, 11%, Nl&,
fh&, hf&, hh&)
'compute the amplitude-ordinate:
DECLARE SUB PCl (Tabl10, Tab120, 11 %, Nl&)
'compute the time-abscissa:
DECLARE SUB PDl (Tadll(), Tad120, 11%, Nl&)
'2d field:
'13th field
'compute one sound-sample:
DECLARE SUB DYNAS13 (113max%, SMP&, C131&, C132&, tI31&, tI32&,
113%, NI3&, fh&, hf&, hh&)
'compute the amplitude-ordinate:
DECLARE SUB PC13 (Tab1310, Tab1320, 113%, NI3&)
'compute the time-abscissa:
DECLARE SUB PDl3 (Tad1310, TadI32(), 113%, NI3&)
'14th field
305
'@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
'Sample-file for output to the converter:
OPEN "C:\SOUND\S351.DAT" FOR BINARY AS #3
SON$ = "S351" 'sound number on disc
'&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
rndj = 401
RANDOMIZE rndj
'rndj initialises the random-number gene-
'rator used through all this programme.
'-32768 < rndj < 32767
'LEHMER'S random-number generators are also used.
'&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
DIM psi%(O TO 31) 'for 32 sequences psi%
DIM chD&(O TO 31) 'the greatest duration-length of a sequence.
'&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
'psi% is the number of a given sequence.
'ysp% is an ordinal number from yspMin% to yspMax% used as an index for psi%.
'DEFINE HERE ypsMin% and ypsMax% and the order ofa freely chosen
'succession of sequences psi% given in the SUB ARCHSEQI(yspMin%,yspMax%)!
'For example:
yspMin% = 1
yspMax% = 7
OPEN "SEQSON" FOR OUTPUT AS #1
WRITE #1, SON$, yspMax%, yspMin%
FOR yspMin% = 0 TO yspMax%
'file to be used in the score
routine.
CALL ARCHSEQI (yspMin%, yspMax%)
WRITE #1, psi%
NEXTysp%
CLOSE #1
'dynMin% and dynMax% (= minimum and maximum values of the dyn%-fields)
'are to be found in PARAG(psi%).
'! IIIl1II1!!1I !!I! IIIII!!! III I! II! II III!!! 1I11! 111111 1I11!!!!!II!!!!!!!! IIII! II!! I! II!!! III! 11111 !!IIII! II!!! II!!!!! II!!!
'=========== COMPUTATION'S BEGINING ===============
ysp% = yspMin%
'**************************************************************************
Ibgl:
CALL ARCHSEQ1(ysp%, yspMax%)
'1!!!!!IIl!!!IlI!!!!!!!! I! I!! I I I!!! I!!!!!!!!!!!!!!!!!! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!! !!!!! I!!!!!! !!!!!!!!!!!!! I I I!! !!!!!!!!
'$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
'Free dimensioning of the tables
,
--_ ...._----.. _ ...._----------------
'Tables for the ordinate values of the Il % segment for cumulation.
DIM Tabll(l TO 2, 0 TO 90): DIM Tabl2(l TO 2, 0 TO 90)'1\.=1 or 2:IjMax%
'= 90
'Tables for the abscissa values of the Il% segment for cumulation.
DIM TadU(l TO 2, 0 TO 90): DIM Tad12(1 TO 2, 0 TO 90)
'Tables for the ordinate values of the 12% segment for cumulation.
~ I M Tab21(1 TO 2, 0 TO 90): DIM Tab22(1 TO 2, 0 TO 90)
DIM Tabl3I(l TO 2, 0 TO 90): DIM Tab132(l TO 2,0 TO 90)
'Tables for the abscissa values of the Il3% segment for cumulation.
DIM Tad131(1 TO 2, 0 TO 90): DIM Tad132(1 TO 2, 0 TO 90)
~ I M Tab141(1 TO 2, 0 TO 90): DIM Tab142(1 TO 2, 0 TO 90)
'************************************************************************
'dyn% = index of the stochastic subroutine DYNAS(dyn%);
'DEB&(dyn%) = ordinal index of the sound-patches of this routine;
'DEBmax&(dyn%) = last sound-patch of this routine;
'DUR&(dyn%,DEB&(dyn% = sound-duration whose ordinal number is
'DEB&(dyn%);
'TH&(dyn%,DEB&(dyn%)) = the SMP& sample at which each sound-patch
'commences;
'SMP& = number of the running sample;
'Ijmax% = number of subdivisions of a waveform time-length.
DIM DEBmax&(O TO 20)
DIM DEB&(O TO 90)
DIM D(O TO 20)
DIM pp(O TO 20)
DIM TH&(O TO 20, 0 TO 90)
DIM DUR&(O TO 20, 0 TO 90)
DIM BED&(O TO 20, 0 TO 90)
DIM U2&(0 TO 20)
DIM V2&(0 TO 20)
DIM sTHend&(O TO 20)
DIM flrt%(O TO 2)
DIM filter%(O TO 20, 0 TO 10)
DIM ralon%(l TO 20)
'max.patch numb.:dynMin%=O TO
'dynMax%=20
'current patch numb.: 0 TO
'DEBmax&(dyn%) =90
"in expon.dens.;dynMin%=O TO
'dynMax%=20
"in Bernoulli dens.;dynMin%=O TO
'dynMax%=20
'patch start:dyn%=O TO '20,DEB&(dyn%)=0
TO 90
'patch dur.:dyn%=O TO '20,DEB&(dyn%)=0
TO 90
'patch param.:dyn%=O TO
'20,DEB&(dyn%)=0 TO 90
'upper mirror size: dynMin%=O TO
'dynMax%=20
'lower mirror size: dynMin%=O TO
'dynMax%=20
'last sample of the considered dyn%.
'final screen or converter filter.
'ten available filters per field (dyn%).
'extcntion of abscissa.
'*************************************************************************
'readings of sequences' data from files written by PARAG(psi%).
,
....--_ ......._---------...._------_ .. _----_ .........----------------------------------..--------------
R$ = LTRIM$(STR$(psi%
prt$ = "prt" + R$: prt$ = prt$ + ".DAT"
QO$ = "ARAGOO" + R$: QO$ = QO$ + ".DAT"
AO$= "ARAGIO" + R$: AO$ = AO$ + ".DAT"
Al$ = "ARAGll" + R$: Al$ = Al$ + ".DAT"
A2$ = "ARAGI2" + R$: A2$ = A2$ + ".DAT"
BO$"ARAG20" + R$: BO$ = BO$ + ".DAT"
'sound-patches data-files.
'general data-file for all
'sequences.
'specific data for 1st
'dyn%-field.
MO$ = 'ARAGl30" + R$: MO$ = MO$ + ".DAT" 'specific data for 13th
'dyn%-field.
MI$ = "ARAG131" + R$: MI$ = MI$ + ".DAT'
M2$ = "ARAG132" + R$: M2$ = M2$ + ".DAT'
NO$ = "ARAG140' + R$: NO$ = NO$ + ".DAT"
'specific data for 14th
'dyn%-field.
'@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
horiz% = 1: e% = 2: ecrvrt% = 3: convrt% = 4 'filter indexes
'general data-files for the dyn%-fields.
' ******************************
OPEN QO$ FOR INPUT AS # 1
INPUT #1, Nmax&, dynMin%, dynMax%, flrt%(l}, flrt%(2}
CLOSE #1
'&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
'specific data-files for each dyn%-field.
,
---------------------------..-_ .... _--------------------
'************************************************************************
OPEN AO$ FOR INPUT AS #1
dyn% = 1
INPUT #1, dyn%, Ilmax%, DEBmax&(dyn%), D(dyn%}, pp(dyn%},
ralon%(dyn%}, filter%(dyn%, horiz%}, filter%(dyn%, e%}, filter%(dyn%,
ecrvrt%), filter%(dyn%, convrt%)
CLOSE #1
OPEN Al$ FOR INPUT AS #1
INPUT #1, AI, Bl, Ull&, Vll&, U2&(dyn%), V2&(dyn%), Rdctl, distrPCl
CLOSE #1
OPEN A2$ FOR INPUT AS #1
INPUT #1, Ad!, Bdl, Udll&, Vdll&, UdI2&, VdI2&, Rdcdl, distrPDl
CLOSE #1
,*************************************************************************
BO$ FOR INPUT AS #1
dyn% = 2
INPUT #1, dyn%, I2max%, DEBmax&(dyn%}, D(dyn%), pp(dyn%),
ralon%(dyn%}, filter%(dyn%, horiz%}, filter%(dyn%, e%), filter%(dyn%,
ecrvrt%), filter%(dyn%, convrt%)
'*************************************************************************
OPEN MO$ FOR INPUT AS #1
dyn% = 13
INPUT #1, dyn%, Il3max%, DEBmax&(dyn%}, D(dyn%}, pp(dyn%},
ralon%(dyn%}, filter%(dyn%, horiz%), filter%(dyn%, e%), filter%(dyn%,
ecrvrt%}, filter%(dyn%, convrt%)
CLOSE #1
OPEN M 1$ FOR INPUT AS # 1
INPUT #1, A13, B13, Ul31&, V131&, U2&(dyn%}, V2&(dyn%}, Rdct13,
distrPC13
CLOSE #1
OPEN M2$ FOR INPUT AS # 1
INPUT #1, Ad13, Bd13, Ud131&, Vd131&, Ud132&, Vd132&, Rdcd13,
distrPD13
CLOSE #1
' * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * + * * * * * * + * * + * * * ~ * * *
OPEN NO$ FOR INPUT AS #1
dyn% = 14
INPUT #1, dyn%, Il4max%, DEBmax&(dyn%), D(dyn%}, pp(dyn%},
ralon%(dyn%}, filter%(dyn%, horiz%}, filter%(dyn%, e%), filter%(dyn%,
ecrvrt%), filter%(dyn%, convrt%}
'*************************************************************************
'+++++++++++++++++++++++++++++++++++++++++++++
'Reading of the starting sampling-points DEB&(dyn%} of
'sound-patches in each dyn%-field.
,
.....................-.. _------------------------------_ .. _------------------_ ....-_.. _-----
OPEN prt$ FOR INPUT AS #1
FOR dyn% = dynMin% TO dynMax% 'loop on the dyn%-fields
'for ignored dyn%-fields. IF pp(dyn%} < = 0 THEN
GOTO Ibgl0
END IF
FOR xi% = 1 TO DEBmax&(dyn%} 'loop on the sound/silent
'patches.
INPUT #1, BED&(dyn%, xi%), TH&(dyn%, xi%), DUR&(dyn%, xi%),
THend&, THsec, DURsec, Thendsec
TELOS = TELOS + DUR&(dyn%, xi%) / 44100
NEXTxi%
INPUT #1, THDUR, THDURpcent, sTHend&(dyn%)
TELOS = 0
'++++++++++++++++++++++++++++++++++++++++++++
'the longest of the dyn%-field durations in this sequence {psi%} is:
,
......... _--......-_........__...._---------------------_ ...._------------_ ..- .............
IF megDUR < = sTHend&(dyn%} THEN
END IF
megDUR = sTHend&(dyn%)
'megDUR is the longest dyn%-field
'duration.
1bglO:
NEXTdyn%
CLOSE #1
'++++++++++++++++++++++++++++++++++++++++++++
chD&(psi%) = megDUR
sDURech = sDURech + chD&(psi%)
DURlept = INT(sDURech / (44100 * 60
DURsec = (sDURech / 44100) MOD 60
'cumulation of the longest
'sequence -durations.
'duration in minutes.
'duration in seconds.
'============================================
megDUR = 0
'$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
FOR dyn% = dynMin% TO dynMax%
DEB&(dyn%) = 1
'starting number of the
'sound or
NEXT dyn% 'silence patch for each
'dyn%-field.
'*************************************************************************
CALL WINDO 'screen window.
SECN = SMP& / 44100 'running sample.
SPM& = SMP& MOD 44100 'running seconds.
PRINT SMP&: PRINT SECN: PRINTsampl&: PRINT sec
dyn% = dynMin%
'*************************************************************************
SMP& = 0
fI% = 0
hf& = 0
Kdyn% = dynMax% + 1
TELEN% = 0
'sample number.
'screen amplitude of
'a current sample.
'converter amplitude of
'a current sample.
'check of the dyn%-fields
'amount still availabble.
'for testing the music-piece
end.
'*************************************************************************
'M A I N PRO G RAM M E
Ibg2:
'This part concerns the computation of the amplitude (ordinate) at
'a given sample SMP& by adding up the sound contributions of all
'dyn%-fields in a row from dynMin% to dynMax% with their
'patches DEB&(dyn%),their starting samples TH&(dyn%,DEB&(dyn%) and
'their durations DUR&(dyn%,DEB&(dyn%).This computation defines
'concurrently the amplitude and time elements of the waveform polygones.
IF DEB&(dyn%) > DEBmax&(dyn%) THEN
IF Kdyn% = dynMin% THEN
TELEN% = 1: GO TO Ibg5
ELSE
GOTO IbgO
END IF
ELSEIF SMP& < TH&(dyn%, DEB&(dyn%)) THEN
GOTO IbgO
ELSEIF SMP& = TH&(dyn%, DEB&(dyn%) THEN
IF DUR&(dyn%, DEB&(dyn% < > 0 AND BED&(dyn%,
DEB&(dyn%) = 1 THEN
GOTO Ibg3 'begining of DYNAS[dyn%l
ELSE fh& = 0: hh& = 0
GOTO Ibg5 'no DYNAS[dyn%l
311
ENDIF
ELSEIF SMP& < = TH&(dyn%, DEB&(dyn% + DUR&(dyn%, DEB&(dyn%
AND BED&(dyn% DEB&(dyn%)) = 1 THEN
, GOTO Ibg4 'continuation of DYNAS[dyn%l
ELSElF DEB&(dyn%) < = DEBmax&(dyn%) THEN
IF DEB&(dyn%) = DEBmax&(dyn%) THEN
Kdyn% = Kdyn% - 1
ENDIF
DEB&(dyn%) = DEB&(dyn%) + 1
GOTOlbg2
ELSEIF dyn% < dynMax% THEMyn% = dyn% + 1
GOTOlbg2
ELSE GOTO Ibg6
END IF
IbgO:
IF dyn% < dynMax% THEN
dyn% = dyn% + 1
GOTOlbg2
ELSE
fh& = 0: hh& = 0
GOTOlbg5
ENDIF
---===========
'==============================---
Ibg3:
'contribution ofa dyn%-field DYNAS[dyn%l at the start:
------------------_ .. ---_ ....----_ ..... ----_ .......---_ ....-------
IF dyn% = 1 THEN
ClePenetr% = 1
CALL DYNASl(llmax%, SMP&, Cll&, C12&, t11&, tI2&, 11%, Nl&, fh&, hf&,
hh&)
GOTOlbg5
GOTOlbg5
,
ClePcnetr% = I
CALL DYNASI3(I13max%, SMP&, C131&, CI32&, t131&, t132&, Il3%, N 13&,
fh&, hf&, hh&)
GOTOlbg5
END IF
____ ........ ______ ............................ ___ .. ______ .......... ____________ oo ................................ _---
Ibg4:
,
'Contribution of a dyn%-field DYNAS(dyn%) after DEB&(dyn%) = 1 is ended.
'This is realized with ClePenetr% = o.
-----------------------------------------------------------oo----------------
,
IF dyn% = 1 THEN
CALL DYNASl(Ilmax%, SMP&, Cll&, CI2&, t11&, tI2&, Il%, N1&, fh&, hf&,
hh&)
GOTOlbg5
GOTOlbg5
CALL DYNASI3(I13max%, SMP&, C131&, CI32&, t131&, tI32&, Il3%, N13&,
fh&, hf&, hh&)
GOTOlbg5
END IF
.. _--.............................._----.................._ .............._-------------._--_oo_---------------------
lbg5: . .
, cumulation of amplitudes (ordinates) at a current sample-point SAMP&.
'============================================
'end-test of the usic-piece:
,
-.........._- --------........ ----._-_oo_-----
IF TELEN% < > 0 THEN
SOUND 500, 500 /200
SOUND 2000, 2000/100
GOTOlbg8
ENDIF
============================================
..
10/0 = fI% + fh& 'to screen.
hf& = hf& + hh& 'to converter.
'incremen.ofthe dyn%-field.
,
---_ ..........--------_ .........--_ ....
IF dyn% < dynMax% THEN
dyn% = dyn% + 1
GOTOlbg2
ENDIF
313
'*************************************************************************
Ibg6:
sflt1:
sflt2:
'screen ordinate
U & = 99: V & = -99: Q = fI%
CALL MIRO(U&, V&) 'mirrors: first pair.
10/0 = Q
'vertical screen filter (flrt%(vertcc%= 1))
,,,,,n""11"nl111""""II"""II!11'nrlllnU"""n"n"ntl"nllllnnlllln'"'"IIII",,!!nn
vertec% = 1
IF flrt%(vertec) = 0 THEN
GOTO sflt1
ENDIF
Q = (ffpreel % + fI%) / 2
Q = (ffpreel % + ffprec2% + ff%) / 3
ffprec2% = ffprec1 %
ffprec1 % = ff%
fl% = Q
ord2% = fl%
fl% = 0
'converter ordinate (for file)
U& = 32767: V& = -32768: Q = hf&
CALL MIRO(U&, V&)
hf& = Q
'vertical converter filter (vertcon%=2)
'""n"nllnlllltlllnnllllnll"nn""nnnnl1lln"nllfl""",","II"",Tllllmmllllllllll
vertcon% = 2
IF flrt%(vertcon%) = 0 THEN
GOTO sflt2
END IF
Q= (hfprec1& + hf&) /2
Q = (hfprec1& + hfprec2& + hf&) / 3
hfprec2& = hfprec1&
hfprec1& = hf&
'final ordinate
,
.....----------------_ ..
'filter
'filter
'filter
'filter
'filter
'filter
'filter
'filter
hf& = Q
'***********************************************************************
IF SMP& < 400000000 THEN
'(400000000 is an arbitrary number.)
'for the screen,ifwe wish to show the resultant:
LINE (abs 1%, ord I %)-(abs2%, ord2%)
absl % = abs2%
ordl% = ord2%
SMP& = SMP& + 1
sampl& = sampl& + I
END IF
'global sampling
'point
'change of sequence.
'******************
IF SMP& < = chD&(psi%) THEN
GOTO Ibg9
ELSE
ysp% = ysp% + 1
IF ysp% < yspMax% THEN 'yspMax% =
'maximum number
GOTO Ibgl 'of sequences.
ELSE
GOTO Ibg8
END IF
END IF
Ibg9:
abs2% = sampl& MOD 639 'global screen
'abscissa.
IF abs2% = 0 THEN
absl% = 0
END IF
'++++++++++++++++++++++++++++++++++++++++++++
'every point is now written in the converter file.
,
sample% = hf&
hf& = 0
PUT #3, ,sample% 'in the converter
'++++++++++++++++++++++++++++++++++++++++++++
'chronologies and beep signals.
,
.._--_ ................................__................................. _--
SEGN = SMP& /44100 'prints the seconds
seed = sampl& /44100
SPM& = SMP& MOD 44100
IF SPM& > = 0 AND SPM& 2 THEN
SOUND 1000, 1000/500
ENDIF
IF abs2% = 0 THEN
SOUND 500, 500/200
SOUND 2000, 2000/500
CALL WINDO: PRINT SMP&: PRINT SECN
PRINT sampl&: PRINT secnd
ENDIF
'&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Ibg7:
Ibg8:
dyn% = dynMin%
ff% = 0: hf& = 0
GOTOlbg2
CLOSE #3
END
'the converter
SUB DYNAS13 (I13max%, SMP&, CI31&, CI32&, tI31&, tI32&, 113%, NI3&,
fh&, hf&, hh&)
'This is the 13th dyn%-field subroutine of the main programme that commands
'the contribution of this dyn%-field to the amplitude-ordinate and the time-
'abscissa of the waveform polygone that are sent both to the screen and the
'digital-to-analog sound-converter into the main programme GENDYl.BAS.
SHARED GlePenetr%, Q13, Qd13, dyn%, DEB&O, Nmax&, dynMax%, TELEN%,
Ml$,M2$
SHARED Tab1310, Tab1320, Tad1310, Tad132(), TH&(), DUR&(), DEBmax&(),
U2&O, V2&O
SHARED ralon%(), horiz%, e%, ecrvrt%, convrt%, filter%()
SHARED aampl, campI, mampl, xampl
SHARED aabsc, cabsc, mabsc, xabsc
STATIC eI3&, pI3&, tI3&, fl3precl&, fl3prec2&, hI3precl&, hl3prec2&
STATICA13, B13, UI31&, VI31&, Rdctl3, distrPC13
STATIC Ad13, Bd13, UdI31&, VdI31&, UdI32&, VdI32&, Rdcdl3, distrPD13
IF ClePenetr% = 1 THEN
'Input of the stochastic-distribution coefficients,of the elastic-
'mirror sizes,ofa reduction factor and of the specific stochastic-
'distribution used for computing the amplitude-ordinates of the
'waveform polygone.
OPEN Ml$ FOR INPUT AS #1
INPUT #1, A13, B13, UI31&, VI31&, U2&(dyn%), V2&(dyn%),
Rdctl3, distrPC13
CLOSE #1
'Same kind of input as above but now, for the time-intervals.
OPEN M2$ FOR INPUT AS #1
INPUT#I,Ad13, Bd13, UdI31&, VdI31&, UdI32&, VdI32&,
Rdcdl3, distrPD13
GLOSE#1
316
Formalized Music
ELSEIF ClePenetr% = THEN
GOTOlbl139
ELSEIF DEB&(dyn%) 1 THEN
GOTOlbl137
END IF
Ib1137:
Ib1131:
N13& = 2: PSET (0, 0): C131& = SMP&
IF N13& MOD 2 = 0 THEN
'K% = alternating switch for cumulating in tables:preced. or present period.
Ib1132: '
K% = 2: GOTO Ib1132
ELSE
K% = 1: GOTO Ib1l35
END IF
'first ordinate of the new period = last ordinate of the preceding one.
Ib1135:
Ib1136:
Ib1l33:
f131rl :
113% = 1
p13& = 0
TabI31(K%, 0) = Tab131(K% - I, Il3max%)
TabI32(K%, 0) = Tab132(K% - I, I13max%)
TadI31(K%, I) = Tad131(K%-1, I13max%)
TadI32(K%, I) = TadI32(K% - 1, I13max%)
GOTO Ibll36
TabI31(K%, 0) = TabI31(K% + I, I13max%)
TabI32(K%, 0) = Tab132(K% + 1, 113max%)
TadI31(K%, 1) = TadI31(K% + 1, 113max%)
TadI32(K%, I) = TadI32(K% + 1, 113max%)
'computing the Imax ordinates.
,
CALL PCI3(TabI310, Tab1320, 113%, N13&)
'computing the Imax abscissa-intervals.
,
--------..-......_--------------.. _-----------------------
CALL PDl3(Tad131 0, Tad1320, Il3%, NI3&)
e13& = Qdl3 'horizontal abscissa filter
IF filter%(dyn%, horiz%) = 0
THEN
GOTO f131rl
ENDIF
e13& = (PDprcl31& + PDprcl32& + Qd13) / 3'filter
PDprcl32& = PDprcl31& 'filter
PDprcl31& = Qd13 'filter
IF N13& MOD 2 = OTHEN
K% = 2: GOTO Ibll34
ELSE
K% = 1: GOTO Ibll34
END IF
============================================
Ib1134:
'Drawing the polygone of period Tl13
,
C132& = C131& + e13&
sc% = 639
t131& = Tab132(K%, 113% - I): t132& = TabI32(K%, 113%)
t13& = t132& - t131&
'LINEAR INTERPOLATION OF ORDINATES
'in-between the abscissa C131& and C132&
Ib1139:
'extension of abscissa (= ralon%)
,
----------------_ ...._------------------------_.
IF filter%(dyn%, e%) = 1 THEN
c13& = ralon%(dyn%)
ENDIF
IF p13& > e13& THEN
GOTO Ibl1310
ELSEIF p13& = e13& AND e 13& < > 0 THEN
GOTO Ibl1310
ELSEIF e13& = 0 AND filter%(dyn%, e%) = 0 THEN
c13& = 1
ENDIF
p13& = p13& + 1
fh& = p13& * t13& / e 13& + t131&
:Attack and decay ofa sound-patch.
DIAFA& = SMP& - TH&(dyn%, DEB&(dyn%
IFDIAFA& = OAND DIAFA <= 500 THEN
fh& = fh& * DIAFA& /500
END IF
DIAFDIM& = TH&(dyn%, DEB&(dyn% + DUR&(dyn%, DEB&(dyn% _
SMP&
IF DIAFDIM& <= 1000 THEN
fh& = fh& * DIAFDIM& / 1000
END IF
'Acoustic Normalisation
,
---..............._--------------...._-------
hh& = fh& * 32767 /100
'screen's vertical filter
,
---------------------------
] F filter%(dyn%, ecrvrt%) = 0
THEN
318
f13Ir3:
fl31r4:
Formalized Music
GOTO f131r3
END IF
fIh& = (fl3precl& + h&) /2 'filter
ffh& = (fl3prccl& + fh&)/2 'filter
ffh& = (fl3precl& + fl3prcc2& + fb&) /3 'filter
fl3prcc2& = f13precl& 'filter
fl3prccl& = fb& 'filter
fh& = ffb& 'filter
,
------------------............... _----
'converter's vertical filter
IF filter%(dyn%, convrt%) = 0
THEN
GOTO f131r4
END IF
hhh& = (h13precl& + hh&) /2 'filter
hhh& = (h13prec1& + h 13prec2& + hh&) /3 'filter
h13prec2& = h13precl& 'filter
h13precl& = hh& 'filter
hh& = hhh& 'filter
ClePenetr% = 0: EXIT SUB
============================================
!b1l310:
END SUB
C131& = C132&
:next segment of the period '1113 or next period.
IF 113% < 1l3max% THEN
113% = 113% + 1: GOTO IblI33
ELSEIF N13& < Nmax& THEN
N13& = N13& + 1: GOTO Ibl131
ELSE TELEN% = 1: EXIT SUB
ENDIF
SUB PCI3 (Tab1310, Tab1320, Il3%, NI3&)
'Subroutine of the 13th dyn%-field that computes the amp!itude-
'ordinate of the vertices for the waveform polygone.
SHARED dyn%, Q, SMP&, fh&, hf&, ClePenetr% M1$ prel3I prcl32 U2&()
V2&() , , , , ,
SHARED aampl, camp!, mamp!, xampl
STATIC A13, B13, UI31&, V131&, RdctI3, distrPC13
IF C!ePenetr% = 1 THEN
'Input of the stochastic-distribution coefficients,of the e1astic-
'mirror sizes,of a reduction factor and of the specific stochastic-
'distribution used for computing the amplitude-ordinates of the
'waveform po!ygone.
INPUT #1, A13, BI3, Ul3l&, V131&, U2&(dyn%), V2&(dyn%), Rdctl3,
distrPC13
CLOSE #1
END IF
K%= 1
ELSE
K%=2
END IF
'LEHMER'S random-number generator:
319
xampl = xamp! * aamp! + campI) / mampl - INTxampl * aamp! +
campi) I mampl) .. mampl
z = xampl/ mampl
'Built-in random-number generator:
"CAUCHY:
z = RND
pi = 3.14159265359#: vang = 2 * pi / 44100
DO WHILEz = 0
z=RND
LOOP
IF distrPC13 = 1 THEN
Cauchy = A13 .. TANz - .5)" pi): Ql3 = Tab131(K%, 113%) + Cauchy
ELSEIF distrPC13 = 2 THEN
"LOGIST.:
L = -(LOGl - z) / z) + B 13) I Al3: QI3 = Tab 131 (K%, 113%) + L
"HYPERBCOS.:
hypc = A13 .. LOG(TAN(z" pi / 2: QI3 = TabI3l(K%, 113%) + hypc
"ARCSINE:
arcsin = A13 .. (.5 - .5" SIN.5 - z)" pi: QI3 = TabI31(K%, Il3%) + arcsin
ELSEIF distrPCI3 = 5 THEN
"EXPON.:
expon = -(LOG(I - z / A13: QI3 = Tab131(K%, 113%) + expon
"SINUS:
sinu = AI3 * SIN(SMP& .. vang" B13): Q13 = sinu 'validate coresp.expression
ENDIF
U& = UI31&: V& = V131&: Q = Q13
CALL MIRO(U&, V&)
Q13 = Q
320
IFK% = 1 THEN
ELSE
END IF
Tab131(2, Il3%) = Q13
Tab131(1, 113%) = Ql3
Q13 = Q13 * Rdct13
'Ql3 = Ql3
Q13 = Tab132(K%, 113%) + Q13
U& = U2&(dyn%): V& = V2&(dyn%): Q = Q13
CALL MIRO(U&, V&)
Formalized Music
'valeur filtree 'filter
Q = (prcl31 + Q)/2
Q13 =Q
'Q = (prcl31 + prcl32 + Q) / 3
'prcl32 = prcl31
prcl31 = Q
IF K% = 1 THEN
Tab132(2, 113%) = Q13
ELSE
Tab132(l, 113%) = Q13
END IF
END SUB
SUB PDl3 (Tad1310, Tad1320, Il3%, N13&)
'Subroutine of the 13th dyn%-field that computes the time-interval
'between two vertices of the waveform polygone.
SHARED Q, Qd13, Il3max%, SMP&, fh&, hf&, ClePenetr%, M2$
SHARED aabsc, cabsc, mabsc, xabsc
STATIC Ad13, Bd13, Ud131&, Vd131&, Ud132&, Vd132&, Rdcd13, distrPD13
IF ClePenetr% = 1 THEN
'Input of the stochastic-distribution coefficients,of the elastic-
'mirror sizes,of a reduction factor and of the specific stochastic-
'distribution used for computing the time-interval in-between
'two verices of the waveform polygone.
END IF
INPUT#1,AdI3, Bd13, Ud131&, VdI31&, Ud132&, Vd132&,
Rdcd13, distrPDl3
CLOSE #1
K% = 1
ELSE
K%=2
END IF
'LEHMER'S random-number generator:
'xabsc=xabsc*aabsc+cabsc)/mabsc-INTxabsc*aabsc+cabsc)!mabsc*mabsc
z= xabsc / mabsc
'Built-in random-number generator:
z=RND
pi = 3.14159265359#: vang = 2 * pi / 44100
DO WHILEz = 0
z=RND
LOOP
IF distrPD13 = 1 THEN
"CAUCHY:
Cauchy = Ad13 * TANz - .5) * pi): Qd13 = Tadl31 (K%, 113%) + Cauchy
ELSEIF distrPDl3 = 2 THEN
"LOGIST.:
L = -(LOGl - z) I z) + Bd13) I Ad13: Qd13 = TadI31(K%, Il3%) + L
"HYPERBCOS.:
hypc = Ad13 * LOG(TAN(z * pi / 2: Qdl3 = Tad131(K%, Jl3%) + hypc
"ARCSINE:
'arcsin =Ad13 *(.5 - .5 * SIN.5 - z) * pi:Qd13 =TadI31(K%, Il3%) + arcsin
"EXPON.:
expon = -(LOG(1 - z)) / Ad13: Qd13 = Tad131(K%, Il3%) + expon
"SINUS:
321
sinu = Ad13 * SIN(SMP& * vang * Bd13): Qd13 = sinu 'validate coresp.expression
ENDIF
U& = UdI31&: V& = Vd131&: Q = Qd13
CALL MIRO(U&, V&)
Qd13 = Q
IFK% = 1 THEN
ELSE
ENDIF
Tad131(2, Il3%) = Qd13
Tad131(1, Il3%) = Qd13
Qd13 = Qd13 * Rdcd13
'Qd13 = Qd13
Qd13 = Tad132(K%, Il3%) + Qd13
U& = UdI32&: V& = Vd132&: Q = Qd13
CALL MIRO(U&, V&)
Qd13 = Q
IFK% = 1 THEN
ELSE
END IF
END SUB
Tad132(2, 113%) = Qd13
Tad132(1, 113%) = Qd13
SOH:S352;yspy; 7;psiY; 8;Pge: 3;dUf,Nin:Sec: 2:13; tot.Nin : Sec: 2:13
129 130 149 150 169 170 189sec
, , , , 1 , , , , , , , 'I
, .... , .... , .... , .... , .... , .... , .... , .... , .... , .... , .... , .... '2
_,,'11111111"'13
I" I I" . I" . I" . I" .. I" . I" ., I" I" . I" .. I' . 14
I' I I I I' I I' I' .. I' .. I' 1 I 15
, .... , .... , .... , .... , .... , .... , .... , .... , .... , .... , .... , .... 16
, , , , I , , I 1 , , 1 '1
I 1 ... 1 ... 1 .... 1 .... I 1 .... 1 .... I' 1 .... I 1 .... 18
I ... 1 .... 1 .... 1 .... , ... 1 .. , .... , .... , .... , .... , .... , .... '9
, , I" , , , , , , , , , '18
....... , '1' , , , , , . , , . , 'II
I' I I I .. I .. 1 I . I .. , . , I , 112
I , , 1 . , . I I I I I I I 113
I I I I I I I ... I .. I . I ... I ... I 114
, 1 , I I' , I , , , 1 , '15
, . , I , ... I . I 1 .... , .... , .... , .... 1 .... , .... 116
SOH:S352;yspl.: 7;psil.: 8;Pge: 2;dup,Min:Sec: 2:13; tot.Min : Sec: 2:13
69 79 89 99 1ml 118 12Qsec
I I - - - - + - - - - + - - - - ~ - - ~ - - - - ~ - - ~ - - - - ~ , --1' 'I
t----_--_- , .... 1 , , 1 , , , , ,2
1---------+-1 I , I .... , . , I 1 . I ---43
I I '1-' '1'" '1'" '1'" ., ... '1'" ., ... 1 .14
1-- I .... 1 .... 1 ... -+ .... I .... I .... I . -' . 1 .... 1 .... I .... 1 .... IS
\---_--_-. 1 .... 1 .... , .... , .. , , I ' , , , , , 1 .... I , 16
1 .... I 1 . 1 .. , .... 17
1 I I I I I I I I 18
-1_11111111119
I .-. I I I I ~ . I ... _ . I .. I ... I . I .. 118
--+-1 _ ... I I , I I 111
""","-+--' I I .. I .. 1 I I , , I' , . 112
I .. I ... I .... , .... I .. 1 , .. , . , , 1 , . 113
I I ~ .. ' 1 .... 1 .... ' .... 1 .... 1 .... 114
~ I 1 .. 1 .. 1 1 .... 1: .... 1 .. .1 .. 115
1---------1 .... , .... 1 . 1 .... I I . I .. 1 .... 116
Two pages of the "score" resulting from the programme reproduced here.
First law
Appendix I
[7] [20]
TWO LAWS OF CONTINUOUS PROBABILITY
Let OA be a segment of a straight line of length l on whieh we place n
points. Their linear density is e = nil. Suppose that land n increase in-
definitely while e remains constant. Suppose also that these points are
numbered AI, A
p
, A
q
, and are distributed from left to right beginning at
the origin O. Let
The probability that the ith segment will have a length Xi between X
and x + dx is
Now the probability Pn, that there will be n points on a segment x, is
given by the recurrence formula
Pn+l ex
p;:=n+I'
Therefore PI = (ex/l )Po. But Po = e-
CX
and
e- CX = I _ ex + (CX)2 _ (CX)3 + ...
I! 2! 3! .
If X is very small and if we denote it by dx, we have
c2(dx)
2
Po = I - c dx + ~ .. '.
323
324
Formalized Music
Since the powers of dx are infinitely small for high values, Po = I - c dx
and PI = c dx Po = c dx. Hence, the probability P
x
is composed of the
probability Po = e-cX, that there will be no point on the segment x, and the
probability PI = c dx, that there will be a point in dx.
APPROXIMATE CALCULATION OF THE SAME PROBABILITY
(FOR CALCULATION BY HAND)
Let there be d points to be placed on a straight line of length t. The
linear density is c = d/l points on length I. If the lengths are expressed in
units v then I = av (a > 0) and cv = d/a points in the unit of length v.
Then XI = iv (i = 0, 1,2,3, ... ), and the probability, the asymptotic
limit of the relative frequency of the segment XI' will be
(1)
We shall now dcfine the quantity t-x
i
. The probability (1) is composed
of the probability Po = e-
CiV
, that there will be no point on Xi> and the
probability PI = ct-x
i
, that there will be a point in t-Xj if (et-xi) 2 is small
enough to be ignored. Set
o < (Ct-Xt)2 < 1O-n.
where n is a sufficiently large natural number; this expression becomes
0< t-Xi < c-
1
.1Q-n/2.
Substitute a constant z for t-x! such that for every Xi
z::; t-Xi < c-
1
1O-
n
/
2

Then equation (1) is written
P
X1
= e-civcz
and must satisfy thc condition
or
i= co
2: e-c1vcz = 1,
i= 0
t= co
Z = 1/c-2: c-civ.
;=0
But since cv > 0, e - cv < 1, so that
i = 00
2: (e-
CV
)! = 1/(1 - e-
CV
)
i=O
and finally
z=---
c
(2)
(3)
Appendix I
325
Now from (2)
_ e- CV lo-n/2
---<--.
e e
Therefore
then
(1 - 1O-n/2) < e-
CV
< 1.
Thus, for cv > 0 we have e-
CV
< 1, and for cv < -log (1 - 1O-n/2) we
have e-
CV
> (1 - 1O-
n
/
2
). And since 0 < 1O-n/2 < 1 we have
1O-(n/2)2 1O-(n/2)3 10 -(n/2)4
-log (1 - 1O-n/2) = 1O-
n
/
2
+ + _-:::--_ +
2 3 4
+ ...
and
1O-n/2 < - log (1 _ 1O-n/2).
In order that e-
CV
> 1 - 1O-
n12
it is therefore sufficient that
Then we may take
cv ::; 1O-n/2.
1 - e-
CV
t-Xj = z = ---
c
(4)
(5)
and substitute this value in formula (1), from which we can now set up
probability tables. Here is an example:
Let d = 10 points as mean value to be spread on a straight line segmeIlt
oflength I = 100 cm. Wc have to definc x and P . as a function of i <rivcn
I Xl , "
that (Ct-XI)2 = 10-
4
is considered to be negligible. -
. From (4), CD = 10-
412
= 0.01 points in v. Now c = d/l = 10/100
pomts/cm, therefore c = 0.1 points/em, v = 0.01/0.1 = 0.1 cm, and x =
O.li cm ~ i mm. !
From (5),
I - e- O.
01
t-Xj = ---;:--=--
0.1
From (1),
(1 - 0.9905)10 = 0.0995 ~ 0.1 cm.
P
X1
= e-
0
.
oli
0.l0.1 = 0.01 (0.099005)i.
For calculation by machine see Chapter V.
Second Law
f(j) d} = (I - d}.
Each variable (pitch, intensity, density, etc.) forms an interval (dis-
tance) with its predecessor. Each interval is identified with a segment x
taken on the axis of the variable. Let there be two points A and B on this
axis corresponding to the lower and upper limits of the variable. It is then
a matter of drawing at random a segment within AB whose length is
included betweenj andj + dj for 0 j AB. Then the probability of this
event is:
Pi = J(j) d} = (I - dj' (1)
for a = AB.
APPROXIMATE DEFINITION OF THIS PROBABILITY FOR CALCULATION
BY HAND
By taking dj ,a's a constant and j as discontinuous we set d} = c,} = iv
with v = aim for i = 0, 1,2,3; ... , m. Equation (1) becomes
(2)
'.l
But
'2c
,I Pi = - (m + 1)
" a
1,
whence
d
. a
U=c=m+I'
On the other hand Pi must be taken as a function of the decimal
approximation required:
2 ( i)
Pi = m + 1 1 - In lO-n (n = 0, 1,2,3, ' .. ).
Pi is at a maximum when i = 0, whence m 2 IOn - 1; so for m =
2lO
n
- 1 we have v = al(2lO
n
- 1) and dj = a/(2lO
n
), and (1)
becomes
1 ( i)
P
j
= Pi = Ion 1 - 2. Ion - 1 .
Appendix II
DEFINITION OF THE SAME PROBABILITY FOR COMPUTER
CALCULATION
327
We know that the computer can only draw numbers Yo at random (of
equal probability) 0 Yo 1. Using the probability law of density Pi
f(j) d}, we have for some interval Xo
. fX
o
. 2xo xg F()
prob. (0 J xo) = f(J) dJ = - - 2 = xo,
o a a
where F(xo) is the distribution function of}. But F(xo) = prob. (0 y Yo)
= Yo' Therefore
2;0 _ = Yo and Xo = a[l y(1 - Yo)],
and by rejecting the positive root, since Xo must remain smaller than a, we
obtain
for all ::5: Xo ::5: a.
Xo = a[1 - yI(l - Yo)]
Appendix II
, [14]
':""1.-
",'2,," "'.' :J!
.'
, Let there be states E
1
, E
2
, E
3
, , Er with r < 00; and let one of these
events necessarily occur at each trial. The probability that event Ek will
take place when Eh has occurred at the previous trial is Phkl
,IPM = 1, with k = 1,2",., r.
k
is the probability that in n trials we will pass from state Eh to state Ek ;
2: pi,nJ = 1, with k = 1,2, ... , r.
k
If for n --+ r:f) one of the tends towards a limit P
hk
, this limit is
expressed by the sum of all the products PhjPjk, j being the index of one of
the intermediate states E
j
(1 j ::5: r):
P hk = Ph1hk + P h2P2k + ... + PhrPrk'
328
The sum of all the limits PilI< is equal to I:
PhI + p
h2
+ p
h3
+ ... + P
hT
J.
\"ie can form tables or matrices Din) as follows:
Din)
pin)
11'
pin)
21,
pin)
2m'
.. "'
.. "'
.. "'
pin)
TI
pin)
Tm
pin)
TT
Formalized AIusic
Regular case. If at least one of the tables Din) contains at least one line
m of which all the elements are positi\'e, then the Ph"J ha\'e limits Ph'" and
among the PhI< there exists at least one, Pm, which has a non-zero limit
independent of n and of h. This is the regular case.
Positive regular case. If at least one of the tables Din) has all positi\'c
clements, then all thc P
hk
ha\'c non-zero limits Pic indcpcndent of the initial
index h. This is (lIe j)ositiuf regular case.
The probabilities P
k
= X
k
constitute the system of solutions of the
r + 1 equations with r unknowns:
XI = XIPll + X2PZI + ... + XTPrl
X2 = Xl Pl2 + X2P22 + ... + XTPTZ
X3 = XI P13 + X2P23 + ... + XriJT3
XT = XlPlT + XZP2T + ... + XTPTT
1 = Xl + X
2
+ ... + Xr
But these equations are not independent, for the sum of the first r equations
yields an identity. After the substitution of the last equation for one of the
first r equations, there remains a system of r equations with r unknowns.
:\Tow there is a demonstration showing that in the regular case the system
has only one solution, also that Din) = Dn (nth power of D).
Appendix III
THE NEW UPIC SYSTEM *
Introduction
UPIC (Unite Polyagogique Informatique du CEMAMu)l is a machine
dedicated to the interactive composition of musical scores. The new and
final version of this system runs on an AT 386 microcomputer connected to
a real-time synthesis unit. The new software offers a mouse-controlled,
"user-friendly" window style graphical interface and allows real-time
drawing, editing and playing of a musical page as well as the recording of a
"performance."
Description
The UPIC is a music composing system which combines a graphic
score editor, a voice editor and a powerful "performance" (or play-back}
system, all sharing the same data. Therefore, all drawing and editing
operations are available while the music plays. All the commands are
mouse driven. A menu command allows one to switch the drawing input
device from the mouse to the digitizer and vice versa.
A UPIC score is a collection ogfnotcs that are called "arcs." An arc is a
pitch (frequency) versus time curve. The frequency variations are continu-
ous and can cover the whole ambitus. The durations can range from 6 ms
to the total duration of the musical page (1 hour maximum).
lCEMAMu (Centre for studies in mathematics and automation of music) ,founded
by Iannis Xenakis in 1965 with grants from the French Cultural Ministry.
*This appendix is freely inspired by a similar paper published by ICMC in
Glasgow, 1990 in "Proceedings," written by Gerard Marino, Jean-Michel Raczinski,
and Marie-Helene Serra ofCEMAMu. My gratitude for their faithful dedication is
herewith expressed. (LX.)
329
Tools are provided for obtaining quantified values of frequency and
duration. In this way, the notion of an arc is an extension of the classical
notion of a note. In addition, each arc has a set of sound attributes that can
be changed real-time, during playback.
Voice editing on the UPIC includes redrawing and redefinition of
waveforms, envelopes, frequency and amplitude tables, modulating arc as-
signment, and modification of audio channel parameters (dynamic and en-
velope). All these operations are feasable during playback and immediately
heard.
Different sound interpretations of the same graphic score may be
tested with the help of arc groups. Groups contain from one arc to the
whole page and allow instantaneous and global modifications of sound par-
ameters (waveform change, transposition, etc.).
During performance, the musician can switch from one page to
another and may control the tempo and play position by moving the mouse
across the page. The resulting live interpretation may be recorded in an
editable object called a "sequence." The tempo and the position in the
sequence is controllable while the sequence is being played.

o Play
@
[8J Pause
Jinuelope
EHU1
filiO n
2111
table
(".; FRQTB12
.9.llIpl. table
EHUTBDB
Channel
rn
-2 dB!
let iiil"
Li"'its------, rfos1t1on--w,
I
l.t.I 11ft : 15 :35 :'9li1
QO : 111 : 00 .00
00
. l:8I.s.top
1011 :12: 115 .2afll I ll.K )
o [ !!.ancel )

reilliil
i L-______________________
Figure 1. Sample screen from UPIC
Appendix III: The New UPIC System 331
Page Drawing and Editing
A maximum of four pages of music can be opened and displayed in
moveable and resizeable windows. Opening a page stored on the disk loads
it into the memory of the real-time unit. Therefore, all the subsequent
operations can be carried out while the page is being played.
Arcs can be drawn by using one of the drawing modes (free hand,
broken line, etc.) If accepted, an arc is inserted in the page as soon as its
drawing is over; if the limit of 64 oscillators is reached, the arc will be re-
fused. At any time, it is possible to modify the set of the default attributes
(waveform, envelope, frequency table, amplitude table, weight, modulating
are, audio channel). One page holds a maximum of 4000 arcs.
Usual editing commands (cut, copy, paste) are available. For each
page, four groups of any number of arcs can be created by using different
types of selection (block, list, criteria) Geometric operations like symmetry,
rotation and vertical alignment can be applied to a group. Instantaneous
modifications of the attributes (waveforms, envelope, frequency table,
amplitude table, weight, modulating are, audio channel) of the arcs belong-
ing to a group can be temporarily applied and saved, if necessary. Further-
more, groups can be instantaneously muted, "solo-ed," and/or transposed.
Voice Edition
Each arc is associated with an oscillator whose configuration is given
by the following arc attributes: waveform, envelope, modulating are, audio
channel. Before being transmitted to the oscillator, the graphic data of the
arc and of the envelope are converted respectively by a frequency table and
an amplitude table.
Waveforms and envelopes can be drawn or extracted from sampled
sound, and normalized.
The contents of the conversion tables are defined either by a drawing
or by a menu command and arc redrawable.
The frequency table definition menu command enables the user to set
the boundaries of the ambitus (in hertz or half-tones) and the musical scale
parameters (tuning note and number of equal divisions in the octave). The
frequency table can be inverted and can be made continuous or discrete. In
the latter case, the steps are the octave divisions. When played with a dis-
crete frequency table, the pitch variations within the arcs follow the
frequency steps of the table.
332
Formalized Music
Performance
Only one page can be played at a time. The four pages maximum in
the window may be chained or not. The user chooses which page to play
simply by clicking on it, stops or restarts the progression of the perform-
ance, defines the time limits of the performance with optional looping.
The tempo and play position can be defined by mouse motions on the
page or by entering their values. All types of motions (forward, backward,
jumps, acceleration, slowing down) within the page are permitted. When
not user-controlled, the page is played at a constant tempo.
A set of channel parameters (dynamic and envelope) is assigned to
each page. The dynamic and envelope of the 16 output audio channels are
real-time controllable during performance. As the channel envelope
spreads over the whole page, it is therefore possible to locally weight arcs
assigned to a given channel.
In the UPIC, a sequence is the recording, during the performance
(controlled or not) of all the successive positions in the page, with a 6 ms ac-
curacy. It holds a maximum of 12 minutes of performance. It is displayable
as a position versus time curve. Any piece of the sequence can be overwrit-
ten by a new recording or redrawn. The performance of a sequence is car-
ried out inside its window with mouse motion controls (like the page itself).
When four pages are loaded, the user has two sequences with which to
work.
Storage
Pages, waveforms, envelopes, conversion tables and sequences are
stored in separate banks (DOS files) on disk. Banks are user-protected.
Copying, renaming, and deleting objects and banks is possible.
The user can load objects that come from different banks. Saving an
object can be done in any bank.
Conclusion
This summarizes the principal characteristics of the UPIC system
today. Additional commands are going to be integrated to the application,
especially sampling utilities (record, play, simple edition fuctions) The syn-
chronization of the performance with an external device as wellas the com-
munication between UPIC and MIDI devices is presently being st-iIdied.
Tools will be provided to allow another application access the data ofUPIC
banks.
The system is being industrialized and will be commercialized in the
course of 1991.
Appendix III: The New UPIC System
TECHNICAL DESCRIPTION
A) Hardware Specifications
Host computer
333
PC-AT 386 with 3 Megabytes memory minimum, hard disk, mouse,
MIDI board, optional digitizer tablet. All Summagraphics compatible digi-
tizers are supported (size AO to A4).
Real-time synthesis unit
64 oscillators at 11.1 kHz with FM
(future extension to 128)
converter board:
4 audio output channels
2 audio input channels
AES/EB U interface
(extension to 4 converter boards)
capacity:
4 pages of 4000 arcs
64 waveforms (4K entries)
4 frequency tables (16K entries)
128 envelopes (4K entries)
4 amplitude tables (16K entries)
2 sequences (12 minutes each, 6 ms accuracy)
B) Software Main Features
Environment DOS with Microsoft WINDOWS 3.x (graphical
multi-application environment with pull-down menus and pop-up
windows)
Storage pages, waveforms, envelopes, frequency tables, amplitude
tables and sequences are stored in separate banks on disk. Banks are
user-protected.
Drawing every object is initialized either by a command or by a
drawing, and is redrawable. Objects are displayed in overlapped, resizable
and zoomable windows.
Edition : several types of selection (block, list, criteria) allow the creation
of up to four groups of arcs per page. Each group can be muted, solo-ed,
graphically transformed and real-time controlled.
Sound :: See C (Real-time controls)
334
C) Real-Time Controls
Page controls
Tempo
Play time interval (with or without looping)
Page switching
Position in the page
For each audio channel: dynamic, envelope
Sequence controls
Tempo
Position
Sequence switching
Group controls
Solo
Mute
Transposition
Intensity
Frequency modulation
Output channel
Waveform (among 64)
Frequency table (among 4)
Envelope (among 128)
Amplitude table (among 4)
Drawing while playing
Formalized Music
While a page is being played, the user can modify its waveforms, en-
velopes and conversion tables.
A new arc can be heard as soon as its drawing is finished.
An existing arc can be redrawn within its endpoints and heard at the
same time.
A Selected Bibliography of
lannis Xenakis *
compiled by Henning Lohner
The bibliography of dle works of lannis Xenakis is arranged as fol-
lows:
1. Works by Xenakis:
1. Books, and
2. Articles in periodicals, booklets, and encyclopedias. In general, the
source of the first printing is given. The writings of Xenakis have been
translated and reprinted many times.
II. Writings about Xenakis:
1. Monographs and collections, the majority of which are dedicated to
Xenakis, and
2. Articles in periodicals, encylopedias. and anthologies.
Works in daily and weekly newspapers, wiili some minor exceptions,
are not included here. That applies, in particular, to ilie numerous festival
catalogs and concert programs, which frequently have original material.
Likewise, record notes by or about Xenakis, record reviews and introduc-
tions to works by Xenakis are excluded.
It should be mentioned here iliat Xenakis, as an independent archi-
tect and long-time collaborator of Le Corbusier is responsible for an exten-
sive, architectural body of work, which is noted in the literature. We can, at
present, include only the most important, useful, interdisciplinary works.
The writings are organized chronologically, and, within each year, ar-
ranged alphabetically. Articles without a specific author are listed by the
initial letter of the tide.
This bibliography is based on the private collection of the author, and
also utilizes other relevant bibliographies on the subject at hand.
The list of abbreviations can be found at the end, p. 364.
*First appeared in Musik Texte No. 13, Cologne 1986, in German, and later updated
in MusiklKonzepte 54/55.
335
336
I. Primary Bibliography
1. Books by Xenakis
Formalized Music
1. 1.: Collected Writings
"Les musiques formelles; nouveaux principes formcls de composition
musicale." Revue musicale 253-254, 1963; published also as:
Musiques formelles. Paris 1963; revised and expanded text in English
translation: Formalized Music. Bloomington and London, 1971.
Musique. Architecture. Collection "Mutations- Orientations" Vol. 11,
edited by M. Ragon, Tournai (Belgium) 1971; 2. expanded and re-
vised edition 1976.
1. 2.: Dissertations
Arts/Sciences-Alliages, these de doctorat es leures et sciences humaines 1976.
Paris 1979; Arts/Sciences: Alloys, English edition, New York 1985.
2. Articles by Xenakis
1955 "Provlimata ellinikis moussikis." Epitheorissi uchnis 9 (Athens 1955):
185-189.
"Le Couvent de la Tourette", in: Modulor 2, by Le Corbusier, Paris
1955.
1956 "La crise de la musique serielle." Gravesaner Bliitter 1 (July 1956): 2-4.
"Wahrscheinlichkeitstheorie und Musik." Gravesaner Blatur 6 (1956):
28-34.
"Brief an Hermann Scherchen." Gravesaner Bltitur 6 (1956) 35-36.
1957 "Le Corbusiers 'Elektronisches Gedicht'/Le Corbusier's 'Electronic
Poem.'" Gravesaner BltterlGravesano Review 3, No. 9 (1957):
43ff./51[(
"Le Couvent d'etudes de la Tourette, oeuvre de Le Corbusier," Art
Chretien No.6, Paris 1957: 40-42.
1958 "Aufder Suche nach einer Stochastischen MusiklIn Search ofStochas-
tic Music." Gravesaner BliitterlGravesano ReviroJ 3, No. 11-12 (1958):
98ff./112Lf.
"De tre parablerna." Nutida Musm 2 (1958-1959).
" G e n ~ s e de I'architecture du pavilion: Ie pavilion Philips a I'Exposition
urnverselJe de Bruxelles 1958." Revue uchnique Philips 20, No. 1
(1958): 10 p.
A Selective Bibliography-Writings 337
"Reflektioner over 'Geste Electronique.'" ("Notes sur un 'Geste Elec-
tronique'" [in Swedish. Nutida Musik 1, March 1958; French origi-
nal: Revue musicale 244 (1959) 25-30.
"Le Corbusier", in: Archiucture, Paris 1958.
"The Philips Pavilion and The Electronic Poem" (summary by
L.C.Kalfl),ArtsandArchitecture 75, Nr.111958: 23.
1960-61 Grundlagen einer stochastischcn MusiklElements of Stochastic
Music." 4 Issues, Gravesaner BlttturlGravesano Review 5-6, No. 18
(1960): 61 [(/84[f; No. 19-20 (1960): 128ff./140ff.; No. 21 (1961):
102ff./113ff.; No. 22 (1961): 131ff./144ff.; French original in part in
Revue d'esthitique 14, No. 3-4 (1961). "Herman Scherchen." In Enzy-
cwpedie de la Musique. Paris, 1961, p. 653.
"Vitruve." In Enzyclopedie de la Musique." Paris 1961, p. 873-874.
"Hermann Schcrchcn", in: Enclopedie de la Musique, Paris 1961: 653.
"Stochastic music", in: (Proceedings of the) Tokyo East-West music enr
counter, Tokyo 1961.
"The riddle of]apan", in: ThisisJapan, Tokyo 1961.
1962 "Debussy a sformaliwwanie muzyki." Ruch Muzyczny 6, No. 16 (1962):
7.
"Elements sur les procedes probabilistes (stochastiques) de composi-
tion musicale." In Panorama de l'art musical ccmumporain. Edited by
C. Samuel. Paris 1962, p. 416-425.
"Un Cas; la musique stochastique." Musica (Chaix) 102 (Sept. 1962):
llfE
"Stochastische Musik/Stochastic Music." Gravesaner Blatur/Gravesano
Review 6, No. 23-24, p. 156ff.(169ff.
"Wer ist lannis XenakislWho is lannis Xenakis." Gravesaner Bliit-
ter/Gravesano Review No. 23-24 (1962): 185/185-186.
1963 "Musiques formelles." Bulletin de souscription, Paris 1963, p. 1.
"Pierre Schaeffer." In Die Musm in Geschichte und Gegenwart (MGG),
Vol. 11, Kasse\1963, p. 1535. German translation by M. Bente.
1965 "Freie Aussprache (Diskussion): I. Naturgetreue Musikwiedergabe, II.
Mathematik, Elektronengehirn u. Musikalische Komposition/Open
Discussion: I. High Fidelity, II. Mathematics, Electronic Brains and
Musical Composition." Gravesaner Blatur/Gravesano Review No. 26
(1965): 3-4/4-5.
338
Formalized Music
"Freie stochastische Musik durch den Elcktronenrechncr/Free Stoch-
astic Music from the Computer." Gravesaner BlliUTIGravesano Review
7, No. 26 (1965): 54ff./79ff.
"La voie de la recherche et de la question." Preuves 177 (Nov. 1965).
"Der Fall Lc Corbusier/Concerning Lc Corbusier." Gravesaner Bliit-
ter/Gravesa1W Review No. 27-28, p. 5-7/8-10.
"La Ville Cosmique." In L'Urbanisrne, 'I11opies et nfalilits by F. Choay,
Paris 1965.
"Zu einer Philosophie der Musicffowards a Philosophy of Music."
Gravesaner Bllitter/Gravesano Review No. 29 (1965-1966): 23 ff./38ff.
French original: "Vers un philosophie de la musique." Revue d'esthe-
tiqW! 21, No. 2--4 (1968): 173-210.
"Lc Corbusier", AuJourd'hui, Art et Architecture No. 51, 1965.
"Motsagelsen musik och maskin", N'I11ida, Musik 9, No. 5-6: 23.
1966 "The Origins of Stochastic Music." (translated and expanded excerpt
from "Musiques formelles"), Tempo 78, p. 9-12.
"Structures hors-temps" The musics of Asia. Papers read at izn inter-
national music .symposium, Manila 1966: 152-173.
1967 "Ad libitum." WOM 9, No.1 (1967): 17-19.
"Iannis Xenakis." In Die Musik in Geschichte und Gegenwart (MGG) Vol.
14 (1967-1968): 923-924; translation by D. Schmidt-Preup.
"Musikalisk axiomatik och formaliseringiAxiomatique et formalisation
de la composition musicale." Fylkingen Bulletin International 2 (1967):
3 p.
"Vers un metamusiquc." La Nef29 (1967): 24 p.
1968 "La Musique et les ordinateurs." Edited by A Capelle. La Quinmine lit-
teraire, 1 March 1968.
"Temoignage d'un createur" (Interview), Pensee et creation, Paris 1968:
78-83.
1969 "Problemy mojej techniki kompozytorskiej." In Horyzonty Muzyki 1,
Warsaw 1969.
"Structures universelles de la pensees musicale." In liberti et organisa-
tion dans le monde actuel, Paris 1969.
"Une note/E.m.a.mu." Revue musicale 265-266 (1969): 8 p.
1970 "Short Answers to Difficult Questions." Composer (USA) 2, No. 2
(1970): 39ff. 1971
"Den Komiska varldsstaden." N'I11ida Musik 15, No.3 (1971-1972): 13-
H.
"Les Dossiers de rEquipe de Mathematique et Automatique Musicales
(EMAMu)." ColoquioArtes 13, No.5 (1971): 40--48.
"Musique ct programmation", ITC (Ingenieurs, Techniciens et Cadres) Ac-
tualizes 2, 1970: 55-57.
"Om Terretektorh." Nutida Musik 15, No.2 (1971-1972): 47.
Preface to L'iniliation musicale des jeunes by M. Gagnard, Paris 1971.
"Stravinsky and Tradition: First Thoughts." In "Stravinsky (1882-
1971), A Composers Memorial." PNM 9, No.2 (1971): 130.
"Structures hors-temps." In The Musics of Asia, Manila 1971, p. 152-
173.
"Une erreur fossile", Pourvous qW! est jesus Christ?, Paris 1970: 150.
1972 "Formalized Music Abstract." InRILM 6, No.3 (1972): 266.
1974 "New Proposals for Microsound Structure" (in Japanese). In Technology
kukan, Transonic 4, Autumn 1974, p. 4ff.
"Propos impromptus, suivis de reflexions en marge." Edited by R.
Lyon. Courier musicale de France 48 (April 1974): 130-133.
1975 ("Wissenschaftliches Denken und Musik-in Greek"). TheatTO 8, No.
46--48 (1975): 14 p.
"The New Music Today." In Inter Nationes, Bonn 21/1975.
"Zur Situation", Darmstlidter Beitrage zur Neuen Musik, Bd.xIV, Mainz
1975: 16-18.
1976 "Archaioteka kai sygchrone mousike, (antike Zeiten und zcitgen.
Musik-in Greek)." Bulletin of Critical Discography 18-19 (January
1976): 374-382.
"Culture et creativite." Culture 3, No.4 (1976): 4 p.
1977 "Des Univers de Son." In Problemes de la musique moderne, by B. de-
Schloezer and M. Scriabine. Paris 2/1977, p. 193-200.
"L'universo politico e sociale." In Musica e politica, La Biennaledi
Venezia. Venice 1977, p. 548-549, translation by A Cremonese.
"Musique et architecture." Artcurial5 (1977): 1 p.
"Nouvelles positions sur la micro-structure des sons", Dossiers-Arts
Plastiques 1, Paris 1977.
340
Formalized Music
"Sur l'architecture au Japon" (inJap.), Ikebana sogetsu No. 113, 1977.
1978 "Centre Georges Pompidou: Geste de lumiere et de son." Le Diatope-
Xenakis, edited by 1. Xenakis. Catalog: Paris 1978.
"Quelques systemes diversifies en composition musicale", ColLoque de
St.Hubert-Le recit et sa repesentatiem, Paris 1978.
1979 "Bela Bartok." Arion (December 1979).
"Opening Address", Proceedings oftlw 1978 International Computer Music
Cemjerence, Vol. 1, C. Roads, Evanston 1979.
"Si la societe etait un homme, ferait plusieurs metiers." Entretien avec
Carine Lenfant. Architecture 7 (August-Sept. 1979).
1980 "Between Charybde et Scylla." Spirali 11 (in Italian, December 1980);
Spirales 1 (in French, February 1981).
"Brief an Karl Amadeus Hartmann" (1956, in English). In Karl
Amadeus Hartmann und die Musica Viva. Mainz 1980; 337.
"Migrazioni nella composizione musicale." In Musica e elaboratore, La
Biennaledi Venezia. Venice 1980.
"Spaces and sources of auditions and spectacles". (English summary,
M. Griech.), Proceedings offirst meeting: enlargenmenJ of theatrical activi-
ties and architectural pactice, International scientific symposium, Volos
1980: 203-212
1981 "Dialexi." In Symbossiem: synchroni techni kai paradossi, convention re-
port. Athens 1981, p. 195ff.
"Homage to Bela Bart6k." Tempo 136 (1981): 5.
"II faut se que <;a change!" In the series "I'IRCAM: un monopole con-
teste par des compositeurs." I.e Malin (26 January 1 981).
"Le miroir du compositeur" (Interview), Spirales No.7, 1981: 20-21.
"On the Nude Body" (statement). In Body Print, edited by M. Tanaka.
Japan 1981.
"Temps en Musique (Time in Music)." Conference given in New
York. Spirales (December 1981).
1982 "Andre Schaeffner." Revue de musique 68 (1982): 387.
"Demain, les compositeurs seront tous des cervaux." Edited by J. N.
Von der Weid. Mcmde de la Musique (Feb. 1982): 66-68.
"n pensiero musicale", Spira Ii No. 41,1982: 44-45.
"La composition musicale est a la fois dependente et indCpendcnte de
revolution technologique des systemes analogiques ou numer-
iques." Son et Image, Video convention report. Porte Maillot, 1982.
Cemference des Joumies d'Etudes (1982).
"Musica e originalira", Numero e suemo, !JJ, Biennale di Venezia, 1982: 41-
42; published in French, "Musique et originalite, Phreatique No. 28,
Friihling 1984: 62-66.
"Polytopes", Festival d'Automne Paris 1972-1982, Paris 1982: 218.
Preface to Informatique musicale et pedagogie, Les cahiers del'arm NO.4 (no
year, ca. 1981-1982): 3.
"Science et technologie, instruments de creation", Colloque natiemal: re-
clwrclw et technologie, Paris 1982.
1983 "II faut se debarrasser des prejuges architecturaux." Les Nouvelles litter-
aires 23-29 Gune 1983): 40-41.
"Perspectives de la musique contemporaine", Echos, No.1 1983: 47.
"Pour saluer Olivier Messiaen", Opera de Paris No. 12, 1983: 6.
1984 "n faut debarasser des prejuges architecturaux." Nouvelles litteraires
(25-30 May).
"Musique et Originalite" in French and English, July-August, 1982 .
Phreatique 28 (I 984).
"The Monastery of La Tourette." In I.e Corbusier Vol. 28. Garland New
York,1984.
"Notice sur la vie et les travaux de Georges Auric", in: Discours pro-
nemces dans La seance publique tenue par l'Acadimie des Beaux-Arts No.6,
Paris 1984: 13-19.
"Un exemple enviable." Revue musicale 372-374 (1981): 67.
"Pour l'innovation culturelle", in: Vous avez dit Fascisme?, by R.
Badinter, Paris 1984: 275-276.
"Un plaidoyer pour I'avant-garde?" I.e Nouvel Observateur, 19-25 Oct.
1984: 97.
1985 "Alban Berg; Ie dernier des Romantiques",!JJ, vie w/l71reile, 7. 7. 1985:
29.
"Le pas d'acier de Paul Klee." I.e Nouvel Observateur (15-20 Novem-
ber).
"Les conditions actuelles de la composition" (Zusammengetragen v.J.-
P. Bady), France Forum No. 223-224,1985: 10-12.
"Music Composition Treks." In Composers and the Computer, edited by
C. Roads. MIT Press, Cambridge, Mass, 1985; already written 1983,
and in French translation by E. Gresset published in Musique et ordi-
nateur". us Viis, 1983.
"Voeux en musique." HamumielPanorama- Musique Qanuary 1985), 1
p.
1986 "Avant-propos", in: H. Scherchen: La direction d'orchestre, Paris 1986:
11-13.
"Briefauszug an Hermann Scherchen" (12.3.1963, frz. im. dt. -000.),
in: Herrmann Scherchen Musiker, Berlin 1986: 95.
"Espace musical, espace scientifique." Le Coumer UNESCO 4. Paris,
1986.
"Hermann Scherchen", Le Monde de UL Musique 89, Mai 1986: 91.
"Microsystemes",June 1986
"Originality in Music and Composition." Fukushima Symposium. Tokyo,
1986.
1988 "Sui tempo." InXenakis, a cura de Enw Restagno. Torino, 1988.
II. Secondary Bibliography
1. Books and Articles that are entirely, or for the most part,
dedicated to Xenakis
1958 Jenneret-Gris, Ch. -E. (Le Corbusier). Le Poeme electronique. Paris,
1958.
1960 Moles, A Les musiques experimentales. Translated from the English by D.
Charles. Paris, 1960.
1963 Richard, A, and Barraud, j., and Philippot, M. P. "Yannis Xenakis et
la musique stochastique." Revue musicale 257 (1963).
1967 Bois, M. "Xenakis." Paris, 1966. German translation: Iannis Xenakis,
der Mensch und sein Werk. Bonn, 1967-1968. Numerous translations
into other languages.
1968 Barraud, j. Pour com prendre les musiques d'aujourd'hui. Paris, 1968.
Charles, D. La pensee de Xena.kis. Paris, 1968.
1969 Diverse authors. "Varese-Xenakis-Berio-Henry' oeuvres etudes
perspectives." Revue musicale 265-266 (1969).' , ,
A Selective Bibliography-Writings
343
Granlet, M., and Roy, J. Journ.ees de musique contemporaine de Paris-25-
31 Octobre 1968." Revue musicale 267 (camet critique, 1969).
Lachartre, N. "Les musiques artificielles." Diagrammes du monde 146
(April 1969): 1-96.
1970 Bourgeois, j. Entretien aveclannis. Paris, 1970.
Rostand, C. Xenakis, biographie et catalogue. Paris, 1970; expanded:
2/1972.
1971 Ewen, D. Composors of Tomorrow's Music. New York, 1971.
Coe, J. W. "A Study of Five Selected Contemporary Compositions for
Brass." Dissertation, Indiana Univ., 1971.
Potter, G. M. "The Role of Chance in Contemporary Music." Disserta-
tion, Indiana Univ., 1971.
1972 Diverse authors. "Iannis Xenakis." L'Arc 51 (1952).
Fleuret, M. Iannis Xena.kis. Paris, 1972.
Tamba, A Soi to Sow (Creation and Creative Ideas by Contemporary French
Composers. (In Japanese) Tokyo, 1972; Dissertation, Paris, 1974.
1973 Sevrette, D. "Etude statistique sur 'Herma,' de Xenakis." Graduate
work, Schola Cantorum. Paris, 1973.
1975 Gallaher, C. S. "Density in Twentieth-Century Music." Dissertaion, In-
diana Univ., 1975.
Halperin, D. "The Musical Oeuvre of lannis Xenakis." (In Hebrew)
Examination work, Jerusalem Univ., 1975.
Revault d'Allonnes, O. La creation artistique et les pro messes de la liberti.
1975.
____ Xenakis: Les Polytopes. Paris, 1975 and Olivier Revault. Les
Polytopes. Paris 1975;]apanese edition, Tokyo 1978.
1978 Fleuret, M. 78-Xena.kis. Paris, 1978.
Roberts, G. M. "Procedures for Analysis of Sound Masses." Disserta-
tion, Indiana, 1978.
Sato, M. /. Xena.kis: sugaku ni yori sakkyoku (I. X: Music Composed through
Mathematics.) (In Japanese). Dissertation, Tokyo Univ., 1978.
1979 DeLio, Th.]. "Structural Pluralism." Dissertation, Brown Univ., 1979.
1981 Various Authors. Regards sur Iannis Xenakis. Paris, 1981.
344
Formalized Music
Bayer, E. De Schiinberg a Cage-Essai sur La rwtitm d'espace stmore dans la
musique ctmtemporaine. Paris, 1981.
Sward, R. L. "A Comparison of the Techniques of Stochastic and Serial
Composition Based on a Study of the Theories and Selected Com-
positions of lannis Xenakis and Milton Babbitt." Dissertation,
Northwestern Univ., 1981.
Vriend,]. "Nomos Alpha for Violoncello Solo (Xenakis 1966); Analy-
sis and Comments." Interface 10, No.1 (1981).
1982 Varga, B. Conversation with Iannis Xenakis. (In Hungarian) Budapest,
1982.
1984 Various Authors. "On lanDis Xenakis." NUlida Musik 28, No.3 (ca.
1984-85).
Matossian, N. Iannis Xenakis. Paris, 1981.
Solomos, G. "Aspects de la musique grccque contemporaine. Memoire
de maitrise." Univ. Paris IV, 1984.
1985 Solomos, G. "Semiologie et musiques actuelles-Nuits de I. Xenakis",
memoire de D. E. A Univ. Paris IV, 1985.
Joseph, S. "The Stochastic Music of Iannis Xenakis: An Examination
of His Theory and Practice." Dissertation, New York.
II. 2.: Magazine Articles and Interviews:
1955 Helm, E. "Donaueschingen Festival Tame Affair." Musical America 75
(1955): 14.
Steinecke, W. "1m Brutofen des Ava ntgardis mus: Donaueschinger
Musiktage 1955." Melos 22 (1955): 327.
1957 Herrmann,]. "Exotik und Mathematik." Musica 11 (1957): 206.
Pringsheim, H. "Munchen." SMZ 97 (1957): 323.
Schmidt-Garre, H. "Strawinsky, Wozzeck, Jazz und junge Genera-
tion." Mews 24 (1957): 330.
Valentin, E. "Neues aus Munchen." NZ 118 (1957) 288.
1959 Franze,.J. "Bericht aus Argentinien." Musica:i3 (1959): 325.
Lindlar, H. "Gelachter undJubelpfiffe." Musica 13 (1959): 310-311.
"Scherchen Conducts Xenakis Work in Florence." WOM 1 (1959): 11.
1960 Galea, A "Eine ganz neue Siebente." NZ 121 (1960) 20.
1961 Charles, D. "lannis Xenakis." In Enzycwpedie de La Musique. Paris, 1961,
p.926.
Davies, M. E. "A Review of the Paris Season." Musical Opinion 84
(1961): 739.
Galea, A "Grenzgebiete der Musik; dreifache Tagung bei Hermann
Scherchen." NZ 122 (1961): 421-422.
v. Lewinski, W. E. "Kompositorische Spiele mit Klang und Farbei bei
Radio Bremen" Melos 28 (1961): 329-330.
Rostand, C. "Avantgarde Season." Musical America 81 (1961): 30.
1962 Kolben, R. "Die drei Gravesaner Tagungen/The Three Gravesano
Conventions." Gravesaner Bliiuer/Gravesano Review 23-24 (1962): 2ff.
119ff.
Pociej, B. "Jesieii 1962-zarys problematyki." Ruck Muzyczny 6, No. 22
(1962): 7-8.
Rozbicki, K. "Na festiwalu bez zmian." Ruck Muzyczny 6, No. 22 (1962):
10.
Stepanek, V. "ST/I0. 1-080262." Hudebny Rozhledy 15 (1962): 64l.
"Xenakis: unjeune architecte grec." Guide du concert 357 (1962): 1456.
1963 Galea, A "Les caprices d'Euterpe." Musica (Chaix) III (1963): 36.
Knessl, L. "Absurdes Musiktheater und Manierismus in Venedig."
Me/os 30 (1963): 309.
Kondracki, M. "list z USA" Ruck Muzyczny 7, No. 19 (1963): 8.
Stibilj, M. "Varsavskajesen 1962." Zvuk 57 (1963): 219.
1964 Harrison, J. "The New York Music Scene." Musical America 84 (1964):
22.
Kassler, M. "Musiques Formelles." (Review) PNM 3, No. 1 (1964):
115-118.
"New Works." Music Journal 22 (1964): 55.
1965 Charles, D. "Aujourd'hui." ReTJUe d'Esthetique 18 (1965): 406-420.
----
. "Entr'acte 'formal' or 'informal' music?" MQ 51 (1965):
144-165.
Finney, R. L. "Music in Greece." PNM 3 (1965): 169-170.
Golea, A, and Hambraeus, B. "Ett stycke samtida fransk musikhis-
toria." Musikrevy 20 (1965): 230.
----
. "Les mauvais calculs de Xenakis." Musica (Chaix) 138
(1965): 10.
Kaufman, W. "To Cover Xenakis Music, Use an IBM." Variety 239 (9
June 1965): 2ff.
Hitchcock, H. W. "New York." MQ51 (1965): 534-535.
Hopkins, B. "Paris Diary." Composer 16 (1965): 29.
McMullen, R. "A Season Strangely Flavored." High Fidelity/Musical-
America 15 (1965): 138.
Slonirnsky, N. "New Music in Greece." MQ51 (1965): 232-234.
Taranu, C. "Aspecte ale evolutiei conceptului despre ritm in muzicase-
colului nostru." (With summary in German, Italian, Russian and
French) Lucrari de Muzicologie 1 (1967): 81-82.
1966 "Iannis Xenakis." Music & Musicians 14 (1966): 15.
Jones, D. "The Music ofXenakis." Musical Times 107 (1966): 495-496.
"Les methodes et les opinions de Iannis Xenakis." Revue musicale de La
suisse romande 19, No.4 (1966): 20.
Morthenson, J. W. "Iannis Xenakis och 'Polla ta Dhina.'" Nutida Musik
9, No. 5-6 (1965-1966): 22-23.
Myers, R. "I'rom Baroque to Computer." Music &' Musicians 15 (1966):
34.
Romano, J. "Musicos de hoy." Buelws Aires Musica 21 (1966): 3 in No.
350, and p. 5 in No. 351.
Ruppel, K. H. "Royan will ein Zentrum moderner Musik werden."
Melos 33 (1966): 159.
____ . "Das 'Sonotron' wird vorgestellt-Iannis Zenakis 'Ter-
retektorh' beim Festival in Royan uraufgefiihrt." Gottesdienst und
Kirche 4-5 (1966): 163-165.
Sadie, S. "English Bach Festival." Musical Times 107 (1966): 697.
Schwinger, W. "Bremer Urauffuhrungen Flucht in die Virtuositat."
Melos 33 (1966): 235.
Takahashi, Y. "Scholia 1-65." Nutida Musik 9, No. 1-2 (1965-1966):
54-55.
"terretektorh" neboli Dalsi vymozenost?" Hudelmi Rozhledy 19, No.1 0
(1966): 317 and No. 11): 320.
"3rd International Festival of Contemporary Art in Royan-April 1st
to 7th." (With translations in French and German)." WOM 8, No.
2-3 1966): 34.
"Yannis Xenakis." San Francisco Symphony, 1 March 1966, p. 30.
347
1967 Dennis, B. "Xenakis 'Terretektorh' and 'Eonta.'" Tempo 82 (1967): 27-
29.
Dibelius, U. "Die Wellen haben sich gelegt-Das XI. Festival fur-
zeitgenossische Musik in Warsehau." Musikalische Jugend 16, No.5
(1967): 2.
Editorial Notes. Strad 77 (1967): 329.
Erhardt, L. "Dziewiee bogatyeh dni." Ruch Muzyczny 11, No. 21 (1967):
13-14.
Haines, E. "Spain." MQ53 (1967)567-577.
Kay, N. "Xenakis "Pithoprakta"." Tempo 890 (1967): 21-25.
Luck, H. "Xenakis und Lutoslawski setzten die Mafistabe-Eindrucke
vom Warschauer Herbst 1967." NZ 128 (1967): 443.
"lannis Xenakis." Music Events 22 (1967): 18.
"lannis Xenakis verkfdrteckning." Nutida Musik 10, No.5 (1966-
1967): 15-16.
Maguire, J. "lannis Xenakis: Formula for New Music." Saturday Re-
view 50 (21 June 1967): 51-53.
____ . "Paris Report: Contemporary Music at Full Tilt." High
Fidelity/Musical America 17 (1967): 30.
"New Philharmonia Concert." Music Events 22 (1967): 29.
Orga, A "Programmed Music." Music &' Musicians 15 (August 1967):
39.
Poeiej, B. "Czas odnaleziony i przestrzen rzeczywista." Ruch Muzyczny
11, No. 22 (1967): 5-7.
Sandin, L. "Reflexioner kring ovanstaende (= Sannolikhetsteori oeh
musik). De tekniska forutsattningarna for Pithoprakta." Nutida
Musik 10, No.5 (1966-1967): 12-14.
Schiffer, B. "Athens." (In English, German and French) WOM 9, No.3
(1967): 37IT.
Smalley, R. "Oxford." Musical Times 108 (1967): 731.
Valek, J. "Nova hudba ve stare zemi." Hudelmi Roddedy 20, No. 10
(1967): 315.
Wallner, B. "Kring centrum-tanken." Dansk Musiktidsskrifl42, No. 1-2
(1967): 10-12.
Walsh, S. "'Pithoprakta' Falls Down." Music &' Musicians 15, February
1967): 43.
348
Formalized Music
W a t ~ r ~ , .E. r:;. "HalVest of the Year; Selected Acquisitions of the Music
DivISion. The Quarterly JounUll of the LilJrary of Congress 214, No.1
(1967): 46-82.
Zoicas, L. T. "Matematica in muzica si unele aspecte in creatia ro-
maneasca contemporana." (With summaries in Russian, German
and French.) Luerari de Muzicologie 3 (1967): 57-65.
1968 '"Akratll'-'Pithopraktll.''' Buffalo Philharmonic (10 March 1968), 21-
25.
Bachmann, C. H. "Zeitgenossische Musik in Venedig." DMZ 23
(1968): 57l.
Brazen, M. "Koussevitzky Commissions." High Fidelity/Musical America
18 (October 1968): 18-19.
Butchers, C. "The Random Arts: Xenakis, Mathematics and Music."
Tempo 85 (1968): 2-5.
Chapin, L. "Xenakis & Balanchine." High Fidelity/Musical America 18
(April 1968): 27.
Drew, J. "Information, Space, and a New Time- Dialectic." Journal of
Music Theory 12, NO.1 (1968): 98-99.
Gelatt, .R. "Xenakis, Penderecki-and Buffalo." High Fidelity/Musical
Amenca 18 (1968): 20ff.
Golea, A "Eine Woche der Neuen Musik in Paris." NZ 129 (1968):
514.
----. "Grand oeuvres a Royan." Journal Musicale Franfais 170-
171 (1968): 14-15.
----. "Zeitgenossische Musik in Royan." NZ 129 (1968): 253-
254.
Hamilt?n, D. "Some Newer Figures in Europe." High Fidelity/Musical-
Amenca 18 (1968): 55.
Heinemann, R. "Festival in Royan." (Also in German and French)
WOM 10, No.3 (1968): 56-57.
----. "Festival mit Konzept." Musica 22 (1968): 265.
Kassler, J. C. "Report from London: Cybernetic Serendipity." Current
Musicology 7 (1968): 47-59.
Monnikendam, M. "Het muziekfeest van Royan." Mens en Melodie 23
(1968): 164-165.
Monteuax, J. "Egy franciaorzagy festival." Muszika 11 (August 1968):
30-31.
Morrison, M. "New Music in Europe." Musicanada 16 (December
1968): 7.
Nestev, 1. "Varshavskie melodii." Sovetskaya Muzyka 32 (1968): 119.
Orga, A "American in London." Music & Musicians 17 (December
1968): 49-50.
Orr, B. "Notes." Composer No. 29 (1968): 44.
Plotkin, F. "Festival of the Arts Today, No.2." High Fidelity/Musical
America 18 (June 1968): 24-25.
Simmons, D. "Concert Notes." Strad 79 (1968): 119.
____ . "London Music." MusicaL Opinion 92 (1968): 119.
Smalley, R. "Royan." Musical Times 109 (1968): 563.
Souster, T. "Oxford Comes to London." Music & Musicians 16 (July
1968): 23.
____ . "Xenakis' 'Nuits.'" Tempo 85 (1968): 5-18.
Terry, W. "World of Dance (Ballet Choreographed by Balanchine)."
Saturday Review 51 (3 February 1968): 48--1.9.
Vandenbogaerde, F. "Analyse de 'Nomos Alpha' de lannis Xenakis."
Mathimatiques et sciences humaines 24 (1968): 15
Wagner, K. "Komponisten aus Wnf liindern im Hamburger 'Neuen
Werk.'" MeLos 35 (1968): 215.
Walterova, E. "Xenakis a zrozeni hudebni reci." Hudebni Rozhledy 21
(1968): 252-254.
"Yannis Xenakis." Nutida Musik 12, No.2 (1968-1969): 16.
1969 Akijama, K. "Iannis Xenakis' Mathematical Thoughts." (In Japanese)
Bijutsu Techo 9 (1969).
Alcaraz, J. A "Musica ellectronica." Heterofonia 2, No.9 (1969): 8-10.
Archibald, B. Review of: Iannis Xenakis, the Man and his Music, M. Bois,
ed. Notes 25, No.3 (1969): 481-482.
Bachmann, C. H. "A1pbach-Venice-Warsaw." (In English, French
and German) WOM 11, No.1 (1969): 55.
Chanan, M. "Maxwell Davies and Xenakis." Music & Musicians 17
(August 1969): 50.
____ ,. "Orchestras at Play." Music & Musicians 18 (September
1969): 59-60.
"Concert Notes." Strad 80 (1969): 181.
Cook, N. "Shiraz." Musical Times 110 (1969): 1165.
"Creations-1969 (On Anaktoria)." CourTier Musicale de France 28
(1969): 248.
"Creations-1969 (On Eonta)." Coumer Musicale de France 26 (1969):
107.
"Creations-1969 (On Nomos Gamma)." Coumer Musicale de France
27 (1969): 187.
"Creations-1968 (On Nuits, Palla ta Dhina, ST/48 and Les Suppli-
antes)." Coumer Musicale de France 25 (1969): 29.
Elliott, C. "Choral Performances 1968-1969: France." American Clwral
Revue 11, No.2 (1969): 39.
'"Eonta.''' New Yom Philharmonic (27 February 1 969): D-F.
Evarts, J. "Paris SMIP." (In English, French and German) WOM 11,
No.1 (1969): 50-51.
Golea, A "Le sixieme festival de Royan." JourlULl Musicale Fratlfais 181
(June 1969): 20.
____ .. "Royan." NZ 130 (1969): 228.
____ ,. "Royan: Das sechste Festival." Orchester 17 (1969): 264.
Heinemann, R. "Sechstes Festival International (Royan)." Musica 23
(1969): 356.
Hiller, L. "The Computer and Music, Ithaca and London 1969."
Within French Experiments in CompuJer Composition, 8 p.
Jungheinrich, H. K. "Ein Festival platzt aus den Nahten-Royan 69."
NMZ 18, NO.3 (1969): 4.
Kassler, M. "Report from Edinburgh." PNM 7, No.2 (1969): 175-176.
Lachartre, N. "Les musiques artificielles." DiagramrtUls du Monde 146
(1969); within lannis Xenakis et La musique stochastique, 13 p.
Maegaard, J. "Vor mand pa parnasset." Dansk Musiktidsskriji 44 (1969):
36-38.
"Many Wonders." The NATS Bulletin 11 (January 1969): 18.
Moor, P. "Royan: High camp, low spirits." High Fidelity/Musical ArtUJTica
19 (July 1969): 27.
Noble, D. "N. Y. Philharmonic (Ozawa)." High Fidelity/Musical ArtUJTica
19 (May 1969): 20-21.
Perrot, M. "Entretien avec lannis Xenakis." Revue musicale 265-266
(1969): 61-76.
Rostand, C. "Royan besticht durch interesante Urauffiihurungen."
Melos 36 (1969): 388.
351
Serres, M. "Musique et bruit de fond-Iannis Xenakis; Pithoprakta."
Critique 261 (1969): 12 p.
Simmons, D. "London Music." Musical Opinion 92 (1969): 619.
. "London Music." Musical Opinion 93 (1969): 124.
----
Soria, D. J. "Artist life." High Fidelity/Musical America 19 (May 1969): 6.
Thomas, S. "London Sinfonietta." Music Events 24 (December 1969):
34.
UDen, J. O. "Krig oeh vetenskap-en myt fOr var tid. Kring Iannis
Xenakis." NuJida Musik 13, No.1 (1969--1970): 12-15.
Zeller, H. R. "Zum Beispiel Xenakis." Melos 36 (1969): 410-416.
Zijlstra, M. "Het Utrechts Studium Generale." Mens en Melodie 24
(January 1969): 16.
1970 Cecchi, C. '''Metastasis' y 'Pithoprakta' de Xenakis." Revista Musical
Chilena 24, No. 112 (1970): 99--102.
Chamfray, C. "Premieres auditions franc;aises (on ST!48)." CourTier
Musicale de France 30 (1970): 73 .
. "Premieres auditions franc;aises (on Persephassa)." Cour-
----
Tier Musicale de France 29 (1970): 24.
Deliege, C. "L'invenzione musicale contemporanea (tentativo disin-
tesi)." Nuova Rivista Musicale Italiana 4 (1970), No.4, p. 637-664;
and No.5, p. 880-897; reprinted from Syntheses 276 (June 1969).
Druey, P. "Geneve." SMZ 110 (1970): 38.
Durney, D. /]ameux, D. "Rencontres avec Iannis Xenakis." Musique
enjeu 1 (1970): 46-65.
Gardiner, B. "Xenakis Talks." Music Events 25 (January 1970): 6-7.
Golea, A "StraBburg Suche nach neuen Formen." NZ 131 (1970): 363.
Harrison, M. "Messiaen, Xenakis." Musical TirtUls III (January 1970):
67.
Komorowska, M. "Muzyczna awangarda na festiwalu." Ruch Muzyczny
14, No.3 (1970): i3.
Krellmann, H. P. "Xenakis-Profile der neuen Musik IV." FonoForum
15 (1970): 212-214.
Ledesma, O. "Festival de Musica Contemporanea." Buenos Aires Musi-
cal 25, No. 42 (1970): 2.
"Le poeme eleetronique-Brusel dukumentace." Hudelmi Rozhledy
23:508-510.
352
Formalized Music
Mache, F. B. "A propos de Xenakis." 1.0 Nouvelle Revue Franraise 35
(1970): 115-122.
"Pro Arte Orchestra (Cavalho)." High Fidelity/Musical America 20 (May
1970): 24-25.
Schaeffer, P. "La musique et les ordinateurs." Revue musicale 268-269
(1970): 57-88.
Schiffer, B. "Interessante Neuheiten bereichern das Londoner
Musikleben." Mews 37 (1970): 22.
Wagner, K. "Grenzen der Konzertsaalmusik Duetsche Erstaufftihrun-
genin Hamburg." Melos 37 (1970): 515-516.
1971 "'Akrata' for Sixteen Instruments." New York Phifharrrwnic (I April
1971): E-F.
Benary, P. "Luzern Internationale Musikfestwochen." SMZ III
(1971): 288.
Bowen, M. "Piano." Music & Musicians 19, May 1971): 61.
Chamfray, C. "Premieres auditions franc;aises, (on Kraanerg and Syn-
aphai)." Goumer Musicale de France 35 (1971): 130.
Feuer, M. "Royan 71." Muszika 14,june 1971): 21-22.
Gill, D. "Stephen Pruslin." Musical Times 112 (1971): 262.
Galea, A "Royan eine Ost-West-Begegnung." NZ 132 (1971): 317.
de La Grange, H. L. "Music of the East at Royan." Saturday Review 54
(26 June 1971): 38.
____ . "Persepolis." Music & Musicians 20 (December 1971): 66-
67.
____ . "Royan 1971." Music & Musicians 19 (June 1971): 47.
Hampel, H. "Maifestspiele Wiesbaden." Oper und Konuri 9 (June
1971): 29.
Heinemann, R. "Musik aus dem Osten." Musica 25 (1971): 364-365.
Hohlweg, R. "1m Trauern war Royan diesmal am besten." MeLos 38
(1971): 250-253.
Konold, W. "Kieler Saison 1970/71 eroffnet mit vieWiItigem Pro-
gramm." Melos 38 (1971): 20-21.
Mache, F. B. "Methodes linguistiques et musicologie." Musique enjeu 5
(November 1971): 76ff.
Monnikendam, M. "Een tripriek der Luzerner Musikfestwoche." Mens
en Mew die 26 (1971): 329-330.
Paap, W. "VrijblUvend musiceren--een wendeIconcert in de Vara
studio." Mens en Mewdie 26 (1971): 353.
353
1972
"Polla ta Dhina (Wonders are Many)." Boston Symplwny (1971): 407-
108.
"Premieres." Music EducatorsJournal57 (May 1971): 90.
Schaeffer, P. "'A propos des In De L'Experience :
L'Ex"'erience HUrrUJine." Revue must-cale 274-275 (1971) .. 55-6 ..
r
. I' Kl ier und Singsllm
mem
Schiffer, B. "Experimente fur K annette. av
Londoner Konzertsalen." Melos 38 (1971): 156.
S I E b rt U
"Luzern Einfuhrung in lannis Xenakis." NZ ee mann- gge e, .
132 (1971): 618-619.
."
Sneerson, G. "Die musikalische 'Avantgarde' der sechzlger Jahre.
Musik und Gesellschaft 21 (1971): 754-760.
"
. "Serialism and Aleatoricism-An Identity of Contrasts.
--(I-n-R-us---::sian) Sovetskaja Muzyka 1 (January 1971): 106-117.. . .
Stuckenschmidt, H. H. "Die Klassiker der modernen MUSlk smd III
West-Berlin erfolgreich." Melos 38 (1971): 208.
Trumpff, G. A "Wiesbaden 75 Jahre Internationale Maifestspiele." NZ
132 (1971): 383.
. 7
Zielinski, T. A "Yannis Xenakis." Ruch Muzyczny 15, No.9 (1971). 1 -
18.
B h C H
"Heraus aus den Theater-Palasten- Pariser Festival ac mann, . .
aufneuen Wegen." NMZ 21 (December 1972): 28.
Cadieu, M. uDa Parigi." Nuova Revista Musicale ltaliana 6 (1972): 600-
601.
Casken J. "Poland." Music & Musicians 21, September 1972): 78-79 ..
,
f . ( Orestie) " Coumer Chamfray, C. "Premieres auditions ranc;alses, on .
Musicale de France 37 (1972): 20.
. "Premieres auditions fram;aises, (on Aroura)." Coumer
--M-us-u-' -aZ-e de France 34 (1971): 78.
Chapman, E. et al. "English Bach Festival." Music Events 27 (June
1972): 10ff.
East, L. "Xenakis." Music & Musicians 20 (August 1972): 61-62.
"First Performances." WOM 14, No.3 (1972): 81.
Fleuret, M. "Xenakis-A Music for the Future." Music & Musicians 20
(April 1972): 20-27.
.
Flothuis, M. "Het muziekfeest der ISCM te Graz." Mens en Melodle 27
(1972): 357-360.
354
Formalized Music
Golea, A "StrafiburglParis: Deutsche Werke beherrschen franzo-
sischen Opernauftakt-Millglilcktes Herbstfestival." NZ 133 (1972):
714-716.
Gradenwitz, P. "Greece." Musical Times 113 (1972): 73.
Griffiths, P. "English Bach." Musical Times 113 (1972): 590.
____ ,. "New Music." Musical Times 113 (1972): 686.
Hoogland, C. "Framtidens fack10r over 2000-ariga ruiner." Nutida
Musik 15, No.2 (1971-1972): 48-5l.
Krellmann, H. P. "Der Mathematiker unter den zeitgenossischen
Komponisten." Melos 39 (1972): 322-325.
Kuhn, H. "Xenakis, oder das Innenleben der Zahlen." Musica 26
(1972): 467-468.
Napolitano, E., and Tonietti, T. "Xenakis tra medioevo e iI-
luminismo." Quadrivium Rassegna Musicale 5 (1972): 45-66.
Russeol, H. "The Liberation of Sound, an Introduction to Electronic
Music." (Englewood CliffS, 1972; within "Xenakis", 8 p.
B. "Englische Orehester erteilen Auftrage an junge Kom-
ponlSten." Melos 39 (1972): 302-306.
____ ,. "Neue gricchische Musik." Orbis Musicae: Studies in Musi-
cology 1 (1972): 193-201.
Simmons, D. "London Music." Musical Opinion 95 (1972): 454.
Sohal, N. "Iannis Xenakis: Formalized Music." (Book review)." Tempo
101 (1972): 53.
Stuppner, H. "Da Darmstadt." Nuova Rivista Musicale ltaliana 6 (1972):
593-596.
Vivier, o. "Iannis Xenakis." In Rizzoli-Ricordi EncicIopedia della
Musica, Vol. 6 (1972).
Wagner, M. "IGNM-Fest 1972 in Graz." OMZ 27 (1972): 617-620.
"Yannis Xenakis am 'Terrctektorh.'" Nutida Musik 15, No.2 (1971-
1972): 47-48.
1973 Blaukopf, K. "Musik iiberschrcitct ihre Grenzen." HiFi Stereophoni.e 12
(1973): 268[
Brown, R. S. "French Music Sinee Debussy and Ravel." High Fitkl-
ity/Musical America 23, September 1973): 50-65.
Browne, R. "Iannis Xenakis Formalized Music." (Book review) Notes
30 (September 1973): 67-68.
Chamfray, C. "Premieres auditions (on Eridanos)." Courrier
Musicale tk France 43 (1973): 105.
355
. "Premieres auditions (on Nomos Alpha)."
-..,------,,--'
Courrier Musicale de France 42 (1973): 68.
. "Premieres auditions fran<;aises, (on Charisma, Linai
-Ago--n-,-M-edea, Polytope de Cluny and Mikka)." Courrier Musicale
de France 41 (1973): 21.
Charles, D. "La musique et l'ecriture." Musique en jeu 13, November
1973): 3-13.
Christensen, L. "Numus-international." Numus-W7 4 (1973): 56.
Chittum, D. "Current Chronicle: Philadelphia." MQ 59 (1973): 125-
133.
Derhen, A "Marcel Couraud Vocal Group." High Fidelity/Musical Amer-
ica 23 (June 1973): 21-22.
Golea, A "Metz Zeitgenossische Musik." NZ 134 (1973): 31.
Griffiths, P. "La Rochelle, Royan." Musical Times 114 (1973): 629.
. "Music in London." Musical Times 114 (1973): 50.
----
Hohlweg, R. "PendeIverkehr zwischen La Rochelle und Royan." Melos
40 (1973): 303-305.
Kaczynski, T. "Misterium Xenakisa w Cluny." Ruch Muzyczny 17, No.8
(1973): 14-15.
McMullen, R. "Paris the new Festival d'Automne." High Fidelity/Musi-
cal America 23 (February 1973): 30-31.
"Premieres." MusicJournal31 (December 1973): 1 Iff.
Schiffer, B. "Paris." Music & Musicians 21 (February 1973): 70-72.
. "Politische Satire auf Garibaldi in Paris." Melos 40 (1973):
----
52-54.
Simmons, D. "London Music." Musical Opinion 96 (1973): 453ff.
Smith, I. S. "Communications." PNM 11, No.2 (1973): 269-277.
Torrens, P. "Avantgarde in Frankreich." Melos 40 (1973): 330-338.
----
. "Panorama de la musique comemporaine en France."
Cou'r'r'Wr Musical de France 42 (1973): 55.
"Tully Hall." Music Journal 31 (November 1973): 36.
Walter, F., and Poulin,]. C. "Geneve." SMZ 113 (1973): 369-372.
1974 Albright, W. "Fourth Festival of Contemporary Organ Music: Hartt
College." Music (Ago) 8 (September 1974): 21-24.
356
Formalized Music
Becker, W. "Da Colonia." Nuova Rivista Musicale ltaliana 8 (1974):
630--Q34.
Chamfray, C. "Premieres auditions franc;aises, (on Evryali and Pollata
Dhina)." Coumer Musicale de France 46 (1974): 53-62.
___ ---:. Premieres auditions fran<.;aises, (on Erikhthon and Nomos
Gamma)." Coumer Musicale de France 47 (1974): 104-105.
"First Performances." WOM 16, No.4 (1974): 62.
Griffiths, P. "lisbon." Musical Times 115 (1974): 683.
____ . "New Music." Musical Times 115 (1974): 496-497.
Jack, A "Avantgarde." Music & Musicians 23 (October 1974): 58-59.
Kirchberg, K. "Zwischen Vergangenheit und Zukunft- Beethoven-
Festund Xenakis-Retrospektive in Bonn." Musiklumdel 25 (1974):
358*359.
"Konzerte: Munchen." OperundXonzert 12, No.3 (1974): 37.
Mlkhe, F. B. "Entretien avec lannis Xenakis." La Nouvelle Revue Fran-
faise 43, No. 257 (1974): 121-128.
"Marie-Franc;oise Bucquet, Piano." High Fidelity/Musical America 24,
February 1974): 20.
Moor, P. ':Xenakis' Latest Gulbenkian Premiere." High Fidelity/Musi-
caLAmeru:a 24, February 1974): 38.
Simmons, D. "London Music." Musical Opinion 97 (1974): 373-375.
Smith, F. J. "Traditional Harmony? A Radical Question." Music Review
35, No.1 (1974): 68ff.
H: H. "Les fondements techniques et mathema-
tlquesde la muslque moderne et specialement chez Xenakis et
Penderecki." Coloquio artes 1I/16, No. 18 (1974): 55-63.
"Urauffuhrungen." NZ 135 (1974): 515.
Zeller, H. R. "Zwischen Elektronischer und Instrumentaler Musik."
Musik und Bildung 6 (1974): 121-128.
1975 Chamfray, C. "Premieres auditions franc;aises, (on Eridanos)." Coumer
Musicale de France 50 (1975): 64.
___ . __ " "Premieres auditions fran<.;aises, (on Antikhthon and Em-
prellltes)." Coumer Musicale de France 51 (1975): 105.
____ . "Premieres auditions fran<.;aises, (on Gmeeoorh and Noo-
mena)." Coumer Musicale de France 49 (1975): 21.
Diemente, E., and Holloway, C. "Gmeeoorh"." Music (Ago) 9 (June
1975): 18.
357
"First Performances: Xenakis at Wigmore Hall." Tempo 112 (1975):
30-31.
Grepen uit het Holland Festival 1975." Mens en Melodie 30 (1975):
226-232.
Griffiths, P. "New Music." Musical Times 116 (1975): 263-264.
____ ". "New Music." Musical Times 116 (1975): 552-553.
____ . "Xenakis: Logic and Disorder." Musical Times 116 (1975):
329-331.
Hill, P. "Xenakis and the Performer." Tempo 112 (1975): 17-22; also
Pruslin, S., Tempo 115 (1975): 54; also Hill, P. Tempo 116 (1976): 54.
Jack, A "lannis Xenakis Visits the English Bach Festival." Music &
Musicians 23 (April 1975): 14ff.
____ . "Modern." Music & Musicians 23, June 1975): 49-51.
Jeudy, H. P. "'A propos des lieux de signifiance' (Xenakis, La Monte
Young). In Recherches et sociologie de La musique." Musique en jeu 18
(April 1 975): 2lff.
Moor, P. "Richter, Xenakis, and a Newcomer Dominate French Festi-
vals." High Fidelity/Musical America 25, November 1975): 33-34.
Naud, G. "Aperc;us d'une analyse semiologique de 'Nomos Alpha.'"
Musique enjeu No. 17, January 1975): 63-72.
Pauli, H. "Xenakis im Gesprach." SMZ 115 (1975): 300-308.
Samana, L. "La Rochelle 1975." Mens en Melodie 30 (1975): 271-274.
Schiffer, B. "Athens." Music & Musicians 24 (December 1975): 60--Q1.
____ . "Athen: Xenakis-Festival." SMZ 115 (1975): 315-316.
____ . "Athen: Xenakis-Auffiihrungcn am FuBe der Akropolis."
Melos/NZ 1 (1975): 475.
__ -,--_ .. "English Bach Festival: Xenakis Exhibition." Tempo 113
(1975): 47-49.
____ . "First Performances: English Bach Festival-Xcnakis and-
Stockhausen." Tempo 114 (1975): 47-49.
W., S. "Iannis Xenakis." In The Inwmational Cyclopedia of Music and
Musicians Vol. 10. London, 1975, p. 2494.
Zaplitny, M. "Conversations with lanDis Xenakis." PNM 14, No.1
(1975): 86-103.
"Zsal koncertowych." RuchMuzyczny 19, No. 22 (1975): 5.
1976 "Brooklyn Philharmonia: Xenakis." High Fidelity/Musical America 26
(August 1976): 24-25.
358
Formalized Music
Cadieu, M. "Da Parigi." Nuova Rivista Musicale Italiana 10 (1976); 493-
495.
Chamfray, C. "Premiere auditions (on Mikka S. )." Coumer
Musical de France 54 (1976); 65.
Editorial notes. Strad 87 (1976); 231.
Emmerson, S. "Sylvia Gualda." Music & Musicians 25 (September
1976); 56ff.
----. "Xenakis and Berio." Music & Musicians 24 Gune 1976):
53-54.
----. "Xenakis." (Interview) Music & Musicians 24 (May 1976):
24-26.
Gradenwitz, P. "Israel, Gaste aus Deutschland und Urauffuhrungen."
Melos/NZ 2 (1976); 396-398.
Griffiths, P. "New Music." Musical Times 117 (1976): 589.
----. "New Music." Musical Times 117 (1976); 246-247.
Roth, D. "Iannis Xenakis on New Music." (Interview) Musical Opinicms
99 (1976); 514[
Schiffer, B. "Jerusalem!TeIAviv Testimonium 1976." SMZ 116 (1976):
191.
---_. "London." SMZ 116 (1976); 105.
----. "Greeks in London (Greek Month at the ICA)." Music &
Musicians 24 (February 1976): 51-52.
---_,. "N'shima." SMZ 116 (1976): 19l.
----. "Xenakis' N'shima." Tempo 117 (1976); 31-32.
Schropfer, L. "Zu lannis Xenakis Kompositionsverfahren." In Pro 26
(1976); 12 p. without pagination.
"Svensk musikvolr 976." (Also in English) Nutida Musik 19, No. 3
(1976-1977): 21-22.
Tonietti, T. "Formalized Music." (Review) Nuova Rivista Musicale Itali-
ana 10, No.1 (1976): 139-140.
Vermeulen, E. "Xenakis-Festival in Midde1burg." Mens en Melodie 31
(1976); 257-260.
Trotter, H. "June !n Buffalo: Second Annual Contemporary Festival
Features XenaklS, Crumb." High Fidelity/Musical America 26 (October
1976): 26-27.
"Xenakis at the City University." Music & Musicians 24 (March 1976):
1 Off.
Zij Is tra, M. "Het componercn van de avant-garde in de 20ste eeuw."
Mens en Melodie 31 (1976): 3.
1977 "Capac Members in the News." (Also in French) Canadian Composer
122 (1977): 36-37.
Chamfray, C. "Premieres auditions (on Akanthos)." Coumer
Musical de France 59 (1977): 106.
----
. "Premieres auditions fran<,;aises, (on Akanthos)." Coumer
"Cite Greek Composer (Beethoven Prize)." BiUboard 89 (28 May
1977).
Dawney, M. "Orchestral Concerts in and around London." Musical
Opinion 100 (1977); 495.
Emmerson, S. "Xenakis, Ligeti and Stockhausen." Music & Musicians
26 (October 1977): 48.
Felder, D. "An Interview with Stockhausen." PNM 16 (1977): 85-101.
Schiirmann, H. G. "Beethovenpreis (1977 fUr Xenakis." Orchester 25
(1977): 546-547.
Schulman, M. "Xenakis: The Capac-MacMillan lectures." (Also in
French) Canadian Composer 123 (1977): 10f.
Tallian, T. "Szazadunk zencje a radioban." Muzsica 20 (October 1977):
33-35.
"Xenakis at the Mount Orford Arts Centre." (Also in French) Canadian
Composer 123 (September 1977); 14.
1978 Baculewski, K. "Kwardrans z lannisem Xenakisem." Ruch Muzyczny
22, No. 24 (1978): 1O-1l.
Chamfray, C. "Premieres auditions franc;aises, on Legend d'Eer)."
Courrier Musical de France 64 (1978): 146.
. "Premieres auditions fran<,;aises, (on Dmaathen, A Colone,
----'
Kottos, Jonchaies, and Epei)." Courrier Musical de France 61 (1978):
20-2l.
Emmerson, S. "Athens." Musical Times 119 (1978): 1070.
Fuhrmann, P. "Sternentheater mit Urknall und Kometen." NMZ 27,
No.2 (1978): 4.
"Iannis Xenakis." In La musica-Dizionario, Vol. 2 (1978).
"Impressions d'antenne." Harmonie-Antenne (October 1978): 26.
M:khe, F. B. "Les mal entendus--compositeurs des annees 70." Rewe
musicale 314-315 (1978; within Enlretien avec Xenakis): 111-150,
160-161.
Matossian, N. "Xenakis." Music & Musicians 26, March 1978): 56-57.
Onnen, F. "Een Xenakis-cyclus in Frankrijk." Mens en Melodie 33
(1978): 164-167.
Piencikowski, R. T. "Paris; Xenakis et I'Ensemble InterContem-
porain." SMZ 118 (1978): 107-108.
Schiffer, B. "Greece." Music [;1 Musicians 27, December 1978): 54.
____ ,. "Xenakis Polytope de Mycenae." Ternpo127 (1978): 44-45.
Schiirmann, H. G. "Beethoven-Preis soil wieder offen ausgeschrie-
benwerden." Orchester 26 (1978): 927.
"Urauffiihrungen." Melos/NZ 4 (1978): 55.
"Xenakis: Jonchaies." Ruck Muzyczny 22, No. 24 (1978): 10.
1979 Bennett, R. M. "Letters C'est si Bonn?" Music [;1 Musicians 27 (August
1979): 4.
"Caught!." Down Beat 46 Uanuary 1979): 36.
Chamfray, C. "Premieres auditions (on Mycenes A)." Cour-
Tier Musical de France 65 (1979): 17.
____ . "Premieres auditions fran<;aises, (on Palimpsest)." Courrier
Dusapin, P., and Rey, A "Xenakis 'Si Dieu existait, il serait bricoleur.'"
Monde de la Musique (May 1979), 5 pp.
Editorial. Heterofonia 12, No. 64 (1979): 2.
Gautier, J. F. "Olivier Messiaen et ses eleves." Spectacle du monde No.
202 Uanuary 1979): 107-110.
"Holland Festival 1979. " Mens en Melodie 34 (1979): 307.
"lannis Xenakis." In Sohlmans Musiklexikon, Vol. 5 (Stockholm 1979):
860-86l.
Matossian, N. "Xenakis Diatope in Bonn." Tempo 129 (1979): 39-40.
Mikulska, M. "Tcoria gier w tworczosci Iannisa Xenakisa." Ruch
Muzyczny 19 (September 1979): 3-5.
Myhill, J. "Controlled Indeterminacy, a First Step Towards a Semi-
Stochastic Music Language." Computer Music Journal 3 (1979): 12-
14.
Papaioannou, J. G. "Die neue griechische Schule: der Weg zur Sclb-
standigkeit." NZ 140 (1979): 34-36.
Varga, B. A "Muzsikusportrek, Budapest 1979." In Unterhaltungmit
Iannis Xenakis (in Hungarian).
Zimmermann, C. "Bonn." Operund Konzenl7, No.6 (1979): 10.
1980 Barry, M. "1945 and After." (Concert review) Music [;1 Musicians 28
(1980): 59ff.
Chamfray, C. "Premieres auditions fran<;aises, (on Anemoessa)." Cour-
Tier Musical de France 69 (1980): 16.
Delio, T. "lannis Xenakis Nomos Alpha: The Dialectics of Structure
and Materials." Journal of Music Theory 24, No.1 (1980): 63-96.
"Dossier: lannis Xenakis-tra Scilla e Cariddi." Spirali (Italian) 3
(December 1980): 5-6.
Jackson, R. "Polarities, Sound-Masses and lntermodulations a View of
the Recent Music." Music Review 41, No.2 (1980): 128ff.
"Konzerte." Oper und Konzeri 19, No.3 (1981): 36-37.
"Le miroir du compositeur." (Interview) Spirali (French) 4 Uuly-
August 1981): 20-21.
Massin, B. "Creer comme on respire." Panorama Musiques 37 (1980):
64-67.
Mikulska, M. "Xenakis w Polsce." Ruck Musyczny 24, No. 12 (1980): 10.
Nyffeler, M. "Beethoven llillt unter die Avant-gardisten; zumdreiBig-
sten Mal Bonner Beethovenfest." NMZ 4 (August-September
1980): 32.
Philippot, M. "lannis Xenakis." In New Grove, Vol. 20. London, 1980,
pp. 559-561.
Rinaldi, M. "Roma e Stresa." Rassegna Musicale Curci 33, No.3 (1980):
49.
Rastoin, B. "Les Enfants devant l'outil musical de Xenakis." Panorama
Musiques 37 (1980): 68.
Rovsing, Olsen, M. "Samtale med Xenakis." (Interview) Dansk Musik-
tidsskrifl54 (1980): 312-314.
Schiirmann, H. G. "Stille-zum Streichquartett; Urauffiihrungen von
Nono und Xenakis beim Bonner Beethovenfest." Orchester 28
(1980): 709-710.
Zeller, H. R. "Musik, die ihren eigenen Raum entwickelt; der Wegvon
Iannis Xenakis." NMZ 29 (October-November 1980): 3-4.
"z sal koncerowych." Ruch Muzyczny 24, No. 13 (1979): 8-9.
1981 Fumagalli, T. "Iannis Xenakis." Musica Viva 5, No. I (1981): 53-57.
Jensen, J. 1. "Koncert-noter fra efterAret." Dansk Musiktidsskrifl 55,
No.4 (1981): 160-161.
van Huijstee, T. "Van Pythagoras naar Xenakis." Mens en Melodie 36
(1981): 408-416.
Schiffer, B. "Xenakis Mists." Tempo 137 (1981): 35-36.
Smith, P. J. "N. Y. Philharmonic: Xenakis' Empreintes (D. S.
premiere)." High Fidelity/Musical America 31 (1981): 24-25.
"Urauffuhrungen." Musikharulel32 (1981): 160.
1982 van Emmerik, 1. "Tiende Rencontres lnternationes te Metz." Mens en
Melodie 37 (1982): 75.
Geysen, F. "Sporen der Niddeleeuwen in hedendaagse Fransekoor-
muziek." (Summaries in English and French). Adem 18, No. 1
(1982): 27.
"lannis Xenakis." In Who's Who in France 1981-1982, p. 1513.
Loubet, E. "Das Interview lannis Xenakis." Musica 36 (1982): 525-
530.
"New Percussion Ensemble World Premiered." Percussion Notes 20,
No.2 (1982): 15.
Norris, G. "Recitals." Musical Times 123 (1982): 559.
"Science et societe-L'informatique musicale." Pour la Science ( S c ~
tifu American 7, French edition, November 1982): 10-11.
Stiehweh, K. "Jubilaum in Metz." DMZ 37 (1982): 114.
"Drauffiihrungen." Musikharulel33, No.3 (1982): 150.
1983 Adam, J.; Cottet, S.; and Doisneau, E. "La passion mathematique."
L'Arc (May-June 1983), 2 p.
Dreyer, M. "Huddersfield." Musical Times 124 (1983): 113.
Frounberg, I. "Teori og praksis i lannis Xenakis kompositionste-
knikker: en eksemplifikation." Dansk Musktidsskrifl 57 (1983): 185-
199.
Frisius, R. "Xenakis und der Rhythmtis." NZ 144, No.4 (1983): 13-17.
Restagno, E. "Da Venezia, La Biennale Musica." Nuova Rivista Musi-
cale Italiana 17, No.1 (1983): 107.
Schiffer, B. "Bath: Xenakis: Arditti." Tempo 146 (1983): 60-61.
____ . "Testimonium VI." Tempo 145 (1983): 49-50.
____ . "Three Books under Review." Tempo 144 (1983): 25--27.
"Xenakis: picklo obmyslone przez ezlowicka." Ruch muzyczny 27, No.
23 (1983): 6-7.
1984 Bodin, L. G. "Warszawahoesten 1983." Nutida Musik 27, No.3 (1984-
1985): 45.
Carl, R. "The Distant Shore Seen from Two Sides (Comparison of
French and American Philosophies in the Composition of Contem-
porary Music)." College Music Symposium 24, No.1 (1984): 148-157.
Concert notes. Strad 95 (1984): 6-7.
Danler, K. R. "Miinchens Musica Viva." Orchester 32 (1984): 757.
Hoffmann, S. "Musik als eine Art Felsbloek; lannis Xenakis bei der
Wintermusik 84." NMZ 33 (April-May 1984): 6.
Frisius, R. "Das Musikfestival in Metz." 6MZ 39 (1984): 42.
Frounberg, I. "Horisontal og vertikal proportionering- en skitse."
Dansk Musiktidsskrifl58, No. 2-3 (1984-1985): 115ff.
Kruse, B., and Ruud, E. "lannis Xenakis til Norge (interview)." BaUade
8, No.2 (1984): 12-17.
Lugowska, M. "sekretarze czyli okresionose i niedookresionose." Ruch
Muzyczny 28, No. 23 (1984): 20-22.
Lunell, H. "ISCM i backspegeIn." Nutida Musik 27, No.3 (1984-1985):
46-47.
Pinguet, F. "'La Degustation (+ entretien avec Iannis Xenakis).' In Un
monde musical metisse." Revue musicale 365--366-367 (1984): 245ff.
Rey, A "Expliquez-Vous." Xenakis, Morule de la Musique (October
1984): 34-40.
"Xenakis herhaalde zich niet." Mens en Melodie 39, February 1984): 92.
"z Xenakisem w Ka7jmierzu." (Interview) Ruch Muzyczny 28, No. 23
(1984): 21.
1985 AhIen, C. G. "Iannis Xenakis." Nutida Musik 28, No.3 (1984-1985).
Vermeil, J. "Le monde cn harmonie--entretien avec lannis Xenakis."
Silences 1 (1985): 91-94.
Vermeil, J. "Les demeures Xenakis." Silences 1 (1985): 201-206.
Wilson, P. "Formalisierte Musik- NaturwissenschaftJiches Denkenin
der Kunst. Gedanken zu den stochastischen Kompositionstheorien
von lannis Xenakis und Clarence Barlow." In Neuland/Anstitze zur
Musik der Gegenwart, edited by H. Henek, Bergisch-Gladbach (1984-
1985): 52-59.
Abbrevations and symbols DISCOGRAPHY
Diss. Dissertation
ff.
MQ
No.
NMZ
NZ
OMZ
PNM
p.
RIIM
SMZ
Univ.
Vol.
WOM
following
Musical Quarterly
number
Neue Musikzeitung
Neue ZeitschriJt fur Musik
Osterreichen MusikzeitschriJt
Perspectives of New Music
page(s)
Repertoire International de Litterature Musicale
Schweizerische MusikzeitschriJt
University
Volume(s)
World of Music
The following list is arranged in alphabetical order by recording company
(and record number). Barring the various format exceptions (CD =
compact disk; MC = cassette) the recordings are presented on long-playing
records. Each entry also indicates the presence of works by anotJler
composer (with an asterisk [*]); the title of the compositions by Iannis
Xenakis; and the performers.
1. Adda81042 *
Children's chorus of Notre-
Mikka, Mikka S
Dame de Paris; Ensemble In-
P.Zukovsky,vl.
strumental de musique
2. Adda 581224 CD *
contemporaine de Paris; K. Si-
Oophaa
monovich, dir.
E. Chojnacka, clay; S. Gualda,
8. Arion ARN 38775 *
perc.
Nuits
3. Ades 14.122-2 CD *
Groupe Vocal de F r a n ~ e ; M.
Metastasis, Nuits
Tranchant, dir.
Orchestra del Siidwestfunk; H,
9. Babel 9054-1 CD
Rosbaud, dir.; Soloists from
Echange, Palimsesl, Waarg, Eonta
the ORTF chorus; M.
A Takahashi, pno; H. Sparnaay,
Couraud, dir.
dB; Asko Ensemble, D. Por-
4. Ades 16005 *
celijn, dir.
Herma
10. Barclay 920217 *
J. Mefano, pf.
Anaktoria, Monsima-A morsima
5. Angel S-36560 *
Octuor de Paris
A trees
11. Bis 256 *
Ensemble Instrumental de
Psappha
musique contemporainc de
G. Mortensen, perc.
Paris; K. Simonovich, dir. 12. Bis 338 *
6. Angel S-36655 *
Keren
Herma
Christian Lindberg, Th.
G. Pludemacher, pf. 13. Bis CD 482 CD
7. Angel S-36656
Ple"iades
Achorripsis, Akrata, Polla ta dhina,
The Kroumata Percussion En-
STIIO
semble
365
366
14. Boite a Musique 070 *
DiaflWrphoses
(electronic composition)
15. CBS 3461226
Akrata
The Festival Chamber En-
semble; R. DufalIo, dir.
16. CBS Sony 32 DC 673 CD *
Psappha
S. J oolihara, perc.
17. CBS Sony 32 DC 691 CD *
Pliades
M. Aruga Ensemble
18. CBS Sony 32DC691 CD
PleIades
Makoto Aruga Percussion En-
semble
19. Candide 31049
Medea, Polytope de Montreal,
Syrmos
Ensemble Ars Nova; ORTF
Chorus; M. Constant, dir.
20. Candide 31000
NOflWS Gamma, Terretektorth
ORTF Philharmonic Orchestra;
Ch. Bruck, dir.
Formalized Music
22. Chant du Monde LDC 278 368
CD*
Eonta
Y. Takahashi, pno; Paris Con-
temporary Music Ensemble,
K. Simonovitch, dir.
Metastasis Pithoprakta
National Orch ORTF; M. Le
Roux, dir.
23. Colosseum 3447253 CD *
Stratigie
Yomiuri Nippon Symphony Or-
chestra; Seiji Ozawa and H.
Wakasugi, dirs.
24. Colombia MS-7281 '"
Akrata
semble; R. Dufallo dir.
25. Connaisseur Musik CP 6 '"
Mikka, Mikka S
P. Zukovsky, vI.
26. Cybernetics Serendipity Music
lCAOl.02 *
Stratigie
chestra; S. Ozawa and E.
Wakasugi, dirs.
27. Decca 411610-1
21. Chant du Monde LDX 78308
(also LDX-A 8368, CD LDC Jonchaies
278368) 28. Decca HEAD 13 (also 591171)
Eonta, Metastasis, Pithoprakta Antikhton, Aroura, Synaphai
Ensemble Instrumental de G. Madge, pf.; New Philhar-
musique contemporaine de monia Orchestra; E. Howarth,
Paris; K. Simonovich, dir.; Or- dir.
chestre National de I'ORTF; 29. Denon CO 73 768 CD
Le Roux, dir.; Y. Takahashi, Ple'iades
pf.; L. Longo, P. Thibaud, tr.; Les Percussions de Strashourg
M. Chapcllier, G. Moisan, J. 30. Denon OX-7063 ND (also CD
Toulon, trh. CO-I052) *
E vryali, H eT'ITUL
Y. Takahashi, pf.
Discography
31. Deutsche Grammophon 2530562
*
NOflWS Alpha
S. Palm, vcl.
32. Disques Montaigne 782002 CD
Okho
Le cercle Trio (perc).
33. EMI Angel EM-85013-5 *
HeT'ITUL
A Takahashi, pf.
34. EMI C0631001
367
E. Chojnacka, clav; Ensemble
lannis Xenakis; H. Kerstens,
dir.
41. Erato Musicfrance 2292 45019-2
CD*
Ikhoor
Paris String Trio
42. Erato Musicfrance 2292 45030-2
CD*
Khoa'i, Kombo'i
E. Chojnacka, clar; S. Gualda,
Achorripsis, Akrata, Polla ta dhina, perc.
ST/IO 43. Erato NUM 75104 *
Children's chorus of Notre-
Kombol
Dame de Paris; Ensemble In- E. Chojnacka, c1av.; S. Gualda,
strumental de musique perc.
contemporaine de Paris; K. Si- 44. Erato STU 70457 *
monovich, dir. Nuits
35. EMI CVB 2190 '"
HeT'ITUL
G. Plude macher, pf.
36. EMI CVC-2086 (also MC
MCV-2086c)
Atrees, Morsima-Amorsima, Nomos
Alpha, ST/4
P. Penassou, vel, : Quatuor
Bernede; Ensemble Instru-
mental de musique contem-
porainc de Paris; K.
Simonovich, dir.
37. EMI Columbia SCXG-55 *
ST/10
Th. Antoniou, dir.
38. EMI HMV CSDG-63 *
Anaktoria
Soloists and Chorus of ORTF;
M. Couraud, dir.
45. Erato STU 70526 (also 9088)
Medea, Polytope de Montreal,
SyrflWs
Ensemble Ars Nova; Chorus of
ORTF; M. Constant, dir.
46. Erato STU 70527/28 *
Kraanerg
M. Constant, dir.
47. Erato STU 70529 (also 91 19)
NOflWS Gamma, Terretektorh
ORTF Philharmonic Orchestra:
Ch. Bruck, dir.
48. Erato STU 70530
Bohor, Concret PH, Diamorphoses,
Orient- Occident
Th. Antoniou, dir. (electronic composition)
39. Elektrola (Hor Zu) 49. Erato STU 70656 (also 9137)
Concret PH, Medea, Orient-Occident Oresteia
(electronic composition) S. Caillat Chorale; Maitrisc de
40. Erato 2292-45030-2 CD Notre-Dame de Paris; En-
semble Ars Nova; M. Con-
A l'ile de Goree, Naama
stant, dir.
368
50. Erato STU 71106 *
Psappha
S. Gualda, perc.
51. Erato STU 71266 *
Formalized Music
57. Finlandia 120366-2 CD *
Anaktoria
Members of t11C Avanti Chamber
Orchestra
Khoa"i 58. Finlandia F ACD 357 >I<
E. Chojnacka, clay. Khoa'i
52. Erato STU 71513 J. Tiensu, clav.
Cendrees,Jonchaies, Norrws Gamma 59. Gaudeamus Foundation - Radio
Gulbenkian Foundation Chorus Net11erland (3 CD, 1988)
of Lisbon; Orchestre National Gmeeoorh
de France; M. Tabachnik, dir.; K. Hoek, Organ
ORTF Philharmonic Or-
chestra; Ch. Bruck, dir.
53. Erato Interfuces (for
Hewlett-Packard)
AiS.
60. Gramavision R2 79440 CD *
Tetras
Arditti String Quartet
61. HMV S-ASD 2441
Morsima-Arrwrsima, Norrws Alpha,
Spyros Sakkas, baritone; Sylvia ST/4
Gualda, percussion; Sym- P P I
. enassou, vc .; Quatuor
phonieorchester des Bay- Berncde; Ensemble Instru-
erischen Rundfunk, dir.
mental de musique contem-
Michel Tabachnik; poraine de Paris; 1<..
Kekuia. S . h d'
ImonOVlC , Ir.
Kolner Rundfunk-Symphonieor-
chester; KOiner Rundfunchor, 62. Mundi HMC 5172 *
dir. Michel Tabachnik. MISts
N'shima. C. Helffer, pf.
Anne Bartolloni, Genevieve
Renon, mezzo-sopranos; En-
semble Instrumentale, dir. Mi-
chel Tabachnik.
54. Ernst Klett 92422 *
Diarrwrphoses
55. Etcetera KTC 1075 *
Kraanberg
Alpha Centauri Ens., R. Wood-
ward, dir.
56. Eterna Stereo 827906 *
Dmaathen
63. Harmonia Mundi HMC 905185
CD
Pleiades
Les Percussions de Strasbourg
64. Hungaroton 12569 (also CD
BCD 12569) >I<
Mists
K. Kormendi, pf.
65. Jeugden Muziek BVHAAST 007
Eonta, Evryali, Herma
G. Madge, pf. P. Eotvos, dir.
Discography
66. Limelight 86047 >I<
Orient-Occident
67. Lyra 251
Eonta, Metastasis, Pithoprakta
Y. Takahashi, pf.; Ensemble In-
strumental de musique con-
temporaine de Paris; K.
Simonovich, dir.
68. Mainstream 5000 *
Herma
Y. Takahashi, pf.
69. Mainstream MS-5008 *
Achorripsis
Hamburger Kammersolisten; F.
Travis, dir.
70. Musical Society MHS 1187 *
Medea, Nuits
Ensemble Ars Nova; ORTF
Chorus; M. Couraud, dir.
71. Musical observations CP 2J6 *
Mikka, Mikka S
P.Zukovsky,vl.
72. Musidisc RC-16013
Anaktoria, Morsima-Arrwrsima
Octuor de Paris
73. Neuma 450-71 >I<
Theraps
R. Black, Cb.
74 Neuma 450-74 CD *
MyceneA
75. Nieuwe Muziek 004
Dmaathen, Epe'i, Palimpsest,
Phlegra
Xenakis Ensemble; H. Kerstens,
dir.
76. Nippon SFX-8683 *
Persepolis
369
77. Nonesuch 32818 (also H-71201) *
Akrata, Pithoprakta
Buffalo Philharmonic Orchestra;
Lukas Foss, dir.
78. Nonesuch H-71245
Bohor, Concel PH, Diarrwrphoses,
Orient- Occident
79. Owl 26 *
Charisma
Jungerman, cl.; Banks, vcl.
80. Performance PER 84061 *
Jonchaies
Leicestershire Schools Sym-
phony Orchestra; P. Fletcher,
dir.
81. Philips 6521020 (also 6718040) '"
Persephassa
Les Percussions de Strasbourg
82. Philips 835485/86 (also A
00565/66 L, 836897 DSY) *
Orient- Occident
83. Philips 835487
Analogique A et B, Concret PH
Ensemble Instrumental de
musique contemporaine de
Paris; K. Simonovich, dir.
84. Philips T 6521045
Persepolis
85. PNM (Perspectives of New
Music) 28 CD
Voyage absolu des Unaris vers
Andrornede
(electronic composition com-
posed on UPIC at CEMAMu)
86. RCA RS 9009 (also RE 25444)
Dikhthas, Embellie, lkhoor, Kollos,
Mikka, Mikka S, ST/4
Arditti Quartet
370
87. RCA Victor JRZ-250I *
Hibiki Hana Ma
88. RCA Victor SJV-ISI3 *
Strategie
chestra; S. Ozawa and H.
Wakasugi, dirs.
89. SaiabertActuels SCD 8906 CD
(dist. Harmonium Mundi)
Oresteia, Kassandra
U. ofStrasbourg Chorus;
Maltrise de Colmar; Altiou
Vocal Ensemble; Ensemble de
Basse-Normandie; D. Debart,
dir; R. Weddle, Vocal dir; S.
Sakkos, bar; S. Gualda, perc.
Sony CBS SONC-I0I63 *
Akrata
semble; R. Dufallo, dir.
90. Teldec 6.42339 AG (also CD
8.42339 ZK) *
Retours- Windungen
The 12 cellists from the Berlin
Philharmonic.
91. TeleclWarner Classics 2292
46442-2 CD *
Eonta.
R. Hind, pno; London Brass
92. Toshiba TA-72034 *
Evryali
A Takahashi, pf.
Formalized Music
93. Vanguard Cardinal 10030
Eonta, Metastasis, Pithoprakta
Y. Takahashi, pf.; Ensemble In-
strumental de musique con-
temporaine de Paris, K.
Simonovich, dir.
94. Varese Sarabande 81060 *
Strategie
chestra; S. Ozawa and H.
Wakasugi, dirs.
95. Wergo WER 6178-2 CD
Akanthos, Dikhthas, Palimpsest, Eper
I. Arditti, vln; C. Helffer, pno; P.
Walmsey-Clark, sop; Spec-
trum Ensemble; G. Protheroe,
dir.
In Preparation:
96. Disques Montaigtne 782xxx 3 CD
Evryali, Mists, Herma, Dikhthas,
Akea, Tetras, ST/4, Mikka, Mikka
liS", Kot/os, No'IrWs Alpha, lkhoor,
Embellie
C. Helffer, pno; Arditti String
Quartet
97. MFA (collection Musique
Franc;ais d'Aujourd'hui)
Charisma
A, Damiens, cI; P. Strauch, vIc.
98. Salabert Actuels SeD 9102 CD
(dist. Harmonium Mundi)
Bokor, La ligende D'Ee1
(electronic compositions)
Iannis Xenakis
Biographical Information
1957: Geneva, European Cultural Foundation Award
1963: Athens, Manos Hadjidakis Award
1963-64: Berlin, Ford Foundation Grant plus Grant from the West-Berlin
Senate
1964: Paris, Musiques Formelles chosen by the Permanent Committee of the
French Book and Graphic Arts Exhibits, to be one of the 50 "Books
of the Year."
1965: Paris, Grand Prize awarded by the French Recording Academy Com-
petition.
1968: Edinburgh, First Prize at the Computer-assisted Music Competition,
IFIP Congress
:Paris, Grand Prize awarded by the French Recording Academy
:London, Bax Society Prize (Harriet Cohen International Music
Awards)
1970 Paris Grand Prize awarded by the French Recording Academy
1971: Tokyo Modern Music Award from the Nippon Academy Awards
1972: London, Honorary Member of the British Computer Arts Society
1974: Paris, Gold Medal Maurice Ravel Award from the SACEM
1975: Honorary Member of the American Academy of Arts and Letters
1976: Paris, Sorbonne, Doctorat es Letters and Humanities
: Paris, National Grand Prize in Music from the French Cultural
Secretary of State
37]
372
Formalized Music
1977: Paris, G r ~ n d Prize, Charles Cros Academy for Recordings (Grand Prix
du President de la Republique in honorem)
:Bonn, Beethoven Prize
:Amste:dam, Edison Award for the best recording of contemporary
mUSIC
1981: Paris, Officier de I'Ordre des Arts et des Lettres
1982: Paris, Chevalier de la Legion d'Honneur
1983: Paris, Member of the Institut de France (Academie des Beaux Arts)
:Berlin and Munich, Member of the Akademie der Kunste
1985: Paris, Officier de I'Ordre National du Merite
: Athens, Medal of Honor of the City
1986: Paris, Ordre National du Merite
1987: Honorary Member of the Scottish Society of Composers
:Grand Prize from the City of Paris
1988: Paris, Nominated to the Victoires de la Musique
1989: Foreign member 0 f the Swedish Royal Academy of Music
1990: Professor Emeritus of the Universitc de Paris I, Pantheon-Sorbonne
:Honorary Doctor of the University of Edinburgh
:Honorary Doctor of the University of Glasglow
Notes
I. Free Stochastic Music
1. Jean Piagct, Le developpement de la notion de temps chez I' enfant (Paris:
Presses Universitaires de France, 1946).
2. 1. Xenakis, Gravesaner Blatter, no. I (1955).
3. 1. Xenakis, Revue technique Philips, vol. 20, no. I (1958), and Le
Corbusier, Modulor 2 (Boulogne-Seine: Architecture d'Aujourd'hui, 1955).
4. I. Xenakis, "Wahrscheinlichkeitstheorie und Musik," Gravesaner
Blatter, no. 6 (1956).
5. Ibid.
6. Ibid.
II. Markovian Stochastic Music-Theory
1. The description of the elementary structure of the sonic symbols that
is given here serves as a point of departure for the musical realization, and
is consequently only a hypothesis, rather than an established scientific fact.
It can, nevertheless, be considered as a first approximation to the considera-
tions introduced in information theory by Ga,bor [I]. In the so-called Gabor
matrix a sonic event is resolved into elementary acoustic signals of very
short effective durations, whose amplitude can be divided equally into
quanta in the sense of inlormation theory. However, these elementary
signals constitute sinusoidal functions having a Gaussian" bell" curve as
an envelope. But one can pretty well represent these signals of Gabor's by
sine waves of short duration with an approximately rectangular envelope.
2. The choice of the logarithmic scale and of the base between 2 and 3
is made in order to establish our ideas. In any case, it corresponds to the
results of research III experimental music made by the author, c.g.,
Diamorphoses.
373
374
Formalized Music
V. Free Stochastic Music by Computer
1. See Gravesaner BlaUler, nos. 11/12 (Mainz: Ars Viva Verlag 1957)
R ' .
2. (V3)e must be equal to the upper limit, e.g., to 150 sounds/sec. in the
case ofa large orchestra.
VI. SymbOlic Music
1. A second-degree acoustic and musical experience makes it necessary
to the Fourier analysis, and therefore the predominance of
frequency III sound construction. But this problem will be treated in Chapter
IX.
2. From previous edition of Formalized Music, another way to map these
same four forms:
y
____ -+ ___ x
Z=x+yi
f] = Z = x + yi = Z = f](Z) = original form
f2 = x --yi = I Z 12/ Z = f
2
(Z) = inversion
f3 = -x - yi = -Z = fiZ) = inverted retrogradation
f4 = -x + yi = -( I Z 12f Z) = fiZ) = retrogradation
Notes 375
VII. Towards a Metamusic
l. Cf. L Xenakis, Gravesaner Blatter, no. 29 (Gravesano, Tessin,
Switzerland, 1965).
2. Cf. L Xenakis, Gravesener Blatter, nos. 1,6; the scores of Metastasis and
Pitfwprakta (London: Boosey and Hawkes, 1954 and 1956); and the recording
by i.e Chant du Monde, L.D.x. A-8368 or Vanguard.
3. I do not mention here the fact that some present-day music uses
quarter-tones or sixth-tones because they really do not escape from the tonal
diatonic field.
4. Cf. Chap. VI.
5. Johannis Tinctoris, Terminorum Musicae Diffinitorum (Paris:
Richard-Masse, 1951).
6. Jacques Chailley, "I.e my the des modes grecs," Acta Musicologica, vol.
XXVIII, fase. IV (Basel Barenreiter-Verlag, 1956).
7. R. Westphal, Aristoxerws von Tarent, Melik und Rhythmik (Leipzig:
Verlag von Arnbr. Abel (Arthur Meiner), 1893), introduction in German,
Greek text.
8. G. Th. Guilbaud, Mathtm.atiques, Tome I (Paris: Presses Universitaires
de France, 1963).
9. AristidouKointiliano ,Peri Mousikes Proton (Leipzig: Teubner, 1963),
at Iibrairie des Meridiens, Paris.
10. The Aristoxenean scale seems to be one of the experimental
versions of the ancient diatonic, but does not conform to the theoretical
versions of either the Pythagoreans or the Aristoxeneans, X(9/8)(9/8) = 4/3
and 6 + 12 + 12 = 30 segments, respectively. Archytas' version, X(7/8) (9/8)
= 4/3, or Euclid's are significant. On the other hand, the so-called Zarlino
scale is nothing but the so-called Aristoxenean scale, which, in reality, only
dates back to Ptolemy and Didymos.
II. Avraam Evthymiadis ETOLXIlLW8T) 'MovaLK1js
(Thessaloniki: OXA, Apostoliki Diakonia, 1948).
12. In Quintilian and Ptolemy the perfect fourth was divided into 60
equal tempered segments.
13. See Westphal, pp. XLVIIff. for the displacement of the tetra chord
mentioned by Ptolemy: lichanos (16/15) mese (9/8) paramese (10/9) trite
(Harmonics 2.1, p. 49).
14. In Ptolemy the names of the chromatic tetrachords were permuted:
the soft chromatic contained the interval 6/9, the hard or syutonon the
iuterval 7/6. Cf. Westphal, p. XXXII.
15. Selidion 1 : a mixture of the syntonon chromatic (22/21, 12/11, 7/6)
and toniaion diatonic (28/27, 7/8, 9/8) ; selidion 2 : a mixture of the soft
diatonic (2l!20, 10/9, 8/7) and the toniaion diatonic (28/27, 817, 9/8), etc.
Westphal, p. XLVIII.
16. Egon Wellesz, A History of Byzantine Music and Hymnography (Oxford:
Clarendon Press, 1961), pp. 7lff. On p. 70 he again takes up the myth that
the ancient scales descended.
17. The same negligence can be found among the students of ancient
Hellenic culture; for example, the classic Louis Laloy in Aristoxene de Tarente,
1904, p. 249.
18. Alain Danielou lived in India for many years and learned to play
Indian instruments. Mantle Hood did the same with Indonesian music, and
let us not forget Than Van Khe, theoretician and practicing performer and
composer oftraditional Vietnamese music.
Cf. WeIIesz. Also the transcriptions by C. H6eg, another great
Byzantinist who neglected the problems of structure.
20. Imagine the bewilderment of the "specialists" when they discovered
that the Byzantine musical notation is used today in traditional Romanian
folk music! Rapports Complbnenlaires du Xlle Congres international des Etudes
byzantines, Ochrida, Yugoslavia, 1961, p. 76. These experts without doubt
ignore the fact that an identical phenomenon exists in Greece.
21. Cf. my text on disc L.D.X, A-8368, issued by Le Chant du Monde.
See also Gravesaner BIiiUer, no. 29, and Chap. VI of the present book.
22. Among themselves the elementary displacements are like the
integers, that is, they are defined like elements of the same axiomatics.
23. Alain Danielou, Northern Indian Music (Barnet, Hertfordshire:
Halcyon Press, 1954), vol. II, p.72.
24. This perhaps fulfilIs Edward Varese's wish for a spiral scale, that is, a
cycle of fifths which would not lead to a perfect octave. This information,
unfortunately abridged, was given me by Odile Vivier.
25. These last structures were used in Akrata (1964) for sixteen winds,
and in Nomos alpha (1965) for solo cello.
Notes
377
VIII. Towards a Philosophy of Music
1. The "unveiling of the historical tradition" is used in E. RusserI's
. f H l' VI "Die Krisis der Europaischen Wlssenschaften und
sense, c, Ulna, . ., . die
die transzendentale Phanomenologie (Eine Emleltung m ..
. '1 h' )" P G omelry (The Rague: M.
phanomenolo81sche Plu osop Ie , ure e
1954), pp. 21-25, and Appendix III, pp. 379-80.
2. Cf. Upanishads and Bhaga di Gila, references by K.
. . . nd B ddh' (New York' Philosophlcal LIbrary,
Coomaraswamy III Hmduzsm a u lsm .
1943).
3. "Perhaps the oddest thing about modern science is its return to
pythagoricism." Bertrand Russell, The Nation, 27 September 1924.
4. In this translation I have considered the original Greek text
translations by John Burnet in Early Greek York: .:r.:endJan
Books, 1962) and by Jean Beaufret in Le Poeme de Parmemde (Pans. P.U.F.,
1955).
1. Elements are always real: (earth, water, air) = fire)
energy. Their equivalence had ready been foreseen by Herachtus.
6. Lucretius, De fa Nature, trans. A Ernout (Paris, 1924).
7. The term stochastic is used for the first time in this work. Today it is
synonymous with probability, aleatory, chance. . .
8 E B I E
l1ments de fa !Morie des probabilitis (Paris: Albm MiChel,
. . ore,
1950), p. 82.
9. Uncertainty, measured by the entropy of information tlleory, reaches
a maximum when the probabilities p and (1- P) are equal.
10. Cf. I. Xenakis, Gravesaner Blatter, nos. 1,6,11/12 (1955-8).
11. I prepared a new interpretation of of limited
transpositions," which was to have been published in a collectIOn III 1966, but
which has not yet appeared. .,
12. Around 1870 A. de Bertha created his "gammes lwmotones pre.mtere.et
seconde," scales of alternating whole and halftones, which would be wnttcn III
our notation as (3n V 30 + 2, 3n V 3n + 1).
13. In 1895, Loquin, professor at the Bordeaux Conservatory, had
alread y preconceived the equality of the twelve tones of the octave.
14. The following is a new axiomatization of the sieves, more natural
than the one in Chaps. VI and VII.
Basic Assumptions. 1. The sensations create discrete characteristics, values,
stops (pitches, instants, intensities, ... ), which can be represented as points. 2.
Sensations plus comparisons of them create differences between the above
characteristics or points, which can be described as the movement, the
displacement, or the step from one discrete characteristic to another, from
one point to another. 3. We are able to repeat, iterate, concatenate the above
steps. 4. There are two orientations in the iterations-more iterations, fewer
iterations.
Formalization. Sets. The basic assumptions above engender three
fundamental sets : Q, L'., E, respectively. From the first assumption
characteristics will belong to various specific domains Q. From the second,
displacements or steps in a specific domain Q will belong to set 1'., which is
independent ofQ. From the third, concatenations or iterations of elements of
L'. form a set E The two orientations in the fourth assumption can be
represented by + and -.
Product Sets. a. Q X L'. Q (a pitch-point combined with a
displacement produces a pitch-point). b. Q X E L'. (a displacement
combined with an iteration or a concatenation produces a displacement). We
can easily identifty E as the set N of natural numbers plus zero. Moreover, the
fourth basic assumption leads directly to the definition of the set ofintegers Z
fromE
We have thus bypassed the direct use of Peano axiomatics (introduced
in Chaps. VI aud VII) in order to generate an Equally Tempered Chromatic
G.amut (defined as an ETCHG sieve). Indeed it is sufficient to choose any
dIsplacement ELD to set 11 and form the product {ELD} X Z. Set
11 (set of melodic intervals, e.g.), on the other hand, has a group structure.
15. Cf Olivier Messiaen, Technique de mon langage musical (Paris:
Durand, 1944).
-J -: r : " , I , I
.
cM.H # E f
C scale Figure 2.
,5" .0
I I f I If., I 10 I' I
_r
I I
-",
I I I I I
Ff" 'i'
B eel D 8Cc.,1>
'" .... f'/
Messiaen Mode N4 Figure 3.
Messiaen Mode N4
Figure 4.
Notes
379
16. " .. therefore tones higher than needed become relaxed [lower], as
they should be, by curtailment of movement; conversely those .lowe.r dIan
needed become tensed [higher], as they should be, by adJun:tIOn of
movement. This is why it is necessary to say dIat tones are constituted of
discrete pieces, since it is by adjunction and curtailment that become
tlley should be. All things composed of discrete pieces are saId to be
numerical ratio to each other. Therefore we must say that tones are also III
numerical ratio to each other. But among numbers, some are said to be in
multiplicative ratio, others in an epimorios [1 + l/x], or others in an epimens
ratio [an integer plus a fraction having a numerator other than one] ;
therefore it is necessary to say that tones are also in these same ratios to each
other ... " Euclid, Katatomi Kanonos (12-24), in Henricus Menge, Phamomena
et Scripta Musica (Leipzig: B. G. Teubner, 1916). This remarkable text already
attempts to establish axiomatically the correspondence between tones and
numbers. This is why I bring it in in the context of this article.
17. Cf. my analysis of Metastasis, in Corbusier, Modulor 2
(Boulogne-Seine: Architecture d'Aujourd'hui, 1955.)
18. Cf. Score by BODsey and Hawkes, eds., and record by Pathe-Marconi
and Angel.
19. Hibiki-Ha11il-Ma, the electro-acoustic composition that I was
commissioned to write for the Japanese Steel Federation Pavilion at the 1970
Osaka World Expo, used 800 loudspeakers, scattered in the air and in the
ground. They were divided into approximately 150 independe.nt
sounds were designed to traverse these groups according to
diagrams. After tlle Philips Pavilion at the 1958 World s :au, tlle
Steel Pavilion was the most advanced attempt at placmg sounds III space.
However, only twelve independent magnetic tracks were available (two
synchronized six-track tape recorders).
20. Mario Bois, Iannis Xenakis: The Man and His Music (New York:
Boosey and Hawkes, 1967).
21. Jean Piaget, Ie Mveloppement de La notion de temps chez i'enfant, and La
representation de l'espace chez I'mfant (Paris. Presses Universitaires de France,
1946 and 1948).
X. Concerning Time, Space and Music
1. Shannon C. and Weaver W.,The Mathematical Theory of Communication
(Urbana: University of Illinois Press, 1949).
2. Eddington, The Nature of the Physical World (New York: Macmillan,
1929).
3. Prigogine, 1., Physique Temps et Devenir (Paris: Masson, 1982).
4. Born, Max, Einstein's Theory of Relativity (New York: Dover, 1965).
5. Morrison, Philip, "The Overthrow of Parity," Scientific American,
April, 1957.
6. Gardner, Martin, "Can Time Go Backward," ScientifIC American, Jan.
1967, p.98.
7. Reichenbach, H., The Philosphy of Space and Time (New York: Dover,
1958).
8. Linde, A D., Physics Letters (1983), 129B, 177.
9. See also Coveney, Peter V.,"The Second Law of Thermodynamics:
Entropy, Irreversibility and Dynamics," Nature N 333 (1988).
10. The idea of the Big Bang, a consequence of the shift (expansion of
the universe) toward the red, is not accepted by all physicists. See Nikias
Stravroulakis, "Solitons et propagation d,actions suivant la relativite
generale," Annales de la Fondation de Broglie 12 N 4 (1987).
11. Russell, B, Introductwn Ii la philosophie mathematique (Paris: Payot,
1961).
12. Cf. chapter 9 in Formalized Music, "New Proposals in Microsound
Structure."
13. C Xenakis, autori vari (a cura di Enzo Restagno) (Torino: KD.T.,
1988).
XI. Sieves
1. Earlier articles on "sieves" by Iannis Xenakis have appeared in
Preuves, Nov. 1965, Paris; liL Nef n 29, 1967, Paris; Revue d'Esthetique vol.
xxi, 1968, Paris; Tempo no 93, 1970, as well as the previous editions of
Formalized Music.
2. As for rhythm outside of Western civilization, cf. AROM, Simha, "Du
pied a la main: Les fondements metriques des musiques traditionelles
d'Afrique Centrale;" Analyse Musicale 1 trimestre, 1988.
Notes
381
3. Let there be (M, I), with M being a composite ofthe form:
M = m
k
* nl ... * rt
It is sometimes necessary and possible to decompose it into :
(m
k
, 1m) n (nl,In) n ... (r
j
, Ir) = (M, I).
4. Euclid's algorithm. Let y, x be two positive whole Begin by
letting D = MOD(y,x), then replace y with x and x with D. If D IS not equal to
0, then start over. But if D = 0, then the last is the largest common
denominator. Let us call Illis last y, D.
take two numbers: y, x
1)
2)
3)
[
D ..... MOD(y, x)
y ..... x,x ..... D

END
example:
y = 30, x = 21
D ...... 9 D ...... MOD(21,9) = 3{ D ...... MOD(9,3) = 0
y ...... 21, x ...... 9 Y ...... 9, x ...... 3 Y 3, x ...... 0
D ...... D ...... D ..... O=O
therefore
D..- y=3
END
5. a modulo b, notated MOD(a, b), is equal to the residue ofthe division
of a by b: a / b = e + r / b where r is this residue, if a, b, e, and r are elements
ofN.
6. * C2, Cl) = 1 represents the integer equadon:
* C2 / Cl = v + 1 / Cl.
Index
Achorripsis, 24, 26-38, 131, 133-43,
295
Alypios, 185
Analogique A, 79, 98-103,105, 109.
See also Markov chain
Analogique B, xiv, xv (illustrations),
79, 103-9. See Markov chain
Anaxagoras. 203
Anaximander, 201
Anaximenes.201
Arc sine function. See Probability
laws
"Arcs," 329-34; (definition), 329
Aristotle, 181
Aristoxenas ofTarent, 269
Aristoxenos, 183-9, 195,202,208,
210
Ataxy, 63, 75-78
Atrees, 136-37, 144
Bachet de Meziriac, 272
Barbaud, P., 133
Barraud,Jacques, 131 , 134
Beethoven, 1 , 164, 169
Bernoulli, Daniel, 205
Bernoulli,Jacques, 206
Bessel function, 247
Big Bang theory, 259, 260, 295
Blanchard, P., 133
Boltzmann, 15,255,257;
theorem, 61
Boolean operations, 209
Borel, Emile, 39, 206
Boudouris, Georges, 133
Brout, R., 260
Brownian movement, 289
383
cauchy, 266, 290,293
cauchy function. See Probability
laws
causality, 257, 258, 259: principle,
1,4,8-9
Chailley, Jacques, 183
CEMAMu (Center for Studies in
Mathematics and Automation of
Music), xii, xiii, 329
Chance, 4, 38-39, 259; definition, 25
Channels (in computing), 329-34
Charbonnier, Jeanine, 133
Combarieu, 183
Computers, 258, 266, 268, 329-34:
hardware, 333-34,software. 333-
34; See also Stochastic music
Concret PH, 43
Coveney, Peter V., 260
Debussy, 5,193,208; Debussian
whole-tone scale, 196
Delayed choice", 257
Descartes, 263: discourse on
method,54
Determinism, 204-5
Diamorphoses, 43
Didymos, 186
Distribution, random-like, 265
Dud, 13-22, 124. See Game theory
Eddington, 255
Einstein, 256, 263
Englert, F., 260
Entropy, 16,61-68,75-78,211,255,
256,257; definition, 61, 186;
mean entropy, 75
Envelopes (in computing), 329-34
384
Epicurus, 24, 205-6,237
Ergodism, 56, 67
Exponential probability function.
See Probability laws
Fermat, 206
Feynman, 257
Fletcher-Munson diagram, 47-49,
243
Flux, 266
Fourier, 258, 266, 293: series, 242
Fractals, sounding, 266, 293
Frechet, Maurice, 79
French Cultural Ministry, 329fn
Frequency and amplitude tables, 329
Fulchignoni, E., 43
Gabor elementary signals, 54, 58,
103
Game theory, 10, 133 ; analysis of
Duel, 13-22; analysis of Slraiigie,
122-23, 125-30; autonomous
music, 1l0-1l ; heteronomous
music, 111-13 ; two-person zero-
sum, 112;
Gauss, 266
Gaussian probability distribution.
See Probability laws
Genuys, Franc;ois, 131, 134
Glissando. See Sound
Graphic score editing, 329
Gregorian chant, 183
Groups, 210; Abelian additive
group structure, 160, 199; Klein,
170
'Guth, Alan, 259
Guttman. N., 133
Gi.inzig, E., 260
Heisenberg, 206 n., 237
Henry, Pierre, 243
Heraclitus, 267
Herakleitos, 202, 259
Henna, 175-77
Index
Hctcronomy. Sec Game theory
HibiJU-lJana-Ma, 269, n.19
I-Iiller, 133
Hindcmith,243
Hucbald, 183,202
Hyperbolic cosine function. See
Probability laws
Interactive composition, 329-34
Isaacson, 133
Isotropy, 14
Kinetic theory of gases, 15,49,95,
205,213,244
Knowledge, 261
Korybantes, 202
Kratylos, 259
La Ugende d'Eer, xii, 293, 296
Lamb's shift", definition, 259
Landmark points, 264, 265
Law, definition, 258
Law oflarge numbers, 4, 8, 16,31,
206
Laws, 267
Le Corbusier, 10
Lee, T. D., 256
Leukippos, 203
Light, velocity of, 256
Linde, 259
Linear probability function. See
Probability laws
Liouville, equation of, 256
Logic, 276
Logistic function. See Probability
laws
Logistic probabilities, 266
Lorenz-Fizgerald, 256
Lucretius, 205
Levy, Paul, 15, 24
Macrocomposition, 22 ; methods,
49. See microsound structure
Index
Macroscopic comp[osition, 256
Markov chain, 73-75. 133,244,248-
49; analysis of compositional ap-
proach, 79-98; entropy of, 86;
matrix of transition probabilities,
74-75,78,82-4, 109; realizatiou of
Analogique A, 98-103; realization of
Analogique B, 103-8; stationary dis-
tribution, 75, 85; use of screens,
79-109
Marino, Gerard, 277, 329fn
Marsima-Amorsima, 144
Mathews, M. V., 133, 246n.
Matrix of transition probabilities.
Sce Markov chain
Maxwell, 15 ; formula, 55
Memory, 258,262,264,266
Messiaen, Olivier, 5, 8,192,208,
210,268
Me/aslais, 2-3, 10
Meyer-Eppler's studies, 244
Microcomposition, 22-23, 50. See
Microsound structure
Microsound structure, 242-54; mi-
crocomposition based on prob-
ability distributions, 24-9;
macrocomposition, 249. See Pour-
ier series
MIDI devices, 332
Modulating arc assignment, 329
Moduli,267
Monteverdi, 208
Moussorgsky, 208
Music, definition and historical back-
ground, 1, 4-5, 8; ancient Greek,
182-85,192; atonal, 4; Byzantine,
182, 186-92,208; electronic, 8, 52,
243; electro-acoustic, 243; electro-
magnetic, 8, 16,43, 103. See
Stochastic music, Serial music,
Game theory
Nomos alpha, 208-9, 219-36
Nomos gamma, 217, 219, 236-41
Non-linear functions, 295
Nothingness, 259, 260,261,267,295
Orient-Occident, 43
Originality, 258
Orphism, 201-2
Parity symmetry, 256
385
Parmenides, 24 n., 202-4,207,209,
259,260,262,263,267
Pascal,206
Peano, 159 n., 194-95,265
"Performance" system, 329
Periodicities, 268
Philippot, Michel, 39-42; Composition
pour double orchestre, 39
Philips Pavilion, 6-7, 10- 11, 43
Piaget, Jean, 5,160
Pierce, J. R., 133
Pitches, 267, 268, 269
Pithoprakta, 15, 17-21
Plato, 1, 179,202,257: Poliiicos, 257,
295
Poincare, H., 206
Poisson law. See Probability laws
Prigogine, I., 256
Probabilies, 256
Probabilistic wave form, 289; See
also Probability
Probability, definition, 207; Cauchy,
246,251-52; distributions, 260; ex-
ponential, 12, 134, 142, 246;
Gaussian, 14-15,33,56, 60, 140,
246; Interlocked densities: ex-
ponential X Cauchy, 251-52; lo-
gistic X exponential, 254;
hyperbolic cosine X exponential,
253-54: laws, arc sine, 246; linear,
13,136; logistic, 246, 250, 291,
293; Poisson, 12, 16,23-25, 29-32,
54,66, 133,246-47; unifonn, 246;
Wiener-Levy, 247; theory, 9, 255
Ptolemaeos, Claudios, 185
Ptolemy, 186, 188
Pythagoras, 1, 207, 209; concept of
numbers, 201 ; Pythagoreans,
185,201,242; Pythagorism, 201,
204; tradition, 193
386
Quantum mechanics, 256
Quantum physics, 257
Quintilianos, Aristides, 185-86
Raczinski, Jean-Michel, 329fn
Raga, 198
Ragas, Indian, 268
Rahn,john, 268 n.
Random walk, 289
Ravel, Maurice, Bolero, 76
Real-time drawing, 329
Reichenbach, Hans, 257
Relativity, 257: theory of, 256
Repetition, 258,259,266,267
Rhythms, 264, 265, 266
Rule See Law
Rules, 260
Russell, Bertrand, 195,265
Scale, 268, 269: diatonic, 269, 276
See also Sieves
Scales of pitch, 264
Schaeffer, Pierre, 243
Scherchen, Hermann, 15, 24, 260
Schonberg, Arnold, 207-8, 243
Screens, 50-79; 108-9; construction,
66-68; definition, 51; elementary
operations, characteristics, 56-58;
linking screens, 69-78; summary,
56. See Markov chains
"Sequence" (in computing), 329
Serial music, 4, 8, 182, 186, 192,
208, 255; composition, 38;
method, 52; system, 204, 243;
Vienna school, 5, 8, 193
Serra, Marie-Helene, 329fn
Shannon, 255
Sieves, 265, 267, 268-76: elemental
displacement, 198; theory, 194-
200; transformations of, 275
Simonovic, C., 134
Sound, clouds, 12; as phase of a
musical work, 22-23; glissandi, 10,
13-14,32-36,55,140,167-68,213;
nature of, 43-50; sonic entities,
Index
definition as phase ofa musical
work and description, 22-25;
sound-points, 12-13; transforma-
tion of sets of sounds, 16; white
noise, 289. See Screens and Micro-
sound structure
Sound masses, 255
Space, 257,259,262,267:discon-
tinuity of, 263
Spatial Ubiquity", 256, 257
Spindel, P., 260
Statistical definition, 16
Steinhardt, Paul, 259
Stochastic, 10,39,43-44, 194, 295-
321; by computer, 131-44, ma-
chine-oriented interpretation of
Achorripsis, 134-43; computer-
programming and musical nota-
tions of, 145-54; construction, 4;
definition, 4; dynamics, 256;
music, 8, 12, 145-53, 182,255:
musical composition,S, 43; laws,
9, 12-16; process, 81, 244;
science, 8; synthesis, dynamic, 289-
93; See also Probability laws: ex-
ponential, Gaussian, linear,
Poisson
Stochastic music, 255, 289 ff., 295 ff.
Slocfws, 4, 92,94
Stoicism, 205
Strategic. See Game theory
Stravinsky, Igor, 5, 192
ST/10-2, 080262, 134, 138-39, 144,
154
ST/48-1,240162, 144
Symmetries, 268, 289; (repetitions),
269: See also Space
Syrmas, 80, 81 n.
Systems, dynamic, 293
Takahashi, Yuji, 175
Tempo, 276 See also Rhythm
Term, 69
Terretektorh, 217, 236-37
Thales, 201
TImbre, 266, 268
Index
lIrne, 256,257, 258,259,261,262,
263,265,266: flux of, 262,263,
264: irreversability of, 256
TInctoris, Johannis, 183
Tonality, 258
Transformation, 69-75, 257; anthro-
pomorphic, 261: definition, 69;
stochastic transformation, 73, 75
Transition, 69, 73-74
Transitional probabilities, 44
Transposition, 276
Tryon, Edward, 259
Uniform function. See Probability
laws
UPIC (Unite Polygogique Informa-
tion), xii, 329-334
Variations, 276
Variety, 61-63
Varese, 8, 243
Vilenkin, Alexander, 259
387
Voice editing (on the UPIC), 329
waveforms, 329-24
Wiener-Levy process, 24, 247
Wittgenstein, 203
Yang, 256
Zarlino, 202

Formalized Music

Uploaded by

Copyright:

Available Formats

Formalized Music

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Formalized Music

Uploaded by

Copyright:

Available Formats

Formalized Music

THOUGHT AND MATHEMATICS IN COMPOSITION

You might also like